Cell Biology International, 1996, Vol. 20, No.

6, 429435

HIGH AND LOW DENSITY WATER AND RESTING, ACTIVE AND

TRANSFORMED CELLS

PHILIPPA M. WIGGINS

Department of Medicine, University of Auckland Faculty of Medicine and Health Science, Private Bag 92019,
Auckland, New Zealand

Accepted 22 April 1996

Resting and active states of cells are described in terms of the expectation, derived from
experiments with aqueous polymers, that they contain two modified forms of water: high
density, reactive, fluid water and low density, inert, viscous water. Low density water
predominates in a resting cell and is converted to high density water in an active cell. It is
proposed that switching from one state to another is an integral part of cellular function.
When this ability is lost cells are transformed either to a state of rigor or to a hyperactive
state in which they no longer depend upon external signals. ? 1996 Academic Press Limited

K: water; cytomatrix; osmosis.

INTRODUCTION High and low density water

Highly charged polymeric solutions and gels have
been shown to contain two modified forms of
water, each of which differs in all measured
properties from normal water (Wiggins, 1994a,b,
1995a,b,c). Low density water is more strongly
hydrogen-bonded, more viscous and, relative to
normal water, is highly selective in its solvent
properties. High density water is more fluid than
normal water and, again, relative to normal water
and to low density water has highly selective sol-
vent properties. Lightly charged, weakly hydrogen-
bonding polyamide gels, on the other hand, contain
principally low density water. Since both intra-
and extracellular matrices consist of mixtures of
polymers presenting to water both highly charged
regions of surface and weakly hydrogen bonding
or hydrophobic regions of surface (proteins, poly-
saccharides, mucins, proteoglycans, polynucleo-
tides) it is highly probable that they, too, contain
both abnormal populations of water molecules.
This expectation has many consequences for
physiological and pathological states of cells, some
of which are outlined briefly in the following
discussion.
10656995/96/060429+07 $18.00/0
These states of water are metastable and transient
but they appear to last long enough to play sig-
nificant roles in biochemical and physiological
processes. Their formation is determined by both
thermodynamic and kinetic considerations. In
simple solutions a gradient in osmolality is abol-
ished by movement of water and solutes until their
chemical potentials are equal throughout. That is
the state of lowest free energy. In aqueous polymer
solutions and gels, however, water is sometimes
prevented from moving to abolish a gradient in its
chemical potential between two regions of freely
exchangeable molecules: it must then adopt a
different strategy for approaching equilibrium.
Figure 1 shows an example which has been dis-
cussed in detail elsewhere (Wiggins, 1990, 1995b).
This represents a protein molecule in solution with
its counter-ions, both positive and negative. Both
the protein molecule and the counter-ions have
kinetic energy but, since the mass of the protein is
so much greater than that of the ions, its movement
through the solution is slower. The result is that
enough ions to neutralize the fixed charges on the
protein move through the solution with it. These
? 1996 Academic Press Limited
430
+ +
+ 1
+ +
+
+
+ +
Fig. 1. A protein in solution with its counter-ions.
ions can exchange rapidly with other ions in the
solution but, whether excess electrolyte is present
or not, there is always a higher local concentration
of ions in the restricted zone round the protein, the
double layer of fixed charges and counter ions, than
in the rest of the solution. It follows that there is
also a gradient of osmolality across the apparent
interface between that zone and the rest of the
solution. The high electric field at the surface of the
protein prevents water from moving to equalize
its activity in the two regions. Experiments with
gels (Etzler and Fagundus, 1983; Wiggins, 1995b)
have suggested that equilibration of water is
approached by means of changes in local density.
Starting at that imaginary interface where
molecules sense their state of disequilibrium and
moving away from it in both directions, the local
density of water increases in the double layer where
its activity is low and decreases in the rest of the
water. Propagation of high density water is limited
by the protein surface; it seems probable that
propagation of low density water is limited by
relative rates of diffusion and by convection and
stirring.
If the protein molecule in contact with water
were absolutely stationary, high and low density
water zones would grow out from its surface with
progressive movement of water molecules together
in the double layer and apart in the external water,
generating, at the rate of self-diffusion of water,
larger and larger zones of high and low density
water. Their size would then be limited by convec-
tion currents, stirring etc. The protein, however,
is not stationary, and as it moves through the
solution relatively slowly it sheds its waters of
hydration, which immediately revert to normal
density while other waters are freshly perturbed by
the protein surface. These dynamics put an upper
limit on the average thickness of modified water
zones associated with the charged regions of the
protein surface. Presumably, it would be a function
Cell Biology International, Vol. 20, No. 6, 1996
+
2
+
+
4 3
+
+
Fig. 2. Growth and shedding of high ( ) and low density water
( ) round a diffusing polyelectrolyte ( ).
1 2
4 3
Fig. 3. Growth and shedding of low density water ( ) round a
hydrophobic molecule ( ).
of mp/mw, the ratio of the mass of the protein to the
mass of a water molecule. Thus the larger the
protein molecule the thicker the zones of modified
water. Even in a gel the matrix has mobility, even if
only of vibration of side chains, so that the extent
of development of high and low density water is
limited. The limitation of growth of modified water
layers is illustrated in Fig. 2.
A similar argument applies to hydrophobic areas
of protein surfaces and hydrophobic moieties of
other molecules, at which water molecules can
make fewer and weaker hydrogen bonds than mol-
ecules in the bulk. Collectively these molecules
move apart from one another, decreasing their
density to compensate for their increased enthalpy,
generating a zone of low density water (see Fig. 3).
In this case the maximal limit to growth of low
density water may be set by the distance over which
hydrogen bonding is cooperative (Barnes et al.,
1979). This maximum may not be reached for a
small molecule with a high diffusion coefficient,
but may be approached by a larger less mobile
hydrophobic surface.
Cell Biology International, Vol. 20, No. 6, 1996
Fig. 4. Aggregation of two hydrophobic molecules in solution:
the hydrophobic interaction. ( =hydrophobic molecule;
=low density water.)
Effects of solutes
The highly selective solvent properties of high and
low density water also affect their lifetimes. When
solutes partition selectively into low density water
(K + , univalent anions, amino acids, compatible
solutes) it increases in volume, if water can move to
follow the concentrating solute, or it reverts toward
normal density; solutes relatively excluded from
low density water (small cations, hydrophobic mol-
ecules) decrease its volume but decrease its density
still further (Wiggins, 1995b). At a charged inter-
face, if low density water accumulates enough
solute to allow it to relax back to normal density,
high density water also relaxes.
Hydrophobic and micro-osmotic forces for
assembly of biopolymers
The hydrophobic interaction has been interpreted
in terms of the growth of low density water at
hydrophobic surfaces (Wiggins, 1990); it is illus-
trated in Fig. 4. Each of two separate hydrophobic
molecules has a zone of high enthalpy water sur-
rounding it. The solution can decrease its overall
free energy in two ways: either two hydrophobic
surfaces align themselves to each other, driving out
some of the high enthalpy water, or the high
enthalpy water expands, decreasing its local density
and chemical potential. Each of these solutions has
its problems: expansion of water is opposed by the
structural entropy term TS; expulsion of water
from between hydrophobic surfaces increases the
entropy of water but decreases that of the solute
molecules.
Figure 4 shows a compromise state of lowest
overall free energy in which some water has been
expelled, decreasing the amount of high enthalpy
water and the residual water has expanded.
431
+
+ +
+
+ + +
+
+
+ + +
+ + + + + +
+ + + + +
++ +
+
+ +
+
+ +
+
+ + + +
Fig. 5. Aggregation of two polyelectrolyte molecules ( ) in
solution: the micro-osmotic force. ( =high density water;
=low density water.)
A similar force for aggregation has been ident-
ified in charged biopolymers (collagen triple helices
and DNA double helices (Rau and Parsegian, 1992;
Leikin et al., 1994) osmotically stressed by poly-
ethylene glycol. It has been suggested (Wiggins,
1995b) that this force, which also increases with
increasing temperature, is due, not to a direct
interactive force between polyelectrolytes, but due
to movement of water outward from between poly-
electrolytes. Parsegian and coworkers have called it
the hydration force; it appears to be the same as the
micro-osmotic force described here. Figure 5 shows
two polyelectrolyte molecules, each surrounded by
a region of water of low activity, containing the
counter-ions in contact with a region of relatively
high water activity; the system can lower its free
energy either by expansion of water in the zones
where its activity is high or, as in the hydrophobic
interaction, by eliminating some high activity
water. In both Figs 4 and 5 the aggregation is such
that the surface area of polymer in contact with
water and therefore the volume of residual high
activity or high enthalpy water is minimized. The
three forces which determine whether polyelectro-
lyte molecules aggregate or not (electrostatic
repulsion, van der Waals attraction and the
micro-osmotic force) are additive.
Figure 6 illustrates some extremely simple exper-
iments with solutions of sodium dextran sulphate
(molecular weight 500,000) which undergo phase
separation as a result of the operation of the
micro-osmotic force. A solution of dextran sul-
phate of about the same water content as a cell or
extracellular matrix (3 g water/g dry weight) is
extremely viscous; addition of NaCl, of any con-
centration, decreases it viscosity but maintains it as
a single solution. Addition of KCl above a critical
concentration (0.7 at 21)C) induces separation
into a clear viscous gel containing most or all of the
432 Cell Biology International, Vol. 20, No. 6, 1996

M KCl
NaCl

M NaCl

Two phases Dextran sulphate Single solution

Top: fluid in water fluid
Lower: gel single solution
viscous

KCl KCl

NaCl NaCl
NaCl NaCl

Fig. 6. The micro-osmotic attractive force, driven by KCl and the repulsive micro-osmotic force, driven by NaCl in dextran
sulphate solutions. ( =dextran sulphate; =high density water; =low density water.)

dextran sulphate and a fluid watery phase; i.e.
dextran sulphate molecules have aggregated into
the gel phase and expelled water. This is the
attractive micro-osmotic force. It is opposed and
the single phase restored by addition of a suitable
concentration of NaCl; this is an example of the
repulsive micro-osmotic force.
The two lower boxes illustrate the roles of the
two populations of water molecules. When NaCl is
added to a solution of dextran sulphate in water it
partitions selectively into high density, low activity
water, generating a fresh osmolality gradient which
is partially abolished by movement of water from
its low density region into the high density region,
thus converting some low density water into high
density water. Addition of KCl to the solution of
sodium dextran sulphate in water, on the other
hand, converts some high density water into low
density, high activity water. Two possible responses
of the system in its approach to its state of lowest
overall free energy are: on the left, it squeezes out
some of the high activity water, yielding two
phases, the extruded water on top and a gel, still
containing some low density water, at the bottom;
on the right all high activity water has been
extruded and the gel contains only normal water.
This is an example of relaxation of high density
water back to normal density as it is no longer
stressed by contact with low density water. These
simple experiments show that because the micro-
osmotic force is not a direct force between one
macromolecule and another it can be either attrac-
tive or repulsive, depending upon the direction of
the osmolality gradient down which water moves.
The direction of that gradient, in turn, depends
critically upon the small solutes present. It is pro-
posed that this versatility, combined with a cells
unique ability to control its cytoplasmic compo-
sition by means of pumps and channels is an
important component of metabolism and homeo-
stasis. It must also be an important factor contrib-
uting to cellular volume regulation (Macknight
et al., 1994).
THE STRUCTURE OF THE CYTOPLASM
There is considerable evidence that cytoplasm is
not a simple solution with proteins, electrolytes and
non-electrolytes all randomly dissolved in all the
water (Pagliaro and Taylor, 1988; Getzenberg
et al., 1991; Knull, 1992; Luby-Phelps et al., 1993).
Cell Biology International, Vol. 20, No. 6, 1996 433

+ + + + + + + + + + + +
+ +
+ + + +
+ + + + + + +
+ + + + + +
+ + + +
+ +
+ + + + + +
+ + +
+ +

+ + + +
+ + + + + + + + + + + +
+ + + + + + +
+ + ++
+ + + + +
+ + + + + +
+ + + + + + +
+ + + + + + + +
+ + +
+ +
+ + + + +
+ + +
+ + +
+ + + ++ +
+ + + + + + + + +
+ + + + + +
+ +
+ + + + + + + +
+ + + + + +
+
+ + + + + + + +
+ + + + +
+ + + + + + +
+ + + + + + + + +
+ + + + + + +
+ + + + + + +

+
+ + + + + + +
+ + + +
+
+ + +
+ + + + + + +
+ + +
+ + +

Fig. 7. Proposed physiological structure of left, a quiescent cell and right, an active cell. ( =protein; =high density water;
=low density water; , =compatible solute.)

Al-Habori (1995) has reviewed evidence for the rest of cytoplasmic water which contains only low
microcompartmentation of the cytoplasm and concentrations of compatible solutes and few ions.
possibly metabolic channelling; i.e. the transfer of Responses to this state of disequilibrium are three-
the product of one enzyme reaction directly to the fold: first, some of the high activity water between
next enzyme in the metabolic sequence without proteins is squeezed out, reducing its volume and
an intermediate sojourn in solution in the main thus decreasing the overall energy of the cytoplas-
body of cytoplasmic water. Drost-Hansen and mic compartment; second, the water remaining
coworkers have written extensively about the between double layers decreases its local free
possible role of vicinal water in cellular processes energy by expanding; third, water in the double
(Clegg and Drost-Hansen, 1991; Drost-Hansen layers increases its local free energy by compacting.
and Singleton, 1992a,b). Clegg (1984, 1992) has In its final state the cytoplasm consists of proteins,
suggested that enzymes are clustered round the surrounded by high density, fluid, reactive water
actin filaments of Porters (1984) microtrabecular containing most of the ions of the cytoplasm,
lattice; he has shown that removal of much of the clustering round actin filaments in such a way that
rest of the water from the cytoplasm has little effect their surface areas are minimized, so that the
upon enzyme activity. Cleggs perception of the amount of residual high activity water which
structure of the cytoplasm is consistent with much expands and becomes viscous and inert is mini-
of the experimental data both from his laboratory mized. This is the quiescent state of a cell illustrated
and from others. Figure 7 illustrates a modification at the left of Fig. 7. The active state of a cell is
of his concept, incorporating the micro-osmotic represented at the right of Fig. 7. Cells are acti-
principles outlined above. vated by influx of ions (Ca2+ and Na + ) which
Intracellular proteins are highly charged and, as induce the repulsive micro-osmotic force, increas-
illustrated in Fig. 1, are surrounded by a zone of ing spaces between proteins, increasing the amount
water in which the counter ions (mainly K + and and fluidity of high density water. This usually
Cl " are confined. In the resting state of a cell Ca2+ stimulates cascades of phosphorylation of proteins
is sequestered in intracellular stores and Mg2+ (Torphy, 1994), which, by increasing charges on
complexed to ATP and other anions. The immedi- their surfaces increase still more the amount and
ate effect of the high local concentration of ions fluidity of high density water and decreases the
surrounding the proteins is to create a water amount, but not the viscosity of low density water.
activity gradient between the double layers and the Although transfer of a phosphate from ATP to a
434 Cell Biology International, Vol. 20, No. 6, 1996

+ + + + + + + + + + +

+
+ +
+ +

+
+ ++

+
+

+
+ +

+
+ + + + + + + ++ + ++
+ + +

+
+

+
+ + + + + +
+ +

+
+ + + + +
+ + +
+ + +
+ +

+
+

+
+ + + + +

+
+ + + + + + + + + +
+

+
+ + + + ++ + +

+
+
+ + +

+
+ +
+ + + + + + + +

+
+

+
+ + + ++ + +
+ + + + +
+ + + +

+
+ + + + +

+
+ + + + +

+
+
+ + + + + + +
+ + +

+
+ + + + + +
+

+ +

+
+ + + ++
+ ++

+
+ + +
+ + + + + + +
+ +

+
+ + +

+
+ + + + +

+
+ + + + +
+

+
+
+
+ + + +

+
++ + + + +

+
+
+ + + + +
+

+
+
+ + + + + + + +
+ + + +

+
+ +

+
+ + + +
+ + +
+ + + +

+
+

+
+ + + + + + + +

+
+
+ + + + + + + + + + +

+
+ + + + + +
+ + + + + +
+ + + + +

++ + + +
+ + +
+ + + + + +

+ +
+ + +

+
+ + + +

+
+ + + + +
+

+
+

+
Fig. 8. Two proposed pathological states of a cell: left, a rigid state of rigor; right, a hyperactive uncontrolled state. ( =protein;
=high density water; =low density water; =compatible solute; # =carcinogen.)

protein does not increase the number of negative Polymers are plasticized by water and with too
charges in the cytoplasm it does increase the little water present they assume a state in which any
number of fixed charges on polymeric surfaces; as movement is extremely slow (Slade et al., 1993).
pointed out earlier, the size of the molecule carry- Residual water associated with the proteins is of
ing a charge is important in determining the extent two extreme forms. The high density water is very
of growth of high and low density water popu- fluid and very reactive because it is rich in free OH
lations. The average viscosity in the cytoplasm of groups and lone pairs of electrons, which are
the active cells is so lowered that diffusion is rapid, the reactive centres of water molecules. It has been
resistance to conformational changes of proteins suggested that this might be a contributory mech-
reduced, and gross protein movements (cytoplas- anism of apoptosis in which sustained increases in
mic streaming, sliding of filaments, etc.) facilitated. intracellular Ca2+ levels lead to fragmentation of
Moreover, there is a fast track through extremely DNA, plasma membrane disruption and irrevers-
fluid water which will selectively retain most ions ible depletion of ATP (Santos-Argumedo et al.,
and many metabolites. 1993). The extreme reactivity of high density water
The active state relaxes back to the quiescent has been demonstrated in unpublished experiments
state when the stimulating ions are pumped back with Dowex AG 1X8 anion exchange resin. The
across the plasma membrane or into cytoplasmic charged groups -N(CH3)3 + are hydrolysed in
stores. This reversal of the repulsive micro-osmotic water, the rate of production increasing with
force restores the tightly clustered quiescent state. conditions which promote high density water.
Figure 8 illustrates two possible pathological An entirely different possible pathological state
states of cytoplasm. Experiments with ion exchange is illustrated on the right of Fig. 8. Carcinogens
resins have shown that Ca2+ ions are the preferred are often either small hydrophobic molecules or
counter-ions at negatively charged surfaces cations, such as Ni2+ . Both species accumulate
(Ca2+ >Mg2+ >K + >Na + ). They induce extreme selectively into high density water and their
forms of high and low density water. The phenom- removal can pose problems. Cells have highly
enon of rigor in muscle occurs when ATP is selective and efficient transport mechanisms for
exhausted and Ca2+ cannot be transported back removal of ions which they commonly encounter
into the sarcoplasmic reticulum. It is a state of and use like Na + , H + and Ca2+ ; they are not likely
extreme and irreversible rigidity (Grinwald and to be as effective for foreign ions like Ni2+ . Small
Nayler, 1981). The left of Fig. 8 illustrates this. The hydrophobic molecules have been shown to have a
extremely high relative activity of water between remarkably high affinity for high density water
double layers results in expulsion of much water from which they are almost impossible to wash out.
with formation of a low water content rigid gel. For example (Wiggins, 1995b), BzOH disappeared
The rigidity of the gel is due more to its low water entirely into both a cation and an anion exchange
content than to the absolute viscosity of the water. resin and could not be eluted with 10 bed
Cell Biology International, Vol. 20, No. 6, 1996
volumes of water. It was finally removed following
neutralization of the gel. A plasma membrane is no
barrier to inward diffusion of small hydrophobic
molecules and their high selective solubility in high
density water is apparently not balanced by an
efficient extrusion mechanism. Cells have only
P-glycoprotein which is induced in drug-resistant
states in tumour cells; it is not clear how effective it
is in normal cells (Critchfield et al., 1994). The
possibility is suggested, therefore, that accumu-
lation of hydrophobic and cationic carcinogens
might transform cells into irreversibly active states
in which they continue to grow and metabolize
without needing external stimuli. The hyperactive
state illustrated in Fig. 8 contains more water and
more Na + ions than normal cells. Accumulation
of small hydrophobic molecules into high density
water increases its selective affinity for small
cations so that the hyperactive state is reinforced
by increases in cytoplasmic concentrations of Na +
and Ca2+ . Both pathological states have lost their
ability to switch between active and resting states.
ACKNOWLEDGEMENT
This work was funded by a grant from the Health
Research Council of New Zealand.
REFERENCES
A-H M, 1995. Microcompartmentation, metabolic
channelling and carbohydrate metabolism. Int J Biochem
Cell Biol 27: 123132.
B P, F JL, N JD, Q JE, 1979. Co- smooth muscle relaxation: challenges to the dogma. Trends
operative effects in simulated water. Nature 282: 459464.
C JS, 1984. Intracellular water and the cytomatrix:
some methods of study and current views. J Cell Biol 99:
167s171s.
C JS, D-H W, 1991. On the biochemistry and 51: 7984.
physiology of water. In Hochachka PW, Mommson TP,
eds. Biochemistry and Molecular Biology of Fishes Vol. 1.
Elsevier Science Publishers, 123.
C JS, 1992. Cellular infrastructure and metabolic organ-
isation. Curr Top Cell Reg 33: 314.
C JW, W CJ, P JM, Y GC, 1994. W PM, 1995b. High and low density water in gels. Prog
Modulation of adriamycin accumulation and efflux by fla-
vonoids in HCT-15 colon cells Activation of P-glycoprotein
as a putative mechanism. Biochem Pharmacol 48: 143745.
435
D-H W, S JL, 1992a. Our aqueous herit-
age: evidence for vicinal water in cells. In Fundamentals of
Medical Cell Biology, 3A. JAI Press, 157180.
D-H W, S JL, 1992b. Our aqueous
heritage: role of vicinal water in cells. In Fundamentals of
Medical Cell Biology, 3A. JAI Press, 181202.
E FM, F DM, 1983. The density of water and
some other solvents in narrow pores. J Colloid Interface Sci
115: 513519.
G RH, P KJ, W WS, C DS, 1991.
Nuclear structure and the three-dimensional organisation of
DNA. J Cell Biochem 47: 289299.
G PM, N WG, 1981. Calcium entry in the
calcium paradox. J Mol Cell Cardiol 13: 867880.
K HR, 1992. Association of glycolytic enzymes with the
cytoskeleton. Curr Topics Cell Reg 33: 15.
L S, R DC, P VA, 1994. Direct measurement
of forces between self-assembled proteins: temperature-
dependent exponential forces between collagen triple helices.
Proc Natl Acad Sci 91: 276280.
L-P K, L F, T DL, 1988. The submicro-
scopic properties of cytoplasm as a determinant of cellular
function. Ann Rev Biophys Biophys Chem 17: 369396.
M ADC, G LGM, P RD, 1994. Prob-
lems in the understanding of cell volume regulation. J Exp
Zoo 268: 8089.
P, T, 1988. Aldolase exists in both the fluid and
solid phases of the cytoplasm. J Cell Biol 107: 981991.
P KR, 1984. The cytomatrix: a short history of its study.
J Cell Biol 99: 3s12s.
P DW, MD A, M M, L-P K,
1993. Cytoarchitecture of size-excluding particles in living
cells J Cell Sci 106: 565578.
R DC, P VA, 1992. Direct measurement of
temperature-dependent solvation forces between DNA
double helices. Biophys J 61: 260271.
S-A L, T C, P G, K PA,
P RM, 1993. A B lymphocyte surface molecule
mediating activation and protection from apoptosis via
calcium channels. J Immunol 151: 31193130.
S L, L H, I J, W M, 1993. The glassy
state phenomenon in application for the food industry. J Sci
Food Agric 63: 133176.
T TJ, 1994. Beta-adrenoceptors, cAMP and airway
Pharmacol Sci 15: 370374.
W PM, 1990. Role of water in some biological
processes. Microbiol Rev 54: 432449.
W PM, 1994a. Osmosis and micro-osmosis. NZ. Sci Rev
W PM, 1994b. Mutual interactions of ions, nonelectro-
lytes and water at charged and hydrophobic surfaces. Trends
Electrochem 3: 129140.
W PM, 1995a. Micro-osmosis, a chaotic phenomenon of
water and solutes in gels. Langmuir 11: 19841986.
Polymer Sci 20: 11211163.
W PM, 1995c. Micro-osmosis in gels, cells and enyzmes.
Cell Biochem Function 13: 165171.