Microbiology (1995), 141,2729-2738
Printed in Great Britain
Polypeptides associated with tufts of
cell-surface fibrils in an oral Streptococcus
Michael W. Jameson,' Howard F. Jenkinson,lt Kyle Parnell*
and Pauline S. Handley2
Author for correspondence: Howard F. Jenkinson. Tel: +64 3 479 7076. Fax: +64 3 479 0673.
e-mail: [email protected]
1
Department o f Oral
Biology and Oral
Pathology, University of
Otago, PO Box 647,
Dunedin, New Zealand
* School of Biological
Sciences, University o f
Manchester, Oxford Road,
Manchester M13 9PT, UK
Cells of the oral bacterium Streptococcus oralis CN3410 produce lateral tufts of
cell-surface fibrils of two lengths. Treatment of cells with trypsin resulted in
loss of the tufts and release of longer fibrils intact. SDS-PAGE analysis of
trypsin extracts containing fibrils revealed two groups of high molecular mass
polypeptides which were denoted group A (molecular mass 227-246 kDa) and
group B (molecular mass 175208 kDa). Antibodies were raised to these two
groups of trypsin-extracted polypeptides (TEPs) and to purified fibrils, and the
reactivities of the three different antisera were found to be similar both on
nitrocellulose blots of cell-surface polypeptides and in ELISA with whole cells.
Similar patterns of TEPs were obtained from cells of a spontaneously derived
mutant strain, KP34V, which lacked the short fibril components of tufts. Cells
of strain KP34V had similar cell-surface hydrophobicity to strain CN3410 cells,
and adhered to the same extent to parotid salivary pellicle or human buccal
epithelial cells (BECs) as the wild-type cells. Trypsin treatment of strain CN3410
cells abolished their surface hydrophobicity and ability to adhere to BECs, but
did not affect streptococcal cell binding to experimental salivary pellicle.
Antibodies to TEPs or fibrils had no effect on cell adhesion to BECs or salivary
pellicle. The results imply that the short fibril components of tufts are not
involved in the cell adhesion properties tested. It is suggested that the TEPs
are components of long fibrils, but they are not determinants of streptococcal
cell adhesion to pellicle or to epithelial cells.
Keywords : Streptococczrs, cell-surface proteins, fibril tufts on streptococci, adhesion of oral
streptococci
INTRODUCTION
Streptococci are found at most sites within the human oral
cavity (Frandsen e t al., 1991) and are amongst the
predominant bacteria in human dental plaque (Socransky
etal., 1977). Successful colonization is due, at least in part,
to the abilities of streptococcal cells to bind to salivary
components within the pellicle formed on hard surfaces,
to host epithelial cells, and to other oral bacteria
(Kolenbrander & London, 1993; Ofek & Doyle, 1994).
Streptococcal cell-surface proteins that function as
t Present address: Until 14January 1996: Nuffield Department of Clinical
Biochemistry, Institute of Molecular Medicine, University of Oxford, John
Radcliffe Hospital, Headington, Oxford OX3 9DU, UK. Tel: $44 1865
222427. Fax: +44 1865 222431.
Abbreviations: BECs, buccal epithelial cells; TEPs, trypsin-extracted
polypeptides.
0002-0010 0 1995 SGM
adhesins may be considered to fall into two structural
groups. The first group comprises high molecular mass
cell-wall-associated polypeptides, the best-characterized
being the antigen 1/11 polypeptides found on a variety of
streptococcal species and involved in binding of cells to
salivary glycoproteins (Bleiweis et al., 1992; Jenkinson,
1994). Other wall-associated polypeptide adhesins include
antigen B (AgB) and antigen C (AgC) on Streptococct/s
salivarim mediating binding to VeilLonella aopica and host
epithelial cells, respectively (Weerkamp e t al., 1986a), and
the CshA polypeptide of Streptococms gordonii involved in
adherence to, or coaggregation with, the oral bacterium
Actinomyes naeslundii (McNab & Jenkinson, 1992a, 1994).
The second group of proteins comprises cell-surfaceassociated lipoproteins. This group includes SarA
(Jenkinson & Easingwood, 1990; Jenkinson, 1992) and
ScaA (Andersen et al., 1993; Kolenbrander etal., 1994) of
S. gordonii, which are involved in coaggregation with A .
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 19:03:13
2729
M. W. J A M E S O N and O T H E R S
naesltlndii, and SsaB of Streptococcus sanguis (Ganeshkumar
e t al., 1991) and FimA of Streptococcu~parasangz/is(Fenno e t
al., 1989; Oligino & Fives-Taylor, 1993), which are
salivary-pellicle adhesins.
A variety of cell-surface structures are elaborated by
different strains of oral streptococci but their functions in
adhesion generally are not well understood (Handley,
1990). Binding of streptococci t o experimental salivary
pellicle may be influenced by the presence of cell-surface
fibrils (Gibbons e t al., 1983; Morris e t al., 1987) or
fimbriae (Fives-Taylor & Thompson, 1985), and ability to
coaggregate with other oral genera was found to correlate
with surface structure morphology (Handley e t al., 1985).
For strains of Streptococcus crista producing polar tufts of
fibrils (Handley e t al., 1991), electron microscopic evidence suggests that these structures are involved in the
formation
of
' corn-cob '
coaggregates
with
Corynebacteritlm matrucbotii (Lancy e t al., 1980; Mouton e t
al., 1980) o r Fusobacterium nucleatum (Lancy e t al., 1983). A
functional role for surface fibrils in cell adhesion has been
demonstrated unequivocally only in S. salivarius HB,
where glycoprotein adhesin AgB and polypeptide A g C
were shown to be separate classes of cell-surface fibril
(Weerkamp e t al., 1986b).
Handley e t al. (1984) described cells of an oral Streptococcus
(strain CN3410) with the Lancefield group K antigen that
carried lateral tufts of long and short fibrils with further
peritrichous fibrils covering the rest of the cells. This
strain conforms t o the biochemical description of the
species Streptococctls oralis (Kilian e t al., 1989). Previous
studies showed that what appeared to be the long fibril
components of the tufts could be released, morphologically intact, from the surface of S. oralis CN3410 by
incubating the cells with pronase (Handley e t al., 1984;
Hesketh e t al., 1987). To try t o define better the structural
components and function of the fibrils in these tufts, we
have extended previous work by isolating fibrils following
treatment of S. oralis CN3410 cells with trypsin and
characterizing biochemically and immunologically the
fibril-associated polypeptides. This paper describes these
trypsin-extracted high molecular mass polypeptides,
which appear to be components of long fibrils but not
determinants of S. oralis CN3410 cell adhesion t o parotid
salivary pellicle o r t o host epithelial cells.
METHODS
Microbial strains and growth conditions. Bacterial strains
utilized in this study were S. oralis CN3410 (previously denoted
S.sanguis biotype I1 by Hesketh e t al., 1987), S. oralis KP34V (a
variant of CN3410 isolated during the course of this study), S.
crista NCTC 12479 and AK1 (Handley e t al., 1991), A . naeslundii
T14V, BE64, WVU627, W1544, EF1006, T F l l and ATCC
12104, and C. matrucbotii ATCC 14266. Bacteria were grown on
TSBY agar containing trypticase soy broth (BBL) supplemented
with 5 g yeast extract (Difco) 1-' and 15 g agar 1-' at 37 "C under
reduced oxygen concentration. Liquid cultures were grown at
37 "C in screw-capped bottles or tubes as stationary cultures in
TSBY medium or, for streptococci only, TY-glucose medium,
pH 7-5, containing 5 g tryptone (Difco) l-', 5 g yeast extract l-l,
4 g K,HPO, 1-' and 8 g glucose 1-l.
2730
Trypsin treatment of cells. Cells in the late exponential phase of
growth (10 ml culture) at an OD,,, of 1.3 (measured using a
Shimadzu model UV-120 spectrophotometer) (approximately
8 x 1O8 cells ml-l) were harvested by centrifugation (5000 g ,
10 min, 4 "C). The supernatant was discarded, the cells were
washed twice by suspension in distilled water and centrifugation, and were then suspended in 10 mM Tris/HCl, pH 7-0
(0.6 ml), containin trypsin (type I11 from bovine pancreas,
Sigma; 0.2 mg ml ). Suspensions were incubated for 1 h at
37 "C, centrifuged (5000g, 10 min, 4 "C) and the supernatants
were carefully removed and centrifuged as before. In some
experiments trypsin inhibitor (Sigma type I-S; 0.1 mg ml-' final
concentration) was added immediately following incubation of
cells with trypsin. For subsequent separation of components by
electrophoresis, a portion of cell-free trypsin extract (containing
fibrils) was mixed with an equal volume of sample buffer
(0.125 M Tris/HCl, pH 6.8, containing 0.02 % 2-mercaptoethanol and 2 %, w/v, SDS) and heated for 10 min at 70 "C. For
larger-scale purification of trypsin-extracted material, cells from
1 1 culture were collected and washed as described above,
suspended in 60 ml10 mM Tris/HCl, pH 7.0, containing 0-2 mg
trypsin ml-', and incubated for 1 h at 37 "C. Cells were pelleted
by centrifugation (SOOOg, 10 min, 4 "C), the supernatant was
removed and recentrifuged, then the supernatant was subjected
to centrifugation at 120000g for 4 h at 4 "C. The pellet was
suspended in 60 ml deionized water, centrifuged as before, and
the pellet was then suspended in deionized distilled water
(0.3 ml). The suspension contained fibril-like material as
visualized by negative staining and electron microscopy.
Preparation of cell-surface or culture-fluid proteins. Cells in
late exponential phase of growth were collected and washed as
described earlier and were suspended at a density of approximately 2 x 10'' bacteria ml-' in one of the following: 0.2 mg
pronase ml-' in 10 mM Tris/HCl, pH 7-5, for 1 h at 37 "C;
0.1 M NaOH for 30 min at 4 "C, then neutralized with HC1
(McNab & Jenkinson, 1992b); 5 M urea in 10 mM Tris/HCl,
pH 7.5, containing 1 mM EDTA for 15 min at 4 "C; sample
buffer containing SDS (see earlier) for 10 min at 70 "C.
Suspensions were then centrifuged (6000 g, 10 min, 4 "C) and a
portion of each supernatant was mixed with an equal volume of
sample buffer in preparation for gel electrophoresis. Culture
fluid polypeptides were precipitated from cell-free culture
supernatant with 80 % (v/v) acetone at -20 "C, collected by
centrifugation and solubilized as described previously by
McNab & Jenkinson (1992b).
Electron microscopy. Streptococcal cells were prepared for
electron microscopy and visualized after staining with 1 %
(w/v) methylamine tungstate as described by Handley e t al.
(1985). To visualize fibrillar material, a 25 pl suspension of
fibrils was applied to an ionized Formvar-coated copper grid,
blotted dry, and then stained with 1 % (w/v) methylamine
tungstate.
SDS-PAGE and electroblotting. Proteins were separated by
SDS-PAGE according to the method of Laemmli & Favre
(1973) and were stained with silver nitrate (Morrissey, 1981) or
with Coomassie brilliant blue. The molecular masses of proteins
were determined by reference to a plot relating distance
migrated to log molecular mass for marker proteins in the range
210-1 5 kDa (Life Technologies). Polypeptides were transferred
from polyacrylamide gel to nitrocellulose membrane by electroblotting in 25 mM Tris, 192 mM glycine, 20 % (w/v) methanol,
pH 8.3 (Towbin et al., 1979), at 20 V cm-' for 1 h at 4 "C,
without prior equilibration of the gel in the electro-transfer
buffer.
Antisera, immunoblotanalysis and ELISA. Antisera to trypsinextracted polypeptides (TEPs) were prepared following pre-
-5
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 19:03:13
Streptococcal cell-surface fibril polypeptides
parative SDS-PAGE and visualizing bands by immersing the
gel in 10 vols 4 M sodium acetate (Higgins & Dahmur, 1979).
The required polypeptide bands were excised, electroeluted
from the gel with a Bio-Rad model 422 apparatus and dialysed
against 5 1 distilled water for 16 h at 4 OC with three changes.
Samples were freeze-dried and suspended in 0.15 M NaCl for
inoculation into rabbits. Fibrils were prepared from cells
following trypsin treatment as described above and collected by
high-speed centrifugation. The preparation was incubated with
10 pg pronase ml-' in 10 mM Tris/HCl, pH 7.0, for 1 h at 37 OC,
which degraded TEP components (see Results) but not the fibril
structures as visualized by electron microscopy of negatively
stained preparations of fibrils. The fibril suspension was then
dialysed against deionized water (10 1) for 16 h at 4 OC, freezedried, and the residue was suspended in 0.15 M NaC1. Antibodies were raised in New Zealand White rabbits by intramuscular injection of approximately 20 pg protein or fibril
preparation, followed by two boosters of 4 pg protein or fibrils
after 3 weeks and 10 weeks.
Antiserum reactivity was determined on nitrocellulose blots of
streptococcal proteins incubated with antiserum diluted 1 :500
or 1: 1000. Antibody binding was detected with peroxidaseconjugated swine antibodies to rabbit immunoglobulin G
(Dako) as described by Jenkinson & Easingwood (1990).
Reactivity with whole cells was determined by ELISA. Briefly,
streptococcal cells from the late exponential phase of growth
were suspended at an OD,,, of 0.5 in PBS (15 mM Na,HPO,NaH,PO,, pH 7-0, containing 0.1 5 M NaCl), added to the wells
of Maxisorp (Nunc) flat-bottom 96-well polystyrene plates
(005 ml per well) and the plates were centrifuged at 800g for
5 min to deposit the cells. Cells were fixed to the plastic surface
by addition of 0.25% glutaraldehyde (0.1 ml per well) for
30 min, washed twice with PBS containing 0.05 % Tween 20
(PBS-Tween), then wells were blocked with PBS containing
0.25 % gelatin (PBS-gelatin) for 16 h at 4 OC. Serial twofold
dilutions of immune sera or preimmune sera in PBS-gelatin
were added to wells, plates were incubated for 1 h at 37 "C, well
contents were discarded and wells were washed four times with
PBS-Tween. Antibody binding was detected with peroxidaseconjugated swine immunoglobulins to rabbit immunoglobulins
(Dako) diluted 1: 1000 in PBS-gelatin (0.05 ml per well). Plates
were developed with 1,Zphenylenediamine as enzyme substrate
and the A,,, was measured.
Estimation of protein and carbohydrate. Protein concentration was determined using Folin reagent with BSA as
standard. Carbohydrate was determined by the phenol-sulphuric
acid method (Dubois e t al., 1956) with glucose as standard.
Adhesion to saliva-treated hydroxylapatite beads. Streptococci were grown to late exponential phase in TY-glucose
medium with [me~'h_yl-~H]thyrnidine
[0.22 MBq (6 pCi) ml-';
85 Ci mmol-'I, harvested by centrifugation (5000 g for 10 min),
washed twice with distilled water and suspended at an OD,,,of
1.0 in KC1 buffer (2 mM NaH,PO,, 5 mM KCl, 1 mM CaCl,,
adjusted to pH 6-4 with NaOH) (Eifert e t al., 1984). The
numbers of cells adhering to hydroxylapatite beads (BDH)
treated with whole or parotid saliva (diluted 1:5 in KC1 buffer)
were measured as described by Hawkins e t al. (1993).
Unstimulated whole saliva was collected from five individuals,
pooled, clarified by centrifugation (lOOOOg, 20 min, 4 "C) and
diluted 1:5 in KC1 buffer. Stimulated parotid saliva was obtained
from one individual using a modified Carlsson-Crittenden
device (Shannon e t al., 1962). In some experiments, cells were
suspended in 10 mM Tris/HCl buffer, pH 7-0, and treated with
trypsin for 60 rnin at 37 "C, or in PBS containing appropriate
dilution of antiserum with end-over-end mixing for 90 min at
20 OC, collected by centrifugation, washed three times by
alternate suspension in KC1 buffer and centrifugation, and then
assayed for adhesion.
Adhesion to buccal epithelial cells (BECs). BECs were collected
from 20 individuals by gently scraping their inside cheek
surfaces with wooden spatulas. Cells were pooled in TSMC
buffer (5 mM Tris/HCl, pH 7.4, containing 100 mM NaC1,
50 mM KC1, 1 mM MgC1, and 1 mM CaCl,), allowed to settle
and the supernatant was decanted from the cells. The BECs
were suspended in TSMC buffer at a density of 1.0 x lo5 cells
ml-' (assessed by direct microscopic count) and 005 ml portions
were distributed into Maxisorp microtitre plate wells. Plates
were centrifuged (800 g, 3 min), incubated for 12 h at 4 OC, and
then the TSMC buffer was aspirated and replaced with TSMC
buffer containing 0-2% Tween 20 (TSMC-Tween) (0.2 ml per
well) for 16 h at 4 OC to block any remaining binding sites on the
plastic. Microscopic examination of the wells revealed that the
BECs formed a monolayer with approximately 80 % coverage
of the plastic surface. The wells were washed twice with TSMC
buffer, 005 ml suspension containing 1.25 x lo' cells of 3Hlabelled streptococci in TSMC buffer was added to each well and
the plates were incubated for 2 h at 20 OC with shaking. The
liquid contents of the wells containing unattached streptococcal
cells were discarded; the wells were washed once with TSMCTween and then twice with TSMC buffer. Bound streptococcal
cells were then removed by suspension in 0.1 M NaOH
containing 0.2 % SDS and transferred to scintillation vials for
radioactivity counting as previously described by Jenkinson e t
al. (1993). In some experiments, bacterial cells were incubated
with trypsin or with antisera as described above, and washed
with TSMC buffer before assaying for adhesion.
Cell-surface hydrophobicity. Late exponential growth phase
cells in TY-glucose medium were collected by centrifugation,
washed twice with PUM buffer (Rosenberg et al., 1980) and
suspended at an OD,,, of 0.5 (approximately 4 x 10' bacteria
ml-'). Streptococcal cell suspension (3 ml) was mixed with
hexadecane (0.3 ml), the phases were allowed to settle, and the
percentage of input cells associated with the organic phase (a
measure of the net surface hydrophobicity of the cell population)
was calculated from the decrease in OD,,, of the aqueous phase.
Coaggregation. The ability of streptococcal cells to coaggregate
with A . nuehndii cells was assessed by light microscopic
examination following mixing of equal volumes of partner cell
suspensions in TSMC buffer each containing about 1 x lo9
bacteria ml-' and incubating for 1 h at 37 OC with shaking. The
ability of streptococci to form corn-cob coaggregates with C.
matruchotii was investigated by mixing a 10 ml culture of
streptococcal cells in exponential growth phase with a 40 ml
exponential growth phase culture of C. matrucbotii in TSBY
medium and incubating with gentle shaking for 12 h at 37 OC.
Coaggregation and the presence of corn-cob formations was
investigated by light microscopy.
RESULTS
Extraction of cell-surface polypeptides and effects on
cell morphology
To begin to characterize the components of surface fibrils
on S. 0rali.r CN3410, a number of agents were examined
for their effectiveness in solubilizing polypeptides and
removing concomitantly the tufts of fibrils from the cell
surface. Incubation of cells with 0.1 M N a O H , which has
been shown to extract surface proteins and polysaccharides from S.gordonii (McNab & Jenkinson, 1992b),
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 19:03:13
2731
M. W. J A M E S O N a n d O T H E R S
...........................................................................................................................................,....,,.,..,..............................................................................................................................................................................
Fig. 1. SDS-PAGE of polypeptides extracted from cells of 5. oralis CN3410 with trypsin (a), pronase (c), or sample buffer
containing SDS (d), and from isolated fibrils with sample buffer (b). In each of the four panels the lanes are composites of
the proteins in the gel stained with silver nitrate (lane 1) and corresponding nitrocellulose blots reacted with antibodies
(diluted 1 :1000) to group A TEPs (lane 2), group B TEPs (lane 3) and purified fibrils (lane 4). TEPs were denoted group A
and group B as described in the text. Molecular masses of marker proteins are indicated (kDa).
did not effect significant release of polypeptides from
strain CN3410 cells as determined by SDS-PAGE. A 5 M
urea extract of cells contained approximately 20 polypeptide bands in the range 15-70 kDa (results not shown).
Neither NaOH nor urea treatment significantly affected
cell-surface fibril morphology. Incubation of cells with
sample buffer containing SDS (see Methods) solubilized a
complex mixture of > 50 polypeptides and also high
molecular mass material that did not enter the separating
gel (Fig. Id, lane 1). Cells that had been treated with SDS
showed partial disruption of cell wall layers and of fibril
tufts (results not shown). Incubation of cells with pronase
completely removed the long and short fibril components
of the tufts (Handley e t al., 1984) but no polypeptide
bands were discernible following SDS-PAGE of pronase
extracts (Fig. lc, lane 1). Incubation of cells with trypsin
resulted in release of intact fibrils (see below) and upon
SDS-PAGE of trypsin extracts several high molecular
mass polypeptides were identified (Fig. la, lane 1) which
we called TEPs. The polypeptides present in the trypsin
extracts were divided into two groups based on their
molecular mass and staining characteristics. Group A of
higher molecular mass comprised three bands (estimated
molecular masses 246 kDa, 238 kDa and 227 kDa) which
stained with silver nitrate but only weakly with Coomassie
blue. Group B comprised four polypeptide bands
(estimated molecular masses 208 kDa, 195 kDa, 183 kDa
and 175 kDa) which stained well with both silver nitrate
and Coomassie blue. These gel patterns were highly
reproducible and were not affected by the addition of
trypsin inhibitor immediately following trypsin digestion
2732
of cells. Little or no material of molecular mass <
100 kDa was visible on SDS-PAGE of trypsin extracts
suggesting that other trypsin-susceptible cell-surface
proteins were extensively degraded. Further incubation of
extracts with fresh trypsin (up to 6 h) did not result in
significant degradation of the high molecular mass TEPs.
Trypsin extracts of cells were estimated to contain about
8 % of total cell protein.
Cells of strain CN3410 carried lateral tufts of short very
fine densely packed fibrils through which longer sparser
clumped fibrils protruded, as well as a peritrichous fringe
of much shorter indistinct fuzzy material which could also
be fibrillar (Fig. 2a). Frequently visible were loose fibrils
around the cell periphery. Treatment of cells with trypsin
for 60 min at 37 "C removed the long and short fibril
components of the tufts from all cells leaving loose fibrils
associated with the cell surface (Fig. 2b). Released fibrils
were collected by high-speed centrifugation of trypsin
extracts, and they appeared by electron microscopy as a
tangled mass of clumped fibres (Fig. 2c). Occasionally,
very thin, possibly single, fibrils could be resolved. The
fibrillar material contained protein and carbohydrate in
the ratio 3.6: 1 based on colorimetric assays. The SDSPAGE profile of proteins extracted with sample buffer
from these fibrils (Fig. l b , lane 1) was similar to the
protein profile of the trypsin extract of cells (that
contained fibrils before high-speed centrifugation), with
some slight differences in migration and staining intensity
of group B bands. These results suggested that the TEPs
were associated with the cell-surface fibrils.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 19:03:13
Streptococcal cell-surface fibril polypeptides
Fig. 2. Transmission electron micrographs of cells or fibrils negatively stained with methylamine tungstate. (a) 5. oralis
CN3410 cells showing tufts comprising short fibrils (large arrows) through which long fibrils protrude (small arrows),
and peritrichous fringe material. (b) 5. oralis CN3410 following incubation with trypsin for 60min a t 37°C; cells are
stripped of their tufts and loose fibrils surround the cells (arrows). (c) Loose fibrils following trypsin treatment as in (b)
comprising material collected by high-speed centrifugation as described in the text (arrows show possible separate
fibrils). (d) 5. oralis KP34V cells showing tufts composed of long fibrils only. Bars, 200 nm.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 19:03:13
2733
M. W. J A M E S O N a n d OTHERS
Table 1. ELISA of immune or preimmune sera activities with immobilized cells of 5. oralis
CN3410 or KP34V
Reactivity (A4%)of serum*
Substrate
S. oralis CN3410
S. oralis CN3410
trypsin$
S. oralis KP34V
+
Group A TEPs
Group B TEPs
Fibrils
Preimmune
058t
0.23
0.34
0.16
042
020
0.05
0.46
042
053
0.06
0.07
* Diluted 6.4 x lo4.
t Means of duplicate samples from a representative experiment that differed by
< &- 5 %.
$Cells incubated with 0.2 mg trypsin ml-' for 1 h at 37 O C , harvested by centrifugation, washed and
deposited onto microtitre plate wells as described in Methods.
Antigenic relatedness of TEPs and fibril proteins
To investigate the antigenic relatedness of group A TEPs,
group B TEPs, and fibrils, antibodies were raised in
rabbits to each of these antigens. Antiserum from rabbits
immunized with group A TEPs reacted most strongly
with group A bands, as might be expected, and also with
group B bands (Fig. la, lane 2). Antibodies to group B
TEPs reacted more strongly with group B bands and also
with group A polypeptides (Fig. la, lane 3). Thus group
A and group B polypeptides contained cross-reactive
epitopes. Antibodies to fibrils reacted with both group A
and group B polypeptides (Fig. la, lane 4). The reactions
of the various antisera with blots of polypeptides extracted
from purified fibrils mirrored the results just described
(Fig. lb). Antibodies to fibrils, but not to group A or
group B polypeptides, reacted weakly with a diffuse band
on nitrocellulose blots of pronase-extracted material
separated by SDS-PAGE (Fig. lc). Each antiserum
preparation reacted with high molecular mass material (>
300 kDa) on blots of proteins extracted from cells with
sample buffer (Fig. Id).
The reactivities of sera with intact cells were determined
by ELISA in microtitre plate wells coated with streptococcal cells. Similar high titres were obtained for each
antiserum, with the anti-group-A polypeptides serum
being most reactive, while preimmune serum was not
reactive (Table 1). The ELISA titres for each of the sera
were > 50% reduced with streptococcal cells that had
been preincubated with trypsin (Table 1). These data
confirmed the antigenic relatedness of TEPs and fibril
components, and showed further that these antigens were
accessible on the cell surface.
Fibril and TEP production by a variant of strain
CN3410
Following routine subculture of S. oralis CN3410 on
TSBY agar a single smaller-colony variant was identified.
This was purified as a spontaneously derived mutant
strain that formed marginally smaller colonies than the
parent strain, but which was otherwise identical to the
2734
taxonomic description of 3. oralis CN3410. Electron
microscopic examination of cells of the mutant strain
(denoted KP34V) showed that they carried peritrichous
fringe material similar to the parent strain; however the
short fibrils were missing from the lateral tufts of fibrils
which, in the mutant, consisted of the long fibrils only
(Fig. 2d). Trypsin-released fibrils from strain KP34V
collected by high-speed centrifugation appeared identical
by electron microscopic observation to those collected
from the parent strain CN3410 and shown in Fig. 2(c).
The SDS-PAGE profiles of TEPs extracted from KP34V
were identical to those from the wild-type strain CN3410
(Fig. 3a). On immunoblots, the TEPs from KP34V
reacted with anti-group-B TEPs serum identically to the
wild-type TEPs (Fig. 3a). They also reacted similarly to
the TEPs from wild-type cells with group A TEPs and
fibril antisera (not shown). In ELISA, anti-serum to the
group A TEPs reacted somewhat less well with KP34V
cells than with wild-type cells (Table 1). Conversely,
antibodies to the group B TEPS and to fibrils both reacted
better with KP34V cells than with the wild-type cells
(Table 1).
Since there were no obvious protein differences between
trypsin extracts of cells of CN3410 and KP34V, we
investigated the profiles of proteins secreted into the
culture fluid. Both strains secreted similar polypeptides as
revealed by SDS-PAGE and staining with silver nitrate
(results not shown) but there was a considerably larger
amount of high molecular mass (> 300 kDa) material
present in the culture fluid of KP34V. This material was
the predominant component on SDS-gels of culture fluid
proteins from KP34V stained with Coomassie blue (Fig.
3b). Nitrocellulose blots of culture fluid proteins from the
two strains gave slightly different reactions with antigroup-B TEPs serum (Fig. 3b) with the high molecular
mass material from strain KP34V reacting comparatively
weakly. This material was therefore possibly not antigenically closely related to the TEPs. Culture fluid
polypeptides from each strain were then prepared, incubated with trypsin, separated by SDS-PAGE, blotted and
reacted with group B antiserum. The resulting immunoblot patterns were identical for each strain with two
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 19:03:13
Streptococcal cell-surface fibril polypeptides
Fig. 3, Comparison of SDS-PAGE and
immunoblot profiles of trypsin-extracted or
culture-fluid polypeptides of 5. oralis
CN3410 and KP34V. (a) TEPs from CN3410
cells (lane 1) or KP34V cells (lane 2) stained
with Coomassie blue, and corresponding
nitrocellulose blots (lanes 3 and 4) reacted
with anti-group4 TEPs serum (1 :500). (b)
CelI-free culture-fIuid polypeptides from
CN3410 (lane 1) or KP34V (lane 2) stained
with Coomassie blue and corresponding
blots reacted with anti-group-B TEPs serum
(lanes 3 and 4).
(c) Culture-fluid
polypeptides corresponding to those in
panel (b) but incubated with trypsin for 60
min
at
37°C
before
SDS-PAGE,
electroblotting and reaction of blots with
anti-group-B TEPs serum (lane 1, CN3410;
lane 2, KP34V). The group A TEPs stained
relatively weakly with Coomassie blue (a,
lanes 1 and 2). The molecular masses of
marker proteins are indicated (kDa).
Table 2. Comparison of cell adhesion properties of 5. oralis CN3410 and KP34V and effects of trypsin or antibodies to
fibrils on adhesion of CN3410 cells
Strain and treatment*
Percentage adhesion ( fSD, n = 4)
to hexadecane
(hydrophobicity)
56.2 f6-10
57.4 f3.10
1-06f0.06
CN3410
KP34V
CN3410 trypsin
anti-fibril serum
(1 :100)
anti-fibril serum
(1 :1000)
preimmune serum
(1 :100)
+
+
+
+
ND
x No. of cells adhering ( fSD, n = 4) to:
Parotid saliva-treated
hydroxylapatite
beadst
BECs$
28.7 f2-40
35.7 f1.70
32.2 f2.20
26.0 f2.30
101 f 4 2 0
9.20 f1.10
0.87 f0.16
121 f3.74
ND
ND
10-6f3.49
ND
28-75 1.10
10.7 f3-00
ND,Not determined.
* Conditions of incubation with trypsin or serum are described in Methods.
t Input number of streptococcal cells 2.5 x lo7.
$ Input number of streptococcal cells 1.25 x lo7.
closely migrating antigenic bands similar in size to the
group A TEPs (Fig. 3c).
Cell-surface hydrophobicityand cell adherence
properties
Cell-surface hydrophobicity is a factor that has been
shown to influence adhesion properties of streptococci,
particularly to salivary pellicle (Doyle e t al., 1990). Surface
hydrophobicities of S. oralis CN3410 and KP34V cells, as
measured by the numbers of bacteria adsorbing to
hexadecane, were found to be virtually identical (Table 2).
Cell-surface hydrophobicity was abolished by treating
cells with trypsin under the standard conditions that
removed TEPs and fibril tufts (Table 2). Cells of S. oralis
CN3410 and KP34V showed similar adhesion abilities to
hydroxylapatite beads treated with human parotid salivary
proteins (Table 2). Tenfold less numbers of cells of both
strains bound to beads treated with human whole saliva
(results not shown). Treatment of strain CN3410 cells
with trypsin had no effect on their adhesion to parotidsaliva-treated beads (Table 2). Incubation of streptococcal
cells with preimmune serum or with antibodies to fibrils
had no effect on their ability to adhere to parotid-salivatreated hydroxylapatite beads (Table 2). Incubation of
cells with antibodies to group A or group B TEPs also
had no effect on their adhesion to beads (not shown in
Table 2). Both strains adhered equally well to human
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 19:03:13
2735
M. W. J A M E S O N and OTHERS
BECs and binding was > 90% reduced following incubation of streptococcal cells with trypsin (Table 2).
However, streptococcal cell binding was unaffected by
incubation of cells with preimmune serum or with
antiserum to fibrils (Table 2), or with antiserum to group
A or group B TEPs (not included in Table 2). Neither
strain CN3410 nor strain KP34V cells coaggregated with
any of seven strains of A . naeslundii that were tested (14V,
BE64, WVU627, W1544, EF1006, TF11, ATCC 12104).
This suggests that, unlike many strains of S. oralis, cells of
strain CN3410 and KP34V do not carry a surface
polysaccharide receptor for A . naeslundii adhesins
(Kolenbrander & London, 1993). In addition, CN3410 or
KP34V cells did not coaggregate with C. matrucbotii,
whereas under similar experimental conditions cells of S.
crista NCTC 12479 and AK1 formed corn-cob
coaggregates with C. matrucbotii.
DISCUSSION
Fibril tufts are some of the largest extracellular structures
produced by streptococci, yet their structural and functional components are unknown. Therefore, we set out to
determine the protein composition and function in
adhesion of the fibril tufts produced by S.oralis CN3410.
This strain was chosen because previous work had
suggested that the fibril tufts of strain CN3410, unlike
those of several other tufted strains, could be removed
morphologically intact by incubation of cells with pronase
(Hesketh e t al., 1987). When these experiments were
repeated and the polypeptide components of the fibrils
were analysed by SDS-PAGE, we could detect no
polypeptide bands. One explanation for this was that
while the fibrils were morphologically intact, the pronase
had nevertheless substantially cleaved the protein components which remained ligated within the fibril structures until solubilized with a denaturant (SDS).
Treatment of CN3410 cells with trypsin also resulted in
release of long fibrils and loss of tufts from the cells.
Trypsin extracts of cells, unlike pronase extracts, contained reproducible SDS-PAGE patterns of proteins and
these were divided conveniently into two groups. The
group of higher molecular mass bands, termed group A
TEPs, stained only weakly with Coomassie blue, which is
a characteristic of some glycosylated polypeptides. When
fibrils were collected by high-speed centrifugation of the
trypsin extracts, and the proteins associated with the
fibrils were solubilized by SDS, an almost identical SDSPAGE pattern of polypeptides was obtained to that from
the crude trypsin extracts of whole cells that contained
loose fibrils. This suggested that the TEPs might be
components of the fibrils. In support of this notion,
antibodies raised to pronase-treated fibrils, which contained no detectable proteins by SDS-PAGE, reacted
with the group A and group B TEPs more or less
identically to the antibodies that had been raised to the
purified TEPs. Thus the pronase-treated fibrils contained
peptide fragments that had common epitopes with the
TEPs. The group A TEPs and group B TEPs were shown
to be immunologically cross-reactive. Therefore the TEPs
are possibly proteolytic fragments of a single gene
2736
product, alternatively they could be derived from the
products of expression of more than one related gene.
Tufts on CN3410 cells were composed of two structurally
distinct types of fibrils, the short fibrils were 159f5-0 nm
and the long fibrils were 289 f15-0nm in length (Handley
e t a/., 1984). It was not possible to deduce from electron
micrographs whether the isolated fibrils contained long
and short fibrils, or long fibrils only. When removed from
the cells and collected by high-speed centrifugation, the
fibrils became aggregated and structures up to a few
microns in length with tapered ends were visible (Fig. 2c).
However, it is likely that the isolated fibrils comprised the
long fibrils only, for two reasons : firstly, short tuft fibrils
are removed and/or degraded by pronase treatment of
strain CN3410 cells before the long tuft fibrils are liftedoff (Hesketh e t al., 1987); secondly, isolated fibrils from
strain KP34V, which lacked the short fibril components
of the tufts, appeared identical in electron micrographs to
those isolated from strain CN3410. The anti-fibril serum
reacted better with KP34V cells than with wild-type cells,
which may indicate that in the absence of the short fibrils
on KP34V cells, long fibril epitopes were more accessible.
The isolation of strain KP34V was also crucial in
establishing that the TEPs were not components of the
shorter fibrils, since SDS-PAGE profiles of TEPs
extracted from wild-type or mutant cells were identical.
Taken together, these results suggest strongly that the
TEPs are components of the long fibrils. We have
attempted to confirm this by electron microscopic observations of cells reacted with anti-TEPs sera and goldlabelled secondary antibody. However these experiments
have been unsuccessful to date because the streptococcal
cells appear to bind avidly conventional blocking agents
such as ovalbumin or bovine albumin, as well as nonspecifically bind the gold-labelled secondary antibody.
In S. crista NCTC 12479 the long fibril components of
tufts have been implicated in conferring cell-surface
hydrophobicity (Busscher e t al., 1991) while the short
fibrils may confer a localized surface negative charge
(Handley, 1991). Trypsin treatment of S. oralis CN3410
cells abolished their hydrophobicity, which corroborates
numerous other findings that polypeptides are determinants of cell-surface hydrophobicity in streptococci
(Doyle e t al. , 1990). However, since trypsin will degrade
many surface polypeptides these data do not imply that
the long fibrils are necessarily hydrophobins. Clearly
though the short fibrils are not determinants of cellsurface hydrophobicity in strain CN3410 since mutant
cells lacking short fibrils were unaffected in their ability to
bind hexadecane. On the other hand, treatment of strain
CN3410 cells with trypsin did not affect their adhesion to
parotid salivary pellicle. This observation is in contrast to
many others which have shown that protease treatment of
streptococcal cells reduces not only hydrophobicity
(Jenkinson, 1986) but also adhesion to salivary pellicle
(Oakley e t al., 1985; Hesketh e t a/., 1987). Despite the
overwhelming evidence that hydrophobic interactions are
important for adhesion of streptococci (Doyle e t al.,
1990), these experiments reveal that for 5. oralis CN3410
hydrophobicity is not an essential attribute for cell
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 19:03:13
Streptococcal cell-surface fibril polypeptides
adhesion to pellicle. This has also been shown recently for
isogenic mutants of S. gordonii deficient in production of
surface polypeptides CshA and CshB. These mutants had
much-reduced cell-surface hydrophobicity but nevertheless adhered normally to experimental salivary pellicles
(McNab et al., 1994). In summary, adhesion of strain
CN3410 cells to parotid salivary pellicle is not sensitive to
trypsin under the conditions employed, is independent of
cell-surface hydrophobicity, and does not involve the
long or short components of the fibril tufts since these
were wholly removed by trypsin treatment of cells.
Binding of strain CN3410 cells to B E G involves different
adhesins from those necessary for binding to parotid
salivary proteins since trypsin treatment of cells abolished
their ability to bind to BECs. However, antibodies to
TEPs or fibrils did not inhibit adhesion of strain CN3410
cells to BECs. This suggests that the TEPs are distinct
from the receptor-binding sites of the adhesins. By
analogy, it has been shown for the Gram-positive oral
bacterium A. naesltlndii that antibodies directed to the type
1 fimbrial structural subunit do not react with the
receptor-binding sites of the type 1 fimbriae (Cisar e t al.,
1991). The involvement of surface structures in adhesin
presentation by Escbericbia coli is well-documented. In
uropathogenic strains expressing P pili the papG gene
product encodes the lectin adhesin which is located at the
tip of the pilus rod (Kuehn e t al., 1992), while in E. coli
expressing type 1 fimbriae the FimH protein is thought to
be responsible for the adhesive properties of the fimbriae,
which are composed of polymers of FimA (Sokurenko e t
al., 1994; Ofek & Doyle, 1994). To elucidate better the
structure and function of fibril tufts in S. oralis CN3410,
we are attempting to isolate the gene(s) encoding the
TEPs. By insertionally inactivating the coding
sequence(s) on the S. oralis chromosome it should be
possible to determine more precisely how the TEPs are
involved in production of fibril tufts and in adhesin
presentation.
ACKNOWLEDGEMENTS
We are most grateful to R. Whiley, The London Hospital
Medical College Dental School, London, UK, for performing
definitive experiments that enabled speciation of strain CN3410.
We thank R. McNab and other colleagues in the Experimental
Oral Biology Laboratory, University of Otago, for helpful
discussions and R. A. Baker for technical assistance. M. W. J.
was in receipt of a University of Otago Postgraduate
Studentship.
van der Mei, H. C. (1991). The relationship between structural and
physicochemical surface properties of tufted Streptococcus sanguis
strains. In Microbial Cellsurface Anahsis, pp. 317-338. Edited by N.
Moses, P.S. Handley, H. J. Busscher & P. G. Rouxhet. New
York : VCH Publishers.
Cisar, J. O., Barsumian, E. L., Siraganian, R. P., Clark, W. B.,
Yeung, M. K., Hsu, 5. D., Curl, 5. H., Vatter, A. E. & Sandberg, A.
L. (1991). Immunochemical and functional studies of Actinomyces
viscosus T14V type 1 fimbriae with monoclonal and polyclonal
antibodies directed against the fimbrial subunit. J Gen Microbioll37,
1971-1 979.
Doyle, R. J., Rosenberg, M. & Drake, D. (1990). Hydrophobicity of
oral bacteria. In Microbial Cell Surface Hydropbobicit_y,pp. 387-41 9.
Edited by R. J. Doyle & M. Rosenberg. Washington, DC:
American Society for Microbiology.
Dubois, M., Gilles, K. A., Hamilton, J. K., Rebers, P. A. &Smith, F.
(1956). Colorimetric method for determination of sugars and
related substances. Anal Chem 28, 350-354.
Eifert, R., Rosan, B. & Golub, E. (1984). Optimization of an
hydroxyapatite adhesion assay for Streptococcus sanguis. Infect Immun
44, 287-291.
Fenno, J. C., LeBlanc, D. 1. & Fives-Taylor, P. (1989). Nucleotide
sequence analysis of a type 1 fimbrial gene of Streptococcus sanguis
FW213. Infect Immun 57, 3527-3533.
Fives-Taylor, P. M. & Thompson, D. W. (1985). Surface properties
of Streptococcussanguis FW213 mutants nonadherent to saliva-coated
hydroxyapatite. Infect Immun 47, 752-759.
Frandsen, E. V. G., Pedrazzoli, V. & Kilian, M. (1991). Ecology of
viridans streptococci in the oral cavity and pharynx. Oral Microbiol
Immunol6, 129-133.
Ganeshkumar, N., Hannam, P. M., Kolenbrander, P. E. & McBride,
B. C. (1991). Nucleotide sequence of a gene coding for a saliva-
binding protein (SsaB) from Streptococcussanguis 12 and possible role
of the protein in coaggregation with actinomyces. Infect Immun 59,
1093-1099.
Gibbons, R. J., Etherden, 1. & Skobe, 2. (1983). Association of
fimbriae with the hydrophobicity of Streptococcus sanguis FC-1 and
adherence to salivary pellicles. Infect Immun 41, 414-417.
Handley, P. 5. (1990). Structure, composition and functions of
surface structures on oral bacteria. Biofouling 2, 239-264.
Handley, P. 5. (1991). Negative staining. In Microbial Cell Surface
Analysis, pp. 6386. Edited by N. Moses, P. S. Handley, H. J.
Busscher & P. G. Rouxhet. New York: VCH Publishers.
Handley, P. 5.. Carter, P. L. & Fielding, J. (1984). Streptococcus
salivarius strains carry either fibrils or fimbriae on the cell surface.
J Bacterioll57, 64-72.
Handley, P. S., Carter, P. L., Wyatt, J. E. & Hesketh, L. M. (1985).
Surface structures (peritrichous fibrils and tufts of fibrils) found on
Streptococcus sanguis strains may be related to their ability to
coaggregate with other oral genera. Infect Immun 47, 217-227.
Handley, P., Coykendall, A., Beighton, D., Hardie, J. M. & Whiley,
R. A. (1991). Streptococcus crista sp. nov., a viridans streptococcus
REFERENCES
Andersen, R. N., Ganeshkumar, N. & Kolenbrander, P. E. (1993).
Cloning of the Streptococcus gordonii PK488 gene, encoding an
adhesin which mediates coaggregation with Actinomyces naeslundii
PK606. Infect Immun 61,981-987.
Bleiweis, A. S., Oyston, P. C. F. & Brady, L. J. (1992). Molecular,
immunological and functional characterization of the major surface
adhesin of Streptococcus mutans. In Genetical& Engineered Vaccines,pp.
229-241. Edited by J. E. Ciardi, J. R. McGhee & J. M. Keith. New
York: Plenum Press.
Busscher, H. J., Handley, P. S., Rouxhet, P. G., Hesketh, L. M. &
with tufted fibrils, isolated from the human oral cavity and throat.
Int J Syst Bacteriol 41, 543547.
Hawkins, B. W., Cannon, R. D. & Jenkinson, H. F. (1993). Interactions of Actinomycesnaeslundii strains T14V and ATCC 12104 with
saliva, collagen and fibrinogen. Arch Oral Biol38, 533-535.
Hesketh, L. M., Wyatt, J. E. & Handley, P. 5. (1987). Effect of
protease on cell surface structure, hydrophobicity and adhesion of
tufted strains of Streptococcus sanguis biotypes I and 11. Microbios 50,
131-1 45.
Higgins, R. C. & Dahmur, M. E. (1979). Rapid visualisation of
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 19:03:13
2737
M. W. J A M E S O N and OTHERS
protein bands in preparative SDS-polyacrylamide gels. AnalBiocbem
93, 1575-1589.
Jenkinson, H. F. (1986). Cell-surface proteins of Streptococcus sanguis
associated with cell hydrophobicity and coaggregation properties.
J Gen Microbioll32, 1575-1 589.
Jenkinson, H. F. (1992). Adherence, coaggregation, and hydrophobicity of Streptococcus gordonii associated with expression of cell
surface lipoproteins. Infect Immun 60, 1225-1228.
Jenkinson, H. F. (1994). Cell surface protein receptors in oral
streptococci. FEMS Microbiol Lett 121, 133-140.
Jenkinson, H. F. 81 Easingwood, R. A. (1990). Insertional inactivation of the gene encoding a 76-kilodalton cell surface polypeptide in StreptococcusgordoniiChallis has a pleiotropic effect on cell
surface composition and properties. Infect Immun 58, 3689-3697.
Jenkinson, H. F., Terry, S. D., McNab, R. &Tannock, G. W. (1993).
Inactivation of the gene encoding surface protein SspA in
Streptococcus gordonii DL1 affects cell interactions with human
salivary agglutinin and oral actinomyces. Infect Immun 61,
3199-3208.
Kilian, M., Mikkelsen, L. & Henrichsen, J. (1989). Taxonomic study
of viridans streptococci : description of Streptococcusgordoniisp. nov.
and emended description of Streptococcus sanguis (White and Niven
1946), Streptococcus oralis (Bridge and Sneath 1982), and Streptococcus
mitis (Andrewes and Horder 1906). Int J J’st Bacteriol39,471-484.
Kolenbrander, P. E. 81 London, 1. (1993). Adhere today, here
tomorrow : oral bacterial adherence. J Bacterioll75, 3247-3252.
Kolenbrander, P. E., Andersen, R. N. & Ganeshkumar, N. (1994).
Nucleotide sequence of the Streptococcus gordonii PK488
coaggregation adhesin genes, scaA, and ATP-binding cassette.
Infect Immun 62, 4469-4480.
Kuehn, M. J., Heuser, J., Normark, 5. & Hultgren, S. J. (1992). P pili
in uropathogenic E. coli are composite fibres with distinct fibrillar
adhesive tips. Nature 356, 252-255.
Laemmli, U. K. & Favre, M. (1973). Maturation of the head of
bacteriophage T4. I. DNA packaging events. J Moi Biol 80,
575-599.
Lancy, P., Jr, Appelbaum, B., Holt, S.C. 81 Rosan, B. (1980).
Quantitative in vitro assay for ‘corncob’ formation. Infect Immun
29, 663-670.
Lancy, P., Jr, Dirienzo, J. M., Appelbaum, B., Rosan, B. & Holt, 5.
C. (1983). Corncob formation between Fusobacterium nucleatum and
Streptococcus sangzfis.Infect Immun 40, 303-309.
McNab, R. & Jenkinson, H. F. (1992a). Gene disruption identifies a
290 kDa cell-surface polypeptide conferring hydrophobicity and
coaggregation properties in Streptococcus gordonii. Mol Microbiol6,
2939-2949.
McNab, R. & Jenkinson, H. F. (1992b). Aggregation-deficient
mutants of Streptococcus gordonii Channon altered in production of
cell-surface polysaccharide and proteins. Microb Ecol Health Dis 5 ,
277-289.
2738
McNab, R., Jenkinson, H. F., Loach, D. M. & Tannock, G. W.
(1994). Cell-surface-associated polypeptides CshA and CshB of
high molecular mass are colonization determinants in the oral
bacterium Streptococcusgordonii. Mol Microbioll4, 743-754.
Morris, J. E., Ganeshkumar, N., Song, M. & McBride, B. C. (1987).
Identification and preliminary characterization of a Streptococcus
sanguis fibrillar glycoprotein. J Bacterioll69, 1 6 4 171.
Morrissey, J. H. (1981). Silver stain for proteins in polyacrylamide
gels : a modified procedure with enhanced uniform sensitivity. Anal
Biochem 117, 307-310.
Mouton, C., Reynolds, H. 5. & Genco, R. 1. (1980). Characterization
of tufted streptococci isolated from the ‘corn cob’ configuration of
human dental plaque. Infect Immun 27, 235-245.
Oakley, 1. D., Taylor, K. G. & Doyle, R. J. (1985). Trypsinsusceptible cell surface characteristics of Streptococcus sanguis. Can J
Microbiol 31, 1103-1 107.
Ofek, 1. & Doyle, R. 1. (1994). Bacterial Adhesion to Cells and Tissues.
London: Chapman & Hall.
Oligino, L & Fives-Taylor, P. (1993). Overexpression and purification of a fimbria-associated adhesin of Streptococcusparasanguis.
Infect Immun 61, 1016-1022.
Rosenberg, M., Gutnick, D. 81Rosenberg, E. (1980). Adherence of
bacteria to hydrocarbons : a simple method for measuring cellsurface hydrophobicity. FEMS Microbiol Lett 9, 29-33.
Shannon, I. L., Prigmore, J. R. &Chauncey, H. H. (1962). Modified
Carlson-Crittenden device for the collection of parotid fluid. J Dent
Res 41,778-783.
Socransky, S. S., Manganiello, A. D., Propas, D., Oram, V. & van
Houte, J. (1977). Bacteriological studies of developing supragingival dental plaque. J Periodontal Res 12, 90-106.
Sokurenko, E. V., Courtney, H. S., Ohman, D. E., Klemm, P. &
Hasty, D. L. (1994). FimH family of type 1 fimbrial adhesins:
functional heterogeneity due to minor sequence variations among
fimH genes. J Bacteriol 176, 748-755.
Towbin, H., Staehelin, T. & Gordon, J. (1979). Electrophoretic
transfer of proteins from polyacrylamide gels to nitrocellulose
sheets : procedure and some applications. Proc Natl Acad Sci U S A
76,4350-4354.
Weerkamp, A. H., Handley, P. S., Baars, A. 81Slot, J. W. (1986a).
Negative staining and immunoelectron microscopy of adhesiondeficient mutants of Streptococcus salivarius reveal that the adhesive
proteins are separate classes of cell surface fibril. J Bacteriol 165,
746-7 55.
Weerkamp, A. H., van der Mei, H. C. & Liem, R. 5. B. (1986b).
Structural properties of fibrillar proteins isolated from the cell
surface and cytoplasm of Streptococcus salivarius (K+) cells and
nonadhesive mutants. J Bacterioll65, 756762.
Received 26 April 1995; revised 28 June 1995; accepted 5 July 1995.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 18 Jun 2017 19:03:13