The Role of Saliva and Serum in Candida albicans Biofilm Formation on Denture Acrylic Surfaces

14 Pages·2009·2.18 MB·English

MICROBIAL ECOLOGY IN HEALTH AND DISEASE VOL.9 : 3548 (1996)
The Role of Saliva and Serum in Candida albicans
Biofilm Formation on Denture Acrylic Surfaces
H. NIKAWA*t, H. NISHIMURAt, T. YAMAMOTOt, T. HAMADAt and L. P. SAMARANAYAKES
TDepartment of Prosthetic Dentistry, Hiroshima University School of Dentistry, 1-2-3 Kasumi Minami-ku,
Hiroshima 734 Japan; $Oral Biology Unit, University of Hong Kong, Prince Philip Dental Hospital, 34 Hospital
Road, Sai Ying Pun, Hong Kong
Received 22 June 1995; revised 2 October 1995
The long term effect of either a salivary or a serum pellicle on Candida albicans biofilm formation on denture acrylic
surfaces was investigated both by quantifying the ATP (adenosine triphosphate) content of the resultant biofilms and
by scanning electron microscopy. When the biofilm formation on saliva-coated acrylic strips was examined, the
yeasts initially colonised this surface at a slower rate than the controls although with increasing incubation time, at
72 h, the ATP content was almost ten-fold higher than the protein-free control strips. Ultrastructural studies
revealed this to be due to cell aggregation and hyphal emergence, phenomena not observed in the controls. As
compared with the control strips, biofilm activity of the serum-coated strips was almost 100-fold greater within 48 h
incubation, and scanning electron microscopy revealed multilayer blastospore-blastospore co-adhesion, germ tube,
hyphal and pseudohyphal emergence and blastospore-hyphal coadherence. Further immunocytochemical observa-
tion revealed that concanavalin-A binding material and fibronectin were involved in biofilm formation on both
saliva and serum coated specimens and, in addition, mannan-binding protein and protein-A binding material also
contributed to the biofilm formation on serum coated specimens.
KEY WORDS: Candida ulbicuns; biofilm; saliva; serum; fibronectin; mannan-binding protein
INTRODUCTION protein adsorbed s~rfaces,~an~d. ~it~ h as been
shown in our studies that salivary and serum
Candida albicans is frequently recovered from den-
pellicles may promote fungal colonisation on den-
ture fitting surfaces which act as a reservoir of
ture lining materials.24 Others have shown that
infection in Candida-associated denture stomatitis
mucinous glycoproteins of human saliva promote
(syn. chronic atrophic ~andidiasis).~A.l~th ough
adhesion of C. albicans to polymethylmethacrylate
successful candidal colonisation of the denture
surface has been recognised as an important step by sugar-specific interactions. lo
In contrast to these observations, a number of
in the pathogenesis of this condition,32 the role
workers have demonstrated that pretreatment of
played by saliva or serum pellicles during the
acrylic strips and/or yeast cells with whole saliva
colonisation process is poorly understood.
decreased the initial adherence of C, albicans to
Components of saliva or serum proteins, such as
mucins, fibrinogen and complement, specifically denture ac' 41l ic, whereas a serum pellicle promoted
adherence.'
bind to Candida blastospores and germ *'9934
It is therefore evident that the relationship be-
t ~ b e s . ~ ,Sp~e,ci~fic. ~int~er actions between manno-
tween the salivary or serum pellicle on denture
protein adhesin of C. albicans and sugar-moiety
surfaces and candidal colonisation has been a
of salivary proteins, including mucins, have also
subject of controversy. This may be attributed in
been demonstrated during candidal adherence to
part to the diverse and complex attributes of saliva
and/or serum, which modulate clearance, aggrega-
*Author to whom correspondence should be addressed at: tion and adherence of microorganisms, and also to
Department of Prosthetic Dentistry, Hiroshima University
various microbial factors, such as specific and non-
School of Dentistry, 1-2-3 Kasumi Minamiku, Hiroshima 734
Japan. specific interactions, cell-growth and coadhesion,
CCC 089 I-O60)[/96/0 10035-14
0 1996 by John Wiley & Sons, Ltd.
36 H. NIKAWA ET AL.
which are involved in fungal colonisation of sur- Standard (JIS). A smooth surface was obtained
faces. Further, there are little data on the long term by compressing the mixture onto glass slides.
activity of Candida biofilms on mucinous surfaces The processed acrylic sheets were cut into
once the initial attachment process is complete 10 x 10 x 0.7 mm pieces.
and it is not known whether the quality of the
primary mucinous layer (i.e. saliva or serum) Saliva and serum
affects the subsequent growth and colonisation of
Unstimulated whole saliva was collected from
the adherent yeasts.
a healthy donor and clarified, according to the
For the current experiments we have adapted
method of Gibbons et al. l2 with modification,
a bioluminescent adenosine triphosphate (ATP)
by centrifugation at 12 000 g for 15 min at 4°C.
assay based on firefly luciferase-luciferin system,
Human serum was purchased from Sigma
which is known as a simple and convenient method
Chemical Co. (St Louis, MO, USA). Whole saliva
for accurate enumeration of viable cells'5 and has and serum were stored at - 25°C before use.
been used previously for gram-positive and gram-
negative bacterial biofilm.'-" The purpose of the
Cell growth and ATP (adenosine triphosphate)
present study was to analyse in detail the role of
analysis
the salivary and serum pellicles in the initial and
To examine the interrelation between yeast
subsequent stages of C. albicans biofilm formation
growth and adenosine triphosphate (ATP) con-
on acrylic surfaces. Preliminary immunocyto-
tent, 100 p1 of yeast suspension, 250 pl of either
chemical studies were also performed to investigate
saliva or serum and 4-0 ml of Sabouraud dextrose
the nature of the Candida biofilm on protein-
medium were dispensed into a series of sterile glass
coated surfaces.
tubes. In the control tubes, saliva or serum was
replaced by an equal volume of sterile distilled
MATERIALS AND METHODS water. These tubes were incubated at 37°C for 6,
12, 24, 48, 60, 72, 96, 120, 144 and 168 h. After
Microorganisms and growth conditions
each of these intervals, the tubes were carefully
Candida albicans GDH 18 and GDH 19, oral vortexed, 1.0ml of each sample was mixed with
isolates obtained from the routine microbiology 4.0 ml of the reagent containin? benzalkonium
services of the Glasgow Dental Hospital and which extracts intracellular ATP3 and allowed to
School, were used in this study. The isolates were react for 5 min in an ultrasonicator. Then the
identified by sugar assimilation test using the API solution was clarified by filtration (pore size
20C system (API Products, Biomeroux, Lyon, 0.45 pm) and the amount of ATP was quantified
France) and 'germ tube' test.36 A loopful of the using a bioluminescence apparatus (ATPA-1000,
yeast was inoculated in yeast nitrogen base me- TOA Electronics Ltd, Tokyo, Japan).' This appa-
dium (Difco, Detroit, USA) containin 250 mM ratus used the firefly-luciferases ystem to determine
glucose and grown aerobically at 37°C.2 8 After the the concentration of cellular ATP and is based
overnight culture, the yeast was harvested in the upon the measurement of light emission produced
late exponential growth phase, washed twice with during the oxidation of luciferin by molecular
sterilised distilled water and resuspended to a final oxygen in the presence of ATP and magnesium
concentration of 105cells/ml by a s ectrophoto- ions. In this system, the light intensity is directly
meteric and haemocytometric means.P 4 proportional to the concentration of ATP.37 A
conventional growth curve was constructed by
Fabrication of acrylic strips haemocytometer counting of yeasts in the control
cultures, which did not exhibit either germ tube
Heat-cured denture acrylic sheets (50 x 50 x
or hyphal formation. The relationship between
0.7 mm) were fabricated according to conventional
the yeasts/ml and ATP content was subsequently
prosthodontic techniques. Briefly, denture acrylic
established.
poly (methylmethacrylate) powder and monomer
liquid (Bio Resin, Shofu, Kyoto, Japan) were
mixed according to manufacturer's directions. The Assay procedures
mixture was packed into the flask, processed in The colonisation assay was conducted as fol-
water tank at 70°C for 90min and then 100°C lows. The acrylic strips were coated with saliva or
for 30min, according to Japan Industrial serum by placing them in wells of Multiwell tissue
SALIVA, SERUM AND CANDIDA BIOFILM 37
culture plates (NunclonR Delta, Nunc, Kamstrup, gold colloidal particle (E-Y Lab. Inc., San Mateo,
Denmark), into which were dispensed 500 p1 of the CA; particle size 20nm, final protein concen-
protein solution per well, and incubating for 1 h at tration 3 pglml) for 1 h at room temperature
37°C. Saliva or serum was substituted with an according to the manufacturer’s direction.
equal volume of sterile distilled water in the con- To examine whether IgG or IgM is involved in
trol wells. After incubation the protein solution fungal biofilm formation or not, each specimen
was aspirated, 50 pl of yeast suspension (1 x lo5 was removed from the wells after the assay (168 h),
cells/ml) was inoculated into each well and the washed with 10mM PBS and immersed in PBS
whole assembly was incubated at 37°C for 2 h to containing 1 per cent bovine serum albumin and
promote yeast adherence and colonisation. Subse- 1: 1 0 diluted protein-A-gold colloidal particle (E-Y
quently, 2.0 ml of Sabouraud broth was carefully Lab. Inc., San Mateo, CA; particle size 30 nm)
dispensed into each well, and incubated for 0, 6, for 1 h at room temperature according to the
12, 24, 48, 60, 72, 96, 120, 144 and 168 h at 37°C. manufacturer’s direction.
Afterwards each specimen was carefully removed, To observe the localisation of fibronectin or
washed thoroughly by rinsing three times for mannan-binding protein, immunocytochemical
a total of 60s with distilled water to remove staining was carried out according to the method
loosely adherent organisms, immersed in 5.0 ml of Bouchara et al.3 Each specimen was removed
of the extraction-reagent (benzalkonium; 37) and from the wells after the assay, washed with 10 mM
allowed to react for 5min with ultrasonication. PBS and immersed in PBS containing 1 per cent
The resultant reagent solution was then filtered bovine serum albumin and either 1:lOO diluted
(pore size 0-45 pm) to clarify, and subjected to mouse antihuman mannan-binding protein mono-
ATP-measurement. To monitor the yeast growth, clonal antibody (IgG1, 131-1, Serotec) or 1:lOO
the total amount of ATP in each well was also diluted mouse antihuman fibronectin monoclonal
quantified. antibody for 30min at room temperature. Each
The assays were carried out on two independent specimen was then washed with PBS and incu-
occasions, with quadruplicated samples on each bated in PBS containing 1 per cent bovine serum
occasion. All the numerical data obtained were albumin and 1:lO diluted goat antimouse IgG-
analysed by analysis of variance (ANOVA) and gold colloidal particle (E-Y Lab. Inc., San Mateo,
Tukey’s multiple range test at 5 and 1 per cent CA; particle size 20 nm) for 30 min at room
levels. temperature.
Specimens were next washed with PBS, fixed in
Ultrastructural observations 2.5 per cent glutaraldehyde and dehydrated as
described above. Each specimen was then sputter-
For ultrastructural studies, specimens were re-
coated with a layer of gold to a thickness of
moved from the wells after the assay, washed with
10 nm3’ and observed under a Scanning Electron
sterile distilled water and fixed in 2.5 per cent
Microscope (JMS-6300, Joel, Tokyo, Japan) using
glutaraldehyde and 1.0 per cent osmium tetroxide.
standard procedures.
Afterwards they were dehydrated through a
graded series of ethyl alcohol (50-90 per cent),
immersed in t-butyl alcohol (three times for RESULTS
10min) stored at 4°C. Each specimen was then
The preliminary experiments were performed to
freeze dried, sputter coated with a layer of gold to
establish the relationship between C. albicans
a thickness of 15-20nm and observed under a
growth in saliva or serum supplemented Sabour-
Scanning Electron Microscope (JMS-6300, Joel,
aud medium, or the latter alone (control), and the
Tokyo, Japan) using standard procedures.
ATP content. As enumeration of yeasts in the
saliva and serum supplemented media was difficult
Immunocytochemical observations
due to mixed blastospore and hyphal phase
For concanavalin-A staining, each specimen was growth, the number of yeasts in the control tube
removed from the wells after the biofilm assay (which grew solely in the blastospore phase) was
(168 h), washed with 10 mM PBS (pH 7.4; contain- used to correlate the cell growth and the ATP
ing 0-8 per cent NaCl, 1 mM CaCI,, 1 mM MgC1,) content of the test media.
and immersed in PBS containing 1 per cent bovine As shown in Figure 1 a to c, there was a good
serum albumin and 1: 10 diluted concanavalin-A- correlation between the yeast numbers or the yeast
38 H. NIKAWA ET AL
and the controls, with increase in cell numbers up
to 48 h and reaching a plateau phase thereafter.
No significant difference in the ATP content was
observed either in the test or the control suspen-
sions, at any incubation interval (GDH 18;
ANOVA; P>0.05). On regression analysis a highly
-cel I number significant positive correlation was observed be-
tween the ATP content of the yeasts and the cell
-dist H,O\,
J numbers, in the test samples with saliva or serum
__ a~; :: ::tent or, the control samples (GDH 18; Figure 1 b,c;
I I I Student t-test; P<O-OI, ~0.98-0-99).S imilarly, a
48 $2 96 120 144 1168
significant correlation between the ATP content of
lncubat ion t ime (hrs)
the yeasts and the cell numbers, in the test samples
- with saliva or serum, or in the control samples, was
b also observed with C. albicans GDH 19 (data not
6,
shown). In all the experiments, the extracellular
ATP content in each sample was less than 0.5 per
cent of intracellular ATP, indicating the reliability
of the ATP measurement.
As these experiments indicated that the ATP
content was directly proportional to the number of
0-0 dist HlO(r=O. 992) yeasts, irrespective of their morphology or the
A ..... ~ A saliva (r=O. 981) growth phase, the former parameter was utilised to
A--& serm (r=O. 994) evaluate the long term growth characteristics and
I I I biofilm formation of C. aibicans on saliva- and
3 5 7 0
serum-coated acrylic strips.
- log (number of cells)
The effect of different protein pellicles coating
the denture material on the growth of these isolates
C
in the suspending medium, and the biofilm devel-
opment on acrylic strips are shown in Figure 2 a,b.
As expected, the yeast growth curves of the test
and the control specimens (containing acrylic
:j
strips) described an almost identical path compar-
able to the foregoing preliminary study (Figure
vn+<- /0.996 2 a,b). Thus, in all three suspensions (containing
the control, saliva- and serum-coated strips), a
M o
-0 r ’ i i 4 i i rapid phase of growth was observed during the
first 12 h, which steadily levelled off to a plateau
log (number of cells) after 48 h, beyond which no further growth was
Figure 1. (a) The growth curve of C. albicum GDH 18 in the perceptible.
control (Sabouraud medium), saliva- and serum-supplemented When the C. dbicans GDH 18 biofilm activity
media as assessed by ATP measurement and haemocytometric
on the control and serum-coated strips was com-
cell counting. (b) The relationship between the cellular ATP
content and haemocytometric cell counts of C. albicunr GDH pared by the ATP assay method, the activity
18 in the control (Sabouraud medium), saliva- and serum- increased rapidly up to 12 h at an exponential rate,
supplemented media. Each experiment was repeated on two closely resembling the yeast growth curves in the
independent occasions with quadruplicated determinations on
suspending medium. However, the increase in the
each occasion. (c) The relationship between the cellular ATP
content and haemocytometric cell counts of C. albicunr GDH biofilm activity of the saliva-coated strips was
18 at the ranges of 40-105 cells relatively slow (GDH 18; Figure 2 a). The increase
in the biofilm ATP content on the saliva- and
serum-coated strips continued steadily up to about
growth in all three media and the ATP content. 72 h although the control biofilm activity peaked
Typical growth characteristics of C. aibicans were after 12 h, reaching a plateau thereafter (Figure
observed in the salivalsemm supplemented media 2 a), with a slight decrease in recorded values.
SALIVA, SERUM AND CANDIDA BIOFILM 39
a Candida albicans CDlf 18 ATP content of
yeas-t susp ension
cont ro I
-............ s aliva
serum
,+/A-~-A-A-~
,
A-A ser m
0
I I I I I
061224 48 72 96 120 144 lA8
lncubat ion time (hrs)
ATP content of
yeast suspension
...v --.. ........-
,V
All' content of
biof i In
cont ro I
v----.vsal iva
-werum
061224 48 72 96 1bO 1b4 lk8
lncubat ion time (hrs)
Figure 2. (a) The growth curve and biofilm activity of C. albicuns GDH 18 on
the control (uncoated, saliva- and serum-coated acrylic strips in Sabouraud
medium as assessed by ATP analysis. The test samples of acrylic were coated
with a pellicle of either saliva or serum prior to incubation. (b) The growth
curve and biofilm activity of C. albicuns GDH 19 on the control (uncoated),
saliva- and serum-coated acrylic strips in Sabouraud medium as assessed by
ATP analysis. The test samples of acrylic were coated with a pellicle of either
saliva or serum prior to incubation. Each experiment was repeated on two
independent occasions with quadruplicated determinations on each occasion
As to the C. al6icans GDH 19, the biofilm activity of the control biofilm was observed after
activity on the control, saliva- or serum-coated 24 h (Figure 3 a), decreased significantly at 48 h,
strips increased rapidly up to 12 h at an exponen- and plateaued thereafter.
tial rate, resembling the yeast growth curves in the When the relative ATP content of the 24 h and
suspending medium. Although the rate of increase 72 h biofilms of both C, aZ6icans GDH 18 and
in the biofilm activity in the saliva-coated strips GDH 19 were compared, a serum pellicle was
slowed down earlier than its counterparts (GDH significantly more effective in promoting fungal
19; Figure 2 b), the biofilm ATP content on both biofilm formation than a saliva pellicle (24 h and
the saliva- and serum-coated strips increased 72 h, ANOVA; P<O.Of ). Interestingly, in the case
steadily up to about 72 h. In contrast, the peak of C. albicans GDH 18, saliva pellicle suppresses
H. NIKAWA ET AL
0 O-l2hrs a 24 h, ANOVA; P<O.Ol). None the less this behav-
12-48hrs iour pattern changes as the biofilm matures. Thus,
48-168hrs at 72 h the biofilm on the serum-coated strips
1 exhibited more than 100-fold increased cellular
kinetics than the control, uncoated strips, and the
activity on the saliva-coated strips exceeded that of
the controls by a factor of ten.
The relative growth rate of yeast suspension and
biofilm formation rate of C. albicuns GDH 18
during the initial (0-12 h), mid (1248 h) and the
final (48-168 h) phases of the study were further
L compared (Figure 3 a). In the absence of a protein
pellicle (control samples) the activity of the biofilm
5
-5 -.€- 4-z = --E .Cz- stolo wyes adstosw onn ctohnes idsaelriavbdlsye raufmter- co1a2t hed assp eocpipmoesneds,
L0L00 r.no- .0>mr- cILm0 -.+n-0 -4.>0m-- DLI0o areltvheorsueg hw adsu roibnsge rtvheed f. inIna l thpeh apseel loicfl et hceo astteudd ys atmhe-
con t ro I saliva serum ples the biofilm metabolism was relatively higher
samp I es samp I es samp I es or similar during the mid phase of the experiment
(12-48 h, Figure 2) when compared with yeast
0 O-12hrs b
growth rate in either the test or the control sus-
12-48hrs
I 48-168hrs pensions. The saliva biofilm demonstrated a
2 significantly high activity during the final phase
(48-168 h) of the study when compard with either
the serum-coated (PeO.01) or control samples
(P<0-05)o,r the growth parameters in any of the
suspensions (WO.01) (Figure 3 a).
Results similar to C. albicuns GDH 18 were
obtained when the growth rate of yeast suspen-
sions and biofilm formation characteristics of
C. albicans GDH 19 were compared, except that,
during the initial period of the study (0-12 h), no
significant differences in the foregoing parameters
---
were observed between the control, saliva- or
serum-coated specimens (Figure 3 b).
SEM Observations
cont r o I saliva serum
sarnp I es samp I es samp I es The processes of biofilm formation was similar
Figure 3. (a) Comparison of C. albicunr GDH 18 biofilm with both C. albicans isolates. During the initial
activity on the control (uncoated), saliva- and serum-coated phase, small numbers of C. albicuns blastospores
acrylic strips and the growth rates on the respective specimens
were seen sparsely attached to both the control and
during the initial (0-12 h), mid (1248 h) and late (48-168 h)
saliva-coated strips and hyphal emergence was
phases of the incubation periods. Asterisks refer to the signifi-
cant differences between the control and test samples (P<O.OS). seldom observed (Figure 4 a,b). in contrast, a large
(b) Comparison of C. albicans GDH 19 biofilm activity on the number of blastospores with and without germ
control (uncoated), saliva- and serumcoated acrylic strips and tubes colonised the serum-coated specimens,
the growth rates on the respective specimens during the initial
though the proportion with germ tubes was very
(0-12 h), mid (1248 h) and late (48-168 h) phases of the
incubation periods. Asterisks refer to the significant differences low during this period (Figure 4 c). During the mid
between the control and test samples (P<O-OS) to late phase (12-168 h), the control specimens
were sparsely colonised with C. albicans blasto-
spores (Figure 4 d), these exhibited blastospore-
or shows no effect on biofilm formation when blastospore coadhesion on longer incubation (not
compared with the protein-free control specimen, shown). During the latter period of the study
during the initial stages of colonisation (i.e. 12- a large population of colonised yeasts also
d
e
at
o
c
m-
u
r
e
s
d
n
a
b)
(a-
v
ali
ntrol (a), sstrips ylic
he co0 acr
t(
n d
oe
n at
oo
atim-c
s
onieru
cold s
18 an
H e)
GDva- (
ns ali
us
bicd),
ulC. rol (
h) of 2 he cont
1t
nitial phase (on onisation
he icol
of s th) of
graphe (72
crohas
ctron mihe late p
et
eld
g n
na
annirips
4. e Scrylic st
rc
ua
Fig(c)
42 H. NIKAWA ET AL.
demonstrated blastospore-blastospore co-adhesion be noted that the phenomenon of adherence may
on the saliva-coated specimens (Figure 4 e), and, in represent only the first. step in the colonisation
addition, hyphal emergence and profuse biofilm processf4 which, as time progresses, leads to a
formation was also observed in the serum-coated formation of a thin biofilm and then a multilayer,
specimens (Figure 4 f). The latter population com- climax community of plaque.
prised a dense network of yeasts, germ tubes, A deeper understanding of adherence and the
relatively young hyphae, and pseudohyphae subsequent behaviour of Candida biofilms requires
(Figures 4f and 5 a to c). Further observation first, the examination of sequential samples over a
revealed that the extracellular matrix mediated period of time and, more importantly, a method
the blastospore-blastospore and/or blastospore- for accurate quantification of yeast biofilm forma-
hyphae co-adhesion on both the saliva- and the tion which may exhibit dimorphic growth patterns
serum-coated acrylic strips (Figure 5 a). In addi- as well as co-adhesion, aggregation and multilayer
tion, hyphae in contact with the surface of acrylic growth over a prolonged period of col~nisation.~~
strips occasionally produced strandlike adhesive The main impediment for such research to date has
material, which has not been previously reported been the lack of a relatively simple assay system to
(Figure 5 b,c). The nature of this surface materials accurately quantify the cell growth. Recently,
which contribute to biofilm formation was evalu- Hawser and Douglasf3 reported fungal biofilm
ated using Con-A, Protein-A, anti-fibronectin formation on catheter material by using dry
monoclonal antibody and anti-mannan binding weight, colorimetric (MTT) and radiometric
protein monoclonal antibody. assays. However, in preliminary studies, we found
With C. albicans GDH 18, Con-A-gold colloidal that the dry weight and MTT assay methods were
staining revealed that saliva- or serum-biofilms less sensitive in detecting the early stage of yeast
were more reactive to Con-A binding as compared colonisation. Further the results of MTT assay
with control biofilms (Figure 6 a to c). Further, were affected by the salivary and serum proteins
ultrastructural studies indicated that fibronectin is (data not shown). Hence we adopted another assay
involved in the biofilm formation on saliva- or system, bioluminescent ATP assay, which has pre-
serum-coated specimens (Figure 6 d,e) and that viously been used to measure bacterial biofilm
human mannan-binding protein and protein-A activity, ',I1 Farber and Wolff" have conclusively
binding material (perhaps IgG and IgM) may shown that the result obtained by this ATP assay
contribute to fungal biofilm development on was consistent with that obtained by either con-
serum-coated specimens (Figure 6 f,g). Control ventional viable counts or radiolabelling methods.
specimen of both isolates were unreactive with any Using this method we noted an excellent correla-
of these materials (Figure 6 a), except for control tion between the yeast cells and the ATP content in
specimen of C. albicans GDH 19 which reacted either the test or the control samples. This was not
with Con-A. surprising as the assay was based on the funda-
mental principles of ATP analysis which states that
the amount of cellular ATP correlates with the dry
DISCUSSION
weight, the volume and the number of viable
A number of experimental approaches have been cell^,^' irrespective of the morphological attributes
made to examine the mechanisms of C. albicans of the yeast.
adherence to solid surfaces, such as denture It is known that some salivas suppress candidal
acrylic.8~'o~'7*20~30~E32v~e3n9 the earliest investiga- growth while others do not33 and hence it is
tors of this topic, using visual quantification of possible that our observations on the lack of effect
adherent yeasts, reported the high affinity of of saliva on fungal growth (Figures 1 and 2) could
C. albicans to denture acrylic and modulation of either be due to the quality of the saliva used
this attachment process due to saliva and serum and/or its high dilution in the incubation medium.
pelli~les.~'.A'~lt hough the latter groups observed A similar explanation could be offered for growth
the initial suppression of C. albicam adherence in serum containing medium as the latter is known
on saliva-coated denture surfaces,32 others have to affect candidal cell kinetics in a variety of
reported the opposite effect.39 While these may ways.'' These findings confirm the results of our
reflect the modulation of candidal adhesion due to previous study where pellicles of salivary or serum
variables such as the quality of the saliva,38 the proteins coating denture material did not essen-
yeast isolates and the growth media, it should also tially affect the rate of fungal growth in the
43
SALIVA, SERUM AND CANDIDA BIOFILM
44 H. NIKAWA ET AL.