.................... ................................................

SYMPOSIUM PRESENTATION

Eric C. Reynolds, PhD

Anticariogenic complexes of amorphous calcium phosphate stabilized

by casein phosphopeptides: A review

ental caries is initiated via the USA and Australia, the prevalence of
Using laboratory, animal, and
demineralization of tooth hard fluorosis was found to have
human in situ caries models, investi-
tissue by organic acids from increased in both fluoridated and
gators have shown that casein phos-
the fermentation of dietary sugar by non-fluoridated area^.^,'^ From a
phopeptide amorphous calcium phos-
dental plaque odontopathogenic bac- recent Western Australian study,
phate complexes (CPP-ACP) exhibit
teria.‘ Even though, in most devel- 48.4% of the 7-8-year-old children
an anticariogenic activity. The
oped countries, the prevalence of examined exhibited fluorosis of their
casein phosphopeptides (CPP) are
dental caries has decreased through maxillary permanent central incisors
produced from a tryptic digest of the
the use of fluorides, the disease (TF fluorosis classification scores of
milk protein casein by aggregation
remains a major public health prob- 1-3), with 43.5% of the children
with calcium phosphate and purifica-
lem.’ In a recent Australian oral showing mild forms (TF scores 1-2)
tion by ultrafiltration. The CPP have
health survey, 61% of 12-year-olds and 4.9% a more severe form (TF
a remarkable ability to stabilize cal-
examined showed signs of caries, score 3).1°This represented a 100%
cium phosphate in sofution and sub-
with a mean Decayed, Missing and increase in the TF score 3 prevalence
stantially increase the level of calci-
Filled Tooth index (DMFT) of By that was found in a similar Western
um phosphate in dental plaque.
the age of 25-29 years, 98% of those Australian study in 1991.11Fluorosed
Through their multiple phosphoseryl
examined were affected by caries, enamel is characterized by the for-
residues, the CPP bind to forming
with a mean DMFT of 11.7, which mation of a more porous enamel
clusters of amorphous calcium phos-
rose to 21.0, with a 6% decayed com- with a subsurface hypomineraliza-
phate (ACP) in metastable solution,
ponent, by 65 years of age. The esti- tion and can occur following either
preventing their growth to the criti-
mated economic burden of treating an acute or chronic exposure to
cal size required for nucleation and
dental caries in Australia in 1991was above-threshold levels of fluoride
precipitation. The proposed mecha-
$471 million, higher than that for during enamel formation.8Riordan12
nism of anticariogenicity for the
other dietary-related diseases inclu- recently reported that TF scores of 2
CPP-ACP is that they localize ACP in
ding coronary heart disease, hyper- or greater were easily noticed by lay
dental plaque, which buffers the free
tension, or ~ t r o k eIn
. ~developing observers, and when the TF score
calcium and phosphate ion activi-
countries, where the availability of was 3, the fluorosis aroused concern.
ties, thereby helping to maintain a
industrialized food products is The increased prevalence of dental
state of supersaturation with respect
to tooth enamel depressing deminer-
increasing, caries prevalence is also fluorosis has been attributed to the
increasing5Recent studies have inadvertent ingestion of fluoride
alization and enhancing remineraliza-
tion. The CPP-ACP, unlike fluoride,
highlighted a number of socio-demo- toothpaste, infant formula consump-
can be added to sugar-containing
graphic variables associated with tion-particularly when the formula
caries risk-high risk being associat- has been reconstituted with fluori-
foods and therefore have commercial
potential as an additive to foods as
ed with ethnicity and low socio-eco- dated water, the use of fluoride sup-
nomic status.6The level of high-risk plements, and a general increase in
well as to toothpastes and mouth-
washes for the control of dental
individuals has remained constant the fluoride level of infant foods and
caries.
even though the overall severity and beverages through processing with
prevalence of disease in the commu- fluoridated water.1° In a recent
nity have decrea~ed.~ Australian consensus conference on
In countries where fluorides are the ”Appropriate Fluoride Exposure
used, it is now recognized that the for Infants and Children”, it was rec-
level of dental fluorosis has ommended that toothpaste manufac-
increased? In recent surveys in the turers produce a 400-ppm fluoride

S SCD Snacinl Care in DentiatN. Val I 8 No I 1998

paste specifically for children and
that fluoride supplements in fluori-
dated areas be phased out in an
approach to reduce the prevalence of
fluorosis.13However, no data exist to
demonstrate the anticariogenicity of
the 400-ppm fluoride paste, and a
recent study by Koch et showed
that a 250-ppm fluoride paste had a
significantly lower anticaries efficacy
than the normal 1000-ppm fluoride
paste. If these measures to limit
fluoride exposure are unsuccessful in
lowering the prevalence of fluorosis
in Australia, then the next recom-
mendation would be the reduction or
elimination of fluoride in the reticu-
lated water supply. It is possible,
therefore, that these measures imple-
mented to reduce fluoride exposure
may in fact produce an increase in
caries prevalence, particularly in the
high-risk groups.
In conclusion, dental caries is still
a major public health problem in
Australia, particularly in ethnic and
lower socio-economic groups, and
may become worse as measures are
taken to reduce fluoride exposure.
This highlights the requirement for
the development of a non-toxic, anti-
cariogenic agent that could be added
to toothpaste, mouthwash, and food
in an approach to lower caries experi-
ence. It would be particularly useful
if the anticariogenic agent was a nat-
ural food derivative, since it would
then be considerably easier to obtain
the appropriate regulatory approval
for the agent as a food additive. With
approval as a food additive, the agent
could be incorporated into sugar-con-
taining foods to target the high-
caries-risk groups.
Anticariogenicity
of dairy products
Except for some reports associating
nursing caries with milk consump-
tion, the food group most recognized
as exhibiting anticaries activity is
dairy products (milk, milk concen-
trates, powders, and c h e e s e ~ ) . ~ J ~ - ~Bos
Using in uitro, animal, and in situ
caries models, investigators have
identified the components largely
responsible for this anticariogenic
activity as casein, calcium, and phos-
phate.16”0 The bovine milk phospho-
protein, casein, which is known to
interact with calcium and phos-
phate31and is a natural food compo-
nent, is an obvious candidate for an
anticariogenic food and toothpaste
additive. In the early studies on the
anticariogenicity of casein, the insolu-
ble acid form (caseinHC1)was used,
which required very high levels for
a~tivity.~~.~~
More recently, lower levels of sol-
uble caseinate have been shown to be
anticariogenic in a rat caries model
when added to the drinking water26
or as an ingredient or supplement in
c o n f e c t i ~ n e r y .The
~ ~ ,level
~ ~ of
caseinate (17% w/w) which signifi-
cantly reduced the cariogenicity of
the chocolate confection was consid-
ered too high by the manufacturers,
due to the unpalatability of the
caseinate. At a much lower, palatable
level (2% w/w), the caseinate did not
significantly change the confection’s
cari~genicity.~~ It was concluded that
casein’s adverse organoleptic proper-
ties and the large amount required
for efficacy precluded its use as a
food or toothpaste additive to lower
the risk of caries.
Anticariogenic casein
phosphopeptides
Using a human intra-oral caries
model, investigators have shown that
digestion of caseinate with trypsin
did not destroy the protein’s ability to
prevent enamel subsurface deminer-
a l i z a t i ~ nTryptic
. ~ ~ peptides of casein
were found incorporated into the
intra-oral appliance plaque and were
associated with a substantial increase
in the plaque’s content of calcium and
phosphate. It was concluded that the
tryptic peptides responsible for the
anticariogenic activity were the calci-
um phosphate sequestering phospho-
peptides. The major casein phospho-
peptides (CPP) released by trypsin
that sequester calcium phosphate are
~ aS1-caseinX-5P (f59-79) [l]and
Bos p-casein X-4P (fl-25) [2],together
with smaller amounts of Bos as2-
casein X-4P (46-70) [3] and Bos as2-
casein X-4P (fl-21) [4].
SCD Special Care in Dentistry, Vol18 No 1 1998 9
Gln-Met-Glu-Ala-Glu-Ser(P)-Ile-
Ser(P)-Ser(P)-Ser(P)-Glu-Glu-Ile-
Val-Pro-Asn-Ser(P)-Val-Glu-Gln-
Lys. - as1(59-79).
Arg-Glu-Leu-Glu-Glu-Leu-Asn-
Val-Pro-Gly-Glu-Ile-Val-Glu-
Ser(P)-Leu-Ser(P)-Ser(P)-Ser(P)-
Glu-Glu-Ser-Ile-Thr-Arg. - p(1-25).
Asn-Ala-Asn-Glu-Glu-Glu-Tyr-
Ser-Ile-Gly-Ser(P)-Ser(P)-Ser(P)-
Glu-Glu-Ser(P)-Ala-Glu-Val-Ala-
Thr-Glu-Glu-Val-Lys.- as2(46-70).
Lys-Asn-Thr-Met-Glu-His-Val-
Ser(P)-Ser(P)-Ser(P)-Glu-Glu-Ser-
Ile-Ile-Ser(P)-Gln-Glu-Thr-Tyr-
Lys. - as2(1-21).
All the CPP contain the phospho-
seryl cluster sequence -Ser(P)-Ser(P)-
Ser(P)-Glu-Glu-.The CPP released by
trypsin are approximately 10%w/w
of caseinate and through their multi-
ple phosphoseryl residues sequester
their own weight in calcium phos-
phate to form colloidal c ~ m p l e x e s . ~ ~
Since the CPP are not associated with
the ~npalatability~~ or antigeni~ity~~
of the caseins and have the potential
for a specific anticariogenicity at least
ten times greater on a weight basis,
then their potential as a food and
toothpaste additive is considerably
better than that of the intact proteins.
A simple and efficient purification
procedure of the CPP has recently
been developed involving ultrafiltra-
tion of amorphous calcium phos-
phate nanoclusters stabilized by the
multiple phosphoseryl-containing
peptides from a tryptic digest of
casein.34The peptides produced by
this procedure have been comprehen-
sively ~ h a r a c t e r i z e d .The
~ ~ ,major
~~
peptides of the preparation are
p(1-25) and as,(59-79)and its deami-
dated forms, with minor amounts of
as2(46-70)and as2(1-21)peptides. The
individual peptides of the prepara-
tion were identified by mass spec-
trometry and amino acid composition
and sequence analyses after purifica-
tion to homogeneity by anion
exchange FPLC and reversed-phase
HPLC.3637Prior to sequence analysis,
the labile phosphoseryl residues were
converted to S-ethyl cysteinyl resi-
dues by P-eliminati~n.~~
(a) Interaction of CPP
with calcium phosphate
The CPP have marked ability to sta-
bilize calcium phosphate in solution.
Using reversed-phase HPLC and
anion exchange FPLC, investigators
have purified the major peptide of a
CPP preparation, as,(59-79),and used
it to characterize its interaction with
calcium phosphate. Solutions con-
taining 0.1% w/v aSl(59-79)at vari-
ous pH, calcium, and phosphate con-
centrations, but constant ionic
strengths, were allowed to attain
equilibrium. Volumes of 10% or less
of each equilibrated solution were
centrifuged through a micropartition
filter with a 2000-Mr exclusion limit.
These filters do not retain calcium
phosphate ions or ion pairs. Calcium
and inorganic phosphate (Pi)concen-
trations in the original solution (to
confirm no precipitation) and the
ultrafiltrate were determined and the
peptide-bound calcium phosphate
calculated. The peptide as1(59-79)
was found to bind, maximally, 24 Ca
and 16 Pi per molecule. The ion activi-
ty products for the various calcium
phosphate phases [hydroxyapatite
(HA); octacalcium phosphate (OCP);
tricalcium phosphate (TCP);amor-
phous calcium phospate (ACP);and
dicalcium phosphate dihydrate
(DCPD)]were determined from the
free calcium and phosphate concen-
trations at each pH by means of a
modified computer program that cal-
culates the ion activity coefficients
through the use of the expanded
Debye-Hiickel equation and takes
into account the ion pairs CaHPO,O,
CaH,PO,+, CaPO,, and CaOH+,the
dissociation of H,PO, and H,O, and
the ionic strength.38The only ion
activity product that significantly
correlated with calcium phosphate
bound to the peptide independently
of pH was that corresponding to ACP
[Ca,(P0,)l,87(HP0,)o,2.xH,0], indicat-
ing that this is the phase stabilized by
as1(59-79).In neutral and alkaline
supersaturated calcium phosphate
solutions, ACP nuclei form sponta-
10 SCD Special Care in Dentistry, Vol18 No 1 1998
neously. The results therefore indi-
cate that as1(59-79),through the mul-
tiple Ser(P) sequence Glu-Ser(P)-Ile-
Ser(P)-Ser(P)-Ser(P)-Glu-Glu, bind to
forming ACP nanoclusters in
metastable solutions, preventing their
growth to the critical size required
for nucleation and precipitation. The
binding of asl(59-79)to ACP results
in the formation of colloidal complex-
es with the unit formula [as1(59-
79)(ACP),], where n is equal to or
greater than one. It is possible that
the predominant form is n = 6, since
as,(59-79)crosslinked with gluta-
raldehyde in the presence of ACP
runs as multimers up to a hexamer
on polyacrylamide gel electrophore-
sis. Interestingly, the synthetic
octapeptide AcGlu-Ser(P)-Ile-Ser(P)-
ser(P)-Ser(P)-Glu-Gl~NHMe~~ binds
only 12 Ca and 8 Pi per molecule, i.e.,
(ACP),, indicating that other residues
and/or conformational specificity is
essential for full ACP binding.
Synthetic peptides corresponding to
the N-terminal sequence Gln-Met-
Glu-Ala-Glu and the C-terminal
sequence Ile-Val-Pro-Asn-Ser(P)-Val-
Glu-Gln of asl(59-79)did not bind
calcium phosphate. A 1.0% w/v CPP
solution can stabilize 60 mM CaC1,
and 36 mM sodium phosphate at pH
7.0 to form colloidal amorphous calci-
um phosphate-CPP complexes (CPP-
ACP) with only 4.94 +- 0.46 mM free
calcium and 4.90 f 0.08 mM free
phosphate. This solution has been
studied in a variety of in vitro, human
in situ, and animal caries models.
(b) Anticariogenicity
of CPP-ACP in the rat
The ability of casein-phosphopeptide
amorphous calcium-phosphatecom-
plexes (CPP-ACP)to reduce caries
activity was investigated with the use
of specific-pathogen-freerats orally
infected with Streptococcus sobrinus
6715WT-13,O. CPP-ACP solutions (100
pL) were applied to the animals’
molar teeth twice daily. Other groups
of animals received 100 pL of solu-
tions containing either 500 pprn F-, the
non-phosphorylated peptides of the
casein tryptic digest, or the synthetic
octapeptide, Ac-Glu-Ser(P)-Ile-Ser(P)-
Ser(P)-Ser(P)-Glu-Glu-NHMe, corre-
sponding to the common sequence in
the major asl-and P-peptides. The
animals consumed a sucrose/gluten-
based diet which did not contain
dairy products. The CPP-ACP signifi-
cantly reduced smooth-surface caries
activity in a dose-response fashion,
with 0.1% w/v CPP-ACP producing a
14% reduction and 1.O% w /v CCP-CP
a 55% reduction relative to the dis-
tilled water controi. A similar but
slightly smaller reduction was found
with fissure caries activity. CPP-ACP
at 0.5-1.0%w/v produced a reduction
in caries activity similar to that of 500-
ppm F-. The anticariogenicity effects
of CPP-ACP and fluoride were addi-
tive, since animals receiving 0.5%
CPP-ACP plus 500 ppm F- had signifi-
cantly lower caries activity than those
animals receiving either CPP-ACP or
fluoride alone. The tryptic digest of
casein with the phosphopeptides
selectively removed showed no anti-
cariogenic activity at 0.5% w/v. The
non-phosphorylated casein peptides
contained a level of arginyl residues
similar to that of the casein phospho-
peptides. The results therefore indi-
cate that the anticariogenicity of the
CPP-ACP is not simply associated
with an arginine-based pH rise.41The
synthetic octapeptide-calcium phos-
phate complex significantly reduced
caries activity, confirming that this
calcium-phosphate-stabilizing portion
of the casein phosphopeptides is asso-
ciated with anticariogenicity.
However, the lower molar specific
activity of the synthetic octapeptide in
binding ACP and in anticariogenicity,
compared with that of the longer asl-
and P-peptides, suggests that other
residues and/or conformational speci-
ficity is required for full activity.
Other possible residues are those
within the sequences -1le-Val-Pro-
Asn-Ser(P)-Val-Glu-Gln-and Glu-
Glu-Leu-Asn-Val-Pro-Gly-Glu- of the
as*-and P-peptides, respectively.
These sequences could have confor-
mational specificity @-turn)required
for full ACP binding, and the
hydrophobic residues could be
involved as attachment anchors at
hydrophobic sites in plaque and the
oral mucosa.
(c) Anticariogenicity of CPP-ACP
in human in situ studies
The ability of the 1.0% CPP, 60-mM
CaCl,, and 36-mM sodium phos-
phate, pH 7.0, solution to prevent
enamel demineralization has been
studied in a human in situ caries
The model has been
described in detail previously29and
consists of a removable appliance
containing a left and right pair of
enamel slabs placed to produce a
plaque retention site. The inter-enam-
el plaque that develops (3-5 mg) is
bacteriologically similar to normal
supragingival plaque.29On frequent
exposure to sucrose solutions over a
three-week period, the levels of
mutans streptococci and lactobacilli
increase, and subsurface enamel
demineralization results in the forma-
tion of an incipient "caries-like"
lesion. Two exposures of the CPP-
ACP solution per day to the right pair
of enamel slabs for 12 subjects pro-
duced a 51 f 19% reduction in enam-
el mineral loss relative to the left-side
(control) enamel. The plaque exposed
to the CPP-ACP solution contained
78 f 22 pmol/g calcium, 52 f 25
pmol/g Pi, and 2.4 f 0.7 mg CPP per
g plaque dry weight compared with
32 k 12 pmol/g calcium and 20 f 11
pmol/g Pi in the control plaque. The
level of the CPP was determined by
competitive ELISA with an antibody
that recognizes both a,,(59-79) and
p(1-25). Electron micrographs of
immunocytochemicallystained sec-
tions of the plaque revealed localiza-
tion of the peptide predominantly on
the surfaces of micro-organisms but
also in the extracellular matrix.
Although these results certainly indi-
cate that CPP are incorporated into
developing dental plaque, the actual
level determined by ELISA would
not be a true representation of that
incorporated due to the breakdown
of the CPP in plaque through the
action of phosphatase and peptidase
a c t i v i t i e ~ .The
~ ~ ,incorporation
~~
the CPP-ACP into the plaque result-
ed in a 144%increase in the plaque
calcium and a 160% increase in
plaque Pi, with a Ca/P, ratio consis-
tent with ACP.
(d) Anticariogenic potential of the
CPP-ACP in a mouthwash study
More recent results from a clinical
trial of a mouthwash used thrice
daily containing 3.0% CPP-ACP, pH
9.0, showed that the calcium content
of supragingival plaque (lower ante-
rior teeth excluded) increased from
177 f 52 pmol/g dry weight to 342 f
89 pmol/g after use of the mouth-
wash for a three-day period, and
inorganic phosphate increased from
306 f 82 pmol/g dry weight to 473 f
125 pmol/g. These results have led to
the proposition that the mechanism
of anticariogenicity for the CPP-ACP
is that they incorporate amorphous
calcium phosphate into plaque,
depressing enamel demineralization
and enhancing remineralization. In
plaque, CPP-ACP would act as a
reservoir of calcium and phosphate,
buffering the free calcium and phos-
phate ion activities, thereby helping
to maintain a state of supersaturation
with respect to tooth enamel. The
binding of ACP to CPP is pH-depen-
dent, with binding decreasing as the
pH falls.
(e) Remineralizationof enamel
lesions by CPP-ACP
An in vitro model system43has been
used to study the effects of CPP-ACP
solutions on remineralization of arti-
ficial lesions in human third molars.
The model involves the preparation
of uniform, reproducible subsurface
enamel lesions which are cut in two,
with one half being used as the con-
trol to the other, which is exposed to
a remineralizing solution. At the end
of the treatment (ten-day exposure),
the lesions are sectioned and subject-
ed to microradiography; their miner-
al content is determined by micro-
densitometry. With this system, the
1.O% CPP-ACP solution used in the
rat caries and in situ experiments
replaced 63.9 k 20.1% of mineral lost.
A 0.1% CPP-ACP solution replaced
of 43.6 * 18.9% of mineral lost.
A series of solutions containing
various amounts of CPP (0.1-l.O%),
calcium (6-60 mM), and phosphate
(3.6-36mM) at different pH values
(7.0-9.0)has now been studied in this
SCD Special Care in Dentistry, Vol18 No 1 1998 11
model system.43The associations
between the activities of the various
calcium phosphate species in solution
and the rates of enamel lesion re-
mineralization for this series of solu-
tions were then determined. The
activities of the various calcium phos-
phate species in solution were calcu-
lated by means of the free calcium
and phosphate concentrations, pH,
and an iterative computational proce-
dure described above. The activity of
the neutral ion species, CaHPO,O, in
the various remineralizing solutions
was found to be highly correlated
with the rate of lesion remineraliza-
tion. The diffusion coefficient for the
remineralization process was estimat-
ed at 3 x m2s1and is consistent
with the coefficients of diffusion for
neutral molecules through a protein
gel/H,O matrix. The rate of enamel
remineralization obtained with the
1.0% CPP-ACP solution was 3.3 x
mol HA/m2/10 days, which is equiv-
alent to rates of remineralization
obtained with constant composition
procedure^.^^ CaHPO? and associat-
ed species after diffusion through the
protein/H,O-filled pores of carious
surface enamel into the body of the
enamel lesion would, by the formation
of Ca2+and ions, increase the
degree of saturation with respect to
HA. The formation of HA in the lesion
would lead to the generation of acid
and phosphate, including H,PO,, which
would diffuse out of the lesion down
a concentration gradient. The results
indicate that the CPP-bound ACP,
CPP[C~,(PO,),.~,(HPO,)~.~.XH~OI,,
acts as a reservoir of calcium phos-
phate ions, including the neutral ion
pair CaHPO,O, which are formed in
the presence of acid. The acid can be
generated by dental plaque bacteria;
under these conditions, the CPP-
bound ACP would buffer plaque pH
and in so doing would dissociate to
calcium phosphate ions, including
CaHPO?. The increase in plaque cal-
cium and phosphate ions and ion pairs
would offset any fall in pH, thereby
preventing enamel demineralization.
Acid is also generated in plaque by the
formation of HA in the enamel lesion
during remineralization. Therefore,
this could explain why the CPP-ACP
solutions are such efficient remineral-
izing solutions, since they would con-
sume the acid generated during enam-
el lesion remineralization by generat-
ing more CaHPO?, thus maintaining
its concentration gradient into the
lesion. These results are therefore con-
sistent with the proposed anticario-
genic mechanism of the CPP, which is
the inhibition of enamel demineraliza-
tion and enhancement of remineraliza-
tion through the localization of ACP at
the tooth surface.
(f) Inhibition of enamel
demineralization by CPP
Roberts45has recently reported that
treatment of enamel slabs in vitro
with a crude preparation of CPP at
0.5% w/v in water significantly
inhibited acid demineralization. The
inhibition obtained with the crude
preparation was approximately one-
third that obtained with a solution
containing 1000 ppm fluoride. In a
similar in vitro study, Roberts45inves-
tigated the effect of a 1%CPP-ACP
solution on 45Cauptake into enamel.
Since the CPP-ACP localized non-
radioactive calcium phosphate at the
tooth surface, this amorphous calci-
um phosphate layer competed with
45Cauptake into the underlying
enamel, which Roberts45assumed
indicated inhibition of enamel re-
mineralization. In experiments where
enamel remineralization has been
measured by subsurface intralesion
mineral deposition, CPP-ACP solu-
tions substantially promoted re-
mineralization (see section e). The
use of a crude CPP preparation at
5.0% w/w in a dentifrice was also
reported by Roberts45as significantly
inhibiting enamel demineralization in
a human intra-oral caries model. The
inhibition obtained in this study was
similar to that obtained with a 500-
ppm fluoride dentifrice. Roberts45
concluded that the in vitro and in situ
caries models predicted anti-caries
activity for CPP.
(g) Potential of CPP to inhibit
plaque formation
Guggenheim et ~ 1 have
. demonstrat-
~ ~
12 SCD Special Care in Dentistry, Vol 18 No 1 1998
ed that cariogenic diets containing
micellar casein or CPP significantly
reduced the numbers of Streptococcus
sobrinus colonizing the teeth of exper-
imental rats. The authors suggested
that this reduction in cariogenic
streptococci was at least partly
responsible for the substantial reduc-
tion in caries obtained with the CPP.
Schiipbach et 1 2 1 . then
~ ~ demonstrated
that incorporation of CPP into sali-
vary pellicle in vitro substantially
inhibited the adherence of Streptococ-
cus mutans and Streptococcus sobrinus.
This result confirms an early study
by Reynolds and W ~ n which g ~ ~
showed inhibition of adherence of S.
mutans to apatite discs by treatment
with casein fractions. Schiipbach et
~ 2 concluded
. ~ ~ that the incorporation
of CPP into pellicle would not only
increase its remineralization potential
but also inhibit incorporation of
cariogenic streptococci.Whether the
inhibition of cariogenic plaque for-
mation is as significant as the inhibi-
tion of enamel demineralization and
promotion of remineralization by the
CPP remains to be established.
(h) Structure of the anticariogenic
casein phosphopeptides
Several phosphoproteins that interact
with calcium phosphates have now
been As well as the
caseins from milk49,they include the
phosphoprotein from saliva
tathe her in^^), egg p h o ~ v i t i nand
~~,
phosphoproteins from mineralized
tissue (dentin pho~phophoryn~l and
bone phosphoproteins, o s t e o p ~ n t i n ~ ~ ,amorphous phases, thereby binding
and matrix-Gla protein53).From the
sequences that are available, a com-
mon feature is apparent:
-Pse-Pse-Pse-Pse-Pse-Pse-Pse-
Phosvitin
-Pse-Asp-Pse-Pse-Pse-Asp-Pse-
Phosphophor yn
-Pse-Met-Pse-Pse-Pse-Glu-Glu-
Riboflavin binding protein
-he-Pse-Gly-Pse-Pse-Glu-Glu-
Osteopontin
-Pse-Ile-Pse-Pse-Pse-Glu-Glu-
asl-casein
-he-Leu-Pse-Pse-Pse-Glu-Glu-
p-casein
-Asp-Pse-Pse-Glu-Pse-
Matrix-Gla protein
Asp-Pse-Pse-Glu-Glu-
Statherin
When Pse is used to indicate a
phosphoseryl residue, the sequences
reveal that these proteins all contain
multiple phosphoseryl and acidic
residues in clusters. The proposed
functions of these proteins are: (i) the
stabilization of calcium phosphate in
solution, preventing spontaneous
precipitation; and (ii) biomineraliza-
tion, where the protein, cross-linked
to a collagen matrix, has been pro-
posed to act as a nucleator/promoter
of crystal growth. It is conceivable
that multiple-phosphoseryl-contain-
ing proteins could, on the one hand,
act as nucleators/promoters of calci-
um phosphate phase separation and,
on the other, as stabilizers of
metastable supersaturated solutions,
depending on concentration and con-
formation of the protein and the com-
position and degree of saturation of
the aqueous phase. An immobilized,
rigid phosphoprotein should bind
strongly in a stereospecific manner to
a particular calcium phosphate
phase, thereby providing a mineral-
ization template. In contrast, the solu-
ble, flexible phosphoproteins-in
particular, phosphopeptides-could
adapt their conformation to a wider
range of surfaces, including those of
spontaneously forming clusters of
calcium phosphate ions in metastable
solution, preventing their growth to
the critical size required for nucle-
ation and precipitation. Protein flexi-
bility in solution is the outstanding
characteristic to emerge from spectro-
scopic studies on phosvitin50and
p h o ~ p h o p h o r y nPhosphorylation
.~~
appears to destabilize secondary and
tertiary structure rather than promote
higher levels of ordering. However, it
is very likely that the flexible phos-
phorylated sequences adapt more
regular conformations when bound
to calcium phosphate. Binding of
Ca2+by phosphophoryn, for exam-
ple, induces a conformational change
with an increase in extended p-struc-
ture as measured by circular dichro-
ism (CD).54Optical rotatory disper-
sion (ORD), CD, hydrodynamic, and
31Pnuclear magnetic resonance
(NMR)measurements of the casein
structures all indicate that as,-casein
and p-casein have a rather open
structure in solution, with many
amino acid side-chains exposed to
solvent and relatively flexible.3231P-
NMR relaxation measurements indi-
cate that Ser(P) residues are relatively
mobile in P - ~ a s e i n . ~ ~
Medium- and long-range nuclear
Overhauser enhancements (noes)
have been demonstrated in 2-D 'H
NMR spectra of as,(59-79) [l]in the
presence of Ca2+,indicating a confor-
mational p r e f e r e n ~ eTwo
. ~ ~ struc-
tured regions were identified with
residues Val72 to Val76 implicated in
a p-turn conformation. Residues
Glu61 to Ser(P)67,which extend over
part of the Ser(P) cluster motif
-Ser(P)-Ser(P)-Ser(P)-Glu-Glu-, are
involved in a loop-type ~ t u c t u r e . ~ ~
Chaplin et ~ 1 have . reported
~ ~ the sec-
ondary structure of the p(1-28) phos-
phopeptide to be composed of p-
structure and random coil, with no
a-helical structure as shown by CD.
However, no measurement in the
presence of calcium or calcium phos-
phate was made in this study.
Wahlgren et aL5* have recently stud-
ied p(1-25) [2] by two-dimensional
NMR spectroscopy and concluded
that the structure was neither that of
a fully-extended peptide nor a ran-
dom coil. Similarly, T ~ u d failed a ~ ~ to
detect medium- and long-range noes
characteristic of structured peptides
in their study of p(1-25). In contrast,
2-D NMR studies on P-casein(1-25)in
the presence of calcium have shown
evidence of medium-range noes at
400 MHz as well as 500 MHz (Huq et
al., unpublished). A medium-range
nOe was observed in the -Ser(P)17-
Ser(P)-Ser(P)-Glu-Glu-21motif region
between the CaH of Ser(P)18and the
NH of Glu20. Further medium-range
noes include one between the CaH
of Ser22 and the NH of Thr24. Pre-
liminary evidence from the lH NMR
spectra of a,-casein(1-21) [4] has
shown that several residues, includ-
ing those around the -Ser(P)-Ser(P)-
Ser(P)-Glu-Glu-motif, are perturbed.
Furthermore, there are medium-
range noes between the NH of
Ser(P)8and the NH of Glull. This is
yet another example of a medium-
range nOe in the -Ser(P)-Ser(P)-
Ser(P)-Glu-Glu-motif. Other exam-
ples of medium-range noes include
that between the NH of Ile14 and the
NH of Ser(P)l6. In summary, the
NMR data indicate that preferred
conformations exist for these pep-
tides in the presence of calcium ions.
Molecular modeling of both aS1(59-
79) and p(1-25), with the constraints
derived from the NMR spectroscopy,
has indicated that the peptides adopt
conformations that allow the gluta-
my1 and phosphoseryl side-chains to
interact collectively with calcium ions.
The relationship between CPP
structure and interaction with amor-
phous calcium phosphate was inves-
tigated by a series of synthetic pep-
tide homologues and analogues
(Reynolds et al., unpublished). These
studies showed that the cluster
sequence -Ser(P)-Ser(P)-Ser(P)-Glu-
Glu- is effectively responsible for the
interaction with ACP, and that all
three contiguous Ser(P) residues are
required for maximal interaction
with ACP. Similarly, the adsorption
of the CPP and synthetic homologues
and analogues onto HA has been
investigated (Reynolds et al., unpub-
lished). These data also confirm that
the Ser(P) cluster sequence is the
major determinant for high-affinity
binding, and that all three contiguous
Ser(P) residues are essential, since the
loss of any one, even when substitut-
ed with a Glu or Asp, resulted in a
lower affinity for the HA surface. The
surface area occupied by the peptides
ap(59-79) and p(1-25) on maximal
binding to HA was commensurate
with the cross-sectional area of their
three-dimensional solution structures.
This result indicates that, under satu-
ration conditions, the peptides are
closely packed on the HA surface.
Interestingly, when these HA adsorp-
tion experiments were repeated with
salivary-coated HA (sHA), the Ser(P)
SCD Special Care in Dentistry, Vol18 No 1 1998 13
cluster motif was still the major
determinant for adsorption, although
the affinities of the peptides for sHA
were slightly reduced by the pres-
ence of the salivary proteins. These
results suggest that the predominant
interaction of the CPP with pellicle
and plaque is likely to be electrostat-
ic, mediated by the Ser(P) cluster
motf of the CPP.
The docking of the peptide Ser(P)-
Ser(P)-Ser(P)-Glu-Glu-onto three
crystallographicplanes of HA-{ 1001,
(OlO), and {OOl)-by computer simula-
tion techniques and the unit cell coor-
dinates of synthetic HA has recently
been investigated (Huq et al., unpub-
lished). These simulation studies re-
vealed that the peptide -Ser(P)-Ser(P)-
Ser(P)-Glu-Glu-is more likely to bind
to the {loo)surface, followed by the
(010) surface. This is in agreement
with the finding that bovine phospho-
phoryn interacts preferentially with
the (100) face of HA (Huq et al.,
unpublished). Further, these results
support those of Furedi-Milhoferet
aL60, who showed that both rat dentin
phosphophoryn and phosvitin, which
also have clusters of Ser(P)residues,
also interact with the apatite-like (010)
face of OCP. The Ser(P)cluster motif
can therefore bind to both (100) and
(010) surfaces, thus allowing for the
deposition of calcium, phosphate, and
hydroxyl ions on the {OOl) surface,
enabling the HA crystal to grow along
the c-axis only. These results therefore
can now explain the c-axis growth of
HA crystals in v i m . Detailed examina-
tion of the computer simulation data
shows that the -Ser(P)-Ser(P)-Ser(P)-
Glu-Glu- conformer with the greatest
relative binding energy is positioned
on the HA surface such that the car-
boxyl groups of the glutamyl residues
and the phosphoryl groups of the
phosphoseryl residues are in proximity
to the HA surface with maximal con-
tact between these groups and surface
calcium atoms, supporting the proposed
solution three-dimensionalstructures.
(i) Interaction of CPP-ACP
with fluoride
The additive anticariogeniceffect of
the 1.0% CPP-ACP and 500-ppm F- in
the rat caries experiments led to the
investigation of the potential interac-
tion between the CPP-ACP and F.
Analysis of the solution containing
1.0% CPP, 60 mM CaCl,, 36 mM sodi-
um phosphate, and 500 ppm F (26.3
mM NaF), pH 7.0, after ultrafiltration
revealed that nearly half of the fluo-
ride ion had incorporated into the
ACP phase stabilized by the CPP to
produce a novel amorphous calcium
fluoride phosphate phase of composi-
tion Ca,(PO,),F x H,O. The identifica-
tion of this novel amorphous calcium
fluoride phosphate (ACFP) phase led
to the proposition that the formation
of this phase is responsible for the
observed additive anticariogenic effect
of CPP-ACP and F. The proposed anti-
cariogenic mechanism of the CPP-ACP
is the localization of ACP at the tooth
surface such that, in the presence of
acid, the ACP dissociates to release
calcium and phosphate ions, increas-
ing the degree of saturation with
respect to HA, preventing enamel
demineralization, and promoting re-
mineralization. The anticariogenic
mechanism of fluoride is now pro-
posed to be the localization of the fluo-
ride ion at the tooth surface, particu-
larly in plaque in the presence of Ca
and phosphate ions. This localization
increases the degree of saturation with
respect to fluorapatite (FA), thus pro-
moting remineralization of enamel
with FA., It is clear that, for the forma-
tion of FA [Calo(P0,),F2],calcium and
phosphate ions must be co-localized in
plaque at the tooth surface with the
fluoride ion. The additive anticario-
genic effect of CPP-ACP and F may
therefore be attributable to the local-
ization of ACFP at the tooth surface by
the CPP which, in effect, would co-
localize ca, Pi, and F. These results
suggest that the CPP may be an excel-
lent delivery vehicle for the co-local-
ization of calcium, phosphate, and
fluoride at the tooth surface in a slow-
release amorphous form with superior
clinical efficacy.
(i)Comparison of CPP-ACP
with calcium sucrose phosphate
(Anticay)
There are some similarities between
14 SCD Special Care in Dentistry, Vol 18 No 1 1998
CPP-ACP and calcium sucrose phos-
phate (CaSP).CaSP is a mixture of
calcium sucrose mono- and diphos-
phates, disucrose monophosphate,
and inorganic calcium phosphate that
contains approximately 11%w / w
calcium and 7.6% w / w inorganic
phosphate.61CaSP has been shown to
decrease tooth enamel demineraliza-
tion, promote enamel remineraliza-
tion, and inhibit the formation of
plaqueb1,similar to the proposed
mechanisms of CPP-ACP. In a two-
year double-blind clinical trial in the
USA, CaSP incorporated into chew-
ing gum was shown to reduce DMFS
by 39% relative to a control gum.,*
This and other clinical trial data led
to the use of CaSP as an anticario-
genic food additive. However, as
explained by Craig61,"the volume of
sales of CaSP proved insufficient to
support continued economic produc-
tion of the material". While there
clearly are some Similarities between
CaSP and CPP-ACP, there are also
some major differences. CPP-ACP
contains 18%w / w Ca and 30% w / w
PO, in the form of amorphous calci-
um phosphate, which represents 2.6
times the amount of calcium phos-
phate of CaSP. The peptides are not
synthetic but have evolved naturally
to carry their own weight in amor-
phous calcium phosphate such that
the CPP have specific conformations
that allow them to stabilize ACP and
to interact with and localize ACP at
the tooth and mucosal surfaces.
Finally, plaque phosphatase and pep-
tidase activity, when degrading the
CPP, ultimately results in a pH rise
through ammonia release.4I Plaque
phosphatase activity on CaSP would
release sucrose, resulting in a plaque
pH fall. Although a direct compari-
son between the efficacy of CaSP and
that of CPP-ACP has not been carried
out, it is interesting to speculate,
therefore, that the efficacy of CPP-
ACP in reducing caries activity
would be superior.
Conclusion
The anticariogenic potential of the
CPP-ACP has been demonstrated in
the rat caries model, the in situ
human caries model, in vitro reminer-
alization models, and human trials.
The CPP-ACP and fluoride were
shown to have additive effects in
reducing caries experience, suggest-
ing that CPP-ACFP would have
potential as a toothpaste additive to
improve the efficacy of the current
fluoride-containing dentifrices. The
CPP-ACP would also have an impor-
tant role in toothpastes either alone
or with low fluoride for children at
risk of fluorosis. The fluoride ion
incorporates into a novel amorphous
calcium fluoride phosphate phase
[Ca,(P0,)5F x H,O] which is stabi-
lized by the CPP, suggesting that the
CPPs are an excellent delivery vehicle
for the co-localization of Ca, F, and
phosphate at the tooth surface in a
slow-release amorphous form, pro-
ducing superior anticaries efficacy.
The proposed anticariogenicity
mechanism for the CPP is that they
stabilize and localize ACP and ACFP
at the tooth surface, thereby buffering
plaque pH and depressing enamel
demineralization and enhancing re-
mineralization. The CPP-ACP anti-
cariogenic activity is greatest when
the peptides are delivered at the
same time as the cariogenic chal-
lenge, and the CPP-ACP are a natural
derivative of milk; therefore, unlike
fluoride, they could be added to
sugar-containing foods. Preliminary
results indicate that the CPP-ACP can
be incorporated into confectionery
without adverse organoleptic effects.
The CPP-ACP therefore could have
an important role as a food additive
for the control of dental caries.
Presented at a symposium entitled "Nutrition
and Prosthodontics-The Inter-relationship of
Diet, Nutrition, and Oral Health", held during
the 74th General Session of the International
Association for Dental Research, March 13,
1996, San Francisco, CA. The symposium and
the publication of these proceedings are sup-
ported by the Block Drug Co.
Dr. Reynolds is Professor of Dental Science
and Director of Research, School of Dental
Science, The University of Melbourne, 711
Elizabeth St., Melbourne, 3000, Australia.
1. Loesche WJ. Role of Strepfptococcusmutans
in human dental decay. Microbiol Rev
 50:353-80,1986,
2. Carr L. Dental health of children in
Australia 1977.1986. Aust Govt Pub Serv,
Canberra, 1987.
3. Barnard P. National oral health survey
Australia 1987-88. Department of Health,
Housing, Local Government and
Community Services. Aust Govt Pub
Serv, Canberra, 1993.
4. Crowley S, Antioch K, Crater R, Conway
L, Matheis C. The economic burden of
diet-related disease in Australia. Paper
prepared for the Nutritional Food and
Nutrition Centre for Health Program
Evaluation and the Australian Institute of
Health, 1992.
5. Thylstrup A, Fejerskov 0.Textbook of
cariology. Copenhagen: Munksgaard,
1986.
6. Spencer AJ, Wright FAC, Brown LM,
Brown LP. Changing caries experience
and risk factors in five- and six-year-old
Melbourne children. Aust Dent J 34:160-5,
1986.
7. Wright FAC. Ministerial review of dental
services in Victoria, Australia, 1986.
8. Ekstrand J, Fejerskov 0, Silverstone LM.
Fluoride in dentistry, Copenhagen:
Munksgaard, 1988.
9. Burt BA. The changing patterns of sys-
temic fluoride intake. J Dent Res 71:1228-
37,1992.
10. Riordan PJ. Dental fluorosis, dental caries
and fluoride exposure among 7-year-olds.
Caries Res 27:71-7,1992.
11. Riordan PJ, Banks JA. Dental fluorosis
and fluoride exposure in Western
Australia. J Dent Res 70:1022-8,1991.
12. Riordan PJ. Perceptions of dental fluoro-
sis. J Dent Res 721268-74, 1993.
13. Riordan PJ. Appropriate fluoride expo-
sure for infants and Children-Consensus
Conference. Perth, 2-3 December, 1993.
14. Koch G, Bergmann-Arnadottir I,
Bjarnason S, Finnbogason S, Hoskuldsson
0, Karlsson R. Caries-preventive effect of
fluoride dentifrices with and without
anticalculus agents: A 3-year controlled
clinical trial. Caries Res 24:72-9, 1990.
15. Reynolds EC, Johnson IH. Effect of milk
on caries incidence and bacterial composi-
tion of dental plaque in the rat. Arch Oral
Biol26:445-51,1981.
16. Rosen S, Min DB, Harper DS, Harper WJ,
Beck EX, Beck FM. Effect of cheese, with
and without sucrose, on dental caries and
recovery of Streptococcus mutans in rats. J
Dent Res 63:894-6,1984.
17. Harper DS, Osborn JC Clayton R, Hefferren
JJ. Cariostatic evaluation of cheeses with
diverse physical and compositional char-
acteristics. Caries Res 20:123-30,1986.
18. Harper DS, Osbom JC, Clayton R,
Hcfferren JJ. Modification of food cario-
genicity in rats by mineral-rich concen-
trates from milk. J Dent Res 66:42-5,1987.
19. Krobicka A, Bowen WH, Pearson S, Yang
DA. The effects of cheese snacks on caries
in desalivated rats. J Dent Res 66:1116-9,
1987.
20. Silva MFdeA, Burgess RC, Sandham HJ,
Jenkins GN. Effects of water-soluble com-
ponents of cheese on experimental caries
in humans. J Dent Res 66:38-41,1987.
21. Schweigert BS, Shaw JH, Zepplin M,
Elvehjem CA. Dental caries in the cotton
rat. VI. The effect of the amount of pro-
tein, fat and carbohydrate in the diet on
the incidence and extent of carious
lesions. J Nutr 31:439-47,1946.
22. Schweigert BS, Potts E, Shaw JH, Zepplin
M, Phillips PH. Dental caries in the cotton
rat. VIII. Further studies on the dietary
effects of carbohydrate, protein and fat on
the incidence and extent of carious
lesions. J Nutr 32:405-12, 1946.
23. Shaw JH. Effects of dietary composition
on tooth decay in the albino rat. J Nutr
41:13-23,1950.
24. Bavetta LA, McClure FJ. Protein factors
and experimental rat caries. J Nutr 63:107-
17, 1957.
25. Holloway PJ, Shaw JH, Sweeny EA.
Effects of various sucrose-casein ratios in
purified diets on the teeth and supporting
structures of rats. Arch Oral Biol3:185-
200,1961.
26. Reynolds EC, del Rio A. Effect of casein
and whey protein solutions on caries
experience of the rat. Arch Oral Biol
29~927-33,1984.
27. Reynolds EC, Black CL. Confectionery
composition and rat caries. Caries Res
211538-45,1987.
28. Reynolds EC, Black CL. The reduction of
chocolate’s cariogenicity by supplementa-
tion with sodium caseinate. Caries Res
21:445-51,1987.
29. Reynolds EC. The prevention of subsur-
face demineralization of bovine enamel
and change in plaque composition by
casein in an intra-oral model. J Dent Res
661120-7,1987.
30. Reynolds EC, Black CL. Cariogenicity of a
confection supplemented with sodium
caseinate at a palatable level. Caries Res
23:368-70,1989.
31. Reeves RE, Latour NG. Calcium phos-
phate sequestering phosphopeptide from
casein. Science 128:472,1958.
32. Swaisgood HE. Chemistry of milk pro-
teins. In: Developments in dairy chem-
istry. Vol. 1. Fox PF, editor. London:
Applied Science Publishers, pp. 1-60,1982.
33. Ametani A, Kaminogawa S, Shimizu M,
Yamauchi K. Rapid screening of antigeni-
cally reactive fragments of a,,-casein
using HPLC and ELISA. J Biochem
102:421-5, 1987.
34. Reynolds EC. Production of phosphopep-
tides. Australian patent application
PK5706,1992.
35. Adamson N, Riley PF, Reynolds EC. The
analysis of multiple phosphoseryl-con-
taining casein peptides using capillary
zone electrophoresis. J Chromatography
646:391-6,1993.
36. Reynolds EC, Riley PF, Adamson N. A
selective-precipitation purification proce-
dure for multiple-phosphoseryl contain-
SCD Special Care in Dentistry, Vol 18 No 1 1998 15
ing peptides and methods for their identi-
fication. Anal Biochem 217:277-84,1994.
37. Adamson N, Reynolds EC. Relationship
between degree of casein hydrolysis and
phosphopeptide release. J Dairy Res
64:505-14,1997.
38. Marsh ME. Binding of calcium and phos-
phate ions to dentin phosphophoryn.
Biochem 28:346-52, 1989.
39. Perich JW, Kelly DP, Reynolds EC.
Efficient solution-phase synthesis of mul-
tiple 0-phosphoseryl-containing peptides
related to casein and statherin. Int J
Peptide Protein Res 40:81-8,1992.
40. Reynolds EC, Cain CJ, Webber FL, Black
CL, &ley PF, Johnson IH, et 01.
Anticariogenicity of tryptic casein- and
synthetic-phosphopeptides in the rat. J
Dent Res 74:1272-9,1995.
41. Reynolds EC, Riley PF. Protein dissimila-
tion by human salivary-sediment bacte-
ria. J Dent Res 68:124-9,1989.
42. Reynolds EC. Anticariogenic phospho-
peptides. US Patent 5015628,1991.
43. Reynolds EC. Remineralization of enamel
subsurface lesions by casein phosphopep-
tide-stabilized calcium phosphate solu-
tions. J Dent Res 76:1587-95, 1997.
44. de Rooij JF, Nancollas GH. The formation
and remineralization of artificial white
spot lesions: A constant composition
approach. J Dent Res 63964-7,1984.
45. Roberts AJ. Role of models in assessing
new agents for caries prevention-non-
fluoride systems. Adv Dent Res 9:304-11,
1995.
46. Guggenheim B, Neeser JR, Golliard M,
Schiipbach P. Salivary pellicle modified
by milk components mediates caries pro-
tection. Caries Res 28:182,1994.
47. Schiipbach P, Neeser JR, Golliard M,
Rouvet M, Guggenheim B. Incorporation
of caseinoglycomacropeptideand
caseinophosphopep tide into the salivary
pellicle inhibits adherence of mutans
streptococci. J Dent Res 75:1779-88,1996.
48. Reynolds EC, Wong A. Effect of adsorbed
protein on hydroxyapatite zeta potential
and Streptococcus mutans adherence. Infect
Immun 39:1285-90,1983.
49. Holt C, van Kemenade MJJM. The inter-
action of phosphoproteins with calcium
phosphate. In: Calcified tissue. Hukins
DWL, editor. Boca Raton, FL: CRC Press,
pp. 175-213,1989,
50. Prescott B, Renugopalakrishnan V,
Glimcher MJ, Bhushan A, Thomas GJ Jr.
A raman spectroscopic study of hen egg
yolk phosvitin: Structures in solution and
in the solid state. Biochemistry 25:2792-8,
1986.
51. Crossley M, Huq NL, Kirszbaum L,
Reynolds EC. Purification and N-terminal
sequence analysis of bovine dentine phos-
phophoryn (abstract). J Dent Res 75(Spec
Iss):154, 1996.
52. Sorensen ES, Petersen TS. Identification of
two phosphorylation motifs in bovine
osteopontin. Biochem Biophys Res
Commun 198:200-5.1994.
53. Price PA, Rice IS, Williamson MK.
Conserved phosphorylation of serines in
the Ser-X-Glu/Ser(P)sequences of the
Vitamin K-dependent matrix Gla protein
from shark, lamb, rat, cow and human.
Prot Sci 3:822-30, 1994.
54. Cookson DJ, Levine BA, Williams RJP,
Jontell M, Linde A, Bernard B. Cation
binding by the rat-incisor-dentine phos-
phoprotein. A spectroscopic investiga-
tion. Eur J Biochem 110:273-8,1980.
55. Humphrey RS, Jolley KW. 31P-NMR
studies of bovine P-casein. Biochim
Biophys Acta 208:294-9,1982.
16 SCD Special Care in Dentistry, Vol 18 No 1 1998
56. Huq NL, Cross KJ, Reynolds EC. A 1H
NMR study of the casein phosphopeptide
(w5,-casein(59-79).Biochim Biophys Acta
1247201-8,1995.
57. Chaplin LC, Clark DC, Smith LJ. The sec-
ondary structure of peptides derived
from caseins: a circular dichroism study.
Biochim Biophys Acta 956~162-72,1988.
58. Wahlgren NM, Leonil J, Dejmek P,
Drakenberg T. Two-dimensional nuclear
magnetic resonance study of the @-casein
peptide 1-25:Resonance assignments and
secondary structure. Biochim Biophys
Acta 1202:121-8,1993.
59. Tsuda S, Niki R, Kuwata T, Tanaka I,
Hikichi K. Proton NMR study of casein
phosphopeptide (1-25):Assignment and
conformation. Magnet Res Chem 28:1097-
102,1991.
60. Furedi-Milhofe H, Moradian-Oldak J,
Weiner S, Vies A, Minz K, Addadi L.
Interactions of matrix proteins from min-
eralized tissues with octacalcium phos-
phate. Connect Tissue Res 30:251-64,1994.
61. Craig GG. The use of a calcium sucrose
phosphates-calcium orthophosphate com-
plex as a cariostatic agent. Br Dent J 725-
8,1975.