Archives of Oral Biology 45 (2000) 277±291

www.elsevier.com/locate/archoralbio

In situ studies of pellicle formation on hydroxyapatite discs

A.M. Vacca Smith*, W.H. Bowen
Center for Oral Biology, University of Rochester, 601 Elmwood Avenue, Box 611, Rochester, NY 14642, USA

Accepted 21 October 1999

Abstract

The formation of acquired enamel pellicle on hydroxyapatite (HA) discs of known surface area carried in the
mouth was studied; discs were carried in the mouth for 30 s, 1, 5, 10 and 20 min. Similar amounts of protein were
found on the discs at each time-point, as determined by ninhydrin analyses. The amounts of amylase and lysozyme
detected remained stable after 5 min of exposure of the discs to the mouth. Assay of the discs for fructosyl- and
glucosyltransferase activities revealed that fructosyltransferase activity increased up to 1 min of exposure to the
mouth and decreased when kept in the mouth for longer periods; glucosyltransferase activity, in contrast, increased
the longer the discs were kept in the mouth. This in situ model provides insight into the activities of various
enzymes during the ®rst 20 min of pellicle formation. The e€ects of rinsing with sucrose and sugar alcohols on
pellicle formation on the discs were also explored. The discs were placed in the mouth for 30 s, 1, 5, 10 and 20 min,
preceded by rinsing with either distilled deionized water, sucrose, sorbitol, xylitol or phosphate-bu€ered saline.
Western blot analyses of disc eluates with antiserum/antibody preparations to various salivary components revealed
distinct patterns of deposition of bacterial and salivary components depending on the composition of the rinse.
These studies con®rm that salivary molecules and bacteria are deposited on apatitic surfaces in a selective manner
and reveal that pellicle formation may be in¯uenced by composition of diet. It is apparent that this in situ model
could be used in screening potential antiplaque agents. 7 2000 Elsevier Science Ltd. All rights reserved.

Keywords: Pellicle; Hydroxyapatite; Discs; Saliva; Glucosyltransferase; Enzymes

1. Introduction. the conformation of micelles (RoÈlla and Rykke, 1994;

The acquired enamel pellicle (Dawes et al., 1963) is
formed by the selective adsorption of salivary and bac-
terial-derived molecules on to tooth and apatitic sur-
faces (Hay, 1967, 1973, 1975; Moreno et al., 1978,
1984, 1987; RoÈlla et al., 1983). Salivary molecules
adsorbed on to tooth surfaces may apparently assume
* Corresponding author. Tel.: +1-716-275-5925; fax: +1-
716-473-2679.
E-mail address: Anne_Vaccasmith@urmc.rochester.edu
(A.M. Vacca Smith).
Rykke et al., 1995). The salivary pellicle serves many
functions, which include lubrication (Tabak et al.,
1982), prevention of crystal growth (Hay et al., 1979)
and modulation of the microbial ¯ora on the tooth
surface (Tellefson and Germaine, 1986; Cowan et al.,
1987; Gibbons and Hay, 1988, 1989; Gibbons et al.,
1990; Schilling and Bowen, 1992; Venkitaraman et al.,
1995).
Several active enzymes have been identi®ed in sali-
vary pellicles formed in vitro, and include, for
example, a-amylase (RoÈlla et al., 1983; Vacca-Smith et
al., 1996a), lysozyme (RoÈlla et al., 1983), and bacterial
glucosyltransferase and fructosyltransferase (RoÈlla et

0003-9969/00/$ - see front matter 7 2000 Elsevier Science Ltd. All rights reserved.
PII: S 0 0 0 3 - 9 9 6 9 ( 9 9 ) 0 0 1 4 1 - 7
278 A.M. Vacca Smith, W.H. Bowen / Archives of Oral Biology 45 (2000) 277±291
Nomenclature
CCP cysteine-containing peptide
PRP proline-rich protein
PVDF polyvinylenedi¯uoride
al., 1983; Vacca-Smith et al., 1996b). Enzymatically
active amylase and glucosyltransferase have also been
detected in pellicles formed in vivo (Scheie et al., 1987;
Al-Hashimi and Levine, 1989).
Techniques employed to study the acquired enamel
pellicle include, for example, light microscopy (Arm-
strong and Hayward, 1968; Lic, 1975), histochemical
staining (Meckel, 1965), amino acid analyses (Mayhall,
1970, 1977; SoÈnju and RoÈlla, 1973; SoÈnju et al., 1974;
Al-Hashimi and Levine, 1989; Lamkin et al., 1991),
electrophoresis (Fisher et al., 1987), chromatographic
and immunological methods (Eggen and RoÈlla, 1984;
Kousvelari et al., 1980; Al-Hashimi and Levine, 1989)
and enzymatic methods (RoÈlla et al., 1983; Scheie et
al., 1987; Vacca-Smith et al., 1996a, 1996b). Results
from these studies provided details about the compo-
sition and activity of the pellicle formed for 2 h or
longer. Information on the phenomena associated with
the very earliest formation of pellicle is sparse. There-
fore, one purpose of this study was to examine the for-
mation of the acquired enamel pellicle in situ, with
particular emphasis on enzyme activities, during the
®rst 20 min of development to determine the steps
involved in its formation from the very earliest stages.
Sucrose is utilized as a substrate by glucosyltransfer-
ase enzymes to form glucan, which in turn provides
bacterial binding sites on tooth surfaces (RoÈlla et al.,
1983; Scheie et al., 1987; Schilling and Bowen, 1988,
1992; Venkitaraman et al., 1995), may in¯uence pellicle
and plaque formation. Glucan production by mutans
streptococci has been shown to be an essential viru-
lence factor for their role in the pathogenesis of dental
caries (Yamashita et al., 1993). Therefore, a second
aim of this study was to determine the e€ects of
sucrose and also explore the e€ects of sugar substitutes
which do not serve as substrates for glucosyltransfer-
ase, on pellicle formation.
2. Materials and methods
2.1. Hydroxyapatite discs
Hydroxyapatite discs with a surface area of
3.06 2 .28 cm2 were custom made by Clarkson Chro-
matography, South Williamsport, PA. Upon receipt of
the discs, they were further sintered at 1200±16008F
for 3 h.
SDS±PAGE sodium dodecyl sulphate±polyacryl-
amide gel electrophoresis
2.2. Formation of pellicle
Before all the assays, the participant, a healthy 33
year old female, refrained from food and tooth brush-
ing for 2 h. For all assays, two discs were held in a
separate side of her buccal sulcus for de®ned periods.
All time-points per assay were collected on the same
day, and all assays were initiated at the same time
each day.
2.3. E€ects of sucrose and sugar alcohol rinses on
pellicle formation
The volunteer rinsed her mouth with 30 ml of dis-
tilled, deionized H2O water for 30 s ®nally spitting it
out. Two hydroxyapatite discs were then held in the
mouth in separate areas of the buccal sulcus for var-
ious times up to 20 min after the water rinse. The discs
were removed from the mouth and washed three times
with distilled, deionized H2O, as described earlier. In
some experiments, the participant rinsed her mouth
with 30 ml of either 29 mM sucrose (Sigma), 55 mM
sorbitol (Sigma), 66 mM xylitol (Sigma), or phosphate-
bu€ered saline (pH 7.5) instead of water for 30 s.
After spitting out the rinse, fresh discs were placed in
the mouth. The sucrose, sugar-alcohols, and phos-
phate-bu€ered saline were without e€ect on salivary
pH after rinsing with the selected agent (data not
shown). All rinses were performed on separate days.
2.4. Western blots
Experiments were performed to identify the com-
ponents of saliva which adhered to the discs. The discs
were carried in the mouth after various rinses as
described earlier and were washed. They were then
placed in 500 ml of SDS±PAGE sample bu€er
(Laemmli, 1970) for 5 min. After incubation, 20 ml of
the solution, which contained material eluted from the
discs, was removed, mixed with equal volumes of dis-
tilled, deionized water, loaded on to a PAGE gel
(10%) and separated by electrophoresis (Laemmli,
1970). To ensure that all protein was removed from
the discs by the SDS±PAGE sample bu€er, ninhydrin
analyses after wet ashing (a procedure which had no
adverse e€ect on the ninhydrin assay) were performed
on the discs directly (Moore, 1968). The gels were
 A.M. Vacca Smith, W.H. Bowen / Archives of Oral Biology 45 (2000) 277±291
transferred to Immobilon-P membranes (Millipore,
Bedford, MA) and were probed with appropriate anti-
serum/antibody according to the method of Towbin et
al. (1979). The blots were developed with the BCIP/
NBT phosphatase substrate system kit according to
the manufacturer's instructions (Kirkegaard and Perry
Laboratories, Gaithersburg, MD).
2.5. Antisera and antibodies
Polyclonal serum was obtained from rabbits immu-
nized with clari®ed whole human saliva, parotid saliva,
CCP-1 and CCP-3, and statherin. Antiserum rasied
against whole human saliva and parotid saliva was
tested and recognized various components in each type
of saliva, while antiserum raised to CCP-1 or CCP-3
reacted only with the target antigen(s). Polyclonal
serum from a goat immunized with histatin was also
used, as were murine monoclonal antibodies to mucin
glycoprotein I and II, proline-rich proteins, cystatin S
and cystain SN. These antisera and antibodies were
found to react only with target antigen(s). The antisera
to CCP-1, CCP-3, statherin and histatin, as well as the
monoclonal antibodies, were kindly provided by Dr
Lawrence Tabak, Center for Oral Biology, University
of Rochester. Commercially obtained antibodies
included antibody to lysozyme, amylase, carbonic
anhydrase I and II (The Binding Site, Birmingham,
UK), and lactoferrin (Cappel, Organanon Teknika,
West Chester, PA). Speci®city of the commercially
obtained anstisera/antibodies was warranted by the
manufacturer and was con®rmed in our laboratory. To
identity bacterial constituents, antiserum from rabbits
immunized with glucosyltransferase enzymes of Strep-
tococcus mutans GS-5, which were adsorbed on to
hydroxyapatite beads, and culture supernatant-¯uid
antigens of S. mutans GS-5 were used. These sera were
found to react only with glucosyl transferase of S.
mutans or with antigens obtained from culture super-
natant ¯uids of S. mutans. Secondary antiserum used
included goat antirabbit; rabbit antigoat; and goat
antimouse IgG±alkaline phosphatase conjugate
(Sigma).
2.6. Total protein adsorbed on to hydroxyapatite discs
with time
Discs were held in the buccal sulcus for 0.5, 1, 5, 10
and 20 min after the volunteer had rinsed and spat out
30 ml of distilled deionized water, 29 mM sucrose
(Sigma, 66 mM xylitol (Sigma), 55 mM sorbitol
(Sigma) or phosphate-bu€ered saline (pH 7.5). Ap-
proximately one week (7 days) elapsed between each
rinse. After removal from the mouth, the discs were
dip-washed three times in 40 ml of distilled, deionized
279
water and were then placed in 1.0 ml of distilled, deio-
nized water. The discs were then placed in an aquaso-
nic water bath, and after a 5 min sonication to remove
adsorbed materials, the eluates from the discs were
transferred to acid-washed Pyrex tubes. The total pro-
tein in the eluates, which is material removed from the
discs by sonication, was determined by the ninhydrin
assay following acid hydrolysis, using glycine (Sigma)
as a standard (Moore, 1968). Again, to ensure that all
protein was removed from the discs by the sonication,
the discs themselves were subject to ninhydrin analyses
following wet ashing.
2.7. Amylase activity on hydroxyapatite discs held in the
mouth for various periods
Discs were held in the buccal sulcus for 0.5, 1, 5, 10
and 20 min. They were removed from the mouth
gently, washed for three dips in distilled, deionized
water (40 ml). The discs were then exposed to a 1%
soluble starch (Sigma) solution (in 10 mmol/l CaCl2,
pH 7.0) for 30 min at 378C. After incubation with
gentle rocking, the liquid contents of each sample were
transferred to separate vials and the quantities of redu-
cing sugars in the sample, which were released from
starch by the amylase bound to the discs, were deter-
mined by the dinitrosalicylic acid assay of Bernfeld
(1955) using maltose as a standard. Controls included
discs that were exposed to the mouth, followed by ex-
posure to 10 mmol/l CaCl2 (no starch) and discs that
were dipped in water and incubated with the soluble
starch solution (no salivary components).
2.8. Lysozyme activity on discs held in the mouth
Discs were held in the mouth for up to 5 min,
removed and washed as outlined in the amylase assay
methods. They were then place in a suspension of
Micrococcus lysodiekticus (Sigma), starting OD700 =
0.400, for 30 min at 378C, so that the lysozyme present
on the discs could degrade the bacterial substrate.
After incubation, the optical density of the liquid
phase of each sample was determined (Germaine and
Tellefson, 1986). The amount of lysozyme activity was
calculated from standard curves constructed from the
results obtained by exposing the M. lysodiekticus to
various concentrations of hen egg-white lysozyme
(Sigma; speci®c activity 70,000 units/mg of protein).
2.9. Glucosyl- and fructosyltransferase activities on discs
held in the mouth
Discs were held in the buccal sulcus and were
removed from the mouth after intervals as already
described, washed as before, and then exposed to ‰14 CŠ
280 A.M. Vacca Smith, W.H. Bowen / Archives of Oral Biology 45 (2000) 277±291

glucosylsucrose or ‰3 H]-fructosylsucrose [®nal concen- were prepared by ®xation in gluteraldehyde, dehy-

tration, 100 mmol/l sucrose, 40 mmol/l dextran 9,000 in
bu€ered KCl (50 mmol/l KCl, 1 mmol/l CaCl2, 0.1
mmol/l MgCl2, 0.35 mmol/l K2HPO4, 0.65 mmol/l
KH2PO4, pH 6.5)] for 4 h at 378C. The discs were
washed with bu€ered KCl, exposed to scintillation
¯uid, and the amount of glucan or fructan formed on
the discs was determined directly by liquid scintillation
spectrometry (Schilling and Bowen, 1988).
dration with ethanol, dried and sputtered with gold
palladium (McCormack et al., 1991; Schilling and
Bowen, 1992; Venkitaraman et al., 1995).
3. Results
3.1. Western blots

2.10. Scanning electron microscopy of pellicle formed
under various experimental conditions
Discs carried in the mouth were examined by scan-
ning electron microscopy with a JEOL 3820. Discs
Western blots were performed to determine the
e€ects of various rinses on the composition of the pel-
licle formed on discs based on the number of immuno-
reactive bands detected in disc eluates. The data
presented here are therefore based on the number of

Fig. 1. Western blots of disc eluates. Discs were held in the buccal sulcus for 0.5, 5, 10, and 20 min after rinsing with either distilled
water (A), sucrose (B), sorbitol (C), xylitol (D), or phosphate-bu€ered saline (E) and were treated with SDS±PAGE sample bu€er.
The samples were electrophoresed and transfered to PVDF membranes and probed with antiserum to CCP-1. In each blot, the ®rst
lane (left to right) is the eluate at 0.5 min, lane 2 is the eluate at 5 min, lane 3 is the eluate at 10 min, and lane 4 is the eluate at 20
min. Molecular weight markers indicated with arrows are (from top to bottom): 197 kDa, 117 kDa, 89 kDa, and 52 kDa.
 A.M. Vacca Smith, W.H. Bowen / Archives of Oral Biology 45 (2000) 277±291 281

immunoreactive bands only; band density was not
measured. The results from Western bloting con®rm
the extraordinary rapidity with which proteins are
adsorbed on to hydroxyapatite in vivo. In some cases,
several bands were detected by antiserum/antibodies in
the vicinity of the putative antigen. These multiple
bands are possibly the result of proteolytic breakdown
of the target molecule, or cross-reactivity with the anti-
serum/antibody preparation (with isoforms of the tar-
get molecule), or both. Fig. 1 shows Western blots of
extracts from discs taken after each rinse, and probed
with the same antiserum, anti-CCP-1 antiserum.
The results from Western blots on eluates from discs
carried in the mouth after a water rinse are shown in
Table 1. Ninhydrin analyses of the discs, after treat-
ment with SDS±PAGE sample bu€er, revealed that
detectable amounts of protein did not remain on the
discs. When probed with antiserum to whole human
saliva components, 10 bands appeared from material
obtained from discs that had been in the mouth for as
little as 30 s after a water rinse; the number of bands
detected increased with exposure to the mouth and did
not increase after 10 min of exposure to the mouth.
Probing blots with antiserum to parotid saliva com-
ponents revealed just seven bands at 30 s of exposure
to the mouth after a water rinse, and these bands
appeared in common with seven of the bands detected
by the antiserum to whole human saliva components.
There were 13 bands of proteins detected in the disc
eluates at 10 min of exposure, after which no ad-
ditional proteins were detected by this assay (Table 1).
Lysozyme, lactoferrin and amylase were detected by
appropriate antibodies in Western blots of eluates
from discs exposed to the mouth for as little as 30 s
after a water rinse (Table 1), an observation consistent
with the enzyme assays performed directly on the
discs. The presence of several bands corresponding to
lactoferrin and amylase suggests the presence of several
isoforms of these molecules in the disc eluates. Carbo-
nic anhydrase II was detected at 30 s of exposure to
the mouth after a water rinse (Table 1), and a second
isozyme of carbonic anhydrase II was detected after 10
min of exposure to the mouth after a water rinse. Car-
bonic anhydrase I was detected only in eluates of discs
carried for 20 min in the mouth (Table 1) after a water
rinse. Mucin glycoproteins were not detected in the
disc eluates until 1 min of exposure to the mouth after
the water rinse (Table 1), and several isoforms of the
cystatins, histatin, statherin and cysteine-containing
peptides were present in material obtained from discs
carried in the mouth for as short as 30 s after a water
rinse (Table 1). CCP-1 seemed to disappear with time,
as seen in Fig. 1(A).
Antiserum to immobilized streptococcal glucosyl-
transferase enzymes revealed a single band on Western
blots obtained from material obtained from discs car-
ried in the mouth for 1 min after a water rinse
(Table 1). When probed with antiserum to S. mutans
GS-5 supernatant-¯uid antigens, two bands appeared
from material obtained from discs held in the mouth
for 5 min after a water rinse (Table 1), once again
demonstrating the rapid appearance of bacterial com-
ponents, in particular glucosyltransferase, in the sali-
vary pellicle.

Table 1
Number of bands detected by Western blot of proteins extracted from hydroxyapatite discs after a water rinsea

Antiserum 30 s 1 min 5 min 10 min 20 min

Whole human saliva 10 11 13 15 15

Parotid saliva 7 10 11 13 13
Glucosyl transferase 0 1 1 1 1
S. mutans GS-5 antigens 0 0 2 2 2
Lysozyme 1 1 1 1 1
Lactoferrin 4 4 4 4 4
Amylase 4 4 4 4 4
Carbonic anhydrase I 0 0 0 0 1
Carbonic anhydrase II 1 1 1 2 2
Mucin glycoprotein I and II 0 1 1 1 1
PRP 0 0 0 0 0
Cystatin S 1 1 1 1 1
Cystatin SN 1 1 1 1 1
Histatin 2 2 2 2 2
Statherin 4 4 4 4 4
CCP-1 2 2 2 2 2
CCP-3 2 2 2 2 2

a
Hydroxyapatite discs were carried in the mouth for 30 s, 1, 5, 10 or 20 min prior to analysis by Western blot.
282 A.M. Vacca Smith, W.H. Bowen / Archives of Oral Biology 45 (2000) 277±291

Rinsing the mouth with sweeteners had a profound
e€ect on pellicle formation as revealed by Western
blots. Results obtained in Western blots of eluates
from discs carried in the mouth after a sucrose rinse
di€ered from those of discs carried after a water rinse,
as seen in Table 2. Antiserum to whole human saliva
or to parotid saliva revealed the same number of
bands on Western blots of material obtained from
discs exposed to the mouth for up to 20 min after a
sucrose rinse (Table 2), which is in contrast to that
observed from discs following a water rinse where the
number of bands detected increased with increased ex-
posure to the mouth. Mucin glycoproteins were not
detected in eluates from discs carried in the mouth
after a sucrose rinse (Table 2), which is in contrast to
data obtained in the water-rinse study (Table 1). Pro-
line-rich proteins and cystatin S and cystatin SN were
detected in Western blots of eluates of discs carried for
30 s after the sucrose rinse but were not detected fol-
lowing longer exposure to the mouth after rinsing with
sucrose (Table 2). These data are in contrast to those
observed after a water rinse (Table 2), in which pro-
line-rich proteins were not detected at all, and cystatins
were detected for periods up to 20 min after exposure
to the mouth. When blots were probed with antiserum
to streptococcal glucosyltransferases and to S. mutans
GS-5 supernatant-¯uid antigens, more bands were
detected more rapidly after a sucrose rinse than after a
water rinse (Tables 1 and 2). The appearance of iso-
forms of CCP-1 seemed to decrease with increase in
exposure to the mouth after a sucrose rinse (Fig. 1B).
Western blots or eluates taken from discs carried in
the mouth after a sorbitol rinse probed with antiserum
to whole human saliva components and parotid saliva
did not show an increase in the number of proteins
deposited on the discs with time (Table 3). The mucin
glycoproteins were not detected in eluates from discs
carried in the mouth after a sorbitol rinse, which is
similar to that observed following a sucrose rinse
(Table 3). Among the cystatins, only cystatin S was
detected in disc eluates carried in the mouth after a
sorbitol rinse, and this molecule was detected in ma-
terial obtained from discs carried in the mouth for 30 s
after rinsing (Table 3). Rinsing with sorbitol resulted
in an increase of deposition of components reactive
with antiserum to S. mutans GS-5 supernatant-¯uid
antigens (Table 3) compared with the number of com-
ponents detected after rinsing with either water or
sucrose (Tables 1 and 2). Four isoforms were detected
by antiserum to CCP-1 (Fig. 1C) in contrast to Wes-
tern blots of disc eluates obtained from discs carried in
the mouth after water and sucrose rinses.
The e€ects of a xylitol rinse on the composition of
the salivary pellicle are shown in Table 4. Lysozyme,
amylase, the carbonic anhydrases, mucins and proline-
rich proteins were not detected in any of the disc elu-
ates, even after the discs were carried in the mouth for
up to 20 min after a xylitol rinse. As observed with the
sucrose and sorbitol rinses, a constant number of
bands was detected by antiserum to whole human
saliva components and by antiserum to parotid saliva
components (Table 4) in material obtained from each
disc carried up to 20 min after a xylitol rinse. Gluco-
syltransferase and culture supernatant-¯uid antigens of

Table 2
Number of bands detected by Western blot of proteins extracted from hydroxyapatite discs after a sucrose rinsea

Antiserum 30 s 1 min 5 min 10 min 20 min

Whole human saliva 14 14 14 14 14

Parotid saliva 6 6 6 6 6
Glucosyltransferase 1 1 1 1 1
S. mutans GS-5 antigens 2 2 2 2 2
Lysozyme 1 1 1 1 1
Lactoferrin 4 4 4 4 4
Amylase 4 4 4 4 4
Carbonic anhydrase I 0 0 0 0 0
Carbonic anhydrase II 2 2 2 2 2
Mucin glycoprotein I and II 0 0 0 0 0
PRP 1 1 0 0 0
Cystatin S 1 0 0 0 0
Cystatin SN 1 0 0 0 0
Histatin 2 2 2 2 2
Statherin 2 2 2 2 2
CCP-1 2 2 2 2 2
CCP-3 2 2 2 1 1

a
Hydroxyapatite discs were carried in the mouth for 30 s, 1, 5, 10 or 20 min prior to analysis by Western blot.
 A.M. Vacca Smith, W.H. Bowen / Archives of Oral Biology 45 (2000) 277±291 283

Table 3
Number of bands detected by Western blot of proteins extracted from hydroxyapatite discs after a sorbitol rinsea

Antiserum 30 s 1 min 5 min 10 min 20 min

Whole human saliva 15 15 15 15 15

Parotid saliva 10 10 10 10 10
Glucosyltransferase 1 1 1 1 1
S. mutans GS-5 antigens 8 6 6 5 5
Lysozyme 1 1 1 1 1
Lactoferrin 1 1 1 1 1
Amylase 1 1 1 1 1
Carbonic anhydrase I 2 2 2 2 2
Carbonic anhydrase II 3 1 1 1 1
Mucin glycoprotein I and II 0 0 0 0 0
PRP 9 9 9 9 9
Cystatin S 1 0 0 0 0
Cystatin SN 0 0 0 0 0
Histatin 2 2 2 2 2
Statherin 13 13 16 16 16
CCP-1 4 4 4 4 4
CCP-3 9 7 7 7 7

a
Hydroxyapatite discs were carried in the mouth for 30 s, 1, 5, 10 or 20 min prior to analysis by Western blot.

S. mutans were detected in the eluates obtained from
discs carried in the mouth for a little as 30 s after a
xylitol rinse (Table 4). Isoforms of lactoferrin and his-
tatin were not detected eluates from discs carried in
the mouth for less than 1 min after rinsing with xylitol
(Table 4). The appearance of isoforms of CCP-1
seemed to decrease with increased exposure to the
mouth (Fig. 1D).
The in¯uence of a phosphate-bu€ered saline rinse
on pellicle formation is shown in Table 5. Lysozyme,
amylase, carbonic anhydrase II, and mucin glyco-
proteins were not detected in Western blots of eluates
obtained from discs carried in the mouth after that
rinse (Table 5). However, activity assays demon-
strated that these enzymes were present on the discs
in small quantities only (data not shown), which
might partially explain why they were not detected by
Western blots. The appearance of isoforms of CCP-1

Table 4
Number of bands detected by Western blot of proteins extracted from hydroxyapatite discs after a xylitol rinsea

Antiserum 30 s 1 min 5 min 10 min 20 min

Whole human saliva 9 9 9 9 9

Parotid saliva 11 11 11 11 11
Glucosyl transferase 4 4 4 4 4
S. mutans GS-5 antigens 3 3 3 3 3
Lysozyme 0 0 0 0 0
Lactoferrin 0 2 2 2 2
Amylase 0 0 0 0 0
Carbonic anhydrase I 0 0 0 0 0
Carbonic anhydrase II 0 0 0 0 0
Mucin glycoprotein I and II 0 0 0 0 0
PRP 0 0 0 0 0
Cystatin S 3 3 3 3 3
Cystatin SN 3 3 3 3 3
Histatin 0 2 2 2 2
Statherin 3 8 8 9 9
CCP-1 3 5 5 5 5
CCP-3 2 2 2 2 2

a
Hydroxyapatite discs were carried in the mouth for 30 s, 1, 5, 10 or 20 min prior to analysis by Western blot.
284 A.M. Vacca Smith, W.H. Bowen / Archives of Oral Biology 45 (2000) 277±291

Table 5
Number of bands detected by Western blot analysis of proteins extracted from hydroxyapatite discs after a phosphate-bu€ered
saline rinsea

Antiserum 30 s 1 min 5 min 10 min 20 min

Whole human saliva 15 15 15 15 15

Parotid saliva 19 19 19 19 19
Glucosyl transferase 4 4 4 4 4
S. mutans GS-5 antigens 4 3 2 1 1
Lysozyme 0 0 0 0 0
Lactoferrin 1 2 2 2 2
Amylase 0 0 0 0 0
Carbonic anhydrase I 1 1 1 1 1
Carbonic anhydrase II 0 0 0 0 0
Mucin glycoprotein I and II 0 0 0 0 0
PRP 0 0 0 0 0
Cystatin S 1 1 0 0 0
Cystatin SN 3 3 2 2 2
Histatin 1 1 1 1 1
Statherin 9 9 9 9 9
CCP-1 4 4 1 1 1
CCP-3 3 4 2 2 2

a
Hydroxyapatite discs were carried in the mouth for 30 s, 1, 5, 10 or 20 min prior to analysis by Western blot.

decreased with exposure to the discs to the mouth

(Fig. 1E).

3.2. Total protein adsorbed on to hydroxyapatite discs

with time

All of the discs used here had the same surface area,
so we were therefore able to measure amounts or ac-
tivities of adsorbed material and compare data from
one rinse with data from other rinses. The amount of
protein adsorbed on to the discs after being exposed to
the mouth for various periods was determined by the
ninhydrin method (Moore, 1968). The results, as
shown in Fig. 2, indicate that the discs became satu-
rated with protein very rapidly after each rinse,
although less protein was adsorbed on to the discs
after the sorbitol rinse than was adsorbed on to the
discs after the other rinses. Ninhydrin analyses of the
discs themselves revealed that all protein had been
removed from the discs (data not shown).

3.3. Amylase activity on hydroxyapatite discs held in the

mouth for various periods
The amount of reducing sugar released from starch
by amylase bound to the discs carried in the mouth
was determined. The results indicate that saturation of
the discs by amylase occurred at about 5 min following
their placement in the mouth (Fig. 3).
Fig. 2. Total protein adsorbed on to the discs. Discs were
held in the buccal sulcus for 0.5, 1, 5, 10 and 20 min after rin-
sing with either distilled, deionized water, sucrose, sorbitol,
xylitol or phosphate-bu€ered saline. Total protein adsorbed
on to the discs was determined by ninhydrin analyses follow-
ing acid hydrolysis (Moore, 1968). The data shown are the
means (mg of protein/disc) for two discs and represent two ex-
periments.
 A.M. Vacca Smith, W.H. Bowen / Archives of Oral Biology 45 (2000) 277±291
Fig. 3. The activity of salivary amylase on the discs. Discs
were held in the buccal sulcus for 0.5, 1, 5, 10 and 20 min.
Amylase activity was determined by the dinitrosalicylic acid
assay of Bernfeld (1955). The data shown are the means for
four discs, expressed as mg/ml of reducing sugar released/disc.
3.4. Lysozyme activity on discs held in the mouth
The activity of lysozyme on the discs held in the
mouth for various periods was also determined. The
results, as shown in Fig. 4, suggest that the maximum
adsorption of lysozyme also occurred within 5 min of
exposure to the mouth.
Fig. 4. The activity of salivary lysozyme on the discs. Discs
were held in the buccal sulcus for 0.5, 1, 5, and 10 min, and
were assayed for lysozyme activity. The data shown are the
means for four discs, expressed as units of lysozyme/disc.
285
Fig. 5. The activities of glucosyl- and fructosyltransferase on
the discs. Discs were held in the buccal sulcus for 0.5, 1, 5, 10
and 20 min, and then assayed directly for glucosyl- and fruc-
tosyltransferase activity. The data shown are the means for
four discs, expressed as mmol polymer formed/disc.
3.5. Glucosyl- and fructosyltransferase activities on discs
held in the mouth
These activities were detected on the discs kept in
the mouth for as little as 30 s. The pattern for adsorp-
tion and activity is quite distinct for these enzymes.
The results, as shown in Fig. 5, revealed an increase in
glucosyltransferase activity with increased exposure to
the mouth. In contrast, fructosyltransferase activity
increased up to 1 min of exposure to the mouth and
decreased with further exposure (Fig. 5).
Fig. 6. Scanning electron micrographs of the control disc.
Scanning electron micrograph of a hydroxyapatite disc that
had not been placed in the mouth.
286 A.M. Vacca Smith, W.H. Bowen / Archives of Oral Biology 45 (2000) 277±291

Fig. 7. Scanning electron micrographs of the discs after water rinse. Discs were carried in the buccal sulcus for 30 s, 5 min and 20
min after rinsing with water. (a) Disc kept in the mouth for 30 s; (b) and (c) Disc kept in the mouth for 5 min; (d), (e) and (f)
Discs kept in the mouth for 20 min.
 A.M. Vacca Smith, W.H. Bowen / Archives of Oral Biology 45 (2000) 277±291 287

Fig. 8. Scanning electron micrographs of the discs after sucrose rinse. Discs were carried in the buccal sulcus for 30 s, 5 min and 20
min after rinsing with 1% sucrose. (a) A disc held in the mouth for 30 s; (b) a disc kept in the mouth for 5 min; (c) discs carried in
the mouth for 20 min.
288 A.M. Vacca Smith, W.H. Bowen / Archives of Oral Biology 45 (2000) 277±291
3.6. Scanning electron microscopy
The discs were also examined by scanning electron
microscopy. Fig. 6 shows the discs before exposure to
the mouth. At 30 s of incubation in the mouth after a
water rinse, long string-like structures of undetermined
origin, along with some amorphous material, appeared
on the discs, as seen in Fig. 7(a), and cocci and
micelle-like components appeared along with the
string-like structures and amorphous material after 5
min of exposure, as seen in Figs. 7(b) and (c). After 20
min of exposure, the cocci, micelle-like components
and amorphous material remained, while the string-
like structures were apparently replaced with short
rods (Fig. 7d±f).
Rinsing with sucrose gave rise to a distinctive pat-
tern of adsorption. (Fig. 8a±e). After 30 s of exposure
to the mouth, a mesh of material thought to be glucan,
cocci and short rods appeared on the discs (Fig. 8a),
and in Fig. 8(b) it can be seen that micelle-like com-
ponents also present with a mesh-like material
appeared at 30 s of exposure. Also, as seen in Fig. 8(c),
long string-like structures appeared at 30 s of exposure
after a sucrose rinse. However, these string-like struc-
tures were free of the mesh-like material. These pat-
terns were seen consistently over the time-span of
exposure to the mouth, as seen in Fig. 8(d) at 5 min of
exposure, and Fig. 8(e) at 20 min of exposure.
4. Discussion
The studies described here show the pattern of pro-
tein deposition and the activities of several salivary-
and bacterial-derived enzymes on hydroxyapatite discs
held in the mouth for 20 or fewer minutes. Our studies
also reveal that distinct patterns of pellicle formation
occur in vivo, and that these patterns are in¯uenced by
dietary constituents.
Many salivary constituents have been identi®ed in
in vitro-formed pellicles. These components include:
mucin glycoprotein I and II, sIgA, amylase, lysozyme,
proline-rich proteins, statherin, cystatins and histatins
(Ericson, 1967; Bennick et al., 1983; RoÈlla et al., 1983;
Tabak et al., 1985; Fisher et al., 1987; Jensen et al.,
1991; Vacca-Smith et al., 1996a), and bacterial gluco-
syltransferase (RoÈlla et al., 1983; Vacca-Smith et al.,
1996b). However, only a few components, such as
sIgA, lysozyme, amylase, IgG. albumin, ®brin and glu-
cosyltransferase, have been identi®ed in in vivo-formed
pellicles (Scheie et al., 1987; Al-Hashimi and Levine,
1989).
Our studies show that the adsorption of protein
on to the discs rapidly reached saturation, in con-
trast to observations by SoÈnju and RoÈlla (1973),
who showed an increase in the adsorption of ma-
terial on to tooth surfaces over a 60 min period.
The discrepancies between our study and the pre-
vious study could be attributed to the time period;
as SoÈnju and coworkers chose 60 min and we chose
20 min, it is possible that they were observing sec-
ondary adsorption, or adsorption in two, three or
more phases.
In our study it is possible that rinsing with test sol-
utions a€ected the composition of secreted saliva.
However, SDS±PAGE of silver-stained gels of whole
saliva taken before and after each rinse indicated that
the rinses apparently did not alter the composition of
the salivary secretions (data not shown). Our studies
showed that the same amount of protein was present
in material removed form the discs carried in the
mouth for various periods after each rinse, although
less protein was present in disc eluates obtained after
a sorbitol rinse than after any of the other rinses.
Similar patterns of protein deposition on to enamel
surfaces were observed by Kuboki et al. (1987) using
X-ray photoelectron spectroscopy. More recently,
however, studies by Lamkin et al. (1996) revealed
that selected salivary molecules, puri®ed, biotinylated
and reconstituted in saliva, showed distinct patterns
of deposition on to hydroxyapatite surfaces in vitro.
Molecules such as amylase, glycosylated PRG and
cystatin showed a hyperbolic, rapid rise of binding
followed by a levelling o€ and then a decline. Other
molecules such as statherin, PRP-3-4 and PIF-f
bound to the beads immediately and amount did not
increase thereafter, and molecules such as PRP-1-2,
PIF-s and histatin bound slowly, followed by a
period of little change, and then an increase of bind-
ing at 60 min of exposure. These observations, along
with our data, suggest that qualitative, and not
necessarily quantitative, di€erences may exist in the
salivary pellicle at a given time. It is conceivable that
pellicle formation is more dynamic than believed
heretofore and that material which adheres may be
removed, replaced, modi®ed, and have additional ma-
terial adsorbed.
Our data show that the hydroxyapatite discs became
saturated with amylase and lysozyme rapidly, similar
to the results observed by Lamkin et al. (1996).
When discs were assayed directly for glucosyltrans-
ferase activity following a water rinse, a distinct pat-
tern was seen from that observed with lysozyme and
amylase. The enzyme activity on the discs increased
with their increased exposure to the mouth and did
not reach saturation in the time that we studied. A
similar pattern of adsorption was observed by Lam-
kin et al. (1996) with respect to the binding of hista-
tin, PRP-1-2, and PIF-s to apatite. In contrast, when
discs were carried in the mouth for up to 20 min
and assayed periodically for fructosyltransferase ac-
 A.M. Vacca Smith, W.H. Bowen / Archives of Oral Biology 45 (2000) 277±291
tivity, a rapid rise in that activity was seen, followed
by a decline, after which activity reached a steady
level.
Diet may a€ect the composition of the pellicle. A
distinct model of deposition was seen after rinses
with sucrose, sorbitol, xylitol or phosphate-bu€ered
saline that di€ered from that observed following a
rinse with water. In Western blots, distinct patterns
of appearance of salivary and bacterial molecules in
disc eluates were seen after each rinse. It is important
to emphasize, however, that the Western blots were
qualitative, the data presented are based on the num-
ber of immunoreactive bands detected in disc eluates
after each rinse, and band density was not measured.
However, the di€ering patterns of deposition were
supported by results from electron-microscopic stu-
dies.
Collectively, these data suggest that oral exposure
to sucrose and sugar alcohols can a€ect the compo-
sition of the salivary pellicle, which in turn can in¯u-
ence which bacteria and bacterial structures bind to
it. Our data show that the rinses we used did not
themselves a€ect the composition of saliva. Several
possibilities may be considered. Perhaps planktonic
molecules in saliva combine with sucrose or the
sugar alcohols or phosphate-bu€ered saline, which
can alter their anities for apatitic surfaces. Alterna-
tively, the sucrose or the sugar alcohols themselves
may bind to the hydroxyapatite, providing novel sur-
faces. Clearly, sucrose could combine with fructosyl-
and glucosyltransferases of saliva. The anities of
molecules in saliva for these newly formed surfaces
may be di€erent from those for bare hydroxyapatite,
thus causing variations in patterns of protein depo-
sition on discs carried in the mouth after either a
water rinse or a sucrose/sugar alcohol rinse. Current
studies in our laboratory suggest that sucrose does
indeed bind to naked hydroxyapatite discs. Alterna-
tively, glucosyltransferase in the pellicle makes glucan
on the discs after a sucrose rinse, and the molecules
on the discs are masked by the glucan or take on a
di€erent conformation in the presence of the glucan
and are not recognized by the antiserum/antibody
preparations.
Our results reveal a novel model for exploring the
sequence of pellicle formation under realistic con-
ditions and support the concept that speci®c salivary
and bacterial molecules show distinct patterns of de-
position and activities on apatitic surfaces (Lamkin
et al., 1996). Some molecules bind to apatite rapidly
and their concentration and/or activity remains con-
stant, others bind slowly and never reach saturation,
and ®nally, some components bind to apatitic
surfaces very rapidly, yet quickly desorb or lose
activity.
In conclusion, these studies show the selective de-
289
position of salivary components, bacteria, and bac-
terial components on to apatitic surfaces held in the
mouth. These studies also reveal the pattern of pelli-
cle and plaque formation varies on apatitic surfaces
and is in¯uenced by exposure to certain dietary com-
ponents. The model described here could be used to
understand further the processes involved in pellicle
and plaque formation and could also be used to
evaluate the potential e€ectiveness of anti plaque
agents.
Acknowledgements
This study was supported by USPHS grant
DEO77907. We also wish to thank Ms Sandy McCor-
mack for her technical assistance with the scanning
electron microscope.
References
Al-Hashimi, I., Levine, M.J., 1989. Characterization of in
vivo salivary-derived enamel pellicle. Arch. Oral Biol. 34,
289±295.
Armstrong, W.G., Hayward, A.F., 1968. Acquired organic
integuments of human enamel: a comparison of analytical
studies with optical, phase-contrast and electron micro-
scope examinations. Caries Res. 2, 294±305.
Bennick, A., Chau, G., Goodlin, R., Abrams, S., Tustian, D.,
Madapallimattam, G., 1983. The role of human salivary
acidic proline-rich proteins in the formation of acquired
dental pellicle in vivo and their fate after adsorption to the
human enamel surface. Arch. Oral Biol. 28, 19±27.
Bernfeld, P., 1955. Amylases, a and b. Meth. Enzymol. 1,
149±159.
Cowan M.M., Taylor K.G., Doyle R.J., 1987. Energetics of
the initial phase of adhesion of Streptococcus sanguis to
hydroxylapatite. 169, 2995±3000.
Dawes, C., Jenkins, G.N., Tonge, C.H., 1963. The nomencla-
ture of the integuments of the enamel surface of teeth. Br.
Dent. J. 115, 65±68.
Eggen, K.H., RoÈlla, G., 1984. Further information on the
composition of the acquired pellicle. In: Guggenheim, B.
(Ed.), Cariology Today. Karger, Basel, pp. 109±118.
Ericson, T., 1967. Adsorption to hydroxyapatite of proteins
and conjugated proteins from human saliva. Caries Res. 1,
52±58.
Fisher, S.J., Prakobphol, A., Kajisa, L., Murray, P.A., 1987.
External radiolabelling of components of pellicle on
human enamel and cementum. Arch. Oral Biol. 32, 509±
517.
Germaine, G.R., Tellefson, L.M., 1986. Potential role of lyso-
zyme activity of in vitro acquired pellicle against
Streptococcus faecium 9790. Infect. Immun. 54, 846±854.
Gibbons, R.J., Hay, D.I., 1988. Human salivary acidic pro-
line-rich proteins and statherin promote the attachment of
290 A.M. Vacca Smith, W.H. Bowen / Archives of Oral Biology 45 (2000) 277±291
Actinomyces viscosus LY7 to apatitic surfaces. Infect.
Immun. 56, 1988.
Gibbons, R.J., Hay, D.I., 1989. Adsorbed salivary proline-
rich proteins contribute to the adhesion of Streptococcus
mutans JBP to apatitic surfaces. J. Dent. Res. 68, 1303±
1307.
Gibbons, R.J., Hay, D.I., Childs III, W.C., Davis, G., 1990.
Role of cryptic receptors (cryptitopes) in bacterial ad-
hesion to oral surfaces. Arch. Oral Biol. 35, 107s±114s.
Hay, D.I., 1967. The adsorption of salivary proteins by
hydroxyapatite and enamel. Arch. Oral Biol. 12, 937±946.
Hay, D.I., 1973. The interaction of human parotid salivary
proteins with hydroxyapatite. Arch. Oral Biol. 18, 1517±
1529.
Hay, D.I., 1975. Fractionation of human parotid salivary pro-
teins and the isolation of an histidine-rich acidic peptide
which shows high anity for hydroxyapatite surfaces.
Arch. Oral Biol. 20, 553±558.
Hay, D.I., Moreno, E.C., Schlesinger, D.H., 1979.
Phosphoprotein-inhibitors of calcium phosphate precipi-
tation from salivary secretions. Inorg. Persp. Biol. Med. 2,
271±285.
Jensen, J.L., Lamkin, M.S., Troxler, R.F., Oppenheim, F.G.,
1991. Multiple forms of statherin in human salivary se-
cretions. Arch. Oral Biol. 36, 529±534.
Kousvelari, E.E., Baratz, R.S., Burke, B., Oppenheim, F.G.,
1980. Immunochemical identi®cation and determination of
proline-rich proteins in salivary secretions, enamel pellicle,
and glandular tissue specimens. J. Dent. Res. 59, 1430±
1438.
Kuboki, Y., Teraoka, K., Okada, S., 1987. X-ray photo-
electron spectroscopic studies of the adsorption of salivary
constituents on enamel. J. Dent. Res. 66, 1016±1019.
Laemmli, U.K., 1970. Cleavage of structural proteins during
the assembly of the head of bacteriophage T4. Nature 227,
680±685.
Lamkin, M.S., Jensen, J.L., Reza Setayesh, M., Troxler, R.F.,
Oppenheim, F.G., 1991. Salivary cystatin SA-III, a poten-
tial precursor of the acquired enamel pellicle, is phos-
phorylated at both its amino-and carboxyl-terminal
regions. Arch. Biochem. Biophys. 288, 664±670.
Lamkin, M.S., Arancillo, A.A., Oppenheim, F.G., 1996.
Temporal and compositional characteristics of salivary
protein adsorption to hydroxyapatite. J. Dent. Res. 75,
803±808.
Lic T., 1975. Pellicle formation on hydroxyapatite splints
attached to the human dentition: morphologic con®r-
mation of the concept of adsorption. Arch. Oral Biol.,
739±742.
Mayhall, C.W., 1970. Concerning the composition and source
of the acquired enamel pellicle of human teeth. Arch. Oral
Biol. 15, 1327±1341.
Mayhall, C.W., 1977. Amino acid composition of experimen-
tal salivary pellicles. J. Periodontol. 48, 78±91.
McCormack, S.M., Tormo, F.J., Featherstone, J.D.B., 1991.
A straightforward scanning electron microscopy technique
for examining non-metal coated dental hard tissues.
Scanning Microsc. 5, 269±272.
Meckel, A.H., 1965. The formation and properties of organic
®lms on teeth. Arch. Oral Biol. 10, 585±597.
Moore, S., 1968. Amino acid analysis: aqueous dimethyl sulf-
oxide as solvent for the ninhydrin reaction. J. Biol. Chem.
243, 6281±6283.
Moreno, E.C., Kresak, M., Hay, D.I., 1978. Adsorption of
two human parotid salivary macromolecules on hydroxy-,
¯uorhydroxy- and ¯uorapatitcs. Arch. Oral Biol. 23, 525±
533.
Moreno, E.C., Kresak, M., Hay, D.I., 1984. Adsorption of
molecules of biological interest on to hydroxyapatite. Calc.
Tissue Int. 36, 48±59.
Moreno, E.C., Kresak, M., Kane, J.J., Hay, D.I., 1987.
Adsorption of proteins, peptides and organic acids on to
hydroxyapatite. Langmuir 3, 511±519.
RoÈlla, G., Ciardi, J.E., Bowen, W.H., 1983. Identi®cation of
IgA, IgG, lysozyme, albumin, a-amylase and glucosyl-
transferase in the protein layer adsorbed to hydroxyapatite
from whole saliva. Scand. J. Dent. Res. 91, 186±190.
RoÈlla, G., Rykke, M., 1994. Evidence for presence of micelle-
like protein globules in human saliva. Coll. and Surf. B:
Bioint. 3, 177±182.
Rykke, M., Smistad, G., RoÈlla, G., Karlsen, J., 1995. Micelle-
like structures in human saliva. Coll. and Surf B: Bioint.
4, 33±44.
Scheie, A.A., Eggen, K.H., RoÈlla, G., 1987.
Glucosyltransferase activity in human in vivo formed
enamel pellicle and in whole saliva. Scand. J. Dent. Res.
95, 212±215.
Schilling, K.M., Bowen, W.H., 1988. The activity of glucosyl-
transferases adsorbed on to saliva-coated hydroxyapatite.
J. Dent. Res. 67, 2±8.
Schilling, K.M., Bowen, W.H., 1992. Glucans synthesized in
situ in experimental salivary pellicle function as speci®c
binding sites for Streptococcus mutans. Infect. Immun. 60,
284±295.
SoÈnju, T., RoÈlla, G., 1973. Chemical analysis of the acquired
pellicle formed two hours on cleaned human teeth in vivo.
Caries Res. 7, 30±38.
SoÈnju, T., Christensen, T.B., Kornstad, L., RoÈlla, G., 1974.
Electron microscopy, carbohydrate analyses and biological
activities of the proteins adsorbed in two hours to tooth
surfaces in vivo. Caries Res. 8, 113±122.
Tabak, L.A., Levine, M.J., Jain, N.K., Bryan, A.R., Cohen,
R.E., Monte, L.D., Zawacki, S., Nancollas, G.H.,
Slomiany, A., Slomiany, B.L., 1985. Adsorption of human
salivary mucins to hydroxyapatite. Arch. Oral Biol. 30,
423±427.
Tabak, L.A., Levine, M.J., Mandel, I.D., Ellison, S.A., 1982.
Role of salivary mucins in the protection of the oral cav-
ity. J. Oral Pathol. 11, 1±17.
Tellefson, L.M., Germaine, G.R., 1986. Adherence of
Streptococcus sanguis to hydroxyapatite coated with lyso-
zyme and lysozyme-supplemented saliva. Infect. Immun.
51, 750±759.
Towbin, H.T., Stoehelin, T., Gordon, J., 1979.
Electrophoretic transfer of proteins from polyacrylamide
gels to nitrocellulose sheets: procedures and some appli-
cations. PNAS, USA 76, 4350±4354.
Vacca-Smith, A.M., Venkitaraman, A.R., Quivey, R.G.,
Bowen, W.H., 1996a. Interactions of streptococcal gluco-
syltransferases with a-amylase and starch on the surface of
saliva-coated hydroxyapatite. Arch. Oral Biol. 3, 291±298.
Vacca-Smith, A.M., Venkitaraman, A.R., Schilling, K.M.,
 A.M. Vacca Smith, W.H. Bowen / Archives of Oral Biology 45 (2000) 277±291 291

Bowen, W.H., 1996b. Characterization of glucosyltransfer- the surface of hydroxyapatite. J. Dent. Res. 74, 1695±
ase of human saliva adsorbed on to hydroxyapatite sur- 1701.
faces. Caries Res. 30, 354±360. Yamashita, Y., Bowen, W.H., Burne, R.A., Kuramitsu, H.K.,
Venkitaraman, A.R., Vacca-Smith, A.M., Kopec, L.K., 1993. Role of Streptococcus mutans glucosyltransferase
Bowen, W.H., 1995. Characterization of genes in caries induction in the speci®c-pathogen-free rat
glucosyltransferaseB, gtfC, and gtfD in solution and on model. Infect. Immun. 61, 3811±3817.