~~~~~~~~~~~~~~~~~~

~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~

~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~

~~~~~~~~~~~~ ~ ~~~~~~~~~~~~~~~~~~~
~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~ ~
~ ~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~ ~~~~~~~~~~~ ~~ ~~~~~~~ ~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~
~~~~~~~~~~~~~~~~~~
~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~
~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~ ~~~~~~~~~~ ~ ~ ~~~~~~~
~~~~~~ ~ ~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~ ~~~~~~~~~~~~~~~~~~~
~~~~ ~~ ~~~~~~~~~~~~~~
~~~ ~~~~~~~~~~~~ ~~~~~~~
~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~~~~~~~~~~~~~~~~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~ ~~~~~~~

~~~~~~~~~~~~~~~~~~~

~~~~~~~~~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~ ~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~ ~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~ ~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~

~
~~ ~~~~~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~~~~~~~~~~~
~~~~~~~~

~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~~~~~~~~~~~~~~~~~~~~~~

~~~~~~~~~~~ ~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~ ~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~
~~~~~~~~~~~ ~
~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~ ~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~ ~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~
~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~
~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~ ~
~~~~~~~~~~~~~~~~~~~~~ ~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~
~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~

~~~
~

~~~~~~~~

~ ~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~ ~ ~ ~~
~~~~ ~~~~~~~~~~

~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~
~~~~~~~ ~~~~~~~~

~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~

~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~ ~~
~~~~~~~~~~~~

~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~ ~ ~~
~~~~~~~~~~~~~~~~~

~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~ ~ ~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~ ~~~~~~~~~~

~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~ ~~~
~ ~~~~~~~~~~~~~~~~~ ~~~~~~~~~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~
~~~~~~~ ~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~
~ ~~~~~~~~

~~ ~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~

~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~
~ ~~~~~~~~~~~~~ ~~~~~~~~~~~~~~~~~~~~~~~

~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~ ~~~
~ ~~~~~~~~~~~~~~~~ ~~~ ~~~~~~~~~~~~~~~~ ~~~ ~~~~~~~~~~~~~~~~~~ ~~~ ~~~~~~~~~~

~~~~~~ ~~~
MAGNETIC NERVE STIMULATION AND
EFFECTS OF MAGNETIC FIELDS ON
BIOLOGICAL, PHYSICAL AND CHEMICAL
PROCESSES

Shoogo Ueno and Masakazu Iwasaka

Institute of Medical Electronics

Faculty of Medicine, University of Tokyo
Tokyo 113, Japan

INTRODUCTION
Biological effects of magnetic fields are classified into three categories; the effects
of (1) time-varying magnetic fields, (2) DC or static magnetic fields, and (3) multiplication
of both static fields and other energy such as light and radiation. For each category, a different
strategic approach is required to shed light on the biomagnetic effects.
Time-varying magnetic fields produce eddy currents which stimulate excitable
tissues at low frequencies. Magnetic brain stimulation can be realized by this effect. The first
part of this paper focuses on magnetic nerve stimulation.
Biological effects of static magnetic fields have been poorly understood. Recognition
of the role of diamagnetic, paramagnetic and ferrimagnetic materials in the body may help
in unraveling the underlying mechanisms. The latter part of this paper focuses on the effects
of magnetic fields on the behavior of diamagnetic water and paramagnetic oxygen. The
embryonic development of frogs, blood coagulation, fibrinolytic processes and other bio-
chemical processes are also observed under strong magnetic fields up to 8 T

BIOMAGNETIC PHENOMENA
Biomagnetic phenomena in different intensities of magnetic fields and their frequen-
cies are shown in Fig. 1. Effects of magnetic fields on living systems and biological materials
have been observed mostly in the range of magnetic fields higher than the earth's magnetic
field. Recently, superconducting magnets are being used in MRI (magnetic resonance
imaging) systems, where human subjects are exposed to static magnetic fields of 1.5 T —2.0
T. More intense magnetic fields (4 T order) are used in advanced MRI systems. We observed
the phenomenon that surface of water is parted by static magnetic fields of up to 8 T Fibrin
Biological Effects of Magnetic and Electromagnetic Fields, Ediled by S. Ueno
Plenum Press, New York, 1996 I
 S. L'eno and M. Iwasaka

10
i
10
Magnetic Stimulation
(^ of the Heart (T=1ms)

-i Parting ot Water
Magnetic Orientation
MRI Magnet
\X^y
J^^/
\ ^
'''"S ^^
.
^ \ Magnetic Stimulation
p ,y^ of,, the Brain (T=0.1mi
Blood Flow Change via
. Magnetic Stimulation
ms)

c
CD Ma#ietophosphene of Sensory Nerves
Q
10 y' 2*
f Ca Release y"
Earth ^}-^ /
Consumer Electronics J
4)
C 10 n Magnetic Storm _ y
a>
(0 Urban Magnetic Fields Hyperthermia

10
D g (MPG) ^""""f^

Heart (MCG)

10
[ J _J Brain (MEG) Mobile Telephone
] Evoked Fields
SQUID > < ^ |
Sensitivitvy^'''<to.,.., I Brain Stem

10 J L
DC rr 1 10^ 10 10"

Frequency of Magnetic Field (Hz)

Figure 1. Biomagneiic phenomena.

fibers, as diamagnetic living materials, are oriented parallel to a direction of static magnetic
field of 1 - 11 T.
The so called magnetophosphene is a visual sensation caused by low frequency
magnetic field exposure which was first reported by D'Arsonval (D'Arsonval. 1896).
Lovsund et al reported that the threshold of magnetophosphene is 10 mT at 20 Hz. and they
claimed that the magnetophosphene is a reasonable example of the study of potential
biological hazardous effects of envirmental magnetic field effects (Lovsund et al, 1980).
Magnetic stimulation of the human brain requires a 1 T order of pulsed magnetic fields for
100 - 200 ^isec. Magnetic stimulation of the heart requires more strongly pulsed magnetic
fields for 1 - 2 msec. In Fig. 1, the frequencies of the pulsed magnetic fields are converted
from pulse width.
There are many studies on the effects of extremely-low-frequency (ELF) magnetic
fields on biological systems from the cellular level to mammals (Adey, 1981. 1993,
Blackman et al, 1985, Liboff et al, 1988, Barnes, 1988, 1990, Tenforde. 1991.1992.
Kirshvink et al, 1992, Liburdy, 1992, Kato et al, 1993). The low levels of ELF electromag-
netic fields are produced by consumer electronics.
The study on biological effects of higher frequency magnetic and electromagnetic
fields has become important in connection with the rapid increase of mobile telephones
(Ghandi. 1992).
Using SQUID (superconducting quantum interference device) techniques, magnetic
fields from the brain (Magnctoencephalogram, MEG), heart (Magnetocardiogram, MCG),
Magnetic Fields on Biological, Pliysical and Chemical Processes 3

and lung (Magnetopulmogram, MPG) can be measured. This chapter, however, does not
include these biomagnetic measurements.

MAGNETIC NERVE STIMULATION

Magnetic stimulation of excitable tissues has been studied by several investigators
(D'Arsonval, 1896, Bickford et al, 1965, Irwin et al, 1970, Maass et al, 1970, Oberg. 1973. ).
Among them, Maass and Asa proposed a transformer type of stimulation in which a nerve
bundle was threaded through a core as the secondary winding (Maass et al, 1970). They
demonstrated that the flux change in the core could be used to excite nerves. Oberg proposed
an airgap type of stimulation, in which a nerve bundle was exposed to alternating magnetic
fields (Oberg, 1973). Induced eddy currents in the membrane tissues could be expected to
stimulate nerves. Although these types of magnetic nerve stimulation were experimentally
demonstrated, the underlying nerve excitation processes following magnetic stimulation
were not yet understood (Ueno et al, 1978).
We carried out an experiment to measure the action potentials of lobster giant axons
under time-varying magnetic fields (Ueno et al, 1981, 1986). The axon membrane was
excited by galvanic stimulation, and the action potential was recorded intercellularly with
micro-electrodes. During the propagation of the action potential along the axon, alternating
or pulsed magnetic fields were applied across the middle of the axon in order to study whether
or not the magnetic fields had any effect on parameters such as conduction velocity and
refractory period of the nerve fiber and amplitude, duration and shape of the action potentials.
The results obtained from the lobster experiments suggest that nerve excitation by
magnetic field influence is mediated via the induction of eddy current in the tissue surround-
ing the nerve. The current density induced depends on geometrical factors as well as the
resistivity of the tissue in which the current flows. In other words, for nerve excitation, the
microscopic eddy currents that flow inside the microstructures of the axon membrane are
not very important. What is important are the macroscopic eddy currents that flow along the
nerve axon and in the tissues surrounding the nerve, as they contribute to the depolarization
of the membrane.
After the study of single nerve axons, we proposed a new type of magnetic nerve
stimulation (Ueno et al, 1984). An insulated magnetic core is implanted into the body with
a nerve bundle positioned on the core aperture. The nerve can be stimulated by the eddy
currents that flow in the body fluids around the core when the magnetic flux in the core is
changed. There is no interlinkage between the core and the nerves. Therefore, this method
is a good example for verifying that the nervous system responds to time-varying magnetic
fields through eddy currents induced in the body.
Magnetic stimulation of the human brain was first reported by Barker et al. (Barker
et al, 1985, 1986, 1987). Parameters of muscular potentials, such as conduction velocity,
latency and amplitudes were studied by a number of researchers by recording the elec-
tromyographic (EMG) responses to stimulation of the motor cortex (Day et al, 1987, Hess
et al, 1987a. 1987b, Mills et all, 1987, Rothwell et al, 1987). Single coils were used for these
studies. Current pulses were passed through a single coil placed outside the head, and eddy
currents induced into the head by the pulsed magnetic fields stimulated the brain. With this
method, however, broad areas of the brain are stimulated simultaneously. We developed a
method of localized magnetic stimulation of the human cortex (Ueno, Tashiro, and Harada,
1988).
The basic idea is to concentrate induced eddy currents locally in the vicinity of a
target by a pair of opposing pulsed magnetic fields which can be produced by a figure-eight
coil. Usingthismethod, we were able to stimulate the motor cortex of the human brain within
4 S. Leno and M. Iwasaka

a 5 mm resolution (Ueno, Matsuda, Fujiki, and Hori, 1989, Ueno, Matsuda, and Fujiki,
1989a). Since the concentrated eddy currents beneath the intersection of the figure - eight
coil flow in a direction parallel to the tangent of both circular coils, vectorial stimulation can
be achieved. The neural fibers can be excited easily when the fibers are stimulated by the
eddy currents that flow parallel to the nerve fibers.
Based upon this principle, we obtained functional maps related to the hand and foot
areas (Ueno, Matsuda, and Fujiki, 1989a, Ueno, Matsuda, and Fujiki, 1989b, Ueno, Matsuda.
and Hiwaki, 1990). We have observed that an optimal direction of stimulating currents for
neural excitation exists in each functional area in the cortex. We have also observed that the
functional maps of the cortex vary with the orientation of the stimulating current. To explain
the mechanism that is responsible for producing this anisotropic response to brain stimula-
tion, we developed a model of neural excitation elicited by magnetic stimulation (Ueno,
Matsuda, and Hiwaki, 1991).
The model explains our observation that the directions of the induced current vectors
reflect both the functional and the anatomical organization of neural fibers in the brain. We
need to emphasize that the point of excitation is not at a point under the intersection with
the coil. Instead it is a few ten mm away from the point of intersection; in the case in which
the coil diameter used is 50 mm, the distance between the coil and the nerve fiber is 10 mm.
This phenomenon was experimentally studied by Nilsson et al in magnetic stimula-
tion of a peripheral nerve (Nilsson et al, 1992). Magnetic stimulations of the human brain
and peripheral nerve were also studied by other groups (Amassian et al, 1989, 1992, Basser
et al, 1991, Maccabee et al, 1990, Olney, et al,1990, Esselle and Stuchly, 1992, Rothwell,
1994, Cohen et al, 1990, Reilly et al, 1989).
Magnetic stimulation of the spinal cord and heart were also studied (Ueno, Matsuda,
and Hiwaki, 1990, Ueno and Hiwaki, 1991, Bourland et al. 1990, 1992. Nyenhuis, 1994.
Hosono, 1992),

PARTING OF WATER BY MAGNETIC FIELD: MOSE'S EFFECTS

Water, which is a diamagnetic material, is one of the most general and abundant
diamagnetic fluid. It is an interesting problem for humans as to whether or not magnetic
fields have any influence on water. Diamagnetic water is often used for calibration and
correction in determining magnetic susceptibilities of materials with the magnetic balance
system. However, little has been observed about the hydrodynamic behavior of water or
diamagnetic fluids, in horizontal magnetic fields of 100 - 1000 T- / m order.
We investigated the dynamic behavior of water in high gradient magnetic fields. We
used a horizontal type of superconducting magnet 700 mm long with a 100 mm diameter
wide bore. The superconducting magnet has a distribution of high gradient magnetic fields
near the center of the magnet, as show in Fig. 2. When the magnet produced 8 T at its center,
the maximum product of the magnetic field and the gradient was 400 T"^ / m at z= 5 mm,
where the z-axis was directed to the bore axis. A water chamber, 50 mm wide, 60 mm high,
and 700 mm long was filled with distilled water.
When the water chamber was inserted into the bore of the magnet, we observed a
phenomenon in which the surface of the water was pushed back by magnetic fields of higher
gradients. Two "frozen" cascades were formed; the surface of the water near the center of
the magnet was parted, and the bottom of the water chamber could be seen. The water level
at both ends of the chamber was raised (Ueno and Iwasaka, 1994 a). We call this phenomenon
the Moses' effects. This phenomenon reminds us of the Biblical story from Exodus, i.e..
Moses parting the Red Sea.
Ma|neti« Fields on Biological, Piiyskal and Chemieal Processes

Water chamber
100 mm

Figure 2. Distribution of mag-

netic fields along the horizonla! or
z-axis in a superconducting mag- -200 0 200
net. Distance z [mm]

We also found the Moses' effect with agarose gels. The agarose solution was allom'cd
to gel in the superconducting magnet. Fig. 3 shows that the surface of the agarose gel was
fortBed in the same manner as in the water experiment.
To measure changes in the water level in the magnetic fields, one of the edges of the
water chamber was positioned at the center of the magnet, and the water level wa.s observed
by a video camera. Magnetic fields in the center of the magnet were changed from 0 to 8 T,
and a decrease in the water level was observed. The changes in the lowest level of water was
-22 mm when magnetic fields were changed from 0 T to 8 T as shown in Fig.4. The decrease
in water level was proportionate to B^ at the center of the magnet. This result shows that the
energy used for the formation of water-walls in a steady state is proportionate to the inag.netic
energy of the water (Ueno and Iwasaka, 1994 b).

"Water-wall"

Figure 3. Moses' effect. A visible ink was added

into the agarose solution, and the solution was al-
lowed to gel in the magnetic fields. 5 cm
 S. Ucno and M. bvasaka

-30

Calculated—>r-'

E -20 -
£
>
I

o3

Figure 4. Change in water surfaee level in

80 magnetic fields.

A Stress analysis was carried out to explain the mechanism of the Moses" effect. The
stress tensor of a diamagnetic fluid (in indicial notation) is given by

T"'„ n, = (p"' + (loH'.' 2 + poMH / 2) n,, 1

p"' = -|i,|MH 2 + |a„ {„" M dH - pgz + const, 2

where n = (n,,nj,n|j) is a unit vector that is perpendicular to the water surface, p'" is the static
fluid pressure, i^,, is magnetic permeability, p is mass per unit volume, g is gravitational
constant, M is the magnetization of diamagnetic fluid and H is a magnetic field (Ueno and
Iwasaka, 1994 a).
On the other hand, the stress tensor that is outside of the diamagnetic fluid is uiven
by

T",jnj=--(p"+H0H2/2)ni. 3

where p" is the environmental pressure.

On the surface of the water, T"",!, n^ equal to T",j Uj. Equations (1), (2) and (3) can be
combined to give

z = h = (Mo / rg) lO" M dH = z,„ Mo H- / 2rg. 4

Decreases in water levels calculated by equation (4) are shown in Fig. 4 and Fig. 5.
The hydrodynamics of diamagnetic fluid in ~40() T ' ' m is comparable with that of
ferromagnetic fluid in weak magnetic fields. Remarkable effects of high gradient magnetic
fields are expected in blood circulation, cerebral spinal fluid system and other intra and extra
cellular solutions.
The magnetic force acting on water molecules is given by

F= i B ^
Mo dz 5

where x is the magnetic susceptibility of water molecules, B is the magnetic flux density.
and Mo is the magnetic permeability of the vacuum. Molecular susceptibility of water is
-12.97 X 10"'' [cgs] at 293 K. The magnetic forces acting on 100 ml of water is shown in Fig.
Magnetic Fields on Biological, Physical and Chemical Processes

£ -10^

-30
Figure 5. Surface profile of diamagnetic water in -300 -200 -100 0 100 200 300
magnetic fields of up to 8 T. Distance [mm]

6. Amaxiniiim magactic force of 0.288 [N], m'hich is about one third of the force of gravity,
is obtained wliere BxclB/dz = 400 T^/DI.
Magnetic levitations of diamagoetic materials such as water, etlianol and acetone i0
magnetic fields of more than 20 T produced by a vertical type of superconducting magnet
were reported by Beaugnoo et al (Beaugnon et al, 1991). Distortion in the surface of liquid
helium by magnetic fields of 6.5 T was also studied (Frost et a!).
Studying the role of diamagnetic fluids in gradient magnetic fields is important in
understanding the mechanism of biological effects of magnetic fields. The phenomena
observed in the present experiments will provide a new aspect to applied pliysics and
biomagnctics.

N a

I -2Q0 0 200

E 0,2 \
o
o \
> 0.1
-^ \
£ \
E 0.0 kL.^ i 1 J^J
-200 0 200
Distance [mm]

IFI '^-

Figure 6. Magnetic force acting on water. IFimax = 0.288 {H / 1 0 0 mi] at C

° S. Ueno and M. Inasaka

REDISTRIBUTION OF DISSOLVED OXYGEN CONCENTRATION

UNDER MAGNETIC FIELDS

It is an interesting question whether or not magnetic fields have any influence on

dissolved oxygen in living tissues and cells. In several reviews and papers, the importance
of the role of oxygen has been suggested by Aceto et al (Aceto et al, 1970), Since oxygen
molecules and dissolved oxygen as paramagnetic molecules are an important role in living
systems, possible biomagnetic effects can be expected if spatial distributions of dissolved
oxygen concentration in the tissues are modulated by magnetic fields.
Based on this assumption, we have measured dissolved oxygen concentration under
magnetic fields using an electromagnet with 1.0 T with a gradient of 10 T/m (Ueno and
Harada, 1982), and a superconducting magnet with magnetic fields of up to 8 T and a gradient
of50T/m(Uenoetal, 1994).
Redistribution of dissolved oxygen was observed during both oxygen desorption and
absorption processes in a magnetic field of I.O T when the surface of the water contacted
the air atmosphere . It is understood that the oxygen in the atmosphere over the surface of
the water is redistributed first by magnetic fields, resulting in oxygen desorption absorption
rates being modulated by the spatial distribution of oxygen concentration in the air atmos-
phere (Ueno and Harada, 1982).
To obtain the direct effects of magnetic fields on the redistribution of dissolved
oxygen in water, we used a horizontal type of superconducting magnet (Ueno et al, 1994).
A closed water chamber 100 % filled with distilled water which was separated from the air
atmosphere, was moved together with a DO (dissolved oxygen) sensor in the direction of
the bore of the magnet. This experiment, however, might include the undesirable factor of
electrochemical processes in the DO sensor possibly being affected by magnetic fields.
The present study focuses on the dynamics of dissolved oxygen in water under
magnetic fields of up to 8 T, with the gradient of 50 T/m. The dissolved oxygen concentration
is measured by a DO meter before and after magnetic field exposures in order to exclude
undesirable interactions.

-200 0 200
Distance [mm)

Superconducting magnet

1 1 \ 1 1 1 100 mm
1 1 2 I 3 I 4 I S I 6 I 7

i:x:i

U U Figure 7. Experimental set-up for redistribution

700 mm of dissolved oxygen in magnetie fields of up to 8 T.
^ >,
 Magnetic Fields on Biological, Physical and Chemical Processes

Figure 8. Redistribution of dissolved oxygen concentration

by magnetic fields of up to 8 T. The concentration of dissolved
oxygen before magnetic field exposure was controlled by 1 2 3 4 5 6 7
introducing oxygen gases into the water. Section number

The experimental set-up is shown in Fig. 7. We used a horizontal type of supercon-

ducting magnet 700 mm long with a bore 100 mm in diameter. When the magnet produced
8 T at its center, the maximum product of the magnetic field and the gradient wa.s 400 T- m
at 5 mm, where the z-axis is the bore axis.
An acrylic water chamber, 700 mm long, 70 mm high, 50 mm wide was 100 "o filled
with distilled water, and the water chamber was sealed with an acrylic cover to separate the
water from the air atmosphere. The water chamber consisted of 7 sections, and the length of
each section was 100 mm. Partitions, i.e., dividers, were removable.
The dissolved oxygen concentration was controlled by introducing oxygen gases into
the distilled water in the range from 8 mg/1 to 26 mg/1. The water chamber without dividers
was placed in the magnet's bore, and the water was exposed to magnetic fields of up to 8 T
for 60 min. After the water chamber was taken out of the magnet, the water in the chamber
was divided by six acrylic dividers into seven sections. The dissolved oxygen concentration
at each section was measured with a Clark type DO (dissolved oxygen) meter.
To obtain dynamic behaviors of dissolved oxygen during and after magnetic field
exposures, time-courses of changes in dissolved oxygen concentration were measured in the
4th section, changing the exposure periods.
Fig. 8 shows the redistribution of dissolved oxygen concentration by magnetic fields
of up to 8 T for 60 min. The dissolved oxygen concentration before magnetic field exposures.
i.e., an initial concentration, was controlled from 11 mg/1 to 22 mg/1. The horizontal axis
shows the section number of each section divided by acrylic dividers. The dissolved oxygen
concentration in sections 3 — 5 at the central part of the magnet was increased.
Fig. 9 shows the normalized data of the redistribution of dissolved oxygen concen-
tration. The data were averaged over 8 trials where the dissolved oxygen concentration at

1.10

o =c 1.05
c =
o >,
» - CO

ot 1.001
Figure 9. Effect of an 8 T magnetic field exposure for 60 .c
min on distribution of dissolved oxygen concentration. The o
data were averaged over 8 trials where the dissolved oxy-
0.95 3 4 5 6 7
gen concentration in section 1 and 7 fell in the range 14mg'l
to 16 mg/1. Section number
10 S. L'eno and M. Iwasaka

Figure 10. Effect of magnetic fields of up to 8 T on changes

in concentration of dissolved oxygen in water. The concen-
tration of dissolved oxygen before magnetic Held exposure
Initial concentration [mg/l] was changed from 8 mg.l to 26 mg. 1,

both edges of the chamber fell in the range of 14 mg/l— 16 mg/l after magnetic field exposures
of 60 min. The results show that the increase in dissolved oxygen concentration was 5 %.
Fig. 10 shows the effect of initial concentration on the redistribution of dissolved
oxygen concentration. When the initial concentration changed from 8 mg/i to 26 mg/l. the
change of dissolved oxygen was 0% to 13 %.
The magnetic force acting on a group of oxygen molecules is described as follows.

X n dB ,. .. . .
— B 3—, (in z-direction)
Ho dz

where x is the magnetic susceptibility of a group of oxygen molecules, B is the magnetic

flux density, and \1Q is the magnetic permeability of the vacuum. For oxygen molecules, x =
3449 X 10^ [cgs] at 300 K. The magnetic force acting on 1 molar of oxygen is shown in Fig.
11.
High gradient magnetic separation (HGMS) has made it possible to separate micron-
sized paramagnetic particles from solution. For instance, Melville et al succeeded in
separating red cells from whole blood by HGMS when the hemoglobin was in the completely
deoxygenated state (Melville et al, 1975) . Friedlaender et al captured small paramagnetic
particles such as Mn203 and CriOj by HGMS (Friedlaender et al, 1978).
The magnetic energy of an oxygen molecule is 3 x 10 '-'' [J] at I T and 2 x 10'-'' [.I]
at 8 r, respectively. Thermal energy at 300 K is 4 x 10 -' [J] . It is estimated that oxygen
molecules drift in a particular direction in magnetic fields of more than 8 T when at least
2000 oxygen molecules aggregate.
Oxygen molecules are attracted by a maximum magnetic force where BxdB/dz is
largest in magnetic fields. When there are two maximum peaks in a magnetic force as shown
in Fig. II, dissolved oxygen molecules are trapped in the area between the two peaks

-200 200 Figure 11. Magnetic force acting on dissoK ed oxy-

Distance [mm] gen molecules per mol.
Magnetic Fieldi on Biological, Physical and Chemical Processes 11

resulting in the redistribution of dissolved oxygen concentration. In other words, spatial

distribution of dissolved oxygen concentration conforms to a single peak distribution in a
steady state, similar to the distribution of magnetic fields, but not to the distribution of
magnetic force. In contrast, in the case with captures of paramagnetic particles by HGMS
techniques, trapped particles accumulate in the vicinity of wires where the magnetic force
is the strongest.
liffects of high gradient magiicric fields on biological systems are expected 'Alien a
living tissue contains a high concentrarioii of dissolved oxygeii. For instance, the oxygen
effect is a piienomeaon when radiation iniury of tissues with a higli concentration of oxygen
becomes serious, compared with the case of a low concentration ofoxygen^ If the beha\iors
of dissolved oxygen could be controlled by magnetic fields, a new type of cancer ilierapy
might be contrived (IJeno and Ffarada, 1986).

EFFECTS OF MAGNETIC FIELDS ON GAS-FLOW AND

COMBUSTION
In relation to oxygen dynamics in water, we observed oxygen dynamics in air.
Combustion is an oxidation reaction which involves both a burning phenomena in the air
and cell respiration in the living body. The effects of magnetic fields on combustion of
hydrocarbons and alcohol with the aid of platinum catalysis have been studied to simulate
in part the oxidation of organic matter in the living body, and we observed that the
combustion velocities of gasoline and alcohol were influenced by magnetic fields (Ueno et
al, 1985, 1986). In order to explain this pheomenon, we examine the effect of gradient
magnetic fields on burning flames. A candle was burned in the airgap between magnetic
poles, and the candle flames were exposed to gradient magnetic fields. During magnetic field
exposures, the flames were pressed down, and the shape of the flames changed like a
mushroom (Ueno and Harada, 1987). The field intensity in the center of the airgap was 1.2
T.
Apart from the combustion experiments, we have observed that the flow of gases
such as carbon dioxide, nitrogen, oxygen and argon are blocked or disturbed by magnetic
fields. A model called a "magnetic curtain" has been introduced to explain these phenomena.
It is assumed tiiat the magnetic curtain is a wall of air caused by magnetic fields as shown
in Fig. 12.
To clarify the mechanism of the magnetic curtain, trajectories of gas flow near and
inside the magnetic curtain are simulated on the basis of molecular dynamics (Ueno and
Iwasaka, 1990, Lfeno et al, 1993). It is assumed that gas molecules such as nitrogen and
oxygen molecules are particles, and these particles move in two-dimensional space. The
Brownian movement of particles in air atmosphere is caused by molecular collisions. We

Gas flow

M-^neti&v
curtai,, \

Magnetic field off f»lEgnetic field on

Figure 12. Formation of the magnetic curtain.
(a) (b)
12 S. Ueno and \1. Iwasaka

Gas flow |oir

G a s flow I 0 1 mm

Figure 13. Trajectories of nitrogen molecules in gradient mag-

netic fields, (a) without magnetic field, (b) with magnetic fields.

assumed that the dynamic movement of paramagnetic molecules such as oxygen is con-
strained by a magnetic force.
Let us consider the condition in which 10 particles start together at an upper starting
line and go down with a velocity of mean free path. Each particle receives random
acceleration through the collision. When a magnetic force is applied to the particle flow, the
velocity component of the z axis is changed by the collision with oxygen particles. Fig. 13
(a) shows the trajectories of gas flow with 10 million collisions. When the region is exposed
to a gradient magnetic field of 225 T-/m, the gas flow is blocked, as shown in Fig. 13 (b).
We have designed an electromagnet with a pair of columnar magnetic poles in which
inner sidepieces were hollowed out. A candle was burned in the hollowed space between the
magnetic poles, and candle flames were exposed to magnetic fields. The flames were

,.r - Magnetic Curtain

^^^^^m
Figure 14. A mouse in a space surrounded by
Magnetic Curtain magnetic curtains.
Magnetic Fields on Biological, Physical and Chemical Processes 13

quenched a few seconds after the onset of field exposures (Ueno, 1989). The interception of
oxygen by the magnetic curtain quenches the flames. In place of the candle, if a mouse or a
human subject is positioned inside the magnetic cartain, as shown in Fig. 14, we have to be
careful about the respiratory system being disturbed in this condition.

EFFECTS OF MAGNETIC FIELDS ON FIBWN POLYMERIZATION

AND FIBMINOLYSIS
Fibrinogen, which is the prime factor in blood coagulation, changes to fibrin
monomer by the action of protease thrombin, and fibrin monomers are polymerized. Fibrin
polymers are diamagnetic materials that arc oriented in a magnetic field (Torbct, 1981,
Yamagishi et al, 1989, Ueno et al, 1993). In the course of the polymerization process, when
a magnetic field of intensity 10 to 20 T is applied, the fibrin fibers orient parallel to the
magnetic fields. When no magnetic field is applied, fibrin fibers become entangled in mesh,
and no orientation is observed. This is an eftect of strong static magnetic fields on the blood
coagulation system.
On the other hand, the fibrinolytic process also occurs in vessels in a mutually
compensating and balanced state with coagulation. Plasmin, which is a serine proteinase,
reacts with fibrins to digest peptide bonds by hydrolysis. Fibrin gels are dissolved when
polymerized fibrins are degraded to fragments by plasmin.
A study on the positive effect of weals magnetic fields (-0.3 T) on fibrinogen
degradation products level in rabbits in vivo was reported (Gorczynska, 1986).
In our study, fibrinolytic processes in magnetic fields were investigated using a fibrin
plate method, whereby mean levels of FDPs (fibrin degradation products) in solutions are
measured.
We investigated the effect of magnetic field exposure at 8 T on the dissolution of a
fibrin-plate (Iwasaka et al, 1994 a). Thrombin (80 nl, 3.1 NIH units/ml) was added to 3-ml
of fibrinogen (4.8 mg/ml in 0.05M Tris-HCl buffer, pH 7.4, containing 0.10 M NaCl). This
solution was divided into two dishes, and incubated at 25 °C for 60 min without magnetic
fields. A hole was made in the centers of the fibrin plates with a gel puncher, and 20 fil of
plasmin (4 casein units/ml) was added to each hole. Fibrin plates were incubated at 37 °C
for 15 hours, either with a magnetic field at 8 T or without magnetic fields, and dissolution
of both types of fibrin was observed. Mean levels of FDFs (fibrin degradation products) in
solutions were obtained. Fig. 15 shows the mean levels of FDPs in samples exposed to an 8
T magnetic field and in those not exposed. IMean levels of FDPs in samples exposed to an 8
T magnetic field were on the average 15 % higher than those not exposed. The same

8.0

c.
S 4.0 n=7
n = 10

Flgire 15. Effects of gradient magnetic fields on fibri- ii

noiysis by fiiasmin. Levels of fibrin degradation products o.O
are shown. Fibrin plates were incubated at 37 °C for 15 Control
hours, witii and without tnagaetic field.s. ~ eo T^/m
14 S. Ueno and M. luasaka

experiment was carried out in gradient magnetic fields (gradient fields: B x dB/dz = 350 -
400 T' m). The mean level of FDPs released in a gradient magnetic field of up to 400 T' /
m was more than 190 % of the control samples. In this case, there are large gradients along
the horizontal line. Solutions containing water, plasmin and fibrin could diffuse to horizontal
directions to promote the reactions of fibrinolysis. The results show that fibrin gels and water
containing plasmin dilTused to less intense magnetic fields, and fibrin dissolved in specific
directions. It is possible to vectorially control the dissolution of a fibrin clot by magnetic
forces.
We also carried out an experiment in order to understand how the fibrin oriented in
a magnetic field dissolves (Ueno et al, 1993). FDPs in the dissolved holes in the fibrin
prepared in either the presence or the absence of an 8 T magnetic field were assayed. Fibrin
gels formed with a magnetic field were more soluble than those formed without a magnetic
field. It was observed that the shapes of holes in dissolved fibrin changed to ellipsoidal
patterns when fibrin plates were formed with a magnetic field at 8 T. The transversal axis of
the ellipse was parallel to the magnetic fields.
As a possible mechanism of these effects, changes in concentrations of diamagnetic
macromolecules in a solution in gradient magnetic fields were examined. We carried out an
experiment to measure changes in the concentration of macromolecules (Iwasaka et al, 1994
b). A solution of fibrinogen and thrombin in a plastic tube made of vinyl chloride, 10 mm in
diameter and 310 min long, was coagulated within 12 hours in magnetic fields up to 8 T (B
-dB/d2<400 f-'m). Coagulated fibrin was frozen at -20 °C, and divided into 1 5 fractions.
Each fraction of the frozen fibrin had the same volume. After incubation with 1ml of 3.3 M
urea for 12 hours at 37 °C, we measured levels of solubilized fibrin polymers as absorbance
at 280 nm. Fig. 16 shows a distribution of concentrat'ons of fibrin in gradient magnetic fields.
The vertical line shows the absorbance at 280 nm corresponding to the concentrations of
fibrin. The absorbance of dissolved solutions at 280 nm obtained from fibrin coagulated in
5-7 I were 20—50% higher than those in 1 T fields. The results suggested that fibrin polymers
in a solution drifted in a specific direction, and concentrations of the fibrin changed, with
the result that the level of solubilized fibrin polymers were inhomogeneous in a chamber. It
is well known that electrophoresis of macromolecules is observed in electric fields. 1 he
phenomenon observed here indicate that "magneto-phoresis" of fibrin polymers occured in
gradient magnetic fields of 400 T'/m.

100 200 300

Distance [mm)
1- Fraction number -15
ri" "1 XXJ__J__L_L_J
Solution of fibrinogen and thrombin in plastic tube
0.9

o 15 0.5 I-
SE

0.0 L . M ._.LA_L I I ' I I I L I..J_

Figure 16. Magnetophoresis of fibrin polymers under
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
magnetic fields of up to 8 T.
Fraction number
Magnetic Fields on Biological, Physical and Chemical Processes 15

When a solution, containing water and diamagnetic macromolecuies is exposed to

gradient fields, magnetic forces act on the water and diamagnetic macromolecuies. The
distance of drifts of a macromolecule in the water in a unit time is described as follows:

/ = ll(Xv-Xw)Vn„A(H=)-dt-2m 7

where t is time, V is volume of fibrin polymer, Xv is magnetic susceptibility of fibrin per

volume, Xw 's a magnetic susceptibility of water per volume, and m is inass of fibrin polymer
per unit volume.
Magneto-phoresis of fibrin polymers in gradient magnetic fields occurred when
magnetic energy -(/(v^Xw)VnnH' II was larger than thennal energy kT. We conclude that
diamagnetic macromolecuies in water drift in a specific direction due to the difference in the
diamagnetic susceptibility of fibrin and water.
It is possible for large macromolecuies, such as fibrins and fibrin polymers, to be
affected by gradient magnetic fields. To investigate the effect of homogeneous magnetic
fields on enzymatic activity of plasmin, we measured a synthetic substrate hydrolysis with
plasmin (Iwasaka et al, 1994 c). As the synthetic substrate, D-Val-Leu-Lys-pNA (S-2251;
molecular weight is 551.5) is a small molecule compared to fibrin, it is expected that the
effects of gradient magnetic fields can be ignored. In order to reduce the effect of gradient
magnetic fields, the center part of the superconducting magnet's bore, where dB/dy < 4 T/m
and B = 8 T was used. The concentration of S-2251 and that of plasmin was 0.1 mM and
0.023 casein units/ml, respectively. After incubation at 37°C 200 ul of 99.9% acetic acid
was added, and the absorbance was measured at 405nm with a spectrophotometer. An 8 T
magnetic field did not have a distinct effect on ether the maximum velocity (Vmax) or the
Michaelis constant (Km). However, when the absorbance changed non-linearly and reached
60—90 % of the maximum absorbance, absorbance of the mixture exposed to 4 - 8 T for 60
min at 37°C was lower than control samples (maximum 11 %). The results show that a
long-term exposure to magnetic fields inhibited the synthetic substrate hydrolysis with
plasmin.

EMBRYONIC DEVELOPMENT OF XENOPUS LAEVIS UNDER

MAGNETIC FIELDS
We have studied a possible influence of intense magnetic fields on the early embry-
onic development of frogs (Ueno et al, 1984, 1990). Some of the most serious hazardous
effects that could be induced by intense magnetic fields are teratogenic etTects on developing
embryos. Embryos ofXenopus laevis were exposed to magnetic fields of up to 8 T for the
period from the pre-cleavage stage to neurula stage. Embryos were then cultured in Brown-
Caston's medium until the feeding-tadpole stage (Ueno, Iwasaka and Shiokawa, 1994).
We carried out an experiment, in which 100 embryos were cultured for 20 hours in
magnetic fields. No apparent teratogenic effects were observed when embryos were cultured
for 20 h from the stage of uncleaved fertilized egg to the neurula stage under magnetic fields
of 8 T We conclude that static magnetic fields of up to 8 T do not appreciably affect the rapid
cleavage and the following cell multiplication and differentiation in Xenopits laevis.
We have also studied the early embryonic development oiXenopus laevis in a 40 nT
magnetic field, or one-thousandths of the earth's magnetic field, and obtained negative
results. Thus, again under this very low magnetic field, fertilized eggs developed normally
and formed tadpoles with no appreciable abnormality.
It is a miracle or paradox that animals such as Xenopus laevis develop normally under
strong magnetic fields, although we have observed very clear biomagnetic effects such as
16 S. L'cno and M. Iwasaka

parting water by magnetic fields, magnetic orientation of diamagneic polymers, and redis-
tribution of dissolved oxygen by magnetic fields. It seems that biological cells arc robust
against magnetic fields.

GENETIC EFFECTS OF MAGNETIC FIELDS ON DROSOPHILA

MELANOGASTER
The genetic effects of magnetic fields on somatic reversion and somatic recombina-
tion in Dmsophila melanogaster have been reported (Levengood, 1966, Mittler, 1971, Koana
et al. 1994, Yoshikawa et al, 1994).
We examined the somatic reversion of the white locus and the somatic recombination
ofmwh and//v genes in Drusophila melanogaster after they were exposed to a static magnetic
field at 8 T for 8 hours or exposed to an alternating magnetic field (3.3 mT) at 20 Hz for 8
hours (Yoshikawa, Iwasaka and Ueno, 1994). We have developed a mutation assay system
in which both types of mutations are genetically combined so as to detect mosaic spots on
eye facets and on wing blades in single individuals. The reverse eye color mutation from
white-ivoiy (w') to wild type fw*) is associated with the loss of a 2.9-kilobase DNA fragment
duplicated in the white locus on the X chromosome. The formation of mutant spots on the
wing blades results from recombination at wing anlage cells in larvae, trans-heterozygous
for the mutations multiple wing hairs (mwh) and/lare (fir) on the 3rd chromosome. The static
magnetic fields with 8 T were effective in producing the eye spots caused by intragenic
mutation and the wing hair spots resulting from somatic recombination. In the exposure
groups, eye spots occurred 1.7 times more frequently than in the control groups, and wing
hair spots 1.4 times more. In contrast, the alternating magnetic fields with 3.3 mT at 20 Hz
were not more efficient for induction of both somatic spots. Mutant clones induced at the
beginning of larval life will be large in size while those produced in the second or third instar
will be successively smaller Hence, classification of mosaic spots into size classes and
subsequent analysis of clone size distribution can be used to estimate the point in time of the
induction of the mutational events. The increased number of mosaic spots in size classes 1,
2, and 3 ^ of the groups that exposed to an 8 T magnetic field were characteristic when
compared to the control groups. It is possible that the period of these increments may
correspond to the third instar of the larval stage. It seems reasonable to conclude that an 8
T magnetic field induced reverse mutation of eye color.
A wing spot test in Drosophila melanogaster was also carried out by Koana et al
(Koana et al, 1994). It was reported that static magnetic fields of 5 T increased chromosomal
recombination 50 %.

MAGNETIC FIELD EFFECTS IN SPIN CHEMISTRY

Possible biomagnetic and chemical effects can be expected when biological systems
are exposed to both static magnetic fields and other energy such as light and radiation (Ueno
and Harada, 1986). Photochemical reactions produced by a radical-pair intermediate in
solution can be exposed to show magnetic field effects that arise from an electron Zeeman
interaction, electron-nuclear hyperfine interaction, or hyperfine interaction mechanism
including an electron-exchange interaction in a radical-pair intermediate. That is, a common
mechanism of magnetic field effects on photochemical processes is that a chemical yield of
the cage- or escape-product comes to show a magnetic field dependence when a singlet-trip-
let intersystem crossing can be subject to magnetic perturbations. The magnetic field effects
Magnetic Fields on Biological, Physical and Chemical Processes 17

on photochemical reactions were verified (Hata, 1976, Tanimoto et al. 1976, Shulten. 1976.
Nagakura and Molin, 1992).
Chemical reactions in the skin and eyes may have potential biomagnetic effects when
the tissues are irradiated by both laser and magnetic fields (Ueno and Harada, 1986). If a
process of enzymatic reactions involve a radical-pair, there is a possibility that magnetic
fields affect the singlet-triplet conversion rate of radical pairs. Some enzymatic reactions
which involve photochemical processes are the candidates for the magnetically-sensitive
enzymatic reactions.
The question of whether magnetic fields affect enzymatic activities or not is of
considerable interest in biochemistry. To clarify the effect of magnetic fields on enzymatic
reactions in living systems, it is important to study the enzymatic reactions that involve
radicals which are generated from non-photochemical processes. We examined the possible
effects of static magnetic fields on biochemical reactions catalyzed by xanthine oxidase and
by catalase.
Haberditzl reported that magnetic fields of up to 6 T enhanced the activity of catalase
4.9 ~ 52% (Harberditzl, 1967). Recently, it has been reported that magnetic fields of 0.10 -
0.15 T affected B,, ethanolamine ammonia lyase (Harkins and Grissom, 1994).
Xanthine oxidase, contained in the liver, lungs, intestine and other organs, catalyzes
the degradation of hypoxanthine to xanthine, and xanthine to uric acid, which is the terminal
waste of purine nucleotides in mammals. During the oxidation of xanthine, the enzyme
releases superoxide anion radicals as intermediates which reduce ferricytochrome c (Fe-'*).
Superoxide anion, as well as any type of free radical, is also paramagnetic. Reduced
cytochrome c (Fe^*) has an absorbance maximum at 550 nm which can be detected by a
spectrophotometer. Superoxide dismutase inhibits the reduction of cytochrome c by accel-
erating the dismutation of superoxide anions to hydrogen peroxide and oxygen, and the
catalase converts hydrogen peroxide into water and molecular oxygen.
We observed that magnetic fields of up to 1.0 T did not alter the reduction rate of
cytochrome c by superoxide anion which was produced by the reaction catalyzed by xanthine
oxidase (Ueno and Harada, 1986). It indicates that neither the electron transfer from xanthine
to molecular oxygen nor the transfer from superoxide anion to cytochrome c was affected
by the magnetic fields of this range.
Catalase, which is an important heme protein in the body used to decompose toxic
hydrogen peroxide, contains a ferric state (Fe^*) iron atom, therefore its magnetic property
shows paramagnetism. The activity of catalase can be inhibited by hydrogen cyanide,
potassium cyanide and others.
In the case of catalase, the influence of magnetic fields on the activity was determined
by the catalytic decomposition of hydrogen peroxide. The rate of disappearance of hydrogen
peroxide was followed by observing the rate of decrease in absOrbance at 240 nm using a
spectrophotometer. Magnetic fields of up to 1.0 T did not alter the rate of decomposition of
hydrogen peroxide catalyzed by catalase (Ueno and Harada, 1986). These biochemical
reactions were also exposed to gradient magnetic fields of the same intensities as used in the
combustion experiments. However, no clear positive effects were observed.

ACKNOWLEDGMENTS
Collaborators in this work include Dr. T. Matsuda, Dr. O. Hiwaki and Dr. M. Fujiki
on magnetic brain stimulation, Dr. H. Tsuda on the hematological experiment. Dr. M.
Nakamura on the enzymatic activity in magnetic fields, and Prof K. Shiokawa for the
embryogenetic experiment. The work was supported in part by grants from the Ministry of
Education, Science and Culture in Japan.
18 S. IJeno and M. Iwasaka

REFERENCES
H. Aceto, Jr., C. A. Tobias, and I. L. Silver. 1970. Some studies on the biological ctTects of magnetic fields.
IEEE Trims, on Xfcign.. MACi-6. 2: .3(i8-."!7.1
W. R. .-Xdey. 1481. Tissue interactions with noninonizing electromagnetic fields. Physiol. Rev.. 61:4.^.'i - .S14.
W. R. Adey. 1993, HIectromagnetic technology and the future of bioelectromagnetics. in Elcciridiy ami
Magnetism in Biology und Mt'dicine, eiiiled hy M. Blank .\0\ - 108.
V. F;. .Amassian. R. Q. Cracco. and R J. Maccabee, 1989, Focal stimulation of human cerebral concx with the
magnetic coil: A comparison with electrical stimulation. Etecli>enccpluilography and cliniial Neuro-
physiology. 74 :4()1-4I6.
\'. H, .'\massian. R. Q. Cracco. R J. Maccabee. and J. B. Cracco. 1992. Cerebello-frontal cortical projections
in humans studied with the magnetic coil. Eleeloenccphalography and clinical Seurophyslologv, 85
:265-:72.
A. T. Barker, R. Jalinous, and I. L. Freeston, 1985, Non-invasive magnetic stimulation of human motor cortex.
The Lancet, i.: 1106-1107.
A.T, Barker. l.L, Freeston. R. Jalinous. and J..'\. Jarratt, 1986, Clinical evaluation of conduction time
measurements in central motor pathways using magnetic stimulation of the human brain. The Lancet.
1.: 1325-1.326.
A.l. Barker. l.L. Freeston. R. Jalinous, and J. A. Jarratt, 1987. Magnetic stimulation of the human brain and
peripheral nervous system: An introduction and results of an initial clinical evaluation. Neiirosurgerv.
20 I No. 1): 100-109.
I'.J, Basser. R. Jalinous. and B. J. Roth. 1991. Stimulation of myelinated nerve axon by electromagnetic
induction. Med. & Biol. Eng. & Computing. 29: 261-268.
F. S. Barnes. 1988, Mechanism of interaction of magnetic fields with biological systems, IEEE Trail'., on
.\^agnelics. 24, 4:2101-2104.
F. S. Barnes, 1990. The effects of the time varying magnetic fields on biological materials. IEEE Trans, on
Magnetics. 25. 5 :2092-2097.
F. Heaugnon and R. Tournier. I'Wl. Levitation of organic materials. Nature. 349. 7.470.
R.Cj Bickford and B.D. Fremming. 1965. Neuronal stimulation by pulsed magnetic fields in animals and man.
Dig. 6th Int. Con/. Med Electr Biol. Eng. 112.
C, F. Blackman. S. G. Bcnane. J R . Rabinowitz. D. H. House, and W. T. Joines. 1985. .'\ role for the magnetic
field in the radiation-induced efJlux of calcium ions from brain tissue in vitro.. Bioelectromagnetics.
6:327-337.
J. D. Bourland.G. A. Mouchawar. J. A. Nyenhuis. L. A.Geddcs, K. S. Foster,and G. RGrabcr, 1990,Transchest
magnetic (eddy-current) stimulation of the dog heart, Med. t& Biol. Eng. & Ctnnputing. 28: 196-19S.
J. D. Bourland, G. A. Mouchawar, J. A. Nyenhuis, L. A. Geddcs, K. S. Foster, and G R Graber. 1992,
Closed-chest cardiac stimulation with a pulsed magnetic field, Med. & Biol. Eng. & Computing. 30:
162-168.
.M. S. Cohen, R. M. Weisskoff, R. R. Rzedzian. and H. L. Kantor. 1990, Sensory stimulation by time-varying
magnetic fields. .Magn. Reson. Med.. 14. 3: 409 -414.
A. DWisonval. 1896, Dispotifs pour la mesure des courants alternatifs des toutes frequences, C. R. .icad. Sci..
48: 4 5 0 - 4 5 1 .
B.L. Day, RD. Thompson, J.R Dick, K. Nakashima, and C D . Marsdes, 1987, Different sites of action of
electrical and magnetic stimulation of the human brain. Neuroscience Letters. 75: 101-106.
K. P. Esselle and M. A. Stuchly. 1992. Neural stimulation with magnetic fields: analysis of induced electric
fields. IEEE Trans. Bioined. Engineering. BME-39: 693-700.
F.J. Friediaender. M. Takayasu. J. B. Rettig, and C.P. Kentzcr. 1978. Panicle flow and collection process in
single wire HGMS studies. IEEE Trans. Magn.. MAG-14 :1158-1164.
J. Frost. M.J. Lea. and R Fozooni. 1991, Distortion of liquid helium surface in inhomogcneous magnetic field.
Cno.ifivHc.v. 3 1 . 8 9 0 - 8 9 1 .
O. P. Gandhi and J. Y. Chen. 1992. A frequency-dependent finite-dift'erence time-domain formulation for
induced current calculations in human beings. Bioelectromagnetics, 13 :543 - 555,
F. Gorczynska. 1986. The effect of static magnetic field on fibrinogen degradation products level in rabhils
with thrombosis, J. Hygiene. Epidemiology. Microbiology and lmmunolog\\ 30. 3: 269-273.
VV. Haberditzl, 1967, Fnzyme activity in high magnetic fields, Nature. 213:72-73.
T. T. llarkins and C. B. Grissom. 1994, Magnetic field effects on B12 ethanolamine ammonia lyase: evidence
for a radical mechanism. Science. 263: 958-960.
N. Hata, 1976. The effect of external magnetic field on the photochemical reaction of isoquinoline N-oxide.
Chem. Lett. :547-550.
Magnetic Fields on Biological, Physical and Chemical Processes 19

C.W. Hess, K.R. Mills, and N.M.F. Murray, 1987a, Measurement of central motor conduction in multiple
sclerosis by magnetic brain stimulation. The Lancet. : 397-419.
C.W. Hess, K..R. Mills, and N.M.F. Murray, 1987b, Responses in small hand muscles from magnetic stimulation
of the human brain, J. Physiol, 388: 397-419.
A. Hosono, T. Goto, F. Okumura. T. Takenaka, M. Yamaguchi, T. Andoh, T. Kawakami. K. Takayama, S. Ueno.
and I. Yamamolo. 1992, Effective combination of stimulating coil for magnetic stimulation, Jpn. J.
Appl. Phys., 31,11: 37.'i9 - 3762.
D.D. Irwin, S. Rush, R. Evering, E. Lepeschkin, D.B. Montgomery, and R.J. Weggcl, 1970, Stimulation of
cardiac muscle by a time-varying magnetic field, IEEE Trans. Magnetics MAG-6; 321-322.
M. Iwasaka, S. Ucno and H. Tsuda, 1994a, Effects of magnetic fields on fibrinolysis,./ .ippl. Phvs.. 75. 10:
7162-7164.
M. Iwasaka, S, Ucno. and H. Tsuda, 1994b, Diamagnetic properties of fibrin and fibrinogen. IEEE Trans, on
Magn., 30, 6: 4695-4697.
M. Iwasaka, S. Ueno, and H. Tsuda, 1994c. Enzymatic activity of plasmin in strong magnetic fields. IEEE
Trans, on Magn.. 30, 6:4701-4703.
M. Kato, K. Honma, T. Shigemitsu, and Y. Shiga, 1993, Effects of exposure to a circularly polarized 50-Hz
magnetic field on plasma and pineal melatonin levels in rats. Bioclcctromagnetics. 14, 2: 97 - 106.
J. L. Kirshvink, A. Kobayashi-Kirshivink,andB. J. Woodford, 1992, Magnetite biomincralizalion in the human
brain, Proc. Natl. .Acad. Sci. USA. (Biophysical. 89, 86: 7683-7687.
T. Koana and M. Nakagawa, 1994, Exposure to static magnetic field increases the frequency of somatic
chromosomal recombination, BEMS Abstract Book of I6ih Annual Meeting , P-96.
W. C. Levengood, 1966, Cytogenetic variations induced with a magnetic probe. Nature. 209. 1009- 1013.
A. R. Liboff and B. R. McLeod. 1988, Kinetics of channelized membrane ions in magnetic fields, Bioelectro-
magnetics. 9: 39-51,
R. P. Liburdy. 1992, Calcium signaling in lymphoctyes and ELF fields: Evidence for an electric fields metric
and a site of interaction involving the calcium ion channel, FEBS Letters. 30!: 53-59,
P. Lovsund, P. A. Oberg. S. E. G. Nilsson, and T. Reuter. 1980, Magnetophosphenes: .^ quantitative analysis
of thresholds. Medical and Biological Engineering and Computing. 28: 326 - 334
P. J. Maccabee, L. Eberle, V. E. Amassian, R. Q. Cracco, A. Rudell, and M. Jayachandra, 1990, Spatial
distribution of the electric field induced in volume by round and figure '8' magnetic coils: relevance
to activation of sensory nerve fibers, Elecloencephalography and clinical Neurophysiology. 76
:13l-141.
J.A. Maass and M.M. Asa. 1970, Contactless nerve stimulation and signal detection by inductive transducer.
IEEE Trans. Magnetics MAG-6: 322-326.
D. Melville, F. Paul, and S. Roath, 1975, High gradient magnetic separation of red cells from whole blood.
IEEE Trans. Magn., MAG-11 :1701-1704
K.R. Mills, N.M.F. Murray, and C.W. Hess, 1987, Magnetic and electrical transcranial brain stimulation.
Physiological mechanisms and clinical applications, Neurosurgeiy. 20 (No.l): 164-168.
S. Mittler, 1971. Failure in magnetism to influence production of x-ray induced sex-linked recessive lethals.
Mutation Research. 13 :287 - 288.
S. Nagakura and Y. Molin (Guest Editors), 1992, Magnetic field effects upon photophysical and photochemical
phenomena. Chemical Physics, special issue, 162,1: 1-234.
J. Nilsson, M. Panizza, B. J. Roth, P. J. Basser, L. G. Cohen, G. Caruso, and M. Hallett. 1992, Determining the
site of stimulation during magnetic stimulation of a peripheral nerve. Electoencephalography and
clinical Neurophysiology , 85: 253-264
J. Nyenhuis, J. Bourland, G. Mouchawar, L. Geddes, K. Foster, J. Jones, W. Schoenlein. G. Graber. and T.
Elabbady, 1994, Magnetic stimulation of the heart and safety issues in magnetic resonance imaging.
in S. Ueno (Ed.) Biomagnetic Stimultion. Plenum Press. New York and London : 75-89.
PA. Oberg. 1973. Magnetic stimulation of nerve tissue, Med. Biol. Eng. 11: 55-64.
R. K. OIney. Y, T. So, D. S. Goodin, and M. J. Aminoff, 1990, A comparison of magnetic and electrical
stimulation of peripheral nerves. Muscle Nerve, 13: 957-963.
J. P. Reilly, 1989, Peripheral nerve stimulation by induced electric currents: exposure to time-varying magnetic
\K\A^.Med.BiolEng.Compiit..No\.21:\0\-\\0
i.e. Rothwell. B.L. Day. P.D.Thompson. J.P.R. Dick, and C D . Marsden. 1987. Some experiences of techniques
for stimulation of the human cerebral motor cortex through the scalp. Neurosuigery. 20 (No, 1):
156-163.
J. C. Rothwell. 1994, Motor cortical stimulation in man, in S. Ueno (Ed.) Biomagnetic Stimultion. Plenum
Press, New York and London : 49-57
20 S. Ueno and M. Iwasaka

K. Shulten, H. Staerk. A. Weller. M. J. Werner, and B. Nickel, 1976, Magnetic field dependence ofthe geminate
recombination of radical ion pairs in polar solvents, Z Phvs. Chem.N. F.. 101: 371-390
Y. Tanimoto, H. Hayashi. S. Naaakura, H. Sakaguchi. and K. Tokumaru. 1976. The external magnetic field
effect on the singlet sensitized photolysis of dibenzoyl peroxide, Chem. Phvs. Leit., 41: 267-269.
T. S. fenforde, 1991, Biological interactions ofextrcmely-low-frequency electric and magnetic fields. Bioelec-
nochcm. Bioencrg., 25; 1 -17.
T. S. Tenforde, 1992, Biological interactions and potential health effects of extremely-low-frequency magnetic
fields from power lines and other comtnon sources. Annu. Rev. Pubt. Health. 13: 173-196.
J. Torbet. M. Freyssinet.andG. Hudry-Clergeon, 1981, Oriented fibrin gels formed by polymerization in strong
magnetic fields,.'V(Ui«v, 289 :91-93.
S. Ueno, S. Matsumoto, K.. Harada, and Y. Oomura, 1978, Capacitative stimulatory effect in magnetic
stimulation of nerve tissue, IEEE Trans. Magnetics MAG-14: 958-960.
.S. Ueno. P. Lovsund. and P.A. Oberg, 1981, On the effect of alternating magnetic fields on action potential in
lobster giant axon. Proceedings ofilw 3ll! i\ordic Meeting on Med. and Biol. Eng.. Liiikoping. Sweden
:262-264.
S. Ueno and K. Harada, 1982, Redistribution of dissolved oxygen concentration under strong DC magnetic
fields. IEEE Trans, on Magn.. MAG-18.6: 1704-1706.
S. Ueno. K. Harada, and K. Shiokawa. 1984, The embryonic development of frogs under strong DC magnetic
fields. IEEE Trans, on Magn.. MAG-20, 5: 1663-1665.
S. Ueno, K. Harada, C. Ji. and Y. Oomura. 1984, Magnetic nerve stimulation without interlinkage between
nerve and magnetic tlux, IFiFE Trans. Magnetics, MAG-20: 1660-1662.
S. Ueno, H. Rsaki, and K.. Harada, 1985, Combustion processes under strong DC magnetic fields. IEEE Trans.
on Magn.. MAG-21, 5: 2077-2079.
S. Ueno, P. Lovsund, and PA. Oberg. 1986. Effects of time-varying magnetic fields on the action potential in
lobster giant axon. Med. & Biol. Eng. & Comput. 24: 521-526.
S. Ueno and K. Harada. 1986, Experimental difficulties in observing the effects of magnetic fields on biological
and chemical processes, IEEE Trans.on Magn.. MAG-22,5: 868-873.
S Ueno and K. Harada, 19X7, Effects of magnetic fields on flames and gas flow, IEEE Trans, on Magn .
MAG-23. 5:2752-2757.
S. Ueno. T. Tashiro, and K. Harada. 1988. Localized stiinulation of neural tissues in the brain by means of a
paired configration of time-varying magnetic fields,./. Appl. Phvs.. 64, 10: 5862-5864.
S. Lcno, 1989. Quenching of flames by magnetic fields,./. Appl. Phvs. , 65. 3: 1243- 1245.
S. Ueno. T. Malsuda. M. Fujiki. and S. Hori. 1989. Localized stimulation ofthe human motor cortex by means
of a pair of opposing magnetic fields. Digests of Inlermag Con/.. CD-10, Washington D.C..
S. Ueno. T. Matsuda, and .M. Fujiki, 1989a. Localized stimulation ofthe human cortex by opposing magnetic
fields, i/i.- .Advances in Biomagnetisin, S.J. Williamson et al., eds.. Plenum Press, New York and
London.
S. Ueno, T. Matsuda. and M. Fujiki, 1989b, Localized stimulation ofthe human brain by a pair of opposing
pulsed magnetic fields. Memoirs ofthe Faculty of Engineering - Kyushu University, 49; 161 -173.
S. Ueno, 1. Matsuda, and M. Fujiki. 1990. Functional mapping ofthe human motor cortex obtained by focal
and vectorial magnetic stimulation ofthe brain. IEEE Trans, on .Magnetics. 26. 5: 1539-1544.
S. Ueno and M. Iwasaka. 1990. Properties of magnetic curtain produced by magnetic fields.,/. Appl. Phvs..
67.9; 5901-5903.
S. Ueno. K. Shiokawa. and M. Iwamoto, 1990. Embryonic development of Xenopus laevis under static
magnetic fields up to 6.34 T../. .4ppl. Phys.. 67, 9: 5841-5843.
S. Lieno, V. Matsuda, and O. Hiwaki. 1990. Localized stimulation ofthe human brain and spinal cord by a pair
of opposing pulsed magnetic fields,./. .4ppl. Phys. 66: 5838-5840.
S. Ueno, T. .Matsuda, and O. Hiwaki. 1991. Estimation of structures of neural fibers in the human brain by
vectorial magnetic stimulation, IEEE Trans, on Magnetics. 27. 6: 5387-5389.
S. Ueno and O. Hiwaki. 1991. Spinal reflex evoked by a pair of opposing pulsed magnetic fields. / .'tppl.
. 69. 8. 6019-6021.
S. lieno, M. Iwasaka, H, Eguchi, and T. Kitajima, 1993, Dynamic behavior of gas flow in gradient magnetic
fields, IEEE Trans, on .Magn., MAG-29, 6: 3264-3266.
S. Ueno, M, Iwasaka, and H. Tsuda, 1993, Effects of magnetic fields on fibrin polymerization and fibrinolysis,
IEEE Trans, on .Magn.. 29, 6; 3352-3354.
S. l.Ieno and M. Iwasaka. 1994a, Properties of diamagnetic fluid in high gradient magnetic fields, ./. Appl.
Phvs. 15. 10; 7177-7179
S. Ueno and M. Iwasaka. 1994b. Parting of water by magnetic Trans, on .Magn., 30. 6: 4698-4700.
Magnetic Fields on Biological, Physical and Chemical Processes 21

S. Ueno. M. Iwasaka, and T. Kitajima, 1994, Redistribution of dissolved oxygen concentration under magnetic
fields up to 8 J.J. Appl. Phys.. 75, 10: 7174-7176.
S. Ueno, M. Iwasaka, and K. Shiokawa, 1994, Early embryonic development of frogs under intense magnetic
fields up to 8 T, J. Appl. Phys., 75, 10: 7165-7167.
A. Yamagishi. T. Takeuchi, T. Higashi, and M. Date, 1989, Diamagnetic orientation of polymerized molecules
under high magnetic field./ Phy.s. Soc. Jpn., 58, 7: 2280 - 2283.
1. Yoshikawa, M. Iwasaka, and S. Ueno, 1994, The genetic effects of magnetic fields using the two systems of
somatic mutation detection in Drosophila Melanogasler, BEMS Abstract Book of I6th Annual
Meeting, J-4-1.
 BIOLOGICAL INTERACTION OF
EXTREMELY-LOW-FREQUENCY
ELECTROMAGNETIC FIELDS

T. S. Tenforde

Health Division
Pacific Noriiiwest Laboratory
Richland, Washincton

INTRODUCTION
Considerable controversy has arisen during the past decade over the question of
whether public and occupational exposures to electromagnetic fields in the extremely-low-
frequency (ELF) range may be linked to adverse health effects, particularly an elevation in
the risk of cancer. Since the original publication by Wertheimer and Leeper (1979), which
reported an apparent association between childhood cancer risk and proximity of homes to
power distribution lines with a high-current configuration, there have been nearly 20
additional reports on this topic. The basic hypothesis resulting from the original study was
that the contribution of power-line fields to the ambient 50 60 Hz tleld level in the home
may be linked to cancer risk by some unknown mechanism(s). Although there were
numerous technical deficiencies in the methods used in the original Wertheimer and Leeper
study, more carefully designed and executed studies in recent years have added some support
to the original hypothesis. For example, in a meta-analysis of 13 published studies Washburn
e! al. (1994) found a statistically significant elevation in the risk of childhood leukemia and
nervous tissue tumors among subjects living in close proximity to electricity transmission
and distribution equipment. A similar association between residential exposures and elevated
cancer risk among adults has not been observed in several independent studies (National
Radiological Protection Board [NRPB], 1992).
In addition to the residential cancer studies, more than 50 publications have appeared
during the past decade on the subject of occupational exposure to ELF fields and cancer risk.
Although the results of these studies exhibit a number of inconsistencies, there does appear
to be an elevation in the risk of leukemia and nervous tissue tumors among electrical workers
(Theriault, 1990; NRPB, 1992). Limited evidence also suggests an elevated risk of breast
cancer among electrical workers (Matanoski et al., 1991; Demers a a!.. 1991; Tyncs ei al..
1992; Loomis , 1994).
A major problem in the interpretation of these epidemiological findings is the lack
of convincing evidence that the underlying cause of the elevated cancer risk is exposure to
Bioloiiiciil Effecfs of Xfaf^nt'tlc and lilcctionhii^iwlic Field.'.. F:dited by S. IJeno
Plenum Press. New York. 1996 2.1
24 T. S. Tcnforde

ELF fields from power lines or in the workplace, rather than exposure to some confounding
variable or set of variables that have not as yet been identified. In general, ELF field
exposures based on short-term measurements have not correlated strongly with cancer risk
among the exposed subjects, although a few exceptions have been reported. The issue of
confounding variables has also been closely examined in several studies, but even when
potential confounders such as chemical exposures have been identified, they have had little
impact on the cancer risk estimates based on living in homes close to power lines or working
in electrical occupations.
Another problematic issue is the fact that a biologically plausible mechanism has not
been identified through which ELF fields at low exposure levels could significantly influence
the development or progression of malignant tumors. As discussed below, the strength of
ELF fields encountered in occupational or public settings is far too low to directly affect
DNA or chroinatin structure. For that reason the emphasis in laboratory studies during the
past several years has been on determining whether 50/60 Hz field exposures can operate
through epigenetic mechanisms to promote the development of tumors from clones of cells
previously transformed either by spontaneous mutation or by exposure to carcinogenic
agents such as chemicals, ionizing radiation, or oncoviruses.
I'he results of several recent studies that address the issue of tumor promotion by
ELF fields are discussed in this article, along with other factors such as changes in gene
expression and endocrine regulation that could influence carcinogenic processes. One aspect
of these laboratory-based studies that merits attention is the fact that field levels required to
elicit reproducible biological responses are typically larger than, and frequently orders of
magnitude greater than, the weak 50.60 Eiz fields to which humans are commonly exposed
in the home or workplace. These exposure levels are generally in the range of 10 to 50 V/m
for the electric field component and 0.1 to 0.3 jiT for the magnetic field component, which
induce electric fields within the body on the order of 5 |.iV/m or less. It is a major challenge
for both laboratory research and biophysical modeling to elucidate plausible mechanisms by
which weak ELF fields could significantly affect biological processes when the signal levels
are comparable to, or less than, the intrinsic physical and biological noise present in living
systems. Several contemporary hypotheses on the physical and chemical mechanisms
through which weak ELF fields could perturb the functioning of living systems are described
and critically evaluated in this article.

PHYSICAL PROPERTIES OF ELF FIELDS

ELF fields in tissue have a long wavelength (-1000 m at 60 Hz) and skin depth
(~150 m at 60 Hz), as a result of which these fields behave as though they are composed of
independent, quasistatic electric and magnetic field components (Tenforde, 1991). As a
consequence, the radiating properties of ELF fields can be neglected in their interactions
with tissue. Another important property of ELF fields is their extremely small energy. For
example, the energy of a 60-Hz photon is 2.5 x 10""' eV, which is 11 orders of magnitude
smaller than the Boltzmann thermal energy, kT (= 2.7 x 10'- eV at 310 K), and 14 orders of
magnitude less than the energy required to break a chemical bond. Substantial laboratory
evidence supports the physical expectation that ELF fields do not disrupt the chemical bonds
in DNA, proteins or other biological molecules. Direct genotoxic effects of ELF fields
leading to cell death, gene mutation, or neoplastic transformation would therefore not be
expected, and in fact, have not been observed.
A third important feature of ELF fields applied to living organisms through air is the
nonthermal nature of their interactions. The highest field in tissue that can be induced by an
ELF field applied through air is about 1 V/m, which leads to a specific energy absorption
Biological Interaction of Kxtrcmcly-Low-Frcquency Electromagnetic Fields 25

rate of 10"* W/kg. This rate of energy deposition is four orders of magnitude less than the
body's basal metabolic rate and produces a negligible rate of temperature rise (about 3 x 10'^
°Cs). The interactions of ELF fields applied to the body through air are therefore of a
nonthermal nature.

BIOLOGICAL EFFECTS AND MEMBRANE INTERACTIONS

Despite the minimal perturbations of molecular structure by environmental levels of

FLF fields, there is abundant evidence for responses to these fields at the tissue and cellular
levels (Nairt'r a/., 1989; Adey, 1990a; Anderson, 1991; Tenforde, 1992). For example, small
functional changes in excitable tissues and neuroendocrine alterations have been reported m
response to ELF fields that induce tissue voltage gradients < 10 mV/m. These alterations
include changes in evoked brain potentials (Graham ct al.. 1990). heart rate (Graham ci ai.
1990), and nocturnal synthesis of pineal melatonin (Wilson ci al, 1981; Lerchl ci al.. 1990;
Kato L'l al.. 1993). In addition, a large number of cellular phenomena, including alterations
in growth rate, gene expression and macromolecular synthesis, have been reported to occur
in response to fields of moderate to weak intensity (Tenforde, 1992). These effects include
alterations in biosynthesis of specific messenger RNA and proteins at field levels below 1
mV in (Goodman and Henderson. 1991), Finally, there is a rapidly growing body of
information that implicates the cell membrane as a primary site of FLF field interactions
(Adey, 1990a; Adey. 1990b; Tenforde, 1992; Tenforde and Kaune, 1987). A wide variety of
cell membrane structural and functional properties have been reported to be altered in
response to ELF fields, including changes in Ca** binding to anionic fixed charges at the
cell surface {e.g.. sialic acid residues of membrane glycoproteins), changes in the transport
of ions such as Ca*^ and the secretion of small solutes such as insulin, and alterations in
ligand-receptor interactions that trigger changes in the biosynthetic and functional states of
cells. The threshold tissue field levels that lead to such effects appear to vary widely
depending upon the end point studied, but in nearly all cases arc < 0.1 V'm. In the specific
case of field-induced Ca*"^ desorption from fixed-charge sites on the membrane surface, the
effective field level has been reported to be as low as 1-10 |.iV m (Bawin and .Adey, 1976;
Blackmantva/., 1985).
The phospholipid bilayerthat forms the structural matrix in membranes of living cells
is an electrical insulator, and the membrane electrical conductivity is about five orders of
magnitude less than that of the extracellular medium or the cytoplasm. As a result, the
membrane of a living cell forms an excellent electrical barrier, as well as a superb chemical
barrier, that mediates cellular interactions with the external environment. For this reason, it
is generally believed that cellular responses to weak ELF fields arc initiated by membrane
interactions that serve as the primary mechanism of field transduction. Under typical
exposure conditions with induced ELF fields in the extracellular medium of < 1 V'm. the
"ieakage" field in the cell cytoplasm is less than 1 | J V m. This conclusion also holds for the
circulating electric fields induced directly in the cell cytoplasm by magnetic induction. For
example, a sinusoidal 60-Hz, 0.1-mT field induces a maximum electric field of 0.2 pV/m in
the cytoplasm of a cell with a 10 jim radius. These considerations reinforce the importance
of the cell membrane in ELF signal reception and transduction.
Another point to be made regarding the role of cell membranes in ELF signal
transduction is the amplification of the extracellular field that occurs across the membrane.
By solving Maxwell's equations for the specific case of a dielectric shell (membrane)
surrounding a spherical conductor (cytoplasm), it can be easily shown that the electric field
across the membrane is greater than that in the extracellular medium by a factor 1.5 R d.
where R is the cell radius and d is the membrane thickness. For a spherical cell with a radius
26 I. S. I enforde

of 10 [xm and a membrane thickness of 5 nm. the field across the membrane is therefore
predicted to be 3000 times greater than that in the extracellular medium. A weak environ-
mental field that induces a voltage gradient of 5 nV/m in the extracellular fluid thus produces
a field of about 15 mV m across the cell inembrane.

ELECTRICAL NOISE IN BIOMEMBRANES

Several physical and biological sources of electrical noise within the cell membrane
may impose a lower limit on the strength of an ELF field that can be recognized as a coherent
signal (Fishman and Leuchtag. 1990; Weaver and Astumian. 1990; Adair, 1991). The four
major sources of electrical noise in biological membranes include: (1) Johnson-Nyquist
thermally-generated electrical noise, which produces a 3 nV transmembrane voltage shift at
physiological temperatures; (2) l/f noise associated with ion current flows through mem-
brane channels, which typically produces a 10 )iV transmembrane voltage shift: (3) "shot"
noise, which results from the discrete nature of ionic charge carriers and is a minor source
of membrane electrical noise; and (4) endogenous biological background fields produced by
electrically active organs such as the heart, muscles and the nervous system, which can
exceed the contribution of physical noise sources by an order of magnitude or more.
Because of these various sources of membrane noise, it is of interest to explore the
minimum strength of induced electric fields in tissue that can achieve a signal-to-noise ratio
greater than one within the cell membrane. For example. Weaver and Astumian (1990)
arrived at an estimate of approximately 0.1 V/m as the smallest applied electric field that
exceeds the Johnson-Nyquist noise signal in the membrane of a single cell with an elongated
cylindrical geometry (such as a fibroblast, neuron, or muscle cell). Larger threshold field
levels, greater than 1 V/m, were predicted for small spherical cells such as lymphocytes.
Further increases in the threshold field level are expected if other sources of electrical noise
within the cell membrane are taken into account.
These calculations of signal-to-noise ratio consider only the transmembrane electric
potential shift introduced by an extracellular field. It has been argued rather convincingly
that the field and noise sources that are most relevant to biological signal transduction are
those that reside within the highly charged electrical double layer that exists at the cell
surface. This double layer, frequently referred to as the Helmholtz-Stern layer, is comprised
of fixed anionic charges on the outer membrane surface and diffusible cations in the
surrounding medium. As described originally by Debye and Hiickel, the average thickness
of the diffuse electrical double layer at cell surfaces is approximately 0.8 nm in a physiologi-
cal medium (Tenforde, 1970). By considering Johnson-Nyquist noise and other sources of
electrical noise within the double layer, it can be concluded that the minimum electric field
required in the extracellular medium to achieve a signal-to-noise ratio greater than one is
about 10 mV m.
Another important factor to be considered in calculating the threshold field level that
exceeds intrinsic electrical noise in biological membranes is the junctional coupling that
occurs between cells in organized tissues. For an aggregate of N cells with electrically
coupled membranes, the threshold field level required to achieve a signal-to-noise ratio of
one is lower than the threshold for single cells by approximately N'-''^ as a result of two
factors: (I) field amplification due to the larger size of the aggregate, and (2) reduction of
membrane voltage noise as a result of the larger capacitance of the aggregate (Weaver and
Astumian, 1992). For example, if an aggregate of 10*' electrically coupled cells is considered,
the threshold field level to achieve a signal-to-noise ratio of one is predicted to be lower by
100,000 than the threshold for a single cell. Because of the junctional coupling exhibited by
most biological tissues, the threshold field to achieve membrane and cellular responses may
Biological Interaction of K\tremclv-l.o«-Frequcncy Klectromagnt'tic Fields 27

therefore be on the order of I i^iV/m as contrasted to the value of about 0.1 V i n or higher
predicted for single cells. A similar conclusion has been reached by Pilla (1993) from an
electrical network model of cells that communicate electrically via Junctional coupling. The
electrical conductivity of gap junctions in cell membranes is, however, lower than that of
the extracellular medium. As a result, the electrical coupling and the gain in signal-to-noise
ratio for cell aggregates in tissue is probably less than that predicted from relatively simple
models.

BIOLOGICAL SIGNAL TRANSDUCTION PATHWAYS AND

POSSIBLE CARCINOGENIC EFFECTS OF ELF FIELDS

The most singly important issue in understanding the pathways by which weak HI^F
signals coidd influence membrane and cellular functions is the elucidation of mechanisms
by which these fields arc transduced within the cell membrane. One approach that has been
taken in addressing this question is to consider (he intricate biochemical pathways that have
evoKed as a mechanism by which a living cell communicates with its extracellular en\ iron-
ment. The binding of a single molecule of a mitogenic substance to a specific receptor w iihin
the membrane triggers a cascade of events that involve conformational shifts in membrane-
associated proteins. These events, in turn, lead to signal transduction and amplification \ la
the production of cytoplasmic second messengers and internal efTectors such as free C'a*^
and protein phosphorylascs (kinases) that regulate DNA transcription and protein biosyn-
thesis (.Alkon and Rasmussen, 1988; Luben, 1991). The end result of a single mitogen
binding event at the membrane surface is thus a cytoplasmic signal that is amplified lo a
level that can produce robust effects on macromolecnlar synthesis and cellular responses
involving significant changes in functional and proliferative states.
The interaction of FI.F fields with biological membranes could, in principle, lead to
alterations in each component of this elegant signaling process that occurs in living cells. A
useful working hypothesis is that the pericellular fields and currents induced by an applied
bLI- field initiate electrochemical events within the cell membrane that are important
elements of the primary signal transduction and amplification process (lenforde, I99.i).
These biochemically-mediated events then produce cytoplasmic second messenger re-
sponses that trigger changes in the biosynthesis of macromolecules and alterations in cellular
growth, differentiation, and functional properties. During the past decade, a grov\ing body
of experimental evidence has been acquired that supports this general picture of the sequence
of events leading to HLF signal transduction and amplification at the cellular level. It has
been demonstrated, for example, that pulsed electromagnetic fields (PKMK) with HLF
repetition trequeneies inhibit the production of cAMP by bone cells in response to the binding
of parathyroid hormone (PTH) to sinTace receptors (Luben ci al.. 1982). Further studies have
shown that the PEMF action leads to inability of the PTIi-receptor complex to activate the
alpha subunit of G protein, thereby interfering with the sequence of events that activates
adenylate cyclase at the cytoplasmic membrane interface (Cain ei <//.. 1987). Other studies
using human lymphocytes have shown that exposure to microwave fields with amplitude
modulation at ELF frequencies leads to the inhibition of non-cAMP-dependent histone
kinases (Byusc'/rt/., 1984).
The possible effects of FLF fields on the kinase-C signaling pathway are of particular
interest because this pathway is known to be activated by the binding of tumor promoters
such as phorbol esters. Activation of this pathway by the binding of a first messenger leads
to a cascade of events that produce activated kinase-C and free cytosolic Ca^* ions (Figure
1). Recent studies have demonstrated a sianificant 70"n elevation of kinase-C activity in
28 I. S. I enforde

Outer Surface

Phosphorylation
Figure 1. Phosphokinase C pathway leading to an elevated intracellular calciinn concentration and increased
kinase activity. The binding of a mitogen to the membrane receptor triggers a meinbrane-associatcd phospholi-
pasc C (PLC) hydrolysis of phosphotidylinositol 4,5-bisphosphatc (PIPT). thereby producing diacylglycerol
(DAG), which activates protein kinase C, and inositol 1,4,5-triphosphate (IP)), which mobilizes intracellular
Ca**. There is also evidence that the conversion by a kinase of IP, to IP4 may lead to an influx of free Ca**
and other ions through membrane channels. These ions, together with calmodulin (CAL). lead to further
activation of kinases.

human HL-60 cells exposed to a 50-Hz pulsed magnetic field (Monti ei al.. 1991). The field
effect was considerably dainped by adding EGTA, a Ca"^^ chelator, to the medium. This
observation is consistent with previous findings that kinase-C activation relies on the
mobilization of Ca** ions. Another relevant study is the recent finding that stimulation of
Ca** uptake into rat thymocytes by the plant lectin, Concanavalin A(Con-A), is significantly
augmented by exposure to a sinusoidal 60-Hz magnetic field (Liburdy, 1992). In lymphoid
ceils the binding of Con-A to surface receptors triggers cytoplasmic signaling events
involved in the kinase-C pathway, including the activation of kinase-C and an elevation of
the cytosolic Ca** concentration. Enhancement of the Con-A effect by a 60-Hz inagnetic
field was shown to be dependent upon the strength of the electric field induced in the
extracellular medium.
Several in vivo studies have been conducted to test the hypothesis that ELF fields
may act as tumor-promoting agents, possibly through the stimulation of proliferation in
transformed cell populations via activation of the kinase-C signaling pathway. These studies
have been conducted primarily with skin, mammary, liver, and brain tumor models in
rodents. In studies on skin tuinor promotion, three studies in mice have shown no promoting
Biological Interaction of Extremely-Low-Frequency Electromagnetic Fields 29

effect of chronic exposure to power-frequency magnetic fields with flux densities ranging
from 0.05 to 2 mT on the development of skin papillomas induced by topical application of
10 nanomole of the carcinogen, 7-12-dimethylbenz(a)anthracene (DMBA) (McLean et al..
1991; Rannuger a/., 1993a; Rannugera/., 1994). In positive control groups of animals, skin
tumor development was observed when the DMBA-initiated mice were treated twice weekly
with topical application of the phorbol ester, 12-0-tetradecanoylphorbol-13-acetate (TPA).
In one experiment the effect of an intermittently applied 50-Hz magnetic field (15 sec on/15
sec off) was tested for tumor-promoting activity in comparison with a continuously applied
field (Rannug et al.. 1994). A small increase in the number of skin tumors was observed in
the mice exposed to intermittent versus continuous fields, but neither group of animals
showed a statistically significant increase relative to control mice treated with DMBA alone.
Hence there was no indication of a tumor-promoting effect with either type of magnetic field
exposure.
Another in vivo study of particular interest in the context of possible ELF field effects
on the kinase-C signaling pathway and tumor promotion is the recent finding by Stuchly et
al. (1992) that a 60-Hz, 2-mT magnetic field exerts a copromoting effect on mouse skin
carcinogenesis. In these experiments skin tumors were initiated by the topical application of
10 nanomole of DMBA, and were then promoted by the application of 4.9 nanomole of the
phorbol ester TPA, once per week for a total of 23 weeks. In the group of mice that received
DMBA and TPA plus exposure to the 60-Hz field for 6 h/d, 5 d/week throughout the tumor
promotion phase, both the percentage of mice with tumors and the number of tumors per
mouse increased more rapidly with time than in the group of mice that received DMB.A plus
TPA alone. For example, at week 18 the percentage of mice with tumors in these two groups
were, respectively, 25% and 8% and the mean number of tumors per mouse were 1.90 x 0.69
(SEM) and 0.65 0.46. By week 23 the differences between the mice exposed to the magnetic
field and the nonexposed group were no longer statistically significant. Other studies by the
same group of investigators had previously shown that a 60-Hz, 2-mT magnetic field acting
alone on DMBA-initiated skin cells does not have a tumor-promoting effect (McLean et al.,
1991). It would be of considerable interest to conduct further experiments in which com-
parative measurements are made of kinase-C activity in the affected skin cells throughout
the tumor promotion phase in the field-exposed and nonexposed groups of mice. Such
experiments would provide a test of the hypothesis that ELF fields may exert a copromoting
effect on tumor development through an influence on the kinase-C signaling pathway
activated by phorbol esters.
Three studies of mammary cancer development in rodents exposed to ELF magnetic
fields following tumor initiation with a chemical initiator have provided some evidence for
a tumor-promoting effect of the fields (Beniashvili et al., 1991; Loscher et al., 1993;
Mevissen et al., 1993). In the first of these studies, rats were injected intravenously with a
50 mg/kg dose of nitrosomethylurea (NMU), which produced mammary tumors in 54% of
the rats (Beniashvili et al., 1991). This percentage increased to 86% in a group of rats initiated
with NMU and exposed to a 50-Hz, 0.2-mT magnetic field for 3 h/d for 5 weeks. The latency
time to tumor development was also found to decrease significantly in the rats exposed to
both NMU and the magnetic field relative to rats that received only NMU. In a second study,
rats that were initiated by oral doses of DMBA (20 mg total dose) were exposed for 23 weeks
to a 30-mT, 50-Hz magnetic field (Mevissen et al., 1993). Although the overall tumor
incidence did not differ significantly between the field-exposed and control groups, there
was a statistically significant increase in the number of tumors per rat in the group exposed
to the 30-mT field. A third study was conducted with the same protocol as the second study
described above, except that a considerably smaller field level of 0.1 mT was used (Loscher
et al., 1993). In this experiment the overall mammary tumor incidence was observed to be
50% higher in the field-exposed rats relative to the control group of animals.
30 1. S. Tenforde

Although most of the studies on chemically-initiated mammary tumor development

in rats exposed to 50-Hz magnetic fields provided evidence consistent with a tumor-promot-
ing effect of the fields, the findings of increased tumor incidence could also be related to a
field-induced suppression of pineal melatonin and a resultant elevation in breast cancer risk
(Tamarkin et al., 1981; Stevens et al., 1992). Further research on the effects of ELF fields
on the development of chemically-induced mammary tumors in rodents, including parallel
measurements of field effects on the level of pineal melatonin in the exposed animals, would
provide useful insights into the pathways by which these fields may exert carcinogenic
effects.
Two series of experiments have been designed to study the effects of chronic exposure
to 50-Hz magnetic fields on the development of chemically-induced liver tumors in partially
hepalectomizedrats(Rannugt';i//., 1993b; Rannugef a/., 1993c). In both studies liver tumor
development was initiated by a 30-mg dose of diethylnitrosamine (DENA) administered
intraperitoneally 24 h following partial removal of liver tissue. Tumor development was
promoted with phenobarbital as a positive control. The development of transformed foci of
liver cells was assayed by histochemical staining for the enzymes y-glutamyl transpeptidase
and glutathione S-transferase. The development of tumor foci in the livers of the chemically-
initiated rats was not significantly influenced by a 3-month exposure to SO-Hz magnetic
fields with flux densities ranging from 0.5 MT to 500 |aT, thus indicating a lack of
tumor-promoting effect by the fields. Similarly, no evidence was obtained for a copromoting
effect in rats exposed to both phenobarbital and 50-Hz fields following tumor initiation with
DENA.
A recently completed study provided evidence that chemically-induced brain tumors
in rats are not promoted by exposure to 50-Hz magnetic fields (Brugere et a/., submitted for
publication). Brain tumors of various histological types, including gliomas and astrocy-
tomas, were induced by intravenous injection of 50 mg/kg ethylnitrosourea (ENU) into
female rats on the 19th day of pregnancy. After weaning the offspring were chronically
exposed to fields of 1, 10 and 100 |iT and studied for survival time. Neither male or female
offspring showed a statistically significant difference in mean survival time relative to
control rats that were exposed to ENU but not to 50-Hz magnetic fields.
Overall, with the possible exception of mammary tumors, the available evidence is
not strong for a promoting or copromoting effect of ELF fields on tumor development in
rodents. However, further studies are needed to clarify the possible influence of ELF fields
on second messenger signaling and endocrine regulation in exposed animals, both of which
are factors that could influence the development of tumors by promoting the proliferation
of initiated cells.

PHYSICAL MECHANISMS OF ELF FIELD INTERACTIONS

In the search for mechanisms by which extremely weak ELF fields could exert
.significant biological effects, a number of innovative models have been proposed over the
course of the last two decades. One class of models that has received particular attention
during the past several years are resonance models that involve the combined action of an
ELF field and the static geomagnetic field. For example, ion cyclotron resonance (ICR) was
proposed by Liboff (1985) as a possible mechanism that could facilitate the transport of ions
such as Ca** through membrane channels in the presence of the geomagnetic field and a
weak ELF field tuned to the ICR frequency. Although some data on Ca** uptake by
lymphocytes (Liboff e^ al., 1987) and diatoms (Smith el ai, 1987) have been cited as lending
support to this model, there are also negative experimental findings (Parkinson and Hanks,
1989; Liboff and Parkinson, 1991; Parkinson and Sulik, 1992). In addition, a number of
Biological Interaction of Extremely-Low-Frequcncy Electromagnetic Fields 31

physical arguments can be raised against the ICR model (Tenforde, 1992), the most serious
of which is the collisional damping of resonant ion motion that is expected to occur in a
condensed phase (Halle, 1988).
Two other resonance models that have been proposed recently are the "quantum
beats" model of Lednev (1991), in which combined static and ELF fields affect vibrational
energy levels and transition probabilities of bound ions (e.g., Ca** ions bound to calmodulin),
and the model of Zhadin and Fesenko (1990) in which the rotational energy levels of bound
ions are pumped by combined static and ELF fields. Shuvalova el al. (1991) have presented
data on the effect of combined fields on the rate of calmodulin-dependent phosphorylation
of myosin that appear to support the predictions of the Lednev model. However, a study
using an optical technique to study Ca*^ binding to calmodulin and to metallochromic dyes
failed to find any effects of combined static and time-varying fields under the resonance
conditions predicted by Lednev's model (Bruckner-Lea et at., 1992). Adair has pointed out
that the Lednev model of resonant field effects is improbable because of the long lifetime
of the excited vibrational states (about 8 sec), during which de-excitation would occur as a
result of collisional damping (Adair, 1992). Similar arguments can be raised against the type
of resonant field effects envisioned in the Zhadin and Fesenko model. In addition, the rate
of transition of a bound ion to an excited state predicted by this model is so low that a
transition would occur only once in several months at typical environmental magnetic field
levels.
A large number of nonequilibrium models have been proposed in which field-induced
structural and functional perturbations result from membrane interactions that exploit the
existence of unstable or metastable states. Examples of such interactions arc dissipative
instabilities involving cooperative transitions of allosteric membrane proteins in the presence
of an applied field. Critical state instabilities in membrane physical properties near the phase
transition temperature can also be amplified by an applied electromagnetic field. A host of
other models involve coherent field interactions that lead to the excitation of oscillating
dipolar modes in membrane proteins or the production of nonlinear oscillations (solitons)
that facilitate vibrational energy transfer in macromolecular structures. A number of reviews
discussing the physical basis of these nonequilibrium models have been published (Adey,
1981; Taylor, 1981; Postow and Swicord, 1986; Tenforde and Kaune, 1987; Adey. 1990a).
The discovery of biogenic magnetite particles in the tissues of a large number of
organisms, including several mammalian species (Kirschvink et al., 1985; Kirschvink,
1989), has led to speculation that oscillatory magnetomechanical forces and torques on these
particles could provide a mechanism for the transduction of signals from ELF magnetic
fields. Of particular interest is the recent demonstration of magnetite crystals in various
anatomic locations within the human brain (Kirschvink, 1992a). Kirschvink el al. (1992b)
have proposed a model in which oscillatory magnetic forces on magnetite particles at ELF
frequencies are visualized as producing the opening and closing of pressure-sensitive ion
channels in membranes. The 60-Hz field level required to overcome the effects of Brownian
motion is predicted from this theoretical model to be on the order of 0.1 mT, which is within
the range of fields at locations close to the surfaces of several types of household appliances
and machine tools.
One difficulty with this model is the sparsity of magnetite crystals relative to the
number of cells in brain tissue. For example, human brain tissue is reported to contain a few
million magnetite crystals per gram, distributed in 5-10 x 10^ discrete clusters (Kirschvink,
1992a). The number of cells in brain tissue thus exceeds the number of magnetite crystals
by approximately a factor of 100. It is therefore difficult to envision how oscillating
magnetomechanical interactions of an ELF field with magnetite crystals could affect a
significant number of pressure-sensitive ion channels in the brain. However, the effects of
such interactions on neural signaling in localized brain regions could possibly result in a
32 T. S. Tcnforde

biological response, although there is no evidence at present to support this hypothesis. The
possibility also exists that some subpopulations of cells may have extremely high concen-
trations of magnetite that experience large forces and torques in a magnetic field of moderate
strength. Evidence for such cells was recently obtained in studies on magnetite in human
leukemic leukocytes (Kirschvink and Kobayashi-Kirschvink, 1993). Further studies are
clearly needed to reveal the biological role of magnetite and the possible mechanisms
through which this mineral could play a role in ELF signal transduction.

CONCLUDING REMARKS
Evidence is growing for a central role of cell membranes in the reception, transduc-
tion and amplification of signals imposed by ELF fields. A major challenge for the future
will be the elucidation of specific molecular pathways through which these fields can
influence transmembrane signaling events and affect the functional and proliferative states
of cells and organized tissues. Of particular importance will be studies on possible mecha-
nisms through which ELF fields may play a role in the development of tumors. At the present
time there is little evidence for a promoting or copromoting effect of ELF magnetic fields
on tumor development, with the possible exception of mammary tumors in which endocrine
alterations resulting from field exposure may play an important role. Further research is
needed to gain an understanding of the ELF signal characteristics that are the most biologi-
cally effective, and to define the threshold field parameters above which predictable
biological responses occur. Recent laboratory studies have provided a number of clues on
the pathways through which ELF fields may operate at the cellular and subcellular levels.
However, a great deal of research lies ahead in order to fully characterize the molecular
substrates of ELF field interactions and the resultant cascade of electrical and biochemical
signals that lead to cellular and tissue responses, including possible carcinogenic effects.

ACKNOWLEDGMENTS
The author thanks Judith M. Christensen for assistance in the preparation of this
manuscript. Research support is received from the U.S. Department cf Energy under
Contract DE-AC06-76RLO 1830 with the Pacific Northwest Laboratory. The Pacific
Northwest Laboratory is operated for the U.S. Department of Energy by the Battellc
Memorial Institute.

REFERENCES
R.K. .'\daii-. Constraints on biological etTects of weak extremely-low-frcqucncy electromagnetic fields. Phys-
/Jtn. .4 41:1039-1048. 1991.
R.K. .Adair, Criticism of Lednev's inechanisin for the influence of weak magnetic fields on biological systems.
Bioeleciromagnetics Ii:231-235. 1992.
W.R. .Adcy, Tissue interactions with nonionizing electromagnetic fields. Physiol. Rev. M:435-514, 1981.
W.R. Adey, Nonlinear electrodynamics in cell membrane transductive coupling. In: Membrane Transport and
In/brmalion Storage, Vol. 4.. R.C. Aloia, C.C. Curtain, and L.M. Gordon, cds., pp. 1-27. Wiley-Liss.
New York, 1990a.
W.R. ,^dey, Electroinagnetic fields, cell inembrane amplification, and cancer promotion. In: E.Ktremely-Lo-iv-
Frequency Eleclroniagnetic Fields: The Question of Cancer. B.W. Wilson, R.G. Stevens, and L.H.
.Anderson, cds., pp. 211 -249. Battelle Press, Columbus, Ohio, 1990b.
D.L. .Alkon and H. Rasmussen. A spatial-temporal model of cell activation. 5c/e«fe 212:'^''8-100.5, 1988.
Biological Interaction of Extremely-Low-Frequency Electromagnetic Fields 33

L.E. Anderson, ELF: Exposure levels, bioeffects, and epidemiology. Health Phys. 61-41-46, 1991.
S.M. Bawin and W.R. Adey, Sensitivity of calcium binding in cerebral tissue to weak environmental electrical
fields oscillating at low frequency. Pmc. Nail. Acad. Sci. fUS.4) 21:1999-2003. 1976.
D.Sh. Beniashvili, V.G. Bilanishvili, and M.Z. Menabde. Low-frequency electromagnetic radiation enhances
the induction of rat mammary tumors by nitrosomcthylurea. Cancer Lea. tL:15-19. 1991.
C.F. Blackman, S.G. Benanc, L.S. Kinney, D.E. House, and W.J. Joincs, Effects of BLF fields on calcium-ion
efflux from brain tissue in vitro. Bioelectromagnetics 6:1 -11, 1985.
C. Bruckner-Lea, C.H. Dumey, J. Janata, C. Rappaport, and M. Kaminski, Calcium binding to metallochromic
dyes and calmodulin in the presence of combined AC-DC magnetic fields. Bioclectromcignetic.s
11:147-162, 1992.
H. Brugere. F. Pupin, and J. Lambrozo, Effects of 50-Hz magnetic fields on ENU-induced brain tumors in rats.
Bioelectromagnetics, submitted for publication.
C.V. Byus, R.L. Lundak. R.M. Fletcher, and W.R. Adey, Alterations in protein kinase activity following
exposure of cultured human lymphocytes to modulated microwave fields. Bioeleciromagnetics
5:341-351, 1984.
C D . Cain, W.R. Adey, and R.A. Luben, Evidence that pulsed electromagnetic fields inhibit coupling of
adenylate cyclase by parathyroid hormone in bone cells. / Bone Min. Re.s. 2:437-441, 1987.
P.A. Demers, D.B. Thomas, K.A. Rosenblatt, L.M. Jimenez, A McTieman. II. Stalsberg. A. Stemhagen, WD.
Thompson, M.F. McCrea-Cumen, W. Satariano, D.F Austin, P. Isacson. R.S. Greenbcrg, C. Key, L.N.
Kolonel and D.W. West, Occupational exposure to electromagnetic fields and breast cancer in men.
Am. J. Epidemiol. 11:340-347, 1991.
H.M. Fishman and H.R. Leuchtag, Electrical noise in physics and biology. Curr Top. Memhr Transport
12:3-35, 1990.
R. Goodman and A.S. Henderson, Transcription and translation in cells exposed to extremely-low -frequency
electromagnetic fields. Bioelectrochem. Bioenerg. 25:335-355. 1991.
C. Graham. M.R. Cook, and H.D. Cohen. Immunological and Biochemical Effect.s of 60-Hz Electric and
Magnetic Fields in Humans. Midwest Res. Inst. Final Rep., Contract No. DE-FC01-S4-CE-76246
(Order No. DE90006671). U.S. Department of Energy, Office of Scientific and Technical Information,
Oak Ridge, Tennessee, 1990.
B. Halle, On the cyclotron resonance mechanism for magnetic field effects on transmembrane ion conductivity.
fi(oe/t'f?romag/K';(rt 2:381-385, 1988.
M. Kato, K. Honma, T. Shigemilsu, and Y. Shiga, Effects of exposure to circularly polarized 50-Hz magnetic
field on plasma and pineal melatonin levels in rats. Bioelectromagnetics 14:97-106, 1993.
J.L. Kirschvink, D.S. Jones, and B.J. MacFadden, eds.. Magnetite Biomineralization and Magneto-reception
in .Animals: A New Biomagnelism. Plenum Press, New York. 1985.
J.L. Kirschvink, Biogenic magnetite and magnetoreception. Bioelectromagnetics lfl:239-259, 1989.
J.L. Kirschvink, A. Kobayashi-Kirschvink. and FB. Woodford, Magnetite biomineralization in the human
brain. Proc. Natl. Acad. Sci. (USA) §2:7683-7687. 1992a.
J.L. Kirschvink, A. Kobayashi-Kirschvink, J. Diaz-Ricci, and S.J. Kirschvink, Magnetite in human tissues: .\
mechanism for the biological effects of weak ELF magnetic fields. Bioelectromagnetics Suppl.
1:101-113, 1992b.
J.L. Kirschvink and A. Kobayashi-Kirschvink, Magnetite biomineralization in human tissues. In: Abstracts of
the XXIVth General Assembly of the International Union of Radio Science. Abstract Kl-8, p. 512.
Kyoto, Japan, August 25-September 2, 1993.
V.V. Lednev, Possible mechanism for the influence of weak magnetic fields on biological systems. Bioelectro-
magnetics ll-l^-15, 1991.
A. Lerchl, K.O. Nonaka, K.A. Stokkan, and R.J. Reiter, Marked rapid alterations in nocturnal pineal serotonin
metabolism in mice and rats exposed to weak intermittent magnetic fields. Biochem Biophys. Res.
Commun. 1M:102-108, 1990.
A.R. Liboff, Geomagnetic cyclotron resonance in living cells. / Biol. Phys. 11:99-102, 1985.
A.R. Liboff R.J. Rozek, M.L. Sherman, B.R. McLeod, and S.D. Smith, '''Ca*^ cyclotron resonance in human
lymphocytes. J. Bioelectr fi: 13-22, 1987.
A.R. Liboff and W.C. Parkinson, Search for ion-cyclotron resonance in a Na*-transport system. Bioelectro-
magnetics 12:17-82. 1991.
R.P Liburdy, Calcium signaling in lymphocytes and ELF fields. FEBS Lett. liLL:53-59. 1992.
D.P. Loomis, D.A. Savitz and C.V. Ananth. Breast cancer mortality among female electrical workers in the
United States../ Natl. Cancer Inst. S6:921-925, 1994.
W. Loscher. M. Mevissen, W. Lehmacher, and A. Stamm, Tumor promotion in a breast cancer model by
exposure to a weak alternating magnetic field. Cancer Lett. 21:75-81. 1993.
34 T. S. Tenfordc

R,A. Luben. C D . Cain. M.C.Y. Chen. D.M. Rosen, and W.R. Adey, Inhibition ofparathyroid hormone actions
on bone cells in culture by induced low energy electromagnetic fields. Proc. Nuti Acml Sci. fUSA)
79:4180-4184. 1982.
R.A. Luben. EtTects of low-energy electromagnetic fields (pulsed and DC) on membrane signal transduction
processes in biological systems. Health Phys. ^ : 15-28. 1991.
G.M. Matanoski, P.N. Breyssc and E.A. Elliott. Electromagnetic field exposure and male breast cancer. Lanccl
212:737. 1991.
J.R.N. McLean. M.A. Sluchly, R.E.J. Mitchell, D. Wilkinson, H. Yang. M. Goddard. D.W. Lecuyer. M.M.
Schunk, E. Callary, and D. Morrison, Cancer promotion in a mouse-skin model by a 6()-H/ magnetic
field: 11. Tumor development and immune response. Bioelcclromui^neiic.s 12:273-287. 1991.
M. Mevissen, A. Stamm. S. Buntenkcitter. R. Zwingelbert. U. Wahnschaffe. and W. Loscher, Et't'ects ofmagnetic
fields on mammary tumor development induced by 7,12-dimethylbenz(a)anthracene m rats. Rioclci -
immciiiiwlics 14:131 -143. 1993.
M.G. Monti. L. Pernecco. M.S. Moruzzi. R. Battini, P. Zaniol. and B. Barbiroli. Effect of ELI- pulsed
electromagnetic fields on protein kinase C activation process in HL-6() leukemia cells../ Buielccii:
1(]:119-130. 1991.
1. Nair. .M.G. .Morgan, and H.K. Elorig. Biolofficul Eth'cls of Power Frei/iieiicv Electric and Mai^iieiic FiclJs.
Rep. No. OT.A-BP-E-53. Office of Technology .Assessment, U.S. Government Printmg Office.
Washington. I X . 1989.
National Radiological Protection Board. Electromagnetic Fields and the Risk of Cancer. Docmnents of the
NRPB Vol. 3. No. 1. 1992.
W.C. Parkinson and C.T. Hanks, Search for cyclotron resonance in cells in vitro. Bioeleclrotnaf^nctiis
111:129-149. 1989.
W.C. Parkinson and G.L. Sulik, Diatom response to e.\tremely-lovv-frequency magnetic fields. Radial. Res.
iii):3 19-330, 1992.
.A.A. Pilla. State of the art in electromagnetic therapeutics. In: Electricity and Magnetism in Biology and
Medicine. M. Blank, ed., pp. 17-22. San Francisco Press, San Francisco, California. 1993.
E. Postow and M.L. Suicord, Modulated fields and "window" effects. In: Handbook of Biological Effects ot
Electrotnagneiic Fields. E. Postow and C. Polk, eds.. pp. 425-460. CRC Press, Boca Raton. Florida.
1986.
;\. Rannug. T. Ekstrom, K.H, Mild, B. Holmbcrg. I. Gimencz-Conti, and T.J. Slaga. .A study on skin tumor
formation in mice with 50-Hz magnetic field exposure. Carcinogenesis 14:57?:-57H. 1993a.
. Rannug. B. Holmberg, T. Ekstrom, and K.H. Mild. Rat liver foci study on coexposure with 50-llz magnetic
fields and known carcinogens. Bioeleeiromagnetics 11:17-27. 1993b.
\ . Rannug. B. Holmberg, and K.H. Mild. .\ rat liver foci promotion study with 50-Hz magnetic fields. Environ
Res. 62:223-229. 1993c.
A. Rannug, B. Holmberg, T. Ekstrom, K.H. Mild, 1. Gimencz-Conti, and T.J. Slaga, Intermittent 50-Hz
magnetic field and skin tumor promotion in SENCAR mice. Carcinogenesis l i : 153-157. 1994
LA. Shuvalova, M.V. Ostrovskaya, V..^. Sosunov, and V.V. Lednev, Influence of a weak magnetic field under
conditions of paramagnetic resonance on therateofcalmodulin-dependent phosphorvlation of mvosin
in solution. Proc. Natl. .icad. Sci. I USSR) 117:227-230, 1991.
S.D. Smith, B.R. McLcod, .A.R. Liboff, and K. Cooksey, Calcium cyclotron resonance and diatom mobility.
Bnielectromagnetics a:215-227, 19X7.
R.G. Stevens. S. Davis. D.B Thomas. L.E. .-Xndcrson. and B.W. Wilson, Electric power, pineal function, and
the risk ofbreasi cancer. FASEB.I. 6:853-860. 1992.
M..\. Stuchly, J.R.N. McLean, R. Burnett, M. Goddard, D.W. Lecuyer, and R.E.J. Mitchcl. Modification of
tumor promotion in the mouse skin by exposure lo an alternating magnetic field. Canier Lett. 65:1 -7.
1992.
L. Vamarkin, M. Cohen, D. Rosclle, C. Reichert, M. Lippman, and B. Chabner, Melatonin inhibition and
pinealectomy enhancement of 7.12-dimethylbenz(a)anthracene-induced mammary tumors in the rat.
Cancer Res. 41:4432-4436. 1981.
L.S. Taylor. The mechanisms of athernial microwave biological effects. Bioeleeiromagnetics 2:259-267. 1981.
1. lenforde. Microelectrophoretic studies on the surface chemistry of erythrocytes. .\dv. Biol. Med. Phvs.
11:43-105. 1970.
LS. Lenforde and W.T. Kaune. Interaction of extremely-low-frequency electric and magnetic fields with
humans. Health Phys. 51:585-606. 1987.
LS. lenforde. Biological interactions of extremely-low-frequency electric and magnetic fields. Bioelectro-
chem. Bioenerg 25:1-17. 1991.
Biological Interaction of Extremely-Low-Frequency Electromagnetic Fields 35

T.S. Tentbrde. Biological interactions and potential health effects of exlremely-low-frequency magnetic fields
from power lines and other common sources. Annu. Rev. Pub!. Hcallh 11:173-196, 1992.
T.S. Tenforde, Cellular and molecular pathways of extremely-low-frequency electromagnetic field interactions
with living systems. In: Eleclricily and Magnetism in Biology and Medicine. M. Blank, ed.. pp. 1 -S.
San Francisco Press, San Franci.sco, California, 1993.
G.P. Theriault, Health effects of electromagnetic radiation on workers: Epidemiologic studies. In: F'roc.
Scientifie Workshop on the Health Effects of Electric and Magnetic Fields on IVorkers, P. J. Bierbaum
and J.M. Peters, cds., pp. 91-124. DHHS (NIOSH) Publ. No. 91-1II, Cincinnati. Ohio. 1991.
T. Tynes, A. .Andersen and F. Langmark, Incidence of cancer in Norwegian workers potentially exposed to
electromagnetic fields. .4m. J. Epidemiol. 136:81-88, 1992.
E.P. Washburn, M..I. Orza, J.A. Berlin, W.J. Nicholson, A.C. Todd, H. Frumkin and T.C Chalmers, Residential
proximity to electricity transmission and distribution equipment and risk of childhood leukemia,
childhood lymphoma, and childhood nervous system tuinors: Systematic review, evaluation, and
meta-analysis. Cancer Causes and Com. 5:299-309, 1994.
J.C. Weaver and R.D. .-Xstumian, The response of living cells to very weak electric tlelds: The thermal noise
limit. Science 247:459-462, 1990.
J.C. Weaver and R.D. Astumian, F.LF biological threshold estimates: .An approach for electric fields based on
noise and mechanism models. Bioeteclromagnclics Suppl. 1:119-138. 1992.
N. Wertheimer and R. Leeper. Electrical wiring configurations and childhood cancer .4m. ./. Epidemiol.
1112:273-284, 1979.
B.W. Wilson, L.E. Anderson, D.I. Hilton, and R.D. Phillips, Chronic exposure to 60-Hz electric fields: Effects
on pineal function in the rat. Bioelectromagnetics Z'.^l^-^Sd. 19S1 [erratum: 4:293, 1983].
M.N. Zhadin and E.E. Fesenko. Ionic cyclotron resonance in biomolcculcs. Biomed. Sci. 1:245-2,50. 1990.
THE EFFECTS OF ELF ON CHEMICAL
REACTION RATES IN BIOLOGICAL
SYSTEMS

Frank S. Barnes

University of Colorado
Department of Electrical and Computer Engineering
Boulder, Colorado, 80309-0425

ABSTRACT
In this paper wc review some of the theory and experiments associated with the effects
of electric and magnetic fields at extremely low frequencies, ELF, on chemical reaction rates.
The paper proposes a simple model for chemical reactions and shows that the effects of weak
electric fields can change the probability that molecules of the reacting materials will
encounter each other as well as shift the barrier energy for the reaction. These etTects occur
primarily in the vicinity of membranes or at boundaries where there are large variations in
current density or the concentration of one of the chemical reactants.

INTRODUCTION
The possible effects of weak electric and magnetic fields on biological systems has
attracted considerable public attention. This is particularly true after the publication of
epidemiological studies which show a correlation between the proximity to power lines and
an incidence of cancer.(l-4) If we are to develop a scientific base for deciding whether or
not there is any reason for these concerns we are going to have to develop a chain of scientific
logic which takes us from the generation of these fields through the coupling of the fields
into the body and their interactions with the biological materials. This logical chain is rather
long as it starts with either calculations or measurements of the fields generated by the power
distribution system and then their coupling into the body. This part of the problem has been
treated in a number of reviews with varying degrees of completeness. (5-8). However it is
to be noted that if you want to know what the values of the electric field are at a given cell
membrane at a given location in the body or at the surface of given organ or gland for a give
external electric or magnetic field we have only a very approximate estimate at this time
(1994). Having estimated the distribution of the fields in the body we then need to measure
or calculate their effect on the biological system. We can start at the level of the forces on
Biological Ejfects of Magnelic and Elevlromagnelic Fields, Edited by S. Ucno
Plenum Press, New York. 1996 37
38 Frank S. Barnes

ions, molecules, magnetite particles etc. and work our way up through the complexity chain
through induced current flows, changes in spin orientations, shifts in energy level, changes
in chemical reaction rates, transmembrane currents, cell growth rates, to changes in the
functioning of biological systems and health effects. These changes occur over a wide range
of times with different time constants being important for different kinds of effects. At the
short end of the spectrum we have atomic and molecular transitions which occur with time
constants as short as 10"'-' seconds and efTects on growth which show up generations later,
which may be days or years. An over all review of these electrically induced effects is given
in reference (9) and some of the direct magnetic field effects are covered in other chapters
in this volume. In this review we will concentrate on those effects of electrical fields which
can be associated with changes in chemical reactions rates. We will start with a brief
discussion of how changes in chemical reaction rates effect the production of a given
biological substance, followed by a discussion of how electrical fields may effect rate
coefficients.

CHEMICAL REACTIONS
Chemical reactions lie at the center of the way biological systems transmit, process
and store information. Chemical reactions are the key steps in the growth of cells, in putting
energy into a usable form for other biological process and in encoding genetic information.
These reactions can be classified in many different ways however, for the purposes of this
discussion we will consider only two types of chemical reactions. The first of these is a
process where two molecules or atoms in solution react to form a product. The second type
of chemical reaction requires the molecule to come in contact with a bound substance. In
order for either of these chemical reactions to occur the two molecules must come together
with enough energy to over come any barrier that may prevent the reaction from proceeding
to a lower energy state or utilize a catalyst. The energy added to an ion in solution by a low
level electric field is usually negligible compared to the thermal energy. For example, the
average drift velocity of a Na* ion in a field of 10^ V/m is estimated to be 5 x 10'' m/s with
a thermal velocity of 4 x 10- m/s then the kinetic energy added by the drift velocity is on the
order of only one part in lO'**, This is truly negligible.
On the other hand, if we look at the effects of the low level electric fields on the
transport of the ions, low level electric fields may have a significant role. Consider the case
of chemical reaction which takes place in a homogeneous fluid if the chemical reaction has
the form of (10,! I)
k,
A + B->C

where (A) and (B) are the concentrations of the two chemical reactants and
where k|is the rate constant for A + B to C.
k, = k ( A ) " ( B r

where n and m are the order of the reaction with respect to the species A and B,
respectively. In order to find the values of n and m one can make the concentration of one
of the reactants small so that k, takes the form

k, = k ( A J " ( B r
k, = k ( B r
The Effects of ELF on Chemical Reaction Rates in Biological Systems 39

where we have made the concentration of {A^) large enough so that it is approximately
constant, and the changes in the reaction rate can be measured by varying B. Many reaction
the rate constants follow Arthenius equation k is given by

k = zpe-f^' «T

where z is the collision frequency for a single ion or molecule, p is the steric factor
which is less than one and reflects the fact that not all collisions occur with the right
orientation of the molecules to react. E' is the activation energy, R is the gas constant and T
is the absolute temperature. This collision frequency is related to the total collision density
Z which is the total number of collisions of the reactants per unit volume per unit time. For
a diffusion controlled reaction (11)

Z = a(8kT/7in) ''^N,2(A)"(B)'"

(J is the reduced mass, and N^ is Avogadro's number, k' is Boltzmans constant, a is

the combined cross section for the atoms or molecules. This collision rate is calculated from
the thermal diffusion of the atoms through a solution and results in a reaction rate when n =
m = 1 which is given by

k, = d[A]/dt = p o(8k'T/7t(i)' - e""^'""^ N,= (A) (B)

The activation energy can take a number of forms depending on the particular reaction
and the environment. In a diffusion controlled reaction where the activation energy is not
important the reaction rate may be approximated by

k, = 8RT/3ri (A)(B)

where r| is the viscosity of the fluid. This expression was derived by calculating the
current flow of one kind of atoms passed the other and the fraction of them which would
have overlapping radii. In this case it is interesting to note the cross section of the atoms has
dropped out of the expression except to the extent that it is included in the viscosity.
The number of particles available to collide with each other is modified if the current
flow is modified by adding a drift or hydraulic flow velocity to the thermal velocity. If we
look at the general problem from the point of view of the conservation of the number of
molecules of a given kind in a small volume then

oN/cT = - N / T + VJ/q

where 1/r is the rate at which the concentration of Ni is consumed, q is the charge on
N, and VJ is the gradient of the current flowing through the volume. For the cases of most
interest to us we will assume are atom is charged or is an ion. Suppose we have a uniform
current of ions flowing through a transparent membrane then the number of molecules per
unit area, n, passing through the membrane would be equal to the density of ions, N| times
the drift velocity ,v, or

nj = V| Nj = J|:q,

For a uniform low level electric field, E, the drift current density. J, is given by

J = I q , V, Ni=Iq,n;N,E
40 Frank S. Barnes

where q; is the charge on the ion and y.,' is the mobility of the charged molecule or
ion. If we look at a more general case the current density is given by the sum of the drift .
diffusion, and hydraulic currents so that

J = I(qiHi'N,E - qDVN, K q(VP)/R'

where D is the diffusion constant and VNj is the concentration gradient, VP is the
pressure gradient, R' is the hydraulic resistance . The third term is the hydraulic flow. This
leads to a complicated expression for N| when J is substituted into the equation for the
conservation of Nj as x may also be dependent on J. Note that to get the desired values of J
we need to sum over the component molecules or ions which are involved in the reaction.
In most descriptions of chemical reactions it is assumed that dominate current is due
to diffusion, however, in the case where electric fields are imposed on the system the drift
current needs to be taken into account. The hydraulic flow term also needs to be taken into
account when we are dealing with blood flow. If we have a homogeneous region where the
collison probability is indpendent of position then to first order T is independent of the current
density. The added drift velocity in the direction of the field leads to an increase in the
collision probability in one direction but this balanced by a reduction on the relative velocity
at ISO". Thus modifications of the chemical reaction rate at low fields may come about
through the gradient of the current density, VJ. If the current density has an AC component
then wc will get an AC component in VJ, f^N/5t and thus N,. The gradients in the current of
one of the reactants typically become most important in the vicinity of a barrier such as a
curved cell membrane where th efiective resistance to the current flow varies rapidly in space
as shown in Fig. 1. Thus, we would expect to see the effects of current gradients to be most
important when we are looking at current flow around or between cells.
I may also be dependent on E at high fields levels. In this case the fields may lead
to a linear increase in the disassociation rate with electric field at field magnitudes as low as
21 kV/cm (12). Additionally, there may be a modification of the number of bound water
molecules associated with each ion or their distribution in space. Thus the effective radius
and mobility are modified and in turn the energy barrier for the disassociation of a charged
molecule or a chemical reaction.
Next, consider the case of the transport of a chraged particle to the surface of a
membrane where it either undergoes a chemical reaction or is bound to the membrane
surface. In this case the rate of arrival of one of the component for our chemical reaction is
determined by the current density. If the total current through the membrane is given by the
Goldman equation (13) and there is a dominate ion in establishing the membrane voltage,V.
such as potasium the current, 1, takes the form

I = I„ (exp qV/k'T - 1)

where V is the potential across the membrane and I;, is a constant. For DC \oltrage
across the membrane we get an exponentially increasing current in one direction and I„ in
the other Normally I„ is small. For a low frequency AC field a power series expansion gives
a DC term which is proportional to the square of the applied AC voltage across the membrane
and compoennts at the fundamental and second harmonic. This form of response has also
been found by Seto and Hsleh (14) for enzyme Tactions which are catalytilized at a substrate
surface at low field strengths At higher field strengths these results show a saturation of the
reaciton rate which is a function of the basic diffusion controlled reaction rate and space
charge limitations at the interface. One would also expect limitations associated with transit
times at high frequencies.
 The Effects of ELF on Chemical Reaction Rates in Biological Systems 41

Another approach to the general problem of the effect of electric fields on chemical
reaction rates is to estimate the fields required for the drift current to be equal to the diffusion
current and thus the electric field on current in the medium at a distance from the barrier. If
we set

E = D/y (VN„.Ni)

If we set D = 1.5 x 10"^cm"sec"' and (i' = 5.2 x IO"*m^ V''sec"' then

3.4 X 10-^E = VN,/Ni

This means that fields of lOV/m which is known to show biological effects on cell
growth rates would correspond to doubling the chemical reaction rate due to diffusion with
a fractional gradient of .1.4 x 10""*.
Another approach to looking at the collision frequency is to look at the distance the
particle travels in a given time t and to make assumption that the number of collisions with
the desired particle A with B is proportional to this distance. The mean squared distance
covered by a particle undergoing a random walk is given by

{<x->y- = 2 ( Dt/n)'^= (2K'Tt/37r-iia)' =

where D is the diffusion coefficient, t is time , K' is Boltzmanns constant, i] is the

viscosity, and a is the radius of the molecule. If we now add to this random walk a drift
velocity which is driven by an applied electric field we will increase the probability of
collisions. The extent to which this probability of collision is increased is proportional to the
distance traveled or
(<x2j>)'-=vt

where v is the drift velocity and (<x'^j> )' - is the average distance travel under the
influence of the electric field, E. At low field strengths v = |a' E. If wc assume the same values
for D,n' and E as above, the time for the drift distance do the electric field to equal the
diffusion distance is about 2 hours. This seems to under estimate the effectiveness of the
electric field, as shorter exposures on the order 20 minutes are known to show biological
effects. This situation would not occur when we have a homogenious solution and the
probability of collision is independent of the position.
If we now return to our fundamental equation for the reaciton rate, we can vary the
energy of the collisions by accelerating the particle through a potential barrier. This in turn
will change the reaction rate by providing part of the energy necessary to initate the reaction.

Impervious
Membrane
Figure 1. Current flow around an impervious membrane.
42 Frank S. Barnes

Variations of this problem have also been treated extensively in references (15-25) for low
levels of oscillating electric fields. These papers show that small fluctuations in the energy
barrier. qV, at a membrane or in the height of the barrier inhibiting the chemical reaction
lead to changes in the rate contant, k. Again a power series expansion gi\es a small change
in the average values of the rate constant, < k >, which is proportional to the square of the
potential variation.
In the forgoing examples we have assumed plane geometry and that the current
is carried by only one type of charged particle. This is unlikely to describe many of
the processes of interest which may involve two or more types of charged particles
and complex geometries. For example if we have a spherical cell with an impervious
membrane in a uniform electric field we would expect the current to flow around the
cell with a non uniform current density distribution near its surface as shown in Figure
1.
In this geometry there are stagnation point with zero current flow at the top and
bottom and the maximum fiow occurs at the sides. This situation is further complicated
by the existence of the Deby Layer where the charges are held in a relatively fixed
position with respect to the membrane and surface and distance from the membrane where
the ions change from fixed layer to a flowing one varies with position. Additionally the
space charge layers may be large enough to change the energy of the ions so that reaction
rate is modified by adding or subtracting energy to the thermal energy of the ions. This
leads to an effective activation energy E' = E + q(l - a)^ where q is the charge and ()) is
the voltage across the bilayer, and a the transfer coefficient. The transfer coefficient takes
into account the fact that the site of the chemical reaction may lake place at various
distances into the space charge layer depending on the site of catalytic protein. This
problem has also been treated in reference (11). Other geometries involving a collection
of cells lead to more complicated current distributions as a large part of any externally
generated field will drive the current between cells and along channels such as blood
vessels.
In addition to changes in the collision frequency the electric field may be significant
in modifying the orientation of large asymmetric molecules with dipole moments and this
in turn may effect the steric factor p in either direction. If the molecules are more favorably
aligned by the field to react, it will increase the probability of a favorable collision. Or the
Fields may orient the molecules in such a direction that the molecules are less likely to react.
The force on a dipole is given by

F = a ' VE

where VE is the electric field gradient and a ' is the dipole inomeni. This force tends
to align the molecule along the field lines. The degree to which this occurs is on the average
proportional to the ratio (x"E/k'T. This is likely to be most important in the large fields that
occur in the vicinity of a membrane, however, to the authors knowledge this effect has ni)l
been studied.
It is interesting to speculate on what the effects of chemical concentration modulation
at the driving frequency of an applied AC signal or its harmonics inay have on biological
systems. Both the ECG and EEG have strong periodic components which have biological
functions. It is reasonable to assume that externally generated chemical signals vvhich are
space and time coherent will be most Important at low levels when the frequency of the
external field corresponds to a natural frequency of the system. (26). It is also likely to be
important when the frequency corresponds to a frequency of a self oscillating chemical
reaction. ( I I )
The Effects of ELF on Chemical Reaction Rates in Biological Systems 43

CONCLUSIONS
For some simple geometries we have shown that chemical reaction rates are functions
of the current flow. The effects of low levels of electric and magnetic fields at low frequencies
are likely to effect chemical reaction rates by changing the encounter rate of the reacting
charged particles rather than their energies. At barriers where the current flow is exponen-
tially related to the applied potential we can expect similar effects on the chemical reaction
rates. This situation is likely to apply for catalytic reactions at a membrane surface, the
binding of charged ions or proteins to the membrane surface and to the transport of ions such
as Ca*"^ through a membrane. The next step in improving our understanding of the effects of
low level electric fields would appear to include the improving of our understanding of how
small periodic changes in chemical reaction rates effect biological systems. It is likely that
the most sensitive systems will be those that involve catalytic reactions and amplification.

ACKNOWLEDGMENT
The author wishes to express his appreciation to George Cole, and Jack Thicsen for
productive conversations. He also wishes to express his appreciation to Dean Astumian for
calling his attention to some of the early work in this field, and to the University of Colorado
and to Samsung for supporting his work in this area.

REFERENCES
I.N. Wertheimer. E Leeper "Electrical Wiring Configurations and Childhood Cancer" .American Journal of
Epidemiology. Vol. 109, No. 3 , 1979
2. D. Savitz, 11. Wachtel. F. Barnes, E. John and J. Tvrdik, "Case-Control Study of Childhood Cancer and
Exposure to 60 H? Magnetic Fields" American Journal Epidemiology. Vol. 128, No. 1. 21-38. 1988
3. S.J. London. D.C. Thomas, J.D. Bowman. E. Sobel, J.M. Peters. "Exposure to Residential Electric and
Magentic Fields and Risk of Childhood Leukeinia" American Journal ofEpidcmiology, Vol. 134.923-937,
1991
4. M. Feychting. A. Ahlbom, "Magnetic Fields and Cancer in People Residing Near Swedish High Voltage
Power Lines" IMM Report 6/92, Karolinska Institute. Stockholm. Sweden
5. F.S. Barnes "Typical Electric and Magnetic Field Exposures at Power Line Frequencies and Their
Coupling to Biological Systems" Advances in Chemistry Series No. 250, Biological Effects of Environ-
mental Electromagnetic Fields. Martin Blank, Ed., Ainerican Chemical Society Books, Washington D.C.
to be published 1995
6. W. T. Kaunc and W. C. Forsythc, "Current Densities Measured in Human Models Exposed to 60 H/
Electric Fields," Bioelectromagnetics. Vol. 6, 13-32, 1985
7. T.S. Tenforde and W.T. Kaune. "Interactionof Extremely Low^ Frequency Electric and Magnetic Fields
with Huinans", Health Physics. Vol. 53. No. 6. 583-606, December 1987. Special Section: Non-Ionizing
Radiation
8. F. X. Hart, K.Cvely, C D . Finch."Use of a Spreadsheet Program to Calculate the Electric FieldCunent
Density Distributions Induced in Irregularly Shaped, Inhomogcneous Biological Structures by Low-Fre-
quency Magnetic Fields" Bioelectromagnetics, Vol. 14. No. 2, 161-172. 1993
9. F.S. Barnes " Interaction of DC and ELF Electric Fields with Biological Materials and Systeius" The CRC
Handbook of Biological Effects of Electromagnetic Fields Charles Polk and Elliot Postov\. Eds.. CRC
Press, Boca Raton, Florida, 1986 and Second ,^ddition to be published 1995
10. Steven S. Zumdahl, Chemistry Second Edition page 546 D.C.Heath Co Lexington Mass. 1989
11. P.W. Atkins. Physical Chemistry Fourth Ed., See Chapter 27 for a more complete discussion. Publisher
W. H. Freeman Co. N.Y. 1990
12. L, Onsager, "Deviations from Ohm's Law in Weak Electrolytes" J. of Chem. Phys. Vol. 2, 599-615. 1934
13. R.J. MacGregor, E.R. Lewis, Neural Modelling, Plenum Press, New York. 1977
44 Frank S. Barnes

14. Y.J. Seto, S.T. Hsieh,"Electromagnetic Induced Kinetic Effects on Charged Substrates in Localized
Enzyme Systems" Biotechnololgy and Bioengineering Vol. XVIII, 813-837, 1976
15. B. Robertson and R.D. Astumian, "Frequency and Amplitude Dependence of the Effect of a Weak
Oscillating Field on Biological Systems" in Charge and Field Effects in Biological Systems-3, M. Allen.
S.F. Cleary, .\. Sowers and D.F. Shillady, F.ds., Birkhauser, 1992
16. R.D. Astumian and B. Robenson. "Nonlinear Effect of an Oscillating Electric Field on Membrane
Proteins" J. Chem. Phys. Vol. 91, No. 8, 15 October 1989
1 7. B. Robertson and R.D. Astumian, "Frequency Dependence of Catalyzed Reactions in a Weak Oscillating
Field" J. Chem. Phys. Vol. 94, No. 11, 1 June 1991
IX. T.^'. Tsong. R.D. Astumian, "Electroconformational Coupling: How Membrane-Bound ATPase
Transduces Energy from Dynamic Electric Fields" Ann. Rev. Physiol. Vol. 50, 273-290, 19X8
19. R.D. Astumian and B. Robertson "Imposed Oscillations of Kinetic Barriers Can Cause an Enzyme to
Drive a Chemical Reaction Away from Equilibrium" J. of the Amer Chem. Soc. Vol. 115, No. 24,
11063-11068, 1993
20. T'.V. Tsong, D.S. Liu, R. Chauvin, "Electroconformational Coupling (ECC): An Electric Field induced
F'nzyme Oscillation for Cellular Energy and Signal Transductions" Bioelectrochemistry and Bioenergel-
ics. Vol. 21. 319-331. 1989
21. D.S. Liu. R.D. Astumian. T.Y. Tsong, ".Activation of Na*and K* Pumping .Modes of (na. K)-ATPase by
an Oscillating Electric Field" J. of Bio. Chem,. Vol. 265, No. 13. 7260-7267. 1990
22. A. Raudino and R. Larter. "Enhancement of Sorption Kinetics by an Oscillatory Electric Field" J. Chem.
Phys. Vol. 98. No. 4. 15 February 1993
23. A. Oraziana, R. Ranjeva and J. Teissie, "E.\ternal Electric Fields Stimulate the Electrogenic Calcium'So-
dium Exchange in Plant Protoplasts" Biochemistry, Vol. 29, 8313-8318. 1990.
24. R.D. .Astumian, J.C. Weaver and R.K, Adair, "Rectificuation and Signal .Averaging of Weak Electric Fields
by Biological Cells" Proc. Natl. Acad. Sci. USA in press 1995
25. R.D, Astumian and M. Bier. "Fluctuation Driven Ratchets; Molecular Motors" Vol. 72, No. 11. Physical
Review Letters, 14 March 1994
26. I .S. Bamcs. "Some Engineering Models for Interactions of Electric and Magnetic Fields With Biological
S),\tems" Bioelcctroniagneties Supplement Vol. 1. 67-85, 1992
 A GROWING SCIENTIFIC CONSENSUS ON
THE CELL AND MOLECULAR BIOLOGY
MEDIATING
Interactions with Environmental Electromagnetic Fields

W. Ross Adey

Veterans Affairs Medical Center and University

School of Medicine
Loma Linda, Calif., 92357

INTRODUCTION
Over the past 15 years, epidemiological studies have raised concerns about possible
health risks related to exposures to electromagnetic fields associated with electric power
transmission, distribution and use; and to various radio and microwave field exposures in
homes and schools, in the workplace, and in the environment.
From the beginning, it has been abundantly clear that energies of tissue components
of these fields are extremely weak; so weak in fact, that credibility of epidemiological reports
of these sensitivities requires fundamentally new concepts of the energetics of living matter.
Here, further laboratory studies will be paramount.
On the other hand, further epidemiological studies appear to hold little prospect of
major progress until a metric for tissue dose can be established. Laboratory studies seeking
this metric now focus on the physics of atomic processes that occur within the framework
of biomolecules.
Over the past decade, remarkable collaborative developments between the physical
and biological sciences have moved these apparently disparate disciplines toward a single
realm of science. Research on fundamental states of matter, in tenns of nonequillbrium slates
and nonlinear electrodynamics, has gone hand in hand with a new vista incorporating these
concepts at the frontiers of cell and molecular biology; and in consequence, there is now in
prospect a new definition of living matter in these physical terms.
Beyond the chemistry of molecules that form the exquisite fabric of living tissues,
we now discern a new frontier in biological organization, perhaps more difficult to compre-
hend. It is based on physical processes at the atomic level, rather than in chemical reactions
between biomolecules; and as we shall see, these physical processes may powerfully regulate
the products of biochemical reactions.
Biological EJfecls of Magnetic and Elearomagnelic Fields, Edited by S. Ueno
Plenum Press, New York, 1996 45
46 VV. Ross Adcv

It will be our purpose here to trace some of the major steps that have led to this broad
synthesis, which describes, at ever finer levels, an hierarchical sequence of events initiated
in biomolecular interactions with environmental EM fields. We will address contributions
of scientific research to some of the more pressing issues facing the electric power and
communication industries, now and in the future.

MAN IN THE ELECTROMAGNETIC ENVIRONMENT

The Natural Low-Frequency Electromagnetic Environment; Changes

Attributable to Use of Electric Power
.Ail life on earth has evolved in a sea of natural low-frequency electromagnetic (EM)
fields. They originate in terrestrial and extraterrestrial sources. Thunderstorm activity in
equatorial zones of South America and Central Africa produces ELF fields that are ducted
worldwide between the earth's surface and the ionosphere. This ELF activity exhibits a series
of peaks, known as the Schumann (1957) resonances, in the spectrum, from 8 to 32 H?. Their
electric components have intensities in air around 0.01 V/m; their magnetic fields reach
intensities of the order of 1-10 nanotesla (0.01-0.1 milligauss), far weaker than the earth's
static magnetic field around 50,000 nanotesla (500 milligauss). The ever-growing use of
electric power over the last century has sharply modified this natural environment in urban
areas (Fig. 1).
Exposure to power-frequency fields far stronger than the natural environment is now
universal in civilized society. Typically, in U.S homes, ambient field levels may be in the
range 0.3-3.0 milligauss (0.03-0.3 microtesla), with substantially higher fields near washing

Rosldentiol Fields

Beoeolti aisttlbutlon lln«

Railtiarisitjvstems
(possengef area)
EtecWc blanket (5 In.)

ElecMc shavef/half dryer (6 In.)

VDTs (coloi) (6 m. to 2 ft.)

Rlgnt-ot-wov 0( 500-kV
transmi^tcm line

Eartfis magnetic Held (OC)

Aluminum smeflets (DC)

1—I I .-J—I—\—I—I—r
lOG 30G lOOG 300G lOOOG
03 3I 10 30 100 300 IG 3G
Magnetic flux daniHy, inMlgouu
O 200 mG ROW limit (moDdmum kaod) in New York and Flortdo
The values given here aie examples of typical oi ovefoge e!(posures thcA may be
encouhtefed in v\e given soutce envlrooment ot ftom use o! cettam opplkjnces, Hlgtief
or kTwer values may occur, the information Is adapted from various s o u c ^ sucti as
EPRI, CXDT, EPA. IITRI. [X)E. BPA. and Washington State Heolth Department.

Figure I. The range of ejectromagnetic field intensities contributed by man-made sources at power line
frequencies, (From Ek'ciric and Mugneiic Fiehis Mechanisms Research. 1994-199H Mulli-vear Program Plan.
US, fJcpartnient of Energy. I944|.
 A Growing Scientific Consensus on tlie Cell and Molecular Biology Mediating 47

machines, refrigerators and air-conditioners. Electric shavers, hair dryers and electric blan-
kets all produce local fields orders of magnitude greater than domestic ambient levels.
Indeed, they may exceed environmental field levels produced by electric power transmission
and distribution systems, but it should be emphasized that exposure to the former may be
for short periods daily, and that they decay rapidly at short distance froin the devices, whereas
in housing near major power lines, exposure to the latter is likely to be for long daily epochs,
typically for many years.

The Biological and Biophysical Dilemma in Reported Tissue Sensitivities

to These Low-Level Environmental ELF Electromagnetic Fields
From the physicist's viewpoint, these are extremely weak fields in terms o^photon
energies necessary to initiate interactions with biomolecular systems. When electromagnetic
radiation is described in terms of these photon "bullets", radiation with wavelengths longer
than the ultraviolet region of the spectrum (350 nanometers), with photon energies less than
about 12 eV (electron volts) has insufficient energy to cause ionization, involving disruption
of atoms. The wavelength of 50 Hz fields is 6,000 km.
For biologists, the functional edifice of living matter has rested on heat exchange as
the sole vehicle in modeling the sequence of animate processes. We are now faced with a
growing realm of biological sensitivities in which tissue heating is not key to the response.
We shall discuss this evidence in detail. Although there has been a persisting view in certain
areas of the physical sciences that nonionizing EM fields are incapable of inducing bioeffects
other than by heating (Foster and Guy, 1986; Foster and Pickard, 1987), evidence to the
contrary now creates a growing international consensus (Adey, 1990, 1993; Grundlcr et al..
1992;Teradaetal., 1994).
For both physicists and biologists, there is the central concern that these sensitivities
involve stimuli below the level of intrinsic energy of thermal atomic collisions (kT). which
has been assumed to set a threshold for all biological stimuli. The physical basis for this

high-frequmcy
therapeutic device

g 10* t

ELF efTecl

10
KIO

10 10 10
SAR |W/kg|
Figure 2. A comparison of absorbed energy from imposed fields at ELF frequencies and from radiofrequency
fields with ELF modulation, in studies where either bioeffects or therapeutic effects were observed. Studies
1-5 refer to therapeutic devices for muscle and joint pains (1 & 2), and to laboratorv' studies of brain calcium
binding and modulation of peripheral vascular mechanisms (3-5). These studies all involved ELF-modulated
radiofrequency fields. Studies 6-10 with ELF fields relate to therapy of wound healing (10) and regulation of
calcium binding in brain tissue (7-9) and in normal and leukemic lymphocytes. (From Terada et al., 1994).
48 W. Ross Adey

sensitivity remains incomplete, but as we shall see, theoretical and experimental evidence
now points to the role offree radicals, formed briefly for nanosecond periods in the course
of all chemical reactions. They offer one mechanism at the microscopic level that transcends
this thermal energy barrier in the first transductive step of tissue-field interaction. Free radical
sensitivity to magnetic fields may extend all the way down to zero field energy (Grundler et
al, 1992; McLauchlan. 1992).
Biomolecular systems exhibit high levels of cooperativity as the basis of responses
to athermal ELF fields, and to radiofrequency fields, amplitude-modulated at ELF frequen-
cies (see Adey, 1990, 1993 for reviews). In brief, this cooperativity appears dependent in
part on coherent states of populations of electric charges on strands of protein. There are
independent reports of biological effects of ELF magnetic fields which show specific
dependence on AC/DC magnetic field combinations.
Theoretical and experimental studies discuss low-frequency resonances for DC
tleld-amplitude and AC field-frequency combinations (LibofT, 1985, 1992), as well as
window phenomena for AC field-amplitude sensitivities. These pioneering studies have led
to a predictive ionic resonance model (Lednev, 1991), based on earlier atomic spectroscopy
theory. Further theoretical and experimental studies by Blanchard, Blackman et al. (1994a
and b) offer an ion paramelric resonance (IPR) model for parallel static and oscillating
magnetic fields, with claimed effects on neurite outgrowth as a function of quasiperiodic.
resonance-based predictions of the IPR model. Theoretical models of these phenomena share
a common problem of competing with thermal and electrical noise inherent in living cells
at normal temperatures. A possible problem for these proposed physical models may relate
to the nanosecond time scales of thermal collisions, which would tend to disrupt millisecond
coherence times essential for resonant interactions with an applied field at frequencies of
1-100 Hz.

MECHANISTIC CONSIDERATIONS AT CELL AND MOLECULAR

LEVELS

Tissue Elements in EM Field Detection; the Role of Cell Membranes

Beyond these physical events in first detection of environmental EM fields in tissues,
there appears to be a general consensus that the site offield action is at cell membranes.
Strands of protein arc strategically located on the surface of cells in tissue, where they act
as detectors of electrical and chemical messages arriving at cell surfaces, transducing them
and transmitting them to the cell interior. The structural basis for this transductive coupling
by these protein strands is well known. Through them, cell membranes perform a triple role,
in signal detection, signal amplification, and signal transduction to the cell interior.
Calcium ions play an essential role at each step in this transmembrane signalling, and
have been used as markers of EM field interactions in a variety of tissues and cell cultures.
Calcium efflux studies, primarily in brain tissue, revealed sensitivities to ELF fields, and also
to LLF-modulated radiofrequency fields (Bawin et al., 1975, 1976; Blackman et al., 1979.
1985 a.b, 1994; Dutta et al.. 1984; Lin-Liu and Adey, 1982). In many instances, these
responses were windowed with respect to field amplitude, or ELF field frequency (or ELF
modulation frequency of radiofrequency fields), or both. More recent calcium influx studies
have created a further consensus on the key role of calcium (Fig. 3).
These studies have focused on cells of the immune system (Lyle et al., 1991;
Walleczek, 1992; Walleczek and Budinger, 1992). In two independent studies with mitogen-
stimulated lymphocytes, exposure to a 3 Hz pulsed magnetic field (in the millitesla range)
A Growing Scientific Consensus on the Cell and Molecular Biology Mediating 49

20.1-30
ES^o.i—'0
E2*o.i-so
Figure 3. Measures of Ca-*(Mir*) uptake ~~>sa.i
in human leukemic Jurkat T-cells chemically o "
activated with a mitogen, and then exposed — z
C 2
to a 3Hz 2mT magnetic field for 2 min.
Sensitivity to the field was determined by 3 c
the initial conditions of the cells. With low
uptake (20,000-30,000 counts.min), there
was a 5% decrease in uptake, but at high
counts, (above 50,000/min) cells were unre- a
sponsive to the field. Results are expressed u
as means and S.E.M.; pcs'min = photon
counts/sec'min. Statistical significance was -2
assessed by the paired t-test and ANOVA.
(From Walleczek. 1994).
Initial Ca** Influx Rat» [ p e t / m i n xlO ]

sharply reduced calcium uptake (Conti et al., 1985a & b; Walleczek. 1992). By contrast, 60
Hz stimulation under the same conditions />7crea.j(j(/calcium influx (Walleczek and Liburdy.
1990). In two other independent studies, both using 3 Hz pulsed magnetic fields, there was
a 55-60% reduction in thymidine uptake into DNA in mitogen-activated lymphocytes (Conti
et al., 1985a & b; Mooney et al., 1986). These observations emphasize the importance oj the
state of activity of cells at initiation of EM field exposure, since cells not already stimulated
with mitogens were unresponsive to the magnetic fields.

Inward Signaling at Cell Membranes; EM Fields and an Enzyme

Cascade
Modulation of cell surface calcium binding by EM fields is consistent with high
cooperativity, since field-induced changes in tissue calciutn efflux are far greater than
accounted for by the energy of the imposed field. It is thus consistent with a major
amplification of the triggering signal. This amplified signal is transmitted along the strands
of receptor proteins to the cell interior, where it activates enzymes located at or near cell
membranes. Enzymes are protein molecules that function as catalysts, mediating chemical
reactions at tissue temperatures that would only occur in their absence at far higher energy
levels (e.g., temperature, pressure). They emerge unchanged from these chemical reactions,
in which they may thus participate repeatedly.
Amplification of the transmembrane signal confinues as it reaches the cell interior.
Many signals carried across the membrane by the receptor strand first interact with a class
of proteins known as G proteins, so named because they are bound to guanosine triphosphate
(GTP). In a further sequence, GTP may stimulate or inhibit the powerful metabolic enzyme
adenylate cyclase (ATP). One receptor protein strand may activate many molecules of G
protein (Luben, 1990), leading to further amplification of the signal {Fig. 4). Magnetic fields
desensitize G protein-linked receptors. Their importance in signaling within cells has been
recognized in the Nobel award (Oilman and Rodbell, 1994).
From their first arrival inside the cell, transmembrane signals activate an enzyme
cascade. For our purpose, we may limit consideration to a series of enzymes identified as
interactive with imposed EM fields. This cascade involves tnetabolic. messenger and cell
growth regulating enzymes.
50 VV. Ross Adc\

PEMF EFFECTS ON PTH RESPONSES IN OSTEOBLASTS

Hormone Binding .Vtta Ooes Mot Chang*
Conformational Changea -PElir Hodlflas
Membrane Fluidity -Not Deteralnad
a Protein Kinetlos 'PBHT May change
Sfleotor Eniyme Activity 'VZHT Does Hot Change
Phosphorylatlon/lnaotivatlon ""Hot Determined

New Synthesis Degradation

Figure 4. Functional diagram of cell membrane, showing sequence of'events associated with modulation of
atlcnvlatc cyclase (AC) activity via G proteins, modeled on hypothetical interactions of pulsed magnetic fields
(72 H/. S gauss peak) with the osteoblast (bone cell) receptor for the parathyroid hormone (PTH) receptor.
Osteoblasts exposed to pulsed magnetic fields for as little as 10 min exhibit a persistent desensitization of the
elfects of PTH on adenylate cyclase (Luben et al.. 1982). Studies using biochemical probes of G protein
coupling (Cain et al.. 19S7) indicate that the ability of bound honnone-receptor complex to activate G protein
<i//;/ii/-subunits is itnpaired by treatment of the bone cell with pulsed magnetic fields. This desensitization of
the PI H receptor results in increased synthesis of collagen (a first step in bone formation) by the bone cell,
and a decreased resorption of bone by bone-destroying osteoclasts (Cain and Luben, 19S7). Both effects would
favor increased hone formation in regions exposed therapeutically to pulsed magnetic fields (Rassett et al..
1482).

a. Cell metabolic activity is fueled by adenosine triphosphate (ATP). Breakdown

of ATP involves the cell inembrane-related enzyme adenylate cyclase. Much evidence of
its sensitivity to EM field exposure has come from studies of its role in fracture healing
(Luben, 1989, 1990; Luben and Cain, 1984; Luben et al., 1982), Signals from horrnones
binding to cell surface receptors activate adenylate cyclase. This activation is strongly
modulated by a variety of low-frequency magnetic fields (Fig. 5). By removal of phosphate
groups, adenylate cyclase converts ATP to cyclic-3',5'-adenosine tnonophosphaie
icAMP). with release of considerable thermal energy. In turn, cAMP continues this enzyme
cascade by transferring phosphate groups to a family of messenger enzymes, the protein
kinases.
b. Messenger protein kinase enzytnes, activated by these membrane-related signals,
can move widely through the cell interior, signalling to many structures, including the cell
nucleus. Their activity is sensitively modulated by ELF and ELF-modulated radiofrequency
fields (Byus et al., 1984; Luben. 1994).
A Growing Scientific Consensus on the Cell and Molecular Biology Mediating 51

1,250

S 1,000
/
As
/
/ -^
/
S 750 /
u
o
"H
E

>; 500

s
250 -
Figure 5. Adenylate cyclase activity in
bone cells, stimulated with parathyroid
hormone (PTH), with and without an im-
posed 72 Hz pulsed magnetic field widely
used in therapy of ununited hone fractures
(from Luben ei al.. 1982), 10
PTH, ng/ml

c. Cell growth and DNA synthesis involve the enzyme ornithine decarhoxylasc
(ODC) in key regulatory steps. ODC synthesizes polyamines (putrescine. spermidine,
spermine) from the amino acid ornithine. The snake-like polyamines have the highest electric
charge/mass ratio of any biomolecule. ODC activity is inodulated by ELF electric and
magnetic fields (Byusct al., 1987; Cainet al,, 1993), and by EI.F-tnodulated radiofrequency
fields (Byus et al.. 1988; Litovitz et al.. 1993).
In cell nuclei, polyamines are essential in DNA synthesis. Elevated levels of polyami-
nes inside cells also triggers transcription oftheproto-oncogenes c-myc and c-fos by c- ras.
Thus, polyamines may participate in a cascade of events leading to communication between
memhrane-hound and nuclear oncogene products fTabib and Bachrach, 1994),
Polyamines are also exported to cell surfaces (Byus et al., 1994: Tjandrawinata et
al., 1994), where they exercise a quite separate regulatory role (for review, see McBain and
Mayer, 1994). acting to either stimulate or inhibit the specific receptor (NMDA receptor) for
the amino acid L-glutamate, which contributes to excitatory synaptic transmission at sites
throughout the brain and spinal cord. Polyamine stimulatory or inhibitory actions depend on
the level of a cell's membrane potential, and on the specific polyamine molecule.
Thus, there emerges an experimental model, in which electromagnetic fields at
athermal levels are first sensed by cell surface receptors, amplified, signalled to the cell
interior, and then, through an enzyme cascade, determine production of chemical modulators,
some of which are returned to the cell surface as their prime site of action. Signals from this
cascade also initiate gene transcription in nuclear DNA (Phillips, 1993).

Outward Signaling through Cell Membranes; EM Fields as Modulators

of Cell-To-Cell Communication in Regulation of Cell Growth
Cells in normal tissue may be considered an organized society, "whispering together"
in a faint and private language. They communicate with their neighbors through chemical
52 V\. Ross \de\

Stimuli, and through electrical fields far weaker than the huge electric barrier of the
membrane potential. Gap-junctions, specialized plaques of protein placed between neigh-
bouring cell membranes, provide electrical and chemical coupling between them (Loewcn-
stein. 1981).
They are perforated by numerous tiny tubes (conncxons) fomiing paths through
which molecules synthesized in one cell reach the cytoplasm of its neighbour. This metabolic
cooperation is essential in the regulation of normal growth (Pitts and Finbow. 1986).
By contrast, "cancer can be regarded as a rebellion in an orderly society of cells.
Cancer cells neglect their neighbours and grow autonomously over surrounding normal cells.
Since intercellular communication plays an important role in maintaining an orderly society,
it must be disturbed during the process of carcinogenesis" (Yamasaki, 1987. 1990). Disrup-
tion of gap-junction communication can lead to unregulated cell growth (for review, see
.Ade>", 1992). This may he reversed in cultures of cancer cells, if they make contact with
normal cells (Newmark. 1987). Specific viral oncogenes cause differential effects on
cell-to-cell communication, in ways that relate to suppression by normal cells of unregulated
cancer cell growth, when the two cell types are grown together (Bignami et al.. 1988).
Introduction of an activated H-ra.v-1 oncogene into epithelial cells significantly reduces
gap-junction communication, possibly from decreased cAMP-dependent phosphorylation
(See Inward signaling above) of the principal gap-junction protein.
Experimental evidence supports a role for EM fields in modulation of this gap-jimc-
tion communication. We shall consider it in the frame of currently accepted models of
carcinogenesis and tumour formation.

Multistage Carcinogenesis: Initiation, Promotion and Progression

There is a consensus that tumor formation involves at least two steps: an early step
oi'initiation and a laicrpromotion effect. A single agent may cause both events, or two or
more separate agents may be necessary, working together in the proper sequence. The time
between exposure to an initiator and appearance of the disease (latent period) is often 20
years or more (Fig. 6).
Initiation involves damage to genetic stores of DNA in cell nuclei, but the changes
are not expressed, i.e., a tumour does not result unless one or more promoting agents act
repeatedly at a later time. Initiation may be a single event, as in exposure to ionizing radiation
from X-rays or atomic explosions, or to certain chemicals, including tar derivatives. Initiated
cells are transformed (mutated). They are cancer cells, but may remain quiescent if no!
stimulated by a promoter.
Promotion results froin agents having little or no initiating action when tested alone,
but markedly enhance tumour yield when applied repeatedly and intermittently following a
low dose of an initiator (Slaga et al.. 1978; Berridge et al.. 1988; Nishizuka, 1984).

Figure 6. Model of multistage carcinogenesis IViiiii stud-

UU U Ii ies in mouse skin, initiation results from onl\ a single
PROMOnON exposure to a carcinogen that appears to damage nuclear
INITIATION PRCX3RESSION
STAGES 1,2,... DN.'\. Promotion invloves multiple exposures at certain
intervals that do not damage DNA directly. Promotion
leads to conversion of benign to malignal tumors. \\ iili
progression increasing the degree of malignancy. (.After
INCREASING MALIGNANCY Weinstein. 1488).
A Growing Scientific Consensus on tlie Cell and Molecular Biology Mediating 53

They do not act primarily on nuclear DNA, and many are known to act by binding
to receptors at cell membranes (Weinstein, 1988).

Epigenetic (Non-Genotoxic) Carcinogenesis: Evidence for

Electromagnetic Field Actions in Tumor Promotion
New lines of research in tumor formation reflect identification of a growing number
of agents, tumor promoters, that appear to play a causal role in human cancer, without direct
action on nuclear DNA stores. Pitot and Dragan (1991) have summarized the importance of
this approach as a cutting edge in current and future cancer research.
In the genotoxic, multistage model of carcinogenesis outlined above, which predicts
multiple mechanisms and multiple defined stages of initiation, promotion and progression,
transitions between successive stages can be enhanced or inhibited by different types of
agents. Development of a fully malignant tumor involves complex interactions between
environmental (chemicals, ionizing and non-ionizing radiation, viruses) and endocrine
(genetic, hormonal) factors (Weinstein, 1988).
The epigenetic (non-genotoxic) model focuses on action of tumor promoting agents,
many in primary interactions with membrane-associated receptors (see Section 3). The most
widely used chemical promoter in experimental cancer studies is the phorbol ester TPA. With
the discovery that TPA activates the membrane-bound messenger enzyme protein kinase C
(PKC) (Castagna et al.. 1982), and subsequent studies indicating that PKC is the major
cellular receptor for TP.^ (Nishizuka, 1983, 1984). a series of bridges now unite research on
tumour promotion, growth factors, signal transduction including the action of EM fields.
and the action of specific oncogenes (Phillips etal.. 1992, 1993; Adey, reviews 1990, 1992).

Experimental Evidence from Cell and Animal Studies Supports a Model

of Joint Actions of Chemical Tumor Promoters and EM Fields at Cell
Membranes
Cell promotion model. When normal cells (fibroblasts) and their daughter cells,
previously mutated with ultraviolet light and behaving as cancer cells with uncontrolled
growth, are placed together in cell culture (co-cultures), contact with the parent cells inhibits
the unregulated growth. The tumor promoter TPA unbalances this contact equilibrium,
allowing return of unregulated growth of mutant daughter cells and formation of tiny tumors
(foci) in the culture dish.
Exposure of this system to a 60 Hz magnetic field (sinusoidal, 0.1 mT, 1 h exposure
4 times daily) increased by 60% the number of TPA-induced foci (p < 0.001) (Fig. 7), with
an approximate doubling of the size and cell density of the foci (Cain et al., 1993). Fields
alone had no effect.
In this system, cells grow to confluence in 10-14 days, and to a mature monolayer in
28 days. Thus, the intermittent exposures occurred while the cells were growing, in the
presence of TPA; suggesting that 60 Hz magnetic fields act in conjunction with chemical
tumor promoters to enhance development and expression of the cancerous daughter cells
scattered amongst normal parent cells. Thisjoint action of cheinical promoters and EM fields
is known as co-promotion.
Cancer promotion in a mouse skin model. A counterpart animal study to the foregoing
cell culture smdy has used mice in an initiation-promotion experiment (McLean et al., 1991).
The skin of the back was initiated with a single subthreshold dose of the carcinogen dimethyl-
benzanthrene (DMBA). The mice were then exposed to a 2 mT 60 Hz magnetic field for 21
weeks, to test whether the field would act as a tumor promoter. No tumors developed in either
 54 \V. Ross Adev

50
A.) NUMBER OF FOCI
X
40 **
a I B FIELD + TPA
30 0 3 SHAM + TPA

20
o
10

0
EXPl EXPS EXP3
100
B.) TOTAL AREA OF FOCI
80 **
" I B FIELD + TPA
60 g 2 3 SHAM + TPA.

40

20

o
EXPl EXP2 EXP3
800
c) TOTAL DENSITY OF FOCI
++
^ 600 -- - Figure 7. Normal fibroblast cells and their daughter
FIELD + TPA niutants were grown together in culture. The parent
F","vj SHAM + TPA _ cells inhibited uncontrolled growth of the mutants.
400 -- This balance was disturbed if a chemical cancer
promoter (TPA) was added in nanomolar concentra-
**
en
a. lOQ
i I tions to the co-culture. Foci ("tumors in a dish")
reappear. Tjeir number, size and density were ap-
proximately doubled in the presence of a 60 Hz. 1
i
EXPl
i
EXP2
ll '
EXP3
gauss magnetic field. (From Cain et al.. 19931.

sham- or field-exposed animals. Two additional groups were then treated weekly with the tumor
promoter TPA. The time to tumor appearance was shorter (hut not statistically so) in the group
exposed to magnetic fields and TPA. This study, which has limitations from small sample size,
also suggests reduced immune surveillance by natural killer (NK) cell activity, which would
otherwise prevent or retard growth of some tumor cells.

From Cells to Tissues to Organ Systems: Evidence for ELF Field

Actions in Immune Responses and Brain Neuroendocrine Mechanisms
Immune suiTeillance protects the body against infection and the creeping tentacles of
cancer, sensing the difference between self and non-self Cellular immunity, as distinct from
humoral immunii)- mediated by antibodies circulating in the blood, is mediated by lymphocytes in
blood and tissues. It includes the natural killer (NK) cells cited above. NK cells are present in most
organs, and without prior immunization, can recognize, bind to and destroy (lyse) malignant cells.
Lymphocytes may also be targeted against tumor cells, again destroying them by actual contact
(allojyeneic cytotoxicity). In cell culture studies, this killing capacity was reduced by 60 Hz electric
fields (Lyleetal., 1988), and by ELF-modulatedradiofrequency fields (Lyleetal., 1983)(F/ij. H).
A Growing Scientific Consensus on the Cell and Molecular Biology Mediating 55

25 I -

16 40 60 80 100
Modulation Frequency (Hz)

Figure 8. The ELF modulation frequency determines the amount by which killing capacity of lymphocytes,
targeted against tumor cells (allogeneic cytotoxicity), is reduced when a inodulated radiofrcquency field (450
MHz, 1.5 mW cm"'') is introduced. The rnaximum reduction, over 20%. occurs with 60 Hz modulation. (From
Lyleetal., 1983).

Cultured human tonsil lymphocytes are also sensitive to ELF-modulated radiofrequency fields,
responding with a sharp but transient reduction in messenger protein kinase activity (Byus et al.. 1984).
Brain neuroendocrine sensitivities to ELF fields have centered around studies of the
pineal gland hormone melatonin. Melatonin is synthesized and secreted with a strong
circadian rhythm, reaching a nocturnal peak around 2.0 a.m. in man and animals (Reiter,
1986). The cycle is variably sensitive to the day/night ratio of light exposure in different
species. The question of its possible susceptibility to a changing electromagnetic environ-
ment has been the subject of intense study (Semm, 1983; Wilson et al., 1981, 1990). Some
results of these studies remain unclear within and between species. The most consistent
results from magnetic field exposures have been in the use of the Djungarian hamster (Yellon,
1994).
Acute exposures of long-day adult hamsters to a 60 Hz magnetic field (0.1 mT, 15
min) 2 h before lights off suppresses the night-time rise in melatonin in the pineal gland and
in the blood. Acute exposures of short-day animals produced similar results (Yellon, 1994).
By contrast, daily exposures of the same type for as long as 3 weeks had no effect.
Beyond diurnal activity rhythms, melatonin is key to a broad range of regulatory
mechanisms (Reiter, 1992), including the immune system, reducing incidence of certain
cancers in mice, and inhibiting growth of breast cancer cells (Hill and Blask, 1988). This
inhibitory action of melatonin on human breast cancer cell growth is reported to be blocked
by 60 Hz magnetic fields at a 1.2 uT threshold level (Liburdy et al., 1994). Moreover, patients
with estrogen receptor-positive breast cancer have lower nocturnal plasma melatonin levels
(Tamarkin et al., 1982).

Tissue Detection of Nonionizing EM Fields at Atliermal Levels; the

Search for the First Transductive Step in Free Radical Mechanisms
We have traced major steps in what has been learned of steps in an energetic hierarchy
that couples to the cell interior cell surface signals originating in ELF fields. Our point of
56 \V. Ross Adc>

departure has not considered the challenging first step in this sequence: the first detection
ofFM fields, often orders of magnitude weaker than the averaged thermal energy AT in the
sensing biomolecular substrate.
Thermodynamic considerations of tissue thermal energy do not necessarily impose
a threshold limit on this primary interaction. Nonthermal states and nonlinear electrodynam-
ics in biological systems offer a number of possibilities (.Adey, 1981, 1993; Adey and
Lawrence, 1984; Grundler et al, 1992; Frohlich, 1988^; Kaiser, 1988; Walleczek. 1994). One
answer to this important question may be found in EM field interactions with free radicals.
Free radicals are atoms or molecules with one or more unpaired electrons, which
typically make free radicals highly chemically reactive (Walleczek, 1994). In chemical
reactions, bonds break and reform. Most chemical bonds between atoms involve paired
electrons with opposite spins, with one electron derived from each partner in the union. The
spinning electron generates a tiny magnetic field. Chemical bonding occurs between atoms
having opposite electron spins, and thus, magnetic fields of opposite polarity.
In a chemical reaction, the bond breaks and each partner reclaims its electron from
the bond, moving away to encounter a new partner. It is now an unattached, highly reactive
free radical. Reforming a bond requires a meeting between two radicals with opposite
electron spins, the union producing a singlet pair. The lifetime of free radicals is typically
short, in the range of microseconds to nanoseconds.
It is in this brief period that irnposed niagnetic fields may alter the rale and amount
of product of a chemical reaction. Since the effect is only on the kinetics of chemical
reactions, they are known as magnetokinetic effects (Sleiner and Ulrich, 1989). They occur
only in nonthermal states of biomolecular systems, defined as an insensitivity to random
thermal interactions during the brief period of their existence (Walleczek, 1994). They are a
consequence of a coherent quantum-mechanical step which accompanies free radical forma-
tion. McLauehlan (1992) has proposed a role for free radicals in mediating biomolecular
interaction with magnetic fields at 50 and 60 Hz electric power frequencies. In his model
(Fig. 9), very low static magnetic fields cause triplet pairs to break and become singlets. ,^t
higher levels around 8 mT, two-thirds of the radical pairs may not react as they would in a
weaker field, "an enormous effect of a small magnetic field on a chemical reaction, and the
effect begins at the lowest applied field stength....The all-important interaction has an energy
very much less than the thermal energy of the system, and is effective exclusively through
its influence on the kinetics; this is counter-intuitive to most scientists."
Although electron spin energies are conserved through thermal collisions, their short
lifetime in relation to thermal collision frequencies has raised unre.solved questions about
the proposed radical-radical interactions. In a synthesis emphasizing nonthermal interactions
of EM fields with cellular systems, Grundler et al. (1992) present models of the sequence of
EM field transductive coupling, based on magnetic field-dependent chemical reactions,
including cytochrome catalyzed enzymatic reactions that involve transient radical pairs and
production of free radicals, such as reactive oxygen and nitric oxide, leading to further highly
cooperative amplifications. They conclude that "imposed fields can be active even at
intensities near zero." In other words, a threshold might not exist in such a system. There is
growing experimental evidence supporting free radical models of magnetic field transduc-
tion in biological systems.
More than 20 enzymes are thought to incorporate radical chemistry in the conversion
ofsubstrates to products (Harkins and Grissom, 1994). As examples, i) static magnetic fields
in the range 0.1-0.15 T decreased the kinetics of the enzyme ethanolamine ammonia lyase
by 25 to 60%, consistent with magnetic field-induced changes in crossing rates between
singlet and triplet radical states; and ii) membrane-related P-450 cytochrome enzymes are
free radical-dependent and can be modulated by the putative P-450 inhibitor econazole.
A Growing Scientific Consensus on tiie Cell and Molecular Biology Mediating 57

il

i
111

f III

c
1-
\ * o c
<a
\ « ff W
~ CJ s

X 01 S. (p / i
V in £ b
o

y '1o
CO r- :

Ov —
-r
',c
01
g
M
CJ
1^
u. £ o
58 W. Ross Ade\

This agent completely inhibited (even reversed) increased calcium influx into human
T lymphocytes induced by a 60 Hz, 2 mT magnetic field (Walleczek et al.. 1994).
Nilric oxide (NO) Is a powerful free radical, synthesized in cells of many tissues, and
as a gaseous molecule, diffusing rapidly beyond the cell of origin. It is an important
physiological modulator in brain and spinal cord, and stimulates vascular tissue, mediating
penile erection. It has been implicated in the pathology of stress diseases, including coronary
artery disease, and in Parkinson's and Alzheimer's diseases of the brain. NO is a modulator
of regularly recurring patterns of rhythmic slow electrical waves (EEG) in brain tissue.
Imposed magnetic fields (1 Hz, 0.1 iTiT|,,p) disrupt the regularity of these fitCJ wave bursts
through a NO-dependent mechanism (Bawin et al., 1994). Pulsed radiofrequency fields
increase NO and cyclic-GMP levels in the rat cerebellum (Miura et al., 1993).

ON THE ISSUE OF CUMULATIVE DOSE; LABORATORY

EVIDENCE

In the most general terms, epidemiological data have suggested a relationship

between duration of exposure to environmental EM fields, typically measured over years,
and the onset of a disease process. Inherently, these long-term exposures are intermittent,
whether measured on a diurnal basis or on a longer time scale. However, use of a Time-
VVeighted-Average (TWA) as an exposure metric has been of limited value in seeking a
precise correlate with disease risk, nor have other measures of the electromagnetic environ-
ment, such as ambient field levels, established a reliable measure of dosc-dcpendence.
Though not yet offering a solution, some laboratory studies may now point the way to future
research.
A recurring aspect of laboratory studies has been observations of O.V- and ()FF-ef-
fect.s. As ON-effects, cell responses to EM field exposure may be transient, even though the
exposure is sustained. This was noted in human lymphocyte protein kinase responses to
ELF-modulated radiofrequency fields; an initial rapid decrease in activity of 60".i within
15-30 min was followed by a return to control levels within an hour, during continuing
exposure (Byus et al., 1984).
In bone cells exposed to pulsed 72 Hz magnetic fields used in fracture therapy, Lubcn
et al. (1982) noted that bone cells are desensitized to the effects of parathyroid hormone
(PTH) during exposure, but regained sensitivity as an OFF-effect. Luben et al. (1994) have
proposed that interference with membrane-mediated signal transduction is a plausible
mechanism by which low energy magnetic fields may influence intracellular processes. Their
study compared actions of 60 Hz, 0.1 mT sinusoidal magnetic fields with actions of the
tumour promoter TPA. Both produced rapid transient increases in the cell membrane of the
messenger enzyme protein kinase C (PKC), but reduced activity in the cell cytoplasm.
Turning off the field led to a recovery of PKC. as determined by a 4-5 fold increase in total
cell PKC activity over a 4 h period; during this exposure period, there was maximum
desensitization of the PTH receptor.
In a different time frame, epidemiological studies have raised questions about
potential health effects of large magnetic transients and heavy starting loads associated with
initial operation of domestic and industrial equipment. Laboratory studies have considered
the coherence time during which an EM field must be sustained at a specific ELF frequency,
before switching to a different but closely related frequency, in order to initiate an enzyme
response. Litovitz et al. (1993) tested ornithine decarboxylase (ODC) responses to ELF
components of an EM field that switched to either 55 or 65 Hz. They reported that the
A Growing Scientific Consensus on the Cell and Molecular Biology Mediating 59

minimum coherence time for this particular response was between 1 and 10 sec to reliably
elicit an enzymatic response.
It is clear that these studies are merely pointers to much needed information on an
exposure metric for health effects in man. Nevertheless, they emphasize the profound
complexity of tissue effects of nonionizing EM fields. Unlike ionizing radiation, where tissue
dose may be calculated relatively simply from the product of tissue radiation intensity and
duration of exposure, the nonionizing metric must take account of profound tissue effects
attributable to intermittency of exposure, and the frequently transient character of the ensuing
biological response.

SUMMARY AND CONCLUSIONS

Over the past 15 years, there have been emergent concerns that the great and growing
use of electric power and radiofrequcncy communication systems throughout the world, with
immeasurable benefits to all mankind, may carry a burden of adverse health effects. In this
same era, research to evaluate levels of possible hazards has followed a dual course, with
epidemiological and laboratory studies proceeding in parallel, but with few options for
coordination.
The scope and content of this laboratory research in bioelectromagnetics has been
significantly fettered in scope and content, because the only major sources of its quite meagre
funding through government agencies has come from mandates in hazard research.
Despite these encumbrances, bioelectromagnetic research appears to have swept
beyond immediate goals in hazard research, to approach a first understanding of the essential
nature of living matter in terms of physical processes at the atomic level, far beyond the
realm of chemical reactions in a biomolecular fabric.
Laboratory studies have identified cell membranes as the primary tissue site of
interaction with environmental electromagnetic fields. They have determined major se-
quences in the coupling of cell surface signals to a cascade of high energy enzymatic
mechanisms inside cells, including mechanisms regulating cell growth. These studies point
to joint actions of chemical cancer promoters and EM fields at cell membranes as key steps
in tumour formation. The role of free radicals in first detection of EM fields at athermal
levels is supported by biophysical models and experimental data.
On the one hand, the importance of this new knowledge emphasizes emergence of
bioelectromagnetics as an interdisciplinary field at the frontier of both physical and life
sciences, holding prospects for major new advances in understanding functions of the human
body in health and disease. On the other, without much further fundamental research, there
are few prospects of developing a metric for tissue dose in EM field exposure; and without
a metric, further epidemiological studies appear to hold little prospect of major progress.
The last decade has seen a progressive erosion of the normal evaluation of new
knowledge by qualified experts in particular fields of medical science and medical practice
through peer review and collegial exchange. In its place, there is the troubling specter of
corporate lawyers importunate in their pursuit of research still in progress, in espousing a
litigant's cause, rather than diligent in legal discovery aimed at establishing scientific
credibility. This is exemplified in publicity afforded courtroom distortions of painstakingly
acquired knowledge about interactions of the human body with environmental electromag-
netic fields. More than 1900 years ago, Tacitus knew well the fate of a Roman society that
had abandoned itself to the machinations of lawyers: "If noone paid for lawsuits, there would
be less of them! As it is, feuds, charges, malevolence and slander are encouraged. For just
as physical illness brings revenue to doctors, so a diseased legal system enriches advocates."
60 VV. Ross Adey

REFERENCES
Adey WR {1981) Tissue interactions with nonionizing electromagnetic fields. Physiol. Rev. 61:435-5l4.
Adey WR (1990) Joint actions of environmental nonionizing electromagnetic fields and chemical pollution in
cancer promotion. Einininmeiual Hecillh Perspectives 86:297-305.
Adey WR (1992) Extremely low frequency magnetic fields and promotion of cancer. Interaction Mccliunisms
of Low-Level Eleclromagnelic Fields in Living Systems. B. Norden and C. Ramcl, eds. Oxford
University Press, pp 23-46.
Adey \VR( 1993) Electromagnetics in biology and medicine. .Worfe'ni Radio Science. H. Matsumoto. cd, Oxford
University Press, pp 227-245.
Adey WR, Lawrence AF (1984) S'imlinear Electrodynamics in Biological Systems. New York, Plenum Press.
Bassctt CAL, Mitchell SN, Gaston SR (1982) Pulsing electromagnetic field treatment in ununited fractures
and failed arthrodeses. JAMA 247:623-628.
Raw in SM, .Adey WR (1976) Sensitivity of calcium binding in cerebral tissue to weak electric fields oscillating
at low frequency. Proc. Nat. Acad. Sci. USA 73:1999-2003.
Bawin SM, Kaczmarek LK, Adey WR (1975) Effects of modulated VHP fields on the central nervous system.
Ann. NY Acad. Sci. 247:74-80.
Bawin SM. Satmary WM, Adey WR (1994) Nitric oxide modulates rhythmic slow activity in rat hippocampal
slices. NewoRepori 5:1869-1872.
Berridge MJ (1987) Inositol triphosphate and diacylglycerol: two interacting second messengers. Ann Rev.
Biochem. 56:159-193.
Bignami M. Rosa S, Falcone G, Tato F, Katoh F, Yamasaki H (1988) Specific viral oncogenes cause differential
effects in ccll-to-ccll communication, relevant to suppression of the transformed phenolype by normal
ccWs.Molec. Carcin. 191:67-75.
Blackman CF (1994) Effect of electrical and magnetic fields on the nervous system. The I'ulnerahle Brain and
Environmental Risks. Vol 3: Toxins in Air and Water RL Isaacson and KF Jensen eds. Plenum Press.
New York, pp 341-355.
Blackman CF, Elder J A. Weil CM. Benane SG. Eichinger DC, House DE(1979) Induction of calcium ion fiux
from brain tissue by radio frequency radiation: efi'ects of modulation frequency and field strength.
Radio Sci. 14:93-98.
Blackman CF, Benane SG, House DE, Joines WT (1985a) Effects of ELF (1-120 Hz) and modulated (50 Hz)
RF fields on the efflux of calcium ions frotn brain tissue, in vitro. Bioelectromagnetics 6:1-12.
Blackman CF. Benane SG, Rabinowitz JR. House DE (1985b) a role for the magnetic field in the radiation-in-
duced efflux from brain tissue, in vitro. Bioelectromagnetics 6:327-338.
Blackman CF. Blanchard JP, Benane SG. House DE( 1994) Empirical test of an ion parametric resonance model
for magnetic field interactions with PC-12 cells. Bioelectromagnelics 15:239-260.
Blanchard JP, Blackman CF (1994) Clarification and Application of an ion parametric resonance model for
magnetic field interactions with biological systems. Bioelectromagnelics 15:217-238.
Byus CV, Lundak RL, Fletcher RM, Adey WR (1984) Alterations in protein kinase activity follow ing exposure
of cultured lymphocytes to modulated microwave fields. Bioelectromagnetics 5:34-51.
Byus CV. Picpcr SE. Adey WR (1987) The effects of low-energy 60 Hz environmental electromagnetic fields
upon the growth-related enzyme ornithine decarboxylase. Carcinogenesis 8:1385-9.
Byus CV. Kartun K, Piepcr S. .Adey WR (1988) Increased ornithine decarboxylase activity in cultured cells
exposed to low energy modulated microwave fields and phorbol ester tumor promoters. Corner Res.
48:4222-4226.
Byus CV (1994) Regulation of the etTlux of putrescine and cadaverine from rapidly-growing cultured
R.'\W-264 cells by extracellular putrescine. Biochem J., in press.
Cain CD, Luben RA (1987) Pulsed EMF effects on PTH stimulated cAMP accumulation and bone resorption
in mouse calvariae. Inieraclions of Biological Systems with ELF, LE Anderson, BJ Kelman, RJ Weigel.
eds. Richland, WA: Battelle Laboratories Press; Conference Publication No. 24. pp 269-278.
Cain CD, Thomas DL, Adey WR (1993) 60 Hz magnetic field acts as co-promoter in focus formation of
C3H10Tl,2cells. Carcinogeiie.-iis 14:955-960.
Castagna M, Takai Y, Kaibuchi K., Sano K, Kikkawa U, Nishizuka Y (1982) Direct activation of calcium-ac-
tivated phospholipid-dependent protein kinase by tumor promoting phorbol esters. / Biol Chem.
257:7847-51.
Conti P. Gigante GE, Alesse E, Cifone MG. Fieschi C, Reale M, Angeletti PU (1985a) A role for calcium in
the effect of very low frequency electromagnetic field on the blastogcnesis of human lymphocytes.
FEBS Lett. 181:28-32.
 A Growing Scientific Consensus on the Cell and Molecular Biology Mediating 61

Conti P, Gigante GE, Cifone MG, Alesse E, Fieschi C, Angeletti PU (l9S5b) Effect of electromagnetic field
on two calciutn dependent biological systems. / Bioelectr.4:2n-27i(>.
Dutta SK, Subramoniam A, Ghosh B, Parshad R (1984) Microwave radiation-induced calcium efflux from
brain tissue, in vitro. Bineteclwmagnetics 5:71-78.
Foster KR, Guy AW (1986) The microwave problem. Sci. Amer 225(.'!):32-39.
Fo.ster KR, Pickard WF (1987) Microwaves: the risks of risk research. Nature London 330:5.'' 1-532.
Frohlich H, ed. (1988) Biological Coherence and Response to External Stimuli. Heidelberg. Springer.
Grundler W, Kaiser F. Keilmann F, Walleczek J (1992) Mechanics of electromagnetic interaction with cellular
systems. Naturwissenschaflen 79:551-559.
Harkins TT, Grissom CB (1994) Magnetic field effects on B|i elhanolainine ammonia lyase: evidence of a
radical nicchanisni. Science 263:958-960.
Hill SM, Blask DE (1988) Effects of the pineal hormone melatonin on the proliferation and morphological
characteristics of human breast cancer cells (MCF-7) in culture. Cancer Res. 48:6121-6126,
Kaiser F (1988) Theory of non-linear excitation. Biological Coherence and Response to External Stimuli. H.
Frohlich, ed. Springer, Heidelberg, pp 25-48.
Liboff AR (1992) The '"cyclotron resonance" hypothesis: experimental evidence and theoretical constraints.
Interaction .Mechanisms of Low-Level Electromagnetic Fields in Living Systems. B. Norden and C
Ramel, eds. Oxford University Press, pp 130-147.
Liburdy RP, Harland JD, Heffeman C, Seeley M, Dunham EE (1994) Inhibition of melatonin's natural
oncostatic action on MCF-7 cells: 60 Hz dose threshold determination. Bioelectromagnetics Society.
16th Annual Meeting, Copenhagen. Proceedings, p 51 (abstract).
Lin-Liu S, Adey WR (1982) Low frequency amplitude-modulated microwave fields change calcium efflux
rates from synaptosomes. Bioelectromagnetics 3:309-322.
Litovitz TA, Krause D, Penafiel M, Elson EC, Mullins JM (1993) The role of coherence tiine in the effect of
microwaves on ODC activity. Bioelectromagnetics 14:395-403.
Loewenstein WR (1981) Junctional intercellular coinmunication: the cell-to-cell communication channel.
Physiol. Rev. 61:829-913.
Luben RA(1989) Effects of low-energy electromagnetic fields (pulsed and DC) on membrane signal transduc-
tion processes in biological systems. Health Physics 61:15-28.
Luben RA, Cain CD (1984) Use of horinone receptor activities to investigate the membrane effects of low
energy electromagnetic fields. Nonlinear Electrodynamics in Biological Systems, WR .Adcy and AF
Lawrence, eds. New York. Plenum Press, pp 23-34.
Luben RA, Cain CD, Chen M-Y. Rosen DM, Adey WR (1982) Effects of electromagnetic stimuli on bone and
bone cells, in vitro: inhibition of responses to parathyroid hormone by low-energy, low-frequency
fields. PTOC. .Vo//. .-trarf. 5r/. fAS/f 79:4180-3.
Luben RA, Hyunh D, Weinshank RL, Smith LE, (1990) Molecular cloning of candidate sequences for the
mouse osteoblast parathyroid hormone receptor. Calcium Regidation and Bone Metabolism. DV
Cohn, FH Glorieux, TJ Martin, eds. Amsterdam, Elsevier pp 39-44.
Luben RA, Morgan AP. Carlson A, Duong M (1994) One gauss 60 Hz magnetic fields inodulate protein kinase
activity by a inechanism similar to that of tumor promoting phorbol esters. Bioelectromagnetics
Society 16th Annual Meeting, Copenhagen. Proceedings p 74.
Lyle DB, Ayotte RD. Sheppard .AR, Adey WR (1988) Suppression of T lymphocyte cytotoxicity following
exposure to 60 Hz sinusoidal electric fields. Bioelectromagnetics 9:303-313.
Lyle DB, Schechter P, .Adey W'R, Lundak RL (1983) Suppression of T lymphocyte cytotoxicity following
exposure to sinusoidally ainplitude-modulated fields. Bioelectromagnetics 4:281-292.
Lyle DB, Wang X, Ayotte RD, Sheppard AR, Adey WR (1991) Calcium uptake by leukemic and normal
T-lymphocytes exposed to low frequency magnetic fields. Bioelectromagnetics 12:145-156.
McBain CJ, Mayer ML (1994) N-methyl-D-aspartic acid receptor structure and function. Physiol. Rev.
74:723-760.
McLauchlan K (1992) Are environmental electromagnetic fields dangerous? Physics World pp 41 -45. .lanuary.
McLean JRN, Stuchly MA, Mitchel REJ, Wilkinson D, Yang H,Goddard M, Lecuyer DW, Schunk M, Callary
E, Morrison D (1991) Cancer promotion in a mouse-skin model by a 60 Hz magnetic field: 11. Tumor
development and immune response. Bioelectromagnetics 12:273-288.
Miura M, Takayama K, Okada J (1993) Increase in nitric oxide and cyclic-GMP of rat cerebellum by
radiofrequency burst-type electromagnetic field radiation../. Physiol. London 461:513-524.
Nishizuka Y (1983) Protein kinase C as a possible receptor protein of tumor-proinoting phorbol esters. J. Biol.
Chem. 258:11442-6,
Nishizuka Y (1984) The role of protein kinase C in cell surface signal transduction and tumour promotion.
Nature London 308:693-698.
<>2 W. Ross Adey

Phillips JL (199}) L-fleets of EM field exposure on gene transcription.,/. Cell. Biochem. 51:381 -386.
Phillips JL. Haggreii \V. Thomas WT, Ishida-Jones T, Adey WR (1992) Magnetic tleld-induced changes in
specific gene transcription. Bioclum. Biophyi. Ada 1132:140-144.
Pitol HC. Dragan YP (1991) Facts and theories concerning the mechanisms of carcinogenesis. FASEB J.
5:22XO-!<.
Reiter RJ (1992) Alterations of the circadian melatonin rhythm by the electromagnetic spectrum; a study in
en\ ironmental to,\icology. Rcgululory Toxicology and Pharmacology 15:226-244.
Schumann WO tl9?7) Uber eleklrische Eigenschwindungen des Hohlraumes Erd-Luft-Ionosphare, erregt
durch Blitzentladungen. Zcits. Angcw. J. Phys. 9:373-378.
Semm P (1983) Neurobiological investigations on the magnetic sensitivity of the pineal gland in rodents and
pigeons. Comp. Biol. Physiol. 159:619-625.
Slaga TJ. Sivak .-\, Boutwell RK, eds. (1978) Mechanisms of Tumor Promotion and Carcinogenesis. Vol 2.
New York. Raven Press.
Steiner UH, Ulrich Th. (1989) .Magnetic field effects in chemical kinetics and related phenomena. Chem. Rev.
89:51-147.
Tabib A. Bachraeh U (1994) Activation of the proto-oncogene c-myc and c-fos by c-ra.v: involveinent of
polyamines. Biochem. Biophvs. /?cv. Communications 202:720-727.
faniarkin L. Han forth D. Lichter A. Demoss E. Cohen M.Chabner B, Lippmann M (1982) Decreased nocturnal
melatonin peak in patients with estrogen receptor positive breast cancer. Science 216:1003-1005.
lerada H. Kitagawa H, Okainoto N. Watanabe S. Taki M, Saito M (1994) An analysis of dose in tissue irradiated
b\ near Held of a circular loop antenna. lEICE Trans. Commun. E77-B:754-76l.
Tjandrawinata RR. Hawel L. Byus CV (1994) Regulation of putrescine e.xpon in lipopolysaccaride or
If N-i,'i(/H»ii(-activated murine monocytic-leukemic RAW 264 cells. J. Immunol. 152:3039-3052.
Wallcc/ek J (1992) Electromagnetic fteld effects on cells of the immune system: the role of calcium
signaling./-.-J5£fl./. 6:3 I 76-3185.
Walleczck J (1994) Immune cell interactions with extremely low frequency magnetic fields: experiinental
\enfication and free radical mechanisms. On the Nature of Electromagnetic Field Interactions nith
Biological Systems. AH Krey. cd. New York, RG Landes Company.
Wallec/ck J. Liburdy RP( 1990) Nonthermal 60 Hz sinusoidal magnetic field exposure enhances'^^Ca'* uptake
in rat thymocytes: dependence on mitogen activation. FEBS Let!. 271:157-160.
Walleczck J. Killoran PL. Adey WR (1994) .Acute 60 Hz magnetic field effects on Ca+ influx in human Jurkat
T-eclls: strict dependence on cell state. Bioelectromagnetics Society, 16th .-Xnnual Meeting, Copen-
hagen. Proceedings p 76 (abstract).
Weiustein IB (1988) The origins of human cancer: molecular mechanisms of carcinogenesis and their
implications for cancer prevention and treatment. Cancer Res. 48:4135-43.
Wilson BW, .Anderson LE, Hilton Dl, Phillips RD (1981) Chronic exposure to 60 Hz electric fields: effects on
pineal functions in the rat. Bioelectromagnetics 2:371-380.
Wilsim BW, .'\nderson LE! (1990) ELK electromagnetic-field effects on the pineal gland. E.xtremely Lou
Lrequcncy Electromagnetic Fields: the Question ol Cancer. BW Wilson, RCi Stevens, LE .Anderson
eds. Columbus, Ohio, Battelle Press, pp 159-186.
Yamasaki 11 (1987) The role of eell-to-cell communication in tumor promotion. Xongenoto.xic Mechanisms in
Carcinogenesis. TE Buttcrworth and TJ Slaga, eds. 25th Banbury Report. Cold Spring Harbor
Laboratory.
">airiasaki 11 (1991) .Aberrant expression and function of gapjunctions during carcinogene.sis- Environ. Health
Perspectives 93:]9\-\91.
Yellon SM (1994) Acute 60 Hz magnetic field exposure effects on the melatonin rhythm in the pineal gland
and circulation of the adult Djungarian hamster../. Pineal Res. 16:136-144.
PHYSICAL MECHANISMS FOR BIOLOGICAL
EFFECTS OF LOW FIELD INTENSITY ELF
MAGNETIC FIELDS

Charles Polk

University of Rhode Island

Kingston, RI 02881

This review consists of four parts. In Part One the experimental evidence for
biological effects due to low intensity, extremely low frequency (ELF) magnetic fields
is reviewed briefly. Part Two is a summary of reasonably well known signal-to-noise
calculations showing that several of the observed cellular effects seem to result from
"signals" that are below thermal noise. However some limits on the applicability of
"standard" thermal noise theory to living systems are pointed out. Part Three describes
theoretical approaches which assume that only the electric field, or electric current, can
act at the cellular level and imply that time varying magnetic fields arc ctTcctivc only
through the electric fields which they induce according to Faraday's law. Part Four, in
contrast, describes theoretical models - and some limited experimental evidence in their
support - that postulate "direct" action of the applied alternating magnetic field in
synergism with the ubiquitous static field.

THE EXPERIMENTAL EVIDENCE

Although there is evidence for biological effects at the tissue and cellular lexel of
very low intensity electric fields, they seem to be influenced by the presence of a static
magnetic field. Furthermore, many of the results showing effects on cells or organ cultures
by apparently extremely low intensity electric fields were obtained by inducing such fields
with time varying magnetic fields. As will be shown below, it is not clear at the present time
whether the time varying magnetic field alone or the induced electric field is responsible for
reported observations. For the discussion in this paper a "low intensity" magnetic field will
be defined as one not much above geomagnetic strength, or weaker, i.e. below about 100 ^T
(= 1 Gauss). Depending upon the experimental arrangement used, the induced electric field
due to a 100 |iT ELF magnetic flux density (B) can be extremely small. Since the induced,
circumferentially directed electric field in a homogeneous circular cylinder of radius a, with
its transverse cross-section perpendicular to the B vector (varying sinusoidally in time at
frequency f) is given by

Biological Effects of Magnetic and Electromagnetic Fields, Edited by S. Ueno

Plenum Press, New York. 1996 63
64 C. Polk

\E\ = nfBr r <a (1)

one obtains, for example, E^ax ^ 57 |iV/m when one applies a 100 jiT 60 Hz B-field to a
commonly used culture plate with 6 mm diameter wells. Several typical observations are
listed in Table 1. It only serves for illustration and is far from being a complete summary of
the existing experimental reports.

SIGNAL-TO-NOISE RATIOS.' MINIMUM ELECTRIC FIELD

REQUIRED TO EXCEED THERMAL NOISE
The signal-to-noise ratio problem is usually approached by applying to the cell
membrane the Nyquist expression for the thermal noise voltage or current in a resistor
(Weaver and .Aslumian, 1990). Thus

T';,=^4kTR8f (2)

7r, = 4kT(\/R)8f (3)

where k = Boltzman's constant, T = absolute temperature, R = resistance and 8/ = bandwidth.

Since the membrane is represented as a resistor in parallel with a capacitor, the steady state
voltage across both must be the same and the thermal energy of the resistor must be equal
to the energy stored in the capacitor

^cyi=^kT (4)

Table I. Reports on in vitro biological effects of time varying magnetic fields that
could induce only very low intensity electric fields

Induced E in B in nT
Author and effect Vm(peak) (peak)
P. Semm, R.C. Bcason (1^)0): Recordings from ophthalmic < (2) 10'" 0.2
ner\e of Bobolink, 0.5 Hz +DC'. 1 sec exposure [assuming I ms
rise time, (i>B''6t) = (2) UHj.
Goodman-Henderson (1991): HL 60 Cells. Increased level of 10^ S
Kfs.\ tran.scripis, 60 Hz, 20 mm. exposure.
D.B. Lyieetal. (1991): T-Lymphocytes. Ca-Uptake, 13.6 Hz <(0.8) 1 0 ' 20
+DC.
M.G. Yost, R.P. Liburdy (1992); Rat thymic Lymphocytes, (2.4) 1 0 ' 42
Inhibition of Ca-uptake, 16 Hz +DC. 60 min. exposure.
S. Mehla et al. (1993): Human Lymphocyte growth, 15 Hz. 60 <(l.5) lO-' 86
Hz, 0 DC. 24 hrs.
M. Kato et al. (1993): Melatonin level in rats. Rotating 50 Hz = 0) 10" = 1
B-fteld.
R.P. Liburdy et al. (1993): Modulation of oncoslatic action of ^(4.5) 10" ^ 1.2
melatonin on breast cancer cells, 60 Hz B-field, 0 DC.
Biological Effects of Low Field Intensity ELF Magnetic Fields 65

and one obtains

r/' AT j ^ 1 ^2
^" = ^ °' '^ = ^ = ^ (5)
where a, and e, are, respectively, the conductivity and dielectric permittivity of the mem-
brane. The amplitude of the applied electric field EQ that is necessary to produce a transmem-
brane potential equal to VF;; is then calculated. Assuming a perfectly homogeneous
spherical cell of inside radius a, with a perfectly homogeneous cell membrane of thickness
A, one obtains for the radially directed electric field inside the cell membrane

£, = ( D - ^ ) c o s e

a^ Y-CT

F--

b=a+ A (6)

where Y =CT2+j(»C2 is the generalized conductivity of the membrane (GJ = conductivity and
e, = £2r£o~ dielectric permittivity). The conductivities of the intercellular medium and of
the cytoplasm are assumed to be both equal to s. The colatitudinal angle q is measured
from an axis parallel to the applied electric field EO. When D « a and s » |Y| equ. (6)
reduces to

3a
E. = -— En cos 0
' 2A " (7)

A common procedure is to compare the peak value of the transmembrane voltage, f, A, given
by equ.(7), i.e. for q = 0, with the noise voltage given by equ. (2) or (5) to find the
value of the applied electric field EO, min that is necessary to give a transmembrane
potential just equal to V „ or 7„. However it is more appropriate to either average the
voltage (ErD) over the surface of the sphere for comparison with F;„ or to compare
the noise current given by equ. (3) with the signal current obtained by integrating the signal

current density YE, over one hemisphere. In either case one must evaluate j ' and the result is

3a jie (8)

This result should also be obtained by equating the energy contained in the electric field
within the membrane to the thermal noise energy

]^t]\\E{Qfdv = ]^kT

E(Qf = El + El (9)
66 C. Polk

where E,, is the average value of the field components in the membrane at r = b and r = a*
shown in Table 2. However a slightly different result (smaller by 13%) is the consequence
of using the approximation Fj, « E^ and using not entirely precise approximations for C or
R in evaluating T"', and T„. This alternate expression for E(, n„,i is

a- 3Ttr. (10)

Using equ. (10) for values of a = 10 fiin, A = 6 nm and £., ^ 6, one obtains

Eo,m„, '^ 2.3 Vim

while equating the peak value (for cos 0 = 1 ) of the transmembrane voltage with V„ given
by equ. (5) would give this value divided by A/3 or 1.3 V/m. Neglecting the non-uniform
distribution of the electric field along the cell surface and assuming long cells of length =
150 |.un and a = 25 |.mi. Weaver and Astumian (1990) obtain E„ „„„ * 8 (10'-) Vm. They also
point out that if the cell membrane somehow can decrease the bandwidth from that given by
equ. (5) (471 Hz when Oi = 10'' S/m and ^^2 = 6) to 10 Hz, the value of E,, „„„ would be
reduced by Mcv^Ti which would give (for the values of a, A and i:, selected above)

E„,„„„* 0.34 V/m

Furthermore, since the applied signal is coherent over time (t) while the thermal noise
is not, the signal-to-noisc ratio would improve by a factor \'//. However, using Weaver's
and Astumian's long cell with E,, „„„ = 0.08 V/m, a 20 minute exposure time (as in the
Goodman-Henderson experiments, 1991) would give EQ ^nn ~ 3 (lO"'') V/m which is still 30
times larger than the 10'^ V/m reported as effective in several experiments. Of course, if the
cell has a mechanism for frequency selection, 8/ might be as low as 1 H/. but that would

Table 2. lilcctrie fields inside and at the surface of thin spherical membrane for field
F,„ applied at r » a in a conducting medium

0 = coUililiidc measurod tVom direction ofl/,, vector

r » b E = r F.„ cos 0 - e E,, sin 0

I =b* E ^ r -, E|| — cos 0 - 0 ^ E„sinO

2 a.\
r=h f" ~ r 4 Fn ^ L-os e - e 1 E„ sin e

r = Li* E * r ^ [in " cos 0 - 0 - E,i —sin 0

r = ;i E . r ^ H„ ^ c o s 0 - 0 ^ E|) — sin 0
oA CTA

a ^ 10 *
\^' = (4) 10 ' a t 60 Hz
A * 6 CTA

Inner radius = a, outer radius = b. A = b - a, membrane conductivity = Oi. membrane dielectric

permittivity = t,. Y = 0^ + 1 m c,, conductivity of exterior (r > b) and interior (r < a) = a. r. 0 and
/ arc. respectively, unit vectors in the radial, colatitudinal and vertical directions. These tbiimilas
do not take into account the counter-ion layer present on biological cell incmhranes.
Biological Effects of Low Field Intensity ELF Magnetic Fields 67

still give E„ = 10""* V/m for the long cell which is not typical of the lymphocytes used in
several experiments. (B cells, for example, are nearly spherical with a = 10 \im).
It has also been suggested by Litovitz et al. (1994) that spatial coherence of the
applied field, as compared with spatially incoherent noise, may contribute to an improve-
ment of the signal-to-noise ratio (S/N). It is postulated that each of a large number of
receptors {Lauffenburger and Linderman, 1993) on a single cell "sees'" a different noise
field while being subjected to a spatially coherent applied field. Each of the affected
receptors would then send the same small chemical "signal" to the cell interior where all
the component signals would be added to produce at total response above noise. However
equations (2), (3), (9) or (10) take this already into account, at least partially, since the
noise energy (1/2) kT was evaluated for the entire cell membrane. We considered the
membrane as a single system and evaluated a total resistance or capacitance to compute
P;,. From equ. (9) it is apparent that if different parts of the membrane were to be
considered as separate, relatively isolated systems of smaller volume than the entire
membrane, the S/N ratio would decrease! The EQ required to overcome thermal noise
would apparently vary inversely as the square root of the system volume. If patches of
the cell membrane of area 6A were to be better described as separate systems (all of
thickness A), one would obtain for each patch

F' ~F ^
OA (11)

if the applied field Eo gave the same transmembrane voltage over the entire cell; but in view
ofthecosO in equ. (7) any independent patch at the equator would require an infinitely large
E,). Considering the fact that the cell membrane is a very poor electrical conductor and is
also very inhomogeneous, it may not be unreasonable to model it as several RC combinations
connected in parallel, each with its own noise source of energy kT''2, and giving a corre-
spondingly lower S/N ratio. These considerations do not imply that the mechanism of
"summation by multiple receptors" to improve the total S/TM ratio is impossible, but do show
that the S/N ratio at each small receptor would be much worse than calculated for the entire
membrane.
Since equ. (9) shows that the S/N improves if the volume of the system is increased,
it is also worthwhile to consider the outside of the cell surface, and in particular the thin
layer where counter-ion motion takes place under the influence of the component, EQ, of the
applied electric field that is tangential to the surface. For the spherical cell considered above,
values of the radial electric field E^ and the tangential electric field E,) are listed on Tabic 2.
Both inside the membrane and immediately adjacent to it on the outside (at r = b" or r = b*
in Table 2)

while the radial electric field immediately outside the membrane is

^ 3Ya ^ „
E,,,^ = El) cos 6
' 2CTA (13)

For typical values such as a = I S/m, |Y| * 10"' S/m, a = 10"'m. A *: 10 '^m, one obtains E^^,^,+
« E,|, therefore EQ can be used for E(9) in equ. (9). The result for the minimum tangential
electric field in a counter-ion layer of thickness d necessary to exceed thermal noise is
68 C. Polk

tan£„ ,^i„ , -L V-H: ,,,,

a 6ni:,i/ (14)

where Ci is the effective dielectric permittivity of the cell due to its counter-ion layer which
can be as high as lO'^nat ELF. Using this value and d = lO""* m. one obtains

k„,F-o,n„n* 5.1 V/m

which is of the same order of magnitude as the applied field required to generate a
transmembrane voltage larger than thermal noise. The appropriate value for d is not really
known and could be larger, but this would still not decrease unEo. mm ^'^O' much. All the
possibilities for improvement by frequency selectivity ( 6 / « l/RC) and integration over
time discussed above also apply here. Nevertheless, while the action of the tangential electric
field may be at least as important as the transmembrane potential in affecting, for example,
chemical receptors on the cell surface, or cell-to-cell communication (Polk, 1992), the S/N
ratio based on the assumption of a thermal noise limit is probably not much ditTerent in the
two cases.
Mathematical models may, or may not be good representations of physical reality. It
is important to realize that the cell membrane, as well as the cell interior and exterior
environment, is extremely inhomogeneous and therefore field expressions (7), (12) or (13)
are at best very crude approximations. Even more important is the assumption of a thermal
noise limit embodied in equs. (2) to (5). The (1/2) kT thermal energy per degree of freedom,
i.e. per variable necessary to describe the state of the system, requires that the "device"
I which may be a cell or cell membrane) does not have a steady input of power. Such an input
- which may be chemical in nature - could produce ordered ion motion that is not small in
comparison with the pre-existing random motion. Chemically initiated spatial and temporal
oscillations of calcium ion Hux, for example, have been observed (Berridge and (iaiionc.
19S,S; Jacob, 1990) and been modeled in terms of non-linear dynamics (Eichwald and Kaiser,
1993; Myer and Stryer, 1988). Applied oscillating fields could very well couple to such
oscillations. "As soon as one introduces an active device, such as a biased diode or a
transistor, the whole argument in terms of thermodynamic equilibrium collapses so that one
must examine the device in terms of internal mechanisms rather than as a black box" (Bell,
1985). This statement, written in relation to analysis of electronic devices, applies equally
to living cells. The thermal noise equations assume completely random motion of charges
and would not be valid if many or most charges are subject to organizing forces in some
preferred direction. l of this must be considered in the search for mechanisms that can be
responsible for biological etTects of electric and magnetic fields.

PROPOSED MECHANISMS FOR ELECTRIC FIELD COUPLING AT

THE CELLULAR LEVEL
Biological effects of electric fields are well known and at higher field intensity levels
- above thermal noise - they are also reasonably well, although not completely, understood.
I'he discussion here is concerned only with the "low level" extremely low frequency (ELK)
effects, tor which examples were given on Table 1, and which are identified as "subtle 'long
term" effects" in Table 3 that summarizes the field amplitudes responsible for different types
of biological action.
Reviews are available where some possible physical mechanisms for the production
of biological effects by low intensity electric fields are discussed in some detail (Cjrundler
et al.. 1992; Polk. 1992). Mechanisms that have been proposed are listed below in Table 4.
Biological Effects of Low Field Intensity EI.F Magnetic Fields 69

Table 3. Amplitudes required for ELF and pulsed electric field effects

V/m Inside
Tissue or Fluid
Electroporation Transient 10'^
Depends on E (Al) Permanent
Cell Rotation in Insulating Fluid 10''
Nerve/Muscle Stimulation initiation of firing 10'
alteration of firing rate 10
Subtle "Long Term" (t > 10 min) 1 0 ' to l O '
Effects (Bone soft tissue repair,
Ca-cfflux, transcription)

All of these mechanisms inust be based on interaction of electric fields E with charges
q (which will be mostly ions in the biological environment) or electric dipoles p. The
pertinent equations are

F = qE (15)

for the force on a single charge, and

F = (p-V)E (16)

T = pxE (17)

for, respectively, the force and torque on electric dipoles. As shown by equ. (16). translatory
motion of dipoles (as in "dielectrophoresis" - see Pohl, 1978) requires a field gradient, while
a uniform field is sufficient to set an electric charge into motion. Within a fluid medium the
motion of charges and dipoles at velocity v will be opposed by viscous forces given by
Stokes' law for linear motion when the (spherical) particle (of radius a) is large in comparison
with any "granularity" in the fluid

F = 67:anv (18)

where r| is the fluid viscosity. Possible attachment of charged particles to structures (e.g.
protein or nucleotide molecules) giving electric forces will quickly lead to non-linear
differential equations for the resulting motion. In addition, conservation of charge will

Table 4. Proposed physical mechanisms for interaction of low intensity ELF electric fields
with living organisms

Mechanisms requiring induced electric fields, or applicable to both electric and magnetic fields:
1. Improvement of S "N ratio for time varying magnetic field by interconnection through "passi\ e"
gapjunctions (Polk. 1W2; Pilla el al., I9Q1).
2. Effect of non-linear gap junctions (Galley, 1993).
3. Response of non-linear and "chaotic" .systems to small inputs (Grundler el al., 1992; Ruelle.
1991; Frohlich. 19X6; Baeriswyl el al„ 1987).
4. Effects due to electric polarization forces (K. McLeod, 1993a).
5. Interaction with endogenous Ca oscillations (Grundler et al., 1992; Walleczek, 1993).
6. Stochastic resonance (Kaiser 1993: Kruglikov and Dertinger 1994),
70 C. Polk

always have to be satisfied, expressed by the relation between current density and volume
charge density p

ot (19)

where the total current density J is not only due to the applied electric field, but will in
general also be due to non-uniform charge concentrations brought about by chemical
processes and leading to a diffusion current density Jj given by

Jd = - q D V C (20)

where C is the ion concentration per unit volume and D is the diffusion constant given by

D = ukT (21)

with u being the hydrodynamic mobility - (velocity/force) = (electric mobility/q).

A schematic diagram of the non-linear processes characteristic for many cells is
illustrated by Fig. 1 (Kaiser, 1994; Eichwald and Kaiser. 1993). In such non-linear systems
transitions occur between different stable states, e.g. from oscillation to no oscillation,
or from oscillation at one frequency to oscillation at a different frequency. In the presence
of noise the phenomenon of "stochastic resonance" may then become important
(McNamara and Wiesenfeld, 1989; Moss, 1991; Benzi et al., 1982) when an exogenous
periodic driving force, such as a weak ELF electric or magnetic field, is applied. If a
bistable or multistable nonlinear system is driven by random fluctuations ("noise") as
well as by a coherent signal, the output, i.e. the transition between states, involves
interaction between noise and input signal. As a consequence the ratio of the output signal
(which, in general, can contain different frequencies than the input signal) to the output
noise can become much larger than the input signal-to-noise (SNR). In fact, as illustrated
by Fig. 2, the output SNR for a fixed input signal first increases with increasing input
noise before it eventually decreases. As pointed out by Kaiser (1993), stochastic resonance
may very well provide a satisfactory explanation of observed biological effects in response
to weak applied electric or magnetic fields, and an experimental example has recently
been described (Douglass ct al., 1993). Since biological entities - cells as well as entire
organs or animals - represent extremely complex interconnected nonlinear systems, it is
likely that exact field-to-chemical reaction or field-to-biological process transduction
mechanisms will not be the same in different systems or under different circumstances.
Therefore several of the mechanisms which have been suggested in the past (Table 4)
and possibly others, may represent the first step in the weak coherent excitation needed
for stochastic resonance. However, the fundamental physical laws summarized by equ.
(15) to (21) will always be involved, albeit in their combination leading most likely

Figure 1. Interaction between extra-cellular signals, membrane

receptors and oscillators in the membrane and the cytoplasm
(illustration from Kaiser, 1994).
Biological Effects of Low Field Intensity ELF Magnetic Fields 71

Figure 2. Output signal-to-noise ratio as a func-

tion of input noise (for a 2 ring laser system, from 0 10 20 30 40 50 60 70
McNamara and Wiesenfeld, 1989). INPUT NOISE DENSITY ( mV/VTlz )

to non-linear differential equations with extreme sensitivity to initial or boundary con-

ditions.
Although involving time varying magnetic fields B, another set of circumstances
should be mentioned which would allow cells or protein molecules to differentiate between
an electric field that is induced by a time varying magnetic field according to Faraday's
law

IE dl = - ^— 1! B ds
(22)

and the electric fields of random orientation due to thermal noise or even larger biological
electric "noise" generated by nerve and muscle action. Since these noise fields are due
to relatively slow charge motion, they are effectively quasi-electrostatic fields E^ for
which

V X E. = 0 (23)

On the other hand for the electric fields induced according to Faraday's law the curl, or
circulation per unit area, is always non-zero

V X E= - —B
(24)

Since most protein molecules, and notably cell surface receptors, as well as DNA
and RNA molecules, are of spiral shapes, the latter electric fields (E) may be
particularly effective in influencing charges or dipoles located on such molecules.
Further consideration of the laws governing electric fields characterized by non-zero
curl in "chiral media" (Lakhtakia and Varadan, 1989) is likely to be useful. In any
case, the paths of currents in biological systems will be very different depending
upon whether the imposed ELF electric fields are due to a primary source that
generates an electric field or are induced by a time varying magnetic field (Polk,
1992; Stuchly, 1994).
72 C. Polk

PROPOSED MECHANISMS FOR ELF MAGNETIC FIELD

COUPLING AT THE CELLULAR LEVEL
Epidemiological findings that suggest the possibility of health effects as a result of
power frequency field exposure (for example, Feychting and Ahlbom, 1993) implicate weak
magnetic rather than electric fields. Several of the biological effects of ELF fields listed on
Table 1, as well as others, exhibit frequency "windows" - that is effects at some frequencies,
but not at others. In general the observed effects do not become stronger as the applied
frequency is increased (Liboff et al., 1983) which might be expected in view of cqu. (1) if
they were linearly related to a magnetically induced electric field. Equ. (1) follows from equ.
(22).
To test whether some cellular effects are due to magnetic or induced electric fields,
a few experiments were performed with exposure dishes that are divided by barriers into
several concentric rings (Misakian and Kaune, 1990). When subjected to an axially oriented,
uniform ELF magnetic field, cells in different rings are then exposed to the same magnetic,
but different electric fields given by equ. (1). Such experiments have shown that at least
some biological consequences of ELF magnetic field exposure were not due to the induced
electric field (Liburdy et al.. 1993; Blackman et al., 1994).
In this context it is useful to point out that the generation of an electric field by a tirne
varying ELF magnetic field is an extremely inefficient way of converting magnetic to electric
energy. If a magnetic fiux density B,„ within a cylindrical volume of radius a contains an
energy (1/2) [B,„'/|.tJ-[volume] and a magnetic flux density Be induces an electric field E so
that (1 2)(;E- integrated over the same volume is equal to the energy corresponding to B,„. it
is easily shown, using equation (1), that the ratio Be/B^ is given by

5.. 2
B„, an/VfX(,e (25)

where ^\.„ is the permeability of free .space. For a radius, a, of 3 cm one obtains al 60 Hz
values of 10** or 10'- depending upon whether one assumes for the dielectric permittivity, e,
that of free space (u,)) or that of muscle tissue at ELF (e a lO^e,)). If biological experimentation
shows that effects are produced by very weak time varying magnetic fields, it is therefore
clearly important to consider all possible mechanisms for direct magnetic field-tissue
interaction and not to assume a priori that biological effects can only be due to the induced
electric fields. Mechanisms proposed thus far are listed on Table 5.

Table 5. Mechanisms proposed for "direct" effects of ELF magnetic field effects
"Cyclotron resonance" (l.iboff, 19X5; Smith et al.. 1987). Also discussed by Durncy et
al.. 1988 and Halle. 1988.
"Paramagnetic Resonance" ot'Ca-ions in Ca-binding proteins (l.ednev, 1991, 199."!i or
other ions (Blanchard and Blackman, 1994), .Mso discussed by Adair, 1992 and
Engstrom, 1995.
Nuclear magnetic resonance (Blackman et al., 1989; Polk, 1992).
Magnetic field effects of free radicals (McLauchlan, 1989; Steiner and Ulrieh, 1989)
Also discussed by Grundlerctal., 1992; Polk, 1992 and Walleczek, 1995.
Interaction with endogenous ferrimagnctic particles (Kirschvink, 1992). Also discussed
by Adair, 199.^ and Polk. 1994.
Biological Effects of Low Field Intensity ELF Magnetic Fields 73

Cyclotron Resonance
Several experiments involving ELF magnetic field exposure of cells (Yost and
Liburdy. 1992; Blackman et al., 1994), marine diatoms (Smith et al., 1987) and rodents
(Thomas et al., 1986) showed a distinct frequency dependence which depended upon the
amplitude of a DC magnetic field that was applied simultaneously and in parallel orientation
with the AC field. Although the attempt to repeat the diatom experiment in another laboratory
was not successful (Parkinson, 1992), there is still a need to explain the other results which
did appear to give a DC dependent "resonance". Liboff (1985) and McLeod and Liboff
(1987) proposed cyclotron resonance in transmembrane channels. Cyclotron resonance of a
freely moving charge is due to the Lorentz force F exerted on a charge q which moves at a
velocity v through a DC magnetic field B(,.

F = qvxBo (26)

If a sinusoidally alternating magnetic field B| is applied in parallel with B,,, the electric field
induced according to Faraday's law will cause the charge to gain energy (from the electric
field) as it moves in circular orbits ofincreasing radius, provided the frequency of B| is given
by

f, = -^^«
2JI m (27)

or an integral multiple thereof with m being the mass of the ion. If the charged particle is
located within a viscous medium, ion-ion and ion-neutral collisions will prevent progression
to larger orbits and, as shown by Durney et al. (1988), resonance effects will not occur unless
the effective collision frequency is smaller than the frequency of the applied field. Since ions
in living organisms are almost always located within a fluid, viscous medium that gives high
collision frequencies, the simple cyclotron mechanism does not provide a satisfactory
explanation of the observed phenomena. In addition, ions in living organisms are likely to
be hydrated so that m in equ. (27) will have to include the mass of the attached water
molecules. Equ. (27) will then give different values depending upon the size of the hydration
sheath.

Ion Parametric Resonance

The "parametric resonance" model, as proposed by Lednev (1991, 1993), postulates
Zeeman splitting by a DC magnetic field Bo of infrared (~ 10'- Hz) vibrational modes of an
ion. The mode separation will be at the ion cyclotron frequency, given by equ. (27). and will
be in the power frequency range for several physiologically important ions when B,, is of
the order of magnitude of the geomagnetic field. DC amplitude-resonance frequency
combinations for some ions are shown on Table 6.
Following earlier work by Podgoretskii and Krustalev (1964), Lednev predicts that
application of an alternating magnetic field B|Cos[(Q|,/n) t] should modify the time average
p of the transition probability p from an excited vibrational state to the eroundstate as gi\en
by

p = K,+(-irK.J„(nB,/B,) (28)

The constants K, and Ki are related by K, = and Ki = which depend upon the original
vibration amplitudes and Q^- = 271/^-. J„ is the n'th order Bessel function and n is an integral
74 C. Polk

Table 6. Ion resonance frequencies for Bo = 50 jiT and required B,, for f ( =60
Hz
f. B„
Charuc Mass B(, = 50 MT i\. = 60 H /
Ion Ratio X 10' Hz MT
f:,.w 5.15 41.0 73.2
40(-..,:.
4.7S 38.0 78.9
Mg^* 7.89 62.8 47.8
p.U ').2S 73.8 40.6
H* 95.79 762.3 3.94

fraction of the resonance frequency fc. Thus the J^ term would indicate a dependence oi'p
upon the ratio of AC to DC field amplitudes, either at the frequency / ( (when n = 1) or at
its subharmonics (n > 1). Expression (28) is the time average of the transition probability
evaluated over a complete period of the applied field; the instaneous value of p is given by

p = A', + K. cos [Q,/ + {iiBiiBa) sin (Q,//;)/] (29)

The principal objections raised by Adair (1992) to the validity of the theory are the
requirement that the vibrating ion not be hydrated if m in equ. (27) is to be known, and that
the residence time of the ion in the excited vibrational state must be at least as long as the
period of the applied oscillating field. Lednev suggested that the calcium ion within the
calmodulin complex may be sufficiently shielded from the aqueous environment to guarantee
adequate lifetime. Within the calmodulin complex the Ca-ion would also not be hydrated.
Lednev t\irthermore assumed that the change in Ca*^"^ transition probability would somehow
affect the ability of the ion to participate in biologically important chemical reactions.
Recently Blackman et al. (1994) subjected "PC-12" cells to combinations of AC/DC
magnetic fields and found that the number of cultured cells with neurite outgrowths
decreased (in comparison with unexposed cells) as a function of the .AC/DC amplitude.
However that behavior was restricted to narrow frequency bands determined by the ampli-
tude of the applied DC magnetic field, suggesting a "resonance" behavior. They also were
able to tit their data to a Bessel function as indicated by Fig. 3. However they assumed a
dependence

p = A'l + A': (-1)" J„ (InBJBn) (30)

where the factor of (2) in the argument of the Bessel function accounts for a dependence on
the B| B|) ratio which differs substantially from that predicted by Ledncv's equ. (28) as
illustrated by Fig. 4. Blanchard and Blackman (1994) presented a modification of the original
Lednev theory which leads to equ. (30). It can be shown however (Polk and Wu, 1994;
Engstrom, 1994, 1995) that the factor of 2 is incompatible with the basic as.sumption of
Zeeman splitting. As indicated on Fig. 3 the data presented by Blackman et al. (1994) also
fit the J„(4B, B|,) function over the range of (BI/BQ) shown in their paper. Since the order n
of .1,, in equ. (28) must be a positive integer, the J(|(4B,,/B„) function likewise does not serve
to support the theory as expressed by equ. (28). However the .I„ result could conceivably be
a con.sequence of equ. (29) in the presence of collision damping or of several closely spaced
vibrational modes since the second term in that equation can be expressed as a series of
Bessel functions including J„(nB|/B„). Blackman et al. (1994) indicate by fairly elaborate
curve fitting to match their data with equ. (30) that the results support resonances or
"windows" corresponding to ions of vanadium, magnesium, manganese and hydrogen. In
view of the uncertainties involved, their conclusion appears to be premature. The authors
Biological Effects of Low Field Intensity ELF Magnetic Fields 75

120

Figure 3. Fit of neurite outgrowth data to J|(2s) and Jfi(4s) functions where s = (B, B,,). Based on experimental
data ( ) shown in Blacliinan et ai. (1994).

0.4- 1 i

0.3 ^ '../..>. -JXsh--^

1 «

0.2- >, \.... -,

\ / '

\V
\\
0.1 ' / ^ / \\ \ \
\ \ .
\ V
0: \ 'V
\ :\
-0.1
/ / V \ 7
">; r: \ \ /
/ * '
-0.2
M 1 / '' ' V ('X

' \ '
7
/
'^ " '
/
' \ ^ '

-0.3 ^ " * \ )

> \ / . /
-0.4
' \: ; / -
*

'1 \\ / ' \ '

/
/ /

-0.5 \ :\ /
^ '\ /
-0.6 1

s^B.IB.
Figure 4. Functions -J|(s) and -J|(2s) appearing in equations (28) and (30): Lednev vs. Blanchard-Blackman
dependence on the s = (B,/B(,) ratio at the first (principal) resonance.
76 C. Polk

recognize that '"additional work is needed to test the agreement between experimental results
and theoretical predictions over an extended range of B^,;^." and that "exposure conditions
should also consider single ion resonances, beginning with hydrogen at n = 1, to establish
what ions in key biological test systems respond consistently with the IPR model". In this
context a quantum mechanics based analysis of the model, including numerical calculations,
by Engstrom (1994, 1995) may help to resolve many remaining questions concerning the
validity of the model. The results obtained by Blackman et al. nevertheless confirm the
importance of considering in ELF exposure experiments the relative orientations, as well as
magnitudes of DC and AC magnetic fields, the applied AC frequency and the charge to mass
ratio of either ions or charge groups that are biologically important in the particular system
under investigation.

Nuclear Magnetic Resonance

Many, but not all, atomic nuclei rotate about their axis. The "spin" angular momentum
quantum nuclear I can have integral or half-integral values between -12.5 and +12.5 (Weast
et al., 1986). Since nuclei also carry an electric charge, rotation about their axis produces a
magnetic moment \x and in the presence of a static magnetic field BQ a torque

T = n X Bo (31)

will tend to orient the spin axis along preferred directions. Transitions between discrete
orientations correspond to changes in energy

^^ = V (32,
Transfer from one discrete energy level to the next higher one requires irradiation by a signal
of the nuclear precession frequency f^ such that

AW = f„h (33)

where h is Planck's constant. From equs. (32) and (33) it follows that

" /)/ (34)

In an earlier review (Polk, 1992) it was pointed out that NMR as a possible mechanism for
biological effects of low intensity alternating magnetic fields, in the presence of the earth's
static field, has both very attractive and very unattractive features. First, the characteristic
precession frequencies, f„, are below 1,000 Hz for many biologically important elements at
earth strength fields. Thus, with BQ = 50 |aT, f^ is equal to 827 Hz for ^Li, 653 Hz for -'Na.
and 99.3 Hz for "'K. Secondly, stability of the resonance lines (or "high Q" of the resonance
peaks), refieeted in the relatively long relaxation times, should make it possible to observe
resonance excitation by weak alternating magnetic fields of the appropriate frequencies.
l.Inattractivc features, which make NMR unlikely as an interaction mechanism are.
first, the small probability of spin alignment in a weak applied field. Thus the ratio R, of the
maximum energy of |.i in a field B,, to the thermal energy is

R = ^
kT (35)
Biological Effects of Low Field Intensity ELF Magnetic Fields 77

For the hydrogen nucleus, which is employed in medical diagnosis at fields of the order of
1 T, R = 3.3 (10-*^) while R = 3.8 (10'°) for 'Li at 100 ^T. Whether this difference by four
orders of magnitude makes biological effects of NMR at such levels of magnetic field
completely impossible is still debatable. Effects could only occur if the resonance phenome-
non would influence extremely critical processes - such as enzyme mediated reactions -
where the effect of a very minute change can be greatly amplified. It is also questionable
how a change in spin state could significantly influence chemical structure - although the
reverse is true, as indicated by the spin-lattice and spin-spin relaxation phenomena (Ando
and Webb, 1983) that make NMR useful for medical diagnosis. The spin-spin relaxation
phenomenon, in particular, is an indication that changes in the spin state of one nucleus do
affect its surroundings. However strong interaction with the nuclear environment, such as
by '"'N nuclei which have an electric quadruple moment (Haken and Wolf, 1984), would lead
to significant line broadening (Paudler, 1971) and thereby to the loss of observable frequency
sensitivity at ELF.
NMR has been suggested as an explanation of relatively low field intensity effects,
notably changes in dielectrophoretic yield and dielectric permittivity, that were observed at
a few kHz (Aarholt et al., 1988). NMR has also been mentioned as a possible explanation
of field effects on the efflux of radio-active calcium ions (''^Ca"^*) from chick brain tissues
(Blackman et al., 1988). The NMR phenomenon is attractive in the latter context, because
the experiments showed sensitivity to steady and alternating magnetic fields which were
mutually perpendicular, as required for NMR excitation. However, it should be noted that
the effect would only occur with radioactive Ca, and not with the non-radioactive a
isotope, which accounts for 97 percent of all Ca occurring in nature. ''"Ca has zero nuclear
spin and therefore no magnetic moment. (Some other biologically important elements with
zero spin are ''C, "'O, '^'Fe, -''Mg and ^^S.) One way of differentiating between an NMR
effect and any type of whole ion-resonance involving Ca, would be to repeat experiments,
which require a radioactive isotope, with both ""^Ca and . These isotopes have the same
spin (-7/2) and practically the same |i (-1.3172 and -1.316 in nuclear magneton units) and
therefore the same NMR frequencies f„. However the ionic masses are sufficiently different
(43/45) to give a 4.5 percent change in the cyclotron resonance frequencv f^ given by equ.
(27).

Magnetic Field Effects on Free Radical Reactions

Free radicals are important in many biological processes (Freifelder, 1982). For
example, free radicals are formed as intermediate products when light is incident upon the
visual pigment Rhodopsin. Chemical reactions that involve free radicals are strongly
influenced by static magnetic fields. This has been known for some time (Hoff et al.. 1977;
Blankenship et al., 1977; Werner et al., 1978). However, more recent work (Hamilton et al..
1988; McLauchlan, 1989) suggests that DC fields as low as 10 Gauss may affect such
chemical processes. By implication, as will be shown below, alternating fields of the same
order of magnitude, acting in concert with steady fields should also influence reaction yields.
Several observations in this area are based on Pauli's exclusion principle, which states
that the electronic states of an atom can only be occupied in such a way that no two electrons
have exactly the same set of quantum numbers. Thus if there are, for example, two valence
electrons in the same orbital, characterized by the same set of orbital quantum numbers, their
individual spin quantum numbers must be +1/2 and-1/2; i.e. their spins must be in opposite
directions. If two electrons in a chemical bond are paired in this manner and if this bond is
broken, for example, by incident light, resulting in two free radicals, subsequent recombi-
nation is only possible if the two electrons preserve this oppositely directed spin. Interaction
with the local magnetic field - due to nuclear magnetic moments or nearby other spinning
78 C. Polk

and orbiting electrons - can, depending upon details of the particular molecular structure,
either favor or destroy opposite spins. In the latter case, of now equally oriented spins,
recombination of the radicals becomes impossible. Conversely, the radical pair may have
been formed from excited atomic states with the unpaired electrons coming from different
atomic shells, In that case they may already have equal spin preventing chemical combination
of intermediate products.
As long as the electrons have opposite spin, the products have "singlet" character,
i.e. the total quantum number J. which characterizes the electron states, is equal to the orbital
quantum number L, since the spin quantum umber S = (1/2) - (1/2) = 0 and J = S + L.
However, as the products diffuse, some fraction will acquire "triplet" character (i.e. the
electron spins may become parallel; then S = + (1/2 + 1/2) = 1 and J can have 3 values, L
+ 1, L - 1 and L, in view of the quantum rules for combination of angular momenta. If the
products were initially in the triplet state, diffusion will have the opposite effect, i.e. partial
conversion from triplet to singlet character.
The singlet and triplet states have, in general, different energies, as indicated at B =
0 on Figure 5. Any magnetic field, including that of nearby magnetic nuclei, will cause triplet
states, T+i and T.,. which have electron spin in the direction of the field, to gain or lose
energy. Therefore the energy levels of these states will separate with increasing B, as
indicated on Figure 5. Interconversion between the singlet and triplet states can occur either
by external energy input, or between T„ and S through a distance dependent "electron
exchange interaction" at any level of B. However, at some critical level of B = B^
Interconversion between T, and S is also possible without external energy input. Hamilton
et al. (1988) have shown experimentally that this level is approximately 1 mT in their
pyrene-dimethylaniline system.
Since interconversion between singlet and triplet states in the direction of greater
singlet product will make possible chemical combination, application of the correct value
B,. will obviously affect the rate of chemical reaction. However it is possible that an ambient
tlux density (from external and internal sources) may have a value B|, which is either slightly
larger or stnaller than the required B,. In that case addition of an alternating field B.^cos wt

MAGN. FLUX
DENSITY

Figures. Effectof magnetic field on radical pair energy levels (from

Bioelectromagnetics S-l, 1992, p,227).
TIME
Biological Effects of Low Field Intensity ELF Magnetic Fields 79

(with a period 2n/w longer than necessaiy for T., —> S) will periodically establish optimum
conditions for conversion. On the basis of theoretical and experimental results (Hamilton et
al., 1988) it is therefore at least possible that combination of DC and AC magnetic fields of
the order of 0.5 mT (=1/2 of the measured B,.) could affect chemical reactions in biological
systems that involve free radicals as intermediate products. However there appears to be no
experimental evidence that alternating fields of a few jiT would be effective. A more detailed
review of this topic has recently been prepared by Walleczek (1995).

Interaction with Endogenous Ferrimagnetic Particles

Kirschvink et al. (1992) have demonstrated that magnetite biomineralization occurs
in the human brain. Kirschvink and Kirschvink (1995) have also found that one line of human
leukemic white blood cells is nearly 100 times more magnetic than brain tissues giving
cellular magnetic moments of ~ 10'^ Am^. Kirschvink (1992) then suggested that the
interaction energy of even weak ELF magnetic fields with magnetic particles of magnetic
domain size would be larger than the thermal energy of (1 /2) kT per degree of freedom within
a system in thermal equilibrium. Using a model of magnetic particles floating in a fluid
medium, Adair (1992) proposed "that a 60 Hz magnetic field weaker than 5 pT cannot
generate significant biological effects at the cell level through action on magnetic elements".
However such a conclusion seems to be at least premature, because not enough is known
about the relation of the magnetite particles to other components of the cell interior, or of
the cell membrane, to formulate a mathematical model that is adequate for precise numerical
predictions.
Any kind of anchoring of the "magnetosomes" to the cell structure would require
description of the dynamics by a non-linear differential equation and even a linearized
equation would not be one with constant coefficients unless the applied alternating magnetic
field is exactly perpendicular to the geomagnetic field and the magnetosome deflection is
very small. In addition, the effective viscosity of the fluid within the cell interior (in the
vicinity of the magnetosome), which is a critical parameter in the Adair model, cannot be
specified very well. But even if this clearly oversimplified model is used, but combined
action by multiple magnetosomes within the cell is allowed, it appears that a signal-io-noise
ratio well above one might be possible with a 60 Hz magnetic field of 2 |iT (Polk, 1994).
Neither estimate can be considered adequate at the present time. However it is likely that
any effect of multiple magnetosomes within a cell would make use of their coherent motion,
thus there would be no significant thermal noise limitation if the energy supplied by an
applied field to a single magnetosome would be much smaller than kT. For single domain
particles of Fe304 the magnetic moment is given by (Frankel, 1986)

H = (volume)(4.8)10^ Am- (36)

and the energy supplying torque to a spherical particle of 60 nm diameter in a 2 pT field

would already be 0.1 kT at 37"C. Therefore it is important to perform further experimental
work on the behavior of such particles within biological cells, or of similar particles that
might be present in cell culture medium as suggested by Kirschvink (private communication,
1994).

CONCLUSIONS
Biological effects of high intensity ELF electric fields, such as muscle stimulation
and - at still higher intensities - cell damage by electroporation and heating, are reasonably
80 C. Polk

well (although not completely) understood. However power frequency effects of magneti-
cally induced electric fields of less than 10 "' V/m or alternating magnetic fields of 1 ess than
about 100 fiT cannot be explained very well at the present time. The biological processes
involved, notably signal transduction at the cell membrane and subsequent biochemical
processes involving "second messengers" are beginning to be identified (Adcy, 1990; Luben,
1993). These processes involve vast amplification of the incident "signal" with metabolically
supplied energy. The physical mechanisms of conversion from an electric or magnetic field
to biochemical process are however unknown at the present time. As a consequence we do
not know which aspect of the applied or environment field is important. Properties such as
polarization, duration and/or intermittency of the signal, frequency and harmonic content
may turn out to be as important as amplitude. The biological, chemical and mechanical
condition of the organism (e.g. stage in cell cycle, stimulation by a mitogen or cell density)
will also influence the effectiveness of the applied field. A key question, which is still only
resolved for a few experimentally observed effects, is whether specific biological results arc
caused directly by the magnetic field or by the electric field that is induced in tissue by a
time varying magnetic field. Particularly for fields larger than about 100 \iJ direct "magne-
tochemical" effects have been suggested and the presence of magnetite particles in some
tissues may also play a biological role. Experimental findings suggest that effects of induced
electric fields are particularly likely when their magnitudes at the tissue or cell level are
relatively "large", of the order of 10' V/m or larger, (McLeod et al., 1993b; Liburdy, 1992;
Lyle et al., 1988), and when multiple cells are connected by gap junctions. Induced electric
fields may also affect the motion of "counterions" on the cell surface and possibly the
structural fluctuations of nucleic acid molecules. While most proteins have a small net
electrostatic charge, nucleic acids are polyelectrolytes with large net charge and are sur-
rounded by counterions (McCammon and Harvey, 1987). However in view of the ineffi-
ciency of electric field induction by an ELF magnetic field, illustrated by equation (25). the
mechanisms for "direct" interaction of time varying magnetic fields with biological proc-
esses need serious consideration. This is particularly necessary when the applied ELF
magnetic fields that produce unambiguous biological effects are of such amplitude and
orientation (in relation to the culture medium or animal) as to produce very small induced
electric fields. Experimental results have suggested that the relative magnitude and direction
of a static magnetic field (such as that of the earth) can determine the biological effectiveness
of a simultaneously present alternating field in some biological systems under controlled
laboratory conditions.

REFERENCES
Aarholt t. Jaberansari M, Jafari-.AsI A H. Marsh P N, Smith C W (1988): N.MR Contlition and Biological
Systems. In "Modern Bioelectritity" (Marino A A, ed.). New York and Basel; Marcel Dekker Inc.
pp.75-104.
-Adair R K (1992); Criticism of Lednev's inechanism for the influence of weak magnetic fields on biological
systems. BioL'k'ctromiignctics 1,3: 2-31-2.35.
.\dair R K (1993): Effects of ELF Magnetic Fields on Biological Magnetite. Biin-lcclromupiflics 14-1. 1-4.
.'Xdey W R (1990): Electromagnetic fields and the essence of living systems. In "Modern Radio Science 1990"
(Bach J. ed.), O.xford UK: O.xford University Press,
Ando 1. Webb G A (1983): "Theory of NMR Parameters." London, New York: Academic Press, pp. X3-l()5.
Baeriswyi D. Droz D. Malaspinas A, Martolini P(eds)( 1987); "Lecture Notes in Physics 284, Physics in Living
matter." Berlin Heidelberg, New York. London. Paris, Tokyo: Springer-Verlag.
Bell D A (1985): "Noise and the Solid State." New York: A Halsted Press Book, John Wiley and S<ms,
Bcrridge M J and Galione A (1988): Cytosolic calcium oscillators (1988); The FASEB .11 2: 3074-3082.
Biological Effects of Low Field Intensity ELF Magnetic Fields 81

Blackman C F. Benane S G, Elliott D J, House D E, Pollock M M (1988): Influence of Electromagnetic Fields

on the EtTlux of Calcium Ions from Brain Tissue in Vitro: A Three-Model Analysis Consistent with
the Frequency Response up to 510 Hz. Bioelectromagnetics 9: 215-227.
Blaekman C F. Kinney L S, House D E, Joines W T (1989): Multiple Power-Density Windows and Their
Possible Origin. B/oe/ec/romagH<?//c.? 10: 115-128.
Blackman C F, Blanchard J P, Benane S G and House D E (1994): Empirical Test of an Ion Parametric
Resonance Model for Magnetic Field Interactions with PC-12 Cells. Bioelectromagnetics 15: 239-
260.
Blanchard J P and Blackman C F(1994): Clarificationand Application of an Ion Parametric Resonance Model
for Magnetic Field Interaction with Biological Systems. Bioeleetromagtu'tics 15: 217-238.
Blankcnship R E, Schaafsma T J, Parson W W (1977): Magnetic field effects on radical pair intermediates in
bacterial photosynthesis. Biochimica et Biophysica Acta 461: 297-305.
Douglass J K, Wilkens L, Pantazelov E and Moss F (1993): Noise enhancement of information transfer in cray
fish mechanoreceplors by stochastic resonance. Nature 365: 337-340.
Durney C H, Rushforth C K, Anderson A A (1988): Resonant dc-ac magnetic fields: calculated response.
Bioelectromagnetics 9(4): 315-336.
Eichwald C and Kaiser F (1993): Model for Receptor-Controlled Cylosolic Calcium Oscillations and for
External Influences on the Signal Pathway. Biophys. Jl. 65: (Nov.) 2047-2058.
Engstrom S (1994): Dynamical properties of Lednev's parameteric resonance mechanism. .Annual Review of
Research on Biological Effects of Electric and Magnetic Fields from Generation, Delivery and Use
of Electricity. DOE, EPRl. EPA, NYDPS, NIEHS. Nov. 6-10, Alberquerque N.M. Available from \V L
.Associates, Frederick, MD 21701. Paper A-21. Also in press (1985): Bioelectromagnetics 16.
Feychting .M and Ahlbom A(1993): Magnetic fields and cancer in children residing near Swedish high-power
voltage lines. Am../. Epidem. 138: 467-481.
Frankel R B (1986): Biological Effects of Static Magnetic Fields. In Polk C and Poslow E (eds.) -'CRC
Handbook of Biological EtTects of Electromagnetic Fields". Boca Raton. EL: CRC Press. Inc. pp.180.
181.
Freifelder D (1982): ""Principles of Physical Chemistry with Applications to the Biological Sciences." 2nd ed.
Boston: Jones and Barlett Publications Inc., p.379.
Frohlich H (1986): Coherent Excitation in Active Biological Systems. In Gutman H, Keyser H (eds) "Modern
Biochemistry." London: Plenum Press.
Gailey P C (1993): A Biophysical Model for the Effects of ELF Fields on Cellular Growth Control. BEMS
Abstracts, 15th Annual Meeting, 1970, pp.70, 71.
Goodman R, Henderson A S (1991): Transcription and Translation in Cells Exposed lo Extremely Low
Frequency Electromagnetic Fields. Bioeleclrochemislry and Bioenergclics 25: 335-355.
Grundler W, Kaiser F, Keilmann F, Walleczek J (1991): Mechanisms of Electromagnetic Interaction with
Cellular Systems. Naturwissenschaffen 79: 551-559.
Haken H, Wolf H C (1984): "Alomic and Quantum Physics". Berlin, Heidelberg, New York. Toronto:
Springer-Vcrlag. pp. 202-208.
Halle B (1988): On the cyclotron resonance mechanism for magnetic field effects on transmembrane conduc-
tivity. Bioelectromagnetics 9: 381-385.
Hamilton C A. Hewitt J P, McLauchlan K E, Steiner U E (1988): High resolution studies of the effects of
magnetic fields on chemical reactions. Molecular Physics 65: 423-438.
Hoff A J, Rademaker H, Van Grondelle R, Duysens LNM (1977): On the magnetic field dependence of the
yield of the triplet state in reaction centers of photosynthetic bacteria. Biochimica et Biophysica Acta
460: 547-554.
Jacob R (1990): Calcium oscillations in electrically non-excitable cells. Biochimica et Biophysica Acta 1052:
427-438.
Kaiser F (1993): Explanation of Biological EtTects of Low-Intensity Electric, Magnetic and Electromagnetic
Fields by Nonlinear Dynamics. Ninth Annual Review of Progress in Applied Computational Electro-
magnetics. Monterey, CA (March 22-26).
Kaiser F (1994): Nonlinear Dynamics and Biophysical Systems: Interaction of Static and Periodic Fields with
Nonlinear Periodic Processes. Kleinheubacher Berichte 37 (German URSl Meeting. Oct. 4-S, 1993).
Kato M, Honma K-i, Shigemitsu T. Shiga Y (1993): Effects of Exposure to Circularly Polarized 50 Hz Magnetic
Field on Plasma and Pineal Melatonin Levels in Rats. Bioelectromagnetics 14: 97-106.
Kirschvink J L (1992): Comment on Constraints on biological effects of weak extremely-low frequency
electromagnetic fields. Physical Review A, 46, no. 4 (15 August), 2178-2184.
Kirschvink J L, Kobayashi-Kirschvink A. Woodford B J (1992): Magnetic Biomineralization in the Human
Brain. Proc. Natl. Acad. Sci. USA. (Biophysics). 89 (86), 7683-7687.
82 C. Polk

Kirschvink J L, Kobayashi-Kirschvink A { 14951: Magnetic Biomineralization in Human Pathological Tissues,

Cliapt. in this book.
Kruglikos I L, Dcrtnigcr H (l9'-)4): Stochastic Resonance as a Possible Mechanism of Amplitlcation of Weak
HIcctric Signals in L i \ i n g Cells. Bii>clt'c!ronicignelic.s\5: 539-5'47.
I.akhtakia A. Varadan V K, Varadan V V ( I 9 8 9 | : "Time-Hannonic Electromagnetic Fields in Chiral Media."
(Lecture Notes in Physics no. .^.15). Berhn. Heidelberg,Nev York. London, Paris, Tokyo. Hong Kong:
Springer-Vcrlag.
Lauffenburgcr D .\ and Linderman J J (1993): "Receptors, Models for Binding, Trafficking and Signaling".
New York, Oxford: Oxford University Press.
Ledncs \ V 11991): Possible mechanism for influence of weak magnetic fields on biosystems. Rii>elt'iiyi>nuii;-
nelics 12: 71-75.
Lcdnc\ V V (1993): Possible mechanism for the effect o f weak magnetic fields on biological systems:
Correction of the basic expression and its consequences. In Blank M (ed.): "Electricity and Magnetism
in Biology and Medicine." San Francisco, San Francisco Press, pp. 550-552.
LibofT.A R. Williams T. Strong D M, Wistar R i r (1984): Time-varying magnetic fields: effects on D N . \
synthesis. Science 223: SI8-820.
Liboff A R (1985): Cyclotron resonance in membrane transport. In C'hiabrera, Nicolini, Schwan (eds.):
"Interaction between Electromagnetic Fields and Cells." London: Plenum, pp. 281-296.
l i b o f f R .-Xand McLeod B R (1988): Kinetics of Channelized Membrane Ions in Magnetic Fields. Hioc/ccim-
nuifiiielic.'t 9: 39-51.
Liburdy K P( 1992): Calcium signal in lymphocytes and ELF fields: Evidence for an electric field metric and
a site of interaction i n \ o h ing the calcium ion channel. FEBS L.cucis 301: 53-59.
Liburdy R P. Sloma T R. Sokolic R. Yaswen P (1993): ELF magnetic fields, breast cancer and melatonin: 60
11/ fields block melatonin's oncostalic action on ER* breast cancer cell proliferation. ,/. Pineal Rc\.
14: 89-97.
Litovit? T A. Montrose C J. Doinov P. Brown K M, Barber M (1994): Superimposing spatially coherent
electromagnetic noise inhibits field-induced abnomialities in developing chick embrios. Hu>elei iri>-
nniiinelic. 15: 105-113.
L.uben R A (1993): Effects o f low-energy electromagnetic fields on signal transduction by Cj protein-linked
receptors. In Blank M (ed.) "Electricity and Magnetism in Biology and Medicine."
l.yle D B, .Xyotte R D. Sheppard A R. .Vdey R \V (1988): Suppression of T-lymphocyte Cytotoxicity Following
Exposure to 60 Hz Sinusoidal Electric Fields. BioeU'Clromugneiics 9: 303-313.
l y l e 1) B. Wang X. Ayotte R D, Shoppard A R. Adey W R (1991): Calcium Uptake by Leukemic and Normal
T-1-yinphocytcs Exposed to Lou Frequency Magnetic Fields. Biocleclromagnelics 12: 145-156.
McCammon J A. Harvey S C (1987): "Dvnamics o f proteins and nucleic acids". Cambridge, London, New
York: Cambridge University Press,
McLauchlan K .'\ (1989): Magnetokinetics. mechanistics and synthesis. Chemi.sln in Britain, Sept. 895-898.
Mel.cod B R. Rubin C I. Donahue H J, Guilak F (1993a): Role of Polarization Forces in Mediating the
Interaction of tow-Frequency Electric Fields with Living Tissue. In Blank M (ed.) "Electricity and
Magnetism in Biology and Medicine." San Francisco: San Francisco Press. Inc. pp. 559-562.
McLeod K J. Donahue H J. Fontaine M-A. Rubin C T (1993b): Cell Density Effects on the Interaction of
l.lectric Fields with Cells in vitro. In Blank M (ed.): "Electricity and Magnetism in Biology and
Medicine." San Francisco: San Francisco Press. 291-294.
McNamara B, Wiesenfeld K (1989): Theory of stochastic resonance. Phy.s. Rev. A 39(9): 4854-4869.
Mehta S. Blackinton D. Wiesendanger M. Wanebo H, Lefebvre M, Cherlin D. Polk C (1993): Accelerated
Human Lymphocvte C/rovvlh in a I CJ, 15 Hz Magnetic Field. In Blank M (ed.) "Electricity and
Magnetism in Biology and Medicine." San Francisco: San Francisco Press. Inc. pp. 637-639.
Misakian M, Kaune W T (1990): Optimal experimental design for in vitm studies with ELF magnetic fields.
BiDclccliomafinclics 11: 251-255.
Moss F (19911: Stochastic Resonance. Ber. Bunsenges. Phys. Chem. 95(3): 303-311.
Myer I, Styer L (1991): Calcium Spiking. Ann. Rev. Biophys. Biophys. Chem. 20: 153-174.
Parkinson W C. Sulik CI L (1992): Diatom Response to Extremely Low-Frequency Magnetic Fields. Radiation
Research 130:319-330.
Paudler W W (1971): "Nuclear Magnetic Resonance." Boston: A l l y n and Bacon Inc. pp. 1-15.81.
Pilla .'\ A. Nasser P R, Kaufman J J (1991): On the Sensitivity o f Biological Cells and Tissues to Weak
Environmental and Therapeutic Electromagnetic Fields. Proc. Ann. Int. Conf. IEEE Enginecrini; in
Mcc/icine and Biology Society. Part 2 5: 1021 -1023.
Podgoretskii and Khaistalev O A (1964): Interference effects in quantum transitions. Soviet Physics Uspekhi
6(5): 682-700.
Biological Effects of Low Field Intensity ELF Magnetic Fields 83

Pohl H A (1978): "Dielectroplioresis." Cambridge, London. New York. Cambridge L'nivcrsity Press.
Polk C (1992): Dosimetry of Extremely-Low-Frequency Magnetic Fields. Bioeteclmma^neiia l.i, suppl. 1:
209-235.
Polk C (1994): Effects of e.\lremely-low-frequency magnetic fields on biological magnetite. Bioclcclromag-
netics 15: 261-270.
Polk C, Wu S-H (1994): AC DC Magnetic Field Synergism: Comments on the Lednev and "IPR" Models.
Annual Review of Research on Biological Effects of Electric and Magnetic Fields from Generation.
Delivery and Use of Electricity. DOE, EPRI, EPA, NYDPS, MEHS. Nov. 6-10. Alberquerque N.M.
Available from W/L Associates, Frederick, MD 21701. Paper .^-20.
Ruelle D (1991): "Chance and Chaos." Princeton: Princeton University Press.
Semm P. Beason R C (1990): Responses to Small Magnetic Variations by Trigeminal System of the Bobolink.
Brain Research Bulletin 25. 735-740.
Smith S D. McLeod B R, Liboff \ R. Cooksey K (1987): Calcium Cyclotron Resonance and Diatom Mobility.
BioL'Icclromagnclus 8: 215-227.
Stcincr U E, Ulrich T (1989): Magnetic Field Effects in Chemical Kinetics and Related Phenomena Clwniical
Rev. 89: 51-147.
Stuchly M A, Xi W (1994): Modeling induced currents in biological cells exposed to low -frequency magnetic
' fields. Phys. Med. Biol. 39:''l319-1330.
Thomas J R, Schrot J. Liboff .A R (1986): Low-intensity magnetic fields alter operant behavior in rats.
Bioeleclmmagneties 7: 349-357.
Walleczek J (1993): Electromagnetic Interactions with Human Lyinphocyte Calcium Regulation: Studies by
Differential Real-Time Fluorcsence Spectroscopy. BEMS Abstracts. 15th Annual Meeting, p. 14.
Walleczek J (1995): Bio-Magnetochemistry: A Possible Paradigm for Understanding sub-kT Magnetic Field
Interaction with Biological Systems. In Blank M (ed.): "Electromagnetic Fields: Biological Interac-
tions and Mechanisms." Washington, DC: ACS Books.
Weast R C. Astle M J. Beyer W H, eds. (1986): "CRC Handbook of Chemistry and Physics." 67th edition.
Boca Raton. FL: CRC Press Inc. pp. B-219 to B-439.
Weaver J C, Astumian D R (1990): The Response of Living Cells to Very Weak Electric Fields: The Thermal
Noise Litnit. Science 247 (26 January): 459-462.
Werner H J, Schulten K., Wellcr A (1978): Electron transfer and spin exchange contributing to the magnetic
field dependence of the primary photochemical reaction of bacterial photosynthesis. Biochirnica et
Biophysica Acta 502: 255-268.
Yost M G, Liburdy R P (1992): Time-varying and static magnetic fields act in coinbination to alter calcium
signal transduction in the lyinphocyte. FEES 296: 117-122.
ELECTROMAGNETIC FIELDS AND
CELLULAR SYSTEMS
Signal Transduction, Cell Growth and Proliferation

Robert P. Liburdy

Life Sciences Division

Lawrence Berkeley Laboratory
University of California
Berkeley, CA 94720

INTRODUCTION
This paper reviews some of the in vitro research conducted in our laboratory to
identify cellular responses to ELF fields. Studies we have conducted support an interaction
site involving ligand-receptor events such as antibody binding to its cell surface receptor
since we have observed changes in receptor-initiated calcium influx in cells exposed to
magnetic fields. These changes suggest that signal transduction (ST) which is initiated by
ligand-receptor binding plays a role in this interaction. Others have also reported that ELF
fields influence enzyme activation, gene expression, protein synthesis and cell proliferation,
all of which are triggered by earlier ST events at the cell membrane.
The concept of ELF fields altering early ST cell membrane events and thereby
influencing intracellular cell function via the ST cascade is a plausible biological framework
currently being investigated for understanding ELF effects on cells. For example, the
consequence of an increase due to ELF fields in mitogenesis, the final endpoint of the ST
cascade, is an overall increase in the probability of mutagenesis and, consequently, cancer
according to the Ames epigenetic model of carcinogenesis. Consistent with this epigenetic
mechanism and the ST pathway to carcinogenesis is recent evidence that ELF fields can alter
breast cancer cell proliferation and can act as a co-promoter in vitro.
In addition to investigating the biological nature of ELF field interactions, we have
conducted biophysical studies to determine whether the electric (E) or the magnetic (B) field,
or if combinations of static B and time-varying B fields represent the exposure metric for
the cell. This dosimetry question relates directly to understanding fundamental interaction
mechanisms and to the development of a rationale for ELF dose-threshold guidelines. An E
field-mediated interaction has interesting consequences for microdosimetry at the cellular
level and is mechanistically consistent with an interaction at the cell surface since the E field
does not penetrate the cell membrane. Recently studies from our laboratory and others have
Biological Effects of Magnetic and Electromagnetic Fields. Edited by S. Ueno
Plenum Press, New York, 1996 85
86 R. P. l.ihurdy

Signal Transduction
Interaction Model
ELF
FIELDS I

Cell Membrane ! Signal Transduction:

" T " ^ " ^ ^ " ^ Calcium Signaling
ST Pathways Receptor Binding

Cell Nucleus | ^ B i o l o g i c a l Consequence: F i g u r e 1. Sequence of m o l e c u l a r e v e n t s t r i g g e r e d b>

_ „ _ ,?. ,^,, E L F field i n l e r a c l i o n w i t h the c e l l m e m b r a n e .
Ceil Proliferation

suggested that an ELF B field by itself or in combination with a static B field may also elicit
cellular effects. Thus, in addition to E-field mediated effects, other interaction mechanisms
as yet not fully understood may operate at the cellular level. In addition to the question of
an exposure field inetric, the biological state of the target cell is important in ELF interac-
tions. Biological factors such as cell type, cell cycle, cell activation, age of donor animal,
passage number of cell line, presence of specific growth/mitogenic factors, temperature,
shape and cell density/packing during exposures have been shown to play a role in mediating
HLF interactions with cells.

FIELD COUPLING AND THE SIGNAL TRANSDUCTION CASCADE

The ST cascade is an important, plausible, common theme in cellular interaction
studies and represents the most plausible biological framework for understanding cellular
responses to LLF fields. This interaction model is shown in Figure 1 in schematic form. ELF
fields influence the ST cascade at the level of the cell membrane and trigger changes in
calcium influx and/or receptor binding. Subsequent events distal to ligand-receptor binding
such as gene expression and protein synthesis which lead to cell proliferation are ultimately
influenced as the initial changes in calcium signaling are propagated down the ST cascade.
Several recent articles dealing with ELF interactions with cellular systems and signal
transduction are recommended for additional reading.''^

THE CELL MEMBRANE AS AN INITIAL FIELD COUPLING SITE

Signal transduction involves cell surface interactions and it is the basis for cell
communication with its environment. ST is a general process in which a ligand molecule
binds to its receptor site on the cell surface and triggers a cascade of biochemical events in
the cell membrane that lead to enzyme activation, gene induction, protein synthesis and
ultimately mitogenesis and cell proliferation.' '" Signal transduction is required for cell
function, growth and differentiation. Two major signaling systems involve receptor tyrosyl
kinases and receptor phospholipid breakdown in the cell membrane. The insulin receptor
protein, for example, is a tyrosyl kinase that is activated by insulin binding. In the second
systein, which is widely distributed and found in all cells that are calcium dependent, a ligand
binds to its receptor and triggers phospholipid breakdown in the cell membrane leading to
the generation of "second messengers" that control a myriad of metabolic, cell growth, and
differentiation events. Many different receptors share this ST pathway for economy. For
example, when Concanavalin A (Con-A) or other mitogen binds to the T-cell receptor (TCR)
of T-lymphocytes it triggers phospholipid breakdown leading to inositol(l,4,5)trisphosphate
(IP,) which in turn elevates intracellular calcium. Importantly, this elevation of calcium ion
Electromagnetic Fields and Cellular Systems 87

Table 1. Mitogen-Activated signal transduction in the

T-lymphocyte
MITOGEN-ACTIVATED SIGNAL TRANSDUCTION
IN THE T-LYMPHOCYTE
RESPONSE TO CONCANAVALIN A FIRST PETECTEP
[034-2]!, Calcium Influx
ipH], 0 - 5 Minutes
IP3Increase
PKC Activation

c-FOS mRNA Increase

c-MYC mRNA Increase
Stimulation of Glycolysis 5 - 50 Minutes
Increase Metabolite Uptalte (e.g. Uridine)
Increased Inositol Incorporation Into IPs

Increased General Protein Synthesis

Increased General RNA Synthesis > 300 Minutes
Increased General DNA Synthesis

concentration is a second messenger event; calcium ions bind to proteins such as calmodulin
and kinases which sustain the ST cascade within the cell that ultimately leads to DNA, RNA.
and protein synthesis, cell proliferation and clonal expansion of the T-cell.
In our laboratory calcium influx has been followed in two ways: using calcium-45
to follow influx and using FURA-2 to monitor in real-time changes in free intracellular
calcium in cells. Extracellular calcium enters the lymphocyte through a channel that is open
when Con-A binds to the T-cell receptor on the cell surface. The first event is Con-A receptor
binding to the extracellular domain of the TCR which activates a tyrosine kinase in the
cytoplasmic domain. This occurs instantaneously and it represents the first coupling event
in the ST cascade. Tyrosine phosphorylation of the G-protein complex leads to activation of
phospholipase C which cleaves phosphoinositol in the cell membrane to yield IP-, and
diacylglycerol (DAG). IP^ opens the calcium channel and also acts as an endogenous calcium
ionophore to release calcium from stores in the endoplasmic reticulum and mitochondria.
DAG activates protein kinase C which among other things regulates hydrogen ion transport
and intracellular pH. Thus within seconds after receptor ligation there are near instantaneous
increases in intracellular calcium and intracellular pH. We have followed both of these
parameters in real-time in lymphocytes exposed to ELF fields. Next, early response ST genes
such as c-MYC are activated within sixty minutes and the cell ultimately progresses into
mitogenesis at times > 300 minutes. This time course is presented in Table 1.
Amplification is a key feature of ST that is relevant to ELF field interactions. The
ST cascade achieves enormous amplification and we have postulated that an ELF-mediated
modification of an early event at the cell membrane involving calcium ion influx could lead
to, and explain, significant changes in subsequent ST events such as gene expression and
cell ' Signal propagation involves at least two, and usually many steps that amplify
the effect of the initial binding event at the cell surface. For example, the binding of one
ligand molecule to its receptor at the cell surface leads to activation of multiple transducer
proteins or to the influx of a large number of calcium ions which will activate intracellular
enzyme molecules.
Evidence for 60 Hz magnetic fields increasing calcium-45 influx during signal
transduction in the lymphocyte has been reported from our laboratory."''^ Abrief60 minute
exposure of rat thymic lymphocytes to a 220 Gauss magnetic field (E.^duced - 1 0 mV/cm) at
37°C was performed in the presence or absence of Con A. Our magnetic field exposure
system and the special multi-ring annular ring petri dishes we designed are described below
in greater detail. Non-activated cells (no mitogen) were unresponsive to the magnetic field;
calcium-45 influx was not altered. When Con A was present, the magnetic field led to an
 R. P. Liburdy

Calcium Uptake in Thymocytes: Effect of a

60 Hz Magnetic Field on ConA Activated Cells

- v

a) U3
Y
0
d
n
-
V?

52
x
5-

4-
O - C o r ~A
m + C o n A
+
C o n A. + MF I

A B C D E

% Con-A Activation 100 30 220 275 75 Figure 2. Cells assessed for calcium up-
% MF Increase 100 275 80 40 350 take in the absence or presence of Con-A,
Exposure Conditions: 220G (rms), 1 mV/cm (max), and in the presence o f con-A and a mag-
15uA,,cm2(max), 370 c ,60 minutes netic field. Exposures at 1.0 mV/cm, 220
p < 0.03, n=3 (unpaired) Gauss, 16 uAlcm2, 20 ug/ml Con-A, 37 +
p < 0.01, n=5 (paired) 0.05"C, 60 minutes.

A. Exposure Geometry
Current

-
Fluorescence Cuvette
Ohm's Law
V = IR,
1
E = VElectrode Spacing
B. Electric Field in Cuvette
Electric field is uniform
throughout center
of Cuvette
carrylng
R-Electrodes
Barr~er embedded
In Agar

Electr~c Figure 3. Schematic diagram of one of our

F~eldLlnes
fluorescence cuvette exposure devices.
fluorescence Cuvette
Electromagnetic Fields and Cellular Systems 89

increase in calcium-45 influx of 50 - 200%. This variation depended on the level of calcium
influx achieved by Con A; weakly activated cells were associated with maximal response to
the field. This relationship is shown in Figure 2.
We have investigated the effect of ELF fields on calcium influx using a different
approach involving real time fluorescence spectroscopy. Real-time changes in intracellular
calcium ([Ca-*],) in cells exposed to 60 Hz fields was reported from our laboratory.' The
advantage of this approach is the ability to observe field effects as they occur in real time.
Thymic lymphocytes were loaded with the calcium sensitive dye FURA-2AM and the cells
exposed to a 60 Hz electric field (1.7 mV/cm) in a electrode cuvette similar to that shown
in Figure 3 while fluorescence was monitored; we have made some electrodes using stainless
steel and platinum foil and we have used some different electrode shapes. A cuvette of this
general design with an optical slit permits cells to be exposed to a uniform 60 Hz electric
field while simultaneous monitoring intracellular free calcium. Figure 4 shows that levels
of intracellular calcium in resting cells (no mitogen) were not apparently influenced by the
field and this is consistent with the calcium-45 influx data for resting cells, mentioned above.
When the mitogen Con-A was added to the cells during field exposure levels of intracellular
calcium began rising and diverged to reach higher values than that for unexposed, Con-A
treated cells. These studies suggest that the initial rise in calcium, the early phase due to
intracellular calcium release over approximatley 100s, was not altered by the field, and that
the steady-state levels of calcium influx were enhanced thereafter. These results are in general
agreement with the the calcium-45 influx studies discu.ssed above in which Con-A treated
cells displayed an increase in calcium-45 influx during field exposure at E,„ju,.„| = 1.0 inV cm.
The biological significance of these real-time fluorescence studies has some rele-
vance to the question of an interaction site. Our observation that the initial rise in intracellular
calcium during Con-A treatment was not influenced by the field indicates that release of
calcium from intracellular stores inside of the cell during ST was not disturbed by the field.
It is known that the initial rise in intracellular calcium during ST is due to release of calcium
from the mitochondria and the endoplasmic reticulum. Our finding that only the plateau
phase was increased during field exposure indicates that influx of calcium through a
ligand-mediated channel in the cell plasma membrane was influenced by the field. This
follows since the plateau phase is sustained by the influx of extracellular calcium across the

IPIotaau Pho8«l

Control
2

o
o

Figure 4. Real-time measurements of Con —A, 1/ig/ml

[Ca*-]| in rat thymocytes during expo- 60Hz
m
sure lo 60 Hz electric fields. The field
effect i.s associated with an increase in _i I L i _
intracellular calcium during mitogen ac- 100 200 300 400 500 600 700 800 9001000
tivation.' 1.7 mV/cm; .'57 + 0.05°C. Time (seconds)
90 R. P. l.iburdy

plasma membrane through a specialized ligand-gated channel. Therefore our real-time study
suggests that the cell membrane and calcium influx was involved in the field interaction.
The above work from our laboratory leads to several important hypotheses about
ELF field coupling to cells: (1) the cell membrane is involved in cellular responses to ELF
fields, (2) ELF fields affect receptor-ligand ST events such as ion transport at the cell surface,
(3) ELF fields can profoundly potentiate suboptimally activated cells during ST.

GENE ACTIVATION
The research discussed above suggests that calcium influx is enhanced during ST in
the presence of ELF fields. We postulated that such ELF field effects on calcium cycling can
play a significant role in triggering mid-stage ST events such as c-MYC niRNA induction
via the ST cascade.-''' This means that ELF field effects on early events such as calcium
signaling are propagated down the ST cascade to alter subsequent mid-stage events. Above
it was mentioned that increases in calcium influx are early events during ST and occur within
the first several minutes whereas the induction of mRNA occurs at least 60 minutes into the
ST cascade.
Within the ST framework gene activation is triggered after an initial event at the cell
surlace such the binding of a hormone or growth factor to its receptor. Since membrane
effects are observed experimentally in our studies of ion transport during signal transduction,
discussed above, we reasoned that ELF fields could subsequently activate genes through the
S r pathway. This is an epigenetic interaction mechanism that is consistent with a wide range
of laboratory evidence arguing against a direct effect of sinusoidal power-frequency ELF
fields on DNA.'^
Our approach to establish a linkage between ELF field effects on early and mid-stage
ST events was to measure both parameters simultaneously in the same cell preparation
exposed to 60 Hz magnetic fields.
To do this we correlated alterations in calcium influx and in c-MYC mRNA induction
measured in the same exposed cell population.''^"' The hypothesis, mentioned abo\e. was
that increases in calcium influx induced by the ELF field would subsequently lead to
increases in cMYC mRNA transcription via the signal transduction cascade. This represents
an attractive and highly plausible biological framework for understanding how a magnetic
field effect at the cell surfiice, i.e. calcium ion influx can influence subsequent events such
as RN.A, DNA, and protein synthesis.
We followed c-MYC mRN.A in these studies because the oncogene MYC plays an
important role in ST. MYC belongs to a set of cellular messengers commonly referred to as
"immediate early response" genes since their expression is activated by a variety of
mitogenic stimuli, independent of tie novo protein synthesis, early during the G,, to (i,
transition of cells from a resting to a growing state.'" The protein products of immediate
response genes such as MYC are thought to facilitate progression of the cell through the cell
cycle and synthesize DNA in S phase. MYC polypeptides play roles in the transcriptional
and post-transcriptional control of other cellular genes and in DNA replication. Strong
evidence indicates that they mediate their function as site-specific DN.A-binding proteins;
cells keep MYC under very tight regulation.
In our experiments rat thymocytes were placed in a special glass annular ring cylinder
and exposed to an incident 220 Gauss 60 Hz magnetic field (E,„ji„,ed = 1.7 mV cm) for 60
minutes. 37C, as described.'"*'^ The cell sample was harvested and split into two aliquots
and simultaneously assessed for calcium influx and for cMYC mRN.A.
Figure 5 depicts the results from three independent experiments in which calcium
influx was measured in thymic lymphocytes in the absence of Con-A, in the presence of
Electromagnetic Fields and Cellular Systems 91

16000
60H2. 22mT, 1.7mV/cn
Con-A, l / i f l / i
o 14000 60min, 37C
p < 0.0001*
O

z
0. 12000
Figure 5. Calcium influx and cMYC o
mRNA transcripts are elevated in the same X
population of mitogen-activated rat thymo-
10000
cytes following 60 Hz magnetic field expo-
sures: calcium influx data. Results from
three independent experiments in which
samples were split for simultaneous assess- o 8000
ment of c a l c i u m influx and cMYC o
mRNA.'-' 1.7 mV cm. 220 Gauss, 27
uA/cnr, 1.0 ug.'ml Con-A. 37 0.05°C, 60 6000
minutes. Mean S.D.

Con-A (1 i-tg/ml), and in the presence of Con-A plus a magnetic field. Non-acthated
thymocytes (no Con-A) exhibited a baseline calcium influx of 10,400 386 cpm 10 cells.
A dose of Con-A at 1 ng/ml did not result in a significant increase in calcium influx. However,
when cells were exposed to a 22mT magnetic field in the presence of Con-A. an approximate
1.5 fold increase in calcium influx was observed. This is consistent with our previous
observations mentioned above of enhanced calcium influx for mitogcn-activatcd thymocytes
exposed to a 22 mT, 60 Hz magnetic It is important to note that mitogen-activated
calcium influx in lymphocytes is dependent on animal age, mitogen dose, and immune .status
of the animal.- In these studies each of the independent experiments employed thymocytes
pooled from three, age-matched animals, and this pooled population responded suboptimally
to a dose of 1 ng/ml of Con-A. This permitted detection of an increase in calcium influx of
Con-A treated cells due to the field. Cells maximally activated by Con-A are at their maximal
dynamic limit and further calcium influx in response to an applied magnetic field is not
possible.
Figure 6 shows Northern blot analyses that were repeated three times on the same
RNA sample from one experiment to demonstrate the precision and reproducibility of this
technique prior to optical density calibration and normalization to message for the house-
keeping gene glyceraldehyde-6-phosphate dehydrogenase (GAPDH), which is not altered
during ST. Northerns were quantitated using a CCD-camera and the digitized images of
bands were integrated on a pixel-by-pixel basis so that [total band gray-scale intensity x
mm-'Jwas computed and this quantity was then linearized by conversion to [OD x mm-].
Panel A shows the images obtained for each Northern. Panel B represents the regions of
interest for the three bands in each Northern. Panel C shows the image after flat-fielding and
thresholding. Panel D shows surface plots of pixel intensity for the bands in panels C. We
compared the mean pixel intensity for these bands using a multivariate analysis of variance
statistical program that analyzed for differences across bands in these experiments. Mean
pixel intensity (arbitrary units S.D.) was -Con-A = 332 128; -^Con-A = 398 48: -^Con-A
plus 60 Hz magnetic fields = 864 181. A statistically significant difference was not detected
between -Con-A and -i-Con-A (p > 0.05). A statistically significant difference was detected
between +Con-A and -i-Con-A plus 60 Hz at the p = 0.0114 level. Thus, we were able to
obtain statistically significant results in the absence of O.D. calibration and normalization
to glyceraldehyde-6-phosphate dehydrogenase (GAPDH), discussed below. Normalization
92 R. p . I j b u r d y

Figure 6. CCD-caniera imaging of c-MYC niRNA Northerns: Comparison of three independent blots of the
same RNA sample from one experiment.'-* Panel A: flat-fielde-d images of c-IVIYC bands. Panel B: Regions of
interest for the bands. Panel C: Threshoided images of the Northern blots. Panel D: .surface plots of pixel
intensity for bands in each image.

is important since it corrects for any generalized, nonspecific changes in total mRNA in the
cells.
In order to determine whether the above effect is specific for the c-MYC niRNA
species or due to a general increase in mRNA abundance, we analyzed levels of RNA
encoding a "hoiiselceeping" protein GAPDH, on the same blot. RNA samples from the above
experiment, plus RNA samples from two additional, independent experiments were ana-
lyzed. For each Northern blot this standardization procedure involved normalizing the
quantitated band intensity (O.D. x .tiim') for c-MYC by the associated band intensity for
GAPDH. We observed that GAPDH mRNA does not vary across treatment groups corre-
sponding to -Con-A, +Con"A, and +Con-Aplus ELF field'^ This indicates that GAPDH is
not affected by the mitogen Con-A or the ELF field and this is not suiprising since GAPDH
is not activated during ST. For this reason we used it is an appropriate control for normalizing
changes in c~MYC mRNA.
Using GAPDH to normalize our c-MYC mRNA data we computed the relative ratios
of c-MYC mRNA abundance to GAPDPI abundance across the three experiments with the
c-MYC/GAPDH ratio for non-activated cells set to one. Figure 7 shows the ratio of c-MYC
mRNA/GAPDH mRNA for cells in the three independent experiments of Figure 5. Levels
of c-MYC mRNA expression in non-activated cells were not observed to be statistically
different than cells treated with the suboptimal dose of 1 pg/ml Con-A. In contrast, cells
treated with Con-A plus magnetic fields exhibited an approximate 3.0-fold increase in
Electromagnetic Fields and Cellular Systems 93

< c
C3 o
Figure 7. Calcium influx and cMYC mRNA
transcripts are elevated in the same population
of mitogen-activated rat thymocytes follow-
ing 60 Hz magnetic field exposures: c-MYC " "5
mRNA Data. Results from three independent H- N

experiments in which samples were split for

O
2t.
simultaneous assessment of calcium influx :;: ^
and cMYC mRNA." Relative transcript lev- o >-
els for c-MYC mRNA were standardized to
the housekeeping enzyme GAPDH by quanti-
tation using coolcd-CCD image analysis. See
Figure 5.
CON A -
60Hz _

C-MYC mRNA transcript levels that was statistically significant. These findings correlate
across treatment groups with that observed for calcium uptake shown in Figure 5. Suboptimal
mitogen treatment by itself did not lead to increases in cMYC mRNA, however the presence
of the magnetic field resulted in an approximate 4-fold increase in cMYC mRNA.
This data provides evidence that ELF effects on these two ST events are linked, and,
based on calcium's role in the ST cascade, this provided strong evidence for an interaction
mechanism in which ELF fields trigger calcium influx at the level of the cell membrane and
this leads to subsequent changes in events in ST pathway. This ST interaction model explains
how ELF field effects have the potential to alter subsequent cellular events in this cascade.
Other laboratories have investigated the effects of fields on cMYC mRNA prior to
our studies. Goodman and Shirley-Henderson reported that cMYC mRNA was enhanced by
magneticfields'**and by 60 Hz sinusoidal electric fields (0.3 x 10"'*V/m).''* Recently Phillips
and colleagues reported that a brief 1 Gauss magnetic field exposure of a T-lymphoblastoid
cell line, CEM-CM3, also increased mRNA transcripts."^" Their report was the first to employ
the nuclear run-off assay technique to assess alterations in specific gene transcription for
MYC, JUN, FOS and protein kinase C.

CELL PROLIFERATION
The final endpoint in the ST process is cell proliferation which results in mitogenesis.
Since cell proliferation is the ultimate endpoint for the signal transduction pathway, it is
important for studies to be conducted to evaluate the effect of electromagnetic fields on cell
proliferation and growth, particularly of cancer cell lines. This we have done for the case of
human breast cancer cells. This is an important question since ELF fields have been
postulated as a significant risk factor in human breast cancer epidemiology.-'-"'
Our studies were recently published and suggest that human breast cancer cell growth
in vitro can be altered by a 60 Hz magnetic field.^' In these studies, melatonin, a hormone
with natural oncostatic activity, was employed to suppress breast cancer cell growth and this
response was investigated during field exposures. This is analogous to the studies discussed
above in which a mitogen such a Con-A is used to trigger the ST cascade and cells are exposed
to 60 Hz fields. This concept is a critical feature in cellular studies since resting or quiescent
cells, as discussed above, do not show sensitivity to an imposed ELF electric or magnetic
94 R. P. IJburdy

field. In general cells appear to be most responsive to fields when they are engaged in signal
transduction mediated by cell membrane events such as receptor binding and calcium
cycling.
We conducted experiments employing a well-characterized human breast cancer cell
line, MCF-7, and we tested whether the growth of MCF-7 cells is suppressed by melatonin
in the presence of a 2 or 12 mGauss 60 Hz sinusoidal magnetic field ' \ Melatonin is a
hormone and natural growth inhibitor of estrogen positive breast cancer cells which is
normally released into the blood stream at night. Melatonin is of interest since (a) it displays
natural oncostatic action towards breast cancer cells,^'' (b) melatonin release into the blood
stream in animals has been reported to be depressed by 60 Hz magnetic fields,^'"'-^ and (c)
60 Hz magnetic field exposures are postulated to be a risk factor in human breast cancer,
mentioned above.

IN VTIRO CELL CULTURE EXPOSURE SYSTEM

Prior to performing in vitro experiments on cell growth we had to innovate an

approach to exposing cells in culture to very uniform ELF magnetic fields while maintained
inside of commercially available incubators. All in vilro experiments in which cells are grown
in commercial incubators and subsequently exposed to ELF fields have to be done very
carefully taking into account several important considerations. In the past in vitro cell culture
experiments have neglected, in general, the fact that all commercial incubators generate
magnetic and electric tlelds due to the operation of the heating and gas exchange electronics
associated with the incubator; and considerable variation can exist across manufacturers.
Since the endogenous magnetic and electric fields incubators can vary greatly, both in spatial
distribution and in time, complete field mapping is essential to fully characterize the interior
of the incubator.
In addition to the question of ensuring that our cells were exposed to a reasonably
uniform ELF magnetic field in our incubators (< 10%), we also addressed the important
question of ensuring that our cells during routine propagation in culture had a known
electromagnetic history. Thus we routinely propagate all of our cells using the system
described below to ensure that the cells have a known prior ELF exposure history before
conducting field exposure experiments.
The approach we use is to shield the interior space of the incubator using 80% nickel
mu-metal.'' One of our mu-metal chambers inside a cell culture incubator is shown in Figure
8 (Co-Nectic .AA Foil, Magnetic Shield Corporation, Perfection Mica Co., BensenviUe, IL).
This mu-metal chamber is ventilated at the upper and lower corners with 2.5 cm diameter
holes that have 2.5 cm extension tubes to minimize entry of stray magnetic fields. The
chamber has a hinged door so that it can be opened and closed in a manner identical to the
commercial incubator door. Our use of a closed, ventilated mu-metal shielding is the best
solution for eliminating all of the endogenous magnetic and electric fields generated by the
operation of the cell culture chamber. There exist time varying magnetic field generated
during the transient operation of the gas solenoids of the incubator and during the transient
heating pulses of the outer water-jacket of the incubator; in addition some incubators have
electrically-heated door panels. All of these fields have 60 Hz and/or ELF components and
their lime profiles can vary significantly depending how the frequently the incubator is
opened. Another significant problem is that these fields can display significant spatial
variation within the incubator. Most if not all incubators have their electronic equipment
mounted on the top of the incubator so that cells placed on the top shelves are exposed to
more intense endogenous fields that cells placed on the very bottom of the incubator. By
using a mu-metal shielded chamber we avoid all of these problems.
Elcctroiiiagnetic Fields and Cellular Systems 95

Figure 8. Exposure coil and mu-nietai

chamber shown inside of a commercial
cell culture incubator."

To expose celis to a uniform ELF magnetic fields we used two identical, double-
wound, four-coil, Merritt exposure systems one of which is also shown in Figure 8.^^ The
figure shows the four-coil Merritt design in which there are four double-wound coils wrapped
around a plastic frame. These coils generate a large and relatively uniform area for exposing
an array of eel! culture plates if Merritt's turns ratio of 26/11/11/26 is followed in mdnding
these four coils (17.0 Q, 6.57 mH). With the center axis of the coil system used as a reference
line and the center point on this axis designated the origin, the four coils had the following
spacing with respect to the origin: 16.7 cm, 4.23 cm, -4.23 cm, and -16.7 cm. Commercially
available standard speaker cable with two parallel wire tracks was used for the double-wrap
cable. Built into the circuit energizing Ihese coils was a switch used to reverse the direction
of the current in one of the wires in the speaker cable comprising the double-wound coils.
When corrent is applied in the anti-parallel configuration (passive) the magnetic fields from
the double-wound coils cancel, and when current is applied in the parallel configuration
(active) a magnetic field is established. This feature can be used to perfomi a control exposure
in one incubator to detennine if coil heating and coil vibration is a factor; relatively high
field exposures greater than approximately 1 Gauss may result in heating and vibration but
this is not the case for fields in the 12 -1 SmGauss range used in our studies. Each coil system
was driven by identical signal generators available from Dynascan Corp., Chicago, IE (B&K.
Precision Model 3020). Each of the four coils comprising the exposure system was shielded
by wrapping the wire bundles in two layers of heavy-duty aluminum foil, with a break of
several inches, to eliminate the electric field compouents generated by current running the
wire wrappings. The magnetic fleld generated by this coil orientation is perpendicular to the
96 R. P. Liburdv

_
1.5 - -
0 M melatonin A

v +

O 3
1.0

0.5
-

k
J^
-

nn 1 1 I 1
Figure 9. Melatonin's action on MCF-"? cell
2 4 e
growth in a 2.0 mG 60 Hz magnetic field
DAYS OF CULTURE

plane of the cell culture plates. Culture plates are positioned on a perforated, Plexiglas
platform that corresponds to the middle plane of the coil. Field uniformity is conservaively
< 10% over the central area where cells are placed. Note that an external temperature probe
is fed through one of the ventilation holes of the chamber to monitor temperature continu-
ously during exposures.

BREAST CANCER CELL GROWTH STUDIES

We tested the hypothesis that 60 Hz magnetic fields block melatonin's growth
inhibitory (oncostatic) action on MCF-7 cell proliferation. We have observed that normal
MCF-7 cell growth is not altered by these magnetic fields in the absence of melatonin.'^ Dr.
D. Blask has reported the melatonin inhibits the proliferation of MCF-7 cells when it is
present in cell culture media at concentrations corresponding to the physiological range of
l O ' to 10'"M.'"' We first performed experiments to confirm melatonin's inhibition of MCF-7
cell growth. Growth curves for MCF-7 cells in the absence or in the presence of 10'"M
melatonin are presented Figure 9. In these experiments cells were placed in an incubator
with the exposure coils energized in the antiparallel configuration (passive) so that a
environmental-level background 60 Hz magnetic field of 2.0 mG was present. This magnetic
field simulates a background field commonly found away from electrical appliances in work
places or homes. Cell proliferation revealed typical exponential growth over the 7-day
period. When melatonin was present, an approximate 25% inhibition of growth was detected
at day 7. These results confirmed Blask's original observation of melatonin's inhibition of
MCF-7 cell growth mentioned above. However, it is important to note that MCF-7 cells from
different laboratories can display heterogeneity in their response to melatonin, and some
subclones can exhibit growth inhibition of up to 70% in the presence of melatonin.^'' Some
MCF-7 subclones may exhibit no sensitivity to melatonin - this may depend on passage
number, serum factors, or other unknown parameters.''
Simultaneously with the 2 mG 60 Hz magnetic field exposures above we conducted
exposures to 12 mG in a matched incubator. We note that 12 mG will be used in the text that
follows with the understanding that mean field values for our experiments fall within the
range from 12 to approximately 15 mG{rms). This field level range represents an environ-
mental level that is at or near the high end of magnetic field strengths people can experience
from exposures to common household or work place appliances'^", much higher field levels
 Electromagnetic Fields and Cellular Systems 97

no malatontn
1.5
— 10 M mdatonln

X in
1.0 -

Figure 10. Melatonin's action on 1V1CF-7 cell 0.0

2 4 6
growth in a 12.0 mG 60 Hz magnetic field.
DAYS OF CULTURE

however, can be encountered very near electronic devices commonly found in the kitchen.
A growth curve for MCF-7 cells in the presence or absence of melatonin in a 12 mG magnetic
field is shown in Figure 10. As can be seen an exponential growth curve for the MCF-7 cells
in absence of melatonin and 12 mG (solid line) was obtained that is superimposable with
that obtained for cells in the presence of melatonin and 12 mG field (dashed line). This data
reveals that the 12 mG magnetic field in the presence of melatonin at 10""M blocked
melatonin's growth inhibition. Compare this result at 12 mG with the approximately 25%
growth inhibition observed in the 2 mG field shown in Figure 9. These two experiments were
performed simultaneously with matched incubators and cells from the same passage.
The above experiments have been repeated a number of times for melatonin at 10'^M
to confirm these results for this physiological concentration of melatonin. Presented in Figure
11 is a summary graph for seven experiments, using cells from different passages, in which
2 mG and 12 mG exposures were performed simultaneously in matched incubators. We
observe that melatonin resulted in an approximate 15% growth inhibition in a 2 mG magnetic
field and that this was blocked by a 12 mG.
Our findings on MCF-7 cells suggest that (a) this in vitro effect is a cellular level
response to magnetic fields involving cell proliferation and growth, (b) it involves an
interaction that requires the presence of melatonin which is a natural oncostalic agent, and

SUMMARY DATA: 7 Exparlmants

p < 0.00001

Figure U, Summary data of 60 Hz magnetic fields Melatonin: None

on Melatonin's action on MCF-7 cell proliferation.
98 R. P. Liburdy

(c) a dose threshold appears to exist between 2 and 12 mG. These findings represent some
laboratory evidence for a cellular level response to ELF magnetic fields that is dependent
on the presence of melatonin a naturally occurring hormone that has relevance to human
breast cancer.
Several models have been proposed to elucidate the link between ELF magnetic field
exposure, melatonin, and breast cancer incidence.-''''-' These models suggest that the most
important aspect of this link is a decrease in melatonin secretion in response to an in vivo
magnetic field exposure concomitant with enhancement in the production of prolactin and
estrogen; the latter events are thought to increase the growth of susceptible breast epithelial
cells.-*' Our laboratory results, in contrast, deal with in vitro exposure of breast cancer cells
and, therefore, relate to cellular events that occur distal to an in vivo effect on the pineal
gland which regulates melatonin's secretion into the blood stream. Keeping in mind the
caveat that our results relate to an in vitro cellular effect, our findings suggest the possibility
of an interaction between ELF magnetic fields; breast cancer cells, and melatonin may exist
at the at the cellular level. This means that in addition to an ELF magnetic field effect on
melatonin release into the blood stream, there might exist an ELF field effect on melatonin's
function at the level of the target cell, e.g. human breast cancer tissue and its proliferation.
This needs to be tested in animal model systems to be verified.

ACKNOWLEDGMENT
Support was provided in part from the Office of Health and Environmental Research,
and the Office of Energy Management, Utilities System Division, of the Department of
Energy, under contract DE-AC03-76SF00098. Support was also provided to RPL by the
N.l.H. through N.C.I, project CA53711.

REFERENCES
i. f'rey, A. Eleclromagnetic Field Interactions with Biological Systems, FASEB JourmA (1993) 7: 272-2X1.
2. Liburdy. R.R, bXF Fields and the Immune System: Signal Transduction. Calcium Metabolism, and
Mitogenesis in i..ymphocyles with Relevance to Carcinogenesis, Inlnlerucdon Mechanisms of Low-level
Eleeiromugneiic Fields in Living Systems [Eds. B. Norden & C. RamelJ, Oxford University Press, N\'.
NY. 1992, pp. 217-2.'?9.
3. Liburdy. R.P.. Calcium Signaling in Lymphocytes and ELF Fields: Evidence for an Electric Field Metric
and a Site of Interaction Involving the Calcium Ion Channel. FEBS Letters (1992) 301; 53-59.
4. Luben. R..^., Effects of Low-Energy Electromagnetic Fields (EMF) on Signal Transduction by G Protein
Linked Receptors. In Eleelrieily unci Magnetism in Biology & Medicine \\:A.. .M. Blank]. San Francisco
Press. San Francisco. 1993, pp. 57-62.
5. Tenforde, T.S.. Biological Interactions of Extremcly-Low-Frcquency Eleclric and Magnetic Fields.
Bioeleclrochemist/y ami Bioeneigetics (1991) 25; 1-17.
6. Tenforde, T.S.. Cellular and Molecular Pathways of Extremely Low Frequency Electromagnetic Field
Interactions with Living Systems. In Electricity and Magnetism in Biology and Medicine [Ed. M, Blank].
San Francisco Press. San Franci,sco, CA, 1993, pp. 1-8.
7. Frey, A. [Editor] On The Nature of Electromagnetic Field Interactions With Biological Systems. R.G.
Landcs Company. Austm, TX, 1994.
X. Liburdy. R.P. Cellular Interactions with Electromagnetic Fields: Experimental Evidence for Field EtTects
on Signal Transduction and Cell Proliferation, In On The Nature oj Electromagnetic Field Interactions
With Biological Systems. [Ed. A. Frey] R.G. Landes Company, .Austin, TX. 1994. pp. 99-125,
9. Morgan, N.G. Cell Signaling. The Guilford Press. NY. 1989.
10. .Altinan, A.. T. Mustelin. KM. CoggeshaU, T Lymphocyte Activation: .\ Biological Model of Signal
Transduction. Crit. Rev in Immunol. (1990)10: 347-391.
Electromagnetic Fields and Cellular Systems 99

11. Liburdy, R.P. & J. Walleczek, The Influence of Magnetic Fields on Calcium Metabolism in the Lympho-
cyte. Annual DOE Conlractor'x Review. Portland, OR, 13-17 September, Abstract A2. 1989.
12. Walleczek, J & R.P. Liburdy, Nonthermal 60 Hz Sinusoidal Magnetic-Field E.\posure Enhances *
Uptake in Rat Thymocytes: Dependence on Mitogen Activation, FEBS Leitcrs (1990) 271; 157-160,
13. Yost. M.G& Liburdy, R.P.,Time-Varying and Static Magnetic Fields Act in Coinbinationlo Alter Calcium
Signal Transduction in the Lymphocyte. FEBS Leilerx {]9')2) 296: 117-122.
14. Scarfi, MR., F. Bersani, A. Cossarizza, etal.,50Hz AC Sinusoidal Electric Fields Do Not Exert Genotoxic
Effects (Micronucleus Formation) in Human Lymphocytes, Radiation Research (1993) 135: 64-68.
15. Liburdy. R.P. D.E. Callahan, J. Harland, et al.. Experimental Evidence for 60 Hz Magnetic Fields
Operating Through the Signal Transduction Cascade: Effects on Calcium Influx and C-MYC mRN.A
Induction. FEBS l.eliers (1993) 334: 301 -308.
16. Liburdy, R.P. D.E. Callahan. T.R. Sloma. et al.. Intracellular Calcium, Calcium Transport, and cMYC
niRN.A Induction in Lymphocytes Exposed to 60 Hz Magnetic Fields: The Cell Membrane and the Signal
Transduction Pathway, In Electricity and Magnetism in Biology and Medicine [Ed. M. Blank], San
Francisco Press, San Francisco, CA. 1993, pp. 311-314.
17. Marcu, K.B., S..A. Bossome & A.}. Patel, MYC Function and Regulation. Annual Rev. Biochemistry
(1992)61: 809-860.
18. Goodman, R. & A. Shirley-Henderson, Exposure of Human Cells to Electroinagnetic Fields: Effect of
Tiine and Field Strength on Transcript Levels, Electro- and Magnetohiology (1992) 11: 19-28.
19. Blank, M., L. Soo, H. Lin, et al,. Stimulation of Transcription in HL-60 Cells by .Alternating Currents
From Electric Fields, In Electricity and Magnetism in Biology and Medicine [Ed. M. Blank], San
Francisco Press. San Francisco, CA. 1993. pp. 516-519.
20. Phillips, J.L., W, Haggren, W.J. Thomas, et al.. Magnetic Field Induced Changes in Specific Gene
Transcription, Biochemica et Biophysica Acta (1992) 1132; 140-144.
21. Stevens. R.G.. Electric Power Use and Breast Cancer: A Hypothesis. Am. J. Epidemol. (1987) 125;
556-561,
22. Tynes. T, & ,\. Andersen, Electromagnetic Fields and Male Breast Cancer, Lancet (1990) 336; 1596.
23. Demers. P.A.. D.B. Thomas, K.A. Rosenblatt, et al.. Occupational Exposure to Electromagnetic Fields
and Breast Cancer in Men. Am. J. Epidemiol. (1991) 132; 715-716.
24. Matanoski. G.M., P.N. Breyese & E.A. Elliot. Electromagnetic Field Exposure and Male Breast Cancer.
Z.(jm-<.'/( 1991)33; 737,
25. Liburdy, R.P. T.R. Sloina, R. Sokolic. et al., ELF Magnetic Fields, Breast Cancer, and Melatonin: 60 Hz
Fields Block Melatonin's Oncostatic .Action on ER-H Breast Cancer Cell Proliferation,./, Pineal Research.
(1993) 14; 89-97.
26. Blask. D.E,. The Emerging Role of the Pineal Gland and Melatonin in Oncogenesis. In E.xtremely Low
Freqiiencv Electromagnetic Fields: The Question of Cancer [Eds. B.W, Wilson. R.G. Stevens. & L.E.
Anderson], Battelle Press, Columbia Press, OH, 1990, pp. 319-325.
27. Wilson, B. W.. C.W. Wright. J.E. Morris, et al.. Evidence for an Effect of ELF Electroinagnetic Fields on
Human Pineal Gland Function,./. Pineal Research, {\<)W) 9; 259-269.
28. Reiter, R.J, & B.A. Richardson. Magnetic Fields Effects on Pineal Indoleamine Metabolism and Possible
Biological Consequences. F^S£SJ,( 1990) 6; 2283-2287.
29. Lerchl, .A., K.O. Hanaka & R.J. Reiter, Pineal Gland Magnetosensitivity to Static Magnetic Fields is a
Consequence of Induced Electric Currents (Eddy Currents). / Pineal Research. (1991) 10; 108-116.
30. Haggren, W., S.M. Yellon. J.L. Phillips, et al.. Effect of Magnetic Field Exposure in .Adult Djungarian
Hamsters on Melatonin Rhythins and Molecular Markers, First World Congress form Electricitv and
Magnetism in Biology and Medicine, Lake Buena Vista, FL, June 14-19. .Abstract P-119, 1992.
31. Stevens, R.G., S. Davis, D.B. Thomas. L.E. Anderson & B.W. Wilson, Electric Power, Pineal Function,
and the Risk of Breast Cancer, FASEB J.(]992) 6; 853-860.
32. Reiter. R.J.. Changes in Circadian Melatonin Synthesis in the Pineal Gland of Animals Exposed to
Extremely Low Frequency Electromagnetic Radiation: A Summary of Observations and Speculation on
Their Implications, In Electromagnetic Fields and Circadian Rhythmicity [Eds. M.C, Moore-Ede. S,S,
Campbell & R.J. Reiter], Birkhauser. Boston. 1992, pp. 13-17.
33. Liburdy. R.P. Cell Culture Incubators, ELF Fields, and Important Considerations for Cell Culture
Exposure and Propagation: An Exposure System for In Vitro Bioelectromagnetics Research, Bioeleclro-
magnetics Society Meeting. Copenhagen, Denmark, June 12-17, 1994, .Abstract P-198.
34. Cos. S, & Blask. D.E, Effects of the Pineal Hormone Melatonin in the .Anchorage-Independent Growth
of Human Breast Cancer Cells (.MCF-7) in a Clonogenic Culture System. Cancer Letters (1990) 50;
115-119,
100 R. P. l.iburdy

35. L'Hermitc-Baleriaux, M. & De Launoit, Y. Is Melatonin Really an In Vitro Inhibitior of Human Breast
Cancer Cell Proliferation? !n Vitro Dev. Biol. (1992) 28A; 583-584.
36. Stuchly, M..^., D.W. Lecuyer& R.D. Mann, H.xtremely Low Frequency Electromagnetic Emissions form
Video Display Termmals and Other Devices, Health Physics (1983) 45; 713- 722.
37. Tcnforde, T.S & W.T. Kaune. Interaction of Extremely Low Frequency Electric and Magnetic Fields with
Humans, Health Physics. (1987) 53; 585-606.
EFFECTS OF ELECTROMAGNETIC FIELDS
ON K^(Rb') UPTAKE BY HeLa CELLS

Hiroshi Miyamoto', Hisao Yamaguchi^, Toshitaka Ikehara', and Yosuke

Kinouchi'

' Department of Life, Environment and Information

Faculty of Domestic Science, Tokushima Bunri University
180 Yamashiro-cho, Tokushima 770, Japan
t of Physiology, School of Medicine
The University of Tokushima
Kuramoto-cho 3, Tokushima 770, Japan
''Department of Electrical and Electronic Engineering
Faculty of Engineering, The University of Tokushima
Minamijosanjima-cho, Tokushima 770, Japan

INTRODUCTION

The effects of static or time-varying magnetic fields on membrane transport of alkali

cations have not been reported more frequently than studies on Ca''* fluxes. Gualtierotti et
al. [1964] observed decrease in Na"^ transport in frog skin on exposure to a static magnetic
field of less than 1 T. CoUis and Segal [1988] reported influences of exposure to a weak
pulsed magnetic field (PMF) on bi-directional -''Na'^ fluxes through rabbit colon epithelium,
finding that the effects on influx and efflux of the ion differed depending on the direction of
the magnetic field. In contrast, Hinsenkamp et al. [1985] and Farndale et al. [1987] failed to
detect any significant effects of exposure to a PMF on the activities of the Na"^-pump and
Na^/K*/Cl"-cotransport of human red cells. Stevenson and Tobey [1985] found that exposure
to a PMF inhibited '''K'^ uptake slightly, but not significantly. As these results are discrepant,
further studies on the effects of various magnetic fields on membrane transport of monova-
lent cations are necessary.
In the present study, we tested the effects on active and passive influxes of Rb* (as a
substitute for K*) into HeLa cells of strong homogeneous magnetic fields. We also tested the
effects on these influxes of strong, time-varying magnetic fields, which have not previous
been investigated extensively, but which could be useful for assessing the physiological
effects at a cellular level of repeated exposures of the human body to a strong magnetic field
over a short period.

Biological EJft'cls of Magnetic ami F.U'clromugfwlic Fields. Edited by S. Ueno

Plenum Press, New York, 1996 101
102 H. Mivamoto et al.

ELECTROMAGMETIC FIELDS
Magnetic fields were produced by an electromagnet designed and set up by Hitachi
Metal Indust. Co. (Tokyo, Japan). The apparatus had two vertically arranged coils, each of
which was attached with a pole piece of round polar face of small (50 mm) or large (100
mm) diameter. The distance between the two polar faces was changeable. A thyristor rectifier
(RCX-100H2, Sansha Electric MFG. Co., Tokyo, Japan), normally used for xenon lamp
ballast, was chosen as a suitable power source of the electromagnet and produced coil
currents of 17 to 100 A (Figure 1 A). In approximate proportion to a current, the magnetic
flux density of 0.3 to 1.7 T was produced with the large polar face, as measured with a digital
gaussmeter(501,Nihondenjisokki, Tokyo, Japan), when the distance between the two polar
faces was 20 mm. At maximum, about 2 T was produced with the small polar face under the
same conditions. The magnetic flux density regulated at 1.5 T at the center of the large polar
face was found to be distributed homogeneously within 45 mm from the center (Figure 1B).
A similar distribution of a magnetic flux density of 1.8 T was observed within 18 mm from
the center of the small polar face. Therefore, culture dishes were placed within these distances
for testing the effects of magnetic fields on cell functions. The electromagnet was fitted with
a water cooling system, so the temperature on the surface of the polar face did not exceed
50°C.
Time-varying magnetic fields were produced by automatic switching of the power
source with an electronic device, which enabled us to change the magnetic fiux density
from 0.07 to 1.7 7 and 1.54 T, respectively, in about 1 s and in the reverse direction in
3 s when the pole pieces with the small and large faces were used. For simplicity, we
took the durations of the switching on and off times, i.e., "on-time" and "off-time", to
be equal. These durations were chosen to be 3 s or longer, since it took less than 3 s for
the eddy current to return to zero after the switching off. as judged from the results in
Figure 2. Because of this sufficiently long interval, the eddy current occurring on switching
off in a preceding cycle of changes in the magnetic flux density did not overlap the

1.S

0.5

c
25 50 75 100
Current (A)
diameter (mm)
B 100
f 50

Figure 1. Characteristics of the electro-

magnet. A, Relation between the magnetic
flux density and current of the power
20 40 60 80 source. B. Distribution of the magnetic flu.x
Distance from the center (mm)
density on the polar faces.
Effects of Electromagnetic Fields on K (Rb ) Uptake by HeLa Cells 103

<
E
40r

20
o

m
0 0--1

-20
0 5 T 10 0 5 T 10
Duration (sec) Duration (sec) >

Figure 2. Changes in the magnetic flux density and the density of the eddy current. A and C, Magnetic flux
densities with the small and large polar faces, respectively. B and D. dB dl and mean values of the eddy current
corresponding to A and C. [Yamaguchi ct al., 1992]

current induced by switching on in the following cycle. Panels A and C of Figure 2 show
examples of the quasi-rectangular forms of changes in the magnetic flux density with
"on-time" and "off-time" periods of 5 s, when the pole pieces of the small (50 mm
diameter) and large faces (100 mm diameter) respectively, were used. The inagnetic flux
density changed from 0.07 to a maximum of 1.77 T in A and to that of 1.54 T in C.
Panels B and D show the spike-like changes in dB/dt and mean values of the eddy current
densities corresponding to the changes in the magnetic fields in panels A and C. The
eddy currents induced by switching on were greater than those induced by switching off
The cycles of changes in the magnetic flux density were estimated to be 6.5 and 8.4 s
from the results in panels B and D. Since the density is proportional to the horizontal
distance / from the center of the culture surface and a change in the magnetic flux density
d5/d/, the density of the eddy current / was calculated by the following equation:

;=y(/-/2)d5/d/, (1)

where y is the conductivity (1.59+0.82 S/m) of the culture medium. The mean value
of the current density in a culture dish was obtainable by the equation.

JJ irdr/l rdr=Y{R:'3)dB!dr, (2)

where W is the radius of the culture surface (16.9 mm).

104 H. Miyamoto er al.

FREQUENCY SPECTRA OF B AND dBIAt

We analyzed the amplitudes of fundamental waves of the magnetic flux density B

and its differential ABIAt as functions of the angular frequency M. Since the magnetic field
used for these analyses changed between nearly 0 and 1.6 T at a cycle Tof 6 sec,

./o=l/r=l/6[Hz]

and

a)o=27t/,| = 27t/r=Tc/3 [rad/sec].

Let

S(y)=ao+a| cos( C0fl/+ e,)+a2Cos(2coQ/+92)+a3Cos(3co„/+e3)+ , (3)

where 9„ indicates the phase of a certain component at time 0 (/ =0). Then, the
amplitudes of the components are regarded as

aipFCOyr, a,=2F(co„)/r. a.=2F(2w(>)/T, a^=2F{3o)„VT.

F{m) corresponds to the frequency of change in the magnetic field /(r) at any time.
We can also represent d5/d/ by a similar expression

d5(0/d/=bo+b,cos( a)o'+ 6|)+b2Cos(2coo/+92)+b3COs(3c)or+63)+ , (4)

where b^ corresponds to a^ in equation (3).

These amplitudes for the magnetic flux density B and its differential dB/dr were
calculated as functions of the angular frequency co and shown in Table 1 and Figure 3A, and
Table 2 and Figure 4A, respectively. Figures 3A and 4A represent different patterns of
changes in the amplitudes between B and dB/dt with increase in co. The amplitude of fi is
relatively large at co=0 and coO, but decreases abruptly with increase in co after 2coO. In
contrast, the amplitude of dB/df is extremely small at 00=0 but becomes maximal at coO. and
then gradually decreases with increase in co. However, the amplitude cannot be ignored even
at higher values of co.

Table 1. Amplitudes of the components of S, i.e., 2F{w)/r,

calculated from the curve of S as a function of time
(0 (rad/sec) 2F(M)/T(Tesla)
0 0 0.894
w„ 1.047 0.852
2a)o 2.094 0.147
3cOo 3.142 0.203
4coo 4.189 0.081
5(0u 5.236 0.097
6(1)0 6.283 0.050
7wu 7.330 0.059
8(.)(, 8.378 0.033
9co„ 9.425 0.033
lOwo 10.472 0.021

=2K/T (T=6 sec).

Effects of Electromagnetic Fields on K*(Rb*) Uptalie by HeLa Cells 105

Table 2. Amplitudes of the components ofdB'dt, i.e., 2F(w)/7",

calculated from the curve of d5/dr as a function of time
CO (rad/sec) 2F(a))/T (Tesla/sec)
0 0 2.203
0J„ 1.047 28.833
2o)„ 2.094 9.512
3o)o 3.142 15.039
4on 4.189 10.378
5wn 5.236 10.511
6(o„ 6.283 10.142
7M„ 7.330 8.314
8co„ 8.378 10.125
9(o„ 9.425 6.330
10(0u 10.472 9.655

=2;rT(T=6sec)

Wc also show B and dS/dras functions of the frequency fin Figs. 3B and 4B. Similar
patterns of change in the amplitudes to those shown in Figs. 3A and 4A are still observed.
The amplitude of the components of B is relatively larger at/=0 and 1/6 Hz and decreases
abruptly after/=2/6 Hz, whereas the amplitude of the components of d5/d/ tends gradually
to decrease with increase i n / These results revealed that B mainly consisted of components
of the lowest frequencies, whereas d5/d/ as well as the eddy currents had more complicated
frequency spectra.

CELL CULTURE

HeLa cells (strain S3) purchased from ICN Biomedicals Inc. (Costa Mesa. CA) were
maintained by serial culture in glass culture flasks containing 10 ml of culture medium
consisting of modified Eagle's minimum essential medium (mMEM) enriched with amino
acids and vitamins [Miyamoto et al., 1976] and supplemented with 10% (v/v) calf serum.
Growing cells were washed with Dulbecco's phosphate-buffered saline (PBS), detached
from the culture flasks and dispersed by digestion with 0.5% trypsin (1:250, Difco Labora-
tories, Detroit, MI), and suspended in the same culture medium at a density of 5x 10"'cells/ml.
Aliquots of 2 ml of the cell suspension were inoculated into plastic culture dishes (35 mm
diameter. Corning Glass Works, Corning, NY) and incubated for 48 h at 37 ° C in an incubator
under humid air containing 5% COj.

CATION FLUX ASSAYS

The culture medium was discarded and the cultures were once washed with mMEM
containing 5 mM RbCl (super-pure grade, Merck, Darmstadt, Germany) instead of 5 mM
KCl (Wako Pure Chemical, Osaka, Japan) and incubated in the same medium containing
RbCl. This medium also contained 25 mM N-2-hydroxy-ethylpiperazine-N'-2-ethanosul-
fonic acid (HEPES, Sigma Chem. Co., St. Louis, MO) to maintain the pH at 7.2. These
procedures were carried out in a special incubator in which the temperature of the air could
rapidly be changed to that required for exposure to the magnetic fields. Rb* was used in
place of K* because the membrane transport of HeLa cells recognizes Rb* and K* equally
[Miyamoto et al., 1978]. Rb"^ influx was assayed for 15 min immediately after medium
106 H. Mivamoto et al.

B
0.8

0.6

^ 0.4

0.2

Figure 3. Amplitudes of the components offl.

0.5 1 .5 .^. The amplitude as a function of to, B. The
f [Hz] amplitude as a function of/.

change, unless otherwise stated, because both active and passive Rb"" influxes take place at
the initial rates for at least 20 min [Miyamoto et al., 1986; Ikehara et al.. 1982 and 1993].
K* efflux was assayed as time-dependent decrease in the cellular K"^ content after
replacement of K* by Rb* (see Figure 11). The rate constant of K* efflux, i.e.. the rate constant
of decrease in cellular K.*, was estimated from the slope of the linear part of the semi-loga-
rithmic plot of the cellular K* content vs. the time after replacement of K*-medium by
Rb*-medium. K^ efflux, i.e., the rate of cellular K* loss, was calculated by multiplying the
rate constant bv the cellular K* content at the beg inning of the assay. Therefore, Rb^ intlux
and K* efflux could be assayed concurrently. Cation fluxes are expressed in nmol'mg
proteinmin.

CHEMICAL ASSAYS

When these flux experiments were finished, the cultures were washed six times with
cold 0.15 M LiCl in 15 s. Volumes of 3 ml of cold deionized and distilled water were added
to the culture dishes, and then the cells were detached from the culture dishes and dispersed
in the dishes with a silicone-rubber policeman. Aliquots of 2 ml of the cell suspension were
sampled and mixed with 2 ml portions of 30 mM LiCl. These mixtures were kept at least
overnight at room temperature and then their cellular contents of Na*, K"" and Rb^ were
Effects of Electromagnetic Fields on K (Rb ) Uptake by HeLa Cells 107

3 0

2 0

^ 1 0

4 8 1 2

u) [rad/s]
25
B
20

t 15

3 10
ft
5
Figure 4. Amplitudes of the componenls of
dB/di. A. The amplitude as a function of w. B. 0
0.5 1 1.5
The amplitude as a function of/
f [Hz]

determined with a flamephotometer (170-30, Hitachi Ltd., Tokyo, Japan) with 15 inM LiCl
as an internal standard. Cellular contents of cations are expressed in nmol/mg protein.
Other samples of 1 ml of cell suspension were mixed with 1 ml of 1 N NaOH for
protein assay by the method of Lowry et al. (1951) with bovine serum albumin (fraction V)
as a standard.
For assay of the cellular ATP content, 2 mi portions of cell suspension in cold water
prepared by the same procedure to those described above were incubated for 5 min in a
boiling water bath to extract ATP, and ATP was measured by the luciferin-luciferase reaction
in a luminescence photometer (Monolight™ 500, Analytical Luminescence Laboratory, San
Diego, CA) as described previously [Ikehara et al., 1990]. The cellular ATP content is
expressed in nmol/mg protein.

CULTURE TEMPERATURE REGULATION

We designed special incubators to keep the temperature of cell cultures strictly
constant during assay of cation fluxes on exposure to an electromagnetic field. One of the
incubators fit for the large polar face is shown in Figure 5. This incubator made of copper
plate (1.5 mm thick) was of 129 mm diameter and 20 mm thickness, so that it could be placed
in the 20 mm gap between two polar faces. The upper portion of the incubator consisted of
four chambers of 37 mm diameter and 13 mm depth to contain four plastic culture dishes.
This portion was covered with a plastic lid (I mm thickness) with four round pads of styrene
foam (35 mm diameter and 3 mm thickness) to cover the culture dishes for adiabatic
purposes. Four thermistor sensors (I mm diameter) connected to a thermometer (D1I7,
108 H. Mivamoto et al.

Upper Face Lower Part

FBT m
Vertical Section
Figure 5. Schematic presentation of a special incubator to maintain the temperature of cultures. B, water bath;
C, cell culture; D, culture dish; L. plastic lid; S, styrene foam; T, thermistor sensor. [Yamaguchi et al., 1992]

Takara Indust. Co., Tokyo, Japan) were inserted to the bottoms of the culture dishes through
pin-holes in the lid and the pads for rnonitoring the temperatures of cultures continuously
during assay of cation fluxes. The lower portion of the incubator consisted of a water bath,
through which water flowed in the directions indicated by arrows and distributed heat evenly.
So, differences in temperatures between any two dishes did not exceed 0.2° C and heating
of the culture dishes by the hot polar faces was prevented. Normally, the temperature of the
dishes was adjusted to be 37° C, but it could be changed from 0 to 50° C. One of the incubators
was placed in the electromagnet and the other was kept outside the magnetic field as a control.
We also made the similar incubators with a smaller diameter of 80 mm to contain only one
dish to fit the small polar face. These incubators were normally used in combination with
the air incubator described before for adjusting the temperature during medium replacement.
This was because it took about 5 min to increase the temperature, from, for example, 37 to
40° C in these incubators, and this delay would influence Rb"^ influxes determined in 15 min.

MICROFLUOROMETRY
The membrane potential of the cells was determined by a modification of the method
of Wright et al. [ 1981 ]. We applied micro-fluorophotometry to cells loaded with a fluorescent
indicator, 3,3"-dipropylthiadicarbocyanine iodide (diS-C3-(5)) (Molecular Probes, Eugene,
OR), keeping the intracellular K* concentration at 139 mM. The cyanine dye was added to
the culture medium at 1 ^M immediately after 2 h exposure to the time-varying magnetic
field. For determination of fluorescence, the excitation wavelength was fixed at 540 nm and
emission was determined in a wide wavelength range from 580 to 680 nm to collect sufficient
energy of fluorescence with a modified fluorescence microscope (Optiphoto, Nikon, Tokyo,
Japan) equipped with a photon counter (PC-545A, NF Circuit Design Block Co., Yokohama,
Japan). The membrane potential was calculated as the K* equilibrium potential in the
Effects of Electromagnetic Fields on K*(Rb*) Uptake by HeLa Cells 109

A: Na+, K«-pump
B: NaVK>/2C|- cotransport
Figure 6. Major pathways of K* up- Q. K* channel
take in the membrane of HeLa cells.

presence of a K* ionophore (4 )iM valinomycin), and the potential was related to the emission
intensity. When 150 mM K"^ was needed, an impermeable anion gluconate was used instead
of CI' as a counter anion to prevent rupture of the cell membrane due to strong cell swelling.
Micorphotographs of cells loaded with the fluorescent dye were taken with the same
fluorescence microscope in the same emission wavelength range at a magnification of 200x
(see Figure 13).
We also monitored changes in the electrical charge on the cell surface with the
fluorescent pH indicator 4-heptadecyl-7-hydroxycoumarin by a modification of the method
of Pal et al. [1983]. Immediately after adding this indicator to the culture medium at 6 pM.
we exposed the cultures to the magnetic field for 2 h, and then determined total fluorescence
energy in an einission wavelength range from 420 to 680 nin with the fluorescence
microscope at excitation wavelengths of 340 and 380 nin. The ratios of the emissions at these
wavelengths, E340/E380, were calculated and the ratios of exposed and control cells were
compared.

HOMOGENEOUS FIELD EFFECTS ON Rb^ INFLUX

Figure 6 is a schematic presentation of major pathways of K* influx through the
membrane of HeLa cells, consisting of an Na-pump [Ikehara et al., 1984a: Ikehara et al.,
1984b], a NaVKV2Cl- cotransport pathway [Miyamoto et al., 1986; Ikehara et al., 1990 and
1993] and a Ca^"^-dependent K'^ channel of the relatively small conductance [Ikehara et al.,
1991; Takahashi et al., 1993]. About 50-70% of the total Rb* uptake was inhibited by 10
mM ouabain and about 70-80% of ouabain-insensitive Rb* uptake was inhibited by 0.1 mM
furosemide, indicating that about 50-70% of the Rb* uptake was mediated by the Na-pump,
and about 20-40% by the cotransport pathway. The fraction of the total K* influx mediated
by K* channel was not determined, since K'^ permeability could be measured exactly only
by the patch-clamp technique.
Figure 7 shows the effects of various magnetic flux densities of homogeneous
magnetic fields from 0 (the level of terrestrial magnetism) to 1.6 T on active and passive Rb*
influxes of HeLa cells at two different temperatures. No significant effects of the magnetic
fields of different magnetic flux densities on total Rb'^ influx were detected without 100 fiM
ouabain but passive influx was observed in the presence of the inhibitor (ouabain-insensitive
influx) at a normal temperature of 37.4° C (panel A, t-test at p>0.05). Therefore, the
difference between the total and ouabain-insensitive Rb* influxes (ouabain-sensitive, active
110 H. Mivamoto ct al.

Rb+ influx
C Total,
A lOOmM ouabain
3 Ouabaln-sensitive

20 ri^
15

10 Q-
^S=%
I 5

i 0
0.5 1.5
1
B
25

20 Lu
'l, I 1 i. ^
6
15 a n
d 0
10 A
V
5 t Figure 7. Effects on Rb'^ influxes into
1 ' HeLa cells of homogeneous magnetic fields
0 1 1
0.5 1 1.5 of various ttiagnetic flux densities from 0 to
1.6 T. A. . B. (42.0+0.1 )°C.
Magnetic (lux density (T)

Rb^ influx
O Total, Control, A lOOmM ouabain, n Ouabain-sensitiV'
Control Control
Total, Exposed
lOOmM Ouabain, Ouabain-sensitiv
Exposed Exposed
2.6 r

3.2 3.3 3.4 3.5

Temperature (*C) 1/T (1000/*K)

Figure 8. Effects of a homogeneous 2 T magnetic field on Rb* influxes at different temperatures. A. Total,
ouabain-insensitive and ouabain-sensitive Rb* influxes. B. Arrhenius plots of data in A.
Effects of Electromagnetic Fields on K (Rb ) Uptake by HeLa Cells 111

influx mediated by the Na-pump) was unaffected. Data in this panel are mean values of
results in six separate experiments expressed in arbitrary units. As a homogeneous magnetic
field of even 2 T did not influence Rb* influx in the absence or presence of ouabain at 37°
C (see Figure 8), homogeneous magnetic fields did not markedly influence the active and
passive Rb+ influxes when the magnetic flux density was less than 2 T. We also tested the
effects of homogeneous magnetic fields of similar magnetic flux densities on ouabain-in-
sensitive Rb"* influx in the presence of 0.1 mM furosemide at 37° C, but again observed no
significant effects of the magnetic fields (data not shown). These results reveal that magnetic
fields of less than 2 T do not significantly influence Rb* influx mediated by the Na*/K*/2C1"
-cotransport or the leakage pathway(s) at normal temperature.
Similarly, no significant effects on Rb* influx of homogeneous magnetic fields of the
same magnetic flux densities were detected at an abnormally high temperature of 42° C
(Figure 7B; t-test at p>0.05). Data in this panel are mean values of results in two separate
experiments.
Kinouchi et al. [1988] have reported theoretical estimations of the Lorentz force for
suppressing diffusion of charged particles such as Na*, K*, Ca-* and CI", and plasma proteins.
They pointed out that the threshold field strength for suppression is so high, (more than lO"*
T) that the Lorentz force does not affect the diffusion of these particles at a magnetic flux
density of a few T. Their estimation is consistent with our finding of no significant effects
on passive Rb* influx at less than 2 T.
Figure 8A shows the effects on active and passive Rb* influxes of HeLa cells of a
homogeneous 2 T magnetic field over a wide range of culture temperatures of 10 to 45°C.
Rb* influxes without ouabain increased linearly with increase in temperature from 10 to
37°C, and decreased at temperatures above 40° C in both control cells (exposed only to the
terrestrial magnetic field) and cells exposed to the magnetic field (upper panel). No
significant differences between the influxes into control cells and those exposed to the
magnetic field were observed at any temperature tested (t-test, p>0.05). Ouabain-insensitivc
influx of about half the total influx increased with temperature from 20 to 37° C. but there
was no significant difference between the values in cultures in control and tested conditions.
Ouabain-sensitive Rb* influx also increased with increase in temperature from 10 to 40° C,
but unlike the ouabain-insensitive influx, it did not decrease markedly at 45° C (lower panel).
Again, there was no significant difference between ouabain-sensitive Rb* influx in control
cells and those exposed to the magnetic field in the temperature range tested. The marked
decrease in Rb* influx without ouabain above 40°C was mainly due to decrease in passive
flux at abnormally high temperatures.
These results show no significant effects on Rb* influx of a strong homogeneous
magnetic field of up to 2 T, implying the absence of effect of steady magnetic fields with a
magnetic flux density of less than 2 T on Rb* influx.
The data in Figure 8 A are represented by Arrhenius plots in Figure 8B. The plots for
ouabain-insensitive Rb* influxes of control cells and those exposed to a magnetic field show^
linear distributions of experimental points from 10 to 37°C along the regression lines
obtained by the least squares method, but above 37°C and especially above 40°C the points
markedly deviated downward from the lines (upper panel). From the slopes of these
regression lines, we estimated the activation energies for ouabain-insensitive (passive) Rb*
influxes to be 60.5 and 51.7 kJ/mol in control cells and cells exposed to the homogeneous
2 T magnetic field, respectively. Statistic analysis verified that the differences between the
variances of the data (F-test), the regression coefficients (t-test) and the intersections of the
regression lines with the ordinate (t-test) in control cells and those exposed to the magnetic
field were all insignificant (p>0.05). We also compared the regression lines for ouabain-sen-
sitive Rb* influxes (lower panel) and found that none of the three parameters were signifi-
112 H. Miyamoto et al.

Table 3. Effects on Rb* influx of strong, time-varying magnetic fields of different

frequencies. [Yamaguchi et al., 1992]
Frequency On- ,Off-time
(Hz) (sec) Rb* Influx (nmol /mg protei n /min) n
Total Passive Active
Control 2 9.5+1.9 10.111.0 7
1:20 10 * 0 * 5
1/10 5 2 7 7 7
1,'6 3 14.5+3.5** 7.8+1.0 6.9+3.6** 7

Values are means+SD. n, number of samples. * and **, Significantly different from values for
control cells at p<0.05 and p<0.01, respectively.

canity different in control and test cells. The activation energies estimated from the regression
lines were 54.9 and 51.4 kJ/mol for control and test cells, respectively.
Tenforde and Liburdy [1988] reported that a membrane consisting of phospholipid
bilayers is markedly influenced by externa! magnetic fields near the phase transition point.
They observed increase in efflux of a chemical agent from liposomes at a ternperature
(40.5°C) close to the phase transition point of the phospholipid on exposure to static magnetic
fields, and found that the effect of the magnetic field was saturated at 0.1 T. Liburdy and
Vanek, Jr. [1985] demonstrated significant increase in permeability to Na"" of red ceil
membranes on exposure to microwaves in the narrow temperature range of 17.7 to 19.5°C.
However, a phase transition point has not been observed in the membranes of living cells of
other than red cells, possibly due to the more complex components of cell membranes,
including various proteins, than of artificial membranes. Consistent with this idea, the
temperature-dependent characteristics of cell membranes are reported to be rnore compli-
cated than those of membranes consisting of pure organic substances [Aceto et al., 1970].
The results in Figure 8 do not support the possibility of membrane phase transition in HeLa
cells between 10 and 37°C.

TIME-VARYING FIELD EFFECTS ON CATION FLUXES

Table 3 shows the effects of the frequency of time-varying magnetic fields of the type
presented in Figure 2A. The effects on Rb'^ influx of magnetic fields with different durations
of "on-time" and "off-time" up to 5 min were tested. However, when these durations were
longer than 30 s. i.e., the frequency was less than 1/60 Hz, there was no significant infiuence
of exposure on any fraction of Rb^ influx (data not shown). With decrease in the durations
to 10, 5 and 3 s, corresponding to frequencies of 1/20, 1/10 and 1/6, respectively, total Rb*
influx without ouabain tended to decrease. There were significant differences between the
influxes into control cells and those exposed to magnetic fields at frequencies of 1/20
(p<0.05) and 1/6 (p<0.01) by Dunnett's t-test. Passive Rb* influx in the presence of ouabain
also tended to decrease with increase in the frequency, but the change was not significant.
Ouabain-sensitive, active Rb* influx was significantly inhibited by exposures to magnetic
fields at these frequencies (p<0.05). Only active Rb"^ influx at 1/10 Hz was not significantly
inhibited in this experiment, probably because of the large coefficient of variance. In another
experiment, in which the coefficient of variance at this frequency was smaller (data not
shown), decrease in Rb* influx was demonstrated to be significant (p<0.05). A frequency of
1/6 Hz was used in following experiments, unless otherwise stated.
Effects of Electromagnetic Fields on K*(Rb*) Uptake by HeLa Cells 113

# no inhibitor, Control
no inliibitor. Exposed
A 1 OOpM ouabain, Control
,^ tOOuM ouabain, Exposed

Figure 9. Effects of exposure to a strong, time-varying mag-

netic field on Rb* accumulation. [Yamaguchi et al., 1992]
Time (min)

For further investigation of the effects of time-varying magnetic fields, the time of
Rb"" accumulation in the cells was prolonged to 2 h and the accumulations in control cells
and those exposed to a magnetic field were compared (Figure 9), The accumulations of Rb*
in control and exposed cells without ouabain were not markedly different in the first 15 min.
This insignificant difference was not consistent with the significant effect in Table 3. because
we used the type of magnetic field shown by Figure 2C (type 2C), which would induce a
weaker eddy current than that induced by the magnetic field of type 2 A used for the results
in Table 3. Since we needed more samples for the present experiment, we used the special
incubator containing four culture dishes with pole pieces with a face of 100 mm diameter.
The accumulafion was significantly inhibited after 30 min, as judged by Dunnett's t-tesi of
one-way layout after analysis of variance (ANOVA test. p<0,05). The inhibitory effect of
the magnetic field became more significant with time until at least 120 min. In contrast, Rb*
accumulation in the presence of ouabain was not significantly infiuenced by exposure to the
magnetic field. Hence, the inhibition of total Rb* accumulation was due to inhibition of
ouabain-sensitive Rb'^ accumulation. These results showing that active, but not passive Rb*
influx is inhibited are essentially consistent with those in Table 3.
As we found that the Na*-pump was inhibited by exposure to the magnetic field, we
next tested the effect of exposure to the magnetic field on the cellular ATP content (Figure
10), Results showed that the ATP contents of cells whose Rb'" accumulations are shown in
Figure 9 were not significantly influenced by exposure to the magnetic field for periods of
up to 120 min, irrespective of the presence of ouabain.
The effects of exposure to the magnetic field on K"" efflux were also tested. The K^
contents of cells exposed for various periods after replacement of K"^ by Rb* in the medium
were plotted on a semi-logarithmic scale and rate constants were determine by the procedures
described before (Figure 11). The plots for both exposed and control cells were linear with

no inltibitor, Control
no intiibilor, Exposed
40 100)iM ouabain. Control
I O O M M ouabain, Exposed

30-

20

<
s
Figure 10, Effects on the cellular ATP content of exposure o
to a strong, time-varying magneticfield.[Yamaguchi et al., O"-
30 BO 90 120
1992] Ttne (min)
114 H. Miyamoto et al.

Control
Exposed

Figure II. Effect on K* efflux of exposure to a

strong, time-var>'ing magnetic tiekl [Yamaguchi
et al.. 1992]

time, but the slope of the regression line for exposed cells was steeper than that for control
cells, and the rate constants for control and exposed cells were determined to be 0,0071 and
0.0105/min, which were significantly different by the t-test (p<0.01). Since the cellular K*
content at the start of exposure was 932 nmol/mg of protein, the K"' effluxes frorn control
and exposed cells were calculated to be 6.66 and 9.79 nmol/mg proteinmin. These results
demonstrate stimulation of K* efflux by exposure to a magnetic field and suggest exposure-
induced decrease in the sum of the intracellular K* and Rb* contents; i.e.. the K"' content of
cells in the nonnal medium.

TIME-VARYING FIELD EFFECTS ON CELLULAR ELECTRICAL

PROPERTIES
We tested the effects of exposure to a time-varying magnetic field on the electrical
properties of cells, using a fluorescent probe of the membrane potential, diS-C3-(5). Results
showed that the fluorescence emission of cells loaded with the dye and exposed to the
magnetic field for 2 h was significantly weaker than that of control cells (Table 4), suggesting
a marked decrease in the emission of dye molecules attached to the cell surface or in the
cells. Inhibition of the Na-pump (Figure 9), decrease in the cellular K* content (Figures 9
and 11) and increase in the cellular Na* content (data not shown) on exposure to the
time-varying magnetic field also support the possibility of the occurrence of depolarization.
Therefore, the exposure might cause extreme depolarization of the cell membrane.
To test this possibility, we examined the ratios of the emissions of cells incubated in
media with various K^ concentrations in the presence of valinomycin to the emission of cells
in normal medium without valinomycin to determine whether the ratio was proportional to
the logarithm of the K* concentration (Figure 12). Result clearly demonstrated a linear
relation of the emission ratio of control cells with the logarithm of the K"" concentration, i.e..

Table 4. Effect of exposure to a strong, time-varying magnetic field on

emission of HcLa cells loaded with the fluorescent indicator of membrane
potential diS-C3-(5). [Yamaguchi et al., 1992]
Experiment Emission Intensity (arbitrary units)
Control (n) Exposed (n)
1 7 (38) * (30)
11 l*t).477 (53) 0.280+0.176* (37)
Values are means SD. n, number of cells determined. *, significantly different at
p<0.001.
Effects of Electromagnetic Fields on K*(Rb*) Uptake by HeLa Cells 115

0 >
E

Figire 12. Effect of exposure to a strong, time-

varying magnetic field on emission of cells loaded
with the fluorescent indicator cliS-C3-(5) and in^
cubatcd in media with various K* eoncentrations
0
in the presence of valinomycin. [Yamagiichi ct a l ,
1992] 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6
Fluorescence ratio Carbitrary ynlts)

the membrane potential. Thus the dye acted as an indicator of the potential in control celts.
In contrast, the emission ratio of exposed cells did not respond to a wide change in tlie K*
concentration in the medium, and the intensity of emission was again very weak. This
suggests that the decrease ia the emission intensity of exposed cells was not due to membrane
depolarization, but probably to an exposure-induced change in electrical charge on the cell
surface, since the dye molecule has a net negative charge,
Microphotograplis of cells loaded with a fluorescent dye taken with the fluorescence
microscope are shown in Figure 13. In control cells (panel A) the fluorescent dye has a more
homogeneous distribution than in exposed cells (panel B), in which the distribution of dye
is uneven and has a granular appearance. This uneven staining with the charged dye suggests
change in the distribution of electrical charge on the celi surface on exposure.

Figure 13. Micorphotographs of HeLa cells loaded

with the fluorescent indicator diS-C3-(5) taken witli a
fluorescence microscope at a magnification of 200
(objective lens 40x). A, Control cells. B, Exposed
cells.
116 H. Mivainoto et al.

Control

iS Exposed

l-lgure 14. Effcccs cf exposure to n >!rc!i"g, dnie-varyiisg

iiiitgneiic field ors eiiiissiim of celh. k'.aded v.'ilh Uie IlLuires-
cent pH indicaror 4-isep!ade>';y!-7-hydrox;'-cCiiiinaiio. Nii-
nieraii in columns indicate luiiabsfs of cciis incasiircci
*Signi}!caiitiy dif&rcrit from vsli:c of eonrro! ceiis at
p<ll!)L [^"EiTsagiiLh! el aL. 1992]

Next we tested the effects of the time-varying magnetic field on the electrical charge
on the cell surface with the fluorescent pH hidicator 4-heptaclecy!-7-hydroxycoiimariii. The
ratio of emissions E340/E380 decreased with increase in pH of the culture medium from 6
to 8 (Figure 14). This resiih implies that the positive charge on the cell surface increases with
increase in pH of the medium. This decrease in the emission ratio with increase in pH of the
medium was also observed in cells exposed to the magnetic field. However, the exposed
cells showed significantly higher values of the emission ratio at all pH values tested (p<0.01
by Dunnett's 1-test after ANOVA test).
The wavelengths of 340 and 380 nm are near the peaks in the spectra of the
undissocaied and dissociated forms of the fluorescent reagent, respectively (Pal et al, 1983).
Therefore, the increase in the emission ratio E340/E380 with decrease in pH of the medium
represents a relative increase in the undissocatcd form of the reagent with decrease in pH in
both control and exposed cells. Exposure to the magnetic field seemed to inhibit the
pH-depcndciit dissociation of li* from the reagent attached to the cell membrane by
increasing the negative charge on the cell surface. These results on the emission ratio arc
consistent with those shown in Table 4 and Figure 13. Thus we concluded that exposure to
the magnetic field induced a relative increase in the negative charge on the cell surface.
An AC current has been demonstrated to decrease the activity of Na^.K^-ATPase in
suspension [Blank and Soo, 1989J. On the other hand, there is a report that exposure to an
ELF electric field increases the negative charge and exposure to an ELF magnetic field
decreases the hydrophobicity of thesurface of Physarumpolycephalum [Marron et ah, 1988],
Also, exposure of cultured U937 cells to a pidsed magnetic field of repeated bursts of 2.5 Hz
has been shown to increase the negative charge on the cell surface [Smith et ah, 1991].
Electro-osmotic flow of medium occurs when an electric field is applied in parallel to the
cell surface, and negatively charged macromolecules are moved on the cells surface and
accumulated on one side of the cell as demonstrated by the accumulation of concanavalin
A receptors on the cathodal side of muscle cells [McLaughlin and Poo, 1981].

_ 1.2
.2
15 ^
_> -—

C.8

I 0S
Figure 15, Reversibility of the effect on Rb* influx of
exposure for 90 inin to a strong, time-varying magnetic
0.2f
field. A, Exposed. B, Exposed and then placed oiit.side the
magnetic field. C, Control.
Effects of Electromagnetic Fields on K*(Mb*) Uptake by HeLa Cells 117

We observed significant iohibitioB of the Na-pump by the strong, time-varying

magnetic field and increase in the negative charge on the surface of IleLa cells. As an
electrical force acting on a charge on the cell surface is proportional to the imposed current
density, the magnitude of the eddy current would play an important role. In fact, we a showed
stronger effect of the magnetic field at higher magnetic flux density (compare the results in
Table 3 and Figure 9). The frequency of change in the field is also important, as shown in
Table 3. Inhibition of the Na-pump by exposure to the time-varying magnetic field would
have a close relation to the eddy current or change in the electrical properties of the cell
surface caused by the current.

RE¥EESIBILITY OF TIME-VARYING FIELB EFFECTS

Finally, wc tested the reversibility of ihe effeets of exposure to the strong, time-var}'-
iog magnetic field on Rb* influx and change in the cellular electrical properties.
Column A of Figure 15 shows Rb* influx into exposed cells in 15 min in Rb* medium
after exposure in normal K*-medium for 75 min (total exposure time, 90 min). Column B
shows that of cells iocubated in Rb*-medium after exposure for 90 min in K^-medium, then
incubation 30 min in K*-medium outside the magnetic field. Column C shows Rb* influx of
control cells. Values are means and SD of Rb* influxes in seven independent experiments,
four dishes being used in each experimental group. Mean Rb"^ influx of exposed cells (column
A) was significantly smaller than that of control cells, as judged by Dunnett's t4est after the
ANOVA test (p<0.01), whereas the mean Rb* influx shown in column B was not (p>0.05).
Similar resuhs were obtained in other experiments. Therefore, the inhibition of active Rb*
influx caused by exposure to the magnetic field was reversible.
Figure 16 shows the reversibility of the effect of the magnetic field on fluorescence
emission of cells loaded with diS-C3-(5). Column A shows the emission intensity of cells
exposed for 90 min, and columns B and C show those of cells exposed for 90 min and then
placed outside the magnetic field for 30 and 60 min, respectively. Column D shows the
fluorescence emission of control cells. The emission intensities shown by columns A and B
were significantly lower than that of control cells according to Dunnett's t-test (p<0.01), but
that shown by column C was not (p>0.05). Therefore, the effect of 90 min exposure was
reversed after 60 min incubation outside the magnetic field.
The results in Figures 15 and 16 demonstrate the reversibility of the effects of the
strong, time-varying magnetic field on both Rb* influx and fluorescence emission of
diS-C3-(5). These results support the idea that inhibition of the Na-pump by exposure to the
magnetic field is closely connected with change in the electrical properties of the cell surface.
However, further investigations are needed for detemiine whether the change in the electrical
properties is directly connected to inhibirion of Na-pump activity, because, the effects of

.5-1-5
c
3

t
I 1
1
Figure 16. Reversibility of the effect of exposure for 90 "-'
min to a strong, timc-var>-ing magnetic field on fluores- g o.5
ceiice emission of cells loaded with diS-C3-(5). A, Ex- "i
posed. B and C, Exposed and then placed outside the "g
magnetic field for 30 and 60 min, respectively, D. Control UJ „
118 H. Miyamoto et al.

exposure may be related to more substantial change in membrane structure than simple
change in the surface properties, since more than 30 min was required for complete recovery.

CONCLUSIONS
Exposure to strong homogeneous magnetic fields with various magnetic flux densi-
ties of less than 1.6 T had no significant effect on either active or passive Rb* influxes into
HeLa cells at normal or high temperatures. Exposure to a similar magnetic field of 2 T at
different temperatures of 10 to 45°C did not cause any change in active or passive Rb* influx,
and no evidence was obtained for the presence of a phase transition point of the cell
membrane between 10 and 37°C.
In contrast, exposure to a strong, time-varying magnetic field of quasi-rectangular
wave form caused significant inhibition of active Rb^ influx when the frequency of change
in the magnetic field was more than 1/20 Hz. Conversely, K* efflux was stimulated, but
passive Rb'^ influx was unaffected. Analyses of the amplitudes of the frequency components
of the time-varying magnetic field B and its differential ABlAt revealed that B mainly
consisted of components with the lowest angular frequencies, whereas dfi/dr contained
components of various frequencies. The inhibition of active Rb* influx was not due to change
in the cellular ATP content.
Results obtained by micro-fluorometry with fluorescent probes of the membrane
potential (diS-C3-(5)) and pH (4-heptadecyl-7-hydroxy-coumarin) showed change in the
electrical properties of the cell surface on exposure to the time-varying magnetic field.
Results suggested an uneven distribution of electrical charge and an increase in negative
charge on the cell surface. The inhibition of active Rb* influx and the change in electrical
properties of the cell membrane were reversible. Further studies are needed to determine
whether change in electrical properties is the direct cause of inhibition of the Na*-pump.

ACKNOWLEDGMENTS
We thank Dr. Elizabeth Ichihara for correcting the English and Mr. T. Masuya for
technical assistance. This study was .supported in part by a Grant-in-Aid for Scientific
Research on a Priority Area (No. 03202136) from the Japanese Ministry of Education,
Science and Culture.

REFERENCES
1. T. Gualtierotti (1964) Decrease of the sodium pump activity in the frog skin in a steady magnetic field.
Physiologist, 7, 150.
2. C. S. Collis, and M. B. Segal (I9S8) Effects of pulsed electromagnetic fields on Na* fluxes across stripped
rabbit colon epithelium. J. Appl. Physiol., 65, 124-130.
It. M. Hinsenkamp, P. Lheureux, D. Martins (1985) Transmembrane Na/K exchanges under electromagnetic
fields. Preliminary study on human erythrocytes. Reconstr. Surg. Traumat.. 19. 63-69.
4. R. W. Famdale, A. Maroudas, and T. P. Marsland (1987) Effects of low amplitude pulsed magnetic fields
on cellular ion transport. Bioelectromagnetics, 8, 119-143.
5. A. P. Stevenson, and R. A. Tobey (1985) Potassium ion influx measurements on cultured Chinese hamster
cells exposed to 60-Hz electromagnetic fields. Bioelectromagnetics, 6, 189-198.
6. H. Yamaguchi, T. Ikehara, K. Hosokawa, A. Soda, M. Shono, H. Miyamoto, Y. Kinouchi, and T. Tasaka
(1992) Effects of time-varying electromagnetic fields on K*(Rb*) fluxes and surface charge of HeLa cells.
Jpn. J. Physiol., 42. 929-943.
Effects of Electromagnetic Fields on K (Rb ) Uptake by HeLa Cells 119

7. H. Miyamoto. L. Rasmussen, and E. Zeuthen (1976) Recording of clonal growth of mammalian cells
through many generations. In D. M. Prescott ed. "Methods in Cell Biology". 13. 107-119. ."Xcad. Press.
Inc. New York.
8. H. Miyamoto. T. Sakai, T. Ikehara, and K. Kaniike (1978) Effect of Rb* substituted for K* on HeLa cells:
cellular content and membrane transport of monovalent cations, and cell growth. Cell Struct. Funct.. ?.
313-324.
9. H. Miyamoto. T. Ikehara, H. Yamaguchi, K. Hosokawa, T. Yonezu. and T. Masuya (1986) Kinetic
mechanism of Na*,K*.Cl'-cotransport as studied by the Rb* influx into HeLa cells: effects of extracellular
monovalent ions. J. Membrane Biol.. 92. 135-150.
10. T. Ikehara, T. Sakai, H. Miyamoto, and K. Kaniike (1982) Interrelation between membrane transport and
the contents of alkali metal cations in HeLa cells. Jpn. J. Physiol., 32, 13-24.
11. T. Ikehara. H. Yamaguchi. K. Hosokawa. A. Takahashi, and H. Miyamoto (1993) Kinetic study on the
effects of intracellular K* and Na* on Na*.K*.Cl' cotransport of HeLa cells by Rb* influx determination.
J. Membrane Biol.. 132. 115-124.
12. O. H. Lowr\'. N. J. Rosebrough. A. L. Farr. and R. J. Randall (1951) Protein measurement with the Folin
phenol reagent. J. Biol. Chem.. 193, 265-275.
13. S. H. Wright. S. Krasne. 1. Kippen, and E. M. Wright (1981) Na*-dependent transport of tricarboxylic
acid cycle intermediates by renal brush border membranes. Effects of fluorescence of potential-sensitive
cyanine dye. Biochim. Biophys. Acta. 640, 767-778.
14. R. Pal. W. A. Petri, Jr., Y. Barenolz. and R. R. Wagner (1983) Lipid and protein contributions to the
iTicmbrane surface potential of vesicular stomatitis vinjs probed by a fluorescent pH indicator, 4-hep-
tadecyl-7-hydroxycoumarin. Biochim. Biophys. Acta, 729, 185-192.
15. T. Ikehara, H. Yamaguchi. T. Sakai. and H. Miyamoto (1984a) Kinetic parameters and mechanism of
active cation transport in HeLa cells as studied by Rb* influx. Biochiin. Biophys. .Acta, 775, 297-307.
16. T. Ikehara. H. Yamaguchi. K. Hosokawa, T. Sakai, and H. Miyamoto (1984) Rb* influx in response to
changes in energy generation: effect of the regulation of the ATP content of HeLa cells. J. Cell. Physiol..
119,273-282.
17. T. Ikehara, H. Yamaguchi, K. Hosokawa, and H. Miyamoto (1990) Kinetic mechanism of ATP action in
Na*-K*-Cl- cotransport of HeLa cells determined by Rb* influx studies. Am. J. Physiol., 258. C599-C609.
18. T. Ikehara. A. Takahashi, H. Yamaguchi. K. Hosokawa, T. Masuya. and H. Miyamoto (1991) Regulatory
changes in the K*. CI' and water contents of HeLa cells incubated in an isosmotic high K*-medium.
Biochim. Biophys. Acta, 1068, 87-96.
19. A. Takahashi, H. Yamaguchi, and H. Miyamoto (1993) Changes in K* current of HeLa cells with
progression of the cell cycle studied by patch-clamp technique. Ain. J. Physiol., 265, C328-C336.
20. Y. Kinouchi. S. Tanimoto. T. Ushita. K. Sato. H. Yamaguchi, and H. Miyamoto 11988) Effects of static
magnetic fields on diffusion in solutions. Bioelectromagnetics, 9, 159-166.
21. T. S. Tenforde. and R. P. Liburdy (1988) Magnetic defonnation of phospholipid bilaycrs: effects on
liposome shape and solute permeability at prephase transition temperatures. J Theoret. Biol., 133.
385-396.
22. R. P. Liburdy. and P. F. Vanek, Jr. (1985) Microwaves and the cell membrane. II. Temperature, pla.sma,
and oxygen mediate microwave-induced membrane permeability in the erythrocyte. Rad. Res., 102.
190-205.
23. H. Aceto, Jr.. C. A. Tobias, and I. L. Silver (1970) Some studies on the biological effects of magnetic
fields. IEEE Trans. Mag. MAG-6, 368-373.
24. M. Blank, and L. Soo (1989) The effects of alternating currents on Na.K-.^TPase function. Bioelectro-
chem. Bioenerg., 22, 3 13-322.
25. M. T. Marron, E. M. Goodman, P. T. Sharpe. and B. Greenebaum (1988) Low frequency electric and
magnetic fields have different effects on the cell surface. FEBS Lett., 230. 13-16, 1988.
26. O. M. Smith. E. M. Goodman, B. Greenebaum, and P. Tipnis (1991) An increase in the negative surface
charge of U937 cells exposed to a pulsed magnetic field. Bioelectromagnetics. 12, 197-202.
27. S. McLaughlin, and M.-M. Poo (1981) The role of electro-osmosis in the electric-field-induced movement
of charged macromolecules on the surfaces of cells. Biophys. J.. 34. 85-93.
 8

EFFECTS OF EXPOSURE TO A 50 HZ
MAGNETIC FIELD ON MELATONIN IN RATS

Masamichi Kato' and Tsukasa Shigemitsu-^

' Department of Physiology

Hokkaido University School of Medicine
Sapporo, 060 Japan
- Abiko Research Laboratory
Central Research Institute of Electric Power Industry. Abiko
Chiba, 270-11 Japan

1. INTRODUCTION
There has been increasing concern over possible human health risks associated with
exposure to alternating current (AC) magnetic fields induced by transmission and distribu-
tion systems. To accurately assess the possible health risk, it is necessary to conduct a wide
range of studies with appropriate exposure systems both for animal and human subjects.
Because AC transmission lines are three phase, the magnetic field at ground level tends to
be elliptically polarized (Deno, 1976) and is represented fairly well by a rotating vector. In
the laboratory, this rotating vector can be simulated by the sum of horizontal and vertical
magnetic fields 90° out of phase. It is very important to allow the simultaneous exposure of
as many animals as possible. An exposure facility must be designed to fill these requirements.
Melatonin, which is synthesized in the pineal gland, has been assumed to have linkage
to the etiology of cancer in at least the following three ways; 1) melatonin itself is oncostatic,
2) melatonin enhances certain facets of immune function and 3) melatonin functions as an
inhibitor of the hypothalamic-pituitary-gonadal axis, hence it may reduce the availability of
hormones that are required for the growth of certain hormone-dependent breast, ovarian and
prostate cancers (Wilson and Anderson, 1990).
Welkeretal. (1983) reported that experimental inversion of the horizontal component
of the natural magnetic field during nighttime led to a significant decrease of pineal
serotonin-N-acetyltransferase (NAT) activity and melatonin content in rats. Since then
several papers have been published investigating the effects on pineal activity of geomag-
netic fields inversion (e.g., Olcese and Reuss, 1986; Olceseet al, 1985; Stehle et al., 1988).
Lerchl et al. (1990) reported that pineal NAT activity of rats was depressed by intermittent
exposure to magnetic fields. Yellon (1991) reported that 15-min exposure to a 60-Hz
horizontal magnetic field, at a strength of 0.1 mT, suppressed the nighttime melatonin rise
of Djungarian hamsters.
Biological Effecis of Magnetic and Electromagnetic Fields, Edited by S. Ueno
Plenum Press, New York, 1996 121
122 M. Kato and T. Shigemitsu

No papers, however, have been published concerning Ihe effects of long-term

exposure to alternating current magnetic fields on pineal gland activity. Therefore, in this
experiment we investigated the effect of exposure to 5()-Hz circularly and linearly polarized
magnetic fields for 6 weeks on melatonin concentration in rats, using a newly de\eloped
exposure facility.

2. EXPOSURE SYSTEM

2.1 System Design and Calculation

Many systems have been proposed for exposure to magnetic fields (Miller et al. 1989;
Wolpaw et al. 1989; Baum et al. 1991; Yasui & Otaka, 1993). A magnetic field can be
generated by using a pair of Hclmholtz coils. This provides a uniform field in an area
perpendicular to and around the center of the axis between paired coils. Due to difficulty in
establishing uniformity of magnetic fields for as large a volume as possible within limited
experimental space, our proposed system consists of five equally spaced square coils,
fomiing the surface of a cube. Initially, we referred to the three-coil design of Cohen's (1992)
exposure system used in the New York Project. Later, we extended our system to five sets
of coils which would establish a relatively large uniformity within a limited experimental
space. Figure 1 shows our proposed loop paired coils for a one-axis magnetic field. The
configuration of five rectangular loops of dimensions d x d and spacing of d/4 between loops
was proposed with different numbers of turns in pairs of loops, for generating a magnetic
field. By superposition of fields obtained from the coil pairs C| and Ci, and single coil C,,
total magnetic fiux density By(x.y,z) at point P(x,y,z) is calculated by using the results given
by Misakian (1984). After connecting five coils in series and adjusting the currents in the
coils, a considerably uniform region can be achieved.
We assume that the current ratio is a:b:c:b:a when the five coils are connected in
series and that the uniform field volume is about 0.6d x 0.6d x 0.6d cube in the center of d
X d X d cubic volume. To obtain the current under this condition, the magnetic fields for Br
(x,y.z) at three points inside the volume are chosen to be nearly the same values. Because of
the cubic symmetry of our proposed configuration, the calculation led to the conclusion that
the current ratio is 11:2:5:2:11, which is the same ratio of turns of coils. Using this current
ratio and assuming d =lm, the uniform field volume is about an 0.6 m cube in the center.
This one-axis system is a suitable exposure system for animal experiments. If there
is a requirement for both horizontal (Bh) and vertical (Bv) magnetic fields, two groups of
coils can be set to provide vertical or horizontal orientations, respectively. Circularly
polarized magnetic fields can be provided by two pairs of orthogonally placed coils.

;
C, It
d/4
Ci li P(x,y,z)
d/4 P(x.Y.z) Figure 1. Coordinate system and
Cj li configuration of the proposed loop
d/4 paired coils for generating a one-
C2 l2 axis magnetic field. (.Shigemitsu cl
d/4 al. Bioelectromatgnetics (1993)
vol 14: 107-116).
Effects of Exposure to a 50 Hz Magnetic Field un Melatonin in Rats 123

Figure 2. View of the exposure system.

2.2 System Construction

Figure 2 shows the exposure system. The system is formed as a 1.86 m cube. Because
of limitation of space, two identical, systems for exposure and sliam-exposure were placed
in the same room.
The floor plan of this room is shown in Figure 3, which gives approximate dimensions
and shows the placem.ent of the two systems. The x, y, z-directions were set in relation to
the wall. The x is west-east while the y is south - north. The two identical systems are 3.7 m
apart from center to center for x and 2.6 m apart for y. Design parameters for magnetic field
generation are as follows:

Horizontals Five 1.7 in square wooden frames at intervals of 0.425 m, with coils
having 66 turns of insulated wire (cross section 2.0 min^) in the outer paired coils, 12 turns

5.93m

Figure 3. Floor plan of the experiineiMal room in which

two exposure systems are placed. (Shigcmitsii. ct al.
Bioelectroiiiagiietics(1993)¥ol. 14: 107-116),
124 IVI. Kato and T. Shigemitsu

in the middle and 30 turns in the center coil. This group is the entire 1.7 m cube. For these
coils, the resistance was 8.0 Q and the inductance was 73.6 mH.

Vertical. Five 1.8 m square wooden fraines at intervals of 0.45 m, with coils having
the same numbers of turns and the same insulated wire as the horizontal coils. This group is
the entire 1.8 m cube, for in these coils the resistance was 13.5 Q and the inductance was
79.7 mH.

Rotating. The cube for the horizontal magnetic field coils fits inside the cube for the
vertical magnetic field coils with planes of the two sets of coils perpendicular to each other.
This arrangement gives the horizontal magnetic field in the x-direction and the
vertical magnetic field in the z-direction. An elliptically polarized magnetic field can be
obtained if each coil is energized separately from the power supply.
Each system has three Im square wooden shelves forming a Im cube. The wooden
shelves are employed as cage supports. These cage shelves are isolated from the coils which
are supported by another wooden frame. This isolation eliminates mechanical vibration of
the cages. Animal-housing space is within the Im cube. Water supply and food pellets are
placed on the top of each cage through a stainless steel grating. The standard wooden chip
is used for bedding. Nine standard cages (30 cmW x 25 cmL x 17 cmH) can be placed on
each shelf However, a total of 24 cages were placed for the actual experiments because no
cage was set on the center of the shelves so as to improve homogeneity of light distribution.
To equalize illumination within the cages during experiments, animal cages were inter-
changed among the three shelves.

2.3 Magnetic Field Distribution

The magnetic field strength and distribution within the animal-housing space was
measured by meters we developed (Shigemitsu et al. 1991). Figure 4 shows the results for
uniformity of the magnetic fields. It shows the variation of the normalized magnetic flux
density as a function of x, horizontal distance from the center of the coil at four different
planes inside the Im cube. Fig. 4A is the results of applying the vertical fields. Fig. 4B is
the results of applying the horizontal ones. The overall uniformity of the magnetic field is
more than 98 % within the animal-housing space of the facility.

2.4 Magnetic Field Circuitry

The power supply is located in the next room. The single phase 50 Hz, 200 V power
is supplied to two inverter units having 1.5 kVA capacity which independently provide power
to both coils. This energizing system can also provide 60 Hz. The system was designed to
produce circularly polarized magnetic fields up to 0.3 mT without heating and vibration.
Due to low heat generation in the coil, a temperature rise inside the animal cage could not
be detected. This system adjusts the phase difference with an accuracy of 5°. The horizontal
and vertical magnetic fields can be energized simultaneously. The exposure room door is
locked to insure safety for laboratory personnel.

2.5 Stray Field

Because the exposure and sham-exposure systems are placed in the same room, the
magnetic field of the exposure system will spill over into the sham-system. Using the
MFM-14 A meter, measurements of the stray field were made at four positions on the middle
shelf of the 1 m' area in the sham system when the exposure system was turned on. The ratio
Effects of Exposure to a 50 Hz Magnetic Field on Melatonin in Rats 125

'Vertical
ax-ax
105 r
- 100?-
95 L
by-by' by-by'
105 r 105 r
100| =» 100 p
95 1 95 I

cy-cy cy-cy
105 [ 105
1001 100?=
95 1 95 L

c-c c-c
105r 105 r
looL 100»-
95 L 95 L

A 0 10 20 30 40 50 0 10 20 30 40 50
Distance from center ot coil (cm) Distance from center of coil (cm)

' Horizontal
ay-ay ax-ax
105r 105|
100?
- 100o=
95L
95 L
by-by' bx-bx'
105r 105r
100<> lOOo-
95 L 95 L

cy-cy cx-cx
105r 105 r
lool- - lOoi-
95 L
' 95 I
c-c c-c
105|- 105
100 fl —7-0= n M
100? ~—0

951- 95

B 0 10 20 30 40 50 0 10 20 30 40 50
Distance from center of coil (cm) Distance from center of coil (cm)

Figure 4. The magnetic field distribution along the x-axis and the y-axis on four different planes inside the 1
m cube of the animal-housing space. The values of the vertical axis are normalized to the values at the center
A: Results of applying the vertical magnetic field. B: Results of applying the horizontal field. (Shigemitsu. et
al. Bioelectromagnetics (1993) vol. 14: 107-116).

of the exposure level to the sham level was overall 50:1. The ratio has a somewhat smaller
gradient, about 2 %, along the diagonal direction in the sham system. The stray field in the
sham system is somewhat distorted to ellipsoidal polarization compared with the circularly
polarized magnetic field in the exposure system.
126 M. Kato and T. Shigemitsu

2.6 Background Level of Magnetic Field

The background level of the 50 Hz magnetic field in the exposure room was measured
by using an MFM - 14 A meter The average level was 0.014 ^T. This value ranges from
0.034 (.iT to 0.01 |.iT depending on whether air-conditioning units are in operation. The value
of the horizontal component of the static magnetic field was in the range of 25-27 |iT and
its orientation was parallel to the y-direction (Gauss Meter HM 201, Mishima Industry). The
vertical component was in the range of 38-42 |aT. There was no difference in static magnetic
field between exposure and sham-exposure locations.

2.7 Harmonic Distortion

By placing the MFM-14 meter on the middle shelf of the exposure system, the
harmonic distortion of the magnetic field was measured. The output of this meter was
analyzed on a signal analyzer (SC-2I00C Iwatsu). The total harmonic distortion was very
small, about 0.03 % (Shigemitsu et al. 1993; Kato et al. 1993).

2.8 Noise
The noise level was measured by using a sound-level meter (Octave band filter
NX-OIA, sound level meter NA-61, Rion). By placing a microphone in the center of the
room and in both exposure and sham - exposure systems, the measurements were made with
the air conditioner turned on or off and the circularly polarized magnetic field turned on at
0.4 mTrms (B|,=B,). Results show that the air conditioner is the main contributor to noise.
The sound level was about 70 dB, which is the general room noise. There was no difference
in the sound level with the exposure system turned on or off

2.9 Light Intensity

Four standard fluoresent lights were installed on the ceiling 3.5 m above the floor
and another two were set up beside each system 1 m apart and 1 m from the floor. Light
scattering panels were set up on the top of the two systems. The overall light intensity was
20.4 - 84.4 lux with 24 animal cages inside the animal-housing space.

2.10 Temperature and Humidity

The temperature and relative humidity inside the experimental room were continu-
ously recorded. The temperature was maintained at 21 2°C. Relative humidity varied from
40 to 60 "o depending on changes in the air-conditioning cycle and the season.

3. MELATONIN CONCENTRATION
By using this newly developed exposure facility and Wistar-King male rats as
subjects, three experiments were conducted to determine whether 6 weeks of exposure to
circularly polarized, horizontally and vertically oriented magnetic fields suppresses mela-
tonin content in plasma and the pineal gland (Kato et al. 1993,1994). Plasma and pineal
melatonin were assayed by radioimmuno-assay (RIA).
In the first experiment rats were exposed continuously to a circularly polarized
magnetic field at 1,5, 50 or 250 ^T (spatial vector rms).
 Effects of Exposure to a 50 Hz Magnetic Field on Melatonin in Rats 127

150
P<0.05

1
100-

12:00 h
Q.

24:00 h

30 30 24 24 22 24 22 24 21 19 23 23 21 19 Sample number
C 0.02 0.1 1 1 5 50 ^T
91/8 90/7 91/4 90/7 90/12 91/4 90/12 Date Of Exp.

Figure 5. Levels of plasma melatonin at 12:00 and 24:00 h at different strengths of the magnetic Held across
time. Sample numbers of the rats, density in microteslas, and date of experimentation are indicated at the
bottom of the figure. For example, 91/8 means the main part of the 6-\veek exposure was carried out during
August 1991: C=control. The ordinate shows melatonin levels in picograms per milliliter The lc\cls deter-
mined at 0.02 j.iT, 90/7; 0.0! |jT, 91/4; and 1 j.iT, 90/12 were from the rats housed in the sham-exposure facility
in which the waveform of the field was ellipsoidal. (Kato, et al. Bioelectroiuagnetics 1199,^) \ol. 14: 97-1061.

Figure 5 summarizes the "no-fields" data and the data obtained by applying different
strengths of the magnetic fields across time. No statistical difference was observed among
the nighttime values for the control, 0.02 (iT, or 0.01 fiT (both stray fields), while a significant
difference was detected between the control and 0.1 |JT exposed animals at 12:00 h. There
was a significant decrease in melatonin content at 12:00 h for fields stronger than 0.1 ^T
and at 24:00 h for fields stronger than 1 |iT, compared with the control. There were no
significant differences among the values at 24:00 h for exposures to fields stronger than 1
I^T. At 12:00 h a significant decrease was observed at 50 /.iT.

Horizontal Vertical

30 30 24 24 24 24 30 30 ie 16 ie 16
C 0.02 1 C 002 1
91/a S3/I 93/1
91/8 91/7 91/7

Figure 6. Concentrations of melatonin in plasma and the pineal gland at 12:00h and 24:00h at horizontally
oriented (Horizontal) and vertically oriented (Vertical) 1 |jT magnetic fields. C is control values collected in
control experiments, (modified from Kato, et al. Neuroscicnce Letters (1994) vol. 168: 205-208).
128 M. Kato and T. Shigemitsu

In the second experiment, rats were exposed to horizontally oriented. 1 nT magnetic

fields (B|,= l |iT, By = 0) and in the third experiment, magnetic field was vertically oriented,
1 ^T (Bh = 0, B^ = 1 )aT). Results from both experiments are shown in Figure 6, in which
melatonin suppression at 24:00h after 6 weeks of exposure was not observed.

4. DISCUSSION
When one thinks about possible mechanisms of field exposure in animals, including
humans, electric current induced in the body under electric and/or magnetic fields is a likely
factor influencing body organs. Eddy currents induced by magnetic fields flow in loops
which are greatest near the periphery of the body. The pineal gland of the rat is located just
beneath the dura and skull, while in other mammals such as the cat, sheep, monkey, baboon
and human, the pineal gland is located deep inside the brain. Considering these anatomical
differences, we suggest that the pineal gland of the rat may be exposed to stronger eddy
currents than in other mammals. Since the biological effects may not be simply explained
by current density alone (K. Isaka, personal communication), other factors such as direction
of the current, phase angles of the field, waveform, speed of onset of the field (Rogers et al.,
1991), and location of the organs should be taken into consideration.
Olcese et al. (1985) compared the effects of magnetic field exposure on melatonin
content between rats with severed optic nerves and intact animals, and Reuss and Olcese
(1986) compared the effects between rats kept in a completely dark room and those kept
under dim red light. Under both experimental conditions there was no effect of magnetic
field exposure on melatonin level in the rats whose optic nerve system was not activated by
light stimulus. Hence they concluded that light stimulation to some extent is essential for
the magnetosensitivity. It has long been known that extremely low frequency (ELF) and
transient magnetic fields of moderate flux densities generate visual phenomena called
magnetophosphenes. Lovsund et al. (1979) reported that magnetophosphenes are generated
in the retina and are in the same channels that normally propagate signals induced by light.
Therefore, it is conceivable that signals set up in the retina by dim red light and magnetic
fields, of which magnetic field alone is subthreshold, eventually reach the pineal cells to
lower the synthesis of melatonin.
In our first experiment of circularly polarized field exposure at I fiJ suppressed
melatonin concentration, while in the following experiments of linearly polarized, vertical
or horizontal, magnetic field exposure it was demonstrated such suppressing effects do not
take place. Because all other obvious experimental conditions, such as geomagnetic fields,
age and strain of the rats, exposure period, etc. were quite similar in all of our experiments,
differences in results obtained with rotating field exposure vs vertical or horizontal field can
reasonably be attributed to differences in magnetic field characteristics.
Whether the primary effects of magnetic field exposure on melatonin production arise
from induced currents in the pineal gland, from alterations in the tonal aspects of neuronal
signaling, or from as yet unrecognized mechanisms, is a question that remains to be resolved
in future studies.

REFERENCES
1. Baum, J.W., Kuehner, .^.V., Benz. R.D. and Carsten, A.L. (1991) A system for simultaneous exposure of
small animals to 60 Hz electric and magnetic fields. Bioelectromagnetics 12:85-99p.
2. Cohen. H.D., Graham. C , Cook. M.R. and Phelps. J.W. (1992) ELF exposure facility for human testing.
Bioelectromagnetics 13:169-182P.
Effects of Exposure to a 50 Hz Magnetic Field on Melatonin in Rats 129

3. Deno,D.W.( 1976) Transmission line fields. IEEE PAS 95:1600-1611.

4. Kato,M., Honma,K., Shigemitsu.T. and Shiga, Y. (1993) Effects ofexposure to circularly polarized 50-Hz
magnetic field on plasma and pineal melatonin levels in rats. Bioeleclromagnetics 14:97-106.
5. ICato,M., Honma.K.. Shigemitsu, T. and Shiga,Y. (1994) Horizontal or vertical 50 Hz. 1 |.iT magnetic field
have no effect on pineal gland or plasina melatonin of albino rats. Neuroscience Letters 168:205-208.
6. Lerchl, A., Nonaka. K.O., Stokkan, K-A. and Reiter, R.J. (1990) Marked rapid alterations in nocturnal
pineal serotonin metabolism in mice and rats exposed to weak intennitlent magnetic fields. Biochem.
Biophys. Res. Commun. 169:102-108.
7. Lovsund, P. Oberg, P.A. and Nilsson, S.E.G. (1979) Influence on vision of extremely low frequency
electromagnetic fields. Acta Ophtalmol. 57:812-821.
8. Miller, D.L., Miller, M.C. and Kaune, W.T. (1989) Addition of magnetic field capability to extremely
low-frequency electric field exposure systems. Bioelectromagnetics 10:85-98.
9. Misakian.M. (1984) Electrical parameters in 60-Hz biological systems and their existing measurement:
A primer. NBS Technical Note 1191 (NTIS PB84-217793).
10. Olcesc, J. and Reuss. S. (1986) Magnetic field effects on pineal gland melatonin synthesis: comparative
studies on albino and pigmented rodents. Brain Res. 369:365-368.
11. Olcese, J., Reuss, S., Vollrath. L. (1985) Evidence for the involvement of the visual system in mediating
magnetic field effects on pineal melatonin symthesis in the rat. Brain Res. 333:3X2-384.
12. Reuss, S. and Olcese, J. (1986) Magnetic field effects on the rat pineal gland: role of retinal activation by
light. Neurosci. Lett. 64:97-101.
13. Rogers, W.R., Reiter, R.L., Smith, H.D. and Barlow-Walden, L. (1991) Under some circumstances,
combined electric and magnetic field exposure reduces serum melatonin concentration in nonhuman
primates. Proc. 1991 Contractors Review, US Dept. Energy A- 26.
14. Shigemitsu, T., Suganuma.H. and Takeshita, K. (1991) Development of power frequency magnetic fiux
meter. CRIEPI, Abiko Research Laboratory Rep No U91041.
15. Shigemitsu, T. and Suganuma, H. (1990) The design of an animal exposure system for investigating the
biological effects of power frequency magnetic field . CRIEPI. .^biko Research Laboratory. Rco No
U99035 (in Japanese).
16. Shigemitsu, T., Takeshita, K.. Shiga, Y. and Kato, M. (1993) 50-H/ magnetic field exposure system for
small animals. Bioelectromagnetics 14:107-116.
17. Stehle, J.. Reuss. S., Schroder, H., Henschel, M. and Vollrath. L. (1988) Magnetic field effects on pineal
N-acetyltransferase activity and melatonin content in the gerbil-role of pigmentation and sex. Physiol.
Behav. 44:91-94.
18. Welker, H.A., Semm, P.. Willing, R.R, Commentz. J.C, Wiltschko. W. and Vollrath, L. (1983) Effects of
an artificial magnetic field on serotonin N. acetyltransferase activity and melatonin content of the rat
pineal gland. Exp. Brain Res. 50:426-432.
19. Wilson,B.W. and Anderson,L.E. (1990)ELF electromagnetic field effects on the pineal gland, in "Ex-
tremely low frequency electromagnetic fields: the question if cancer" Wilson B.W., Stevens. R.G. and
Anderson, L.E. (eds) chap 8 pp. 159-186.
20. Wolpaw, J.R., Seegal, R.F. and Doweman, R. (1989) Chronic exposure of primates to 60 Hz electric and
magnetic field: 1. Exposure system and measurements of general health and performance. Bioelectro-
magnetics 10:277-288.
21. Yasui, M. and Otaka, Y. (1993) Facility for chronic exposure of rats to ELF magnetic field. Bioelectro-
magnetics 14:535-544.
22. Yellon, S.M. (1991) n acute 60 Hz magnetic field exposure suppresses the nighttime melatonin rise in
the pineal and circulation of the adult Djungarian hamster Proc. 1991 Contractors Review, US Dept.
Energy A-25.
INVESTIGATION OF EXPOSURE TO
EXTREMELY LOW FREQUENCY (ELF)
MAGNETIC AND ELECTRIC FIELDS
Status of Laboratory Animal Studies

Larry E. Anderson

Bioelectromagnetics, Biology and Chemistry Department

Battelle. Pacific Northwest Laboratories
Richland, Washinaton 99352

ABSTRACT

There is now convincing evidence from a large number of laboratories, that exposure
to extremely low frequency (ELF) magnetic and electric fields produces biological responses
in animals. Many of the observed effects appear to be directly or indirectly associated with
the neural or neuroendocrine systems. Such effects include increased neuronal excitability,
chemical and hormonal changes in the nervous system, altered behavioral responses, some
of which are related to sensing the presence of the field, and changes in endogenous
biological rhythms. Additional indices of general physiological status appear relatively
unaffected by exposure, although effects have occasionally been described in bone growth
and fracture repair, reproduction and development, and immune system function. A major
focus of ongoing research in the laboratory is to determine whether the epidemiological-
based suggested association between EMF exposure and risk of cancer can be supported in
studies using animal models. Three major challenges exist for ongoing laboratory research:
I) knowledge about the mechanisms underlying observed bioeffects is incomplete. 2) we do
not as yet understand what physical aspects of exposure produce biological responses, and
3) health consequences resulting from ELF exposure are unknown. .Although no animal
studies clearly demonstrate deleterious effects of ELF fields, several are suggestive of
potential health impacts. From the perspective of laboratory animal studies, this presentation
will discuss biological responses to ELF magnetic and/or electric field exposures.

INTRODUCTION

Over the past several decades, the likelihood for humans and animals to be exposed
to ELF magnetic (H) and electric (E) fields has increased substantially. Furthermore, the
Biolo^u-al Ff'fccty of Magnetic atii! f-'h'ciromagnctk Fields. Edited by S. L'eno
Plenum Press. Kew York. 1496 131
132 L. E. Anderson

level of exposure has increased by orders of magnitude over the natural background of
generally very low F, and H fields, Such elevated fields span all frequency ranges; however,
specific interest is currently focused on the potential biological impact of increased power
frequency (50 and 60 Hz) fields. Detailed reviews on the bioeffects of ELF fields arc
available'-'.
In the past twenty-five years, research programs throughout the world have made
significant progress in describing interactions between H and E fields found in the environ-
ment and living organisms. Also addressed have been questions regarding biological cfiects
from such fields, both real and potential, and whether such effects are permanent or transient,
detrimental or beneficial. In H and E field studies that have been reported, exposure levels
have been utilized from 0.1 to 30 millitessla (mT) and from a few volts/meter (V/m) to more
than 100 kV/m. Similarly, a broad range of biological endpoints have been examined for
evidence of possible ELF field exposure effects.
It should be noted that most indices of general physiological status appear to be
relatively unaffected by exposure to ELF fields. It is also generally recognized that exposure
to such fields does produce responses in specific biological systems, however, health
implications for humans and animals have yet to be determined. Where experimental effects
are demonstrated, the mechanisms of interaction between the field and the organisms remain
largely unknown. For instance, it is not known whether observed effects result from fields
acting at the surface of the body, E fields induced in the interior of the body, or from H fields
penetrating the body.
Even though the effects of H and E fields in humans are of primary importance, many
areas of biological interest are more effectively investigated using other animal species.
Results from experimental laboratory research in rats, mice, birds, swine, and nonhuman
primates form the basis for this paper, with emphasis on those areas which appear to be most
sensitive to ELF fields and which may provide information of most relevance to human ELF
exposure.

NERVOUS SYSTEM

Many of the biological effects which have been reported in animals or humans
exposed to ELF H or E fields appear to be associated with the nervous system. Such system
responsiveness to fields might be anticipated since the nervous system plays a basic role in
the interaction of animals with their environment. Indeed, other biological systems may be
influenced indirectly by ELF exposure through neural/neuroendocrine (hormonal) functions.
Reported nervous system effects from ELF exposures include changes in behavioral and
activity response; chemical changes in nerve cells; changes in the excitability of nerves;
altered neurotransmitter and neurohormone levels; and disruption of biological rhythms.

Behavior. Among the most sensitive measures of perturbation in a biological system

are tests which determine modifications in the behavioral patterns of animals. Behavioral
studies in several species provide evidence of E field perception, as well as indications that
behavior can be altered by exposure. The threshold of detection of 60-Hz E fields has been
reported to be between 4 and 10 kV/m in rats'*. E-field thresholds in other animal species,
including m i c e \ pigs'', and birds , have been reported in the 25- to 35-kV/m range.
Perception of H fields has not been observed in the submillitesia range**, however, visual
perception in the form of phosphenes has been demonstrated with H fields above 20 mT in
humans''"'.
.A.voidance behavior of animals has been investigated at several E-field strengths.
With some exceptions, animals generally avoid exposure to E fields over 50-75 kV/m".
Exposure to Extremely Low Frequency (ELF) Magnetic and Electric Fields 133

Under those levels, changes in physical activity responses have been reported although the
changes are usually transitory [see review '^]. Effects of H fields on behavior are curious in
that many of the investigations performed at low field intensities showed behavioral
alterations (primarily activity changes) [see review . In contrast, studies conducted at
higher H-field intensities demonstrated no evidence of effects on animal behavior.

Neurochemistiy. A number of studies have investigated the central nervous system

(CNS) for changes in various chemicals in the brain upon exposure to ELF fields. In general,
these studies report small, albeit highly variable changes in certain neurotransmitters'''. The
data provide limited evidence that exposure to E fields in the ELF range may cause slight
changes in nervous system function. The number of experiments is not large, and there are
significant questions about the strength and validity of several of the studies. Recent
experiments suggest significant changes in norepinephrine content in specific brain regions
of the hamster exposed to 60-Hz magnetic fields'^.

Neurophysiology. In the area of neurophysiology a confusing array of studies claims

both effects and noneffects of ELF-field exposure. A case in point is the commonly used
measure of general CNS activity, the electroencephalogram (EEG) where significant altera-
tions have been reported in some studies but not in others'"*. In an assessment of a more
specific electrical "fingerprints" of the brain, evoked responses, no effects caused by
exposure were observed in visual evoked response"', however, changes have been demon-
strated in somatosensory evoked response''.

Biological Rhythms. A number of investigations have been conducted to examine

the effects of ELF fields on natural biological rhythms. The phase and duration of activity
and rhythms of oxidative metabolism have been shown to be shifted in male mice by
exposure to ELF fields''\ Wilson and coworkers" measured the changing levels of indolami-
nes and enzymes in the pineal glands of E-field exposed rats. A significant reduction in the
normal night-time rise of melatonin and it's associated synthetic enzymes was observed in
rats exposed to either 1.5 or 40 kV/m. In more recent studies, nocturnal pineal components
in mice and rats were shown to be sensitive to rotated H fields*"and 50-Hz magnetic fields"'.
It is difficult to interpret the possible health consequences from the work thus far
published on circadian or biological rhythm effects from ELF exposures. However, it is
evident that E and H fields may alter the biological timing mechanisms in mammals. It is
possible that such changes could mediate significant alterations in other areas where effects
have been observed (e.g. behavior, reproduction and development). In addition, plausible
mechanisms involving ELF-induced hormonal changes have been proposed as a possible
basis for increased risk of adverse health outcomes^''.

REPRODUCTION AND DEVELOPMENT

It is generally assumed that the developing organism, including pre- and postnatal
mammals, are more sensitive to physical or chemical agents than are adult animals. This
greater sensitivity, when it occurs, is thought to originate in subtle effects on the processes
and controls which guide the developing cellular interactions. A number of studies have been
conducted to examine the effects of ELF exposure on reproduction and development of both
mammalian and nonmammalian species. Most of the nonmammalian studies have been
performed in birds, either chickens or pigeons. E-field exposure of chicks at several field
strengths, both before and after hatching, did not produce any significant effects in viability,
morphology, behavior, or growth-^ There have been some studies that reported deleterious
134 L. E. Anderson

effects of E fields on postnatal growth and survival in prenatal mammals, however, these
studies are countered by others in which rats, rabbits, or mice were exposed to broad range
of field strengths with no evident effects on reproduction, survival, and growth'^'''*.
Few studies have been performed to examine the effects of ELF H fields on growth
and development. In the most comprehensive study published to date, no reproductive or
developmental effects of 60-Hz H fields were observed^' These results have been generally
supported by a recent study performed in Finland'*. A great deal of interest has been
generated by reports from Delgado's lab; that significant increases in malformation rates
were observed in chick eggs exposed to low levels of pulsed magnetic fields-'. Subsequent
efforts have partially confirmed these results-*.

BONE GROWTH AND REPAIR

In animals exposed to 60-Hz E-fields, it appears that bone growth in rats and mice,
per se, was not affected by exposure to 100 kV/m, however, bone-fracture repair was
retarded'^'^. Pulsed H-field exposure of bone produces quite different responses, with en-
hanced repair^".

IMMUNOLOGY
Exposure of animals to electric fields does not appear to affect the immune system.
In a comprehensive investigation of the immune system, no effects of exposure at very low
field strengths (150-250 V/m) in mice or rats were observed^'. In contrast to the apparent
lack of E-field influence in vivo on the immune system, magnetic fields arc reported to
strongly affect system responses to mitogens and antigens^'.

CARCINOGENESIS AND MUTAGENESIS

Because of the increasing number of epidemiological reports of positive correlations
between ELF fields and cancer, considerable research interest has been generated concerning
a possible connection between H-fields and cancer. To date there are few published labora-
tory animal studies that bear directly on this question, however, an increasing number of
investigations are now being conducted.
A number of animal models can be utilized in the laboratory to investigate the issue
of EMF and cancer. The selection of a specific model is principally dependent upon the
hypothesis chosen to investigate a specific underlying mechanism. For example, if one
desires to test EMF for its potential to be a complete carcinogen (an agent, that by its
application alone, can give rise to the development of cancer), 1 1/2 to 2 years of exposure
of mice or rats to the EMF is necessary. Throughout that period of time exposure of the
animals to other possibly confounding agents must be kept to a minimum. In such a study,
animals are observed during the major portion of their lifetime and the occurrence of tumors,
in number, type, and time of development are the biological endpoints of interest. Complete
carcinogen studies require several dose groups and a relatively large number of animals,
particularly if the natural occurence of a tumor type is low. Clearly, studies evaluating
complete carcinogenicity are complicated and expensive because of the length of time and
the number of animals involved.
Carcinogenesis is generally recognized as a multistep process, therefore, another
approach is to assume the EMF acts either as an initiator or a promoter where a two phase
Exposure to Extremely Low Frequency (ELF) Magnetic and Electric Fields 135

protocol is required for testing. "Initiation" is defined as a genotoxic event in which the
carcinogen causes a direct affect on the DNA. "Promotion" is operationally defined, as an
enhancing agent that is applied subsequent to initiation over a protracted time period.
Promotion is tied to a number of subcellular events that are usually non-genotoxic and is
involved in the conversion of initiated cells to cancerous cells. To evaluate the agent (EMF)
as an initiator, one high dose of EMF would be given followed by repeated exposure to a
model promoter (e.g. 12-0-tetradecanoylphorbol-13-acetate. TPA). If, on the other hand.
EMF were to be investigated as a potential promoter, the animals would be treated with a
potent initiator of cancer (e.g. 7,12-dimethyl benz[a]anthracene, DMBA), and subsequently
exposed to EMF over a period of several months. These initiation/promotion approaches use
fewer animals and involve shorter experimental time and less cost than complete carcino-
genicity studies. However, a specific initiation/promotion model is usually restricted to
evaluating one or few specific cancers and may provide only limited information on possible
biological mechanisms of EMF and development of cancer.

Complete Carcinogen Studies. Few truly life-long animal studies examining EMF
as a complete carcinogen have yet been completed, although several are underway (in the
US, Italy, Japan, and Canada). However, several studies designed to evaluate EMF as a
promoter of cancer have contained control groups that were exposed to EMF without being
treated with chemical carcinogen (initiator). These studies include a mammary tumor
promotion study in rats", a lymphoma study in mice''*, and a mouse skin tumor promotion
study-*^. A major deficiency of using such "add-on" studies to evaluate complete carcino-
genicity is the small group sizes involved. The Beniashvili study found an increase in
mammary gland tumors in rats exposed to 20 ^iT for 3 hours per day compared to unexposed
animals. The other two studies reported no increase in tumors with long term exposure to
magnetic fields (500 or 50 \xT and 15 (xT, respectively).

Tumor Initiation Studies. No tumor initiation studies have yet been reported in the
literature. There is very little motivation for such studies because of the very weak energies
involved in extremely low frequency electric and magnetic fields: energies that are too weak
to break chemical bonds. Furthermore, invitro studies have provided no evidence that DNA
molecules can be damaged by exposure to 50/60 Hz EMF.

Tumor Promotion Studies. Despite the obvious need for EMF promotion studies,
based on the suggested associations between EMF and cancer found in the epidemiological
results, relatively few animal experiments have been completed. Skin tutnor promotion, after
initiation with DMBA, was examined in mice exposed to a 2 mT, 60 Hz continuous magnetic
field, 6 hr/d, 5 days/wk for up to 21-23 weeks'*. None of the exposed or sham exposed mice
developed papillomas. When magnetic fields were combined with application of TPA, a
slightly earlier development of tumors was observed in the magnetic field-exposed animals' .
Rannug and coworkers have conducted both skin tumor and liver foci studies in
Sweden . in the two-year skin tumor promotion study, mice were initiated with
DMBA then exposed to 0.5 mT or 50 ^T, 50 Hz magnetic fields for 19-21 hr/d. No evidence
of a field-exposure effect was observed in either systemic or skin tumor development or in
skin hyperplasia. In the liver foci study, rats were treated with similar EMF exposures over
a 12 week period. The exposed animals showed no differences in foci development from the
sham exposed rats. In animals treated with a chemical promoter (phenobarbital) as well as
the magnetic field, foci formation was slightly inhibited when compared to initiated-only
animals.
In a series of experiments conducted in Germany'"''*'''-'*', perhaps the strongest
evidence for the carcinogenicity potential of EMF in rats has been reported. Generally, rats
136 L. E. Anderson

were exposed for 3 to 4 months to 50 Hz magnetic fields ranging from 0.1 to 30 mT. Initiation
was accomplished with repeated oral doses of DMBA and subsequent development of
mammary tumors was determined. In some of the experiments, the tumor incidence was
increased in EMF exposed animals. In other experiments of this series, the number of tumors
per tumor bearing animal was increased but the total tumor incidence was not affected. In
still other efforts at replicating the initial results, non-statistically significant trends were
seen but the clear enhancement of tumor development with EMF exposure was not observed.
These apparent differences in results from the same research group may reflect differences
in animal response at the different field intensities used or may simply be a reflection of the
differences in group size between experiments. Several efforts are now underway to try and
replicate this important studies on EMF and mammary cancer.
Prior to the Mevissen study, a group in Georgia also examined mammary carcino-
genesis in EMF-exposed animals that were initiated with N-nitroso-N-methylurea'\ In the
groups of animals exposed to a 20 |iT, 50 Hz magnetic field for 3 hr/d, for the lifetime of
the animals, there was an increased incidence of NMU-induced mammary tumors over the
sham animals or animals exposed to only 1/2 hr of EMF per day.

SUMMARY AND CONCLUSIONS

Numerous studies have been initiated to determine the nature of the physical
mechanisms involved in ELF field-induced effects and to what extent an electrical environ-
ment containing H or E fields poses a health hazard to living organisms. The biological
effects reported in many of the experiments have not confirmed pathological effects, even
after prolonged exposures to high-intensity H (10-mT) and high-strength E fields (100 k V/m)
fields. However, the question of whether an occurrence of a "biological effect" from field
exposure constitutes a health hazard has yet to be answered. Areas in which effects have
been demonstrated appear to be associated primarily with the nervous system. In addition,
in several instances, unconfirmed or controversial data exist where observed effects may be
due to the fields (e.g., changes in brain chemistry and morphology, alterations in reproduction
and development). It is not yet known whether confirmed or putative effects are due to a
direct interaction of the electric field with tissue or to an indirect interaction, e.g.. a
physiological response due to detection and/or sensory stimulation by the field. Future
research thrusts, at the minimum, need to address the following topics: 1) an investigation
of what aspects of field exposure constitutes dose to living systems; 2) a determination of
the interaction mechanisms between ELF fields and animals/humans; 3) an exploration of
the implications of observed effects on nervous system function; 4) an evaluation of the
potential effects of ELF fields on reproduction and early development; and 5) a determination
of whether cancer promotion and/or progression is influenced by ELF fields.
Research sponsored today in several countries is designed to provide data to help set
maximum levels of ELF field intensities which are biologically acceptable for both workers
and the general public. This research also addresses the broader issues of increased electro-
magnetic fields in the environment from a variety of sources, and the potential health
implications for humans.

ACKNOWLEDGEMENTS

This work was supported by the Electric Power Research Institute (RP 2965-16) and
the U.S. Department of Energy, Office of Energy Management, under contract DE-AC06-
Exposure to Extremely Low Frequency (ELF) Magnetic and Electric Fields 137

76RLO 1830. Portions of this paper were previously published in Electricity and Magnetism
in Biology and Medicine, M. Blank, Ed. San Francisco Press, 1993.

REFERENCES
1. .Anderson, L.R. Biological effects of extremely low frequency electromagnetic fields: In vivo studies. In
Proceedings ol llie Scientific Work.sliop on tite Health Effects of Electric and Magnetic Fields on Worlicrs.
Public Health Service. NIOSH, Cincinn, OH, pp 45-89, 1991.
2. Anderson, L.E. and W.T. Kaune. Electric and magnetic fields at extremely low frequencies. In M.J. Suess.
Ed., Nonionizing Radiation Protection. Second edition. World Health Organization. Geneva. 175-243,
1989.
3. Shcppard, A.R. and M. Eisenbud. Biological Effects of Electric and Magnetic Fields ol Extremely Low
Frequency. New York Univ. Press, New York, 1977.
4. Stern S., V.G. Latics.C.V. Stancampiano.C. Cox and J.O. deLorgc. Behavioral detection of 60-Hz electric
fields by rats. Bioelectromagnetics 4:215-247. 1983.
5. Sander R.. J. Brinkmann and B. Kuhne. Laboratory studies on animals and human beings exposed to
50-Hz electric and magnetic fields. In International Conference on Laige High Voltage Electrical Systems.
September 1-9, Paris, Abstract number 36-01, 1982.
6. Kaune W.T., R.D. Phillips, D.L. Hjercsen, R.L. Richardson and J.L. Beamcr A method for the exposure
of miniature swine to vertical 60-Hz electric fields. IEEE Trans Biomed Eng BME 25:276-283, 1978.
7. Graves H.B., J.H. Carter, D. Kellmcl, L. Cooper, D.T. Poznaniak and J.W. Bankoskc. Perceptibility and
electrophysiological response of small birds to intense 60-Hz electric fields, IEEE Trans Power .Appar
Sysl. PAS 97:1070-1073. 1978.
8. Lovely. R.H.. J.A. Creim, W.T. Kaune. M.C. Miller. R. D. Phillips and L.E. Anderson. Rats are not aversive
when exposed to 60-Hz magnetic fields at 3.03 mT. Bioelectromagnetics 13:351-362. 1992.
9. Barlow. H.B., H.I. Kohn and E.G. Walsh. Visual sensations aroused by magnetic fields, .\m. J. Physiol.
148:376. 1947,
10. Lovsund, P., P.A. Oberg. S.E.G. Nilsson and T. Reuter. Magnetophosphenes: a quantitative analysis of
thresholds. Med. Biol, Eng. Compul. 18:326, 1980.
11. Hjercsen D.L., W.T, Kaune. J,R. Decker and R.D. Phillips. Effects of 60-Hz electric tields on avoidance
behavior and activity of rats. Bioelectromagnetics 1:299-312. 1980,
12. Lovely, R.H. Recent studies in the behavioral toxicology of ELF electric and magnetic fields. In M,E.
O'Connor and R,H. Lovely. Eds,. Electromagnetic Fields and Neiiroheharioral Function. I.iss Inc. New
York. pp327-348. 1988.
13. Tenforde T.S, Interaction of ELF magnetic fields with living matter In C, Polk, and E, Postow, Eds.,
Handbook of Biological Effects of Electromagnetic Radiation. CRC Press, Boca Raton, PL, pp 197-225,
1986,
14. Anderson, L.E, Interaction of ELF electric and magnetic fields with neural and neuroendocrine systems.
In B.W. Wilson, R.G. Stevens, and L.E. Anderson, Eds., Extremely Lon- Frequency Electromagnetic
Fields: The Question of Cancer. Battelle Press, Columbus OH, ppl39-158, 1990.
15. Wilson, B,W.. J.E. Morris, L.B. Sasser, L.E. Anderson and K. Matt. Changes in the hypothalamus and
pineal gland of djungarian hamsters from short-term exposure to 60-Hz magnetic fields. DOE EPRl
Contractors Review, Savannah, GA, Oct. 3 l-Nov 4, 1993.
16. Jaffe R./\.,C.A, Lopresti, D.B. Carrand R.D. Phillips. Perinatal exposure to 60-Hz electric fields: Effects
on the development of the visual-evoked response in the rat. Bioelectromagnetics 4:327-339, 1983.
17. Wolpaw. J.R., R.F. Secgal, R.I. Dowman and S. Satya-Murti. Chronic Effects of 60 H: Electric and
Magnetic Fields on Primate Central Nervous System Function. Final Report to the New York State Power
Lines Project, New York. 1987.
18. Ehret C.F, R.S, Rosenberg, G.A, Sacher, P.H. Duffy, K.R. Groh and J.J. Russell. Biomedical Effects
.Associated with Energy Transmission Systems: Effects of 60-Hz Electric Fields and Circadian and
Ultradian Physiological and Behavioral Functions in Small Rodents. Washington, DC. .Annual Report to
U, S, Department of Energy, 1980.
19. Wilson B.W,, L,E. Anderson, D.I, Hilton and R.D. Phillips. Chronic exposure to 60-Hz electric fields:
Effects on pineal function in the rat. Bioelectromagnetics 2:371-380, 1981.
20. Lerchl, A,, K.O. Nonaka, K.A, Stokkan and R.J, Reiter Marked rapid alterations in nocturnal pineal
serotonin metabolism in mice and rats exposed to weak intennittent magnetic fields. Biochem. Biophys,
Res. Commun. 169:102-108, 1990.
138 L. E. Anderson

21. Kato, M., K. Honma, T. Shigemitsu and Y. Shiga. Effects of exposure to a circularly polari/ed 50-Hz
magnetic field on plasma and pineal melatonin levels in rats. Bioelectromagnetics 14:107-116. 1993.
22. Stevens, R.G.. S. Davis. D.B. Thomas, L.K. Anderson and B.W. Wilson. Electric power, pineal function,
and the risk of breast cancer. FASEB 6:853-860. 1992.
23. ChemolT N.. J. M. Rogers and R. Kavel. A review of the literature on potential reproductive and
developmental toxicity of electric and magnetic fields. Toxicology 74:91-126, 1992.
24. Rommcreim, D.N.. R.L. Rommereim, R.L. Buschbom, M.R. Sikov and L.E. Anderson. Reproduction,
growth and development of rats during chronic exposure to multiple field strengths of 60-Hz electric
fields. Fund. .Appl.Toxicol. 14:608-621, 1990.
2.'i. Rommereim. D.N.. R.L. Rommereim. D.L. Miller. R.L. Buschbom and L.B. Anderson. Developmental
toxicology evaluation of 60-Hz horizontal magnetic fields in rats. Curr. Top. Occup. Hea. (in press),
26. Huuskonen, H.. J. Juutilainen and H. Komulainen. Effects of low-frequency magnetic fields on fetal
development in rats. Bioelectromagnetics 14:205-214, 1993.
27. Delgado J.M.R., J. Leal, J.L. Montcagudo and M.G. Gracia. Embryological changes induced by weak,
extremely low frequency electromagnetic fields. J. Anat. 134:533, 1982.
28. Berman, B., L. Chacon, D. House, B.A. Koch, W.E. Koch, J. Leal, S. Lovtrup, E. Mantiply, A.II. Martin.
G.l. Martucci, K.H. Mild, J.C. Monahan, M. Sandstrom, K. Shamsaifar, R. Tell, M.A. TriUo. A. L'beda
and P. Wagner Development of chicken embryos in a pulsed magnetic field. Bioelectromat;nelics
11:169-187, 1990.
29. McClanahan B.J. and R.D. Phillips. The influence of electric field exposure on bone growth and fracture
repair in rats. Bioelectromagnetics 4:11-20, 1983.
30. Bassett, C.A.L., S.N. Mitchell and S.R. Gaston. Pulsing electromagnetic field treatment in ununited
fractures and failed arthrodeses. JAMA 247:623-628, 1982.
31. .Vlorris J.L, and R.Ll. Phillips. Eflects of 60-Hi? electric fields on specific humoral and cellular components
of the immune system. Bioelectromagnetics 3:341-348, 1982.
32. Conti P.. G.E. Giganti, M.G. Cifone, E. Alessc, G. lanni, M. RealeandP.U. Angeletti. Reduced mitogenic
stimulation of human lymphocytes by extremely low frequency electromagnetic fields. FEBS Lett
162:156-160, 1983.
33. Beniashvili. D.Sh.. V'.G. Bilanishvili and M.Z. Menabde. Low-frequency electromagnetic radiation
enhances the induction of rat mammary tumors by nitrosomethyl urea. Cancer letters. 61:75-79, 1991,
34. Svedenstal. B.M. and B. Holmberg. Lymphoma development among mice exposed to x-rays and pulsed
magnetic fields. Int. J. Radiation Biol. 64:119-125. 1993.
35. Rannug. .'\.. T. Ekstrom. K. Hansson Mild, B, Holmberg, I. Gimencz-Conti and T.J. Slaga. .\ study on
skin tumor formation in mice with 50 Hz magnetic field exposure. Carcinogenesis 14:573-578. 1993.
36. McLean. J.R.N., M.A. Stuchly, R.E.J, Mitchel. D. Wilkinson. H. Yang. M. Goddard. D.W. Lecuyer. M,
Schunk. E, Callary and S.D. Morrison. Cancer promotion in a mouse-skin model by a 60 Hz magnetic
field:ll. Tumor development and immune response. Bioelectromagnetics 12:273-287. 1991.
37. Stuchly, MA.. J.R.N. McLean. R. Burnett. M. Goddard, D.W. Lecuyer and R.E.J. Mitchel. Modification
of tumor promotion in the mouse skin by exposure to an alternating magnetic field. Cancer Letters 65:1 -7.
1992.
38. Rannug. .^.. B. I lolmberg and K. Hansson Mild. A rat liver foci promotion study with 50 Hz alternating
magnetic fields. Environ. Res. 62:223-229. 1993.
39. Rannug. A., B, Holmberg, T. Ekstrom and K. Hans.son Mild. Rat liver foci study on coexposure with 50
Hz magnetic fields and known carcinogens. Bioelectromagnetics 14:17-27, 1993.
40. Mevissen, M., A. Stamm. S. Buntcnkotter, R. Zwingelberg, U. Wahnschaffe and W. Loscher. Effects of
magnetic fields on mammary tumor development induced by 7,12- dimethylbenz[a]anthracene in rats.
Bioelectromagnetics 14:131-143, 1993.
41. Loscher. W.. M. .Mevissen, W. Lehmacher and A. Stamm. Tumor promotion in a breast cancer model by
exposure to a weak alternating magnetic field. Cancer Letters 71:75-8l, 1993.
42. Loscher. W..L. Wahnschaffe, M, Mevissen. A. Lerchland A. Stamm. Effects ofweak alternating magnetic
fields on nocturnal melatonin production and mammary carcinogenesis in rats. Oncology 51:288-
295.1994.
43. Mevissen, M.. M. Kietzmann, and W. Loscher. In vivo exposure of rats to a weak alternating magnetic
field increases ornithine decarboxylase activity in the mammary gland by a similar extent as the
carcinogen DMBA. Cancer Letters 90:207-214, 1995.
 10

MILLIMETER-RESOLUTION DOSIMETRY
FOR EM FIELDS FROM MOBILE
TELEPHONES AND POWER LINES

O. p. Gandhi and J. Y. Chen

University of Utah
Department of Electrical Engineering
SakLakeCity, Utah, 84112

INTRODUCTION
Increasingly finer-resolution anatomically based heterogeneous models of the human
body are being used for calculations of induced electric fields, current densities, and specific
absorption rates for electromagnetic exposures from extremely low frequencies (ELF) to
microwave frequencies [1,2]. Because ofthe need to extend dosimetric calculations to higher
frequencies of several gigahertz, and for small-volume exposure devices such as mobile
telephones, hair dryers, hair clippers, etc., resolutions on the order of millimeters are needed.
A new mm-resolution model based on the MRI scans of an adult male volunteer has,
therefore, been developed [1, 3, 4]. This model, with a resolution of 3 mm in the vertical
direction and a pixel size of 1.974 x 1.974 mm has, to date, been used to calculate coupling
of EM fields from handheld wireless communication systems operating at cenlerband
frequencies of 835 and 1900 MHz [4], and for spatially varying 60-Hz magnetic fields of a
hair dryer and a hair clipper [5].

THE FINITE-DIFFERENCE TIME-DOMAIN METHOD

The numerical method used for all of the calculations given in this paper is the
finite-difference time-domain (FDTD) method. This method was first proposed by Yee [6]
and later developed by Taflove and colleagues [7-10], Holland [11], andKunzand Lee [12],
We have extended the method for calculations ofthe distributions of electromagnetic (EM)
fields and SARs in anatomically based models of the human body for whole-body or
partial-body exposures due to far-field or near-field irradiation conditions [13-16] and for
electromagnetic pulse (EMP) exposures [17]. In this method, the time-dependent Maxwell's
curl equations

Biological Effects of Magnetic and Electromagnetic Fields, Edited by S. Ucno

Plenum Press. New York. 1996 139
 140 O. P. Gandhi and J. Y. Chen

V X E = -^li ^ , V X H = CTE + E

(1)

are implemented for a lattice of siibvolumes or "cells" that may be cubical or

parallelepiped with different dimensions 5„ Sy, and 5^ in x-, y-, orz-dircctions, respectively.
The details of the method are given in several'of the above referenced publications and will,
therefore, not be repeated here,
lo the FDTD method it is necessary to represent not only the scatterer/absorber
such as the human body or a part thereof, but also any near-field source/s such as an
antenna and handset of a wireless telephone by means of the volume-averaged electrical
properties (e„ o). The source-body interaction volume is subdivided into the Yee cells.
The interaction space consisting of several hundred thousand to a few million cells is
truncated by means of absorbing boundaries. Initial fields often assumed to be sinusoidally
varying are tracked in time for all cells of the interaction space. The problem is considered
completed when a sinusoidal steady-state behavior for E and H is observed for the
interaction for single-frequency irradiation or the incident pulse has died off for transient
exposures.

MILLIMETER^MESOLUTION MODEL OF THE HUMAN BODY

We have developed a millimeter-resolution model of the human body from the
magnetic resonance imaging (MRI) scans of a male volunteer of height 176.4 cm and weight
64 kg. The MRI scans were taken with a resolution of 3 mm along the height of the body
and 1.875 mm for the orthogonal axes in the cross-sectional planes. Even though the height
of the voluDteer was quite appropriate for ao average adult male, the weight was somewhat
lower than an average of 71 kg, which is generally assumed for an average male. This

Figure I. ,-\ rypical crosh section of the milUtTieter-rf'so'ulioti model of the human bodv. Th; ticular cross
seel icr. is for la>xr r;0, 4-), which is 12 cm below the top of the head. Shown arc the eonioyi the eyes, the
opiic nerves,, cerebeiUuii, etc, against tiit* b,:ickgTijijRii of a gnd o.f ceifs oiMiaiciisions l.9"4 974 mm.
Millimeter-Resolution Dosimetry for EM Fields 141

problem can, to some extent, be ameliorated by assuming that the cell dimensions for the
cross sections are larger than 1.875 mm by the ratio of (71/64)' - = 1.053. i.e.. 1.974 mm
instead of 1.875 mm. Using a software package from the Mayo Clinic called ANALYZE,
the MRI scans were converted into images involving 29 tissue types whose electrical
properties (s^, a) can then be prescribed at the frequency of interest. Shown in Fig. 1 is a
typical cross section of the new model of the body. This is a section through the eyes and
the cerebellum. Also shown in Fig. 1 are the contours for the various features of this cross
section such as the eyes, ear, optic nerve, cerebellum, etc.

SAR DISTRIBUTIONS FOR MOBILE TELEPHONES

Cellular telephones and mobile wireless communication systems are being intro-
duced into society at a very rapid rate. This has resulted in a public concern about the health
hazards of RF electromagnetic fields that are emitted by these devices. We have used the
head and neck parts of the above described millimeter-resolution model of the body to study
electromagnetic absorption for a number of mobile telephones operating at centerband
frequencies of 835 MHz, and some devices that may be designed for use at 835 and 1900
MHz, respectively. The peak 1 cm^ mass-normalized rates of energy absorption (specific
absorption rates or SARs) may be compared with the ANSI IEEE C95.i-I992 RF Safety
Guidelines [18]. These safety guidelines are given in terms of maximum permissible
exposures (MPE) of electric and magnetic fields, or of power density for controlled and
uncontrolled environments. Though simple to use for far-field, relatively uniform exposures,
the MPE limits are not easy to use for highly nonuniform fields such as in the near-field
region of a mobile telephone. An alternative procedure given in the following [18] has,
therefore, been suggested to satisfy the safety guidelines for uncontrolled environments
which are defined as situations where there is exposure of individuals who have no
knowledge or control of their exposure.
-An exposure condition can be considered to be acceptable if it can be shown that it produces S.ARs
"below 0.08 W/kg, as averaged over the whole body, and spatial peak SAR values not exceeding
1.6 W'kg, as averaged over any 1 cm' of tissue (defined as a tissue volume in the shape of a cube)".

Since negligible SARs result for the lower regions of the body, we have used
the upper 42-cm height of the body involving head, neck, and the upper torso. Taken
from references 19-22, the dielectric properties used for the various tissues at center
frequencies of 835 and 1900 MHz are given in Table 1. To simulate a handset that
is typically tilted forward by about 33° for a vertically erect head, we have modified
the MRI-based model so that it is tilted forward by 33°. With this forward tilt ac-
complished by displacing each successive layer of the MRI scans separated by 3 mm
back by 1 cell, i.e., 1.974 mm, the handset may then be held in the vertical position
for SAR calculations. The vertical orientation of the handset and the antenna allows
a more accurate modeling of their shapes and dimensions. For commercially used
telephones 1-10, given in Tables 2 and 3, we have used the X-ray pictures of the
various antennas to model the exact dimensions of the radiating elements. Models for
each of the antennas and the handsets were assumed to be covered with insulating
materials of prescribed dielectric constants (typically 4.0 and 3.0). Since the coatings
of these materials are thinner (generally 1 mm) as compared to the cell size of 1.974
mm, effective dielectric constants K^ given by the following equation are used for
the cells covering the antennas and the handsets.
142 O. P. Gandhi and J. Y. Chen

Table 1. Dielectric properties of the various tissues assumed for the model of the head and
neck at midband mobile telephone frequencies of 835 and 1900 MHz
Old values for 835 New values for 835 New values for 1900
MHz [19 -21] MHz [22] MHz [22]
Tissue
Type* Tissue Sr o Sr CT e, a
1 muscle 51.0 1.35 50.0 1.08 48.0 1.54
2 fat 7.2 0.16 11,0 0.17 11.0 0.28
3 bone 7.2 0.16 21.0 0.33 16.5 0.45
4 cartilage 7.2 0.16 37.0 0.80 36.0 1.22
5 skin 35.0 0.60 35.0 0.60 55.0 0.55
6 brain 43.0 0.86 41.0 0.86 41.1 1.14
II blood 64.0 1.25 55.0 1.86 54.0 2.27
12 eye 70.0 1.90 70.0 1.90 70.0 2.70
17 cerebrospinal 76.0 1.75 78.0 1.97 78.0 2.76
tluid (CSF)
18 vitreous humor 73.0 1.90 67.0 1.68 74.0 2.35
14 sclera/comea 52.0 1.80 51.0 1.13 50.0 2.30
20 lens 45.0 0.75 33.0 0.79 42.0 1.20
* These numbers correspond to the tissue types used for the model of the whole body.

5 £r
K= [E r (8 - w) + w]

(2)

where w is the thickness of the coating in millimeters and 8 is the dimension of the
Yee cell which may be 8^, 6y, or 8^, depending on the surface. Because of the proximity of
the hand to the telephone, it is essential to also model the hand for numerical calculations.

Table 2. Calculated and measured SARs for the various commercial telephones
for the maximum radiated power of 0.6 W or 600 mW at 835 MHz
Peak 1 -cell (11.7 mg of tissue) local SARs
Brain Ear
Calculated
whole-body-
Telephone averaged SAR
No. mW,/kg W/kg Calculated W 'kg M easured W.'kg
1 2.35 1.29 3.60 1.56-2.80
2 1.58 0.97 2.20 1.61-2.91
3 1.60 1.33 2.97 2.80-5.30
4 1.58 0.59 1.87 1.94-2.84
5 2.13 1.99 4.03 1.82-3.57
6 0.89 0.51 1.72 0.80-1.70
7 0.81 0.49 1.59 0.47-1.28
8 1.09 0.31 0.96 0.70-0.84
9 2.30 0.70 1.96 0.86-1.92
10 0.88 0.51 1.63 1.81-3.07
4* 1.60 0.35 2,48 1.94-2.84
* Values calculated for telephone no. 4 using the new tissue properties at 835 MHz given
in Table 1.
Millimeter-Resolution Dositnetrv for EM Fields 143

Table 3. Peak SARs for 1 g of tissues of the head, brain, and "hand"
for the maximum telephone power of 0.6 W at 835 MHz
Peak SAR for any 1 cm' of tissue^
Telephone of the head of the brain of the "hand"
No. W/kg W/kg W/kg
1 0.57 0.26 0.66
2 0.38 0.21 0.41
3 0.51 0.28 0.59
4 0.28 0.13 0.49
5 0.69 0.41 0.71
6 0.26 0.10 0.15
7 0.26 0.10 0.09
8 0.16 0.06 1.90
9 0.48 0.16 0.33
10 0.25 0.14 0.24
4* 0.35 0.15 0.49
* Defined as volume in the shape of a cube.
* Same as for Table 2.

For the present calculations we have modeled the hand by a region of 2/3 muscle-equivalent
material of thickness 1.974 cm (10 5^) wrapped around the handset on three sides, with the
exception of the side facing the head, up to two-thirds of the height of the handset.
To normalize the calculated SAR distributions for peak operating powers of 600 m\V
or 125 mW for mobile telephones operating at 835 and 1900 MHz, respectively, the
procedure was as follows. A sinusoidal driving voltage of 3 V rms was assumed across the
gap for the cell representing the driving point. Starting with this prescribed unifonn
z-directed E-field of 1000 V/m rms, the various converged components of E and H were
calculated for all the cells of the modeled region. From the calculated tangential magnetic
fields for the cells representing the antenna, we obtained the z-directed current distributions
along the lengths of the various antennas from the expression I = - f H d_. The z-directed
current calculated for the cell corresponding to the driving point was then used to calculate
the complex feedpoint impedance Z - V/I = R + jX and, hence, the power 1/2 I I * R fed into
the antenna. The calculated SARs were subsequently scaled for maximum antenna powers
of 0.6 and 0.125 W, respectively.

Table 4. Comparison of the powers absorbed and peak SARs for the X<'4 and 3 /J8
antennas at 835 MHz. Time-averaged radiated power = 0.6 W. Assumed size of the
handset is 2.5 x 6 x 15 cm
Peak 1-cm^ SAR W/kg % power absorbed by
Antenna
Length Tilt Head Brain "hand" head and neck
X/4 0° 1.53 0.97 17.0 63.1
(596 mg)* (877 mg)*
/,/4 33° 1.21 0.54 15.0 50.8
(549 mg)* (713 mg)*
3 >./8 0° 0.67 0.35 16.5 42.5
(526 mg)* (877 mg)*
3;./8 33° 0.74 0.68 12.5 34.4
(795 mg)* (877 mg)*

* Actual weight of 1 cm of volume.

144 O. P. Gandhi and J. Y. Chen

Table 5. Comparison of the powers absorbed and peak SARs for the >./4 and 3 L'8
antennas at 1900 MHz. Time-averaged radiated power = 125 mW. Assumed size of the
handset is 2.5 x 6 x 15 cm
Peak 1-cm' SAR W/k g % power absorbed by
Antenna Head and
Length Tih Head Brain "Hand" "Hand" Neck
}J4 0= 0.86 0.28 0.46 22.5 46.9
(.'>03 mg)* (818 mg)
k:4 33.3° 0.86 0.49 0.41 22.1 45.1
(503 mg)* (818 mg)*
3 X:'S 0° 0.71 0.24 0.39 17.7 48,8
(503 mg)* (853 mg)*
3 >./8 33.3° 0.85 0.38 0,38 17.1 46.3
(561 mg)* (878mg)»

* Actual weight of 1 cm- of volume.

The salient features of the results obtained for some of the mobile telephones studied
to date are given in Tables 2 to 5. For each of the telephones we have selected the side of the
head for which the SAR was higher because of the proximity of the antenna to the head. In
Table 2, for the telephones 1, 2, 3, and 5, we have considered the telephone placed against
the left ear, while for telephone 4, we have considered the telephone placed against the right
ear in order to obtain the highest possible SARs for the model. Because of the symmetrical
placement of the antennas in telephones 6-10 and for the }J4 and 3 X/S antenna devices
assumed for the data given in Tables 4 and 5, the choice of the side of the head was immaterial.
Even though older values of £r and a given in Table 1 [19-21] were used for the
calculations given in Tables 2 and 3, newer values for both 835 and 1900 MHz, given in
Table 1, have recently become available [22], and these have been used for the calculated
data given in Tables 4 and 5, respectively. An important thing to note for the newly obtained
dielectric properties given in Table 1 is that the values of £, and o, both for skull/bone and
cartilage, are much higher than those given previously [19-21]. Using telephone 4 as a test
case, we find slight alterations in the calculated values of the SARs when new values of Er,
a are used at 835 MHz. The data calculated for telephone 4 using the new values of the tissue
properties are given in the last rows of Tables 2 and 3, respectively. As expected, the peak
1 -cell SAR for the higher-conductivity cartilage of the ear is somewhat higher, and the 1 -cell
SAR for the brain is somewhat lower (see Table 2) because of the additional shielding
provided by the higher-conductivity skull.
We have examined the effect of the antenna length {X/4 vs. 3^/8) on the power
absorbed by the head and neck and have determined the peak 1-cm' SARs anywhere in the
head, including the ear, and the tissues in the brain. The salient features of the calculated
results at 835 and 1900 MHz are given in Tables 4 and 5, respectively.
It is interesting to note that the powers absorbed by the head and neck and the peak
1-cm" SARs are lower for a 3>./8 antenna vis vis a X/4 antenna at 835 MHz (Table 4), while
these quantities are almost the same for corresponding lengths of antennas at 1900 MHz
(Table 5). This may be due to the fact that a 3?./8 antenna is substantially longer (13.5 cm)
at 835 MHz than that for 1900 MHz (5.9 cm). The peak current region for a 3/./8 antenna is
a length of >L/8 higher up on the antenna as compared to the A./4 antenna where it is at the
base of the antenna and, hence, very close to the ear. For a longer 3A./8 antenna at 835 MHz,
the peak current region is, therefore, somewhat removed from the ear and the head, which
results in lower absorbed power and SARs as compared to the X/4 antenna, while a similar
effect is not obtained for the 3?t/8 antenna at 1900 MHz where the high-current region is still
Millimeter-Resolution Dosimetry for EM Fields 145

awfully close to the head. Similar arguments can also be given for the calculated lower S ARs
for 33f tilted antennas vis vis the vertically held antennas, since the somewhat longer
antennas at 835 MHz get further away from the head, which results in lower SARs for tilted
antennas at this frequency, while a similar effect does not occur at the higher frequency of
1900 MHz because of the smaller lengths of the antennas.

INDUCED ELECTRIC FIELDS AND CURRENTS FOR POWER

FREQUENCY EMFs

With several epidemiologic studies linking electromagnetic fields (EMFs) with

higher rates of incidence of cancer, there is an increasing concern in the public mind
regarding the potential health effects of these fields. An important issue that has come up is
that coupling of the EMFs to the tissues in the human body is poorly understood for
power-line-related frequencies between 10 and 1000 Hz. In the past, a mannequin covered
with copper foils has been used to measure the induced currents for the various sections of
the body by using breaks in the copper foil [23]. A homogeneous saline-filled, reduced-scale
model of the human body (height = 45 cm) has also been used for exposure to vertical electric
fields associated with power transmission lines [24]. Assuming a rotationally symmetric
model of approximate human dimensions (height = 1.83 m or 6 ft) which is represented by
36 spheres, DiPlacido et al [25] have calculated the induced currents for the various sections
of the body for grounded and ungrounded conditions. While the data generated by these
authors are important first steps, and provide approximate understanding of the induced
currents, it is felt that heterogeneous models will provide a proper representation of the
internal organs and tissue properties leading to knowledge of induced currents and internal
electric fields. Such models will also permit inclusion of both the electric (E) and magnetic
(H) fields for a variety of exposure conditions rather than uniform E-field exposures that
have mostly been considered.
We have previously considered a 1.31-cm (nominal 1/2") resolution anatomically
based model of the human body for exposure to purely electric, purely magnetic, and
combined electric and magnetic fields at 60 Hz [26]. For this model we have calculated the
induced electric fields and current densities for exposure to electric and magnetic fields (E
= 10 KV/m, vertically polarized; B = 33.3 nT, oriented from arm to arm) that may be
encountered under ultra-high-voltage power lines. Since the heterogeneous model was
obtained from anatomical sectional diagrams of several cadavers available in a book on
human anatomy [27], it is obviously not as satisfactory as the present millimeter-resolution
model obtained from the MRI scans of an individual. Also, resolution of 1.31 cm for the
various cells is not as good as the 1.974 » 1.974 oo 3 mm cell size used for the present model.
We feel that the present model identifying single tissues for each of the cells is capable of
modeling the tissue interfaces and identifying regions of peak induced E-field and current
densities much better than the previously used coarser model [26]. And yet, it is not possible
to use the millimeter-resolution model of the whole body because of the very large number
of cells (over 6 million) and the large cpu memories that are not easily available to us today.
Recognizing this limitation of the computer resources, we have combined the 3 x 3
X 2 ceils of the mm-resolution model to obtain a new model with resolution of 5.922 x 5.922
X 6 mm along x-, y-, and z-directions, respectively. We have used the previously outlined
frequency scaling approach and the FDTD method [26] to obtain the induced currents for
grounded conditions of this model for exposure to 60-Hz EMFs with incident electric fields
of 10 KV/m (vertically polarized) and for magnetic fields of 33.3 )iT polarized from side to
side of the model. Taken from references 28-30, the conductivities taken for the various
146 O. P. Gandhi andJ. Y. Chen

Table 6. Conductivities of the various tissues

assumed for power-frequency EMFs
Tissue Conductivity
muscle (a,) 0.068
muscle (a^) 0.068
muscle (o,) 0.86
fat'bonc'canilagc 0.04
skin 0.11
nerve 0.12
intestine spleen pancreas 0.12
heart 0.11
blood 0.60
parotid gland 0.11
liver 0.13
kidney 0.16
lung 0.04
bladder 0.10
cerebrospinal tluid (CSF) 1.66
eye humor 1.66
eye sclera 0,11
eye lens 0.11
stomach 0.11
erectile tissue 0.11
prostate gland 0.11
speririatic cord 0.11
testicle 0.11
compact bone 0.04
ligament 0.11
pineal gland 0.12
pituitary gland 0.12
brain 0.12

tissues are given in Table 6. The vertically directed current 1^ passing through the various
sections of the model is shown in Fig. 2. Though the current variation is very similar to that
given earlier [26] and, in fact, also by Deno a number of years ago [23], it is now possible
to identify the regions of the peak induced E-fields and current densities and the magnitudes
of these internal quantities with a degree of precision that was not possible in the past.
In Figs. 3 and 4, we give the calculated maximum magnitudes of the induced electric
fields Ej = (E + E + E)' - and current densities Jj for the various cross sections of the body,
respectively. It is interesting to note that local induced current densities may be as high as
20 mA/m- for the head and trunk and up to nearly 100 mA/m- for the lower regions of the
body (Fig. 4). These are considerably higher than 4 and even 10 mA/m- that have been
suggested in the various safety guidelines [31, 32].

CONCLUSION
Numerical methods have matured to a level that they are being increasingly
used by many laboratories for dosimetric calculations for important and meaningful
bioelectromagnetic problems. For certification of mobile telephones to be within the
ANSI/IEEE C95,1-1992 RF Safety Guidelines, the approach discussed in this paper
may be quite useful. We should also be able to use the numerical approach outlined
here to understand coupling of power-frequency high-magnetic-field sources such as
Millimeter-Resolution Dosimetry for EM Fields 147

;—-T--1 - v r - r - T rr^ri vv

i:ii;

-.1 .- M M
50 100 ir)U 200

Current 1-, ((.lA)

Figure 2. The calculated vertical current passing through the various layers of a 5.922 x 5.922 x ^ mm
MRI-based grounded model of the human body exposed to EMFs at 60 Hz. E = 10 kV m (\ertical) and B =
.l,"?..! |_iT (from side to side of the body).

175

15C

1 00

75

50

25

0 3 0,5 0 (i

Peak electric field

Figure 3. The calculated variation of maximum induced electric fields for each of the cross sections of the
5.922 X 5.922 x 6 mm resolution grounded model. Incident fields are the same as for Fig. 2.
148 O. P. Gandhi andJ. V.Chen

0 U) :^() :!() -10 r> 70 80 <jf)

Peak current density (mA/m-)

Figure 4. The calculated variation of maximum current densities induced for each of the cross sections of the
5.922 X 5.922 x 6 mm resolution grounded model. Incident fields are the same as for Fig. 2.

hair dryers, hair clippers, electric shavers, etc., to the human head. Of particular
interest would be the induced EMFs and current densities for the pineal gland which
has been alleged to be involved in the biological effects of power-frequency HMFs.
With the resolution of the present models being on the order of 11.7 milligratns of
tissue for each of the cells of dimension 1.974 x 1.974 x 3 mm, it is possible to
define even small glands, such as the pineal, with a great deal of precision. The
numerical models tnay also be used for the design/assesstnent of important bioinedical
devices such as implantable cardiac defibrillators, etc.

REFERENCES
1. O, p. Gandhi, "Some Numerical Methods for Dosimetry: Extremely Low Frequencies to .Microwave
Frequencies," Radio Science. Vol. 30, pp. 161-177, January/February 1995.
2. O, P. Gandhi, "Numerical Methods for Specific Absorption Rate Calculations," in Biological Effects and
Medical Applications of Electromagnetic Energy, O. P. Gandhi, Editor, Prentice-Hall, New Jersey, 1990.
3. J. N. Lee and O. P. Gandhi, "Models of the Human Body; A Historical Perspective," invited paper
presented at the Radio-Frequency Radiation Dosimetry Workshop, Brooks Air Force Base, Texas,
December 8-9, 1992; to appear in the Proceedings of the Workshop.
4. O. P. Gandhi. J. Y. Chen, and D. Wu, "Electromagnetic Absorption in the Human Head for .Mobile
Telephones at 835 and 1900 MHz," Proceedings of the International Symposium on Electromagnetic
Compaiibiliiy. (EMC '94 ROMA), Vol. I, pp. 1-5, September 13-16, 1994.
5. D. Wu, O. P. Gandhi, and J. Y. Chen, "Electric Field and Current Density Distributions Induced in a
Millimeter-Resolution Model of the Human Head and Neck by Magnetic Fields of a Hair Dr\'er and an
Electric Shaver," paper presented at the Sixteenth Annual Meeting of the Bioelectromagnetics Society.
Copenhagen, Denmark, June 12-17, 1994.
6. K. S. Yee, "Numerical Solution of Initial Boundary Value Problems Involving Maxwell's Equations in
Isotropic Media," IEEE Transactions on Antennas and Propagation. Vol. AP-14, pp. 302-307. 1966.
Millimeter-Resolution Dosimetry for EM Fields 149

7. A. Taflove and M. E. Brodwin. "Computation of the Electromagnetic Fields and Induced Temperatures
Within a Model of the Microwave Irradiated Human Eye," IEEE Tiansaciions on Microwave Theory und
Techniques. Vol. MTT-23, pp. 888-8%, 1975.
8. A. Taflove and M. E. Brodwin. "Numerical Solution of Steady-State Electromagnetic Scattering Problems
Using the Time-Dependent Maxwell's Equations," IEEE Transactions on Microwave Theory and Tech-
niques. Vol. MTT-23, pp. 623-630, 1975.
9. .\. Taflove, "Application of the Finite-Difference Time-Domain Method to Smusoidal Steady-State
Electroinagnetic-Penetration Problems," IEEE Transactions on Electromagnetic Compatihility, Vol,
EMC-22, pp. 191-202, 1980.
10. K. Umashankar and .\. Taflove, '".A Novel Method to Analyze Electromagnetic Scattering of Coinplex
Objects," IEEE Transactions on Electromagnetic Compatibility. Vol. EMC-24. pp. 397-405, 1982.
11. R. Holland, "TH REDE: A Free-Field EMP Coupling and Scattering ; Transactions on Nuclear
Science. Vol. NS-24, pp. 2416-2421. 1977.
12. K. S. Kunz and K. M. Lee, "A Three-Dimensional Finite-Difference Solution of the External Response
of an .Mrcraft to a Complex Transient EM Environment. The Method and Its Implementation," IEEE
Transactions on Electromagnetic Compatibility. Vol. 20, pp. 328-332. 1978.
13. D. M. Sullivan, O. P. Gandhi, and .\. Taflove, "Use of the Finite-Difference Time-DoiTiain Method in
Calculating EM Absorption of Man Models," IEEE Transactions on Biomedical Engineering. Vol.
BME-35. pp. 179-186, 1988.
14. J. Y. Chen and O. P. Gandhi, "RF Currents Induced in an Anatomically Based Model of a Human for
Plane-Wave Exposures 20-100 MHz." Health Physics, Vol. 57, pp. 89-98, 1989.
15. J. Y. Chen, O. P. Gandhi, and D. L. Conover, "SAR and Induced Current Distributions for Operator
Exposure to RF Dielectric Scalers," IEEE Transactions on Electromagnetic Compatibility. Vol. 33, pp.
252-261. 1991.
16. O. P. Gandhi, Y. G. Gu, J. Y. Chen, and H. I. Bassen, "SAR and Induced Current Distributions in a
High-Resolution Anatoinically Based Model of a Human for Plane-Wave Exposures 100-915 MHz."
Health Physics. Vol. 63, pp. 281-290, 1992.
17. J. Y Chen and O. P. Gandhi, "Currents Induced in an Anatomically Based .Model of a Human for Exposure
to Venically Polarized EMP," IEEE Transactions on Microwave Theory and Techniques. Vol. 39, pp.
31-39, 199l'.
18. ANSI IEEE C95.1-1992, "American National Standard— Safety Levels with Respect to Human Expo-
sure to Radio Frequency Electromagnetic Fields, 3 kHz to 300 GHz," published by the Institute of
Electrical and Electronics Engineers, Inc., 345 East 47th Street, New York. New York. 10017.
19. C. C. Johnson and A. W. Guy, "Nonionizing Electroinagnetic Wave Effects in Biological Materials and
Systems," Proceedings of the IEEE. Vol. 60, pp. 692-717, 1972.
20. M. A. Stuchly and S. S. Stuchly, "Dielectric Properties of Biological Substances - Tabulated,"./oi/rHu/ of
Microwave Power. Vol. 15, No. l,pp. 19-26, 1980.
21. P. J. Dimbylow and O. P. Gandhi, "Finite-Difference Tiine-Domain Calculations of SAR in a Realistic
Heterogeneous Model of the Head for Plane-Wave Exposure from 600 MHz to 3 GHz," Physics in
Medicine and Biology. Vol. 36, pp. 1075-1089, 1991.
22. C. Gabriel, personal cominunication.
23. D. W. Deno, "Currents Induced in the Human Body by High Voltage Transmission Line Electric Field —
Measurement and Calculation of Distribution and Dose," IEEE Transactions on Power Apparatus and
Systems. Vol. 96, pp. 1517-1527, 1977.
24. W. T. Kaune and W. C. Forsythe, "Current Densities Measured in Human Models Exposed to 60-Hz
Electric Fields," Bioelectromagnetics. Vol. 6, pp. 13-32, 1985.
25. J. DiPlacido, C. H. Shih, and B. J. Ware. "Analysis of the Proximity Effects in Electric Field Measure-
ments," IEEE Transactions on Power Apparatus and Systems, Vol. 97, pp. 2167-2177, 1978.
26. O. P. Gandhi and J. Y. Chen, "Numerical Dosimetry at Power-Line Frequencies Using Anatoinically Based
Models," Bioelectromagnelics Supplement I, pp. 43-60, 1992.
27. .A. C. Eycleshymer and D. M. Schoemaker, A Cross-Section Anatomy, Appleton-Century-Crofts, New
York, 1970.
28. L. .\. Geddes and L. E. Baker, "The Specific Resistance of Biological Materials A Compendiuin of
Data for the Biomedical Engineer and Physiologist," Med. and Biol. Engng.. Vol. 5, pp. 271-293, 1967.
29. E. Zheng, S. Shao, and J. G. Webster, "Impedance of Skeletal Muscle from 1 Hz to I MHz." IEEE
Transactions on Biomedical Engineering, Vol. BME-31, pp. 477-481. 1984.
30. C. H. Dumey et al., Radiofrequency Radiation Dosimetry Handbook, Second Edition, Report SAM-TR-
78-22. prepared for USAF School of Aerospace Medicine (AFSC), Brooks Air Force Base, Texas, 78235,
Mav 1978.
150 O. P. Gandhi and J. Y. Chen

31. S. G. Allen. J. H. Bernhardt. C. M. H. Driscoll, M. Grandolfo. G. F. Mariutti. R. Matthes. A. F. McKinlay,

M. Steinmetz, P. Vecchia, and M. Whillock. "Proposals for Basic Restrictions tor Protection Against
Occupational Exposure to Electromagnetic Nonionizing Radiations. Recommendations of an Interna-
tional Working Group Set L'p Under the .'\uspices of the Commission of European Communities." Physicu
Medica. Vol. VII. pp. 77-89, 1991.
32. IRPA, "Interim Guidelines on Limits of H,\posure to 50/60-Hz Electric and Magnetic Fields." Health
Physics. Vol. 58. pp. 113-122. 1990.
 11

CYLINDRICAL MODEL FOR NEURAL

STIMULATION WITH MAGNETIC FIELDS

Maria A. Stuchly' and Karu P. Esselle^

' Department of Electrical & Computer Engineering

University of Victoria
P.O. Box #3055, M.S. #8610
Victoria, B.C., Canada, V8W 3P6
' Electronics Department
Building EGA
Macquarie University
Sydney, NSW 2109, Australia

INTRODUCTION
Electrical excitation of motor neurons in the brain cortex, or peripheral neurons, and
observation of evoked responses have been extensively used in research and medical
practice. Until mid 1980's, excitation had been obtained with current pulses produced either
by implanted or external electrodes applied near the neuron. Magnetic field stimulation offers
the advantages that it is a non-invasive, non-contact method which produces minimal
discomfort to the patient as only low density current flows through skin pain receptors during
the procedure (Amassian et al., 1989; Barker et al., 1987; Freeston et al., 1984). Magnetic
stimulation of brain, spinal cord, and peripheral nerves has been used to diagnose various
medical conditions associated with the abnormal conduction of motor pathways (Barker et
al., 1987; Chokroverty, 1990; Evans et al., 1988; Hallett and Cohen, 1989). It has also been
used in mapping the motor-cortex (Cohen et al., 1988; Benecke et al., 1988).
In magnetic stimulation, the magnitude, sign and time-course of the spatial derivative
of the induced electric field along the axis of the axon (nerve fiber) determine whether
stimulation occurs and where along the axon it occurs (Basser and Roth, 1991; Roth and
Basser, 1990). Depending on the sign of the field derivative the axon is depolarized (negative
derivative) or hyperpolarized. All neurons within the volume where the electric field
derivative is negative and above a threshold value for the given neurons are stimulated. In
clinical practice, it is important to control the location and size of this volume. Furthermore,
it is important to control the location of the volume where hyperpolarization occurs, as it
may block the propagation of the action potential.
The induced electric fields have been modeled for some tissue geometries. Two
approaches have been used to compute the spatial distribution of induced electric fields from
Biological Effects of Magnetic and Electromagnetic Fields, Edited by S. Ueno
Plenum Press, New York, 1996 151
152 M. A. Stuehlv and K. P. Ksscllc

coils. Simplified tissue models such as homogeneous semi-infinite plane (Esselle and
Stuchly, 1992), cylinder (Esselle and Stuchly, 1994), and sphere (Eaton. 1992) have been
used in analytical techniques. Homogeneous and heterogeneous models of varied degree of
anatomical fidelity have been used in numerical techniques, (Branston and Tofts, 1991;
D'Inzeoet al., 1992; Roth et al., 1990. 1991: Ueno et. a!., 1992). The numerical techniques
are the preferred solution when a final accurate analysis is required. However, for compara-
tive evaluation of various exposure conditions and coil configurations analytical solutions
otTer considerable advantages. Firstly, the computer resources and time requirements for
analytical .solutions are significantly smaller than for numerical solutions. This alone makes
analytical solutions exceptionally suitable for optimization of coils used in various medical
applications. Another advantage is that useful physical insight can be gained by in\ estigating
mathematical expressions associated with analytical solutions.
A cylindrical volume conductor model provides a good representation of geometries
associated with stimulation of peripheral nerves. This paper describes the method and results
of coil optimization for this model. An analytical technique is used. A comparison with
numerical computations for the same geometry is also given.

THEORY
Electric Field Due to a Coil Element
This shows a coil carrying a time-varying current /, and an intmitely-long homoge-
neous tissue cylinder with a radius a. The coil has an arbitrary shape and A'number of turns.
Vector iU' represents a small element of the coil located at ) First, we derive an
expression for the electric field produced by this coil element assuming that it is located on
the.v-r plane, ie.cp' = 0 Later, we will generalize it for any tp'The total electric field is obtained
by integrating this expression along the coil.
Under quasi-static conditions (Roth et al., 1991). the primary electric field produced
by the coil-elements is given by (Roth et al., 1990):

li„N{dI/dl)dI' (1
dE" =
AnR

Figure 1. Geometry of the problem.

Cylindrical Model for Neural Stimulation with Magnetic Fields 153

where |.i„ is the permeability of free-space and R is the distance between the coil element and
the point where the field is calculated, (r,(p,z), given by:

R=Jr^+ r'^ -2rr'cos(t) + (z -:'

(2)

The secondary quasi-static electric field, which is a result of the surface charge at the
tissue-air interface, is derived from a scalar potential funcion y as

dE' =-Vv|/
(3)
The function \\i is defined only inside the cylinder and it satisfies Laplace's equation:

VV = 0 (4)

Under quasi-static conditions, the electric field component normal to the tissue-air

dt =-V\i/
interface should be zero inside the tissue cylinder (Polk and Song, 1990). That is:

(5)

When tp' = 0 substitution of (1) and (3) in (5) gives the following boundary condition for
V|/:

{dE"+dE').rl__, = 0 (6)

where dl',. and dl'^ are the - and cp- components of dJ ', respectively.

HoN{dl I dt){di;con<t> + dl'^ sin <p)

1^ JKR

To obtain the solution distance R is expanded in terms of modified Bessel functions.

Since (p is a solution of Laplace's equation, it can be expressed as (Esselle and Stuchly. 1994):

<P = J " £ c o s A ( z - 2')[/l„,(A)cosw0 + fi,„(A)sinm(;>]/„,(AryA, (7)

where.4„,(A.) and B„{}.) are the unknowns to be determined. The boundary condition (equation
5) is used to determine the unknown coefficients.

The Total Electric Field

The total electric field, £, produced by the coil can be obtained by integrating dE
along the coil. For example, the z-component of the electric field, £., is given by
154 M. A. Stuchlv and K. P. Essellc

ii„N{di I dt)di:
4KR

+J"AXsinA(z - z')[\(A)cos m(0 - f)

+BjXyinm{^-0')]l„XXr)dX} ^^^

where dl'. is the r-component of dl ' .

Detailed derivation and closed form expressions for the electric field are given
elsewhere (Esselle and Stuchly, 1994). Because the coefficients that require integration
(summation) of Bessel functions are independent of coordinates of the point in which the
electric field is computed, they can be determined once for a given geometry and position
of the coil, and the results can be stored in memory and re-used in computations of the fields
in multiple locations. This feature and the closed form expressions for the induced fields
facilitate fast computations. For instance, calculation of axial electric field at 500 points takes
less than 5 minutes on a 25 MHz 80486-based PC.

RESULTS
The electric field patterns obtained by this method are qualitatively and quantitively
similar to those previously computed by Roth et al., (1990). using the finite difference
technique. For example, a 5 cm diameter cylinder was exposed to a 5 cm diameter circular
coil carrying a current rising a 100 A/)as. When this lO-turn coil was perpendicular to the
cylinder and the gap between the two was 1 cm, the electric field at a point 6.25 mm beneath
the surface was 39 V'm. When the same analysis was repeated using our method, the result
was 41 V/m.
Stimulating coils of circular, square, double-square (DS) and quadruplesquare (QS)
shapes were analyzed. The coils analyzed were placed parallel or perpendicular to the
cylinder axis or for DS and QS coils with its wings tilted (better conforming to the cylinder
surface). For all results presented, the conducting (tissue) cylinder diameter was 8 cm.
Circular coil diameter or the length of each square coil section was 5 cm. The nerve, parallel
to the cylinder axis, was assumed to be located I cm below the cylinder surface. The
minimum distance between the coil and the cylinder surface was 1 cm. The time rate of
current change in the coil was 100 A/|j.s.
Figures 2 to 6 show contours of the electric field and its spatial derivative along the
direction of the nerve for various coil configurations. The surfaces shown are cylindrical and
are located 1 cm below the surface of the cylinder, i.e. where the nerve to be stimulated is
located. Table 1 gives a summary of the maximum values of the electric field and its
derivative at the nerve location for various coils illustrated in Figures 2 to 6.

DISCUSSION
The analysis of the cylindrical model confirmed the previous findings from a planar
model regarding properties of various stimulating coils (Esselle and Stuchly, 1992a, 1992b).
Cylindrical Model for Xeiral Stimuiallon with Magnetic Fields 155

..4C:'.'

60 90 : 2G
i fde
(deo)
Figure 2. a, Eieclric iielci i;'-iiiid its derivative dEJb::^ fh Fkctric field iLoii a sortace cm beneath ihecyliiider
S'drfiict; for a iioriz-OPtaiiy fnerited circiiinr Cdii, dia. 5 Ci)i. .V = Id.

Single CO! IS of circiiiar or square shape produce two maxima and rwr> minima or8£V5r (Figs.
2 and 3). ..Analogously to the previous findings the localion of the si.imiilation con be easily
and reasonabiy well deicrmincd by Visuai iiispection for square coils, but not for circular
coils. This is becau^JC ibr square coils dse niaxima of j 6Z:-.'(t I occur midemeath the coil
corners clo.se to the coil and somewhat outworth for greater distances (deeper nerves). Sc}uare
coiis also produce a stronger stirnidus for the same current pulse liable 1). Reorienting a
circular or stpiare eoil wiih respect to ihe tissue cylinder, so n is perpendicular, results in
only one polarization and one hyperpolarization spot But the stimulus is much wcalcer, 0.6
kV-m*- compared witir 2.1 kV.'m-' for the sarTic coil, parallel lo the cylinder.
As illiislrated in Figure 5, for a double coil there is one maximum and one minimum
of 6tf ftc. Their magnitudes arc n.early double thai ofa single coil with the same number of
turns in each section (but twice the total iuunbcr of turns) (Table !). This ieafure makes these
COJIS more advantageous than suigle cods. .A quadruple coil is even belter, as it can be seen
frojTi Figure 6 and Tabic 1. There is only one location of the grealeyt 5Zy.''5e, negative or

Table 1. The maximum tLand 8EJSz at the nerve iocation for various coils
Coil f}£, / dz
(V/in) (kV/m^)
single-circular, parallel 78.5 1.8
N = 10
single-square, parallel, 91.0 2.1
N = 10
single-square, perpendicular. ,50,4 0.8
N = iO
double-square, parallel. 167.7 .1.8
N = 2 x 10
quad-square, tilted 110°, 79..i 4.7
N=4X5
156 M. A. Stuchlv and K. P. Essclle

Ej, !jt r-'icm, ffs"!^ honz«"slas sq^ce <:» 1, ni«.>."9l C v*/'''

£
" 0

-2

' /CI '30 ; 83 5C

7^ :*dec)

Figure ?.. a. Electric tlcld c- and its dc!ivsn!>c (>£V& b. Electric field E. for a horizcnialiy oriented square
coii, side .1 en-. N= Kl

l\? at r=3rnn, f*;jfr. «eft;^a^ sgjore eo' , rray-50.^ v^-V &l^/dz af r^3cm. ^rom veriicol sqyore coil, ma5<-S2* v/m^^

E
" 0

30 0' ;8D 90 180

idsc'. » (deg)

Figure 4. a. Electric fieitl £-. and its <i-;ri\;iiive O£J5Z . b. Electric field £_ for a vertically oriented square coil,
side 5 em, A' = 10.
Cylindrical Model for Neural Stimulation with Magnetic Fields 157

DS, ii fjl '-~^>;;ir. 'i^'^gO'", ?=C mcx--'- 67,7 V/n^

b "'
-V/

' "

.
1

" .

; "" -- -

' ''"..--
r'i\ i
v--% - 1 ; :
'V
V^ y , ' " " -

' ^ - ..„''

i K.

^
.0 50 1 80 i rX: 1 80
r ^ '^-^-^i;

Figure S. a, Ficctfic iieid £_ and ITS iicrivative SE/Sz. b. Electric iieid £- for a horizoriaily orienled double
square cell, side ? CIT!, A'= 2 x '.il

Qi, EEOS f™5csn, ^™1 IQB, s^O, TS3*-7'i.^'.'/T QS, dE^/a£ «! r^3£n^ ^ ^ ; 1 0^ s^O, mo»--l,6= k'//"-

30 60 90 1 20 1 50 1 BO
#> ( d e g )

Figure 6. a. Eicctrsc tleid E. and its ilar- aii^ e d..Sr. b^ Electric field t for a quadruple square coil, side 5 cm.
the angle between the 'ifcl'.tjos I iO" inx^y Piane (Fig. 1), ,V= 4 x 5 .
158 M. A. Stuchlv and K. P. Essellc

Table 2. The niaximuiTi E. and hEJdiz for various coils and a heterogeneous model of a
human forearm

Coil dE. 1 dz
(Vim)
single-circular, parallel. 40.1 0.85
N = 10
single-square, parallel, 48.1 1.05
N = 10
double-square, parallel, -iS.S 1.25
N = 2 x 10
quad-square, tilted 110'. 22.1 1.01
N = 4x5

positive, d e p e n d i n g on the directions of the currents in the coil sections. T h e r e f o r e , there is

no ambiguity with respect to the nerve depolorization and its location, once the windings of
the four sections of the coil are properly connected together. Tilting the sections of the coil
to foriTi 110" rather than 180" increases the magnitude of the stimulus.d/T./Sz
Since the anatomy of a human arm is heterogeneous rather than homogeneous,
differences in magnitudes of the electric field and its derivative can be expected from those
shown in Table 1. This issue has recently been addressed (D'lnzeo et al. 1994); An admitance
method (numerical) was used to analyze an anatomically correct three-dimensional model
of a human arm. The resolution was 2.5 mm, and conductivities of the following tissues were
represented: skin, fat, muscle, bone, tendon and nerve. The results are given in Table 2. Lower
values of both the electric field and its spatial derivative are clearly evident when compared
with the corresponding values in Table 1. The reductions are due to the electric charge at the
interfaces between various tissues. However, the superior performance of quadruple coils
remains unaffected.

ACKNOWLEDGEMENT
This work was supported by a grant from the National Sciences and Engineering
Research Council of Canada (NSERC) and by a NATO travel grant.

REFERENCES
1. .'\nuissian, V.E., Cracco, R.Q., and Maccabee, J.P., 1989, "Focal stimulation of human cerebral cortex
with the magnetic coil: .\ cotiiparison with electrical stimulation", Electroencephal. din. Seiirophysiol
74:401-416.
2. Barker, .\.l., E-reeston, I.L., Jalinous. R.,and Jarratt, J.A., 1987, "Magnetic stimulation of the human brain
and peripheral nervous system: An introduction and the results of an initial clinical evaluation". ./,
.\V»mv»g 20:100-109.
3. Barker. A.T., Jalinous, R,. and Freeston, I.L., 1985, "Non-invasive stimulation of human motor cortex",
i.aiuct i:l 106.
4. Basser, RJ. and Roth, B.J., 1991, "Stimulation of myelinated nerve axon by electromagnetic induction".
Med. Biol. Eng. Compul. 29:261-268.
5. Benecke. R., Meyer. B.C., Schonle. P.. and Conrad, B., 1988. "Transcranial magnetic stimulation of the
human brain: Respon.ses in muscles supplied by cranial nerves", E.rp. Bruin Res. 71:623-632.
6. Branston, N M. and Tofts, RS., 1991, "Analysis of the distribution of currents induced by a changing
magnetic field in a volume conductor", Phys. Med. Biol., 36:161-168.
7. Chokroverty, S.. 1990. "Magnetic Stimulation in Clinical Neurophysiology", Boston. M.A . Butieywoiks
Cylindrical Model for Neural Stimulation with Magnetic Fields 159

8. Cohen, L.G., Leili, S.,and Hallett, M.. 1988, "Noninvasive mapping of human motor cortex with magnetic
stimulation", iVt'((m/ogv. 38: 1: 386-387.
9. D'lnzeo, G., Esselle, K.P.. Pisa, S. and Stuchly, M.A.. 1994. "Coinparison ofhoinogeneous and hetero-
geneous models for coil optimization in neural stimulation". Radio Science. 29:in press.
10. D'lnzeo, G., Giacomozzi. C . and Pisa, S. 1992, "Analysis of the stimulation of a ner\c fiber surrounded
by an inhoniogeneous. anisotropic, and dispersive tissue". ACES.I., 7: 180+.
11. Eaton, H., 1992. "The electric field induced in a spherical colume conductor from arbitrary coils:
applications to magnetic stimulation and M F G " , Medical Biol. F.iig. Comp.. 30:433-440.
12. Esselle, K..P. and Stuchly. M.A., 1992a, "Neural stimulation with magnetic fields: analysis of induced
electric fields", IEEE Trans. Biomed. Engineering. BME-39:(i93-700.
13. Esselle, K.P. and Stuchly, M.A., 1992b, "Simplified analytical solutions for magnetic stimulation of
neurons", .-ICfS./., 7:162-178.
14. Esselle, K.P. and Stuchly, M.A., 1994, "Quasi-Static Electric Field in a Cylindrical Volume Conductor
Induced by External CoilsF", IEEE Trans. Biomed. Engineering. BME-41:151-158.
15. Evans, B.A.,Litchy, VV.J.and Daube, J.R., 1988,"Theutility of magnetic stimulation of routine peripheral
nerve conduction studies". .Muscle & .\'er\'e. 11:1074-1079.
16. Hallett, M. and Cohen. E.G.. 1989, "Magnetism: A new method for stimulation ofner\e and brain"../.(,V/,1.
262:538-541.
17. Polk. C. and Song. J.H., 1990, "Electric fields induced by low frequency magnetic fields in inhoniogene-
ous biological structures that are surrounded by an electric insulator". Bioelectmmagneiics 11:235-249.
18. Roth. B.J. and Basser. P.J.. 1990, " A model of the stimulation of a ner\e fiber by eleclroniagnctic
induction", IEEE Trans. Biomed. Engineering. BME-37:588-597.
19. Roth, B.J., Cohen, E.G., Allett, M., Friauf, W., and Basser, PJ.. 1990. " A theoretical calculation of the
electric field induced by magnetic stimulation of a peripheral nerve". Muscle & Nerve. I 3:734-741.
20. Roth, B..1.. Cohen, E.G.. and Hallett. M., 1991, "The electric field induced during magnetic stimulation".
Magnetic Motor Stimulation: Basic Principles and Clinical Experience F.EG Suppl.43:268-278.
21. Roth, B.J., Saypol, .I.M.. Hallett. M.. and Cohen, E.G.. 1991. W theoretical calculation of the electric
field induced in the cortex during magnetic stimulation", Elearoenceplml. Clin. \'ciiroplivsiol.. 8 1:47-56.
22. Ueno, S., Tashiro. T.. and Harada, K., 1988. "Localized stimulation of neural tissue in the brain by means
of a paired configuration of time-varying fields",./. .Applied Physics. 64:5862-5864.
 12

THE APPLICATION OF
ELECTROMAGNETIC ENERGY TO THE
TREATMENT OF NEUROLOGICAL AND
PSYCHIATRIC DISEASES

Brandon Chandos', ArifuUa Khan-^, Henry Lai\ and James C. Lin"**

' Department of Neurology, RG-27

^Department of Psychiatry and Behavioral Science, RP-10
^Department of Pharmacology, SJ-30
University of Washington
Seattle, WA 98195
''Department of Electrical Engineering and Computer Science
(M/C 154) 851 South Morgan Street
University of Illinois
Chicago, IL 60607

INTRODUCTION

Recent progress in the application of electromagnetic energy has revolutionized

many areas of medicine. Nuclear magnetic resonance imaging (MRI) continues as a leading
noninvasive modality for diagnostic examination of tissues to detect or localize disease.
Other emerging functional imaging modalities include magnetocardiography and magneto-
encephalography. Magnetic stimulation for the diagnosis of neurological disorders repre-
sents yet another noncontact application of electromagnetic energy that is gaining in
popularity (Lin, 1993). Magnetocardiography (MCG) is a promising noninvasive modality
for obtaining functional information on the electrical activity of the heart. The small
biomagnetic signals (I pT or less) are recorded using multi-channel superconducting
quantum interference devices (SQUID) with subjects in the supine position and the SQUID
sensors placed directly over the thoracic surface (Stroink, 1989). It has been explored as a
mapping tool to localize noninvasively cardiac electrical sources responsible for atrial flutter
and ventricular fibrillation (Ribeiro et al., 1992, Williamson et al., 1989). In a like manner,
high resolution magneloencephalography (MEG) measures the minute magnetic fields
generated by the ionic currents in the brain (Barth, 1993). MEG correlates well vvith abnormal

To whom correspondence should be sent.

Biological Effects of Magnetic and Eleciroinagiietic Fields, Edited by S. Ueno

Plenum Press, New York, 1996 161
162 B. Chandos et al.

electrical activity measured by electroencephalography (EEG), and sometimes shows abnor-

malities not seen on EEG. It has been successfully used to localize cerebral sources of
electrical activity (Pantevetal., 1991;Sandykand Anninos, 1992; Squires etal., 1990; Stefan
et al., 1989). MEG is more accurate in spatial localization, as the magnetic field passes
through the scalp and skull unimpeded, while the electrical signal of EEG is attenuated and
dispersed.
In magnetic stimulation of the nervous system, a time-varying magnetic field,
produced by passing a current through a wire coil, gives rise to an induced electric field in
proximity to excitable tissues of the central and peripheral nervous system. The currents are
delivered as 50-200 microsecond pulses with peak values of several thousand amperes. The
technique is used to study the neuromuscular system in humans (Amassian et al., 1989; 1990:
Chokroverty, 1990; Evans, 1991; Ueno et al. 1991). The principal advantages of magnetic
stimulation are that it is noninvasive and less painful than applying electrical currents through
surface electrodes, and it has the ability to reach nerves lying well below the skin surface.
The main disadvantages are that magnetic stimulation is not selective enough to restrict the
region of excitation and it has relatively poor controllability and reproducibility.
This paper reviews the pathophysiology of several common neurologic and psychi-
atric diseases, and the effect of electromagnetic field (EMF) on these conditions. Studies of
this kind are sparse and are mostly not yet replicated in a different laboratory. Other recent
diagnostic and therapeutic applications of electromagnetic energy have been described
elsewhere (Hand and Cardossi, 1993; Lin, 1993 a; b).
The brain is an electric organ. Its major cellular components, the neurons, can be
viewed as combinations of capacitors in parallel. The lipid dielectric between the neuronal
membranes maintains a potential of-60mV in the cell's resting state. The change in potential
is mediated by transmembrane flow of ions. Electromagnetic fields of different frequencies
could affect this system by induced-current or direct field effect on molecular interactions.
The therapeutic effect of EMF has been studied on a variety of neurologic diseases. In these
studies, extremely low frequency (ELF) magnetic or electric fields have been used. Thorough
clinical trials assessing the true efficacy of EMF in specific neurological or psychiatric illness
are lacking. However, experimental models and some studies on humans offer intriguing
results.

EPILEPSY
Epilepsy is a neurologic disorder characterized by paroxysmal attacks of brain
dysfunction due to excessive neuronal discharge. This abnormal activity is often associated
with trauma to brain tissue, or to genetic predisposition. The physical and cognitive
manifestations of these discharges vary depending on their locations in the brain, and to what
extent the abnormal activity spreads to adjacent tissue. This abnormality depends on the
extracellular environment, the supporting cell populations, and interaction with other neu-
rons (Schwartzkroin, 1993). The electronic structure of the neuron is composed of both
"passive" membrane structural components and an active component of transmembrane
potential mediated by intra- and extracellular ion concentrations and their fluxes. The ions
involved in electrical activity are potassium, sodium, chloride, and calcium. Anticonvulsant
medications affect the ion channels and modulate the repetitive discharges which occur in
hyperexcitable neurons during epilepsy.
A recent case report involved treatment of a 20 year old female with idiopathic
seizure disorder, with onset in childhood (Sandyk and Anninos, 1992). She was having
generalized convulsions and averaged three times a day while on anticonvulsant medi-
cations with therapeutic serum levels. She also had intellectual decline and violent
Treatment of Neurological and Psychiatric Diseases 163

behavior directed against herself and others. An interictal EEG was normal. External
magnetic fields were applied for 2 minutes on 3 consecutive days. The patient was
continued on her anticonvulsant medication (Valproic acid). She was seizure free for 2
weeks. She was then given a device to administer "magnetic smoothing" treatments at
home each evening. At a 3 month follow-up, her seizure frequency declined to 1 per
week, and her aggressive behavior had abated. Time spent in sleep also increased 1-2
hours a day.

MULTIPLE SCLEROSIS
Multiple sclerosis (MS) is classified as a demyelinating disease, with onset in young
adulthood. Diagnosis can be made on a clinical basis. Initial physical complaints are often
paresthesias and visual changes: decreased visual acuity or diplopia. Other neurologic
systems may be affected, and the patient may have weakness of one or more extremities,
with loss of balance and difficulty or inability to walk. Urinary incontinence, fatigue and
depression are often features. The disease is characterized by a waxing and waning course
of either of two patterns. One is acute exacerbations of symptoms with resolution, and the
other is a chronic progressive course. Diagnosis can be confirmed by Magnetic Resonance
Imaging (MRI) of the brain which on T2 signal will show demyelinating lesions, often
involving the corpus callosum. Visual evoked potential studies show slowing of cortical
electrical response to visual stimuli.
The exact mechanism of the disease is unknown. It is generally thought to be an
autoimmune response, perhaps related to past viral exposure. Some patients respond favor-
ably to treatment with steroids during an exacerbation (methylprednisolone or ACTH). It is
not clear that the demyelinating lesions on MRI account for the clinical course of the disease,
and there is speculation that the clinical manifestations are due to impaired serotonin activity
(Sandyk and lacono, 1993). A cohort study of 25 patients with multiple sclerosis found pineal
gland calcification in 96%, abnormal melatonin levels in 52%, and abnormal alpha-melano-
cyte stimulating hormone in 70% (Sandyk and Awerbuck, 1992).
Several cases of treatment of multiple sclerosis with magnetic fields have been
reported in detail (Sandyk 1992; Sandyk and lacono, 1993; Sandyk, 1994). The main
features of these cases will be mentioned here. The patients were age 50-64 with a history
of MS who were referred for experimental treatment based on their poor physical condition
and failure to respond to other treatments. They were treated with an external array of
magnetic field generating coils at 7.5 pT, 5 Hz, for durations of 20 to 30 minutes. After
a single treatment, and over a period of several days, the patients reported significant
(50-60%) improvement in all symptoms. This included improved vision, balance, sleep,
clearness of thinking and speech, One measure of improvement was pre- and post-treat-
ment drawing tests for assessment of motor control and psychological constructional
ability. Patients were asked to draw an Archimedes spiral and a house. In pre-treatment
drawings, the lines were irregular and showed poverty of content. In post-treatment
drawings the lines were more regular and the pictures more detailed. More objective
evidence of a change in neurologic function was obtained in one patient by measurement
of visual evoked potentials (VEP). Prior to treatment, the latency response time of the
occipital cortex to stimulation of the eyes with a flashed checkerboard light pattern was
slowed in both eyes to an abnormal range. The VEP was remeasured and showed a
normalization of latency 24 hours after treatment. Up to a 3 month follow-up was done
showing the improved condition to be stable.
164 B. Chandoset al.

PARKINSON'S DISEASE

Parkinson's Disease is a movement disorder witii peaic onset in the sixth decade of
life. It is characterized clinically by a resting tremor, muscle rigidity, akinesia, stooped
posture and masked faces. The disease is progressive and can be debilitating. There is no
clear etiology for this disease. Pathology shows loss of pigmented cells in the substantia
nigra and other pigmented nuclei (Adams and Victor, 1993). This is associated with a loss
of dopamine producing cells. The mainstay of symptomatic treatment is dopamine replace-
ment therapy with the dopamine precursor L-dopa.
The symptoms of Parkinson's Disease are associated with an increase in activity of
striatal neurons which project to the globus pallidus. The system is normally balanced by
the inhibitory nigral dopamine input interacting with the excitatory cortical glutamate input.
In this disease the excitatory input dominates, and may contribute to induced cell death along
the striatal pathway (Mitchell et al., 1994). The substantia nigra has elevated iron levels
compared to other areas of the brain which is bound by neuromelanin, with electron
paramagnetic resonance showing unbound irons to be in paramagnetic valence states (Zccca
and Swartz, 1992). Iron contributes to oxidative stress in the brain tissue by being a catalyst
for lipid peroxidation reactions, with resulting alteration in calcium hemostasis (Youdin and
Ben-Shachar. 1991).
There are reports of low-level magnetic fields improving the symptoms of Parkin-
son's Disease. (Sandyk, 1992). We will summarize one typical case. A 71 year old man had
a 3 year history of progressive symptoms of Parkinson's Disease. At the time of treatment
he had moderate generalized bradykinesia, mild resting tremor, frontal lobe release signs,
oily face and forehead, hypophonic speech, decreased blink rate of 5-8/min, micrographia,
cogwheel rigidity, and stooped posture. He had been treated with the medication bro-
mocriptine (16 mg/d) with mild improvement of symptoms. He was treated with a sham
exposure session with no change in symptoms. He was then treated with a 2-7 Hz magnetic
field at 7.5 pT for seven minutes. Ten minutes after initial treatment, the patient's blood
pressure dropped from 140/80 to 130/70 mm Hg. Heart rate dropped from 76/min to 68/min.
Thirty minutes after treatment facial muscles showed improved expression and blink rate
had increased to 18-21/min. His voice was louder and he was less bradykinetic. Posture had
improved. There was complete resolution of his hand tremor. The initial benefit lasted 4
days. The patient assessed his improvement at 60%, which remained stable for three months
with weekly magnetic treatments.
Further assessment of cognitive improvements were done using picture drawing and
a word-fluency test (Sandyk, 1994). Subjects were asked to produce a list of words beginning
with the letters "S" and "C". Parkinson patients tended to show a decline in the number of
words produced; about half of that expected when compared with controls. After a series of
two magnetic field treatments, word fluency on the same tasks improved to normal range.
When drawing objects such as a daisy flower or a bicycle, the pre-treatment subjects had
drawings with irregular lines and poverty of content. Post-treatment subject showed
smoother lines and more detailed drawings; an indication of improved tremor and visuocon-
structional functioning. Micrographia has also been shown to improve with magnetic field
treatments (Sandyk and locono, 1994).

STROKE
A stroke is a sudden focal neurologic deficit. This is usually caused by the
interruption of oxygenated blood flow to an area of the brain by an embolus or thrombosis
Treatment of Neurological and Psychiatric Diseases 165

of an artery. The neurons in that region undergo ischemic necrosis. Treatment of stroke
was traditionally providing supportive medical care and treating the risks for vascular
disease or cardiac disease. It was suggested that the neurons surrounding the core area
of ischemic damage were being injured by glutamate which was being released by the
excess calcium released by the dying cells (Choi, 1990). These surrounding cells, dubbed
the "ischemic penumbra", were thought to be potentially salvageable. Indeed, several
pharmacologic agents have been known to decrease the volume of brain injury, and
clinical trials are ongoing (Ginsberg, 1993). Many of these agents block the release of
excess glutamate by preventing calcium influx to neurons. The presence of iron in the
brain also contributes to propagation of injury by acting as a cofactor in the lipid
peroxidation reaction. This produces malignant free radicals which the brain is poorly
equipped to deal with, due to the low level of the endogenous free radical scavenger
superoxide dismutase (Davalos, 1994). This mechanism is also an area of research in
pharmacologic intervention.
Electromagnetic fields have been shown to alter calcium flux in the brain (Bawin and
Adey, 1976; Blackman, 1988). This led to attempts to reduce the amount of brain injury from
stroke in animal models using electromagnetic fields. Using a pulsed electromagnetic field
(parameters: 27.1 MHz, 585 W peak power, 65 us pulses, 400 pulses per second) on rats
undergoing permanent middle cerebral artery occlusion, it was found that animals treated
for 2 hours had reduced edema, but no significant reduction in the volume of injury
(Rappaport and Young, 1990). In a more recent study, using a model of transient ischemia
in the rabbit, certain zones of ischemia in the treated group showed marked improvement,
but overall injury was not significantly different (CJrant, 1994). The field used was 2.8 mT
at 75 Hz with pulse width of 1.3 ms. Of additional interest was that somatosensory evoked
potentials were done to assess ischemic change, and the magnetic field treated animals tended
to show less attenuation of conduction compared to controls.

ELECTROMAGNETIC FIELDS, MELATONIN PRODUCTION AND

PSYCHOLOGICAL DISORDERS
Secretion of the hormone melatonin by the pineal gland in vertebrates is affected by
exposure to EMF (Reiters, 1992; 1993). Visible light had been shown to inhibit melatonin
synthesis. Acute exposure to light induces aT 1/2 of 10 minutes. During periods of darkness
the levels increase 5-10 fold. This response is mediated via the visual system. Light on the
retina of the eye is transduced to an electrochemical signal communication with the pineal
gland via suprachiasmatic nuclei of the hypothalamus, preganglionic neurons in the inter-
mediolateral column of the upper thoracic spinal cord, and post ganglionic neurons in the
superior cervical ganglion. The release of norepinephrine in the pineal gland augments the
production of cAMP and melatonin, and may be mediated by protein kinase C (Ca'"
activated).
Different magnetic field manipulations have been shown to reduce melatonin levels
by 50% (Reiters, 1992; 1993) Experiments in rodents have shown intermittent static
magnetic fields interfered with the nocturnal metabolism of melatonin. An inversion of the
geomagnetic field reduces pineal cAMP and melatonin production. Instantaneously inverted
fields with induced current depress pineal melatonin production. Since 50-60 Hz fields also
affect melatonin production, conceivably, magnetic fields can be used to alter melatonin
metabolism for the treatment of various psychological disorders, such as changes in circadian
cycle.
166 B. Chandoset al.

ELECTRICAL STIMULATION IN MENTAL ILLNESS

Although controversial, as many of 50,000 Americans receive electroconvulsive

therapy (ECT) annually so that they obtain relief from several depression and related illness.
Because of such frequency of use, its persistence in clinical practice, and surrounding
controversy we will review ECT in some depth followed by sections on use of ECT in mental
illness.
In 1938, Cerletti and Bini from Italy proposed the use of electrical stimulus to induce
grand mal seizures, which from 1934 until then were induced by drugs such as camphor, to
"cure" schizophrenia. Both patient and physician acceptance was so high that the newly
"discovered" ECT rapidly spread through war-torn Europe and North America within a few
years. Research in the past 25 years has led to the conclusion that ECT is probably the most
effective for depression although it may also be effective in many types of other mental
illnesses such as mania, schizophrenia, delirium, and neuroleptic malignant syndrome (Kahn
etal., 1993).
Besides induction of grand mal seizures, electrical stimuli were used to induce
electro-narcosis and electric sleep to treat mental illness. However, neither of these attempts
were shown to be effective when studied systematically (Similar to insulin coma, psycho-
surgery or hydrotherapy). Type of electrical stimulus and its delivery for ECT have under-
gone considerable development. Research in the I960's and 1970's convincingly showed
that ECT's unwanted effects such as loss of memory, disorientation, and delirium were
significantly related to the amount of electrical energy used to induce seizures (Fink, 1989).
Because of concern over ECT's adverse effects, newer and possibly more benign electrical
stimulus were considered. The past 20 year's research has focussed ways to deliver the lowest
but effective amount of electrical energy to induce seizures. The present commercially
available ECT machines offer the following types of delivery system. Instead of sine waves
the newer machines deliver alternating current (40-180 Hz) as brief pulses (20 to 100 mA)
ranging in duration from 0.5 to 2 ms. The electrical charge can be delivered at intervals of
0.5 to 3.0 seconds. Administration of 8 to 12 ECT's in a period of 3 to 4 weeks leads to
patient improvement.
As a rule, such transcranial stimulation needs to overcome a resistance of 100 to 300
ohms, and in humans grand mal seizures are induced with a electrical current of about 150
to 250 mA. However, recent data suggest that men generally need a larger charge, and there
is a linear relationship between increasing age and electrical charge to induce seizure. A
significant proportion of older persons need 2 to 5 fold higher energy.
One of the unresolved issue is the mechanism by which ECT works. Accepted
scientific rationale relies on the assumption that seizures per se may be the relevant factor.
This is based on the finding that seizures induced by such chemical means as camphor,
pentyeneterazol, indoklon, or ECT are only effective when they lead to grand mal seizures.
However, recent data suggest that induction of a grand mal seizure with very low levels of
electrical stimulus (50 to 100 mA) by placing electrodes on one side of the head may not
lead to any patient improvement (Sackeim et al, 1991).
Although many psychological theories were proposed for the reason by which ECT
works, they have lost support since many excellent studies have shown that a electrical
stimulus followed by a grand mal seizure are needed for clinical improvement. Thus, most
physicians consider that biological effects are the key to ECT's effects. Attempts at a detailed
study of human brain at autopsy in patients who have had ECT shown no structural changes
(Devanand et al, 1994). Moreover, large spike-like increases in most brain neurotransmitters
such as norepinephrine, serotonin, acetylcholine, and dopamine occur immediately after
ECT (Khan et al, 1993). Repeated ECT administration, as in routine medical practice, leads
Treatment of Neurological and Psychiatric Diseases 167

to a decrease in concentration of receptors for most of these neurotransmitters and is one of

the leading contenders as to how ECT works since most antidepressants also lead to such
effects. An alternative possibility is the stimulation and re-regulation of hypothalamic-pitui-
tary endocrine functions since ECT leads to major changes in this system. A third possibility
includes reorganization and resynchronization of circadian rhythms as a possible mechanism
by which ECT works.
An alternative approach to induce seizure for the treatment of depression is the use
of rapid-rate transcranial magnetic stimulation (George and Wassemian. 1994). This ap-
proach could also allow selective stimulation of brain sites which are involved in depression
and thus reduce side-effects, e.g. memory deficits, due to electrical disruption of function in
unrelated sites.

CONCLUSION
While the application of ELF electromagnetic fields to neurologic and psychiatric
disease may appear promising, it is in its infancy. A great deal of research must be conducted
to demonstrate its safety and efficacy. The mechanism of interaction of magnetic fields with
the nervous system is still unclear. Apart from seizure induction, the flux density used in the
cited studies was very low (pT), the effects seemed to occur very fast and were long-lasting.
Magnetites have been identified in cells in the human brain (Kirschvink et al., 1990),
however, it has been calculated that magnetic fields higher than 5mT are required to generate
sufficient effect on these particles (Adair, 1993). There may be a common pathway as to how
EMF improves the conditions in the mentioned diseases. The role of calcium is fundamental
in neuronal functioning and may play a key role in elucidating mechanisms of effect. It's
involvement was mentioned in all the diseases, and could probably be associated with the
neurotransmitter glutamate. Iron also plays a key role in the pathophysiology of neurologic
disease (Sachdew, 1993), but to what extent magnetic field affects iron metabolism in the
main remains to be elucidated. Endogenous opioids may also play a role. Magnetic fields
have been shown to activate endogenous opioids in the brain (Lai, 1993). In addition,
naltrexone has been shown to attenuate the antiparkinsonian effects of magnetic field
(Sandyk, 1994). Magnetic field also shows an effect on immune responses, which may be
significant in multiple sclerosis (Tanovic, 1991).
An important criticism of the human case reports is the lack of control subjects. Even
though, the treatments included sham exposures, late improvement in tasks could be in part
from practice. In most cases, the small numbers yield weak statistical power. Animal models
of stroke are variable in reproducibility, and further systematic studies of different treatment
parameters would be beneficial. While some mechanisms of action have been proposed, their
substantiation is needed.

REFERENCES
Adair, R. K., 1993, Effects of ELF magnetic fields on biological magnetite. Bioelectmmagiieiics 14; 1-4.
Adams. R., and Victor, M., 1993, Degenerative Diseases of tlie Nervous System. In: Principles of Neurology,
,5th ed. New York, N.Y. McGraw-Hill; 978-979.
Amassian, V. E., R. Q. Cracco, and J. P. Maccabe, 1989. Focal stimulation of human cerebral cortex with the
magnetic coil; A comparison with electrical stimulation, Electroencep/m! din Neyrophvsiol. 74:
401-416.
Amassian, V. E., R. Q. Cracco. and P. J. Maccabee, 1990, Perspectives in stimulation of human nei%'ous system
with magnetic coils. In: M. E. O'Connor etal.. eds.. Emerging Electromagnetic Medicine. New York.
Springer-Vcrlag: 171-204.
168 B. Chandoset al.

Barth, D. S., 1993, Magnetoencephalography. In; Wyllie E. The Treatment of Epilepsy. Principles and Practice
Philadelphia PA. Lea & Febiger: 285-297.
Bawin. S. M., and Adey. W. R.. 1976. Sensitivity of calcium binding in cerebral tissue to weak environmental
electromagnetic fields. frac.Vuf/.'irarfSf/. 7.3: 1999-2003.
Blackman, C. E., 1988. Influence of electromagnetic fields on the afllux of calcium from brain tissue in vitro.
Bioeteciromagnetic.s 9; 215-227.
Choi, D. W., and Rothman. S. M.. 1990, The role of glutamate neurotoxicity in hypo.xic-ischemic neuronal
iiea\)c\. Anna Rev neurosci. 13: 171-182.
Chokroverty, S., 1990, Magnetic Stimulation in Clinical Neurophysiology. Boston. Butterworth.
Davalos. .4., ct al.. 1994, Iron-related damage in acute ischemic stroke. Stoke 25: 1543-1546.
Devanand. D. P. et al.. 1994. Does ECT alter brain structure'Mm 7 Mm/i 151: 957-970.
Evans. B. .A.. 1991. Magnetic stimulation of the peripheral nervous system, J C//M Meurophysiol. 8: 77-84.
Fink M.. 1979. Convulsive Therapy: Theory & Practice. New York: Raven Press.
George. M. S.. and Wassermann, E. M.. 1994. Rapid-rate transcranial magnetic stimulation and ECT.
Convulsive Thcrcipv 10: 251-254.
Ginsberg, M.. 1993. Emerging Strategies for the Treatment of Ischemic Brain Injury. In: Molecular and Cellular
Approaches to the Treatment of Neurologic Disease. Edited Waxman, S. G. New York. N.Y Raven
Press.
Grant. G.. Cadossi, R.. and Steinberg. G., 1994, Protection against focal cerebral ischemia following exposure
to a pulsed electromagnetic field. Bioelecttvmagnetics 15: 205-216.
Hand. J. W.. and Cardossi. R.. 1993. Therapeutic applications od electromagnetic fields. In: Review of Radio
Science 1990-1992. Edited Stone, W. R. Oxford, U.K.. Oxford University Press.
Jancovic. B.. et al. 1991. Magnetic fields, brain and immunity: Effect on humoral and cell-mediated immune
responses. Inlern J Neumscienc. 51: 25-43.
Khan. .\.. Mirolo. M. H.. Hughes, D., and Bierut. L., 1993, F-lectroconvulsive therapy. Psvch Clin N Ainericu
16:497-513.
Kirschvink. J. L.. Kirschvink. A. K. and Woodford, B., 1990. Human brain magnetite and .squid magnetonietry.
Ann Conf IEEE Engin Med Biol Soc. 12: 1089-1090.
Lai. II.. et al.. 1993. EfTects of 60 Hz magnetic field on central cholinergic systems of the rat. Biocleclromag-
neiic.s 14: 5-15.
Lin. J. C . 1993a, Diagnostic applications od electromagnetic fields. In: Review of Radio Science 1990-1992.
Edited Stone, W. R. Oxford, U.K. Oxford University Press.
Lin. J.C., 1993b, Microwave technology for minimally invasive interventional procedures. Chinese J. Med
Biol. Engg. 13:293-304.
Mitchell. I., et al.. 1994. Glutamate-induccd apoptosis results in a loss of striatal neurons in the Parkinsonian
rat. .\'eiiroseienee 6i: 1-5.
Pantev. C.. Gallen. C . Hampson, S., Buchanan, S., and Sobel. D. 1991, Reproducibility and validity of
neuromagnctic source localization using a large array biomagnctometer. Am JEEG Teehnol. 31:
83-101.
Rappaport 2. and Young, W., 1990. EtTect of pulsed electromagnetic fields on calcium tissue changes in focal
ischemia. Neiirologiecil Reseiireh 72: 95-98.
Rcitcr. R, J.. 1992. Alterations of the circadian melatonin rhythm by the electromagnetic spectrum: A study in
environmental toxicology. Regul Tox and Pharm. 15: 226-244.
Rcitcr. R. J.. 1993. Electromagnetic fields and melatonin production. Biomed & Pharmaeother. 47: 439-444,
Ribeiro. P. C. ct al. 1992, Detection of reentry currents in atrial flutter by magnetocardiography, IEEE Tnin\
Biomed Eng. 39: 818-824.
Sachdev. P.. 1993. The neuropsychiatry of brain ITOTI. J Neuropsychiatrie Clinical Netiroseienee 5: 18-29.
Sackheim. 11. A.. Devanand D, R. and Prudic J.. 1991, Stimulus intensity, seizure threshold and seizure
duration: Impact on the elTicacy and safety of ECT. Psych Clin M America 14: 803-845.
Sandyk. R.. and Anninos. P. A.. 1992. .Attenuation of epilepsy with application of external magnetic fields: \
case report. Intern J Neiiroscience 66: 75-85.
Sandyk. R.. 1992. Successful treatment of multiple .sclerosis with magnetic fields. Intern J Xeiimscience 66:
237-250.
Sandyk. R., and Awerbuck, G., 1992, Nocturnal plasma melatonin and alpha-melanocyte stimulating homione
levels during exacerbation of multiple sclerosis. Intern J Neuroscienee 67: 173-178.
Sandyk. R., and lacono, R., 1993. Resolution of longstanding symptoms of multiple sclerosis by applications
of picotesla range magnetic fields. Intern J Sciiroscience 70: 255-269.
Sandyk. R., 1994, Rapid normalization of visual evoked potentials by picotesla range magnetic fields in chronic
progressive multiple sclerosis. Intern J Neuroscienee 11: 243-259.
Treatment of Neurological and Psychiatric Diseases 169

Sandyk. R., 1992. Magnetic fields in the therapy of Parkinsonism. Inlern J Seumscience 66: 209-23.5.
Sandyk. R.. 1994, Improvement in word-fluency performance in Parkinson's disease by administration of
electromagnetic fields. Intern J Neiimscience 11: 23-46.
Sandyk, R., and locono, R., 1994, Reversal of micrographia in Parkinson's disease by application of picotesal
range magnetic fields. Inlern J Neuroscience 11: 11-S4.
Sandyk. R., and locono, R.. 1994, Naltrexone attenuates the antiparkinsonian effects of picotesla range
magnetic fields. Intern J Neuroscience 78: 111 -122.
Schwartzkroin. P. A., 1993, Basic Mechanisms of Epilepetogenesis. In: Wyllie. E., The Treatment of Bpilepsy:
Principles and Practice. Philadelphia Pa. Lea & Ferbiger: 83-98.
Squires, K. C , Buchanan. S., and Schwartz, B. 1990, Localization of cortical function by analysis of
multi-channel magnetic field recordings, J Japan Biomagnet Bioelectromagn. Soc. 3: 1.S9.
Stefan, H., et al. 1989, Biomagnetic multichannel measurements in the preoperative diagnostic evaluation of
ejiWcpsy. Electrnmedica, 57: 129-133.
Stroink, G. 1989, Principles of Cardiomagnetism. In: S. J. Williamson et al. eds.. .advances in Biomagnetism,
New York. Plenum Press: 47-56.
Ueno. S., Tashiro, T. Kamisc, S. and Harada, K. 1988, Localized stimulation of neural tissues in the brain by
means of a paired configuration of time-varying magnetic fields. J Appl Physics 64: 5862-5864.
Ueno. S., Masuda, T. and Fujiki, M. 1991, Functional mapping of the human motor cortex obtained by focal
and vectorial magnetic stimulation of the hxmn.IEEE Trans Magn 26: 1539-1544.
Williamson. S. J., Hoke. M., Stroink, G.. and Kotani, M. eds. 1989. Advances in Biomagnetism. New York.
Plenum Press.
Youkim. M. B. H., Ben-Shachar, D., and Riederer, P., 1991, Iron in brain dysfunction with emphasis on
Parkinson's disease. Eur Neurol. 31: 34-40.
Zecca, L.. and Swartz. II. M., 1993, Total and paramagnetic metals in human substantia ncgra and its
neuromelanin. yA'fHra/ Transm. 5: 203-213.
 13

GENETIC ANALYSIS OF IRON

BIOMINERALIZATION

Tadashi Matsunaga, Chikashi Nakamura, and Yuko Hotta

Department of Biochemistry
Tokyo University of Agriculture and Technology
2-24-16 Naka-cho, Tokyo, Japan

1. INTRODUCTION

Some species of bacteria possess the ability, termed a taxis, to move from an
unsuitable environment to a more suitable one. Chemotaxis is the ability to detect a specific
chemical substance and then swim either away from or towards it. A similar ability with
respect to light is termed phototaxis. Some of bacteria swim in the direction of a magnetic
field, and they were therefore called magnetotactic bacteria. The intracellular magnetic
particles synthesized by a magnetotactic bacterium form a chain-like structure which
functions like a single bar magnet, causing the bacterium to orient in an external magnetic
field. As a result of this orientation, the bacterium is directed to swim towards microhabitats
of lower oxygen concentration deeper in the sediment of lakes, swamps and other aquatic
habitats due to the inclination of the earth's magnetic field. North-seeking bacteria swim
downwards in the northern hemisphere and upward in the southern hemisphere (south-seek-
ing bacteria do to opposite). However, culture of magnetotactic bacteria is now possible, and
cultured magnetotactic bacteria swim towards both poles. Magnetotactic bacteria have also
been discovered which can grow under conditions of high oxygen concentration, although
they do not synthesize magnetic particles under these conditions. Since the discovery of
these bacteria, the category of magnetotactic bacteria appears less clearly defined, and
perhaps a more useful category would be magnetic bacteria, that is, all bacteria which
synthesize magnetic materials. A further area of uncertainty is the reason why magnetic
bacteria synthesize magnetic particles. One suggestion is that under anaerobic conditions it
is necessary for these bacteria to metabolize iron, and magnetic particles are simply
by-products of the metabolism. However, research has not yet provided any clear information
about bacterial magnetic particle synthesis or related aspects of metabolism.
Magnetic materials have also been found in migratory fish, such as salmon and tuna,
and in other higher organisms. Magnetic materials have been extracted from the heads of
these types offish, and it has been proposed that these materials function as compasses to
enable the fish to determine direction in order to migrate vast distances across the open ocean.
It has been suggested that the magnetic material may form a composite structure which
Biological Effects of Magnetic and Electromagnetic Fields, Edited by S. Ueno
Plenum Press, New York, 1996 171
172 T. Matsuga et al.

funciions as a mfigiiefosensory organ, ilowever, it is unclear whether these inagnetoseiisory

organs aciually exist. There is stili iilt'c known about the possession and synthesis of
magnetic materials in fish, but ihey arc vciy similar to ihe magnetic particies of tiiagneiic
bacteria, and is has therefore been suggested that there is an e'."oliitionary relationship
between synthesis of triagnetic materials in bacteria and in animals. The evolutionary
relationship between niagsictic bacteria and higher animals has not been confirmed, but the
mutual ability to synthesize magnetic rnarerials is an important point they have in common.
We hope that by invesiigaiing the geaes that code for enzymes and proteins involved in the
.synthesis of magnetic maierials, the connection between magnetic materials in various
organisnis can, be elucidated.
Some bacteria isolated from fresh water synthesize intraceliidar magnetic particles
\vhich consist of magnetite. MagneiaspinHuifi magneiotacticum MS-I (1), Mas^netospirii^^
luti! sp, MGT-! (15). Magaeiospif'iliuin sp, A^4B-l (Fig.!) (14) and Magnetospiiiiium
gryphiswaidense (20), Phylogeneiic analysis of 16S rR'NA sequences oiMagneiuspihijum
sp. iiave shown their close evoluucnary relationships to some photosynthetic bacteria (2,9).
We arc presently using AMB-l as a model system K>r the study of magnetite
biotmnerahzation at the molecular genetic level(l3). In Oils review, we describe a genetic
approach to mvestigatuig AMIh4, and analysis of the Mag.4 protcui of .\MB-1, which is
thought to be invcived in the rriagneiic particles synthesis.

2. GENE TRANSFER AND TRANSPOSON MUTAGENESIS IN AMB-1

Because a system for gene transfer into magnetic bacteria has not been studied, a
basic molecular genetic approach was required. Among magnetic bacteria, Magnetospirillmn
magneotacticuin MS-1 was the first isolated, but it can only grow under microaerobic
condhions. Magnetospirillum sp. strain AiVIB-1, however, can grow under microaerobic or
aerobic conditions, and forms colonies on agar. To investigate the mechanism of magnetite
biominerali^ation in magnetic bacteria, we chose AMB-1 as our study material.
To search for genes involved in magnetite synthesis, we used transposon mutagenesis
of AJVlB-1. Transposon Tni is inserted into the genomic DNA at only one point, and
transconjugants have kanamycin resistance. When one of the genes involved in magnetite
synthesis is cleaved by TnJ, the traosconjugant will lose the ability to synthesize magnetite.

Figure 1, Transmission electron micrograpti

of Magnetospirillum sp. AMB-1. Scale bar
indicated 1 \xm.
Genetic Aialvsis of Iron Bioiniieralizatlon 173

Figure 2. Southern hybridizaiion analysis ofpRK4!5 isoSuted froin transi-oara«.i.inL L.isie:-, : \ l , X-Pstl : 1,2.
pRK4 15 purified from E. co!: by uilracetuniiigauon; 3,4, r.RK41S jsoiaied IrGsi! A;vtB-! imiisconjiiganl: 5,6.
pRK4i,'5 frDni..\MB-; tnifisionrifd arsd rc:selaied iVorn £. tx;,'< ; 7, gc-ionsic D \ . \ iioinvuki type AK'1B-1. DNA
sairifjics i)i iafics i ,3,:5,"' v,,"^rc digcsicd v-iih A'cuR! aiiij saiiipirs in lane; 2,4.6 v-cre digested with 5>Mal. Probe
: purified [)RK4!5.

In this section, gene transfer into AMB-1, by conjugation and nonmagnetic

transposon mutagenesis of AM.B-1, arc dt,scussed.

2-1. Conjugal Gene Transfer between E. coli and AMB-1

We chose a broad-host-range mobilization system to attempt gene transfer in AMB-1.
We used E. coli SI7-1 as the liost in the gene transfer system. E. coli SI7-1 contains the
broad-host-range plasmid RP4 (tra^Jm its genomic DNA, and can transfer plasmids contain-
ing mob* to donors such as AMB-1. The broad-host-raoge plasmid, pKT230, is derived from
RSFIOIO and belongs to the IncQ group of incompatible plasmids. pRK415 is also a
broad-host-range plasmid, is derived from RK2, and belong to the IncP group of incompat-
ible plasmids. Both plasmids were transferred into AMB-1 using SI7-1 as a donon Each

Figure 3. Optimal mating ratio and mating

time of pKT2.10 transconjugation. Mating
time; (O) 2 hrs, ) 4 hrs, (D) 6 hrs, ) 8 hrs li 2i 30
Mating ratio; S17-1/AMB-1. Mating ratio
174 T. Matsuga et al.

Figure 4. Transmission electron mi-

crograph of transposon nonmagnetic
mutant, NM5. Scale indicates 1 }im.

transconjugant was confirmed to have antibiotic resistance, and was sliowii by southern
hybridization to maintain the plasmid(Fig. 2 ) (13).
Moreover, we optimized transfer of pKT230 by performing conjugation at different
condetions. Optimum conditions for this vector were at a donor-recipient ratio of 10:1 and
for6hat25°C(Fig. 3 ). ForpRK415,they were at the ratio of 100:1 for 6 hand the maximum
efficiency was 5.0 x 10'^(transconjiigants / donors). Therefore, all conjugations of AMB-l
were done under the conditions.
This efficiency is fairly high when compared with Aquaspirillum species (5), so this
transfer method is considered useful for future transposon mntagenesis.

2-2. Random Transposon Mutagenesis of AMB-l and Cloning of

Ctiroinosomal Regions Involved in Syntiiesis of Magnetic Particles
Selection of mutants which lose magnetite synthesis ability is the easiest way to find
genes iovoived in toe synthesis of magtietic particles in magnetic bacteria. However, mutants
due to IJV or chemicals have various mutations in dieir genomic DNA, so it is difficidt to
find which mutation is involved in magnetite synthesi.s. On the other hand, transposons arc
inserted at only one point oi the gcnoiriic DNA. Transposon Tni also gives kanamycin
resistance, and the fragrnenl whci-f the transposon is ioserted is detected by southern
hybridization. Therefore, we next made a transposon mutant of AMB-l by conjugation with
pSL;Plil21 (22). pSUPi021 contains I n J and mob* to transfer in. AMB-l.
We selected 'l'n.5 transconjiigants by kananiycm resistance. Among 118 colonies
of transconjugants, 5 were found to be completely nonmagnetic while 2 showed reduced
magnetite production compared with the wild type. The nonmagnetic transconjugants
were denoted NMl, 'NM2, NM3, NM5, and NM7. The nonmagnetic phenotypes of
these mutants were confirmed by transmission electron microscopy (Fig. 4). \'iOreover,
genomic DN.'\ from each mutant was probed with a Tni-specific Xhol restriction
fragrnenl purified from pSUP1021 (Fig. 5). In each lane, the probe hybridized with
only one band. This means that insertion by TnJ was at only one point in the genomic
DN.'\ of each mutant.
Three TnJ-containing chromosomal £eoRI fragments were cloned into E. coli DH5a,
lliree clones, plasmids pCM. pCN3, and pCNS, were obtained from the nontnagtietic
transcoriiugantslSiMl, NM3, and MM5, respectively.
Genetic Analysis o f I r c i i Biomiiicralization 175

figure 5. Southeni h>bridiza!!Oii analysis of £foi<i^4igestcd genomic DXA from notimago utants with
Tii.5 is a probe^ l.iines : M. l^/fwdiH; L pSUPiU2i; 2, wild tvpc; 3, NEvli; 4. \7vl2; 5. NM3; 6, KM5; 7. KM7.

3. GENETIC ANALYSIS OF magA GENE

As indicated in the previous sectioo, 5 nonmagnetic mutants of MagnetospiriUum sp.
strain AMB-1 were obtained by transposon mutagcEesis, and tliree Tn5-contaiiiing genomic
DNA fragments were clooed. Among these clones, pCNS had an 8.8-kb EcoVJ fragment
which contained inserted transposon TnJ. About 3-kb of the genomic DNA fragment from
NM5 was sequenced and two open reading frames were analyzed. Genetic analysis of these
two genes is discussed in this section.

3-1. Sequence of Genomic DNA from AMB-1 Nonmagnetic Mutant

2975 r>p of the genomic DKA fragment fromi the TnJ Hanking region of Ni\15 was
sequenced. The physical map of this region is shown in Fig. 6 . The transposon msertion site
is also shown. Two open reading frames with putative ribosomal binding sites winch
resemble a SI) sequerice (21) were ibund, hi Fig. 6 , the black hne represents one of the open
reading frames. ORF1, named magA. which is inteniipled by the Tn5 insertion and followed
by 0 R F 2 , which overlaps magA h\ 25 bp. Fig. 7 shows the niicleotidc sequence of the
genomic DNA fiagmsnt containing inagA and 0 R F 2 . Witlrin magA gene, there is a TnJ target
sequence, TTCTGACC, at nucleotide (ntj 1524 which was duplicated by TnJ integration in
the inutagcnizcd genome of NM5. The magA. gene L'J05 bp in size and a putative pronsoter

2W5bp

13«51jp 6tl6liD
O E F1 (magA ( OR fUmagli)

Ecom Putative X EimM BamHl

pronioter
Tn5
Figure 6. lAicalization of the TnJ insertion site within tlie open reading frame ORF I in ttie mutagenized
genomic fragment from nonmagnetic mutant NMS.
176 T. Vlatsujia et al.

CCCGCTGTGCATGCAAACCCCCAALlfiMtCCGCCAGTCICCMIAaiCOCCTACGCGATTCTACTltlfiiAIAGAATATSAGCAAIACCTGCTTTCGly: 800
-35 -10 -35

S t a r t ORTl(MaffA)
TTTTATGCAGCCGTCGTCACATTICCGCCACATCGGGGCrTTAGACCGGGGGCIGCCTICGTaAIAAOMaiCGICCATSGAACTGCATCATCCC 900
-10 net Go Leo Hs Hs FVo 6

CAACTGACCTATGCCGCCATCGTCGCCCTGOCCGCCGIGCTGTGCGGCGGGATCATGACGCGCCIGAAGCAGCCGGCCGTCGTCGGCTACATCCTGGCGG 1000
Go L«u Thr Tyr Ao Alo [ I * Vol A Q Lao Aa Ao Vol Leo CyS Giy G'y f1e1 i e t Thr Arg Leu Lys Gn Pro Aio Vol Vol Gly Tyr He Leu Aio 39

GGGTGGTGCTGGGACCCAGCCGCTTCGGGCTGGTGAGCA4CCGCGACGCCGTGCCCACCi:TGGCCGAGTICGGCGTGCTGATGCTGCTGTTi;£Iiilt£i 1100
Gy Vol Vol Leo G'y fVo 5«r G'y fbe G'y Lao Vo) Ser Asn Arq Aip A'o Vol Ao Thr Leu Ao C'u Pha G'y to! Leu net Leu Leu Phe Vo; Me Gy 73

PrUurl P>tl
CATGAAGCTGGACATCATCCGCtTTCTCGAAGTGTGGAAGACGGCCATCTTCACCACGGTTCiaCAGATCGCCGGCAGCGTGGGCACGGCCCTGCTGCTG 1200
l e t Ly5 Leo Asp [le [le Arg Pt^e Loo G o Vol Trp Lya Thr A a lie Phe Ttir Thr Vol Lao Gn He Alo Gy Ser Vol Cy Thr Au Leu Lao Leu 108

CGTCACGGCCTGGCCTGGAGCCTGGGGCTGGCGGTGGTGCTGGGCTCTGCCGTGGCGGTGTCGTCCACCGCCGTAGTGATCAAGGTGCTGGAATCCTCGG 1300
Arg H'8 Gly Lao Gy Trp Ser Leo Gy Leo Ao Vol Vol Leo G'y Cya Ao Vol A'o Vol Ser Ser Thr A'n Vol Vol He Lys Vol Leo Co Ser Ser 139

ACGAGCTGGACACGCCGGTCGCCCGCACCACCCTTGGCATCCTGATCGCCCAGGACATGGCGGTGCTGCCCATGATGCTGGTGCTGGAATCCTTCGAGAC HOO
Asp Giu Leu Asp Thr Pro Vol G y V ^ Thr Thr Leu G'y lie Leu He Aio G'n Asp Hat Ao Vol Vol Pro Net llat Lao Vol Lao Gio Ser Pha Gio Thr 173

SnsI
CAAGGCGCTGCTGCCCGCCCACATGGCCCGGGTGGTGCTGTCGGTGCTGTTCCTGGTGCTGCTGITCTGGTGGCTGTCCAAGCGCCGCATCGACCTGCCa 1 5 0 0
Lys Ao Leo Leo Pro Ao Asp net Ao Arg Vol Vol Leo Ser Vol Leo Pha Leu Vol Lao Leu Pha ' r p > p Leu Ser Lys Arg V g lie Asp Leu Pro 206

TnS t a r g a t a l t a
ATCACCGCCCOGCTTTCCCCCGAIlLIfliLLTTGCCACCCIGTCGACCCTGGCCTGGTGnTCGGCACCGCCGCCATCICCGGCGTGCIGGACTrGTCGC 16O0
[le Thr Ao Arg Lao Sar Arg Asp Sar Asp Leo Aio Thr Leu Ser Thr Leu Aio ' r p Cys Pha G y Thr Ao Ao lie Sar G y Vol Lao Asp Leo Sar 239

Prl]Bar2
CCGCCTATGGCGCClTCCTCGqCGGCGTGGIGCTCGGCAATTCCGCCCAGCGCGACATGCTGTTGAAGCGTGCCCACCCCATCGGCAGCGTGCTGCTGAT 1700
Pro AiQ Tyr G'y Ao Pha Lao Gy Gy Vo' Vol Leu G'y Asn Ser A o G n V g Asp i t t Leu Leu Lys V q Ao Gn Pro [le Gy Sar Vol Lau Leo Net 273

GGTGTTCTTCCTGTCCAICGGGCIGCTGCTCGACTTCAAGTICAICTGGAAGAATCTGGGCACCGTTCICACCCTGCTGGCCATGGTGACCCTGTTCAAG I BOO
Vol Phe Phe Leu Ser lie G y Leo Leu Leo Asp Pha Lys Phe He V p Lys Asn Leo G y Thr Vol Leo Thr Leu Leo Ao Nat Vol Tnr '^oo Phe LyS 306

ACCGCGCTGAACGTCACGGCGCrGCCCCTGGCCCGCCAGQACTGCCCCAGCOCCTrCCTGGCCGaCGTGGCCCIGOCCCAGATCGCCGACTTCTCGTTCC 1900
Thr Ao Leu Asn Vol Thr Ao Leo Arg Leu Ao Arg G r Asp Trp Pro Ser Ao Phe Leu A'o Gly Vol AIO Leu Ao Gn [le G'y G'o Phe Sar Phe 339

TGCTGGCCGAGACCGGCAAGGCGGTCAAGCTGATCAGCGCCCAGGAGACCAAGCTGGTGGTGGCGGTCACCGTGCTGTCCCTGGTGCIGTCGCCGTTCTG 2OO0
Leu Lau Ao G'u Thr G y Lya Alo Vol Lys Lau [le Sar Ao Gn Go Thr Lya Leo Vol Vol Alo Vol TN- Vol Leo Sar Leo Vol Leu Sar Pro Phe Trp 373

GCTGTTCACCATGCGGCGCATGCACCGGGTGGCGGCGGTGCATGTCCATTCGTTCCGCGAICTGGTCACGCGGCTGTATGGCGACGAGGCCCGCGCTTTC 2100
Leu Pha Thr Nat Arg Arg Nat Hs Arg Vol Ao Ao Vol Hs Vol Hs Sar Ph« Arg Asp Leu Vol Thr Arg Lao Tyr G y Asp Ciu Ao Arg Ao [Tie q06

9D S t a r t ORTS
GCCCGCACCGCGCGGCGGGCCCGTGTGCTGGTGCCGCGTGGTTCCTCGAGGGATGACCCCAATGCCGGACCTGGCTCTGGAATTTGACATCCCCGGCATC 2200
Alo Arg Thr A'o Arg Arg A'o V q Vol Lau Vol JVg Arg G'y Ser > p Arg Asp Asp Pro Asn Ao Gy Pro G'y Ser G'y He 'ISt
Net Pro Asp Leu A'o Leo G'u Phe Go He Gy G'y He 13

GTCTCCGGCATCGACCAGGTGGCCCGCGGCCCCCTGGCCGGGCCGGTGCTGGCCGCCGCGGTGATCCTCGACCCCCCCCOCCTGCCCAAGACGCTGCTGG 2300

Vol Cya Gly [le Asp Go Vol Gy Arg G'y Pro Leo Ao G'y Pro Vol Vol Ao A'o Alo Vol [le Leu Asp Pro Aio Arg Leu Pro Lya Thr Lau Lau 48

AGCGGCTGGACGATICCAAGAACCIATCCAAGCGTAACCGCCAAGAACIGGCCGAACTGGTGCITCGCCACCCCCATCCTTGGCTTCGGCGAGGCTTCGGT 2100

Go Arg Leu Asp Asp Ser Lys Lys Leo Ser Lys fifq Asn Arg Go Go Leu Ao Go Leo Vol Pro Ao Thr Ao He Leu Gy Phe Gy Gio Alo Ser Vol 80
P«tl
GGAGCAGATCGACCGGATCAACATCCTGCAGGCGACGTTTCTCGCCATGCGCCGCOCCTAIGACGCCCTGGGGCGGCAAIGCGCCCATGCGCrCGTCGAC 2600

G0 Gio He Asp Arg He Asn [la Leo Gn Ao Thr Phe Leu Ao Net Arg Arg A'o Tyr Asp A Leo Gy Arg Go Cys A'o H s Ao Leu Vol Asp 113

CGCAACCGGCCGCCGCGGCTGCCCTGCCCGGTCCGCTCCGTGGTGGGGGGCGACGGCATCTCATTG'CCATCGCCCCCGCCTCGGTGGTGGCCAAGGTGC 2600

G'y Asn V g Pro Pro Gy Leu Pro Cys Pro Vol V g Cys Vol Vol Gly G'y Asp Gy He Ser Leu Ser lie Ao Ao Ao Ser Vol Vol Ao Lys Vol 146
ICORI
GGCGCGACGCCATGATGGCCGAICTGGCCCGCGCCCACCCCGAATICGGCTGGGACAGGAACGCCGGATACGGTACGGCCGAGCAICTGCACGCCCTGAA 2700

V g V g Asp Alo Net Net Ao Asp Lao A'o V g Ao H s Pro G o Phe Gy Trp Go V g Asn Ao G'y Tyr Giy Thr Aio Go His Leo Asp Aio Leo Lys 180

ACGCCTCGGTCCCACCCCCCATCACCGACGCICCTTCGCCCCCGTGGCCCAATATATGCIGTTCTAACCTGIGGATAAGCCACTACCTGTGGCAATGAAT 2800

V g Leu Gly Pro Thr Pro H s N s V g A-g Ser Pha A o ("Vo Vol A o G'n Tyr Net Leo Phe 201

Figure 7. Nucleotide sequence of leh mulagenized genomic fragment of NM5. Arrows indicate inverted
repeats and hairpin loop.
Genetic Analysis of Iron Biomineralization 177

region is located 75 bp upstream from the start codon ofmagA at nt 883. Two sets of putative
-35 and -10 domains are shown as boxed regions and inverted repeats occur near the -35 and
-10 domains which are indicated by arrows in Fig. 7 . However, no promoter-like region was
found upstream from the start codon of 0RF2 at nt 2162. Furthermore there is a region of
dyad symmetry at nt 2227 downstream from the stop codon of inagA which resembles a
Rho-independent terminator (19). No terminator structure was found downstream from the
stop codon of ORF2. The hypothetical protein encoded by the mag A gene. MagA. consists
of 434 ainino acids with a predicted molecular weight of 46.8 kDa. 0RF2 consists of a 606
bp section which would encode a 21.6 kDa protein.

3-2. Homology Analysis of magA Gene and ORF2

MagA was found to have high homology with the KefC protein from E. coli (16),
with 25.4 % similar amino acid residues. Fig. 8 shows an alignment of MagA and KefC.
KefC functions as a potassium efflux protein for control of turgor. Moreover the NaH-an-
tiporter, NapA of Enterococcus hirae (26) is homologous to MagA, with 24.1 °o similarity.
The similarity between KefC and NapA has been described (18). These proteins form a group
of cation efflux antiporters. This result suggests that the magA gene may encode a cation
efflux transport protein.
Moreover, KefC consists of two domains, a hydrophobic membrane-binding domain,
and a strongly hydrophilic carboxy-terminus. The region homologous with MagA corre-
sponds to the hydrophobic domain, which is thought to be a potassium-translocating channel
domain (16). Furthermore the hydropathy plot of MagA was compared with KefC. and a
similar pattern of hydrophobicity of amino acid sequence was observed (Fig.8) . The
sequence of hydrophobic amino acids could form eight to ten transmembrane a- helices.
The similarity and structure suggest that MagA may also be located in the membrane of
AMB-1.
The putative protein encoded by ORF2 was also analyzed. A strong homology (47.2
% similarity) with E. coli RNase HII (8, 25) was exhibited (Fig. 9 ). RNase HIl degrades the
ribonucleotide moiety on RNA-DNA hybrid molecules. This strong homology suggests that
the ORF2 coding protein has a similar function to RNase Hll.

3-3. Measurement of magA Promoter Activity

The function of the promoter which was found upstream from magA was investigated.
A 540 bp Kpn\-Psi\ fragment containing the promoter region was cloned upstream from the
CAT reporter gene of the promoter probe vector, pXC AT, in the orientation which the putative
promoter transcribes the CAT gene. This plasmid was designated pMKR pXC.AT and pMKP

Table 1. Average CAT activity of exponentially growing

transconjugant cells

pXCAT pMKP

AMB-I N. D. 45.7

E. coli 2.0 2.0

unit; nunit/cell, 1 unit of CAT activity acetylates

10 nmol chloramphenicol per minute. N. D.; not detected.
 "^ T. Malsuga et al.

kTuple: 2; Gap Penalty: 4; Gap Length Penalty: 12

Seq1(1>435) Seq2(1>620) Similarity Gap Gap Consensus
MagA AMB-1 KefC E.coli Index Number Length Length
(3>420) (1>422) 25.4 8 10 425
(3>420) (1>422) 25^4 8 10 425

MagA_AMB-1 LHHPELTYAAIVALAAVLCGGMMTRLKOPAVVGYILAGVV 42
::: : I. . I: I : : I; I: : : : : : II: : : I: I I i I :
KofC_E. coll MDSHTLIQALIYLGSAALIVPIAVRLGLGSVLGYLIAGCI 10

nagA_AMB-l LGPSGFGLVSNRDAVATLAEFGVLflLLFV IGMKLDI IRFL 82

I I: I. : I I : . : : : : : 11:11 1 1 : 1 1 : 1 1 : 1.
<efC_E, coll IGPWGLRLVTDAESILHFAEIGVVLMLFIIGLELDPQRLW 80

MaqA_AMB-1 EVWKTAIFTTVLQIAGSVGTALLLRHGLGWSLGLAVVLGC 122

: :: ::; II::: : I:: I ::: II. : : I. : I:
KefC_E. coll KLRAAVFGCGALQMVICGGLLGLFCMLLGLRWQVAELIGM 120

MagA_AMB-l AVAVSSTAVVIKVLESSDELDTPVGRTTLGILIAQDMAVV 162

: : I: I I I I: : : : : : : : : . I: : I I: : 1:11:1::
KefC_E. coll TLALSSTAIAMQAMNERNLMVTQflGRSAFAVLLFaDIAAr 160

MagA_AMB-1 P---MMLVLESFETKALLPA-DMARV-VLSVLFLVLLFWW 197

I I::: I: .:::.: I ::. : I: :: I. II I. .
KefC_E. coll PLVAMIPLLATSSASTTMGAFALSALKVAGALVLVVLLGR 200

MagA_Af1B-l LSKRRIDLPITARLSRDSDLATLSTLAWCFGTAAISGVLD 237

:: I: : I: : 11 : :. : : : I: . ; II : : : . :
KefC_E. coll YVTRP-ALRFVARSGLREVFSAVA-LFLVFGFGLLLEEVG 238

MagA_Af1B-l LSPAYGAFLGGVVLGNSAQRDMLLKRAQP IGSVLLnVFFL 277

11:1:1111 11:1 : I: : I ; I. : : : : I : : : I I: : I I
KofC_E. coll LSMAMGAFLAGVLLASSEYRHALESDIEPFKGLLLGLFFI 278

MoqA_>\MB-l SI GLLLDF-KF IWKNLGT VLTLLAMVTLFKTALNVTALRL 316

: I II . . :: I: . I ; I I: : . :: : I. :. I: : I
KofC_E. coli G V G M S I D F G T L L E N P L R I V I L L L G F L M K I A M L W L I A R P L 318

MagA_AMB-l ARQDWPSAFLAGVALAQIGEFSFLLAETGKAVKLISAQET 356

: : . : ::: . I I: I: I: : 1 I: I : : : : :: :: : . :
KefC_E, coll aVPNKQRRWFA-VLLGQGSEFAFVVFGAAQMANVLEPEWA 357

MagA_;^MB-1 KLVVAVTVLSLVLSPFWLFTMRRMHRVAAVHVHSFRDLVT 396

I : :::: II :: I:. 1.. I :. ::. ::: :: ::
KofC_E. coll KSLTLAVALSMAATPILLVILNRLEQSSTEEAREADEIDE 397

MagA_AMB-l RLYGDEARAFARTAR-RARVLVRRG 120

: :. :. : I: I :: :: I: I: I
KofC_E, coll EQPRVIIAGFGRFGQITGRLLLSSG 122
Figure 8. Amino acid sequence homology of MagA with KefC of £. coli. Identical amino acids are indicated
as bar. Equivalent amino acids are shown as one or two dots. Two dots indicate higher similarity than one dot.
Homology exists between the MagA protein and the KefC hydrophobic region (16).

were transfeired into wild type AMB-1 cells by conjugation (13, 23) and change in CAT
activity with cell growth was measured (24). Table 1 shows average CAT activities of AMB-1
transconjugants containing either pXCAT or pMKP in log-phase. The pXC AT transconjugant
did not show CAT activity. In contrast, high CAT activity, 44.8 nunit/cell was measured in
Genetic Analysis of Iron Biomineralization 179

kTuple: 2; Gap Penalty: 4; Gap Length Penalty: 12

Seq1(1>202) Seq2(1>213) Similarity Gap Gap Consensus
MagB_AMB-1 RNH2_E.coll Index Number Length Length
(4>199) (2>194) 47.2 2 5 197
(4>199) (2>194) 47.2 2 5 197

MagA_AMB-l LALEFEIGGIVCGIDEVGRGPLAGPVVAAAVILDPARLPKTLLERLDDSKKLSKRNREEL 63
: . : : : 1:1 I I I I I I I I : I: I I : I I I I I I I I I : : : I: I I I I I I : : : I: : I
RNH2_E. Col I lEFVYPHTOLVAGVOEVGRGPLVGAVVTAAVILDPAR PIAGLNDSKKLSEKRRLAL 57

MagA_AMB-l AELVPATAl-LGFGEASVEEIDRINILQATFLAMRRAYDALGRECAHALVDGNRPPGLPC 122

: I. : : : : I : . I: I: . : I I I I I I: I I : I I I: I I: : : I : : : . : : I: I I I I: I: I I :
RNH2_E. Col I YEEIKEKALSWSLGRAEPHEIDELNILHATMLAMORAVAGLHIAPEYVLIDGNRCPKLPM 117

MagA^MB-1 PVRCVVGGDGISLSIAAASVVAKVRRDAMMADLARAHPEFGWERNAGYGTAEHLDALKRL 182

I : : : I I : I I; : : : I: I I I: : I I I : I I I: 11: I; : : : I. . I. : : : : I I : I I I I : : I .
RNH2_E. Col I PAMAVVKGDSRVPEISAASILAKVTRDAEHAALDIVFPQYGFAQHKGYPTAFHLEKLAEH 177

MagA_Af1B-l GPTPHHRRSFAPVAQYM 199

I: I: I I I I 11: I I : :
RNH2_E. C o l l GATEHHRRSFGPVKRAL 191
Figure 9. Amino aced sequence homology of the MagB protein with RNase HII of £. coli. Identical amino
acids are indicated with a bar. Equivalent amino acids are shown as one or two dots. Two dots indicate higher
similarity than one dot (8).

the cells of the pMKP transconjugant. This result suggests that the 540 bp fragment cloned
from upstream of the magA gene contains an active promoter which is functional in AMB-1
cells. Escherichia coli transconjugants containing pXCAT and pMKP were also analyzed for
magA promoter activity. Both transconjugants exhibited little CAT activity. The inagA
promoter does not function in E. coli.

4. EXPRESSION OF PROTEIN ON THE SURFACE OF AMB-1

MAGNETIC PARTICLES USING magA GENE
Attempts to express proteins on the surface of bacteria using various genes which
encode proteins located in the outer membrane have been made. For example, using oinpA
or iga gene, proteins have been expressed on the cell membranes of gram-negative bacteria
(6, 7, 11). Moreover, 'phage display', which is the method of forming fusion proteins on the
surfaces of bacteriophages by fusing with coat protein, has been applied to gene screening
(3,4,12).
The putative protein of MagA seemed to be localized on the cell membrane froin the
homology between magA and kefC gene. We try to localize MagA protein by fusion gene
method and immnoelectron microscopy localization method. Since magnetic particles in
AMB-1 is also covered with membrane, MagA fusion protein will also be localized on the
surface of AMB-1 magnetic particles.

4-1. Localization of magA Protein

In order to investigate localization of the MagA protein, the gene fusion method of
magA and luciferase was employed. The firefly luciferase gene, luc (Toyo Ink co. ltd., Tokyo,
Japan), was used as the fused reporter gene. The strategy effusion gene construction is shown
180 T. Matsuga et al.

JamHl Sal 1
^ " ^ BamHI So/ I
(pRK145 (10.5 kb)j) -t /«c(i.7kb) ^i

Figure 10. Construction of plasmid containing the magA-liuiJt'nise fusion gene.

in Fig. 10 . inagA is 1305 bp long. Firstly, luc was cloned into the plasmid vector, pRK415
(Tc', lacZ, mob*) (10), which is a broad-host-range vector for gram-negative bacteria, and
was maintained in Magnelospirilliim sp. AMB-1. 886 bp of the EcoR\-Ncol fragment
containing the iiiagA promoter region and excluding the magA open reading frame was
connected with the luc gene. This plasmid was designed pKPL. Then, 1135 bp of the
EcoR]-Sph\ fragment was ligated with the luc gene after blunting treatment. The Sph)
restriction enzyme site is in the hydrophilic amino acid coding region in magA. By this
ligation the magA-luc fusion gene was successfully constructed, with no translation frame
shift. The constructed plasmid was termed pKML. Both plasmids were transferred into wild
type AMB-1 cells by conjugation.
The luciferase activities of cytoplasm, cell membrane and bacterial magnetic particles
(BMPs). of transconjugants were measured. The luciferase assay of each fraction was carried
out using PicaGene(Toyo Ink co.. ltd., Tokyo, Japan). The luminescence of the luciferin-lu-
ciferase reaction was detected using Luminescence Reader. pK.PL and pKML transconju-
gants were grown to stationary phase. Cells were harvested and resuspended in 10 mM Tris
buffer (pH 7.0). Harvested cells were broken by sonication, and BMPs were collected using
a magnet. Cells which were not lysed were separated from the cell debris by cenirifugaiion
at 5000 X g for 20 min. Then the supernatant was ultracentrifuged at 96,600 x g for 1.5 h to
separate the membrane fraction as a precipitate from the supernatant containing the cyto-
plasmic fraction.
Results of the luciferase as.says on pKPL and pKML transconjugants are shown in
Table 2 . Luciferase activity was detennined from the integrated luminescence yield for 3
minutes per ng of protein. In the pKPL transconjugant the highest luciferase activity was
Genetic Analysis of Iron BioniineralMaticin 181

Table 2. Luminescence yields of cell

fractions of transconjiigants

Cell fraction pKPL pKML

cytoplasm 35.2 (1.9

cell membrane 2,6 13,0
magnetic particle 0.2 1.3

Unit: kcounl/mg prolan, Lndferase activity was determitied from

tiic iiiiegfiited lummesceoce yield for 3 minutes.

detected in the cytoplasmic fraction. Lucifcrase is a soluble protein, so it should localize in

the cytoplasmic space when it ex.presses independeotly. ID contrast, most of the lucifcrase
activity ia the pKML transconjugant localized in the membrane fraction. This result indicates
that the MagA protein locah'zes in the cell membrane, and the /Mc-connected site in MagA
protrudes from the membrane. Moreover, the BMP fraction showed weak lucifcrase activity.
This resuh suggests that the MagA-Luc fusion protein is expressed on the BMP surfaces.
Iron must be translocated across the BMP membrane during the crystallization of
magnetite. We consider that MagA is probably an iron channel protein which is integrated

h-

Figure 11. Transmission electron

micrographs of BMPs extracted
from transconjiigants which were
treated immunogold. Photograph
A shows BMPs of the pKMLiran-
scotijugant. B shows BMPs of the
pKPL transconjugant, Gold collo-
dion particle has 15 nin diameter
in size.
182 T. Matsuga et al.

in the cell aod BMP membranes of AMB-1, and translocates iron from the environment into
the cytoplasmic space across the cci! membrane, and from the cytoplasm into the lipid
vesicles of BMPs across the BMP membrane. In order to elucidate the synthesis mechanism
of the lipid vesicles of BMPs. it is important to know whether the MagA protein localizes
on the outer or inner membrane. We assume that the inner membrane invaginates to form
the lipid vesieles of BMPs.

4-2. Detection of Luclferase Protein on the Bacterial Magnetic Particles

To detect the localization of the MagA-Lueiferase fusion protein, immunoelectron
microscopy localization method was used. Anti-Luciferase polyclonal antibodies which bind
to gold particles (15 nm) were used to localize the fusion protein at the ultrastiiictural level.
BMPs were extracted from pKFL or pKML transconjugants of AMB-1 by French press and
magnetic separation, and resuspended in PBS buffer (pH 7.4). Then the suspension was
incubated for 1 hour after addiog anti-Luciferase antibody and was washed twice. Samples
were observed by electron microscope. The results are shown in Fig. 11 . Antibody-gold
particles were bound to magnetic particles extracted from pKML transconjugants. This
means that the MagA-Luciferase fusion protein is localized on the lipid membranes of BMPs.
Moreover, it will be possible to express various proteins on the surfaces of magnetic particles
using fusion proteins with MagA.

cell niembrape

BMPs
magnetite

lipid layer

2. Development of new protein

1. Proiuction of BMPs engineering system
posses new fnnction
Clem&gt m4
scpmrnmn of proteirs

- O

i
Production of IgG-Hni BMPs and
nil!* prmUjrtlon of BMPi
Csllectlon «t BMPi l!j- magnet
Collecltoiiofprotmii

construction of new immunoassay system

Development of microniacliine bound to
protein whicii enable to induce by magnet

Defelopment of Moreacter by
eniyme-bind BMPs
Figure 12. Teciinolo|ica! application of proiein product on BMPs.
Genetic Analysis of Iron Biomineralization 183

4-3. Protein Expression on the Surfaces of Magnetic Particles

Examples of protein expression of the surface of magnetic particles is shown in Fig.
12 . In a previous novel fluoroimmunoassay method using BMPs, fluorescein isothiocyanate
(FITC)-conjugated monoclonal antibody was immobilized onto BMPs using a heterobifunc-
tional reagent, N-succinimdyl-3-(2-pyridyldethio)propionate (SPDP) (17). However, conju-
gation of antibodies to magnetic particles is difficult and the quantity of antibody per
magnetic particle was unstable. On the other hand, magnetic particles from MagA-Antibody-
fusion-gene transconjugants will be useful, and will eliminate the necessity conjugate
antibodies and particles chemically.
Moreover, expression of fusion proteins with MagA protein can be applied to the
system of protein production. Magnetic particles with fusion proteins can be collected easily
and quickly by magnet after breaking the cells. We will be able to obtain purified proteins
by this system by inserting a gene between magA and the target gene in order to cleave the
target protein from the fusion product.

5. SUMMARY

Establishment of gene transfer has enabled study of analysis of genes in magnetic

bacteria and protein production on BMPs using magnetic bacteria as the host celt. Gene
transfer into magnetic bacteria, and transposon mutagenesis, are very useful for future
analysis of the mechanism of magnetic particle production. The magA gene was the first
discoverd gene involved in magnetite biomineralization in magnetic bacteria, and the
product of magA seems to be a kind of cation channel protein localized on the cell and
magnetic particle membranes. Gene analysis ofMagnetospirillum sp. AMB-1 will determine
the mechanism.
Moreover, protein expression on the surfaces of magnetic particles using the fusion
gene method was studied. Magnetic particles which have fusion proteins on the surfaces can
be collected very easily, and production of purified protein will be applied in future.

REFERENCES
1. Blakemore, R.P., D. Maratea, and R.S. Wolfe. 1979. Isolation and pure culture of a freshwater magnetic
spirillum in chemically defined medium. J. Bacteriol. 140:720-729.
2. Burgess, J.G., R. Kawaguchi. T. Sakaguchi, R.H. Thomhill, and T. Matsunaga. 1993. Evolutionary
relationships among Magnelospirillum strains inferred from phylogenetic analysis of I6S rDNA se-
quences. J. Bacteriol, 175:6689-6694.
3. Corey. D.R., A.K. Shiau, Q. Yang, B.A. Janowski. and C.S. Craik, 1993. Trypsin display on the surface
of bacteriophage. Gene 128:129-134.
4. Djojonegoro. B.M., M.J. Benedlk. and R.C. Willson. 1994. Bacteriophage surface display of an immu-
noglobulin-binding domain of Slaphylococcus aureus protein A. Bio Technology 12:169-172.
5. Eden, P.A. and R.P. Blakemore. 1991. Electroporation and conjugal plasmid transfer to members of the
genus Aquaspirillum. Arch. Microbiol. 155:449-452.
6. Francisco, J..'\., R. Campbell, B.L. Iverson, and G. Georgiou. 1993. Production and fluorescence-acti-
vated cell sorting oiEscherichia colt expressing a functional antibody fragment on the external surface.
Proc. Natl. Acad"^ Sci. USA 90:10444-10448.
7. Georgiou, G., H.L. Poetschke, C. Stathopoulos, and J.A. Francisco. 1993. Partial application of engineer-
ing Gram-negative bacterial cell surfaces. Tibtech 11:6-10.
8. Itaya, M. 1990. Isolation and characterization of asecondRNase H (RNase HU) of Escherichia coli K-\2
encoded by the rnhB gene. Proc. Natl. Acad. Sci. USA 87:8587-8591.
184 T. Matsuga el al.

9. Kavvaguchi. R., J.G. Burgess, and T. Matsunaga. 1992. Phylogeny and 16S rRN'A sequence of Muf^iiC'
tospinlhiin sp. AMB-1. an aerobic magnetic bacterium. Nucl. Ac. Res. 20:1140.
10. Keen. N.T., S. Taniaki. D. Kobayashi. and D. Trollinger. 1988, Improved broad-host-range plasniids for
DNA cloning in Gram-negative bacteria. Gene 70:191-197.
11. Klauser, T., ) . Pohlner, and T.F. Meyer. 1990. Extracellular transport of cholera to\in B subunit using
Seisseiki ]g.\ protease b-doniain; conformation-dependent outer membrane translocation. The IIMBO
Journal 9:1991-1999.
12. Lownian, H.B.. S.H. Bass, N. Siinpson, and J.A. Wells. 1991. Selecting high-affinity binding proteins bv
monovalent phage display. Biocheinistry 30:10832-10838.
13. Matsunaga. T., C. Nakainura, J.G. Burgess, and K. Sode. 1992. (.iene transfer in magnetic bacteria;
transposon mutagenesis and cloning of genomic DNA fraginents required for magnetosome synthesis. J,
Bacteriol. 174:2748-2753.
14. Matsunaga, T., T. Sakaguchi, and F. Tadokoro. 1991. Magnetite formation by a magnetic bacterium
capable of growing aerobically. Appl. Microb. Biotechnol. 35:fi5l-()55.
15. Matsunaga. T.. F. Tadokoro, and N. Nakainura. 1990. Mass culture of magnetic bacteria and (heir
application to tlovv type immunoassays. IEEE Trans. Magnet. 26:1557-1559.
16. Munro. A.VV.. G.Y. Ritchie, A.J. Lamb, R.M. Douglas, and I.R. Booth. 1991. The cloning and DNA
sequence of the gene for the glulathione-regulatcd potassium-efflu.x system KefC of HKclu'rii hia coli.
Molec. Microbiol. S:607-616.
1 7. Nakamura. N.. J.G, Burgess. K. Yagiuda, S. Kudo. T. Sakaguchi, and I, Matsunaga, 1993, Detection and
removal l-'schericliici coli using tluoroseein isothiocyanate conjugated inonoclonal antibody immobili/ed
on bacterial magnetic particles, .Anal, Chem, 65:2036-2039,
IS. Rei/er. J . A. Reizer. and M.ll. Saier Jr. . 1992. The putative Na* H* antiporter (Nap,-\) o\' ljiu'r(>cc>c< us
hinic is homologous to the putative K^/H* antiporter (KefC) of Escherichia coll. FEIMS Microbiology
letters 94:161-164,
19. Rosenberg. M, and D. Court. 1979. Regulatory sequences involved in the promotion and terrmnation of
RN.-\ transcription. .Annu. Rev. Genet. 13:319-353.
20. Schleifer. K.H., D. Sehulcr. S. Spring. M. Wcizencgger. R. Amann. W. Ludvvig, and M. Kohler 1991. The
genus Miigiu'lospirilhim. gen. nov. Description of Miignel<?spirillum gnp/uswa/dense and transfer of
Aqiui'.piiillum miigiietoiacticum to Mugnetospirilliim magnelotucliciim. comb, nov, Syst. :'\ppl. Micro-
biol. 14:379-385.'
21. Shine, J, and L. Dalganio. 1974. Determinant of cistron speeitleily in bacterial ribosomes. Nature
254:34-38.
22 Simon, R.. .M, O' Connell, M. Labes. and A. Puhler. 1986. Plasmid vectors for the genetic analysis of
/?/)ir<;/i/(/and other Gram-negative bacteria. Meth. Enzyinol. 118:640-659.
23, Simon. R,. U, Priefer, and A, Puhler. 1983. A broad host range tiiobilization system for in vivo genetic
engineering : transposon mutagenesis in Gram negative bacteria. Bio Technology 1:784-791,
24, Sode, K,. N. Hatano. and M. Tatara, submitted. Cloning of a marine cyanobacterial promoter for foreign
gene expression using a promoter probe vector. .Appl. Biochem. Biotechnol,
25, Toniasiewicz, H.G, andC.S, Mchenry, 1987, Sequence analysis ofthe/:r,v(7;c7((/(/i;iv)//i//i(;/r gene, Jomnal
of Biotechnology 169:5735-5744,
26, Waser, M,, D. lless-Bienz. K. Davies, and M. Solioz. 1992. Cloning and disruption of putative \aH-aii-
tiportcr gene of Enlcrococciis liirtw. The Journal of Biological Chemistry 267:5396-5400,
 14

EFFECTS OF STATIC MAGNETIC FIELDS ON

ERYTHROCYTE RHEOLOGY

Takeshi Shiga', Masaharu Oi<azai<i-, Nobuji Maeda\ and

Akitoshi Seiyama'

' Dept. Physiol., Med. Sch., Osaka Univ.

Suita, Osaka 565 Japan
'Dept. Chem., Natl. Ind. Res. Inst.
Nagoya, 462 Japan
-Dept. Physiol., Sch. Med., Ehime Univ.
Ehimc 791-02 Japan

ABSTRACT
The acute influence of an external static magnetic field on the erythrocyte rheology
is briefly reviewed. 1) The magnetohydrodynamic action rnay be effective to alter the fast
blood flow in aorta at a field over 5 T, but no experimental study has been made. 2) The
diamagnetic interaction, between the normal blood component and uniform magnetic field,
has been found: i.e., the biconcave disk-shaped erythrocytes and platelets orient their flat
plane parallel to the field direction above several T. 3) The paramagnetic interaction, effective
only on a slow flow of deoxygenated erythrocytes, affects the distribution of erythrocytes
within a vessel by the paramagnetic attractive force, which is proportional to the product of
[flux density] x [spatial gradient], the magnetic susceptibility of erythrocytes, and the
reciprocal of the flow velocity. Our model study showed that the paramagnetic interaction
may be effective under an anemic situation and under an inhomogeneous magnetic field over
above 100 T'/m for the product of [flux density] x [spatial gradient].

INTRODUCTION
The static magnetic field has been known to interact with the cellular components of
blood: for example, the magnetic separation of phagocytes loaded with ferromagnetic
material [1], the orientation of sickled erythrocytes in a uniform magnetic field of 0.35 T [2,
3], and so forth. An external magnetic field may affect the blood flow enen in the normal
state through three distinct physical mechanisms: 1) The magnetohydrodynamic action: i.e.,
when charged particales pass at a high speed through a strong magnetic field, a force against
the flow develops (Lorentz force). Lorentz force is predicted to increase the pressure drop
Biological Effecls of Magnetic and Eleclromagnetic Fields. Edited by S. Ueno
Plenum Press. New York. 1996 185
186 T. Shiga et al.

by 10% in the ascending aorta under an uniform magnetic field of ca. 5 T [4]; but no
experimental proof has been given yet. 2) The diamagnetic interaction, which affects the
orientation of erythrocytes in an uniform magnetic field with normal discocytes at several T
[5], 3) The paramagnetic interaction, which acts only on the paramagnetic erythrocytes at
an inhomogeneous magnetic field with a strong spatial gradient. The first trial to trap the
Malaria-infected erythrocytes from flowing blood was not satisfactory [6], but later the idea
is successfully applied to the magnetic separation of paramagnetic erythrocytes [7].
Among these mechanisms, we have concentrated our attention to the paramagnetic
attraction of flowing erythrocytes under an inhomogeneous magnetic field [8,9.10,11], in
order to establish the physical characteristics of the phenomena, to find out the experimental
condition for the basis of safety criteria and for possible animal experiments, and to apply
the techniques for the hemorheological study. We have already demonstrated the magnetic
attraction of a flow line of the paramagnetic erythrocytes, using three experimental proce-
dures; i.e., I) displacement of a narrow stream line of erythrocyte suspension in a wide
laminar flow [8,9], 2) a small change in the distribution of erythrocytes flowing through a
cylindrical vessel [10], and 3) an acceleration of the erythrocyte sedimentation rate [11], due
to an inhomogeneous magnetic field.
This paper briefly reviewes these interactions between erythrocytes and magnetic
field, with emphasis on our own studies of paramagnetic attractive interaction.

A. MAGNETIC PROPERTIES OF ERYTHROCYTES

Blood contains various electrolite ions and negatively charged proteins. The eryth-
rocyte is a biconcave disk-shaped cell containing hemoglobin as a major constituent, which
occupies a volume fraction of 4 5 % (called hematocrit, Ht) thus determines the rheological
properties of blood. The surface of erythrocyte membrane is covered with negatively charged
polysaccharides.
The erythrocytes have a unique feature, i.e., the magnetic susceptibility of hemoglo-
bin can be varied by changing the valency and the spin state. The deoxygenated erythrocytes
are paramagnetic due to ferrous deoxy-hemoglobin (S=2), while the oxygenated erythrocytes
are diamagnetic due to oxyhemoglobin (S=0); and the erythrocytes containing high- and
low-spin (ferric) methemoglobin are both paramagnetic (S=5/2 or 1/2, respectively) but with
different magnetic susceptibility (Fig. 1).

Fe(ll) Fe (III)
Deo.vy ONV H2O CN
-?
Energy -f -4-
Diagram -t

-M- -M-
SPIN HJGH LOW HJGH LOW
^.9 0 5.9 1.7
2slSfS+\)

Figure 1. The spin states of hemoglobin (Hb) derivatives. Depending on the ligand to the 5th position (L) the
spin slate of central iron ion varies: i.e., (a) deoxy-Hb is in ferrous high spin state [S=2], (b) oxy-Hb is
diamagnetic ferrous [S=0], (c) metHb-H20 is in ferric high spin state [S=5 2], and (d) metHb-CN is in ferric
low spin state [S=l'2].
Effects of Static Magnetic Fields on Erythrocyte Rlieology 187

Four types of erythrocytes, containing various spin states of hemoglobin, were

prepared from freshly drawn human blood after removal of plasma and buffy coat [8-11],
(a) The washed erythrocytes were rc-suspended in an isotonic buffered saline ( 90 niM NaCI.
5 mM KCl, 50 mM Na-phosphate, 5.6 mM glucose; pH 7.4). and employed as the erythro-
cytes with oxygenated hemoglobin, (b) The suspension was deoxygenated by adding sodium
hydrosulfite. to obtain the erythrocytes containing deoxy-hemoglobin ; (c) after an addition
of sodium nitrite to oxidize the hemoglobin in erythrocytes, the erythrocytes were washed
and resuspended at pH 5.7 to establish methemoglobin in high spin ferric state, or (d) the
erythrocytes containing methemoglobin were treated with potassium cyanide and washed
with buffer at pH 7.4 to obtain the erythrocytes with low spin methemoglobin.

B. MAGNETOHYDRODYNAMIC ACTION (LORENTZ FORCE)

A static magnetic field interacts with flowing blood in a vessel, because blood
contains various ionic species.

B-I. Action on Blood Flow. When a solution, containing the ionic species (electric
charge: q), e.g., blood, flows rapidly at a velocity (u) in a tube (diameter: d) placed
perpendecularly to a static magnetic field (magnetic flux density: B), the Lorentz force (FB)
appears towards the direction perpendicular to both the flow direction and the megnetic field
as follows:
FB = q ( u x B).
Blood is a conducting fluid (electrical conductivity, o ), thus an electric current (j )
is generated in a homogeneous magnetic field :y = a (u x B) . The current produces a force
(/"[) to reduce the blood flow,
/ i = y X B = a (u X B) X B
Combining the Navier-Stokes equation and the Lorentz force, a calculation predicted
that 10 % retardation of blood velocity may occur with ascending aorta under the static field
of 5 T [1], but no change with small arteries [12]. However, no experimental proof has yet
been made.

B-2. Flow Potential. In addition, the induced electric potential (V) is provided by
blood flow with a diameter (d) in a static magnetic field: V = |u| |B| d. The amplitude of
T-wave in electrocardiogram can be modified by the superposition of the induced potential
(synchronizing to the pulse flow) due to aortic blood flow in a static magnetic field
[13,14,15].

C. DIAMAGNETIC INTERACTION
Most of the biological materials are diamagnetic, and for a rod-like macromolecule
placed in a strong uniform magnetic field the diamagnetic anisotropy may lead an orienta-
tional change along its rotation axis. In the case of erythrocyte, a disc-like material (volume,
V) with the anisotropic diamagnetic susceptibility ( perpendicular to and parallel with the
disc surface, d under an uniform magnetic field (H), the field-induced energy (U) is
given by the following equation:
Vi = - (MVfl.) ixr + ^xcQs^ Q ) ,
where A/ = Xa~ /Crand 9 is the angle between H and the disc plane of erythrocyte as
shown in Fig. 2.
188 r. Shij>a ct al.

Figure 2. Definition of anisotropic susceptibilities ( , and Xr> t'"' 'he

Magnetic Field disk-sliapcd erythrocyte with reference to the external magnetic field.

However, the magnetic interaction energy is theoretically predicted to be small

compared with the thermal energy kT, thus the degree of orientation of only 1 % was achieved
for calf thumus DNA at 14 T [16]. For the blood cotTiponents. Torbet et al [17], later
Yamagishi et al [18], demonstrated the orientation of fibrin polymerization in a strong
magnetic field.
Recently, Yamagishi et a! [19] have shown an orientation of normal, biconcave
erythrocytes under a high uniform field (the etTect saturates above 4 T): i.e.. human
erythrocytes are oriented with their disk plane parallel to the magnetic field direction.
Furthermore, the spin state of hemoglobin inside the erythrocytes does not influence the
degree of erythrocyte orientation in a strong magnetic field. For example, the deoxygenated
erythrocytes contain the paramagnetic hemoglobin (high spin ferrous Fe, S=2 ) while the
oxygenated cells possess diamagnetic hemoglobin (low spin ferrous, S=0), the degree of
orientation is unaffected by such changes in spin state of hemoglobin-Fe. Therefore, Higashi
et al [20] explained the magnetic orientation of normal discocytes takinging account for the
contribution of anisotropic susceptibility of lipid bilayer membrane and transmembrane
proteins. Metnbrane phospholipids are oriented with their hydrocarbon chain perpendicular
to the magnetic field and the anisotropic diamagnetic susceptibility (Ax) is about I x K) -''
emu/tnolecule [21,22]. On the other hand, the a-helix portion of the transmembrane proteins
(e.g.. Band 3, glycopholins, etc) is oriented their long axis (the longitudinal direction of the
helix-chain) in parallel with the magnetic field and A^ is about 7 x 10'^" emu/peptide group
[23], Comparing these two opposite forces, the estirnated contribution of total inetnbranc
phospholipid (13 X 10'-- emu/cell) is greater than that of transmcmbane proteins (7 x 10'--
emu/cell), thus intact erythrocytes oriented their disk plane parallel to the magnetic field.
Further, the observed, anisotropic diamagnetic susceptibility of the erythrocytes (Ay = 8 x
10'-- emu/cell) agrees with the above rough estimation [20, 24].
Similar phenomenon is also observed with human platelets [25], which are oriented
their disc surface parallel to the magnetic field direction. In the case of platelet , however,
the obserbed A^ cannot be explained by counting only the estimate of tnernbrane lipids but
the contribution of cytoskeletal microtubules should be accounted for.
In addition, a death was reported, possibly due to the dislodgment of intracanial
aneurysrn clip placed in 1.5 T magnetic field [26].

D. PARAMAGNETIC INTERACTION
Early in 1946, Heidelberger et al [6] made the first trial to separate or concentrate
the Malaria infected erythrocytes (paramagnetic due to deoxygenation and/or oxidation of
hemoglobin) from the un-infected (presumably diamagnetic) erythrocytes with an inhomo-
geneous magnetic field. They could concentrate the paramagnetic erythrocytes in sorne
extent, and finally they wrote that the Malaria-infected erythrocytes were separable frorn
Effects of Static Magnetic Fields on Erytlirocyte Riieology 189

unparasitized erythrocytes in 5 to 25 blood samples in a strong, unsymmetrica! magnetic

field, but the method does not appear suited to the preparation of large quantities.
In 1975, Melville et al [7] succeeded to separate the paramagnetic erythrocytes from
the flowing blood in a column filled with stainless-steel wire, which produces a strong
inhomogeneous field in an externally applied uniform magnetic field. Various methods of
the magnetic separation have been widely applied in biotechnology [27]: e.g.. Rous and
Beard (1933) used magnetic separation technique to select the living Kupffer cells which
phagocyted ferromagnetic iron oxide [28]. Similar techniques arc employed for separating
the activated leucocytes which phagocyted iron granules [1].
We have been studying the effect of paramagnetic attractive force on the flowing
erythrocytes under an inhomogeneous magnetic field. Our aims for the following three types
of hemorheological experiments were (i) to establish a theoretical basis for the safety criteria
of static magnetic field on the blood circulation and (ii) to find the appropriate conditions
for the possible animal expertiments under magnetic fields. The essential parts of our results
are summerized: in the order 1) displacement of a narrow stream line of erythrocyte
suspension in a wide laminar flow [8,9], 2) a small change in the distribution of erythrocytes
flowing through a cylindrical vessel [10], and 3) an acceleration of the erythrocyte sedimen-
tation rate [11],

D-1. the Displacement of Erythrocyte Stream Line in Laminar Flow [H. 9]. (i) U s-
ing the fiow channel shown in Fig. 3, the position of the narrow stream line was measured
on microscopic photographs before and during the application of the magnetic field and the
displacement of the stream line due to the magnetic attraction was obtained by comparing
the two values. Fig. 4 shows the displacement of the stream line of erythrocyte suspension,
whose diameter is ca. 80 (.im in a wide laminar flow (Fig. 3a). The displacement is detected
only for paramagnetic erythrocytes, but not for the oxygenated erythrocytes containing
diamagnetic oxy-hemoglobin. The displacement became larger with the increase of: (a) the
product of magnetic field strength and its spatial gradient, (b) the paramagnetic susceptibility
of hemoglobin in erythrocytes, and (c) the reciprocal of the flow velocity. Further, it
increased in parallel with (d) hematocrit of the suspension,
(ii) The forces acting on a flowing paramagnetic particle are shown in Fig. 5. The
force F,„.,g is due to the attractive magnetic interaction,
F,nug = % V ( H . V ) B + q ( v x B + E)
the first term is obtained by differentiating the potential energy (cp = Vx HB'2) of a
paramagnetic particale in the magnetic field, using V x B (and H) =0. Here V (nabia) is a
differential operator. This term describes the magnetic interaction of erythrocytes (suscepti-

(A)

Figure 3. Flow channels and an iron blocl< attaclied to an electromagnet, (a) For tlie observation of the
displacement of an erythrocyte stream line in the laminar flow [8,9]. The dimension of the channel was 4 x 0.4
X 150 mm, and the width of the stream line (shown with a dotted line) was ca. 80 \\m. (b) For the detection of
excess flow of erythrocytes to the side branch [10] The inner diameter of the main cylinder was 2.0 mm and
that of the side branch 1.0 mm. The inhomogeneous magnetic field was made using an electromagnet and an
iron block with one side tapered, to which the channel or the tube is attached lightly. The magnetic flux density
around the model channel was measured point by point with a gauss meter, then the spatial gradient and the
"averaged" value of the product, B x dB/dz, were calculated [8,9,10].
190 T. Shiga et al.

0)CY CN-MET DEOX¥ MET

imttlc:)
. If 1 K\.^i
0
1 1^ \ 1
150
t 1 1 11 -^ Bz-dBz/dz ^
0 05 1-Omm

Figure 4. Displacement of tlie erj'throcyte sfrcam line. Top panels show the position of erythrocyte stream line
in the absence of magneticfield;bottom panels are those in the presence of a magneticfield,with erythrocytes
(liemalocrit of 5 %) containing [from left] (a) oxy-henioglobin, (b) low spin methemogiobin, (c) deoxy-hemo-
globin, and (d) high spin methemogiobin, The microphotographs were taken at the position y=l50 mm. The
unfocused spots were small dust attached outside of the Oow channel. Standard experimental conditions: flow
velocity 0,7 mni,'s, average value of B x dB/dz 29 T'/m [For details, see ref 9],

bitity X and volume V) with the external magnetic field (magnetic flux density B, magnetic
field intensity H). The second term is the mteraction of erythrocytes (charge q) with the
electric fields v x B (due to Lorentz's Force) and E (due to Hall effect). Neglecting the
Lorentz force and the induced Hall effect, because they are small for a slow blood flow,
F,„ag=xVn-'(B.,xdB,./dz),
where x is the paramagnetic susceptibility of material, V the volume, ji the magnetic
permeability of the solvent, B^ and dBj. ./dz magnetic fl.ux density and its gradient with
z-coordinate (z is the direction of the magnetic attraction), respectively.
The frictional Stokes force, Ff, gives a resistance for the displacement,
Ff = 6 It T| R u,
where u is the velocity of displacement, and R the hydrodynamic radius of the moving
particle in a medium of viscosity ri. Ffis equilibrated to F^^g within a short time. From these
two equations, the velocity of displacement (u) is obtained, and by integrating u along the
path (y-axis) the estimate of displacement (L) can be calculated,

L = j u dt = f" F„, ,,{53iii|i Rv) dy

O yl

= V X / (ft II h ^iRv) f B, (dB, /dz) dy ,

where Y] and Y, are the y-coodinates of the particular erythrocytes at t=0 and at t=T
when the observation was made, respectively, and v is the flow velocity (y-direction), thus
dt=dy/V.

Fmag

I ,ig : llgure 5, Displacement of the paramagnetic erythrocytes due to the interaction

I ! with an inhomogeneous magnetic field in a flow experiment (sec text).
Effects of Static Magnetic Fields on Erythrocyte Rlieology 191

Although the displacement of the erythrocyte stream line occurred in accord with the
above theory, the observed displacement is much larger than the calculated value for a single
erythrocyte. Moreover, the displacement becomes large with an increase in the hematocrit
of erythrocyte suspension. These results indicate that a group of erythrocytes (in a "volume")
is attracted as a whole by the magnetic force, as discussed previously [8,9]. Actually a good
agreement was obaing between the observed and theoretical values of displacement, when
we calculate the displacement for a volume of spheric droplet, of which the diameter R is
the same as the diameter of narrow stream line and the magnetic susceptibility x as that for
the flowing erythrocyte suspension.

D-2. Asymmetric Distribution of Flowing Erythrocytes in a Vessel [10]. (i) With

the setup shown in Fig 3b, which is similar flow channel tested by Heidelberger et al in 1946
[6] but with much higher magnetic field, a displacemet of erythrocytes due to a strong
inhomogeneous magnetic field was observed [ 10]. In order to separate the paramagnetic cells
from blood, an extremely high gradient magnetic field was necessary [7]. As we have shown,
with paramagnetic erythrocytes containing the high spin methemoglobin , the skimmed
erythrocytes into the side branch of the flow channel of Fig 3b increased by the magnetic
attraction [10]. The degree of attraction increased linearly (a) with the magnetic susceptibility
of erythrocytes, (b) with the product of magnetic field strength and its spatial gradient, but
it saturated above ca. 20 T"/m, and (c) with the reciprocal of the flow velocity. Further, (d)
a peculiar hematocrit dependence was found: i.e., at the hematocrit of ca. 5% the maximal
attraction took place, while at 45% no attraction was detected. In the case of mixed
suspensions containing erythrocytes with high spin methemoglobin (paramagnetic) and
oxygenated erythrocytes (diamagnetic), the attraction reached maximum at the "partial
hematocrit" for the paramagnetic erythrocytes of ca. 5% and remained nearly constant with
a further increase of the "partial hematocrit" [10].
(ii) In order to quantify the degree of the magnetic attraction of deoxygenated
erythrocytes, we have to measure anaerobically the amount of ervthrocytes skimmed into
the side branch . Therefore, a spectrophotometer with long optical fiber was attached at the
rectangular side ranch just below the magnetic field, taking the difference spectra before and
after exposure to the magnetic field. The changes for the skimmed erythrocytes (AV) are
defined as : AV= (amount with field )/(amount without field) - 1.
The following results were obtained [29]. AV increased proportionally to (a) the
product of magnetic field strength and its spatial gradient, but saturated at weak magnetic
field (20 T'/m ); (b) the magnetic susceptibility of hemoglobin derivatives in erythrocytes;
and (c) the reciprocal of the flow velocity, (d) The effect is maximal at the optimal hematocrit
of 5%. Further, (e) AV varied with the degree of deoxygenation, proportionally to the
estimated magnetic susceptibility.
(iii) These phenomena can be interpreted as due to the asymmetric distribution of
flowing erythrocytes made by its paramagnetic interaction with the magnetic field in the
cylindrical vessel. The saturation phenomena, at high magnetic field and'or at high hema-
tocrit, may be arisen from the collision among erythrocytes and to the vessel wall, which
produces the local eddy currents and turbulent flow, and the magnetic field-induced asym-
metry in the distribution tends to disappear. Therefore, as the collision rate increases with
increasing the magnetic field and/or the hematocrit, the magnetic effect (asymmetric distri-
bution) is apparently suppressed.

D-3. Acceleration of Erythrocyte Sedimentation Rate [11]. The sedimentation rate

of paramagnetic erythrocytes in the Westergren tube increased in a spatially inhomogeneous
magnetic field, but not in a homogeneous field. Moreover, no magnetic effect was detected
for diamagnetic erythrocytes. This phenomenon is similar to the so-called Boycott effect
192 T. Shiga etal.

[30], i.e., the accelerated sedimentation in an inclined cylinder [31]. However, the detailed
mechanism of Boycott effect is not fully understood.

D-4. Summary' for Paramagnetic Attraction. The eftect of an external inhomogene-

ous magnetic field on the flow of erythrocytes containing paramagnetic hemoglobin was
systematically studied with three experimental setups. The results are summarized as
follows:
1) Paramagnetic erythrocytes are attracted by an inhomogeneous magnetic field.
2) The degree of attraction is proportional to (a) the product of magnetic field strength
and its spatial gradient, (b) the magnetic susceptibility of hemoglobin, and (c) the reciprocal
of the How velocity. The attraction is detectable with the flow channels under an inhomo-
geneous magnetic field above 10 T-/m.
3) The degree of attraction depends on the hematocrit, with the maximuin effect at
ca. 5%.
4) Such asymmetric distribution of venous, deoxygenated blood flow in an inhomo-
geneous magnetic field of above 100 T-/m may induce some biological effect.

CONCLUDING REMARKS

The effect of a static magnetic field on erythrocyte rheology was summarized. In

order to detect circulatory effects by the magneto-hydrodynamic action, a considerably
strong field will be needed. Minor effects due to the diamagnetic interaction may appear at
the field strength of several T. On the other hand, the paramegnetic attraction affects only
the venous blood flow, when the product of magnetic field strength and its spatial gradient
exceeds above 100 T-'m, especially in anemic situation under slow flow.

ACKNOWLEDGEMENTS

Our works were supported in part by the grants from the Ministi-y of Education,
Science and Culture of Japan, Nissan Science Foundation, and Ehime Health Foundation.

REFERENCES
l.Levinc. S. . 1955, Magnetic techniques for in vitro isolation of leucocytes. Science. 123: 185-1X6.
2. Murayama. M.. 1965, Orientation ofsickled erythrocytes in a magnetic field. Nature. 206: 420-422.
3. Brody..\.S., Sorette.M.R. Gooding.C.A., Listerud, J., Clark,M.R., Mentzer.W.C. Brasch, R.C. and James,
T.L., 19S5, Induced alignment of llowing sickle erythrocytes in a magnetic Held: a preliminary report,
Invest.Radiol., 20: 560-566.
4. Chen. I.I.H and Saha. S.. 19X4 , Analysis of an intensive magnetic fiels on blood flow. J.Bii)elcctricity,.^
:293-298.
5. Yainagishi.A.. Takeuchi,T., Higashi ,T. and Dale.M., 1992. Diamagnetic orientation of blood cell.s in higli
magnetic field. Physica B, 177:, 523- 526.
6. Heidelbcrger.H.. Meyer.M.M. and Damarest.C.R., 1946 . Studies in human malaria. 1. The preparation
of \accines and suspensions containing plasmodia. J.Immunol,,52: 325-330.
7. Vlehille, D.. Paul. F, and Roath.S., 1975 , Direct magnetic separation of red cells from whole blood.
Nature, 255 : 706.
8. Okazaki,.Vl.. Maeda.N. and Shiga.!.. 1986 . Drift of an erythrocyte flow line due to the magnetic field
E,\perientia. 42: 842-843.
9. Okazaki.M.. Maeda.N. and Shiga,T,, 1987, Effect of an inhomogeneous magnetic field on flowing
erythrocytes. Eur. Biophys.J.. 14 :139-145.
Effects of Static Magnetic Fields on Erythrocyte Rheology 193

10. Okazaki.M., Kon, K. Maeda,N. and Shiga,T., 1988, Distribution of erythrocyte in a model vessel exposed
to inhomogeneous inagnetic field. Physiol.Chem.Pliys.Med.NMR,20 : 3-14.
11. Okazaki.M., Seiyama, A.. Maeda,N. and Shiga,!., 1991 . Boycott effect with vertical cylinder for
paramagnetic red blood cells under the inhomogeneous agnetic field. J.Colloid Interface Sci,, 146:590-
593.
12. Chen, 1.1.II. and Saha, S., 1985 , Analysis of an intensive magnetic field on blood flow: part 2.
J.Bioelcctricity, 4: 55-61.
13. Togawa.T., Okai.O. and Oshima.M., 1967 , Observation of blood flow E.M.F. in externally applied strong
magnetic field by surface electrode. Med. Biol. Engng., 5: 169-170,
14. Okai.O.. Togawa,T. and Oshima,M., 1971 , Magnetorheography: nonbleeding measurement of hlood
flow. .I.Appl. Physiol..30: 564-566.
15. GatTey.C.T., Tenforde.T.S. and Dean,E.E., 1980 . Alterations in the electrocardiograms of baboons
exposed to DC magnetic felds. Bioeleclromagnetics, 1: 209-211.
16. Maret.G., v.Schickfus.M.. Mayer.A. and Dransfeid.K., 1975 . Orientation of nucleic acids in high
magnetic fields. Phys. Rev. Lett., 35: 387-400.
17. lorbet,.!., Freyssinet,J.-M., and Hudry-Clcrgcon.G., 1981 , Oriented fibrin gels formed by polymerization
in strong magneic fields. Nature, 289: 91-93.
18. Yamagishi.A., Takctichi,!,, Date,M, and Higashi.T.. 1989 . Polymerization of biolgical molecules under
high magnetic fields. Physica B. 155:433-436.
19. Yamagishi..'\., Takeuchi.T.. Higashi.T. and Date,M.. 1990 . Magnetic field effect on the polymerization f
fibrin fibers, Physica B,:164, 222-228,
20. Higashi,T,, Yamagishi.A.. Takeuchi.T.. Kawaguchi.N., Sagawa.S., Onishi.S,. and Date.VI.. 1993, Orien-
tation of erythrocytes in a strong magnetic field. Blood. 82:1328-1334.
21. Rosenblatt,C. Yager.P. and Schoen.P.F... 1987 . Orientation of lipid tubules by a magnetic field.
Biophys.J., 52: 295-301.
22. Sakurai.l, Kawamura.Y, Ikegami.A. and Iwayanagi.S., 1980 , Magneto-orientation of lecithin crystals.
Proc.Natl.Acad,Sci„77:7232''-7236.
23. Pauling,L., 1979 . Diamagnctic anisotropy of the peptide group. Proc.Natl,.^cad.Sci.. 76:2293-2294.
24. iligashi.T., Sagawa.S,. Kawaguchi.N. and Yamagishi.A.. 1993 . Effects of a strong magnetic field on
blood plateels. Platelets, 4: 341-342.
25. Higashi.T. Yamagishi.A., Takeuchi,T. and Date.M., 1995 , Effects of a strong magnetic fields on
erythrocyte rheology. Bioclectrochem. Bioenerg.. in press.
26. Klucznik,R,P,. Carrier.D..A.. Pyka.R.. and Haid,R.VV., 1993 . Placement of a ferromagnetic intracerebral
aneurysm clip in a magnetic field with a fatal outcome. Radiology, 187: 855-856.
27. Dunlop, E,H,, Fciler.W,.'\, and Mattione,M,J., 1984 , Magnetic separation in biotechnology, Biotech.
Advs,, 2:63-74,
28. Rous. P. and Beard.J,VV,. 1933 . Selection with the magnet and cultivation of "'reliculo-endothelial" cells.
Science, 77: 92,
29. Shiga,T, Okazaki.M,. Seiyama,A. and Maeda.N., 1993 , Paramagnetic attraction of erythrocyte How due
to an inhomogeneous magnetic field. Bioclectrochem. Bioenerg.. 30: 181-188.
30. Boycott. A.E., 1920 . Sedimentation of blood corpuscles. Nature. 104 : 532.
31. Dobashi.T. Goto.H.. Sakanish^.A. and Oka,S.,l987 . Erythrocyte sedimantation rate. II. Effects of tilted
angle in saline solution. Biorheology, 25: 157-164.
 15

PUBLIC HEALTH ON ELECTROMAGNETIC

FIELDS AND MAGNETIC SHIELD OF
LINEARMOTORCAR (EDS) MAGLEV

M. Nakagawa, M. Ikehata, and T. Koana

Railway Technical Research Institute

Hikaricho 2-chome, Kokubunji
Tokyo, 185 Japan

1. SAFETY
1.1. Denotation of "Safety" as a Legal Term
s a legal term, the Labor Safety and Hygiene Law (LSHL; enforced October, 1972)
uses it with a denotation of laborious disaster prevention. Activities for disaster prevention
of workplace are divided into safety management and hygienic management by a narrow
sense. Here, we think that it is prevention of an emergent danger with a task of "safety" of
a narrow sense concerning LSHL, and we are allowed to think it is prevention of a health
impairment that occurs chronically with an issue of "hygiene". We state about "safety"
related to magnetic field or electromagnetic field. Safety issues about a case of MR! taking
a superconducting magnet (SCM) as an example, are as follows; 1. a strong magnetic field
attracts an iron-made oxygen bomb and a stretcher (simple bed with wheels that carries a
patient prostrate) or causes a hammer and operative instruments to fly. injuring personnel
nearby; 2. danger of an asphyxia for a personnel nearby, when a SCM is quenched under
insutFicient ventilation; and 3. when dB / dt change in magnetic field of super-high-speed
MRI equipinent is large, induced currents stimulate nerves, muscles and directly act on the
heart.
In an industrial scene, we can easily understand "safety" issue as an accident which
is visible directly. "Safety" issue is generally well known, since we were concerned about it
earlier than about "hygiene" issue. As we are interested in weak influences of the factors that
shift and persist chronically, the present revised Act for Labor Safety and Hygiene (ALSH)
puts stress on a hygienic matter. We see not only "safety" matter that the ALSH treats, but
also "hygiene" issue. We discuss on safety issue with a wide denotation that includes
professional and general personnel.

Biological Effects of Magnetic uiui Elcctrornagnclic Fields, Fdiled by S. Ueno

Plenum Press. New York. 1996 195
1% M. Nakagawa et al.

1.2. Division of "Safety" Conception with Hygienic Management

When we speakc of safety as a basis to hygienic management we mean that we do
not meet an accident, and that there is no worry even though we are exposed to various
factors over a long term. To be specific, we aim at each factor by which we can manage to
hold the exposure amount below the level that causes no health impairment, when an 8-hour
per day exposure is supposed to last 30 years. A theme that comes next is what level of
exposure amount is safe and over what level is harmful. As the basis to determine the
exposure-dose level, the concept called dose-response relation [I] is known for hygienic
management. This concept shown in Figure I indicates that as exposure dose increases, the
hurnan response progresses from the first step of homeostasis that can be compensated, to
sickness able to recover from, to sickness unable to recover from, and finally to death. How
to divide the levels of this stimulus factor is a difficult job involving not a few physical
factors which are unlike chemical materials. With a chemical substance, for example, a
carbon monoxide, combines with hemoglobin and makes the whole body short of oxygen,
division between harm and no harm is very clear. However, physical stiinuli, except strong
ones like radioactive rays, show no clear-cut border between safety and danger.
Recently an interest is aroused in influence of the factors such as non-energy or
non-ionizing agents. Because an action of such factors is extremely weak, dose-response
relation is hardly found in many cases. For such weak physical factors, we can count other
factors besides electromagnetic fields that we treat here. These are radio-frequency radiation
(especially 300 MHz from 30 kHz), ELF radiation (from 3 to 300 Hz or up to 3 kHz), cold
stress, and change of pressure, and distinguishing between safe and unsafe is considered is
also a difficult job within either of them. As far magnetic exposure dose (frankly speaking
on this concept there is no definition as yet), we try to think of 4 steps in evaluation of the
responses; i.e. 1. safe without specific anxiety, 2. indistinct, 3. harmful, and 4. dangerous.
However, from a point of view of hygienic management, a magnetic issue is considered an
object that involves a safe and indistinct step. Then, there remains the division of these first
and second steps in a very ditTicult situation of these days. About how to distinguish between
second and third step is similarly difficult.

Death
t
UJ
(n
Malcompensation 2
O
D.
CO
UJ
Disease

Wellness

DOSE-3J
i) i2> v3' Figure I. Dose-response relation, indicating changes
CO c D D upward in response as magnitude of dose is increased.
in
A level of non response exists below a threshold dose
level of above zero. Threshold dose of electromagnetic
tTelds with no health risk is also unknown yet.
Linearmotorcar (EDS) Maglev 197

2. PROGRESS OF MAGNETIC APPLICATION

2.1. Classification of Magnetic Fields

Application of magnetic fields has advanced with scientific-technological progress.
Energy mployed for the machines utilizing magnetic fields gets very large. With utilization
of magnetic fields extended further, we have to think about safety in various applied scenes.
Today, concern for magnetic fields as a matter of safety may be involved in an
application of static-magnetic field; 1. a super-strong magnetic field of the equipment using
a superconducting magnet, in an application of a time-varying magnetic field, 2. the magnetic
field applying an induced cirrent by large dB/dt, 3. a time-varying magnetic field of ELI'
(extremely low frequency), and 4. high-frequency electromagnetic fields. In these four large
classifications, however, as we think about safety of magnetic fields to the human body, we
see that mechanism of each action is different. Next, we consider for what the magnetic field
in each equipment is used.

2.2. Equipment Using Static Magnetic Field

It seems to be thought generally that a static-magnetic field does not affect deleteri-
ously a living body. That might be a mere concept, however, in the age when there exists
none of such gigantic equipment using a static-magnetic field with several T (tesia).
As an opportunity that average people encounter today, it is a rare case that MRI
(magnetic resonance imaging) equipment for the whole body of 4 T is trially made. A lot of
MRI equipment with 0.5 T- and 1.5 T-superconducting magnet arc sold as mass production
machines. We were not able to imagine that we might have an opportunity of our head and
body trunk being exposed to a magnetic field of 0.5 T, before MRI has come to be actually
used for medical diagnoses. As other large superconducting magnet equipment, there are
EDS (linearmotorcar)-maglev, MHD (magnetohydrodynamics). SMES (superconducting
magnetic energy storage), and a fusion reactor of a tokamak- and a mirror- type. However,
only some specialists have an opportunity of access to strong magnetic field of the central
portion. For others the equipment using a large magnetic fiux density are NMR (nuclear
magnetic resonance), MRS (magnetic resonance spectroscopy) and hybrid type super-strong
magnetic field equipment. However, with these equipment, there seems to be no chance of
an experimenter being exposed to a super-strong magnetic field. The reason is due to this
structure with a super-strong magnetic field being installed in a secluded narrow space.

2.3. Magnetic Fields of Large dB/dt

Generally such a magnetic field is called a pulse magnetic field and it is intended for
a special purpose. As one expected to have an effect on a living body, the first equipment
produced for a medical treatment of bone-fracture was Bi-Osteogen system. This experimen-
tal equipment was built 20 years ago[2] to apply electric currents in the human body induced
by a lime-varying magnetic field. Early in 80's devices began to be made which could
generate so large induced currents as to stimulate nerve and muscle[3]. Barker et al. [4] after
that stimulated a human cerebral cortex and invented use of this equipment for pathological
diagnosis of the nerves. It is a merit of the machine that it can stimulate nerve-cells of cerebral
cortex from the outside without craniotomy, free from any unpleasant action of an amperage
such as in direct stimulation by electrodes. As an application of this merit, a direct stimulation
of volitional muscles and/or a respiratory muscle (diaphragm) is also tried. Further, an
198 M. Nakagawa et al.

apparatus for magnetic stimulation is produced in large quantities as a clinical test machine
(e.g. SMN-1100 of Nihon Kohden Corp.)

2.4. ELF Time-Varying Magnetic Fields

ELF (extremely low frequency)-band indicates a frequency range of 3 kHz from 3
Hz in electric waves. As frequency of a power transmission line is 50 Hz or 60 Hz, a
time-varying magnetic field affecting the residence near the transmission lines belongs to
this frequency band. Additionally, there are EMF of low-frequency band from the heaters of
an industry use, harmonic ingredient that occurs from transport facilities, and other noise-like
electromagnetic fields with this band.

2.5. High Frequency Electromagnetic Fields

When high-frequency electromagnetic field is used only near the generative source,
the electromagnetic field has a disposition that differs from that of electromagnetic waves
used for communications covering a long distance. Electromagnetic field within a distance
of ' 2 from a wave source is called a near field. There, however, seems to be no unity about
the definition and character of this field.
Usually, as high-frequency electromagnetic field sources, the cite an induction heater
(180 Hz ~ 1 MHz) applying Joule's heat of induced currents, a dielectric heater (3 MHz ~
100 MHz), and a microwave heater (2.45 GHz, 5.85 GHz) utilizing dielectric loss. Repre-
sentative equipment of the former is an electric furnace used for iron manufacture and metal
processing, those of the latter are a plastic sealer, diathermy, and microwave oven. As
electromagnetic waves of far field with high energy area, there are microwave communica-
tion and radar. There are some cases where a radar is treated as belonging to ELF band
because it is remodulated with ELF.

3. EMF ENVIRONMENTS AND THEIR SAFETY

3.1. Static Magnetic Field

3.1.1. Static Magnetic Field of Superhigh Intensily. Though a superstrong static

magnetic field is used widely with application of a superconducting magnet, the studies of
a super-strong magnetic field for its influence on a living body are surprisingly few. The
reason might be speculated that the interest in recent years is concentrated on influences of
ELF electromagnetic fields.
An obvious effect of a static-magnetic field on composing ingredients of biota is to
cause the diamagnetic macromolecules to be oriented in the magnetic direction. For this
mechanism, the following phenomenon is attributed: that when a diamagnetic substance has
different magnetizing rates by crystal axes, this material orients itself along the weak
magnetized axis to an outside magnetic field direction. In a gigantic molecule structure with
a similar configuration of repeating diamagnetic substance such as biomembranes and DNA,
an orienting power times the integer of a molecule number has a chance to turn itself toward
the direction of a magnetic field. Murayama [5] is known as the person who reported on this
phenomenon in early stage. A theory of nuclear acids orientation is discussed by Maret et
al. [6]. Recently super-strong magnetic fields are used in the experiments, some reports
saying that a blood clot gets weak by an orientation of fibrin at coagulation [7].
Lincarmotorcar (EDS) Maglev 199

The probable effects of this mechanismon a living body are a change of metabolism
accompanied with mechanical distortion of biomembrancs, or direct effect on gene functions
and expression. It is difficult to prove a substantial effect of static-magnetic field changing
the serum-chemical substances found in mice exposed for long term to a super-strong
magnetic field [8], though some relation could be suspected. An effect on genes is also one
of large themes in recent years. However, the effect seems not to be so strong as to be
suspected from the mechanisin in a static-magnetic field.A growth of strong bacteria on
fertility is not affected even in a strong magnetic field of 11.7 T, unless the condition of
cultures is so specific [9]. We found that the effect of a magnetic field is extremely weak by
our sharp mutation detection system of pyrimidine dimers using mutant fruit fly that lacked
repair functions of DNA damage. A mutation rate caused by 24-hour serial exposures to 0.6
T, was almost the same value of 5-s irradiation (12.5 Jm"-s"') at a distance of 50 cm from a
usual 15 W germicidal lamp [10],
A real case that human whole body, especially head and body trunk, is exposed to a
strong magnetic field is a medical diagnosis using MRI. Though usual flux density of mass
production machines is 0.50 - 1.5 T, it is said that three trial production equipment of MRI
for a whole body diagnosis with 4 T are available in the world. Schenck concerned in
development of 4 T-MRI for the whole body of General Electric Corp. tested exposures of
11 volunteers for a year according to a protocol of Pennsylvania University [11]. The total
exposure time of these volunteers was 150 hours that is less than 1 hour to 40 hours per one
volunteer. Though healthy men and women were selected for the test, nothing objective was
found by a check before and after the exposure in a medical examination by a medical doctor.
However, some personnel conspicuously complained of unusual conscious perception in a
reply to the questionnaire. There were reactions of vertigo, nausea, and metallic taste that
differed with 2 groups of 4 T and 1.5 T depending of the strength of a magnetic field
significantly, while there was no difference about headache, vomiting, tinnitus, balance
difficulties, numbness, or hiccuping. Magnetophosphenes were also noticed in both groups.
However, the perceptions felt at 4 T were clinically all light grade, and they did not occur at
all unless the subject abruptly moved the head inside a magnet or they were weak if ever
occurred.
"Safety" issue of magnetic field in medical examination and equipment treatment
seems to remain usually at the extent comparable with any other high-energy equipment.
After all, comparing electromagnetic fields for X-ray CT, electromagnetic field is recognized
to be safer. Even though an exposure to a super-strong magnetic field of 1.5 T of MRI is
brief and repeated examinations may be a few, we can say that safety in medicine is not a
matter of extent.

3.1.2. Estimation of Minimum Amount of Static Magnetic Exposure That Produces a

Detectable Effect on Mammals, and Calculation of Safety Exposure Levels.

(!) Introduction. In recent years, we occasionally come close to a strong magnetic

field. For example, magnetic resonance imaging (MRI) technique is practiced in many
hospitals for diagnosis. Some of the newest models of MRI device are equipped with
superconductive magnet, and they generate 1.5-T flux density. Advance in medical engineer-
ing is so rapid that we have little time to discuss potential effect of strong magnetic field on
human body before MRl-diagnosis is widely spread. In contrast, the US Department of
Energy proposed very low flux densities as safety standard for workroom environment. The
strict standard is, however, sometimes criticized for hampering normal operation in factories
and laboratories equipped with strong magnets. The author shall illustrate biological effects
of magnetic fields, estimate minimum amount of magnetic exposure that produces a
detectable effect on mammals, and propose a tentative safety standard.
200 M. Nakagawa el al.

Inconsistent results on biological effects of magnetic fields have been reported from
experiments involving various combinations of experimental variables, such as the species
of animals, exposure environment, frequency and flux density of the magnetic field, duration
of exposure and biological end point. Among these variables, the most critical factor appears
to be the amount of exposure as defined by flux density and exposure duration, with more
weight on the latter. In general, brief exposure of biological materials to a static magnetic
field, even under highly intense flux densities, has produced no biological effects. Null
examples are the nerve-conduction velocity after a 20-30 min exposure to 1.2 T [ 12] and the
growth rate of mammalian cells exposed for 2 h to 1.75 T [13]. In contrast, positive results
have been found in studies in which biota were exposed to a static magnetic field for a
prolonged period: e.g., Drosophila for 3 days to 0.44 T [14], mice for 35 days or more to
0.42 T [ 15], mice for 30 days to 1.6 T [ 16], rabbits for 5 weeks to 0.06 T [ 17], and mice for
4 weeks to 0.42 T [ 18]. F.ven at high densities used by many investigators, the static magnetic
field is probably too weak to cause substantial changes in animals exposed only for a short
period. Therefore, lengthy exposures and prolonged periods of observation after exposure
are indispensable if one is to obtain reliable results by which the existence of an effect can
be ascertained or denied. Magneto-electromotive force, magnetic anisotropy [19], and
modification of chemical reactions [20] are regarded as the primary factors involved in the
mechanism underlying the biological effects of static magnetic fields. Even a flux density
of 0.2-0.5 T is effective at the molecular level. However, there is a wide gap between the
understanding at the molecular level and that at the whole-body level. Selye [21] demon-
strated that the homeostatic responses of organisms to external stimuli follow a characteristic
pattern, which he named "general adaptation syndrome" (GAS). The progress of GAS takes
rather long time, so a persistent observation for days or weeks is necessary to ascertain this
pattern. Barnothy & Barnothy [22], Barnothy & Siimcgi [15], Laforge [23], and Nakagawa
& Matsuda [24] have demonstrated that reaction of animals to a static-magnetic field shows
the GAS-type pattern.
The activity of an animal is depressed immediately after a stressful stimulus is given
(shock phase), then it recovers beyond the original level of activity (counter-shock phase).
Both phases constitute "the stage of alarm reaction"; a characteristic feature of the GAS.
When the intensity of stimulus is moderate or exposure period is short, specific reactions
disappear after the counter-shock phase without showing a "going-to-death pattern" de-
scribed by Selye. Therefore, identification of "stage of alarm reaction" following a magnetic
exposure was taken as a criterion for successful detection of a biological effect of magnetic
field.

(2) Method and Ccixc Traces. Barnothy & Barnothy [22] identified a typical pattern
of the GAS al the "stage of alarm reaction" by a body mass change in mice that had been
exposed to a field at 9,400 Oe (ca. 0.94 T) for 4 consecutive days. They specified it as a
"shock effect" caused by magnetic field.
Figure 2 illustrates the result of our experiment. It shows changes in the body mass
of mice that were subjected to a field of 0.6 T for 16 h a day for 4 days [14]. The double
dotted line indicates data obtained from control mice and the solid line from exposed ones.
The changes of body mass caused by the exposure were rather small, and the weight returned
to pre-exposure level on the second day. This is probably because exposure amount (which
is operationally defined here as the product of magnetic-fiux density and exposure duration)
was less than that employed by Barnothy and Barnothy [22]. The magnitude of mammals"
response to the magnetic field, in general, seems to be positively related to the exposure
amount, and there may be a threshold of exposure amount below which animals do not show
an observable reaction to the magnetic field.
Linearmotorcar (EDS) Maglcv 201

1,04

1,02

1.00

0.98

Figure 2. Rate of changes in body mass of mice

in the exposure week, under three conditions of
control (O), 0.6 T magnetic field . and 4°C 0.96
temperature (A), with half bars of the standard Mon, Tue, Wed. Thu Fri
error. Day

Figure 3 shows ftequency of shock received in Sidnian's conditioning schedule. Rats

in the experimental group were exposed to 0.6 T for 16 h a day for 4 days [24]. The frequency
of shoci^ received was counted during the final 30 min of a one-hour session and averaged
for each experimental condition (exposed/control, 10 rats each). Asterisks mean a significant
difference between the two conditions (*: p<0.05, **; p<0.01). Shocks received by the
exposed rats were more frequent than those by the control rats during and afier the exposure
period.
Author filed the reports in which magnetic-field density and minimum exposure time
have been determined with certain effects produced at such intensities or densities.

I3) Results and Conclusions. There are only nine reports that found the GAS-like
responses of mammals by magnetic exposure: five from our laboratory; A[26]. B[27]. D
[28], F[24] and G [24], and four from others; C [23], E [29], II [22] and I [16] as shown in

o Control
E
o A Exposed

42

Figure 3. Frequency of shucks received in the final 30 min of sessions in the SA schedule [13]. Exposed rals
(A) received more electroshocks than did control (O) during the exposure period and thereafter. (*: p<0.05.
.
202 M. Nakagawa et al.

Figure 4. The minimum total exposure durations employed in these studies are plotted as a
function of flux density. A power model without a constant term was used to search for a
regression equation. It was found that the solution was y=34x" ^' **. Although this solution is
a preliminary one due to the small number of data, half of the exposure amount indicated by
the above-mentioned function (y=17x*''"* ) can be considered the best estimation now-
available of the minimum exposure that produces detectable biological effects. However, in
order to think about safety standards of magnetic fields, we are compelled to push these
values lower. In that case, generally we decide the values as one tenth of the real effect. We
reached the standard values taking about one tenth of a dotted line in Figure 4 (y=l .7x"^ ).
According to this inference, these values are 7.6 h with 0.05 T, 5.4 hours with 0.1 T, 2.4
hours with 0.5 T, and 1.7 h with 1 T and 1.2 h with 2 T.
It is difficult to estimate the effects of high-density magnetic fields, since there is no
report on mammals exposed to a magnetic field over 2 T in which prolonged observation
was carried out to obtain the GAS patterns. It seems that a considerable amount of exposure
is required to make the changes on molecular level appear outside of the body exceeding the
limits of homeostasis. As demonstrated in Beischer's report [21 ], he did not detect any effect
on mice during one hour exposure to a magnetic field at extremely high flux density of 14
T. A certain length of exposure time would be needed for mammals, even a magnetic field
of ultra-high flux density, because magnetism, unlike ionizing radiation, does not destroy
biological structures instantaneously.

xA

150-
>B

as
Q

loo-
'i \ "C
(D CO

xE
X 9-
Q>

se- \ X D ^*^*„_ XH
o
X
XF y=34x-'"'«

i ^ Safely zone proposed

""y="l7x"'''^"

0.5 1.0 1.5 2,0

Magnetic field density (tesia)

Figure 4. Minimum exposure duration was plotted as a function of flux density employed in the studies in
which a GAS-like pattern was observed in mammals in response to the magnetic field. Responses observed in
Studies A to 1 are as follows: A: ln(HD-cholesterol)/ln(total cholesterol) in rabbits, B: food consumption b\'
mice, C: delayed reinforcement of low rate by rats, D; urination in mice, E; body mass in mice, F-,G; operant
responding by rats, H; body mass in mice, 1; adrenal corticosterone in mice. The power model of regression
on density of magnetic field and exposure time (y=ax''), half of which was presumed to be the minimum amount
of exposure that produces a detectable effect in combination with exposure time against flux density of
magnetic field. Values of about one-twentieth (a=1.7, b=-0.5) can thus be used as safety standards for whole
body exposure.
Linearmotorcar (EDS) Maglev 203

This chapter was rewritten based on the paper "Mammals' response and adaptation
to static magnetic fields as a nonspecific stressor" [31].

3. 1. 3. Some Recommended Safety Standards for Static-Magnetic Field. By 1986,

some so-called safety standards on magnetic field proposed were all the objects regarding a
static-magnetic field. It seems likely that these standards have been influenced by a series
of results from the study with extremely prolonged exposure period by Barnothy et al. at the
beginning of work on this area [15, 22]. They considered that the research results at the
beginning stage had evidence that a magnetic field caused plenty of organic changes.
On the other hand, reports on effects of static MF from the United States in recent
years and experimental reports concerning MRI diagnosis are almost negative. These reports
were based on the experiments that were faithfully performed according to a method of
radiobiology. relying on the concept of safety in clinical radiology. Consequently, clinical
radiobiology seems to adopt that of kind of a clinical medicine. That is to say. clinical
medicine takes precedence in a medical application of magnetic fields. It can be said that
this concept became a big propulsive power, boosting a clinical application of MRI (NMR-
CT) that appeared as a powerful tool in X-ray diagnosis. Table 1 shows supposedly main
examples of safety standards published so far. In 1970's, we thought that conventional
research results were added for consideration. And after early stage in 80's. we could hardly
find a report on this issue discussing from standpoints of environmental hygiene and
ecological ideas, The limited marginal curve of Nakagawa (figure 4) by magnetic field
strength and exposure time was drawn using only the endpoints that noticed a pattern of
GAS (general adaptation syndrome) after a long-term observation made before and after the
exposure period. This limited exposure time is quite long as stated before. Values E
(Nakagawa) in Table 1 shows one tenth of that time for a reference.Three years after that
recommendation ACGIH published a value 60 mT, time-weighted average of a day. as a
threshold limited value of static-magnetic field as given in the table (values F). Safety
standards of a static-magnetic field look rather rised back.

3. 2. Magnetic Field with High dB/dt

Goodman et al. [32] reported series of results that indicated that comparatively weak
(< 0.2 mT) low-frequency (15 Hz or 72 Hz)time-varying magnetic fields increased a
transcription of RNA, which explains an effect of the Bi-Osteogen System described before.
We think that weak time-varying magnetic fields come to be considered to promote cellular
proliferation from the series of their studies. We refer to evaluation on this matter later, which
is related to the effects of ELF electromagnetic fields. However, there remain difficult points
in plausible estimation. We do not know clearly about what level of dB.dt of a time-varying
magnetic field affects deleteriously a living body.
An obvious effect of MF with high dB/dt is to stimulate nerves and muscles as MF
is intended for this purpose. Ueno et al. recorded smarting pain and decrement of blood fiow
when they brought the palm close to the coil of an induction heater with 3.8 kHz(16 to 48
mT)[33]. T ( rising time ) was 130 |a s, and dB/dt was 1.5 x 10- T s"' with a strength of 20
mT. We have tested whether it is effective or not to use as a respirator to stimulate a diaphragm
with a pulse magnetic field applied from the exterior of lateral abdomen [34]. The condition
for the abdominal muscles and deep diaphragm to be contracted enough, in our equipment,
was that dB/dt = 5.6 x lO-T s"' with a rising time of 250 ]i s.
Figure 5 shows the concept of a safety zone of dB/dt from the 6th plan of lEC's draft
[35] for making a safety treating plan regarding MRI for a medical diagnosis. Our equipment
[34] also holds in C area of this figure, as an output or total energy is large enough. To
stimulate a phrenic nerve at a cervical portion with MF of this strength is rather harmful
204 M. Nakagawa et al.

Tabic 1. Some safety standards and guidelines on static-magnetic fields recommended

today

A. Safety standards by Bcischcr (1962)

Less than ^ d/y 5,000 G

Duration of 15 min 20.000 G

B. Safety standards by Vyalov (1967)

Field Gradient

Whole body 300 Oe lOOOe/cm

Hands 700 Oc 10-20 Oc/cin

C. Safety standards by Stanford Linear Accelerator Center (1971)

Extended periods (h) Short periods (min)

Whole b<xly or head 200 G 2,0<X)G

Arms and hands 2,000 G 20,000 G

D. DOE» interim guidelines (1979)

For 8 h work day For exposure:<1 h For exposure <10 min

Whole b<xly 0.01 T 0.1 T 0.5 T

Extremities 0.1 T IT 2T

'.. Safety standards by Niikagawa (1986)

<0.05 T <0.1 T <0.5 T <1 T <2T

Whole btxly 7.6 h 5.4 h 2.4 h 1.7h 1.2h

F. Threshold limit values by ACGIH (1989)

Time-weighted average ceiling

for a work day

Whole body 60 mT IT
Extremities 600 mT 2T
T(tesla)=IO'G(gauss)==F 10''Oc(cx;rsted). * Department of Energy.

because a sternocleidomastoid muscle contracts spastically and the vagus of the cervical
portion is also stimulated. dB/dt and also duration time T are related with a stimulation of
the living body. Nerve and tiiuscle cannot respond to electromagnetic fields, when the x is
short and so the frequency is high, even though dB/dt might be large. This is the reason why
Figure 5 illustrates a longer portion than 1 |.t s. From a point of view of safety for heart
muscles, an experimental stimulation was carried using a dog [36]. After tetnporarily
stopping the canine heart with a strong stimulation of the vagus, it was possible to move it
again by stimulating with a pulse magnetic field. However, a persistent stimulation as long
as 10 ms is necessary to cause a fibrillation with an electric stimulation of a normal heart.
Lincarniotorcar (EDS) Maglev 205

10' -

Figure 5. Fimelion on the safety concept of lime- 10

varying magnetic field on (rising time)and ciB/'dt. 10 10' 10' 10' 10* ^ ^ f c

Values applied in MRl used nowadays are witliiii

A-area, but, the power used in future^type MRl
witli echo planar method is said to have cases
exceeding C-area where muscles and/or nerves
are stimulated by induced currents in a patient's
body 134],
A ; NOMAL OPERATIING MODE
B ; FIRST LEVEL COI^TROLLED OPERATING MODE
C : SECOND LEVEL COIMTROLLED OPERATING MODE

This energy is so great thai tlB/dt needs to be as high asfix10^ T s^'. Even a future-type MRl
(echo planar imaging) demanthng very high energy cannot come up to that power. It is
supposed difficult to cause cardiac arrest with magoetic stimulation by a real equipment.
As a problem of an unexpected obstacle arising from a large dB/dt of magnetic field,
we cite the case of a cardiac pacemaker. There are many investigations reporting that
temporary abnormal conditions of pacing often cannot be noticed by a pacemaker user. It is
reported that it is feared that an elimination equipment of EAS (electronic article sur\'eil-
lance; cheap watching patch to be attached to the recent products to protect from robbery)
system causes malfunctioning of a mode switch of pacemakers [ 37].

3.3. ELF-Electroiiiagiietic Fields

Thoiiab a ffeque^scy hand is of the feghesi inicrcsl. the opiniori is probably divided
over iL As f!)r clctaromugceuc fleids of the levels of iraiisnnssic'ri isnes, not a fev.' reports
state ihat leukemia isic!dcru:.e mcreaaed in children wiio live under powerinics, or ihar cance;
and leukemia occur freMiiesith' at eieciricai workplace, and only cpideitiioiogical surveys
iridicaie positive slew [381. On die odicr hdiui, if we want to ascertain on this mailer by
exijerirnents. we wiil get aiinwc iiwariabij' negative resufi cunccrning r-ancer nihiation or
proinoiion under the cffcci. oi"\\eaiv eletlroiuaciscti.- field wiili flux dcosiiy of less than 2 ~
5 mT [39]. Therehus:, many expt-riineura! srieniists negaic it.
r igiirc 6 cites sostse grapiis fivsrr! safety gaidciines piopc'sed ncjwadays about electro^^
matiiiciic lleids rajistng uOirs fllF (sdtta lo\v treoueocy; to micfowave. In the first hall ot
80"s safety valtses on higher phase cf tVcqiJcocv were published and sbiltedio the lower one
around 1990 We might say that these vah.;cs are sate enoegh cotisidernig the experirneiira!
i^vKlence. fRff\ ilnfcrnatioiial Radiadosi ProtectiOti Agency, see F>g. 6) is re\ie\viiig die
values ai present in order to add and draw new lines Tor tower and higher freqaency bands,
ihoiigh ir iias published relativefy lower guidehaes. Furthermore, such vtihics that Florida
and New Yorkrecommetsded eonit aroand tise figures measured under the '-eai transmission
hues, or Sweden ts ready iior to recommerKi strictly !ow values coinparafsle with Swoc:ii;.h
regulatson on stray ilelds from VDT. Howcvcis Japanese oilicial opiniori reported in recent
vears states ihat icporis by epideniiologtcal method are loo defective to saiisiy these data for
ludinncni uFcariecT-ielatiag efiects of ELF-eieciTomagnedc iaeids [40].
206 M. Nakagawa et al.

NOTE
(a) : Occupational
(b) : Public

Sweden
(VOT.1990)

Figure 6. Main recommendation on safety standards or guidelines of electromagnetic fields with frequency
of ULF (ultra low frequency) to microwave range. Some researchers in advanced industrial countries think
these values are rather high compared with those of real fields where positive results are reported on high risk
of leukemia or some kinds of tumors by epidemiological methods.
Linearmotorcar (EDS) Maglev 207

3.4. High Frequency Band

A dielectric heating effect of a high-frequency EMF has been known from early stage,
and in a corresponding plan for safety, ANSI (American Standard Research Institute) made
a recommendation on safety standard in 1966. The facilities with high-power output
exceeding this standard are not a few, and plastic sealers used in industry may be one of the
objects calling for careful watch at present [41]. A short wave diathermy and microwave
diathermy for clinical use are said to have surprisingly large straying electromagnetic field
around the cable [42]. Furthermore, in highly energized MRI, RF output for signal giving
and receiving is so large as to exceed the standard. In the case exceeding the safety standard
in a medical examination and treating equipment, with the same concept as for a static-mag-
netic field of MRI, "the medical treatment business" is managed with an idea to prefer one
of higher merit. On safety in heat generation an lEC draft mentioned before [35] is reviewed,
and equipment will be employed under proper watch of a medical doctor.

4. INTRODUCTION FROM OUR DATA

4.1. Case of Bacterial Mutation Assay (Ames Test)

This test is one of the most commonly used assays in genetic toxicology to detect
mutagenicity and carcinogenicity of chemicals newly produced, drugs and environmental
samples. We applied this method in estimating the ability to detect genotoxicity for some
kinds of electromagnetic fields. Until now, some studies reported about weak time-varying
magnetic field using bacterial mutation assay at 20nT of 1 OOHz (Juutilainen and Liimatainen
[43]) and 20mT of 0.3Hz (Moore [44]). The results suggest that those magnetic fields
revealed no mutagenic activity. Our report shows the result about mutagenicity test used
50Hz-timevarying and I.3T-static magnetic fields.
The tests were carried out with the plate incorporation method using preincubation
modification described by Maron and Ames [45]. We used two types of magnetic field
exposure systems. One is for a static magnetic field with ESR (electron spin resonance)
equipment i.e. JES-RE2X (JEOL). This magnet can generate a quite homogeneous static
field up to 1.3 T. The other system (made by IDX) is for a time-varying magnetic field, and
has a capability to generate the flux density of up to 35 mT and has a frequency of zero to
100 Hz. For the first stage experiment, we used a commercial 50 Hz frequency in east area
of Japan. Scale and both sections of generating coil for time-varying field are shown in Figure
7. Figure 8 indicates a specially made shaker used in the JES-RE2X, which can be shaken
in the field and the other side of the rod is also connected for the sham exposure incubation.
For the preincubation method the shaker maintains the temperature of bacterial suspension
tubes both in exposure and sham space at 37+0.1 . These system is very unique and ensure
the high quality condition of these bacterial mutation assay.
Strains used for the test were Salmonella typhimurium TA98 and TA100. The cultures
of test strains were incubated overnight at 37°C until the cell concentration ranged 1-3x10''
viable cells per ml. Cultured cell suspension (0.1 ml) is added to 0.1 M-phosphate buffer
(0.5 ml) in sterilized tubes. Half of the tubes were incubated shakenly at 37°C in a 1.3 T-static
or 50-Hz-various density of time-varying magnetic field for 20 min and the others in a sham
space for the control. After that 2 ml of molten top agar was added to each of the tube and
the contents were immediately poured onto Vogel-Bonner E minimal agar plates. In the static
magnetic field test, experiments and also carried out for 100 min exposure to magnetic field
by the same procedure. The plates were incubated upside down at 37°C for 48 h, and the
numbers of His* revertant colonies were counted. Table 2 shows the result of the test in 1.3
208 M. Kakagawa el al.

Figure 7. Generating coil for time-varjing magnetic field,

T-static magnetic field. Table 3 indicates tlie result of the test at various flux den^sities of
time-varying magnetic field. In botli data, no significant difference between exposed and
control groups was observed, Tiiese results suggest that 1.3 T-static and up to 35mT-50Hz
time-varying magnetic fields have no mutagenic activity for these bacteria. However,
Shimizu et al, [46] reported a high density static magnetic field (up to 11 T) which affected
mutagenicity of some chemical mutagens and its effect depended on density of magnetic
flux m the Ames test with modifiedfnot shaken) preincubation method. Now we made a new
5 T-supcrcondiicting magnet. This magnet generates homogenious static magnetic field in
20cm-diameter and 20cm-long space, and is placed in the constant temperature room. We

Figure 8. Tlie slialcer (left) and temperature-conirolled water circulating system through the AC magnet.
Lincarmotorcar (EDS) Magkv 209

+1
o

+1 Q

<N
00
<N
s
-a
+1 +1 +1 +1
O X! s
^
\c r j §
oc v-,
~ <

+1 +1
oo ^
— (N
+ 1 +1 1

3 - s r
ft 3i
a i
i 1
1 1 1
3
:3
3
k.
i
T) i/i

; *
^ 1
0

s 6p ,
U c«
< .i. ^ so
8
:
sr
«a oa
a.
^ E5 fS
210 M. Nakagawa ct al.

o d — ^ i n v S o o o d o ^

0
+1 +1 -l-l +1 +1 +1 +1 +1
in t^ ir-, -^ t^ Tt -^ v^^
a.
X
i> \o sc v^ v^ r- n~- ^
-a
c
'"
1—
8
c
c < "O
u H ot/1
5Q i
— — ! —
3
.s
+1 +1 +1 +1 +1 +i +1 +1 +! 1
30
o ^o sc r"^ (N t*-' <N ri <5
c SL

o tN r- —1 q q
W^ r"' r-^ r-^ f^'
+1 + 1 +1 +1 +1
>£l \C u-i
+1
>n r^,
*

ri r-i <N r^i

I
q ^
^
00 (N
— + 1 + 1 +1
+1 +1 +! +1 +1 —'
^
' O
\o Tf
*
^ I
r*', -^ rr -^ —
0^ in
(N f^, (N rJ IN
1
c> u
u
t/1 g
u E
s 'S
o
c
-3
s
0
1
S 1
g

Si
X. 5
«
t3
o
i. si
A
f!
1
^
Linearmotorcar (EDS) Maglev 211

can perform both of plate incorporation and preincubation method in it. We have started a
test for detection of mutagenicity under combination of high static magnetic field with
chemical mutagens.

4.2. Detection Of Mutagenic Activity Using Fruit Fly (Somatic Cell Test)
4.2.1. Method and Result. To estimate genetic effects of static magnetic field we
applied the mutant fruit fly Drosophila melanogaster that lacked repair function of damage
to their cellular DNA (deoxyribonucleic acid). The mutant was sc z' w^^^ mei-9" mei-41 ""^
that was defective in both excision repair and post-replication repair. Young larvae of mutant
and normal genotypes were exposed to 0.6 T magnetic field for 24 h, and then allowed to
continue development in normal culture condition until they molted and finally emerged
from pupa cases. After their eclosion, the number of surviving adults was counted. Table 4
shows the adults of mutant genotype (male) which decreased by about 8 %, while normal
siblings (female) remained unchanged. When FM6 with normal repair function was used
instead of the mutant, its survival rate was not affected by magnetic field. It is inferred that
exposure to static magnetic field caused damage to larval cellular DNA, and that somatic
cells without normal DNA repair function failed to continue cell division, which resulted in
developmental lethality of mutant larvae. DNA damages occurring in normal larvae must
have been repaired, so that their survival rate was not altered. Similar effect was observed
when mutant larvae were subjected to ultraviolet (UV) light of conventional 15 W germicidal
lamp at 50 cm distance for 5 s (12.5 Jm'^). We estimated that mutagenic activity of 0.6 T
static magnetic field was the same as that of 0.14 mJm"' UV lights.

4.2.2. Discussion

(1) Detection and Estimation ofDna Damaging Effect. It was found that repair-de-
fective, mutagen-sensitive larvae were sensitive to a 0.6T static magnetic field. When first
and second instar larvae were exposed for 24h, about 8% of mutagen-sensitive males failed
to grow up to adult, while female siblings with normal repair function eclosed normally. As
a result, the male/female ratio showed a statistically significant decrease compared with the
non-exposed control (Table 2). It is inferred that static magnetic field caused, directly or
indirectly, damages to cellular DNA of the exposed larvae, which were repaired in female
somatic cells, but could not be repaired in male cells. As DNA molecules with damages could
not be duplicated normally, male somatic cells with damaged DNA may not divide, which
finally resulting in developmental lethality of mutagen-sensitive male larvae.

Table 4. Ratio of survivors of mutant and normal genotypes

Genotype sc z! w*" met-.go mei-41'" FM6

Condition 0.6 T for 24 h unexposed 0.6 T for 24 h unexposed
Mutant (male) 7,938 8,514 13,000 7,920
Normal (female) 7,135 7,038 13.085 7,936
M/F 1.12 1.22 1.01 1.01
Odds ratio (95 % CI) 0.88-0.96 - 0.96-1.04 -
212 M. Nakagawa et al.

The apparent genotoxicity of static magnetic field was compared with that of UV
light. When second instar larvae were irradiated with various doses of UV light, the viability
of the females with normal repair function was more than 90% of untreated control at 180s
irradiation (450 Jm"-). Mutagen-sensitive males were, on the other hand, sensitive to even
10s irradiation. The number of male survivors decreased exponentially as the irradiation
period was lengthened. Exposure of the mutagen-sensitive larvae to 0.6T magnetic field for
24h and to 12.5J m'- UV caused the same decrease of male survivors. Assuming a linear
relationship between the UV effect and exposure dose, genotoxicity of a 0.6T static magnetic
field was estimated to be the same as that of 0.14mJ m ' s' UV light.
It can be argued that exposure to magnetic field gives general physiological effects
on larvae rather than effects of DNA damage. If female larvae were more resistant to such
stress than males, the exposure to magnetic field should decrease male/female ratio. This is
not probable, if not impossible, because FM6 males with normal repair function were not
killed by the exposure to magnetic field. The relative viability of FM6 males to y f:= females
is similar to, but a little bit smaller than that of sc z' w^^*^ mei-9^ mei-41'^'' males, the former
are expected to have similar or higher sensitivity to general physiological effects than the
latter. Therefore, the resistance of FM6 males to a 0.6T magnetic field strongly suggests that
repair-defective males were actually killed by induced damages in cellular DNA. rather than
due to general physiological effect of the exposure to which males were more sensitive than
females.

f2j Possible Mechanisms of DNA Damaging. However, it is not probable that a 0.6T
static magnetic field directly attacked DNA molecules forming pyrimidine dimers or other
DN.A damages. Dissociation energy of 4 to 5 eV is necessary to disrupt chemical bonds and
form a pyrimidine dimer but the interaction of an electron spin with a 0.6T magnetic field
creates electromagnetic energy of 10 ' of it. Other kinds of DNA damage need 0.1 to 10 eV
and they are impossible to occur too. It is possible, on the other hand, that exposure to a 5T
magnetic field suppresses thephotoreactivation process. In somatic cells of sc z' w*"' mei-9''
mei-4l"^ males, excision repair and post-replication repair are defective while photoreacti-
vation, another DNA repair function, is still active. It is well known that some enzyme
reactions are affected by static magnetic field exposure[47], Drosophila photolyase can be
another example of magnetic field sensitive enzyme.
Another possible hypothesis is that the magnetic field interacts with mutagenic free
radicals which appeared occasionally in larval somatic cells. It is well known that static
magnetic field affects singlet-triplet transition in the unpaired electron. This effect is
especially prominent in a reaction in micelle. Tanimoto et al. [48] showed that the life time
of 2-naphthylphenylmethyl radicals in polyoxyethylene(44)dodecanol (Brij 35) micelles
was elongated by several times in a 0.6T static magnetic tleld. As micelle can be considered
a simple model of cell, it is possible that the life time of radicals in living cells is also affected
by static magnetic field. If its life is elongated, the same amount of radicals may have more
mutagenic activity.
Still another possibility is that the membrane permeability of cells is affected by
magnetic field, which results in influx of mutagenic agents which exist in the haemolymph
at the background level. Permeability change by magnetic field exposure has been reported
in cultured cells [49] and in liposome vesicles [50]. Although experiment was done at a
temperature near the phase transition region of lipid bilayer, this hypothesis is not absurd at
physiological temperature. This hypothesis predicts that the dose-response relationship
between flux density and genotoxicity is not linear. Deformation of lipid bilayer or channel
protein causes influx of mutagens while the flux density is not too high. When the flux density
exceeds a certain level, an over-deformation may occur which prevents influx of chemicals.
This is compatible with the "window" theory of the magnetic field effect. This should be
Linearmotorcar (EDS) Maglcv 21.1

confirmed by an experiment at a higher flux density using a superconducting magnet. For

the time being, we can only say that static magnetic field has. in a practical sense, a weak
genotoxicity.

(3) Sensitivity and Limitation of Dna Repair Test. The average dose of genotoxic
UV component in the sunlight is 10 to 20 kJ m"^ in Japan. As an exposure to a 0.6T magnetic
field for 24h and to 12.5J m"" UV caused the same decrease of male survivors, the
genotoxicity of a 0.6T static magnetic field is considered to be 1/2 to 14 of that of sunlight.
It is surprising that such weak genotoxicity could be detected by an individual animal test.
In normal cells, almost all DNA damages induced by a weak mutagen are repaired by DN.A
repair functions. Occasional error in the post-replication repair process causes mutations
which could be detected by mutagenic activity tests such as the sex-linked recessive lethal
test. In our experimental system, on the other hand, we detected DNA damage which was
much more frequent than occasional error in the repair process. The mutant used in our
experiment was defective in both excision repair and post-replication repair. In Escherichia
coli. a single damage in the whole genome was sufficient to kill a mutant cell that lacked
both of excision repair and recombination repair (bacterial version of the post-replication
repair, see ref [51]). Developmental lethality of individuals occurs at a high frequency if
DNA damages are induced in even a few percent of larval or embryonic cells. This
"amplification" makes the sensitivity of our test system much higher than in vitro tests, liven
if repair-deficient Xeroderma pigmentosum cells were used, it is hardly possible to detect a
mutagenic activity that causes DNA damages in a few percent of the cultured cells.
However, it is not clear whether this genotoxicity is equal to mutagenic activity.
Mutations arc created by occasional errors in the DNA repair process. Our test animals
carried mei-41'^'' mutation and therefore were defective in the error-prone post-replication
repair, so that mutation could not occur. The DNA damages caused by UV light are mainly
pyrimidine dimers which are known to be repaired, in the cells with normal repair functions,
by excision repair and/or post-replication repair process. Errors are expected to occur at a
frequency proportional to the number of damages and therefore the genotoxicity determined
by the DNA repair test using a repair defective strain can be considered equal to, or at least
proportional to the mutagenic activity of UV light. Mutagenic activity of chemicals whose
DNA damaging mechanisms are known is also determined by this test. The genotoxicity of
magnetic field determined by our test system will be also considered a mutagenic activity,
when its DNA damaging mechanism will have becoine clear.

5. STRUCTURE OF SUPERCONDUCTING MAGLEV TRAIN AND

MAGNETIC FIELD SHIELDING

5.1. Introduction

Powell and Danby bore the original conception of a magnetically suspended train in
1966 [52]. In Japan study on conception design of EDS (electro-dynamic suspension)-
maglev began from 1962, and it led to construction of the preliminary test track in Miyazaki
in 1975. Vehicle MLUOOl with some research staff on board recorded a speed of 400 km h
in 1987. In June 1990 the Ministry of Transport Japan declared the construction of an
experimental track for maglev in Yamanashi adjoining Tokyo authorized, which will go into
practical operation in future. Test train will consist of 3 and 5 vehicles and have ability of
developing a maximum speed of 550 kra'h.
214 M. Nakagawa et al.

Here, we introduce a first trainsct to be used for the ranning test in Yamaaashi aiming
at business opcraiion of the next phase. Wc state position and size of the SCM isupercon-
duciirig magnet), related posirioii of a propulsion coil, and shielding method to control a
leakage iiiagnedc ilux inside of the car and near tiis door where passengers get on and off.

5.2. Principle ofMagHetic Propulsion and Le¥itatioii

Linearmotor may be equated to a stator of the conventional motor being spread out
on the level, and the vehicle ivith magnet seremg as a rotor. To propel liDeamiotor smoothly,
e frequency of 3-phase allernaiing current has to match the polarity of the magnets on
vehicle, I .inearmotor N-pole muss always be located at the same position of the vehicle where
N-pole attracts the S-pole of the vehicle, and at the same time linearmotor S~pole aller the
N^poIe repels the S-pole of the vehicle, fhis relation holds for the linearmotor S-polc, too.
Because supercondiiciixx- magnets at^e arranged with speeifie pitch and the polarities of the
magnets alternate, some ripples (iime-varyirig magnetic fields) remain in the vehicle. In order
to niai<e these ripples as smai! as pos.siblc, creativity vvas displayed in various ways. In
Yamanashi test track,, "a formation wdh 240-degrces double layer placement of alienated
poles"' will be adopted for cod arrangement and wiring of propulsive linearmotor. This
arrangement of coii-positions is shown m Fig, 9 As shown ai right side, jt is a complicated
structure with levitaiing and guidance coils aaiTanged on the front side of the surt'ace where
propulsive coils are set i,ti double layer A mechanical meaning of this ripple decremciit plan
is to reduce vibration of SCM, to improve stability and prevent a quenching. The feature of
"guidance" is that combmed coHs and wning are made to cooperate to bring the vehicle from
exiremely right or left side track to the center of gaideway by virtue of electromagnetic force.
The principle of leviiaiion is rather simple. With propulsion of the vehicle, magnetic
flux density of SCM changes relative to cross section of the levitating and gisiding coils
arranged along the side-beam of the guideway. so that induced currents in the ccnl pmduce
new magnetic fiu.x and cause the vehicle to float. Fig. iO shows the relation between SCM
on the vehicle and the propulsion coil and levitation-guidance eoils along the gtjideway. An
arrangement of side wall leviiation in the left is adopted in Yamanashi test track. While the
speed IS low, the vehicle runs on wheels. As die speed reaches 1.50 kni/'h approximately,
however, the vehicle can levitate just by a repulsion of magnet fluxes with induced currents
in the surface coils.

5.3. a Tminset Using Yamanashi Test Track. The first trainset will be composed of
three vehicles as shown in Fig. i 1, their production to be completed in 1995. There are 3
cars that is a Kofu-side leading car, a standard middle coach, and a Tokyo-side leading car,
and every vehicle is combined with connecting bogies. Big feature with respect to outside

Ground coll for suspension and guidance

Ground coil for prepuIsienX^ .,-''''<5'
Side wall \ V ' - ' S - ^ i - ' f \y

Figure 9, Arrangement of positions and wiring

of propulsive linearmotor used in Yamanaslii lest
rnick, named "formation with 240-dcgrecs double
layer placement of alienated poles".
Linearmiitorcar (EDS) Maglev 215

Suspension coil
(side wall
levilatioii'
Cryoslal

Supereonducting

Hgure 10. A relation with SCM on the

vehicle and propulsion and Icvitation
coils along ihe guideway. An arrange- Propulsion1 col
coil ;
ment of side wall levilation in Ihe left is
Suspension coil
adopted in Yamanashi test track. (Facing lemtatlon)

appearance is a forefront shape. The Kofu-side leading car "double-cusp" and the Tokyo-side
"aerowedge" . Performance of these two forefront shapes will be tested by real riinoing. One
aerodynamic brake is provided above everj' bogie. Fig. 12 is an. imaginative figure drawn
with computer graphics seeing from the Tokyo-side leading car. With no driver aboard, there
is no window in front of the leading car.
In the machine chamber of Tokyo-side forefront, batteries for main power supply,
gas turbine generator, DC/AC converter, and a lielium buffer tank are mounted. Middle coach
has a passenger room at the center and an end body .structure above the bogie. The connecting
division is provided for a person to pass through the center, and the exterior of passage makes
the machine chamber. In the machine chamber at Tokyo-side bogie, we load an aerodynamic
brake equipment, helium buffer tank, and inverters for machine power supply. In the
ICofu-side machine chamber, we load a helium buffer tank, ventilator, and air conditioning
outdoor machines. Large fairing for both sides of the connecting portion is done to rectify
the air flow aerodynamically.
About the magnetic shield inside of the car, less than 1 mT in passenger rooms, and
less than 20 mT on the passage of the connecting division, are considered design features.
Fig. 13 shows the arrangement of shielding materials with pure iron of industrial use at
connecting portion. We confirmed the real shield effect by a mock-up of a full size by
comparing with calculated values and investigating whether they were controlled at marked
levels of a magnetic shield. An equipment shown in Fig. 14 is the conception figure of a
boarding bridge to uoanable passengers to approach stray magnetic fields from outside of
the vehicle. As the maglev has SCMs at both exterior sides of connecting portion, it allows
passengers to avoid from the outside of the bogies. Two doors of the equipment are designed

2^600

srrpT
i£53iiSiiES2..., dtzii.l tZufczJL

Kofu side Tokyo side

Oouble-cusp type - SCM r Aerodynamic braise Aerowedge type

./'"^^ '"*; iOOaOS.^SOQC.^15 i'i!^^! l^'^^ 000030000000 I 1.0 00 0 D 0 O n

Figure It. The fmi trainscf is composed of three vehietes now in production for completion in 1995.
216 M. iN'akagawa ft al.

Figure 12. \ constnicted 3<ar tramset for "s'aniiinashi 'est track seeing frijiri ihe uikyo-side leading car.
aerowedge-shaped.

to open only to guide passengers to the door of vehicle when they get oo or off the train. The
Iloor of the equipnient consists of a partial bridge, which sirves as to shield people against
the magnetic field from under the floor. Moreover, end of doors of tlic equipment are
designed in the shape of a cross section closely adhering to tlie vehicle with an increased
shielding effect.

5.4. Theme of the Futere

Given today's safety concept of electromagnetic fields, we do not bcheve that many
people are of the same opinion necessarily. A consensus seems to be reached on the difference
in an exposure amount of EMF among general public, occupational workers, and on medical

Mock op

(Ferromagnetic materiaij ^;^^ _.^#='-' "''

Figure 13. An arrangement of shielding materials with pure iron for indu.strial use at connecting portion.
Lincarmotorcar (EDS) Magiev 217

Shield Wall

Boarding Form

Controller

Figure 14. A conception figure of boarding bridge not allowing passengers to approach stray magnetic fields
from outside of the vehicle.

treatment use. It is not fully agreed, however, that any level of magnetic field should be
regulated for the vehicles utilized by people for short time. Levels of a time-varying magnetic
field of electrified railroads and electric vehicles are quite large comparing with those of
power transmission lines. As for the evaluation of an exposure amount of electromagnetic
fields, it is not argued yet that an exposure amount might be multiplied by a simple magnetic
flux density (whether it is a total amount of magnetic flux or not that a whole-body receives)
and by time (or the square of it or a square root of it). Therefore, about the safety of electric
trains nobody can say anything definite. Either it is not solved about ELF frequency whether
more than a certain Hz should be considered controversial (e.g. whether we should think that
the effect of less than 3 Hz is as weak as a static-magnetic field in IEEE frequency
designation).
We can say about safety of EDS magiev that magnetic shield marks meet the values
in Table 1 of a static-magnetic field. About the ripples mentioned before, however, magnitude
and frequency of them may become an issue. Frequency, strength, and exposure time must
be reviewed for the safety of a time-varying magnetic field, and further arguments are going
to begin from now. In parallel to the argument, we are progressing in development of magiev
such as the reform of a running mechanism, countermeasures for magnetic resistivity,
stabilization of the SCM, and development of power supply method. In the same way as in
development of electric vehicles, we expect that scientific technology and safety aspects
keep abreast of magiev development.

REFERENCES
1. Hatch.T.F., 1972, Permissible levels of exposure to hazardous agents in industry. J. Occup. Med. 14:
134-137.
2. Bassett, C.A.L., Pawluk, R.J., Filla, A. A., 1974, Acceleration of fracture repair by electromagnetic fields:
A surgically non-invasive method. Ann. NY. Acad. Sci. 238:242-262.
3. Poison, M.J.R.. Barker, A.T, Freeston, I.L., 1982, Stimulationofnervetrunks with time varying magnetic
fields. Med. Biol. Eng. Comput. 20: 243-244.
218 M. Nakaga«a el al.

4. Barker, AT.. Jalinous. R.. Frecslim, I.l... I9S5, Non invasive magnetic slinuilationcirimman motor cortex.
Lancet : 1106-1107.
."i. Murayama, M,. I96.'i. Orientation ot'sickled erythrocytes in a magnetic field. Nature 206: 420-422.
6. Vlaret. G.. Schickfus. .VI.\'., Mayer. ,'\.. Dran.sfeld, K.. 1975. Orientation of nucleic acids in high magnetic
fields, Phys. Rev. Lett, .l,"!: .197-400.
7. V'amagishi, A., Takeuchi, T.. Higashi. T., Date, M.. 1990, .Magnetic fields effect on the polymerization of
fibrin fibers, Physica B 164: 222-228.
8. Shimi/u, H., Suzuki, Y., Shimizu, K., Fukumoto, M., Okonogi, H., Nakagavva, M., 199.i, Biological
effects of high magnetic field (2); change in lime-course of mice serum LDH, Jpn. J. Ind. Health .'55: S248.
9. Okuno. K.. Tuchiya, K.. Ano, T., Shoda, M., 1993, Effect of super high magnetic field on the growth of
Hschelichia coli under various medium compositions and temperatures, J. Ferm. Bioeng. 7.'i: l().i-l()6.
10. Koana, K., Ikehata. M., Nakagawa, M., (submitted). Estimation of genetic effects of static magnetic field
by somatic cell test using mutagen sensitive mutants of Drosophila melanogastcr. Bioelecirochein.
Bioenerg.
11. Sehenck, J.F., 1992, Health and physiological effects of human exposure to whole-dody four-tesla
magnetic fields during MRL Ann. NY. Acad. Sci. 649: 281101.
12. Schwartz. J.-L.. 1978. Influence of a constant magnetic field on nervous tissue: 1. Nerve conduction
velocity studies, lELL Trans. Biomed. Eng. BME-2S: 467-47-1.
1-1. Nath. R.. Schulz, R.J., and Bongiorni, P., 1980, Response of inainmalian cells irradiated with 10 M\'
.X-rays in the presence of a uniform 20-kG magnetic field. Int. J. Radiat. Biol. IS: 2X1092.
14. Mulay, 1.1.., and Mulay, L.N., 1964, Effect on Drosophila melanogastcr and S--17 tumor cells, postulates
for magnetic field interaction-s. in Biological Effects of Magnetic Fields (ed Barnolhy, M.I .1 146-1 169
(Plenum Press. New York &. London).
1-S. Barnothy, ,M.F., and Siimegi. 1.. 1969. Effects of the magnetic field on internal organs and the endocrine
system of mice, in Biological Effects of Magnetic Fields \ol.2 (ed Barnothy, M.F.) 10.1-126 (Plenum Press.
New York & London).
16. Kuz"initia. Z.F.. 1984, Corticosterone content in mouse adrenals under the prolongcil intensive infiiiencc
of a constant magnetic field, Kosm. Biol. Aviakosm. Med. 18:87-89.
17. Pautrizel, R., Prior. M.. Dallocio. M.. and Crokett. R.. 1972. .Xetion d'ondes electromagnetiques et dc
champs magnetiques sur les modifications lipidiques chez le Lapin par I'adminislration d"um regime
alimentaire hypercholesterole, C. R. Acad. Sci. Paris D. 274: 488-491.
18. Barnothy, M.F., 1964, h possible efTect of the magnetic field upon the genetic code, in Biological EtTecis
of Magnetic Fields (ed Barnothy. M.F) 80-89 (Plenum Press. New York & London).
19. Maret, G., and Drasl'eld. K., 1977. Macromolecules and membrane in high magnetic fields. Physica
86-88b:1077-108-l.
20. Tanimoto.Y.. Takashima. M.. and Itoh. M.. 1984. Magnetic field effect on the hydrogen abstraction
reaction of xanthone in sodium dodecyl sulfate micellar solution. J. Phys. Chem. 88: 60-'i-1-6056.
21. Selye. IE, 1946, The general adaptation syndrome and the diseases of adaptation, J. Glin. Endocrin. 6:
11 7-2.10.
22. Barnothy, J.M.. and Barnothy. M.F.. 1963. Secon-day minimum in the growth curve of mice subjected to
magnetic fields. Nature 200:189.
23. Laforge, H., Moisan, M.. Champagne, F.. and Seguin, M.. 1978, General adaptation syndrome and
magnetostatic field; effects on sleep and delayed reinforceinent of low rate. J. Psychol. 98:49-55.
24. Nakagawa. M., and Matsuda. Y, 1988, .A strong static-magnetic field alters operant responding by rats,
Bioelectromagnetics 9:25-37.
25. Nakagawa. M.. and Muroya. H., 1981, Static-magnetic field affected on emotional behavior and sense of
equilibrium of mice, in Abstracts of 54th Meeting of Japan Association of Industrial Health, 380-381.
26. Nakagawa, M.. Muroya, H., Matsuda,Y., and Tsukamoto, H., 1980, Effects of static inagnetic field on
some lipid and protein metabolic processes of rabbit. J. Transprt. Med. 34:376-384.
27. Nakagawa, M., 1980, Food consumption of mice in the static magnetic fields of moderate strength, .Ipn.
J. Ind. Health 22:280-281.
28. Nakagawa, M., and Hosoya, M., 1975, Effect of magnetic fields on the amounts of electrolytes in urine
of the iriice, Med. Biol. (Tokyo) 90: 307-311.
29. Barnothy, J.M., 1963, Growth-rate of mice in static magnetic fields, Nature 200:86-87.
30. Beischer, D.E., 1964, Survival of animals in magnetic field of 140,000 Oe. in Biological Effects of
Magnetic Fields (ed Barnothy,M.F) 201-208 (Plenum Press, New York).
31. Nakagawa, M.. 1990, Mammals' response and adaptation to static magnetic fields as a nonspecific
stressor, Phvsica B 164: 213-216.
Linearmotorcar (EDS) Maglcv 219

32. Goodman, R., Bassett. C.A.L.. Henderson, A.S., 1983, Pnlsing electromagnetic fields induced cellular
transcription. Science 220: 12S3-1285.
33. Ucno, S., Lovsund. P..Obery. P..'\.. 1986. Effects of alternating inagnetic fields and low-frcquenc\ electric
currents on human skin blood tlovv, Med. Biol. Eng. Comput. 24: 57-61.
34. International HIcctrotechnical Commission, 1993, Medical Electrical Equipment Part 2: Particular Re-
quirements for the Safety of Magnetic Resonance Equipment for Medical Diagnosis (Draft, 6D), lEC
601-2-XY.
3-^. Nakagawa, M., Takebayashi, H., 1990, Diaphragm pacing by pulsated magnetic trains, Jpn. .1. Med.
Electron. Biol. Eng. 28: 234-239.
36. Mouchawar, G.A., Bourland, J.D.. Nyenhuis. J.A., et al., 1992, Closed-chest cardiac stimulation with a
pulsed magnetic field. Med. Biol. Eng. Comput. 30: 162-168.
37. Dodinot, B., Godenir, J.-P., Costa. A.B., 1993, Electronic article suveillance: A possible danger for
pacemaker patients. Pace 16: 46-.53.
38. Broadband Magnetic Fields: Their Possible Role in EMF-Associatcd Bioeffecls. DOT ERA ORD-93 29,
DOT-VNTSC-ERA-93-17. final Report. August 1993.
39. Murphy. ,I.C.. Kaden. D..-\.. Warren, J., Sivak. A.. 1993, Power frequency electric and magnetic fields: .\
review of genetic toxicology. Mut. Res. 296: 221-240.
40. Study Committee for EtTects of Electromagnetic Fields in Department of Resource and Energy, 1993,
Review Report on the Effects of Electromagnetic Fields.
41. Kolmodin-Hedman, B.. Mild. K.H., Hagberg. M., etal.. 1988, Health problems among operators of plastic
welding machines and exposure to radiofrequency electromagnetic tields. Int. .'\rcli. Occup. En\iron.
Health 60: 243-247.
42. Stuchly, M.A.. Repacholi, M.H., Lecuyer, D.W., Mann, R.D.. 1982, I:xposurc to the operator and patient
during short wa\e diathermy treatments. Health Phys. 42: 341-366.
43. Maron, D.M., ./Vmes, B.N.. 1983, Revised methods for the Salmonella mutagenicity test, Mut. Rcs. 113:
173-21.'^.
44. Juutilaincn, ,1., Liimatainen. A., 1986, Mutation frequency in Salmonella exposed to weak 100-117
magnetic fields. Hercdit. 104: 145-147.
45. Moore. R.L.. 1979, B.N.. 1983. Biological efTects of magnetic fields: studies with microorganism. Can.
Microbiol. 25: 1145-1151.
46. Shimizu, H., Suzuki. Y.. Vlasizu, N.. Hayashi, K.. Hashida. C . .Ando. IT. 1989. EtTects of static magnetic
field on mutagenicity o^ Salmonella I., in Abstracts of 62th Meeting of Japan .Association of Industrial
Health. 243.
47. Nossol. B.. Buse. G.. Silny, .1., 1993, Influence of weak static and 50-Hz magnetic fields on the redox
activity of cytochrome-C oxidase, Bioelectromagnetics 14: 361-372.
48. Tanimoto, Y., Jinda. C . Fujiwara, Y. Itoh, M., Hirai, K., Tomioka, H., Nakagaki, R.. Nagakura. S.. 1989.
Lasar flash photolysis studies of the magnetic field effects on the hydrogen abstraction reaction of
2-naphthylphenylcarbene in micellar solution, J. Photochem. Photobiol.. k. 47: 269-276.
49. .Aoki. H.. Yamazaki. H.. Yoshino, T., Akagi, T., 1990, EtTects of static magnetic fields on membrane
permeability of cultured cell line. Res. Comm. Chein. Pathol. Pharmacol. 69: 103-106.
50. Liburdy. R. P., Tenforde. T.S., Magin, R. L., 1986, Magnetic field-induced drug permeability in liposoine
vesicles. Radiation Res. 108: 102-111.
51. Howard-Flanders, P., Boyce, R. P., 1966, DNA repair and genetic recombination: studies on mutants of
Escherichia coli defective in these processes, Radiation Res. Suppl. 6: 156-184.
52. Powell. J. R., Danby, G. R., 1966, High-speed transport by magnetically suspended trains. .ASME
Publication (preprint) 66-VVA,RR-5:1-11.
 16

RECENT BIOLOGICAL STUDIES RELEVANT

TO CARCINOGENESIS

N, A. Cridland, Z. J. Sienkiewicz, C. I. Kowalczuk, and R. D. Saunders

NRPB
Chilton, Didcot
0X11 ORQ, United Kingdom

INTRODUCTION
Technological advance and changing social behaviour have been responsible for
driving the massive increase in the production and consumption of electricity during the 20th
century. One inevitable consequence of this process has been increased environmental
exposure to electromagnetic fields both at work and in the home. While there have been
many undoubted benefits from the widespread use of electricity, there has been growing
concern, especially during the last thirty years, that prolonged exposure to electromagnetic
fields at even low levels could have detrimental consequences on human health, and in
particular could be associated with some cancers. This possibility has aroused persistent
interest and concern, and continues to be investigated by a substantial number of laboratories
throughout the world. Much of this work has been comprehensively reviewed elsewhere
(Sienkiewicz, e/a/, 1991: Saunders, e? a/, 1991:NRPB, 1992; Cridland. 1993: WHO. 1993).
and it is clear that many of the older observations were essentially phenomenological in
nature. Within the last few years, however, a more coherent research strategy based on
powerful cellular and molecular approaches has started to emerge (Sienkiewicz. el al, 1993).

ELECTRIC AND MAGNETIC FIELDS UP TO 100 kHz

Initiation
The available evidence suggests that exposure to extremely low frequency (ELF)
fields is not genotoxic (Murphy et al, 1993; McCann el al. 1993) and is therefore unlikely
to be directly associated with initiation. Although a few studies have reported a higher
incidence of chromosome aberrations following exposure to pulsed (Garcia-Sagredo et al,
1990; Khalil and Qassem, 1991) or intermittent (Nordenson et al, 1994) magnetic fields,
most recent studies have failed to find any evidence for chromosome aberrations or DNA
damage after exposure to 50 Hz electric fields at up to 20 kV m ' or magnetic fields at up to
Biological Etfects of Magnetic and Electromagnetic Fields, Edited by S. Ueno
Plenum Press. New York. 1996 221
222 N. A. Cridland ct al.

5 mT (Garcia-Sagredo el al. 1990; Garcia-Sagredo and Monteagudo, 1991; Livingston ei al,

1991;Saalmane/a/. 1991;Scarfie/a/, 1991,1994;Fioraniefa/, 1992; Fairbairn and O'Neill,
1994; Nordenson el al. 1994). In agreement with previous studies using microbial systems,
Tabrah et al (1994) found that exposure to 0.2 mT. 60 Hz magnetic fields did not affect the
mutation frequency in the Ames test; azide-induced mutations, however, were increased by
exposure to the magnetic field. The frequency of dominant lethal mutations was not increased
in male mice exposed for 14 days to 50 Hz electric fields at 20 kV m ' (Kowalczuk and
Saunders, 1990).

Tumour Promotion
A number of studies have examined the possible effects of electromagnetic fields on
proliferative responses at the cellular level, as an indicator of tumour promotion. This
evidence is considered here at the various stages in the signalling pathways which control
cellular behaviour.

Early Signals. A number of cell signalling pathways produce transient increases in

the intracellular concentration of free Ca"*, initially by stimulating release from intracellular
stores, and subsequently by influx across the cell membrane from the extracellular fluid. A
number of studies have sought to investigate the possibility that electromagnetic fields act
to stimulate calcium ion movements, and thereby influence signalling pathways. Using rat
thymocytes incubated in the presence of , Walleczek and Liburdy (1990) reported that
a 22 mT, 60 Hz magnetic field increased the influx of Ca"* which resulted from stimulation
with the mitogen Concanavalin A (Con A). In an extension of this work, annular ring dishes
were used to expose the cells to different induced electric fields at the same magnetic fiux
density; the induced current depends on the radius and is therefore larger in the outer rings
of such dishes. These experiments indicated that the influx of calcium depended on the
induced electric field and demonstrated that the direct application to the culture of a similar
electric field produced the same effect (Liburdy, 1992). Furthermore, the fluorescent dye,
Fura-2 was used as a probe to make real-time measurements of intracellular free calcium
concentration. The data indicated that exposure to a 60 Hz electric field in the culture of
170 mV m"' increased the influx of calcium ions from outside the cell, but did nol affect
release from intracellular stores.
Exposure of Jurkat human lymphoblastic T-cells to a 50 Hz, 100 nT magnetic field
has been reported to elicit an increase in intracellular free Ca-^* concentration which was
similar in magnitude to that induced by stimulation with an anti-CD3 antibody (Lindstrom,
et al. 1993). Preliminary reports from other groups using fluorescent indicators to monitor
intracellular ion concentrations in Jurkat cells have indicated that the response to a 60 Hz.
2 mT magnetic field (inducing an electric field of 1.8 mV m ' in the medium), may be
dependent on the biological status of the cells (Walleczek el al. 1994). This observation is
in agreement with previous findings from the same authors on the Con A-induced calcium
response of rat thymic lymphocytes exposed to 3 Hz pulsed magnetic fields (Walleczek and
Budinger, 1992); significant effects on Ca-* influx were observed at peak flux densities of
6.5 and 28 mT, and appeared to be field strength dependent. Exposure of HL-60 human
leukaemia cells to complex fields generated in a magnetic resonance imaging unit has been
reported to induce a small increase in intracellular free Ca'* concentration, as monitored
using the fluorescent indicator indo-1, (Carson et al. 1990).
It has been reported that ^-^Ca** uptake was increased following exposure of a variety
of lymphocytic cells to combined static and time-varying magnetic fields 'tuned' to resonant
conditions (Lyle et al, 1991; Yost and Liburdy, 1992), although neither of these studies
examined the dependence on resonance, for example, by 'detuning' the fields. Furthermore,
Recent Biological Studies Relevant to Carcinogenesis 223

two other studies found that exposure of either human (Prasad et a/, 1991) or mouse (Coulton
and Barker, 1993) lymphocytes to resonant fields did not affect * uptake or intracellular
free Ca""" concentration respectively.

Effects on the Nucleus. Early reports of increased RNA synthesis from several chro-
mosomes in cultured Sciara copwphilia salivary glands exposed to a variety of ELF
magnetic fields (see Goodman and Henderson, 1991) has recently been extended (CJoodman
et al. 1992) to include an analysis of transcription from the right arm of chromosome 3 in a
similar preparation from Drosophila melanogaster. Exposure to a 72 Hz pulsed magnetic
field at a flux density of 3.5 mT has been reported to induce a short-term increase in the
synthesis of both total and messenger RNA (mRNA) in CCRF-CEM lymphoblastoid cells
(Phillips and McChesney, 1991). Exposure of HL-60 cells to a 1 mT, 60 Hz sinusoidal
magnetic field was reported to produce a temporally similar, though quantitatively smaller,
increase in total RNA synthesis which was dependent on the magnitude of the induced
electric field (Greene el al. 1991).
The significance of the reported short-term increases in gross transcription is difficult
to assess. Clear and unequivocal evidence for effects on the transcription of specific genes,
particularly those known to be important for regulating cellular behaviour would be of far
greater importance. Unfortunately the evidence for this is much less convincing. In particu-
lar, the absence of internal loading controls renders the data extremely difficult to interpret.
Furthermore, the published studies have mostly been based on dot-blots rather than more
analytical assays such as Northern blots or RNAase protection assays, and examples of raw
data have rarely been shown. Even where the evidence for magnetic field effects on gene
expression appears to be reliable, the magnitude of the response tends to be small and should
be viewed in the context of the response to other agents. For example, in cultured fibroblasts,
c-myc expression can be induced about 20-fold by serum (Campisi ci aL 1984). 40-fold by
platelet-derived growth factor, 15-fold by fibroblast growth factor and 10-fold by the
chemical tumour promoter 12-0-tetradecanoylphorbol-13-acetate (TPA) (Kelley et al.
1983). Similarly, treatment of B-lymphocytes with the mitogen lipopolysaccharide results
in a 20-fold induction oi c-myc expression, whilst treatment of T-lymphocytes with Con A
elicits a 10-fold induction (Kelley et al, 1983). For the proto-oncogenes c-fos (Shah el at.
1993) and c-/w« (Stein <??o/, 1992), inductions of up to 200-fold have been reported following
treatinent with serum and ultraviolet radiation respectively. Thus the biological significance
of much smaller inductions by EMF must be questionable.
It has been reported that exposure of HL-60 human lymphocytic cells to either pulsed
(1.5 - 72 Hz, 0.38 - 19 mT peak) or sinusoidal (5 - 150 Hz, 0.57 - 570 |uT rms) magnetic
fields increased accumulation of transcripts from the P-actin, histone H2B, c-myc. c-src. and
P-tubulin genes (Goodman e/cf/1989, 1992,1994; Wei e/a/, 1990; Goodman and Henderson,
1991; Gold et al. 1994); one gene, a-globin, did not respond to magnetic tlelds. but
expression of this gene is normally cell-type specific anyway. Exposure to a 5.7 nT rms,
60 Hz sinusoidal magnetic field has also been reported to induce expression of the heat shock
protein gene, hsp70 in HL-60 cells (Goodman et al, 1994), and large T antigen in SV40-
transformed human fibroblasts (Gold et al, 1994). In general these data suggest the existence
of frequency, time and field strength windows for magnetic field effects on transcription.
However, excessive reliance on the use of dot-blots, and the absence of internal loading
controls renders interpretation of these data difficult, if not impossible. Moreover, the results
for histone are puzzling since the expression of histones is normally tightly regulated within
the cell cycle, whilst the reported increase in accumulation of mRNA from the proto-onco-
gene c-myc may have little significance since the c-myc gene in these cells is abnormally
regulated following a gene amplification event.
224 N. A. Cridland et al.

There is some support for magnetic field effects on the expression of c-w vc, and other
regulatory genes, from work in other laboratories. For example, exposure of CEM-CM3 T
lymphoblastoid cells to a 100 \xV. 60 Hz magnetic field for up to 2 hours resulted in transient
changes in the expression of the c-myc, c-fos, c-jun, and protein kinase C (PKC) genes
(Phillips el al, 1992). In this study the principle approach was to assess the rate of
transcription directly using nuclear run-off assays rather than indirectly by estimating the
accumulation of transcripts, a parameter which may be influenced by other processes.
Slot-blot analysis of cytoplasmic RN A from the same cells indicated that accumulation gave
a reasonable estimate of transcription. In general exposure was reported to increase tran-
scription from the c-fos, c-myc, and PKC genes, whilst the effect on expression oic-jun was
variable and dependent on cell density. Time course data indicated that these responses were
all observed within an hour, whilst longer exposure was reported to inhibit expression of
PKC. It should be noted, however, that in the absence of appropriate internal loading controls
it is difficult to reliably assess such small responses (generally around 2-fold).
Effects on gene expression, if they exist, are unlikely to result from direct effects of
the applied field on transcription complexes, but would instead result from activation of an
appropriate signalling pathway. Data on early signalling events (see above) indicate that
there could be effects on calcium ion influx, and it is therefore logical to investigate possible
links between this and the activation of genes such as c-myc. Liburdy, et al (1993a) have
reported such an association in cultured rat thymocytes stimulated with a suboptimal
concentration of Con A and exposed to a 22 mT, 60 Hz magnetic field for 60 minutes.
Calcium ion influx and accumulation of c-mvc RNA were measured in the same cells, and
whilst both were increased by a combination of magnetic field exposure and suboptimal Con
A, neither agent on its own produced an effect. This study appears to have been better
conducted than most; c-myc transcripts were quantitated by Northern analysis, and glycer-
aldehyde-3-phosphate dehydrogenase was used as an internal loading control. However,
although this study provides evidence for an association between calcium ion influx and
c-myc expression in response to magnetic fields, it does not demonstrate that one leads to
the other. It is to be hoped that future extensions to this work will address this question,
possibly through the use of appropriate inhibitors and ionophores.
The activation of a gene in response to earlier signalling events occurs as a result of
changes in the interaction of protein transcription factors with regulatory elements in the
promoter region of the gene. There may be several such elements in the promoters of genes
which are responsive to a number of signalling pathways (activated by different stimuli),
and identification of the element conferring responsiveness to a particular stimulus may be
helpful in elucidating the pathway through which the signal is transduced. It has been
reported that a magnetic field responsive element resides between positions -353 and -1257
upstream of the c-myc PI initiation site (Lin, et al, 1994). The element was identified by
deletion analysis of the human c-myc promoter fused to a chloramphenicol acetyl transferase
(CAT) reporter gene, and mediated transcriptional activation in HeLa cells exposed to 60 Hz
magnetic fields at 8 yCY, but not 80 )uT. However, 900 base pairs is a relatively large fragment
of DNA, and there therefore remains considerable scope for defining the location of the
element more precisely. Moreover, it should be noted that the CAT activity appeared to be
extremely low for all the constructs, suggesting that they were all essentially unresponsive
under the conditions tested.
In contrast to those discussed above, other studies have failed to find consistent
effects on proto-oncogene expression. For example, Parker and Winters (1992) examined
transcription in a variety of human and mouse cell lines exposed to 0.1 mT, 60 Hz magnetic
fields and found no evidence for increased accumulation of mRNA from the proto-oncogenes
c-fos, c-myc, c-/a/ and c-ras; there was also no effect on transcription from hsp70 or, in the
case of cells infected with either mouse mammary tumour virus or murine sarcoma virus.
 Recent Biological Studies Relevant to Carcinogenesis 225

viral genes. It should be noted, however, that in addition to suffering many of the methodo-
logical flaws criticised above, the minimum exposure time would have been too long to
detect transient effects.
In another study, it was found that exposure of HL-60 cells to a 1 mT, 60 Hz field for
up to 90 minutes did not affect the accumulation of either c-niyc or p-actin transcripts: the
steady state level of the 28 S ribosomal RNA was also unaffected (Greene el al, 1993). Using
a unique assay involving pulse labelling of cellular RN.A in combination with a nuclease
protection step, it was also shown that the transcription rates for c-mvc and P-actin were
unaffected by exposure. There did, however, appear to be an effect on synthesis of the 45 S
precursor ribosomal RNA, although this appeared to be associated with a concomitant
reduction in the half-life of the 45 S fraction and its mature products the 18 S and 28 S RNAs.
A number of studies currently in progress which appear to have been carefully
designed and performed, have not yet detected any effect of exposure to magnetic fields.
Preliminary reports indicate that in one study accumulation ofc-myc RNA was not affected
by exposure of HL-60 cells to 60 Hz magnetic fields at 1 and 10 jiT (Lacy-Hulbert el al,
1994), whilst in another careful and extensive study, neither c-myc nor p-actin appeared to
be affected by exposure of either HL-60 or Daudi cells to 60 Hz fields at flux densities from
8 nT to 1 mT (Saffer and Thurston, 1994). Similarly, it has been reported that exposure to
50 Hz magnetic fields at cither 6 |uT or 2 mT did not affect the interleukin-3 induced
expression of c-/av, c-jiin orjun-B in FDCP-mix murine haemopoietic stem cells (Reipert ct
al, 1994).
There is some evidence from studies of calcium influx that magnetic field effects on
cells are the result of the induced electric field (see above). It is therefore pertinent to consider
whether electric fields applied directly to the cultures can affect gene expression. It has been
reported that exposure of HL-60 cells to electric fields of 0.3 - 3 V m ' induced expression
of the c-myc, histone H2B, and hsp70 genes; cells were exposed to frequencies between 15
and 120 Hz (Blank e/a/, 1992; Goodman era/, 1994). Under the same conditions, transcrip-
tion from the p2-microglobulin was unaffected. It should, however, be noted that the
magnitudes of the reported responses were modest and are therefore of questionable
significance.

Effects on Cell Proliferalion. Recent studies of DNA synthesis, a measure of cell

growth, have failed to establish effects in CCRF-CEM cells exposed to a 3.5 mT magnetic
field pulsed at 72 Hz (Phillips and McChesney, 1991), or HF-19 normal human fibroblasts
exposed to a 2 mT, 50 Hz field (Cridland et al, 1993). The proliferation of K562 human
myeloid leukaemia cells was similarly unaffected by exposure to 50 Hz electric fields of 20
kV m"' (in air) or magnetic fields of 200 jiT (Fiorani et al, 1992). In contrast, it has been
reported that a 14 Hz electric field of 10 ^iV m"' applied directly to the culture increased both
DNA synthesis and insulin-like growth factor-II (IGF-Il) production in TE-85 human
osteosarcoma cells (Fitzsimmons et al, 1992). Exposure to a 2 mT, 50 Hz magnetic field
induced a small but significant increase in several measures of proliferation, although this
was dependent on the field generating system (Schimmelpfeng and Dertinger. 1993).
Several studies have been performed under resonant conditions, with the fiux density
of the static field and the frequency of the time varying field "tuned" for Ca-* ion resonance.
Under these conditions, rabbit-ligament fibroblast proliferation was dependent on the
amplitude of the time-varying field, an effect which could be abolished by detuning the static
field (Ross, 1990). In another study, the proliferation of 3T3 mouse fibroblasts was increased
in a resonant field, whereas human Raji lymphoma cells were unaffected. The proliferation
of the human cells could, however, be inhibited by exposure to field conditions which were
tuned to K^ (Rochev et al. 1990). Exposure of FDCP-Mix murine haemopoietic stem cells
226 N. A. Cridland et al.

to a various field conditions, including fields tuned to Ca-"" did not affect proliferation or cell
viability (Reipert ef a/, 1992).
A recent study of ornithine decarboxylase (ODC) activity in L929 mouse fibroblasts
indicated that basal activity was elevated by 55-65 Hz magnetic fields at flux densities of
1-100 |.iT (Litovitz et a!. 1991). The process appeared to be dependent on the maintenance
of a coherent signal for at least 10 s.

In Vivo Stiulies. Tumour promotion has also been examined directly using animal
carcinogenesis models. Three studies have employed the well-established mouse skin
promotion model. McLean etal (\99\) initiated skin tumours in groups of 32 juvenile female
SENCAR mice by painting with 7,12-dimethylbenz(a)anthracene (DMBA). Exposure to
60 Hz magnetic fields at 2 mT for 6 hours a day over 21 weeks did not affect tumour
development.
Other studies have investigated whether magnetic field exposure can exert promoting
effects on the development of skin tumours in female NMRI (Rannug et al. 1993a) and
SENCAR (Rannug et al, 1994) mice. Tumours were initiated in 7-8 week old mice by
painting shaved dorsal skin with DMBA and groups of 30 (NMRI) or 40 (SENCAR) mice
were assigned to each treatment. Exposure to 50 Hz magnetic fields at fiux densities of 50
or 500 ).iT commenced one week after initiation and continued for 2 years. Fields were either
continuous or intermittent (15 s on/off SENCAR mice only) and mice were exposed for 19
hours a day (21 hours at weekends). Coinparison of exposed animals with control mice,
which had received only DMBA, did not reveal any significant differences in either the
number of tumour bearing animals or the total number of tumours per group thus demon-
strating that magnetic fields do not act as tumour promoters in this system; treatment with
the chemical tuinour promoter TPA as a positive control greatly enhanced the yield of
tumours. These findings were further confirmed by analysis of epithelial hyperplasia, a
marker for tumour promotion in mouse skin. In contrast to TPA, which produced a marked
hyperplastic response both on its own and in combination with DMBA. magnetic fields did
not have any effect.
Exposure to a 20 kHz. 15 \iT (peak to peak) pulsed magnetic field did not affect the
development of lymphoma in female CBA mice which had previously been exposed to
5.24 Gy of x-rays; magnetic field exposure commenced when the mice were 40 days old and
continued for the remainder of their lives (SvendenstSl and Holmberg. 1993). It should be
noted, however, that the high incidence of x-ray-induced lymphomas would preclude the
detection of magnetic field effects in this study. Exposure to the magnetic field alone did not
affect the spontaneous lymphoma incidence although it did significantly decrease survival;
the siirv ival data for exposed mice were unusually variable thus casting doubt on the validity
of the latter finding.
Three studies have investigated the possibility that magnetic fields might promote
the de\ elopment of mammary tumours initiated with DMBA in female Sprague-Dawiey rats.
In these studies exposure to magnetic fields started at the time of the first treatment with
DMBA and continued for 91 days, at which time the animals were sacrificed for his-
topathological examination. In one of these studies, a group of 99 rats exposed to a 100 |.iT,
50 Hz field exhibited a significantly higher tumour incidence than controls after 8 weeks of
treatment, a trend which continued for the remainder of the exposure (Loscher et al. 1993).
The number of tumours per tumour bearing rat was the same for both groups, although the
final tumour size was greater in exposed animals. In the other studies, however, groups of
18-36 rats which had been exposed to various magnetic fields did not exhibit significant
differences in tumour incidence when compared with controls; some measures were altered
in individual experimental runs, but these changes were not consistent across the study
(Mevissen et al, 1993; Loscher et al, 1994). The field conditions included a 15 mT static
Recent Biological Studies Relevant to Carcinogenesis 227

field, a homogeneous 50 Hz field of 30 inT, and a gradient 50 Hz field of 0.3 - 1 ).iT. It is

possible that small field-dependent effects would not have been delected due to the restricted
sample size used in the latter studies (Loscher and Mevissen, 1994).
Beniashvili et a/ (1991) reported that exposure of rats to a 50 Hz. 20 ).iT magnetic
field for 3 h per day enhanced the induction of mammary tumours initiated with ni-
trosomethyl urea (NMU). Exposure apparently enhanced the rate of tumour development in
addition to increasing both the number of rats with tumours and the number of tumours per
rat increased. A more complete description of the experimental protocol would, however, be
useful for a proper evaluation of this study.

Co-Promotion
The possibility that magnetic fields may act as tumour co-promoters by enhancing
the inhibitory effect of tumour promoters on gap junctional communication, thereby releas-
ing premalignant cells from the inhibitory effect of adjacent normal cells has been investi-
gated. Cain et al (1993a) employed an experimental model in which untransformed mouse
fibroblasts inhibit focus formation by transformed cells, an effect which can be relieved by
the addition of the chemical tumour promoter 12-0-tetradecanoylphorbol-13-acctatc (TPA)
to the culture medium. Exposure to a 100 |iT, 60 Hz field for four 1 hour periods every day
for 29 days significantly enhanced focus formation by sub-optimal concentrations of TPA.
although the effect was not apparent at concentrations of TPA which induced little or no
focus formation. It should be noted, however, that there was considerable interexpcrimental
variation. Moreover, preliminary results from the same authors (Cain el al. 1993b) indicated
that in an extension to this study 60 Hz magnetic fields of 1 - 200 (iT inhibited rather than
enhanced focus formation in a field strength-dependent manner.
Magnetic field co-promotion has also been investigated in a whole animal study
(Stuchly et al, 1992). Tumours were initiated in 6 week old female SENCAR mice by
painting DMBA onto shaved dorsal skin and promoted by application of a sub-optimal dose
of TPA at weekly intervals thereafter. The mice were split into two groups of 48. one of which
was exposed for 6 hours per day and 5 days per week to a 2 mT. 60 Hz magnetic field, whilst
the other was sham exposed in the same room. The results for the first 24 weeks of the
54 week study indicate that exposure produced a significant increase in the rate of tumour
development but not in the yield of tumours at the end of the study. However, preliininary
reports from the same authors have indicated that the co-promotion effect is not reproducible
and may have been confounded by variable exposure to light (Thansandote et al. 1993).
Earlier studies had failed to establish a direct tumour promoting effect of magnetic fields
under the same conditions (see above).
Magnetic field exposure does not appear to have any effect on the formation of
chemically-induced preneoplastic lesions in rat liver (Rannug et al. 1993b. 1993c). Groups
of 9 - 10 male Sprague-Dawley rats were partially hepatectomised and treated 24 hours later
with the chemical initiator diethylnitrosamine (DENA).
Rats were exposed to 50 Hz magnetic fields for 19 hours per day (21 hours at
weekends), starting one week after initiation and continuing for 12 weeks. After sacrifice of
the animals, preneoplastic lesions, characterised by expression of y-glutamyl transpeptidase
(GOT) and the placental form of glutathione S-transferase (GST), were detected by histo-
chemical staining of liver sections. There were no consistent differences in the number of
preneoplastic lesions identified in rats exposed to fiux densities between 0.5 and 500 |.iT
when compared with unexposed rats; treatment with the chemical tumour promoter pheno-
barbital increased the yield of preneoplastic lesions almost 20-fold (Rannug, et al. 1993b).
Similarly, exposure to magnetic fields at flux densities of 0.5 and 500 pT did not have any
consistent effect on tumour promotion by phenobarbital in this system (Rannug et al. 1993c).
228 N. A. Cridland et al.

Peptide Hormones
Another avenue whereby exposure to electromagnetic fields could influence tumour
promotion is via an effect mediated by circulating peptide hormones. The best developed
hypothesis relates to effects mediated by melatonin. This hormone is produced by the pineal
gland in a distinct circadian rhythm controlled by the photoperiod: synthesis and secretion
of this hormone is high at night and minimal during the day.

Melatonin. Stevens (1987) first suggested that chronic exposure to electric fields
may affect melatonin secretion and thereby increase the risk for breast cancer. This possibility
has aroused wide interest and attention and produced a large number of reviews, monographs
and books on this topic (for example, Gupta el al, 1988; Wilson el al, 1990a; Stevens el al,
1992, Moore-Ede <?f a/, 1992).
Melatonin is known to affect seasonal (reproductive) rhythms, although more re-
cently it has been implicated in a range of other physiological functions, and several possible
mechanisms have been proposed which suggest that melatonin may be able to influence
oncogenesis, and particularly breast cancer. The most convincing of these suggests that
decreased melatonin levels, which cause elevations in circulating estrogen and progesterone
and so increase cellular proliferation within the stem cell population of the breast, would
increase the risk of cancer of these cells (Cohen el al, 1978). It must also be remembered
that significant increases in estrogen concentration would also increase the proliferation of
any estrogen-responsive tumour cells. An alternative mechanism suggests that melatonin
directly suppresses the growth of (estrogen receptor-positive) tumours (Blask and Hill, 1986;
Hill and Blask, 1988). In addition, there is some evidence to suggest that melatonin acts as
a scavenger of free radicals (Tan e/o/, 1993a) and so protects DNA molecules from oxidative
damage (Tan et al, 1993b). In this case, reductions in melatonin could increase the likelihood
of the initiation of cancers, and possibly of their promotion. To be effective as a radical
scavenger, however, melatonin is required at a supraphysiological concentration of about 20
|iM. This is many orders of magnitude greater than normal serum melatonin concentrations
of 100 - 400 pM. For comparison, glutathione, an established radical scavenger which is
claimed to be less eflective than melatonin, is present in human skin at concentrations of
about 1 mM. For these reasons, it is possible that only melatonin in pharmacological
quantities would provide protection against free radicals. Finally, it is possible that melatonin
may have a modulatory effect on immune function (see Stevens et al, 1992) and decreases
in circulating melatonin could compromise immune surveillance and leave the animal more
vulnerable to carcinogenic attack.
All the proposed mechanisms require that exposure to electromagnetic fields cause
reductions in pineal or serum levels of melatonin. However the experimental evidence to
support such a statement is somewhat equivocal: some studies have reported an effect, but
others have failed to find consistent field-dependent changes in pineal function. This may
mean that either the effect is not very robust or it may only occur under highly specific
conditions which have yet to be fully described.
Studies using 60 Hz electric fields (reviewed in Wilson and Anderson, 1990) provided
the first indication of a field-dependent effect on pineal function. Suppression of the normal
nocturnal rise in the production of melatonin by the pineal gland was observed in adult rats
exposed for three weeks to electric fields in the range of 2-40 kV m ' : normal melatonin
rhythms returned within three days of cessation of the field. However, the results were
somewhat variable, and magnitude of the effects were independent of field intensity. Similar.
if more modest effects have been reported in rats exposed to electric fields from conception
until 23 days after birth. In contrast, a more recent study (Sasser el al, 1991) reported that
exposure to electric fields failed to modify nocturnal levels of melatonin in adult male and
Recent Biological Studies Relevant to Carcinogenesis 229

female rats. It is of interest to note that the exposure parameters were identical to those used
in the previous studies which found positive effects. Similarly, no change in pineal function
was found in ewe lambs following chronic exposure to a combined 60 Hz electric and
magnetic field (at 6 kV m"' and 4 |iT) associated with a 500 kV transmission line (Lee et aK
1993). It is possible that the original positive observations were the result of some artifact,
possibly associated with the methods used to assay melatonin levels (Reiter, 1993) and may
not be related to exposure to the electric field/?e/-.ve. Few attempts to replicate the original
findings using electric fields have been made in recent years.
Exposure to magnetic fields has also been reported to disrupt pineal function. A
number of studies have found that inversion of the geomagnetic field or exposure to pulsed
static magnetic fields modifies nocturnal melatonin metabolism in a variety of rodents (see
Reiter, 1993). In contrast, only a handful of studies have investigated the effects of power
frequency magnetic fields, and the results are somewhat variable. Chronic exposure to a
rotating 50 Hz magnetic field has been reported to depress both nocturnal and diurnal pineal
and serum melatonin levels in albino rats exposed at between I and 250 |.JT (spatial vector
rms) (Kato et al. 1993) and in pigmented rats exposed at 0.02 and 1 (,iT (Kato a al. 1994a).
The unusual exposure conditions appear to limit the relevance of these studies to humans,
and exposure to a (more usual) horizontal or vertical field at 0.02 and 1 pT was found to be
without effect in albino rats (Kato et al, 1994b). As part of an experiment to determine if
magnetic fields affect mammary carcinogenesis induced by application of DMBA, albino
rats were exposed to a 50 Hz field al 1 fiT for 8 - 9 weeks (Loscher el al. 1994). This treatment
caused a significant but rather conservative reduction in nighttime serum melatonin levels.
The effects of DMBA itself on pineal physiology were not described.
Occasionally large, but inconsistent effects have been reported in Djungarian ham-
sters (Yellon, 1994). An initial experiment found that the nighttime peak of pineal and serum
melatonin levels was both reduced and delayed following exposure to a 0.1 mT field for 15
minutes beginning 2 hours before dark, although replicate experiments failed to reproduce
these effects, and in one experiment, exposure had no effect on melatonin levels. The
differences in responsiveness were attributed to differences in age of subjects and other
seasonal factors. No other study has indicated that such a brief exposure before dark could
be disruptive to melatonin rhythms, and replication of this study by an independent labora-
tory appears warranted.
Very few studies appear to have used non-human primates; one preliminary study
(Rogers et al, 1991) reported inconsistent changes in serum melatonin levels in baboons
exposed to a combined electric and magnetic field (at 30 kV m"' and 0.1 mT). With few
laboratories possessing the necessary facilities to work with primates, it is important that
this study appears in the peer-reviewed literature.
It has not been established that exposure to electromagnetic fields is able to affect
the circadian production of melatonin from the pineal in humans, although several large scale
investigations are underway to help resolve this issue. The available data are very limited.
Wilson el al (1990b) reported a reduction in urinary levels of a stable metabolite of melatonin
in some volunteers following nighttime exposure to a very weak (0.4 - 0.7 ).iT) 60 Hz
magnetic field from an electric blanket. The results were highly variable, and were only
found in 7 out of 28 subjects, so the robustness of the effect and the implications for human
health are not at all clear. It is recognised that individuals exhibit wide differences in
sensitivity of melatonin levels to light at night, and it may be that individuals will also show
similar differences in sensitivity to electromagnetic fields, although further investigations
are required before any firm conclusions can be drawn.
Of more direct relevance to carcinogenesis, Liburdy el al (1993b) investigated
whether a 60 Hz magnetic field could influence the inhibitory effects of melatonin
on the growth of cultured breast cancer cells. Growth curves for estrogen receptor-
230 N. A. Cridland el al.

positive MCF-7 breast cancer cells were determined over 6 - 7 days in the presence
or absence of melatonin and magnetic fields. It was found that exposure to the magnetic
field did not affect the growth of cancer cells in the absence of melatonin. The presence
of melatonin at physiological concentrations (1 nM) produced a small but significant
inhibition of cell growth which was abolished by exposure to a 1.2 (JT field. However,
it should be noted that the melatonin-induced growth inhibition was much smaller in
magnitude than the variation in growth between subsequent passages of the cells.
Despite these limitations, this study is potentially important in that it describes a
direct effect of a magnetic field on the interaction of melatonin with cancer ceils,
although unlike most other studies in this area, the mechanism does not invoke an
effect on melatonin synthesis or secretion.

Opioids. It is possible to speculate that other polypeptides apart from melatonin

could be affected by exposure to electromagnetic fields and possibly lead to an increase in
carcinogenic risk. For example, there is good evidence to indicate that exposure to weak
magnetic fields can inhibit the functioning of endogenous opioid systems and the actions of
exogenous opiate agonists in mice (see Kavaliers and Ossenkopp, 1992). It is therefore of
interest to note that endogenous opioids, particularly the pentapepetide [Met'^]-enkephalin.
have been shown to inhibit the proliferation of murine neuroblastoma cell lines (Zagon and
McLaughlin, 1989a) most likely by depressing DNA synthesis and mitosis. Importantly,
blockade of the action of these opioids appears to accelerate tumorigenesis and increase cell
proliferation (Zagon and McLaughlin, 1989b). It is therefore plausible to consider that
exposure to magnetic fields could affect carcinogenic events via field-induced changes in
the endogenous opioid systems although this has yet to be demonstrated.

Tumour Progression
ELF fields could affect tumour progression via changes in immune function. How-
ever, exposure of animals to magnetic fields does not appear to result in any significant
inhibition of immune responsiveness (Putinas and Michaelson, 1990; Morris et al. 1990).
Exposure for 7 days to a 50 Hz magnetic field at 20 mT did not affect haematocrit values or
either total or differential white blood cell counts in mice (Lorimore et al, 1990). In addition,
assays of bone marrow stem cells and myelomonocytic progenitor cells failed to reveal
significant effects on cell population dynamics, although subtle effects could not be ruled
out. McLean el a/ (1991) examined the effects of exposing mice to a 2 iriT magnetic field
for 6 h'day after treatment with DMBA. No effects were observed on mononuclear cell
counts, and natural killer cell activities in spleen and blood, or on spleen size, and it was
suggested that magnetic fields did not affect the immune response. However, in animals also
treated with the tumour promoter TPA exposure to the field resulted in both a greater number
of enlarged spleens, and a greater number of mononuclear cells per spleen. Moreover,
although mononuclear blood cell count was not significantly increased as a group, it was
high in three out of ten aniinals. The authors interpreted their results as suggesting that
magnetic field exposure compromised the animals' immune surveillance thereby reducing
its effectiveness in combating tumour growth.
Possible effects on tumour progression have been examined directly. The incidence
of leukaemia in a strain of leukaemia-prone mice was unaffected by periodic exposure to a
6 mT magnetic field pulsed at either 12 or 460 Hz (Bellossi, 1991). The mice were exposed
to the magnetic field for 30 minutes twice a week, and leukaemia incidence was studied over
five generations of the mice.
Recent Biological Studies Relevant to Carcinogenesis 231

RADIOFREQUENCY FIELDS ABOVE 100 kHz (INCLUDING

MICROWAVES)

Initiation
A series of reports from one group in particular have indicated that exposure of either
Ch inese hamster fibroblasts (Garaj-Vrhovac €t ol^ 1990; 1991) or human lymphocytes
(Garaj-Vrhovac et al, 1992) to 7.7 GHz radiation at 0.5 - 30 mW cm"- for up to 60 minutes
produced a statistically significant increase in the number of chromosome aberrations. It
should, however, be noted that temperature was not actively controlled, and small tempera-
ture rises at the surface of the samples were reported. It would appear likely that temperatures
within the cultures may have been elevated. Long-term irradiation of mice with 2.45 GHz
at an SAR of about 1.2 W kg"' was reported to induce gross DNA reaiTangcmcnts (Sarkar a
aL 1994). In contrast to these results, exposure of Chinese hamster ovary cells to 2.45 GHz
radiation at an SAR of 33.8 W kg"' for 2 hours did not produce a nonthermal increase in the
number of chromosome aberrations, even in the presence of genotoxic chemicals (Kerbacher
c/a/. 1990).

Promotion

Membrane Effects. Ion fluxes through the membrane constitute important signalling
mechanisms and rely on ion gradients built up by the action of transmembrane ion pumps.
A number of reports have suggested that radiofrequency (RF) radiation may be capable of
affecting the activity of such pumps. It has been reported, for example, that the human
erythrocyte Na'^,K*-ATPasc can be induced to pump Na* ion by application of 1 MHz electric
fields at 2 kV m ' (Liu et al, 1990). Another report has suggested that irradiation of a
Na^K^-ATPase with 9.14 GHz RF at an SAR of 20 W kg"' generally elevated its activity,
although the activity was depressed by irradiation at 25"C (Brown and Chattopadhyay, 1991).
This observation is consistent with earlier observations that RF irradiation may depress the
activity of Na*,K*-ATPase at temperatures which produced conformational changes in the
active site.

Gene Expression. The possible effects of RF and microwave radiation on gene

expression have not been thoroughly investigated. One recent study found that exposure of
glioma cells to continuous wave (CW) radiation of either 27 MHz or 2.45 GHz resulted in
a dose dependent athermal change in gross transcription as measured by incorporation of
e (Cleary et al. 1990a). Transcription was elevated at an SAR of 25 W kg"', but
appeared to be unchanged or even depressed at higher SARs. No eftect of frequency was
obvious, and the reported effects appeared to persist over a relatively long lime. It should be
noted, however, that RF radiation at 2.55 GHz has been reported to alter gene expression in
a bacterial system by a mechanism which was apparently thermal in nature, despite a gross
temperature rise of only 0.1°C (Saffer and Profenno, 1992). Hence any reports of altered
transcription following exposure at high SARs should be treated with caution.

Proliferation. Athermal effects on the rate of DNA synthesis have been reported
following exposure of glioma cells (Cleary et al, 1990a) or human lymphocytes (Cleary et
aL 1990b) to either 27 MHz or 2.45 GHz at SARs of 25 W kg"'. However, higher SARs
resulted in a smaller response, and in some cases actually depressed DNA synthesis. In
contrast, athermal exposure to 2.45 GHz radiation has been found not to affect transformation
of human lymphocytes in vitro, although thermal exposure enhanced transformation to the
232 N. A. Cridland et al.

same extent as conventional heating (Czerska ef al, 1992). In the same system, exposure to
pulsed fields significantly increased transformation under both thermal and athermal condi-
tions.
Krause et al (1990) have reported that exposure of mouse fibroblasts to 915 MHz
amplitude modulated microwaves at an SAR of 3 W kg' increased ODC activity by 2- to
3-fold. To put this in perspective it should, however, be noted that phorbol ester treatment
and serum stimulation both produced increases of over 20-fold under the same conditions.

Tumour Progression
It has been suggested that microwaves may affect tumour progression, and one
mechanism for this could involve impairment of the immune system which normally plays
a role in preventing tumour development. Veyret et al (1991) reported changes in the
antibody responses of mice following repeated exposure to very low level pulsed 9,4 GHz
microwaves, amplitude-modulated at 14-41 MHz, at an SAR of 0.015 W kg"'. The direction
of the observed effect appeared to be dependent on the frequency of amplitude modulation.
A recent preliminary report suggested that daily exposure of groups of 44 Fisher 344
rats to either CW, or pulse modulated 915 MHz microwaves did not enhance the growth of
brain tumours induced by direct injection of RG2 tumour cells (Salford el al. 1992).
Exposures which were for 7 h/day and 5 days/week started 5 days after inoculation of tumour
cells into the head of the right caudate nucleus. Rectal temperatures were monitored before,
during and after exposure using an optical temperature device, and were not altered by
exposure. Animals were sacrificed after 3 weeks and brains were examined histopathologi-
cally.

SUMMARY
There is no convincing evidence that ELF electric or magnetic fields cause genetic
damage and it is therefore extremely unlikely that they could have any effect on the initiation
of cancer. It is generally accepted that if these fields do affect carcinogenesis it is likely to
be at the level of promotion. This possibility has been investigated at the cellular and
subcellular levels, principally by exploring the possibility that cell signalling pathways may
be affected leading to increa.sed cellular proliferation. Many of these studies have centred
on changes in calcium signalling, whilst others have followed the signalling pathways to the
nucleus and examined the expression of proto-oncogenes known to be involved in regulating
cellular behaviour, Possible effects on cellular proliferation have been assessed directly and
animal carcinogenesis models have been employed to assess the potential for electromag-
netic fields to act as promoting agents. Overall, the available experimental evidence remains
contradictory and does not provide a clear indication that electromagnetic fields affect
tumour promotion. Further studies are required to clarify the situation with respect to putative
effects on cellular processes.
There is limited recent evidence suggesting that ELF fields may act as co-promoters,
essentially enhancing the effects of chemical tumour promoters, possibly via an effect on
cell-cell communication. These studies are, however, at best preliminary, and will need to
be both repeated and extended before any firm conclusions can be drawn.
Exposure to electric or magnetic fields has, under some circumstances, been reported
to inhibit the night-time synthesis of melatonin, which may have effects on the growth of
certain tumours, and this has therefore been suggested as a route by which electromagnetic
fields could influence tumour promotion. However, this effect has not always been success-
fully replicated, and it is possible that the reported changes in melatonin are more attributable
 Recent Biological Studies Relevant to Carcinogenesis 233

to the way the samples are collected, stored or analyzed than exposure to electromagnetic
fields. Overall, the link between exposure to electromagnetic fields and depression of
melatonin levels must remain tentative. Effects mediated by endogenous opioids have been
postulated, but have not been demonstrated.
It has been suggested that ELF magnetic fields could affect tumour progression via
a suppression of the immune system. There is, however, very little evidence for effects on
the immune system and the positive results which have been reported are tentative and do
not appear to have been independently replicated.
There is no convincing new evidence that athermal exposure to RF induces genetic
damage and thus exposure is unlikely to initiate carcinogenesis. The evidence for effects on
ion pumps or cell proliferation, which could constitute a mechanism for influencing tumour
promotion, is extremely limited. There is no new convincing evidence that RF irradiation
can affect tumour progression.

REFERENCES
Bellossi, A., 1991. Effect of pulsed magnetic fields on leukemia-prone AKR mice. No-cffcct on mortality
through five generations. Leukemia Res. 15:899-902.
Beniash\ili. D. Sh., Bilanishvili, V. G.. and Menabde, M. Z., 1991, Low-frequency cleettoiiiagnctic radiation
enhances the induction of rat mammary tumours by nitrosomethyl urea. Cancer Leil. dl: 75-79,
Blank. M.. Soo, I.., Lin, H., Henderson, A. S.. and Goodman, R., 1992,Changes in transcription in llL-60 cells
following exposure to alternating currents from electric fields, Bioelct li'oclieiii. Bii>ciieri;el. 28:.''01-
309.
Blask, D. H.. and Hill, S. M., 1986, Effects of melatonin on cancer: Studies on MCF-? human breast cancer
cells in culture,./. Neural. Tran.sm. 21 (Suppl):433-449.
Brown. H. D., and Chattopadhyay, S. K.. 1991. Ouabain inhibition of kidney .'\TPase is altered by 9.14 GHz
radiation. Biot'lectromagiielic.s 12:137-143.
Cain, C. 0., I homas, D. L., and Adey, \V. R., 1993a, 60 Hz magnetic field acts as a co-promoter in focus
formation of C3H: IOTl/2 cells, Carcinogeiie.^h 14:955-960.
Cain, C. D., Thomas, D. L., Jimenez. L., Jones. R. .^..and Adey, W. R., 1993b. Dependence of 60 Hz magnetic
field strength on copromotion of focus formation, in vitro. IN Ihe Annual Rcxiew of Research on
Biological Effects of Electric and Magnetic Fields from the Cieneration. Deli\ery and Use of
Electricity. Savannah. Georgia. 3 1 October - 4 November
Campisi. J.. Gray. H. H., Pardee, .\. B.. Dean. M., and Sonenshcin, G. I',, 1984. Cell-cycle control o(c-m\c
but not c-^as expression is lost following chemical transformation. Ceil 36:241-247.
Carson. J. L.. Prato. F. S., Drost, D. J., Diesbourg. L. D., and Dixon, S. J.. 1990. Tiinc-\arying magnetic fields
increase cytosolic free Ca^* in HL-60 cells. Am. J. Physiol. 259: [Cell Phv.siol. 28):C687-C692,
Cleary, S. F.. Liu, L-M., and Merchant. R. E.. 1990a, Glioma proliferation modulated in vitro by isothermal
radiofrequency radiation exposure. Radial. Res. 121: 38-45.
Cleary. S. P. Liu. L-M.. and Merchant. R. E.. 1990b, In vitro lymphocyte proliferation induced by radio-fre-
quency electromagnetic radiation under isothermal conditions. Bioclettronuifiiietics I 1:47-56.
Cohen. M.. Lippman, M., and Chabner. B., 1978. Role of pineal gland in aetiology and treatment of breast
cancer iudrcV 2:814-816.
Coulton. L. .\.. and Barker. A. '['.. 1993. Magnetic fields and intracellular calcium: effects on lymphocytes
exposed to conditions for 'cyclotron resonance', Fhvs. Med. Hied. 38:347-360.
Cridland. N. A., 1993, Electromagnetic fields and cancer: A review of relevant cellular studies, Chilton.
NRPB-R256, HMSO, London.
Cridland. N. A. Cragg, T A., and Saunders, R. D.. 1993, Proliferative responses of human fibroblasts exposed
to ELF magnetic fields. IN Electricity and Magnetism in Biology and Medicine (M Blank, editor) San
Francisco, San Francisco Press, pp 628-632.
Czerska. E. M.. Elson, E, C , Davis, C. C . Swicord. M. L., and Czerski. P.. 1992. Effects of continuous and
pulsed 2450-MHz radiation on spontaneous lymphoblastoid transformation of human lymphocytes
in \\lTO. Bioeleelroniagiieties 13:247-259.
Fairbairn, D. W. and O'Neill, K.. L., 1994. The effect of electromagnetic exposure on the formation of DN.A
single strand breaks in human cells. Cell. Mol. Biol. 40:561-567.
234 N. A. Cridland et al.

Fiorani, M., Cantoni. O.. Sestili, P.. Comi. R.. Nicolini, P., Vetrano, F., and Daclia. M., l')92. Electric and or
magnetic field effects on DNA structure and function in cultured human cells, Miiial. Res. 2X2:25-29.
Fitzsimmons. R. J., Strong, D. D., Mohan, S., and Baylink, D. J., 1992, Low-amplitude, low-frequency electric
field-stimulated bone cell proliferation may in part be mediated by increased R i F - l l release,./ Cell.
Physiol. 150;84-S9.
Garaj-Vrhovac, V., Horval, D.. and Koren, Z., 1990. The effect o f microwave radiation on the ceil genome,
Xhiiai. Res. 24.t;87-93.
Garaj-Vrho\ac, V,, Horvat, D., and Koren, Z. 1991, The relationship between colony-forming ability.
chromosome aberrations and incidence of micronuclei in V79 Chinese hamster cells exposed to
microwave radiation, Mutal. Res. 263:143-149.
Garaj-Vrhovac, V., Fu_i_, A., and Horvat, D., 1992, The correlation between the frequency of micronuclei and
specific chromosome aberrations in human lymphocytes exposed to microwave radiation in vitro,
Miiuil. Res. 281:181-186.
Garcia-Sagredo, J. M., and Monteagudo, J, L., 1991, Effect of low-level pulsed electromagnetic fields on
human chromosomes in vitro: Analysis of chromosomal aberrations, lleyedilas. 1 1.5:9-11.
Garcia-Sagredo, J. ,V1., Parada 1,. A., and Monteagudo, J. L., 1990, Effect on SCE in human chromosomes in
\ i t r o o f low level pulsed magnetic field. Environ. Mol. Mutagen. 16:185-188.
Gold, S., Goodman, R., and Shirley-Henderson, .A., 1994, Exposure of Simian Virus-40-transformed human
cells to magnetic fields results in increased levels of T-antigen mRNA and protein. Bioelecironuii;-
/if/fV.v 15:329-336.
Goodman, R., Blank, M., Lin, H., Dai, R., Khorkova, O.. Soo, L., Weisbrot, D., and Henderson, A., 1994,
Increased levels of hsp70 transcripts induced when cells are exposed to low frequency electromagnetic
fields, Bioeleclrochem. Bioenergel. 33:115-120.
Goodman, R.. and Henderson, A. S., 1991. Transcription and translation in cells exposed to extremely low-
frequency electromagnetic fields. Bioeleelroehem. Bioenerget. 25:335-355.
Goodman. R,. Wei, L-X., Bumann, J., and Henderson, A., 1992, Exposure to electric and magnetic ( E M ) fields
increases transcripts in HL-60 cells: does adaptation to EM fields occur? Bioeleelroehem. Bioenerget.
29:185-192.
(ioodman, R., Wei, L-.X.. .Xu, J - C , and Henderson, ,\.. 1989, Exposure o f human cells lo low-frequency
electromagnetic fields results in quantitative changes in transcripts, Bioehiin. Bioplns .Utu 1009:216-
220.
Goodman, R., Weisbrot. D.. Uluc, .-X.. and Henderson, .A., 1992, Transcription in Drosophiki melaiioguster
salivary gland cells is altered following exposure to low-frequency electromagnetic fields: .Analysis
of chromosome 3R, Bioeleetromiigneties 13:111-118.
Greene, J. J., Pearson, S. L., Skowronski. W. J., Nardonc, R. M., Mullins, J. M., and Krause, D., 1993,
Ciene-specific modulation of RN.A synthesis and degradation by extremely low frcquencv eleciromag-
netic fields, a-//. ,V/o/. S/W. 39:261-268.
Greene, J. J., Skowronski, W. J., .Mullins, M., Nardonc. R. .M., Penafiel, M., and Meister, R., 1991, Delineation
of electric and magnetic field effects of extremely low frequency electromagnetic radiation on
transcription, fl;(«/(t'Hi. Biophys. Res. Coinm. 174:742-749.
Gupta, D., .'\ltanasio. A., and Reitcr, R. J., (editors) 1988. The Pineal Gland and Cancer. Tubingen. I^rain
Research Promotion-
H i l l , S. M.. and Blask, D. C , 1988, Effects of the pineal hormone melatonin on the proliferation and
morphological characteristics of human breast cancer cells (MC F-7) in culture, Cuneer Res. 48:61 2 1-
6126.
Kavaliers. M., andOssenkopp, K-P, 1992, Magnetic fields, opioid systems, and day-night rhythms of behav ior
IN Electromagnetic Fields and Circadian Rhythmicity ( M . C. Moore-Ede. S. S. Campbell, and R. J.
Reiter, editors), Boston, Birkauser, pp 95-117.
Kato, M., Honnia, K., Shigemitsu. T., and Shiga, Y , 1993. ElTects of exposure to a circularly polarised 50-H7
magnetic field on plasma melatonin levels in rats, Bioeleetroinagneties 14:97-106.
Kato, M., Honma, K., Shigemitsu, T , and Shiga, Y., 1994a, Circularly polarised 5()-H7 magnetic field exposure
reduces pineal gland and blood melatonin concentrations o f Long-Evans rats, Nettrosei. Lett. 166:59-
62.
Kato, M., Honma, K., Shigemitsu, T , and Shiga, Y , 1994b, Horizontal or vertical 50-Hz. 1-^T magnetic fields
have no etTect on pineal gland or plasma melatonin concentrations of albino rats, Neurosei. Lett.
168:205-208.
Kelly, K., Cochran, B. H., Stiles, C. D., and Leder, P., 1983, Cell-specific regulation of the c-mi'c gene by
lymphocyte mitogens and platelet-derived growth factor. Cell 35:603-610.
Recent Biological Studies Relevant to Carcinogenesis 235

Kerbacher. J. J., Mcltz, M. L.. and Ervvin, D. N., 1990, Influence of radiofrequency radiation on chromosome
aberrations in CHO cells and its interaction with DNA-damaging agents, Radial. Res. \ ^Vi 11 -? 19.
Khalil. A. M., and Qassem. W., 1991, Cytogenetic effects of pulsing electromagnetic field on human
lymphocytes in vitro: Chromosome aberrations, sister-chromatid exchanges and cell kinetics. Mural.
/?«-. 247:141-146.
Kowalczuk, C. I., and Saunders, R. D., 1990, Dominant lethal studies in male mice after exposure to a .''O-Hz
electric field. Rioclcctromagnelics 11:129-137.
Krause, D., Brent, J. A.. Mullins. J. M., Penafiel, L. M.. and Nardone, R. M.. 1990. Hnhancemenl of ornithine
decarboxylase activity in L929 cells by amplitude modulated microwaves. IN .abstracts. 12th Annual
Meeting of the Bioelectromagnetics Society, San .Xntonio, Texas. June 1990. p 94.
Lacy-Hulbert, .\., Wilkins, R.. Metcalfe, J. C , and Hesketh, R.. 1994. Quantitatixe northern analysis o{iiivc
gene expression in ELF field-exposed HL60 Cells. IN .Abstracts. 16th Annual Meeting of the
Bioelectromagnetics Society. Copenhagen. June 1994. p >2.
Lee, J. M., Stormshak, F.. Thompson, J. M.. Thinesen, P., Painter. L. J., Olenchek, E. (i.. Hess. F). L.. Forbes,
R.. and Foster. D. L.. 1993, Melatonin secretion and puberty in female lambs exposed to en\ ironmental
electric and magnetic fields, Biol. Reprod. 49:857-864.
Liburdy, R. P.. 1992, Calcium signalling in lymphocytes and ELF fields. Evidence for an electric field metric
and a site of interaction involving the calcium ion channel, FEBS Lett. 301 :.'i3-.S9.
Liburdy, R. P., Callahan. D. E., Harland, J.. Dunham. E.. Slonia, T. R., and Yasvven, P. 1993a. Experimental
evidence for 60 Hz magnetic fields operating through the signal transduction cascade. Effects on
calcium infiux and c-MYC mRNA induction, FEBS Ull.. 334:301-308.
Liburdy, R. P., Sloma. T R.. Sokolic, R., and Yaswen, P.. 1993b. ELF magnetic fields, breast cancer, and
melatonin: 60 Hz fields block melatonin's oncostatic action on ER^ breast cancer cell proliferation,
/ Pineal Res. 14:89-97.
Lin, FL, Goodman, R., and Shirley-Henderson, A., 1994. Specific region of the c-myc. promoter is responsive
to electric and magnetic fields, J Cc//, Bioehem. 54:281-288.
Linstrom, E., Lindstrom. P.. Berglund. A., Mild, K. II., and Lundgren, E,, 1993, Intracellular calcium
oscillations induced in a T-cell line by a weak 50 Hz magnetic field, J. Cell. Pliysinl. 156:395-398,
Litovit/. T. A.. Krause, D.. and Mullins, J. M., 1991, Effect of coherence time of the applied magnetic field on
ornithine decarboxylase activity. S/oc/it'm. Biophxs. Res. Coinin. 178:862-865.
Liu, D-S., Astumian, R. D..andTsong, T. Y., 1990, Activation of Na* and K* pumping modes of(Na.K)-.\TPase
by an oscillating electric field../. Biol. Chem. 265:7260-7267,
Livingstone, G. K., Witt. K. L., Gandhi, O. P.. Chatterjee. 1.. and Roti Roti. J. L,. 1991, Reproducli\ e integrity
of mammalian cells exposed to power frequency electromagnetic fields. Eiuinw. .\!ol. .\ltaai;en.
17:49-58.
Lorimore. S. A., Kowalczuk, C. L. Saunders, R. D., and Wright. E. G.. 1990. Lack of acute effects of 20 niT.
50 Hz magnetic fields on murine haemopoiesis. Int../. Radiut. Biol. 58:713-723,
Loscher. W., and Mevissen, M., 1994, Animal studies on the role of 50 6()-Hertz magnetic fields in carcino-
genesis. Life Set. 54:1531-1543,
Loscher, W., Mevissen. M,, Lehmacher, W., and Stamm. .A., 1993, Tumour promotion in a breast cancer model
by exposure to a weak alternating magnetic field. Cancer Lett. 71:75-81.
Loscher. W,, Wahnsehaft'e. U,. Mevissen, M., LerchI, A., and Stamm. .'\.. 1994, Effects of weak alternating
magnetic fields on nocturnal melatonin production and mammary carcinogenesis in rats. Oneo/oi^y
51:288-295.
Lyle. D. B., Wang, X., .Ayotte. R. D,, Sheppard, A. R., and Adcy. W, R., 1991. Calcium uptake by Icukaemic
and normal T-lymphocytes exposed to low frequency magnetic fields, Bioeleclri>tnai;netics 12:145-
156.
.McCann, J., Dietrich, F., Raffcrty, C , and Martin, A. O.. 1993. .\ critical review of the genotoxie potential of
electric and magnetic fields, Miilal. Res. 297:61-95.
McLean, J. R. N., Stuchly. M. A., Mitchel, R. E. J., Wilkinson, D.. Yang, H.. Goddard. M,. Lecuyer. D. W..
Schunk, M.. Callary. E.. and Morrison, D., 1991. Cancer promotion in a mouse-skin model by a60-Hz
magnetic field: 11. Tumour development and immune response, Bioeleetromagneiies 1 2:273-287,
Mevissen, M,. Stamm, .\.. Buntenkotter, S., Zwingleberg, R., Wahnschaffc. LJ., and Loscher. W.. HW."!. EtTects
of magnetic fields on mammary tumor development induced by 7.12-dimethylbenz(a)anthracene in
rats, Bioeleelminagnetics, 14:131-143.
Moore-Ede, M. C . Campbell. S. S., and Reiter, R. J., (editors) 1992. Electromagnetic Fields and Circadian
Rhythmieity, Boston, Birkauser,
236 N. A. Cridland tt al.

Morris, J, F., Sasser, L. B., Buschbom. R L.. and Anderson, L. E., 1990. Natural immunity in rats exposed to
60-Ilz magnetic fields. IN Project Resumes, DOE Annual Review of Research on Biological EtTccls
of ?0 and 60 Hz Electric and Magnetic fields, November 1990, Denver, Colorado, p A-2.1.
Murphy. J. C . Kadan. D. A.. Warren. J., and Sivak. A.. 1993, Power frequency electric and magnetic fields: A
review of genetic toxicology. Miilal. Res. 296:221-240.
Nordenson. I., Hansson Mild, K., Andersson,G.,andSandstr6m, M., l994.ChroiTiosomal aberrations in human
amniotic cells after intermittent exposure to fifty Hertz magnetic fields, Bi(>e/ectiomijf;nciic.s L'>:291-
301.
NRPB, 1992. Electromagnetic fields and the risk of cancer Report of an Advisory Group on Non-ionising
Radiation, Doc .\'RPB. 3.(1)1-138.
Parker, J. E., and Winters, W.. 1992. Expression of gene-specific RNA in cultured cells exposed to rotating
60-Hz magnetic fields. Bioclwm. Cell Biol. 70:237-241.
Phillips, J. L., Haggren, W., Thomas, W. J., Ishida-Jones, T., and Adey, W. R., 1992, Magnetic field-induced
changes in specific gene transcription, Biochim. Biophys. Acta 1132:140-144.
Phillips, J. L., and McChesney, L., 1991, Effect of 72 Hz pulsed magnetic field exposure on itiacroinolecular
synthesis in CCRF-CEM cells, Ciincer Biochem. Biophys. 12:1-7.
Putinas. J., and Michaelson. S. M.. 1990. Humoral responsiveness of mice exposed to 500 (.iT. 60 Hz magnetic
field. RioeU'clriiclu'in. Bioeiwrg. 24:371-374.
Prasad. .\. V.. Miller, M. W., Carslensen. E. L., Cox, Ch., Azadniv, M., and Brayman, A., 1991, Failure to
reproduce increased calcium uptake in human lymphocytes at purported cyclotron resonance exposure
conditions. Rcidiai. Environ. Biophys. 30:305-320.
Rannug, .A., Ekstrom, T., Hansson .Mild, K., Holmberg, B., Giinenez-Conti. I., and Slaga, T. J., 1993a. .\ study
on skin tumour formation in mice with 50 Hz magnetic field exposure. Carcinogenesis I4:573-57X,
Rannug, .A., Holinberg, B.. Ekstrom, and Mild, K. H., 1993c, Rat liver foci study on coexposure with 50 Hz
magnetic fields and known carcinogens, Bioelectromagnelics 14:17-27.
Rannug, .X., Holmberg, B., Ekstiom. T., Hansson Mild, K., Gimenez-Conti, 1.. and Slaga, T. J.. 1994.
Intermittent 50 Hz magnetic field and skin tumour promotion in SENC.AR mice. Carcinogenesis
15:153-157.
Rannug, A.. Holmberg. B.. and Mild. K. H.. 1993b, .A rat liver foci promotion study with 50 Hz magnetic fields.
Environ. Res. 62:223-229.
Reipert, B.. .Allan, D., Dale. R. E.. and Dexter. T. M.. 1992. Interaction of low-frequency, low-intensity
electromagnetic fields with haemopoietic stem cells, IN Abstracts, The First World Congress for
Electricity and Magnetism in Biology and Medicine, Orlando. Florida. June 1992. p 54.
Reipert, B., .Allan. D., Dale, R. E.. and Dexter. T. M.. 1994. Interaction of low-frequency, low-intensity
electroinagnetic fields with haemopoietic stein cells. IN .Abstracts, 16th .Annual Meeting of the
Bioelectroinagnetics Society, Copenhagen, June 1994, p54.
Reiler. R. J., 1993, A review ofneuroendocrine and neurochemical changes associated with static and extremely
low frequency electromagnetic field exposure. Integr Physiol. Behav. Sec 28:57-75.
Rochev. Yu. .A.. Narimanov. A. A.. Sosunov. E. .A.. Kozlov. A. N.. and Lednev. V. V.. 1990, EtTects of weak
magnetic field on the rate of cell proliferation in culture, Sludia Biophysica 135:93-98.
Rogers, \V. R.. Reiter. R. J., Stnith, H. D., and Barlow-Walden, L., 1991, Under some circumstances, combined
electric and magnetic field exposure reduces seruin melatonin concentration in nonhuman priinates.
IN Project Resumes, DOE .Annual Review of Research on Biological Effects of 50 and 60 llz Electric
and Magnetic Fields, November 1991, Milwaukee. Wisconsin, p A-26.
Ross, S. M., 1990. Combined DC and ELF magnetic fields can alter cell proliferation. Bioeleclromagnelics
11:27-36.
Saalman, F.. Onfelt, A., and Gillslcdt-Hedman. B.. 1991, Lack of c-mitotic etTects in V79 Chinese hamster
cells exposed to 50 Hz magnetic fields. Hioeleclrochem. Bioenerg. 26:335-338.
Saffer, J. D., and Profenno, L. A., 1992, Microwave-specific heating aftects gene expression, Bioeleclronuig-
iierics 13:75-78.
Saffer, J. D., and Thurston, S. J., 1994, .Alteration of gene expression by ELF fields. IN .Abstracts, 16th Annual
Meeting of the Bioelectroinagnetics Society, Copenhagen. June 1994. p 54.
Salford. E.G.. Brun. .A.. Eberhardt, J. L.. and Persson. B. R. R.. 1992, Development of rat brain tumours during
exposure to continuous and pulsed 915 MHz electromagnetic radiation. IN Abstracts, The First World
Congress for Electricity and Magnetism in Biology and Medicine, June 1992, Orlando, Florida,
p27,
Sarkar, S., .AH, S., Behari, J., 1994, Efi'ect of low power microwave on the mouse genome: .A direct DN.A
analvsis, .Mumr Res. 320:141-147.
Recent Biological Studies Relevant to Carcinogenesis 237

Saunders, R. D., Kowalczuk. C. I , and Sienkiewicz.Z. J., 1991, Biological effects of'exposure to non-ionising
electromagnetic fields and radiation; III. Radio-frequency and microwave radiation. Chilton, NRPB-
R240. London. HMSO.
Sasser, L. B., Morris, J. E.. Buschbom, R. L., Miller, D. L., and Anderson. L. E.. 1991. Effect of 60 Hz electric
fields on pineal melatonin during various times in the dark. IN Project Resumes. DOE Annual Review
of Research on Biological Effects of .'50 and 60 Hz Electric and Magnetic Fields. November 1991.
Milwaukee. Wisconsin, p A-24.
Scarfu M. R., Bersani, P.. Cossarizza. A.. Monti, D., Castellani. G.. Cadosi. R., Franccschetti, G., and
Franceschi, C , 1991. Spontaneous and mitomycin-C-induced micronucici in human lymphocytes
exposed to extremely low frequency pulsed magnetic fields, Biochem. Biophys. Res. Comm. 176:194-
200.
Scarfi, M. R., Lioi, M. B.. Zeni. O.. Franceschetti. G.. Franceschi. C , and Bersani. F.. 1994. Lack of
chromosomal aberration and micronucleus induction in human lymphocytes exposed to pulsed
magnetic fields. Miilal. Res. .Wr 129-133.
Schimmelpfeng, J., and Dertinger. H., 1993, The action of 50 Hz magnetic and electric fields upon cell
proliferation and cyclic .^MP content of cultured mammalian cells. Bioeleclrochem. Bineneri;e!.
30:143-150.
Sienkicwicz. Z. J.. Kowalczuk, C. 1.. and Saunders, R. D., 1991, Biological effects of exposure to non-ionising
electromagnetic fields and radiation: II. Extremely low frequency electric and magnetic fields.
Chilton, NRPB-R239, London, HMSO.
Sienkicwicz, Z, J., Cridland, N. A., Kowalczuk, C. I., and Saunders, R. D., 1993, Biological effects of
electromagnetic fields and radiation. IN Review of Radio Science 1990 - 1992 (\V. R. Slone. editor)
Oxford, Oxford University Press, pp 737-770.
Shah, G., Ghosh, R.. .Amstad, P. A., and Cerutti, P. A., 1993, Mechanism of induction ofc-/(),v by ullraviolel B
(290 - 320 nm) in mouse JB6 epidermal cells. Caneer Res. 53:38-45.
Stein, B.. Angel. P., van Dam, H.. Ponta. H.. Herrlich, P, van der Eb, A., and Rahmsdorf, W. J., 1992,
Ultraviolet-radiation induced c-/w» gene transcription: two .XP-l like binding sites mediate the
response. Phnlochem. Pholohiol. 55:409-415.
Stevens. R. G.. 1987. Electric power use and breast cancer: A hypothesis. Am. J. Epidemiol. 125:556-561.
Stevens. R. G.. Davis, S., Thomas, D. B., Anderson, L. E.. and Wilson, B. W., 1992, Eleclric pouer. pineal
function, and the risk of breasl cancer, F.4SEB J. 6:853-X6().
Stuchly, M. A., McLean. J. R. N.. Burnett. R.. Goddard, M.. Lecuyer. D. W., and Mitchel. R. H. J., 1992.
Modillcation of tumour promotion in the mouse skin by exposure to an alternating magnelic field.
Caneer Leil. 65:1-7.
Svedenstal. B-M..and llolmberg, B., 1993, Lymphoma development among mice exposed to .X-rays and pulsed
magnetic fields, Inl. J. Radial. Biol. 64:119-125.
Tabrah. F. L.. Mower. H. F., Batkin. S.. and Greenwood, P. B., 1994, Enhanced mutagenic effect of a 60 Hz
time-varying magnetic field on numbers of azide-induced TAIOO revertant colonies. Bioeleelmmag-
neiics 15:85-93.
Tan, D-X., Chen, L-D., Poeggeler, B., Manchester, L. C , and Reiler, R. J., 1993a, Melatonin: .X potent,
endogenous hydroxyl radical scavenger. Endocrine J. 1:57-60.
Tan. D-X.. Poeggeler, B., Reiter. R. J., Chen, L-D., Chen, S.. Manchester, L. C , and Barlow-VValdcn. L. R..
1993b. The pineal hormone melatonin inhibits DNA-adduct formalion induced by the chemical
carcinogen safrole in vivo, Caneer Leil. 70:65-70.
Thansandotc, A.. McLean, J. R. N., Lecuyer, D. W., Davidson, C Goddard. M. J., and Burnett. R., 1993. The
effect of 60-Hz magnetic fields and ambient light on tumour co-promotion in SENC.'^R mouse skin.
IN Abstracts, XXIVlh General .Assembly of the International Union of Radio Science. Kyoto. Japan,
.August 25th - September 2nd.
Veyret, B.. Bouthet, C , Deschaux, P., de Seze, R., Gcffard, M., Joussol-Dubien, J., le Diraison, M., Morcau,
J-M., and Caristan, A.. 1991, Antibody responses of mice low-power microwaves under combined
pulse-and-amplitude modulation, Bioeleetromagnelies 12:47-56.
Walleczek, J., and Budinger. T. F, 1992, Pulsed magnetic field effects on calcium signalling in lymphocytes:
dependence on cell status and field Intensity, FEBSLett. 314:351-355.
Walleczek, J.. Killoran. P L.. and Adey. W. R.. 1994, 60-Hz magnetic field effects on Ca-* (Mn-^) influx in
human Jurkat T-cells: Strict dependence on cell state. IN .Abstracts, 16th .Annual Meeting of the
Bioelectromagnetics Society, Copenhagen, June 1994. p 76.
Walleczek, J., and Liburdy, R. P., 1990, Nonthermal 60 Hz sinusoidal magnetic-field exposure enhances ""Ta-*
uptake in rat thymocytes: Dependence on mitogen activation, FEBS Leil. 271:157-160.
238 N. A. Cridland et al.

Wei, L-X., Goodman, R., and Henderson, A., 1990, Changes in levels of c-myc and histonc H2B following
exposure of cells to low-frequency sinusoidal electromagnetic fields: evidence for a window effect.
Bioclcctromagnt'livs 11:269-272.
WHO. 199.1, Electromagnetic Fields (300 Hz to 300 GHz). Environmental Health Criteria 137, Geneva, World
Health Organization.
Wilson. B. W.,and .Anderson, L. E,, 1990, ELF electromagnetic-field effects on the pineal gland. IN E.\tremel\
Low Frequency Electromagnetic Fields: The Question of Cancer (B. W. Wilson, R. G. Stevens, and
L. F. .Anderson, editors), Columbus, Ohio, Battelle Press, pp 159-186.
Wilson, B. W.,Stevens,R.G., and Anderson. L. H, (editors) 1990a, Extremely Low Frequency Electromagnetic
Fields: The Question of Cancer, Columbus, Ohio, Battelle Press.
Wilson. B. W.. Wright. C. W., Morris, J. E.. Buschbom, R. L., Brown D. P, Miller. D. L., Sommers-Flannigan,
R., and .Anderson, L, F., 1990b, Evidence for an effect of ELF electromagnetic fields on human pineal
gland function,,/. Pineal Res. 9:259-269.
Yellon, S, M.. 1994, Acute 60 Hz magnetic field exposure effects on the melatonin rhythm in the pineal gland
and circulation of the adult Djungarian hamster, J. Pineal Res. 16:136-144.
Yost, VI. G., and Liburdy, R. P., 1992, Time-varying and static magnetic fields act in combination to alter
calcium signal transduction in the lymphocyte, FEBS LeII.. 296. 117-122.
Zagon, L S., and McLaughlin, P. J., l9S9a, Endogenous opioid systems regulate growth of neural tumor cells
m culture. Brain Res. 490:14-25.
Zagon. I. S.. and McLaughlin, P. J., 1989b. Opioid antagonist modulation of murine neuroblastoma: ,A profile
of cell proliferation and opioid peptides and receptors. Brain Res. 498:16-28.