1444 DOI 10.1002/pmic.200900611 Proteomics 2010, 10, 1444 1454
RESEARCH ARTICLE
Proteomics of drug resistance in Candida glabrata
biofilms
C. Jayampath Seneviratne1, Yu Wang2, Lijian Jin1, Y. Abiko3 and Lakshman P. Samaranayake1
1
Oral Biosciences, Faculty of Dentistry, The University of Hong Kong, Hong Kong
2
Department of Pharmacology and Pharmacy, The University of Hong Kong, Hong Kong
3
Biochemistry and Molecular Biology, Nihon University School of Dentistry at Matsudo, Chiba, Japan
Candida glabrata is a fungal pathogen that causes a variety of mucosal and systemic infections Received: August 21, 2009
Revised: December 15, 2009
among compromised patient populations with higher mortality rates. Previous studies have
Accepted: December 20, 2009
shown that biofilm mode of the growth of the fungus is highly resistant to antifungal agents
compared with the free-floating or planktonic mode of growth. Therefore, in the present
study, we used 2-D DIGE to evaluate the differential proteomic profiles of C. glabrata under
planktonic and biofilm modes of growth. Candida glabrata biofilms were developed on
polystyrene surfaces and age-matched planktonic cultures were obtained in parallel. Initially,
biofilm architecture, viability, and antifungal susceptibility were evaluated. Differentially
expressed proteins more than 1.5-fold in DIGE analysis were subjected to MS/MS. The
transcriptomic regulation of these biomarkers was evaluated by quantitative real-time PCR.
Candida glabrata biofilms were highly resistant to the antifungals and biocides compared with
the planktonic mode of growth. Candida glabrata biofilm proteome when compared with its
planktonic proteome showed upregulation of stress response proteins, while glycolysis
enzymes were downregulated. Similar trend could be observed at transcriptomic level. In
conclusion, C. glabrata biofilms possess higher amount of stress response proteins, which
may potentially contribute to the higher antifungal resistance seen in C. glabrata biofilms.
Keywords:
2-D DIGE / Biofilm / Candida glabrata / Microbiology / Stress response
1 Introduction glabrata infections have the highest mortality rate among
infections due to non-albicans Candida species. The repor-
Candida glabrata has emerged as a major pathogen among ted mortality associated with C. glabrata candidemia is
compromised patient groups such as HIV/AIDS patients, approximately 50% among cancer patients and as high as
transplant recipients, and patients receiving chemotherapy 100% among bone marrow transplant patients [4].
[1 3]. Candida glabrata infections rank second to those of Candida infections primarily begin with adherence and
Candida albicans among all forms of candidiasis [4], and C. colonization on an artificial or a biotic host surface. This
process leads to the formation of surface-attached commu-
nities known as biofilms, which are structured communities
Correspondence: Professor Lakshman P. Samaranayake, Oral
of microorganisms that are encased in a matrix of exopoly-
Biosciences, Faculty of Dentistry, The University of Hong Kong,
meric substances. Biofilms of microorganisms display
Prince Philip Dental Hospital, 34 Hospital Road, Hong Kong
unique characteristics that confer a survival advantage over
E-mail: [e-mail]
their planktonic or free-floating counterparts [5]. Formation
Fax: [phone]
of biofilms is the predominant mode of growth of micro-
Abbreviations: CLSM, confocal laser scanning microscopy; MIC,
organisms in nature, and it is estimated that at least 65% of
minimum inhibitory concentration; SEM, scanning electron
all microbial infections are related to biofilms [6]. Previous
microscopy; Q-RT-PCR, quantitative real-time PCR; XTT, tetra-
studies have shown that biofilm-forming ability is a major
zolium salt 2,3-bis(2-methoxy-4-nitro-5-sulfophenyl)-5-[(pheny-
virulence attribute of C. glabrata, and a major contributing
lamino)carbonyl]-2H-tetrazolium hydroxide; YNB, yeast nitrogen
factor is higher antifungal resistance [7 10]. Candida
base medium
& 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.proteomics-journal.com
Proteomics 2010, 10, 1444 1454 1445
glabrata possesses both innate and acquired resistance 2.3 Biofilm formation
against antifungal drugs due to its ability to modify ergo-
sterol biosynthesis, mitochondrial function, or antifungal Yeast cells were resuspended in YNB supplemented with
efflux, and higher antifungal resistance within biofilms 100 mM glucose and adjusted to a cell density of 1.0 107
makes C. glabrata biofilms even more difficult to manage. cells/mL to achieve optimal biofilm formation [14, 15].
The molecular mechanisms underpinning drug resistance Candida glabrata biofilms were formed in wells of polystyrene
of C. glabrata within biofilms are still elusive but need to be culture plates (Iwaki, Tokyo, Japan) as previously described
understood to allow the development of new management [16]. For the antifungal susceptibility assays, 100 mL of the
strategies against C. glabrata infections. One can utilize the 1.0 107 cells/mL cell suspension was placed in each well of
  hypothesis-free  system biology tools of   proteomics, 96-well plates. For all other experiments, 1 mL of cell
genomics, and bioinformatics  to investigate the molecular suspension was placed in each well of 12-well plates. As
machinery of the biofilm mode of the growth of C. glabrata negative controls, some wells received no Candida suspen-
and hence can propose a possible mechanism of higher sion. Plates were first incubated at 371C in a shaker at 75 rpm
antifungal resistance. So far, only a few studies have explored for 1.5 h to allow yeast to adhere to the well surface. The
the proteome of C. glabrata [11, 12], and there appear to be no medium was then removed by aspiration, each well was
studies on C. glabrata biofilms. In a previous study, we washed with 1 mL of PBS to remove non-adherent cells, and
performed a comparative proteomic analysis on the plank- 2 mL of YNB with 100 mM glucose was pipetted into each
tonic and biofilm modes of C. albicans [13]. We found that well. Plates were then incubated at 371C with shaking for
C. albicans biofilms are associated with increased anti-oxida- 48 h. For the proteomics and transcriptomic analyses, age-
tive capacities, which we hypothesized are associated with matched planktonic cultures in test tubes were prepared in
higher antifungal resistance during the biofilm growth. parallel and incubated in an orbital shaker at 371C.
Therefore, the aim of this study was to compare the proteome
of planktonic and biofilm modes of C. glabrata to explore the
mechanisms that might contribute to the virulence of this 2.4 Antifungal susceptibility of planktonic and
organism. Furthermore, we compared the differential biofilm modes of C. glabrata
proteomic expression of C. glabrata with those of C. albicans
from our previous study in an attempt to obtain a wider view Antifungal susceptibility testing was performed on both
of higher antifungal resistance in Candida biofilms. planktonic and biofilm samples of C. glabrata. Four anti-
fungal drugs commonly used to treat oropharyngeal and
systemic candidiasis were selected for the study: caspo-
2 Materials and methods fungin (Merck, USA), which is a relatively new echino-
candin; nystatin, a polyene (Sigma, USA); amphotericin B, a
2.1 Organisms and growth conditions polyene (Sigma); and ketoconazole, an azole (Sigma). The
antifungal agents were prepared as previously described
We used two C. glabrata strains: a reference laboratory strain [17].
of C. glabrata ATCC 90030 from the archival collection of the For the planktonic samples of C. glabrata, the minimum
Oral Biosciences Laboratory of the Faculty of Dentistry, and inhibitory concentration (MIC) of each drug was determined
a wild-type strain C. glabrata Cg-5, which has been according to the relevant protocol of the Clinical and
previously characterized [9]. The identity of the yeast isolates Laboratory Standards Institute [18]. Antifungal susceptibility
was confirmed using the commercially available API32C for C. glabrata biofilms was performed using standard XTT
identification system (bioMérieux, Marcy l Etoile, France). reduction assay, as described previously [16, 19]. In brief,
Candida glabrata strains were subcultured on Sabouraud s after 48 h of biofilm formation, the medium was removed
dextrose agar (Gibco, Paisley, UK) and maintained at 41C and biofilms were washed with 100 mL PBS to remove non-
during the experimental period. The purity of the cultures adherent cells. The stock solutions of drug were diluted with
was confirmed periodically by Gram staining and the germ- RPMI 1640, supplemented with 2% glucose to obtain drug
tube test. concentrations ranging from 100 to 0.1 mg/mL for caspo-
fungin, 240 to 0.225 mg/mL for amphotericin B, and 64 to
0.125 mg/mL for nystatin and ketoconazole. After 100 mL of
2.2 Preparation of standard yeast cell suspension drug solution had been added to each well, plates were
incubated at 371C for 24 h and the metabolic activity of
Candida were grown in Sabouraud s dextrose agar at 371C fungal cells was determined by the XTT assay. The MIC
for 18 h and then inoculated in yeast nitrogen base medium for drug activity against Candida biofilms was defined as
(YNB, Difco, USA) supplemented with 50 mM glucose [14]. the lowest drug concentration that reduced the optical
After overnight culture in a rotary shaker at 75 rpm, yeast density in XTT readings by 50% compared with the drug-
cells were harvested in the late exponential growth phase free control. Each experiment was repeated three times
and washed twice with 20 mL of PBS (pH 7.2, 0.1 M). with four replicates.
& 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.proteomics-journal.com
1446 C. J. Seneviratne et al. Proteomics 2010, 10, 1444 1454
2.5 Oxidative stress tolerance of planktonic and care, USA. Sequencing-grade trypsin was from Promega
biofilm modes of C. glabrata (Madison, WI, USA).
For sample labeling, 50 mg of the lysate was labeled with
Oxidative stress tolerance of the planktonic and biofilm 400 pmol of cyanine dyes, Cy3 or Cy5, according to standard
modes of C. glabrata was evaluated using two clinical protocols. Cy2 was used to label the internal standard, which
biocides: NaOCl and H2O2. The MICs of these two biocides comprised equal amounts of proteins from all samples. The
in planktonic and biofilm modes of C. glabrata were deter- labeling was terminated by the addition of 1 mL of 10 mM
mined with the relevant protocol of the Clinical and lysine. The labeled samples were then randomly mixed to
Laboratory Standards Institute and the XTT reduction assay, allow every gel to contain 50 mg each of Cy2-labeled internal
respectively. standard and Cy3- or Cy5-labeled samples. The labeled
samples were randomly mixed to allow every gel to contain
50 mg each of Cy2-labeled internal standard and Cy3- and
2.6 Scanning electron microscopy and confocal Cy5-labeled samples. For the first dimension separation, the
scanning laser microscopy labeling mixture was applied to three Immobiline DryStrips
(18 cm, pH 4 7 linear) by cup loading with a total running
Candida biofilms were developed for 48 h on polystyrene time of 55 kV/h of the IEF. The second dimension was
discs under similar conditions as described above and carried out with three 12.5% SDS-PAGE gels, and gel
processed for scanning electron microscopy (SEM) and images were subsequently acquired at the recommended
confocal scanning laser microscopy (CLSM), as described wavelengths by using a Typhoon 9410 high-performance gel
previously [13]. The topographic features of the biofilm were and blot imager (GE Healthcare). In total, nine images with
visualized by SEM (XL30CP; Philips) in high-vacuum mode good separation qualities were analyzed using DeCyder
at 10 kV, and images were processed for display with differential in-gel analysis and biological variation analysis
Photoshop software (Adobe Systems, Mountain View, CA, software programs (GE Healthcare). For analysis, a gel with
USA). For CLSM imaging, Candida biofilms were gently the most detected protein spots was chosen as the master
washed twice with PBS and stained with SYTO-9 and image, against which spots of all the other gel images were
propidium iodide stains in the Live/Dead BacLight Viability matched. The internal standards (Cy-2 images) were inclu-
kit (Molecular Probes, Eugene, OR, USA). Biofilms were ded in the analysis procedure to eliminate technical varia-
incubated with the two stains for 20 min in the dark at 301C tion. Highly reproducible protein spots that up- or
before the CLSM. Subsequently, images of the stained downregulated greater than 1.5-fold between planktonic
biofilms were captured using a CLSM imaging system versus biofilm proteomes with a Student s t-test p-value of
(FLUOVIEW FV 1000; Olympus, Tokyo, Japan). less than 0.05 were considered for MS identification.
2.7 Proteomic experiments 2.9 MS
The C. glabrata Cg-5 isolate was selected for comparative For protein identification, the cyanine dye-labeled gels were
proteomic studies using an optimized protein extraction subsequently stained with modified silver staining, as
method that had been used previously for proteomic studies described previously [22]. Differentially expressed spots were
of C. albicans biofilms [13]. Briefly, cell pellets were resus- selected after DeCyder analysis of the DIGE images and
pended in 500 mL of b-mercaptoethanol buffer [20] and an were excised from the gels. In-gel trypsin digestion was
equal volume of 0.5-mm diameter glass beads. Cells were performed manually according to a previously described
mechanically disrupted by seven cycles of shaking in a protocol [23]. The peptides were extracted from the gel
Vortex Genie-1 mixer (Scientific Industries, USA) for 30 s pieces, concentrated, and desalted with the ZipTip kit
followed by 3 min of cooling on ice. Cell extracts were (Millipore, USA), and 1 mL of each sample was mixed with
centrifuged for 13 200 rpm for 15 min and the supernatant 1 mL of matrix solution containing 0.01 g/mL CHCA for MS
was quantified for protein content by the Bradford assay analysis in a Voyager-DE STR MALDI-TOF MS system and
(Bio-Rad, Hercules, CA, USA). an ABI 4800 MALDI-TOF/TOF MS/MS (Applied Biosys-
tems, Foster City, CA, USA).
Peptide mass lists and peptide fragment sequences were
2.8 2-D DIGE generated for protein identification using Data Explorer
software version 4.9 and 4000 series ExplorerTM software V
2-D DIGE was performed as previously described [21]. All 3.5, respectively. For MS analysis, laser intensity of 2500 was
2-D DIGE reagents and equipment for isoelectric focusing used and eight sub-spectra with 50 shots each were acquired
(including IPGphor, Immobiline DryStrip kit, Immobiline for each sample spot. Calmix 1 and 2 (Applied Biosystems)
Drystrips (18 cm, pH 4 7 linear gradient), IPG buffer, and were used for external calibration with mass tolerance of
the Ettan DALT system) were purchased from GE Health- 100 ppm. Keratin and trypsin autodigestion peaks were
& 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.proteomics-journal.com
Proteomics 2010, 10, 1444 1454 1447
excluded. Peak filtering was from 800 to 4000 Da with S/N ABI PRISM 7900HT sequence detection system using Syber
filter 410 for database searching. For MS/MS analysis, laser Green (SYBR Green PCR master mix; Applied Biosystems).
intensity of 3100 was used. The precursor tolerance was The primers (Sigma) are listed in Table 1. Prior to the
0.2 Da and the MS/MS precursor resolution was set at 350. experiment, several housekeeping genes were tested for
Twenty-five sub-spectra with the total of 2500 shots were their stability across the samples. URA3 was selected as the
acquired for each sample spot with metastable suppressor most stable candidate and therefore used for normalization.
on. Mass tolerance of fragment ion was 0.1 Da. Each 20-mL PCR reaction contained 10 mL SYBR, 1 mL of
cDNA, 2 mL of primer mix, and 7 mL of double-distilled
water. Step 1 took place at 501C for 2 min, step 2 at 951C for
2.10 Database search 10 min, and step 3 at 951C for 15 s and 601C for 1 min for 40
cycles. Melting curve analysis and gel electrophoresis was
An in-house MASCOT v2.1 (Matrix Science) searching performed to confirm specificity of the product. Relative
engine (http://www.matrixscience.com/search_form_se- transcriptomic expressions of the selected genes were
lect.html) was used to identify candidate proteins against the analyzed using DDCt Pfaffl method taking their PCR effi-
NCBInr database (Fungi, protein entries 467215). These ciencies into account [25].
candidates were subsequently clarified with the C. glabrata
genome database (www.genolevures.org). The criteria
used in the search were as follows: modifications including 3 Results
Cys as S-carbamidomethyl derivate and Met as oxidized
methionine, one missed cleavage site, pI 3 0, and a protein 3.1 Antifungal susceptibility and oxidative stress
mass range from 10 to 500 kDa [21]. For protein identifica- tolerance of planktonic and biofilm modes
tion from both MS and MS/MS, probability scores were
taken to be significant if the p-value was o0.05. For all Antifungal susceptibility testing of the planktonic and
significant identifications, both protein and total ion scores biofilm modes of C. glabrata showed that Candida biofilms
were above or equal to the 95% confidence interval. To are considerably more resistant to antifungals than their
assign putative molecular functions for identified C. glabrata planktonic counterparts, irrespective of the class of the
proteins, we searched the following bioinformatics data- antifungal (Table 2). Moreover, C. glabrata biofilms were
bases: Saccharomyces cerevisiae genomic database (www. less susceptible to the clinical biocides (Table 2). Hence, it
yeastgenome.org/), C. albicans genomic databases (www. seemed that C. glabrata biofilms were more tolerant to the
candidagenome.org and http://www.broad.mit.edu/annota- oxidative stress than its planktonic counterparts.
tion/fgi/), and the MIPS database (http://mips.gsf.de/
genre/proj/yeast/index.jsp).
3.2 Microscopy of C. glabrata biofilms
2.11 Isolation of total RNA, cDNA synthesis, and Both SEM and CLSM images showed mature biofilm
quantitative real-time PCR architecture in C. glabrata biofilms by 48 h. Multilayered
C. glabrata biofilms were composed of blastospores
Total RNA was extracted from both C. glabrata Cg-5 and
ATCC planktonic and biofilm samples, as described
previously [24]. Briefly, the SV Total RNA isolation kit
Table 1. Primer sequences of the selected genes for transcript-
(Promega) was used according to the manufacturer s
omic analysis using Q-RT-PCR
instructions, and RNA purity and integrity were quantified
Gene Primer sequence
in a NanoDrop ND-1000 spectrophotometer (NanoDrop
Technologies). Gel electrophoresis was also performed to
URA3 F 50-GGGCTTTGACTGGCTAATAATGAC-30
URA3 R 50-CCAAGTGCATCGCCTTTATCA-30
verify intact RNA. cDNA was synthesized by RT-PCR at
PEP4 F 50-AAGAAGGAAAAATTGACCAAGGAA-30
431C for 90 min in a 20-mL reaction volume containing 1 mg
PEP4 R 50-GCCTTCTCATATTGACTCACGTACTT-30
of total RNA, 1 mL (200 U) of Superscript II (Gibco-BRL),
HSP12 F 50-TGAATCCTACGCAGACACTGCTA-30
0.5 mg of oligo dT-primer, first-strand buffer, 10 mM DTT,
HSP12 R 50-CGGCATCGTTCAACTTGGA-30
and 1 mM dNTPs. A control reaction was performed without
TRX1 F 50-CGAAAAGTTCGCTGCTGAATACT-30
reverse transcriptase for all the isolates to verify the absence
TRX1 R 50-TCTGGCAACTCGTCGACATC-30
of genomic DNA contamination. FBA1 F 50-CCAGCTTACGGTATCCCAGTTG-30
FBA1 R 50-TACCATCGTACCATGGCAATAGC-30
Quantitative real-time PCR (Q-RT-PCR) was used to
ENO1 F 50-GTGTCATGGTTTCCCACAGATCT-30
evaluate gene expression levels of the proteins that had
ENO1 R 50-AGTTCTCAAACCGACGACCAA-30
shown significant up- or downregulation at transcriptomic
GPM1 F 50-GCTGACTCCCCATACTCTCAAAA-30
level. Tests were carried out in duplicate in at least three
GPM1 R 50-TCAATGACCAAAGCCAAAGATTC-30
separate experiments. Q-RT-PCR was carried out with the
& 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.proteomics-journal.com
1448 C. J. Seneviratne et al. Proteomics 2010, 10, 1444 1454
Table 2. Antifungal and clinical biocide susceptibility of plankto- considering the availability of annotated information and
nic and biofilm modes of C. glabrata
evolutionary proximity of C. glabrata and S. cerevisiae, iden-
tified proteins were named according to the S. cerevisiae
Minimum inhibitory concentration
database (www.yeastgenome.org).
C. glabrata Planktonic Biofilm
Biological data mining with bioinformatics databases
strain mode mode
revealed that the studied proteins belonged to several func-
tional categories, including stress-response proteins,
Antifungal
enzymes related to carbohydrate metabolism, and trans-
Caspofungin ATCC 0.2 100
porters (Table 3). Among the upregulated protein biomar-
Cg5 0.8 4100
Amphotericin B ATCC 0.23 16 kers were a number of stress-response proteins such as heat
Cg5 0.93 32
shock protein-12 (Hsp12p), cytoplasmic thioredoxin iso-
Nystatin ATCC 2 16
enzyme (Trx1p), alkyl hydroperoxide reductase (Ahp1p),
Cg5 4 32
vacuolar aspartyl protease (Pep4p), aldehyde dehydrogenase
Ketoconazole ATCC 0.125 464
(Ald2p), and alcohol dehydrogenase isoenzyme III (Adh3p).
Cg5 0.25 464
Among the downregulated proteins were several biomarkers
5-FC ATCC 0.8 4420
corresponding to key enzymes involved in glycolysis. These
Cg5 1.6 4420
included fructose-1,6-bisphosphate aldolase (Fba1p), glycer-
Biocide
aldehyde-3-phosphate dehydrogenase (Thd3p), phosphogly-
NaOCl ATCC 0.01% 0.30%
cerate mutase (Gpm1p), enolase (Eno1p), and alcohol
Cg5 0.01% 0.30%
dehydrogenase (Adh1p). On the contrary, transaldolase,
H2O2 ATCC 0.02% 2.10%
Cg5 0.02% 4.30% which belongs to the non-oxidative branch of pentose
phosphate pathway, was identified among the upregulated
MIC is indicated as mg/mL for antifungals and as a percentage
proteins.
(%) for biocides.
Some of the proteins, such as enolase (Eno1p), transal-
dolase (Tal1p), fructose-1,6-bisphosphate aldolase (Fba1p),
embedded in thick extracellular matrix in both SEM and
alcohol dehydrogenase (Adh1p), and glyceraldehyde-3-
CLSM images (Supporting Information Figs. 1 and 2).
phosphate dehydrogenase (Tdh3p) had different isoforms
with a slightly different Mr and pI (Fig. 1B, Table 2). These
results suggest that these biofilm-related proteins were
3.3 2-D DIGE profiles of C. glabrata biofilms
subjected to posttranslational modifications, which may
result in different functional assembly of the isoforms.
As shown in the representative images (Figs. 1A and B),
protein spots were evenly distributed across the 4 7 pI range
in the first dimension and between 15 and 160 kDa after
3.5 Transcriptomic analysis by Q-RT-PCR
12.5% SDS-PAGE in the second dimension. Approximately
900 spots were detected in each image by DeCyder DIA
To evaluate the relative transcriptomic expression of the
image analysis. Comparison of the differentially expressed
up- and downregulated proteins, Q-RT-PCR was carried
proteins in the biofilm and planktonic modes of C. glabrata
out using SyberGreen assay (SYBR Green PCR master
using the DeCyder software revealed that there were 17
mix; Applied Biosystems) for both C. glabrata Cg-5 and
upregulated and 7 downregulated protein spots (1.5 times or
C. glabrata ATCC strains, which we have previously shown
more) (Fig. 1C). These protein spots were examined further
to be comparable biofilm formers [26]. The transcriptomic
by MS/MS analysis.
level of all the selected genes in both the species showed
similar trends to those in the proteomic analyses. For
instance, HSP12, PEP4, and TRX1 were upregulated,
3.4 Tandem mass spectrometric identification of
whereas ENO1, FBA1, and GPM1 were downregulated at
differentially expressed proteins
the transcriptomic level, thereby corroborating proteomic
findings (Fig. 2).
MS and MS/MS were used to study the 17 upregulated and
7 downregulated proteins in C. glabrata biofilms (Fig. 1B,
Table 3). To avoid confusion regarding the identity of the
4 Discussion
proteins, nomenclatures of the proteins were retrieved from
four relevant bioinformatics databases (Table 3). Although
Candida glabrata is the second most prevalent fungal
C. glabrata falls in a similar genus to C. albicans, it exhibits
pathogen in humans after C. albicans. As compromised host
substantial differences in its structure and biological prop-
populations have been growing worldwide, the prevalence of
erties. Furthermore, in the taxonomic lineage, C. glabrata is
C. glabrata infections, particularly of systemic origin, has
closer to S. cerevisiae non-pathogenic yeast. Therefore,
also been increasing in the recent years [4]. Biofilm-forming
& 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.proteomics-journal.com
Proteomics 2010, 10, 1444 1454 1449
Figure 1. 2-D DIGE images (overlay (A) and silver stained (B)) of biofilm proteome of C. glabrata. (A) Protein lysates (50 mg) labeled with
Cy3, Cy5, or Cy2 (for internal control) were mixed and separated using pH 4 7 isoelectric focusing strips in the first dimension and 12.5%
SDS-PAGE in the second dimension. (B) Annotated spots are those found to be up- or downregulated at equal or over 1.5-folds between
biofilm and planktonic experimental groups. These proteins were subsequently identified by MS/MS (Table 3). (C) Three-dimenional
images of differentially expressed proteins of planktonic and biofilm 2-D DIGE gels analyzed by DeCyder software.
ability is known to be a major virulence attribute of biofilms are highly resistant to antifungal agents, as are
Candida species including C. glabrata. In the present study, C. albicans biofilms [16]. Thus, it is tempting to speculate
C. glabrata biofilms were much more resistant to four that a common mechanism exists in the biofilm mode of
clinically used antifungal agents than their planktonic Candida that confers higher antifungal resistance. Exploring
counterparts. Moreover, C. glabrata biofilms were more the molecular mechanisms of biofilm growth of these
tolerant than planktonic cells to the oxidative stress gener- Candida species may help in the understanding of the
ated by two clinically used biocides. These results agree with generic and specific pathways pertaining to higher anti-
those of previous studies on C. glabrata biofilms [7, 27]. fungal resistance, but there appear to be no studies in the
Hypothesized mechanisms to explain the higher antifungal literature on the proteomic profiles of C. glabrata biofilms.
traits seen in the biofilm mode of Candida include robust Limited data on biofilms of other Candida species suggest
biofilm architecture, decreased metabolic activity, altered that protein profiles are differentially expressed in proteome
gene expression, extracellular matrix, presence of   persister of biofilms and planktonic cultures [20, 32 34].
cells  , and higher anti-oxidative capacities [28 31]. However, Unlike traditional 2-DE, 2-D DIGE offers increased
the exact mechanism by which fungi acquire higher resis- capacity for spot matching and accurate quantitative analysis
tance in the biofilm mode of growth is yet to be elucidated. of multiple groups of samples with relative ease. Therefore,
Cellular imaging obtained by SEM and CLSM showed 2-D DIGE technology was used in the present study to
that C. glabrata biofilms are exclusively composed of examine differential protein expression between the two
blastospores that are devoid of hyphal cells, in contrast to growth modes of Candida. The 24 spots that were up- or
C. albicans biofilms, which show a predominance of hyphae. downregulated by more than 1.5 times were identified by
Hence, the architecture of the C. glabrata biofilms is both peptide mass fingerprinting and peptide fragment
distinctly different from that of C. albicans biofilms, as we sequencing (Table 3). Pathway analysis of the protein
have observed previously [13]. Nevertheless, C. glabrata changes suggested a downregulation of the glycolytic
& 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.proteomics-journal.com
1450 C. J. Seneviratne et al. Proteomics 2010, 10, 1444 1454
lated during stress conditions in a wide range of micro-
organisms [37, 40], including C. albicans [13]. Taken
together, the results on protein biomarkers imply the exis-
tence of a common pathway in the biofilm mode of both
C. glabrata and C. albicans species that confers higher anti-
oxidative capacity, regardless of architectural and morpho-
logical dissimilarities.
The C. glabrata biofilm proteome also showed differential
expression of other stress-response proteins  for instance,
Hsp12p, which is induced by various forms of stress,
including heat shock, oxidative stress, and osmotic stress
[41]. Previous studies have shown that Hsp12p is an
important stress-response protein in other fungal species
such as S. cerevisiae [35, 37, 38]. Intriguingly, HSP12 gene
has also been shown to be upregulated in azole antifungal
resistance of C. albicans [42]. Cytoplasmic aldehyde dehy-
drogenase (Ald2p) is another stress-response protein that we
found to be upregulated in C. glabrata biofilms. Under stress
conditions, Ald2p, together with Ald3p, removes the accu-
mulated toxic metabolite acetaldehyde by converting it to
Figure 2. Transcriptomic analyses of the selected genes that
acetyl Co-A. Hence, Ald2p is an important enzyme in the
were up- or downregulated at the protein level using Q-RT-PCR.
yeast stress response [43].
Note the transcriptomic expression of TRX1 of Cg5 was
When cells are exposed to stress conditions, particularly
approximately similar between planktonic and biofilm modes
(log10 0.00470.001).
those inducing oxidative stress, molecules such as nucleic
acids, lipids, proteins, and carbohydrates become oxidized,
pathway. At the same time, several proteins in the biofilm and the accumulated oxidized versions need to be removed
that were associated with stress response were upregulated. fairly efficiently. Therefore, protein degradation, in addition
Transcriptomic regulations of identified protein biomarkers to repair, plays a key housekeeping role in eliminating
evaluated by Q-RT-PCR in both C. glabrata reference and oxidized proteins which is mediated by vacuolar proteinases.
wild-type strains corroborated our proteomic findings. The protein Pep4p is vital for vacuolar proteinase activity
Hence, expression patterns of both mRNA and protein and enables fungi to survive under conditions of oxidative
showed similar trends. stress [44]. In the present study, Pep4p was upregulated in
In previous studies, upregulated proteins in C. glabrata C. glabrata biofilms, again consistent with the major role
biofilms that were associated with stress responses (Hsp12p, played by oxidative stress-response mechanisms in the
Trx1p, Ahp1p, Pep4p, and Ald2p) were shown to play an biofilm mode of Candida growth.
important roles in the stress responses of yeast [35 38]. We The findings of this study indicate that glycolysis and
previously found that oxidative stress-response proteins hence energy production is downregulated in the biofilm
were upregulated in C. albicans biofilms and suggested that mode of growth. Several glycolytic enzymes, including
these anti-oxidant biomarkers may contribute to the higher Fba1p, Tdh3p, Gpm1p, and Eno1p, were downregulated in
antifungal resistance of C. albicans biofilms [13]. Therefore, C. glabrata biofilms compared with planktonic cultures.
it is important to establish the functional roles of identified Fba1p catalyzes the hydrolysis of fructose-1,6-bisphosphate
biomarkers in the C. glabrata biofilm proteome and compare into glyceraldehyde-3-phosphate and dihydroxyacetone
them with those in C. albicans. phosphate, Tdh3p catalyzes the phosphorylation of glycer-
One of the upregulated stress-response proteins in the aldehyde-3-phosphate into 1,3-bisphosphoglycerate, Gpm1p
C. glabrata biofilm proteome is Trx1p, a component of the catalyzes the conversion of 3-phosphoglycerate into 2-phos-
oxidative stress defenses of the yeast. The thioredoxin redox phoglycerate, and Eno1p catalyzes the conversion of
system, which includes Trx1p, is able to protect cells from 2-phosphoglycerate to phosphoenolpyruvate. The limited
reactive oxygen species and plays a key role in defending the available genomic and proteomic studies on Candida
cell against stress, particularly oxidative stress, thereby biofilms agree with the notion that metabolic activity
maintaining a reduced environment within the cell [38, 39]. decreases as biofilms mature [45]. The development of a
This molecule is also upregulated in the C. albicans biofilm biofilm community into a three-dimensional multilayered
proteome [13]. Hence, it is tempting to speculate that Trx1p structure leads to the deprivation of nutrients to the bottom
is a part of an oxidative stress-response pathway that, in layers. It can be surmised that cells across the biofilm
turn, plays a role in the biofilm mode of growth of Candida. exhibit differential metabolic activity: while the bottom
Another upregulated oxidative stress-response protein in layers are in a state of quiescence, metabolic activity of
C. glabrata biofilms, Ahp1p, has been found to be upregu- middle and top layers may still be in an active state. Indeed,
& 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.proteomics-journal.com
Proteomics 2010, 10, 1444 1454 1451
Table 3. Differentially expressed proteins identified by MS/MS
Spot Protein Identifier Gene Gene Gene Unique Sequence Biofilm/ Factor
no. name Cg Sac Ca peptide coverage Planktonic
(%)
289 Cytoplasmic aldehyde gi|50287753 CAGL0F07777g ALD2 ALD4 26 50 Up 1.9
dehydrogenases
379 b subunit of the F1 sector of gi|50288781 CAGL0H00506g ATP2 ATP2 20 46 Up 2.2
mitochondrial ATP
synthase
436 Enolase/phosphopyruvate gi|50289857 CAGL0I02486g ENO1 ENO1 22 53 Up 2
hydratase
511 Golgi vesicle protein of gi|50292761 CAGL0L00891g GVP36 orf19.1236 14 55 Up 1.7
unknown function
550 Adenosine kinase gi|50285923 CAGL0C04983g ADO1 ADO1 12 39 Up 1.9
566 Transaldolase gi|50285355 CAGL0B03069g TAL1 TAL1 10 30 Up 1.7
575 Transaldolase gi|50285355 CAGL0B03069g TAL1 TAL1 12 43 Up 1.7
554 Glyceraldehyde-3- gi|50288681 CAGL0G09383g TDH3 TDH3 16 62 Up 1.6
phosphate
dehydrogenase
743 Protein component of the gi|50292587 CAGL0K11748g RPS11B orf19.4149.1 9 51 Up 1.5
small (40S) ribosomal
subunit
760 Alkyl hydroperoxide gi|50294878 CAGL0M11704g AHP1 AHP1 7 45 Up 1.5
reductase
762 Peptidyl-prolyl cis-trans gi|50288871 CAGL0H01529g CPR5 CYP5 3 16 Up 1.6
isomerase
793 Vacuolar aspartyl protease gi|50294061 CAGL0M02211g PEP4 APR1. 4 12 Up 7.2
(proteinase A)
799 Heat shock protein-12 gi|50290911 CAGL0J04202g HSP12 HSP12 11 84 Up 2.3
801 Peptidyl-prolyl cis-trans gi|50292417 CAGL0K09724g FPR1 RBP1 9 67 Up 1.5
isomerase
803 Alcohol dehydrogenase gi|68476713 CAGL0I07843g ADH3 ADH2 6 27 Up 2.4
isoenzyme III
804 Cytoplasmic thioredoxin gi|50291653 CAGL0K00803g TRX1 TRX1 8 88 Up 1.5
isoenzyme
805 Cytoplasmic thioredoxin gi|50291653 CAGL0K00803g TRX1 TRX1 4 56 Up 2.2
isoenzyme
478 Enolase/phosphopyruvate gi|50289857 CAGL0I02486g ENO1 ENO1 20 49 Down 1.6
hydratase
509 Fructose-1,6-bisphosphate gi|50292893 CAGL0L02497g FBA1 FBA1 19 63 Down 2.2
aldolase
513 Fructose-1,6-bisphosphate gi|50292893 CAGL0L02497g FBA1 FBA1 17 54 Down 2
aldolase
526 Alcohol dehydrogenase gi|50290317 CAGL0I07843g ADH1 ADH2 14 59 Down 1.5
529 Alcohol dehydrogenase gi|50290317 CAGL0I07843g ADH1 ADH2 14 59 Down 1.7
598 Glyceraldehyde-3- gi|50290597 CAGL0J00451g TDH3 TDH3 13 53 Down 1.8
phosphate
dehydrogenase
713 Tetrameric gi|50287073 CAGL0E06358g GPM1 GPM1 22 64 Down 1.7
phosphoglycerate
mutase
Proteins were named according to the S. cerevisiae genomic database (www.yeastgenome.org), or C. glabrata genomic database
(www.genolevures.org). To avoid confusion in nomenclature, gene names according to NCBInr_200705 FASTA (Identifier), C. glabrata
genomic database (Gene Cg), S. cerevisiae genomic database (Gene Sac), and C. albicans genomic database (Gene Ca)
(www.candidagenome.org) databases are also provided.
Candida biofilms have been found to contain heterogeneous glyoxylate pathway, which is important in producing cell-
cells with variable metabolic activity [46, 47]. Reduced wall carbohydrates and exopolymeric substances [48].
metabolic activity may reflect energy and nutrient conser- The protein Atp2p, which was upregulated in C. glabrata
vation to avoid stress inducers, as well as activation of the biofilms in this study, has been shown to be upregulated
& 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.proteomics-journal.com
1452 C. J. Seneviratne et al. Proteomics 2010, 10, 1444 1454
during pseudohyphal growth of C. albicans. Furthermore, in bloodstream, invasive, and colonizing strains and differ-
ences between isolates from three urban teaching hospitals
via links with Efg1p, Cph1p, and Tec1p, Atp2p may be
in New York City (Candida Susceptibility Trends Study, 1998
connected to downstream signaling events of the MAP
to 1999). Antimicrob. Agents Chemother. 2002, 46,
kinase pathway, which is important for biofilm formation,
3268 3272.
stress response, and virulence [49, 50]. However, it is note-
[3] Trick, W. E., Fridkin, S. K., Edwards, J. R., Hajjeh, R. A.,
worthy that C. glabrata does not produce true hyphal forms.
Gaynes, R. P., Secular trend of hospital-acquired candide-
In our previous study on C. albicans biofilm proteome,
mia among intensive care unit patients in the United States
which used the same methodology as in this study, a
during 1989 1999. Clin. Infect. Dis. 2002, 35, 627 630.
number of upregulated biomarkers were associated with
[4] Li, L., Redding, S., Dongari-Bagtzoglou, A., Candida glab-
hyphal proteins, including Sti1p, Ald5p, Cit1p, Mdh1p,
rata: an emerging oral opportunistic pathogen. J. Dent. Res.
Mcr1p, Egd1p, Ynk1p, and Cyp1p [13]. These protein
2007, 86, 204 215.
biomarkers were not present in the C. glabrata proteome,
[5] Costerton, J. W., Lewandowski, Z., Caldwell, D. E., Korber,
suggesting that species-specific molecular mechanisms may
D. R., Lappin-Scott,H. M., Microbial biofilms. Annu. Rev.
operate in Candida biofilms to preserve morphological and
Microbiol. 1995, 49, 711 745.
architectural variability of different species. These findings
[6] Potera, C., Forging a link between biofilms and disease.
validate our methodology to explore the proteomics of
Science 1999, 283, 1837 1839.
Candida biofilms.
[7] Jain, N., Kohli, R., Cook, E., Gialanella, P. et al., Biofilm
Many studies suggest that Candida in the biofilm mode
formation by and antifungal susceptibility of Candida isolates
possesses higher antifungal resistance than in the plank-
from urine. Appl. Environ. Microbiol. 2007, 73, 1697 1703.
tonic mode. However, the exact molecular mechanism
behind this phenomenon has yet to be described. In the [8] Lewis, R. E., Kontoyiannis, D. P., Darouiche, R. O., Raad, I. I.,
Prince, R. A., Antifungal activity of amphotericin B, fluco-
present study, we have shown that changes in expression
nazole, and voriconazole in an in vitro model of Candida
profiles of various stress-response proteins are associated
catheter-related bloodstream infection. Antimicrob. Agents
with the biofilm mode of growth. Cannon et al. suggested
Chemother. 2002, 46, 3499 3505.
that drug resistance of Candida is in fact a mechanism to
[9] Thein, Z. M., Samaranayake, Y. H., Samaranayake, L. P., In
cope with stress [51]. Therefore, it is conceivable that the
vitro biofilm formation of Candida albicans and non-albi-
biofilm mode of growth, stress response, and antifungal
cans Candida species under dynamic and anaerobic condi-
resistance of Candida may all have a generic link that ulti-
tions. Arch. Oral Biol. 2007, 52, 761 767.
mately helps the fungi to survive in a hostile environment.
[10] Tumbarello, M., Posteraro, B., Trecarichi, E. M., Fiori, B. et al.,
From the evidence so far furnished from proteomic studies
Biofilm production by Candida species and inadequate
of C. albicans and C. glabrata biofilms, we hypothesize that
antifungal therapy as predictors of mortality for patients with
antifungal resistance is not a primary trait of the biofilm, but
candidemia. J. Clin. Microbiol. 2007, 45, 1843 1850.
a   bonus trait  that Candida acquires through its stress
[11] Schmidt, P., Walker, J., Selway, L., Stead, D. et al., Proteo-
response, which represents its integral arsenal to combat
mic analysis of the pH response in the fungal pathogen
and survive in a wide range of habitats. Further experiments
Candida glabrata. Proteomics 2008, 8, 534 544.
are needed to explore the potential of this new insight to
[12] Stead, D., Findon, H., Yin, Z., Walker, J. et al., Proteomic
devise more efficient antifungal strategies that will bring
changes associated with inactivation of the Candida glab-
clinical and other translational benefits.
rata ACE2 virulence-moderating gene. Proteomics 2005, 5,
1838 1848.
We thank Alan Wong, Lawrence Luk, and Priscilla Leung for
[13] Seneviratne, C. J., Wang, Y., Jin, L., Abiko, Y., Samar-
technical assistance and Dr. Trevor Lane for editorial assistance.
anayake, L. P., Candida albicans biofilm formation is asso-
This work was supported by the Hong Kong Research Grants
ciated with increased anti-oxidative capacities. Proteomics
Council, RGC no. HKU 7624/06M.
2008, 8, 2936 2947.
[14] Jin, Y., Samaranayake, L. P., Samaranayake, Y., Yip, H. K.,
The authors have declared no conflict of interest.
Biofilm formation of Candida albicans is variably affected
by saliva and dietary sugars. Arch. Oral Biol. 2004, 49,
789 798.
5 References
[15] Jin, Y. Y. H. K., Samaranayake, Y. H., Yau, J. Y., Samar-
anayake, L. P., Biofilm-forming ability of Candida albicans is
[1] McMullan, R., McClurg, R., Xu, J., Moore, J. E. et al., Trends unlikely to contribute to high levels of oral yeast carriage in
in the epidemiology of Candida bloodstream infections in cases of human immunodeficiency virus infection. J. Clin.
Northern Ireland between January 1984 and December Microbiol. 2003, 41, 2961 2967.
2000. J. Infect. 2002, 45, 25 28.
[16] Seneviratne, C. J., Jin, L. J., Samaranayake, Y. H., Samar-
[2] Safdar, A., Chaturvedi, V., Koll, B. S., Larone, D. H. et al., anayake, L. P., Cell density and cell aging as factors modu-
Prospective, multicenter surveillance study of Candida lating antifungal resistance of Candida albicans biofilms.
glabrata: fluconazole and itraconazole susceptibility profiles Antimicrob. Agents Chemother. 2008, 52, 3259 3266.
& 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.proteomics-journal.com
Proteomics 2010, 10, 1444 1454 1453
[17] Anil, S., Ellepola, A. N., Samaranayake, L. P., The impact of [32] Martinez-Gomariz, M., Perumal, P., Mekala, S., Nombela, C.
polyene, azole, and DNA analogue antimycotics on the cell et al., Proteomic analysis of cytoplasmic and surface
surface hydrophobicity of Candida albicans and Candida proteins from yeast cells, hyphae, and biofilms of Candida
tropicalis in HIV infection. Mycopathologia 2002, 153, albicans. Proteomics 2009, 9, 2230 2252.
179 185.
[33] Mukherjee, P. K., Mohamed, S., Chandra, J., Kuhn, D. et al.,
[18] National Committee for Clinical Laboratory Standards. Alcohol dehydrogenase restricts the ability of the pathogen
Reference Method for Broth Dilution Antifungal Suscept- Candida albicans to form a biofilm on catheter surfaces
ibility Testing of Yeasts, 2nd Edn., Approved Standard M27- through an ethanol-based mechanism. Infect. Immun. 2006,
A2, NCCLS, Wayne, PA, USA 2002. 74, 3804 3816.
[19] Ramage, G., Vande Walle, K., Wickes, B. L., Lopez-Ribot, J. [34] Thomas, D. P., Bachmann, S. P., Lopez-Ribot, J. L., Proteo-
L., Standardized method for in vitro antifungal susceptibility
mics for the analysis of the Candida albicans biofilm life-
testing of Candida albicans biofilms. Antimicrob. Agents
style. Proteomics 2006, 6, 5795 5804.
Chemother. 2001, 45, 2475 2479.
[35] Chauhan, N., Inglis, D., Roman, E., Pla, J. et al., Candida
[20] Vediyappan, G., Chaffin, W. L., Non-glucan attached
albicans response regulator gene SSK1 regulates a subset
proteins of Candida albicans biofilm formed on various
of genes whose functions are associated with cell wall
surfaces. Mycopathologia 2006, 161, 3 10.
biosynthesis and adaptation to oxidative stress. Eukaryot.
Cell 2003, 2, 1018 1024.
[21] Wang, Y., Lam, K. S., Lam, J. B., Lam, M. C. et al., Over-
expression of angiopoietin-like protein 4 alters mitochon- [36] Kim, H. J., Rakwal, R., Shibato, J., Iwahashi, H. et al., Effect
dria activities and modulates methionine metabolic cycle in
of textile wastewaters on Saccharomyces cerevisiae using
the liver tissues of db/db diabetic mice. Mol. Endocrinol.
DNA microarray as a tool for genome-wide transcriptomics
2007, 21, 972 986.
analysis. Water Res. 2006, 40, 1773 1782.
[22] Yan, J. X., Wait, R., Berkelman, T., Harry, R. A. et al., A
[37] Kim, I., Yun, H., Jin, I., Comparative proteomic analyses of
modified silver staining protocol for visualization of
the yeast Saccharomyces cerevisiae KNU5377 strain against
proteins compatible with matrix-assisted laser desorption/
menadione-induced oxidative stress. J. Microbiol. Biotech-
ionization and electrospray ionization-mass spectrometry.
nol. 2007, 17, 207 217.
Electrophoresis 2000, 21, 3666 3672.
[38] Chae, H. Z., Chung, S. J., Rhee, S. G., Thioredoxin-depen-
[23] Wang, Y., Xu, L. Y., Lam, K. S. L., Lu, G. et al., Proteomic
dent peroxide reductase from yeast. J. Biol. Chem. 1994,
characterization of human serum proteins associated with
269, 27670 27678.
the fat-derived hormone adiponectin. Proteomics 2006, 6,
[39] Trotter, E. W., Grant, C. M., Thioredoxins are required for
3862 3870.
protection against a reductive stress in the yeast Sacchar-
[24] Samaranayake, Y. H., Dassanayake, R. S., Jayatilake, J. A.
omyces cerevisiae. Mol. Microbiol. 2002, 46, 869 878.
M. S., Cheung, B. P. K. et al., Phospholipase B enzyme
[40] Jean, D., Briolat, V., Reysset, G., Oxidative stress response in
expression is not associated with other virulence attributes
Clostridium perfringens. Microbiology 2004, 150, 1649 1659.
in Candida albicans isolates from patients with human
[41] Perez-Torrado, R., Bruno-Barcena, J. M., Matallana, E.,
immunodeficiency virus infection. J. Med. Microbiol. 2005,
Monitoring stress-related genes during the process of
54, 818 825.
biomass propagation of Saccharomyces cerevisiae strains
[25] Pfaffl, M. W., A new mathematical model for relative
used for wine making. Appl. Environ. Microbiol. 2005, 71,
quantification in real-time RT-PCR. Nucleic Acids Res. 2001,
6831 6837.
29, e45.
[42] Karababa, M., Coste, A. T., Rognon, B., Bille, J., Sanglard,
[26] Seneviratne, C. J., Silva, W. J., Jin, L. J., Samaranayake, Y. H.,
D., Comparison of gene expression profiles of Candida
Samaranayake,L. P., Architectural analysis, viability assess-
albicans azole-resistant clinical isolates and laboratory
ment and growth kinetics of Candida albicans and Candida
strains exposed to drugs inducing multidrug transporters.
glabrata biofilm. Arch. Oral Biol. 2009, 54, 1052 1060.
Antimicrob. Agents Chemother. 2004, 48, 3064 3079.
[27] Al-Dhaheri, R. S., Douglas, L. J., Absence of amphotericin B-
[43] Navarro-Avino, J. P., Prasad, R., Miralles, V. J., Benito,
tolerant persister cells in biofilms of some Candida species.
R. M., Serrano, R., A proposal for nomenclature of aldehyde
Antimicrob. Agents Chemother. 2008, 52, 1884 1887.
dehydrogenases in Saccharomyces cerevisiae and char-
[28] Ramage, G., Saville, S. P., Thomas, D. P., Lopez-Ribot, J. L.,
acterization of the stress-inducible ALD2 and ALD3 genes.
Candida biofilms: an update. Eukaryot. Cell 2005, 4,
Yeast 1999, 15, 829 842.
633 638.
[44] Marques, M., Mojzita, D., Amorim, M. A., Almeida, T. et al.,
[29] Lopez-Ribot, J. L., Candida albicans biofilms: more than
The Pep4p vacuolar proteinase contributes to the turnover
filamentation. Curr. Biol. 2005, 15, R453 R455.
of oxidized proteins but PEP4 overexpression is not suffi-
[30] Nobile, C. J., Mitchell, A. P., Genetics and genomics of cient to increase chronological lifespan in Saccharomyces
Candida albicans biofilm formation. Cell. Microbiol. 2006, 8, cerevisiae. Microbiology 2006, 152, 3595 3605.
1382 1391.
[45] Yeater, K. M., Chandra, J., Cheng, G., Mukherjee, P. K. et al.,
[31] Chandra, J., Mukherjee, P. K., Leidich, S. D., Faddoul, F. F. Temporal analysis of Candida albicans gene expression
et al., Antifungal resistance of candidal biofilms formed on during biofilm development. Microbiology 2007, 153,
denture acrylic in vitro. J. Dent. Res. 2001, 80, 903 908. 2373 2385.
& 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.proteomics-journal.com
1454 C. J. Seneviratne et al. Proteomics 2010, 10, 1444 1454
[46] LaFleur, M. D., Kumamoto, C. A., Lewis, K., Candida albi- [49] Monge, R. A., Roman,E., Nombela, C., Pla, J., The MAP
cans biofilms produce antifungal-tolerant persister cells. kinase signal transduction network in Candida albicans.
Antimicrob. Agents Chemother. 2006, 50, 3839 3846. Microbiology 2006, 152, 905 912.
[47] Uppuluri, P., Sarmah, B., Chaffin, W. L., Candida albicans [50] Singh, V., Sinha, I., Sadhale, P. P., Global analysis of altered
SNO1 and SNZ1 expressed in stationary-phase planktonic gene expression during morphogenesis of Candida albi-
yeast cells and base of biofilm. Microbiology 2006, 152, cans in vitro. Biochem. Biophys. Res. Commun. 2005, 334,
2031 2038. 1149 1158.
[48] Brown, A. J., Odds, F. C., Gow, N. A., Infection-related gene [51] Cannon, R. D., Lamping, E., Holmes, A. R., Niimi, K. et al.,
expression in Candida albicans. Curr. Opin. Microbiol. 2007, Candida albicans drug resistance another way to cope with
10, 307 313. stress. Microbiology 2007, 153, 3211 3217.
& 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.proteomics-journal.com