Proteomic analysis of cytosolic proteins associated with petite mutations in Candida glabrata C .

The incidence of superficial or deep-seated infections due to Candida glabrata has increased markedly, probably because of the low intrinsic susceptibility of this microorganism to azole antifungals and its relatively high propensity to acquire azole resistance. To determine changes in the C. glabrata proteome associated with petite mutations, cytosolic extracts from an azoleresistant petite mutant of C. glabrata induced by exposure to ethidium bromide, and from its azole-susceptible parent isolate were compared by two-dimensional polyacrylamide gel electrophoresis. Proteins of interest were identified by peptide mass fingerprinting or sequence tagging using a matrix-assisted laser desorption/ionization tandem time-of-flight mass spectrometer. Tryptic peptides from a total of 160 Coomassie-positive spots were analyzed for each strain. Sixty-five different proteins were identified in the cytosolic extracts of the parent strain and 58 in the petite mutant. Among the proteins identified, 10 were higher in the mutant strain, whereas 23 were lower compared to the parent strain. The results revealed a significant decrease in the enzymes associated with the metabolic rate of mutant cells such as aconitase, transaldolase, and pyruvate kinase, and changes in the levels of specific heat shock proteins. Moreover, transketolase, aconitase and catalase activity measurements decreased significantly in the ethidium bromide-induced petite mutant. These data may be useful for designing experiments to obtain a better understanding of the nuclear response to impairment of mitochondrial function associated with this mutation in C. glabrata.


Introduction
The incidence of life-threatening fungal infections that are mainly caused by the Candida species has increased dramatically in the past decades along with the development of antibiotic treatments, the widespread use of immunosuppressive therapy, and the emergence of the AIDS epidemic (1).Among the causative agents of these infections, Candida albicans remains by far the most frequent, but infections due to other Candida species are being reported increasingly (2,3).For instance, C. glabrata has recently emerged as a significant pathogen in various hospital settings, where it is responsible for an increasing number of systemic infections and candiduria (4,5).In a recent study conducted in the United States, C. glabrata ranked second among the causative agents of fungemia, accounting for 21% of all Candida bloodstream isolates (6,7).The rise in the number of C. glabrata systemic infections is due to the poor intrinsic susceptibility of this yeast to azole antifungals and to its propensity to acquire azole resistance (5,(8)(9)(10)(11)(12).
The mechanisms of resistance to azole antifungals have been studied mainly in C. albicans and can be categorized as i) changes in the cell wall or plasma membrane, which lead to impaired azole uptake; ii) alterations in the affinity of azoles for their target Erg11p (lanosterol 14α-demethylase) or increase in the cellular content of Erg11p due to mutations in or overexpression of the ERG11 gene, respectively, and iii) increased efflux of the azole drugs mediated by membrane transport proteins belonging to the ATP-binding cassette (ABC) transporter family (CDR1 and CDR2) or to the major facilitator superfamily (MDR1 and FLU1).For instance, the CDR1, CDR2 and MDR1 genes have been shown to be overexpressed in many azole-resistant isolates, and deletion of these genes resulted in hypersensitivity to azoles (13).However, different mechanisms including overexpression of genes encoding the efflux pumps and overexpression or point mutations in the ERG11 gene frequently combine, resulting in a stepwise development of azole resistance over time (14).In addition, compensatory pathways that involve alterations of specific steps in ergosterol biosynthesis have been documented as resistance mechanisms to both azoles and polyene antifungals (15).
More recently, increased levels of expression of the ABC transporter genes C. glabrata CDR1 (CgCDR1) and CDR2 (CgCDR2) were detected in azole-resistant isolates of C. glabrata (16)(17)(18).Furthermore, the azole resistance of C. glabrata petite mutants obtained by exposure to fluconazole or induced by ethidium bromide (ETB) was shown to be associated with the up-regulation of the nuclear genes CgCDR1 and CgCDR2 (19,20).However, due to the numerous cross-talks between the nucleus and mitochondria, petite mutations may also lead to the deregulation of the expression of other nuclear genes (21).For instance, an increased cellular content of free ergosterol due to a defect in sterol esterification has been reported in petite mutants (19).Similarly, changes in the biochemical composition of the cell wall associated with a lower cell surface hydrophobicity were also observed in mutant cells, as well as an increased expression of the CgEPA1 gene encoding a lectin involved in adherence to epithelial cells (22).
In the present study, changes in the C. glabrata proteome associated with petite mutations and azole resistance were examined by investigation of an azole-susceptible wild-type isolate and an ETB-induced azole-resistant petite mutant.

Yeast strains and culture conditions
The study was carried out using a C. glabrata clinical isolate designated 90.1085, which was obtained at the Laboratory of Parasitology and Mycology of Angers University Hospital, Angers, France, from a urine sample collected in 1990, and a derived petite mutant induced by exposure to the intercalating agent ETB (Sigma-Aldrich, USA).The mutant presented cross-resistance to azoles due to overexpression of the CDR1 and CDR2 genes.The parent isolate and its ETB-induced petite mutant were maintained by biweekly passages in yeast extract-peptone-glucose (YEPD) agar containing 5 g/L yeast extract, 10 g/L peptone, 20 g/L glucose, 0.5 g/L chloramphenicol, and 20 g/L agar.Mutant cells were subcultured on yeast extract-peptone agar containing 2% glycerol as the sole carbon source to ascertain their respiratory deficiency.Both isolates were preserved in 20% glycerol at -80°C.

Cytosolic protein extraction
For the isolation of cytosolic proteins, blastoconidia were washed in deionized distilled water and resuspended in 10 mM Tris-HCl, pH 7.4, containing 300 U RNase and a mix of protease inhibitors (1 mM phenylmethylsulfonyl fluoride, 1 mM ethylenediamine tetraacetic acid and 1 µM pepstatin).Cells were disrupted using glass beads in a Braun homogenizer with cooling CO 2 and the suspensions obtained were centrifuged at 12,000 g for 30 min at 4°C.The pellets were discarded, and the supernatants were centrifuged for 1 h at 75,000 g.The resulting supernatants, which correspond to the cytosolic extracts, were desalted by dialysis and protein content was measured using the Bradford (Bio-Rad Laboratories, USA) and the BCA protein assay kit (Thermo Scientific, USA).
For the determination of the total number of spots, analytical gels were silver-stained as described by Bjellqvist et al. (24), fixed first in methanol-acetic acid-distilled water (4:1:5) for 1 h and then in ethanol-acetic acid-distilled water (0.5:0.5:9) overnight.The gels were rinsed with 7.5% acetic acid and incubated in 1% glutaraldehyde containing 0.5 M sodium acetate for 30 min.The gels were then extensively washed with water and stained with an ammoniacal silver nitrate solution for 30 min.Gels were washed and color was developed in 0.01% citric acid containing 0.1% formaldehyde.Staining was stopped with 5% Tris in 2% acetic acid.
Preparative gels for further analysis of proteins by mass spectrometry were stained with colloidal Coomassie blue R-250 (Bio-Rad).The gels were destained using several changes of distilled water over a 2-h period.Finally, spots were excised and in gel-digested for analysis by matrix-assisted laser desorption/ionization tandem time-of-flight mass spectrometry (MALDI-TOF/TOF MS).

Image analysis
Silver-and Coomassie-stained gels were captured with an Image Scanner (Amersham Biosciences).Different  images were processed for detection, volumetric quantification, matching, and editing of molecular masses and pI of spots, using ImageMaster 2-D Platinum software (Amersham Biosciences).Proteins were considered to be differentially increased or decreased if the protein spot in the ETB-induced mutant showed statistically significant differences of 2-fold or more in their mean spot volume on at least three of four gels (P < 0.05, t-test) compared to its parent isolate.

In-gel tryptic digestion
Digestion was performed by the method of Pitarch et al. (25).Protein spots were excised from Coomassie-stained 2-D gels and transferred to 0.5-mL tubes.Gel fragments were destained with acetonitrile (ACN), washed twice with 50% ACN in 25 mM ammonium bicarbonate (AmBic), and vacuum-dried.The proteins were then reduced with 10 mM dithioerythritol in 25 mM AmBic for 30 min at 56°C and subsequently alkylated with 55 mM iodoacetamide in 25 mM AmBic for 20 min in the dark.Next, the gel fragments were washed with 25 mM AmBic and ACN and dried under vacuum.All gel fragments were incubated with 12.5 ng/µL sequencing grade trypsin (Promega, USA) in 25 mM AmBic overnight at 37°C.Peptides were then extracted from the gel fragments with 50% ACN, 1% trifluoroacetic acid in 25 mM AmBic, and finally with 100% ACN.The extracts were pooled, and concentrated by evaporation of the solvent with a SpeedVac apparatus (Thermo Fisher Scientific, USA).

MALDI-TOF/TOF MS
Peptides were applied onto a MALDI plate after co-crystallization with and α-cyano 4-hydroxycinnamic acid (CHCH) matrix (Sigma-Aldrich).One microliter of each sample with 0.4 µL 3 mg/mL CHCH matrix in 50% ACN and 0.01% trifluoroacetic acid was spotted onto a MALDI plate.MS/MS sequencing analyses were carried out using a MALDI-TOF/TOF MS 4700 Proteomics Analyzer (Applied Biosystems, USA).

Database search
Peak lists from all MS/MS spectra were submitted to a database search using an in-house copy of MASCOT, version 3.1 (Matrix Science Inc., USA).The following criteria were used for all database searches: a minimum signal-to-noise ratio threshold of 5-10; mass values in the 0-60 Da range and masses within 20 Da of the precursor ion mass were excluded.A maximum of 60 peaks per spectrum were included as product ions.The mass tolerance was ±75 ppm for MS data, ±200 ppm for MS/MS precursor ions, and ±250 ppm for MS/MS product ions.The sample was searched against the NCBInr database (October 4, 2007).The ExPASy (Expert Protein Analysis System) proteomics server of the Swiss Institute of Bioinformatics (SIB) was used for the analysis of protein sequences.

Measurements of transketolase and aconitase activity
To estimate the transketolase activity, samples were added to a cuvette containing buffer (50 mM Tris-HCl, pH 7.6), 2 mM ribose 5-phosphate, 1 mM xylulose 5 phosphate, 5 mM MgCl 2 , 0.2 U/mL triosephosphate isomerase, 0.2 mM NADH, and 0.1 mM thiamine pyrophosphate.Reactions were initiated by the addition of cytosolic extracts at 37°C.The aconitase activity was determined by addition of the samples to a cuvette containing buffer (50 mM Tris-HCl, pH 7.4), 0.6 mM MnCl 2 , and 0.5 mM ferrous ammonium sulfate.Reactions were initiated by the addition of cytosolic extracts at 37°C.Transketolase and aconitase activities were measured by spectrophotometry (340 nm) (Thermo Spectronic Genesys 10 uv), and data are reported as ng product•min -1 •mg total protein -1 .Total protein content of cytosolic extracts was determined by the method of Bradford.Each experiment was repeated three times.

Measurements of catalase activity
Catalase activity was measured by the method of Aebi (26), which is based on the principle that the absorbance will decrease due to dismutation of H 2 O 2 at 240 nm (UV-visible spectrophotometry).The amount of H 2 O 2 converted into H 2 O and oxygen in 1 min under standard conditions is accepted as enzyme reaction velocity.Data are reported as µmol H 2 O 2 metabolized•min -1 •mg total protein -1 .

Statistical analysis
Statistical analysis for the enzyme activity data was performed using the Student t-test, with the level of significance set at P < 0.05.For identification of protein, the Mascot protein score reports a match as significant if it has a match with less than 5% chance of being a random hit.Protein scores are derived from ion scores as a non-probabilistic basis for ranking protein hits.Protein score is -10*Log(P), where P is the probability that the observed match is a random event.Individual protein scores >50 indicate identity or extensive homology (P < 0.05).

Results
In order to identify differences in protein expression between the parent isolate and its derived petite mutant, differences in protein profiles between cytosolic protein extracts from cultures of the two isolates were examined.Equal amounts of each protein extract (parent isolate and ETB-induced petite mutant) were submitted to 2-D electrophoresis.Multiple gels from three independent experiments were run for each sample to ascertain reproducibility.One representative gel of each extract was used as a reference 2-D map (Figure 1).Analysis of the virtual image generated using the ImageMaster Platinum software and based on the overlay of the reference 2-D maps permitted the observation of significant differences in the protein pattern between the two isolates (Figure 2).In the silver-stained 2-D gels, 664 and 626 spots were visualized for the parent isolate and its ETB-induced petite mutant, respectively (data not shown).Although lower numbers of spots were observed on the Coomassie-stained gels, 364 in the parent isolate and 326 in the ETB-induced petite mutant, the main differences remained the same as expected (Figure 2).For protein identification, the spots were digested ingel and analyzed by MALDI-TOF/TOF MS.Bioinformatic analysis of MALDI-TOF spectra permitted the identification of a total of 65 different proteins in the parent isolate and 58 in the ETB-induced petite mutant corresponding to the 140 and 121 spots, respectively, numbered in Figure 1.A list of identified proteins is presented in Table 1, together with their accession numbers, peptides matched, sequence  1.  coverage, peptide score, and assignment to the strains.
Although most proteins (N = 120) were detected in 2-D gels of both extracts from strains, quantitative changes were shown for some of them.Quadruplicate gel images for each biological sample (at least three protein extracts for each cell type) were obtained and quantitatively analyzed using the ImageMaster 2-D Platinum software to select statistically significant changes (P < 0.05, t-test).We only considered differences of at least 2-fold in protein expression levels.At least 12 matched spots were higher in the ETB-induced petite mutant (Table 2) and 24 were diminished (Table 2), corresponding to 9 and 11 distinct proteins, respectively.Additionally, some qualitative differences in protein patterns between the two strains were also observed, and the corresponding proteins are listed in Table 3.For example, five proteins involved in carbohydrate degradation, three hexokinases and two glyceraldehyde-3-phosphate dehydrogenases 1 (spots 37-39, 87, 88) were not detected in the cytosolic extract from the ETB-induced petite mutant.Transketolase (spot 85), transaldolase (spot 90), alcohol dehydrogenase (spot 63), inorganic pyrophosphatase (spot 86), and acetyl-coenzyme A synthetase 1 (spots [13][14][15], also involved in the carbohydrate pathway, were identified only in the parent isolate.The heat shock proteins, Hsp104, Hsp70 and Hsp60 (spots 3-5, 17, and 26, respectively), were exclusive to the parental strain.In contrast, a single hexokinase (spot 1M), which was not detected in the parent isolate, was identified in the ETB-induced petite mutant.
In order to confirm whether some enzymes were essentially decreased in ETB-induced petite mutant, we measured total transketolase, aconitase and catalase activity in both extracts.As expected, we found a significant decrease in enzymatic activity of these three enzymes in the cytosolic extract from the ETB-induced petite mutant (P < 0.0001, t-test; Table 4).

Discussion
Mechanisms of antifungal resistance in C. glabrata are being elucidated at the molecular level.Azole resistance, which commonly occurs in patients receiving fluconazole for prophylaxis or therapy, is usually associated with increased mRNA levels of the ATP binding cassette transporters, CgCDR1, CgCDR2 and PDH1 (16).However, the number of molecular events analyzed in previous studies was usually limited to the overexpression of genes encoding lanosterol demethylase or some efflux pumps and to mutations in the ERG11 gene.Furthermore, putative changes in other enzymes of the ergosterol pathway were not investigated in most of these studies.27) demonstrated increased levels of the products of the genes CgCDR1 and ERG11 using proteome analysis, confirming the up-regulation of these genes.Previous studies from our group and others have revealed that petite mutations, which are caused by the partial or total loss of mitochondrial DNA, are associated with a cross-resistance to almost all the azole drugs due to an increased expression of CgCDR1 and CgCDR2 (17,20).Although CgERG11 expression was not affected in petite mutants, mutant cells showed a marked increase in free ergosterol content (20).The present study was designed to provide additional data using a new methodological approach regarding changes in the expression of nuclear genes induced by the impairment of mitochondrial function.Here, the proteomic analysis of cytosolic proteins of an ETB-induced petite mutant is described and compared to that of the parent strain.
The experiments revealed a huge variety of proteins in the azole-susceptible parent isolate and its derived petite mutant.Changes were observed in the protein pattern in association with the petite mutation, including the low expression of some proteins involved in carbohydrate metabolism.Some of these proteins were identified as aconitase, phosphoglycerate mutase, glyceraldehyde-3-phosphate dehydrogenase 2, and pyruvate kinase 1.Moreover, three isoforms of hexokinase and two of glyceraldehyde-3-phosphate dehydrogenase were not detected in the ETB-induced mutant.Similarly, a transketolase, a transaldolase and a 6-phosphogluconate dehydrogenase, which play a key role in the regulation of the pentose-phosphate pathway, were down-regulated or not detected in mutant cells.These results suggest a significant decrease in enzymes associated with the metabolic rate of mutant cells aside from the mitochondrial loss, and are therefore in agreement with the limited growth of petite mutants on YEPD agar plates and the increase in the generation time observed in a previous study (22).Although the amount of numerous glycolytic enzymes was lower in mutant cells, an up-regulated isoform of hexokinase was identified, suggesting a possible mechanism of compensation for energy production from glucose.
Sulfur amino acid biosynthesis is peripherally linked to ergosterol biosynthesis.Homocysteine is required for the biosynthesis of S-adenosylmethionine, which is necessary for the ability of sterol C-24 methyltransferase to convert zymosterol to fecosterol (28).The up-regulation of cystathionine beta-synthase and S-adenosylmethionine synthetase detected in the ETB-mutant may therefore impact the ergosterol biosynthesis pathway in azole resistance, since these enzymes are involved in homocysteine metabolism and S-adenosylmethionine synthesis, respectively.
Certain heat shock proteins including Hsp60, Hsp70 and Hsp104, were also shown to be down-regulated or not detected in the ETB-induced mutant.Hsp104, a cytosolic chaperone system member of the AAA+ protein family, is directly involved in the refolding of heat-denatured proteins (29), but full activity of this protein requires cooperation with the Hsp70 chaperone system (30).In 2006, Matsumoto et al. (31) demonstrated that mutant yeasts lacking both the cytosolic Hsp70 genes SSA1 and SSA2 presented numerous changes in gene expression with the up-regulation of genes involved in the stress response, protein synthesis and ubiquitin-proteasome protein degradation.A proteome analysis of the yeast strain ssa1/2 was also carried out and revealed up-regulation of elongation factor eIF-5A proteins and stress-inducible proteins (31).These stress proteins have been shown to play a direct role in the repair of macromolecular complexes involved in the RNA metabolism of yeast cells conditioned to environmental stresses (32).
In Saccharomyces cerevisiae, heme, a molecule that indicates oxygen level, binds to and activates Hap1 (33).Hsp70 also promotes regulation of Hap1, required for the activation of aerobic genes (expressed only in the presence of oxygen), including those required for respiration and for the control of oxidative damage, and also indirectly represses hypoxic genes (low-oxygen conditions) by activating ROX1 (34).Interestingly, among hypoxic genes that are down-regulated in the Hap1-deficient mutant are genes encoding flavohemoglobin and catalase, which are both down-regulated in the petite mutant together with low levels of Hsp70 (35).Down-regulation of cytosolic Hsp70 in mutant cells is not surprising since they were shown by transmission electron microscopy to be devoid of mitochondria.Indeed, this protein is important for the maintenance of competence for the importation of some mitochondrial precursor proteins (36).The absence of Hsp104 and cytosolic Hsp70 might also lead to a decrease in thermotolerance in petite mutants, which could contribute to their reduced growth rate.Conversely, another heat shock protein, Hsp12, was shown to be up-regulated in mutant cells.Interestingly, in C. albicans, the transcriptional factor Tac1p was shown to be responsible not only for the overexpression of the genes CDR1 and CDR2, but also for the up-regulation of other genes including HSP12 (37).The gene HSP12 encodes a protein responsible for a shift from the carbohydrate to the lipid metabolism.The present results suggest that HSP12 overexpression in C. glabrata petite mutants, their resistance to azoles and possibly the down-regulation of their carbohydrate metabolism could be regulated by a common transcriptional factor similar to Tac1p.Recently, a zinc-finger protein homologous to Tac1p, called Pdr1p, has been identified in C. glabrata, acting as a transcriptional regulator of a pleiotropic drug resistance network (38,39).Disruption of the gene PDR1 largely reversed azole resistance in azole-resistant isolates with high expression levels of CDR1 and microarray analysis demonstrated an overexpression of several genes including HSP12 in a laboratory mutant resistant to fluconazole (40).
Although quite descriptive by having as its main objective the determination of the protein changes induced by petite mutations, the present study provides new insights into the current understanding of the relationships between the impairment of mitochondrial function and azole resistance of petite mutants.The results suggest that regulation of the expression of genes encoding the efflux pumps by certain by-products of mitochondrial metabolism could occur through the regulation of the transcription factors of these genes.Further experiments will be performed to confirm and extend our results by quantifying the expression level of genes encoding up-regulated proteins by PCR, particularly the HSP12 gene, and to establish the role of the transcription factor Pdr1p in the protein changes observed in mutant cells.

Figure 1 .
Figure 1.2-D gels of cytoplasmic protein extracts of the Candida glabrata parent isolate and its ethidium bromide (ETB)-induced petite mutant separated on immobilized pH gradient (IPG) strips covering the pH range of 3 to 11.The gels were stained with colloidal Coomassie blue.A, Parent isolate; B, ETB-induced petite mutant.The protein spots that were identified are numbered and listed in Table1.

Figure 2 .
Figure 2. Two-dye overlay (parent isolate/ethidium bromide (ETB)-induced petite mutant).A, Virtual image generated using the Im-ageMaster Platinum software and based on the overlay from 2-D maps of Candida glabrata isolates where the blue spots indicate the ETB-induced petite mutant, and the red spots, the parent isolate.B, Venn's diagram presenting the total number of spots visualized in each 2-D map.IPG = immobilized pH gradient.

Table 1 .
Proteins identified by two-dimensional sodium dodecyl sulfate-polyacrylamide gel electrophoresis and matrix-assisted laser desorption/ionization tandem time-of-flight mass spectrometry.
www.bjournal.com.brBraz J Med Biol Res 43(12) 2010 www.bjournal.com.brBraz J Med Biol Res 43(12) 2010 a Primary accession numbers from the Candida glabrata NCBI blast database (http://www.expasy.org/tools/blast/).b Mascot protein score obtained for the protein indicated.Individual protein scores >50 indicate identity or extensive homology (P < 0.05).c Description of C. glabrata or Saccharomyces cerevisiae homologous proteins as denoted in the NCBI blast database.d Relative amount of the protein spot in the ethidium bromide (ETB)-induced petite mutant compared to its parent isolate (parent isolate/ETB-induced petite mutant).e Spots present exclusively in the parent (P) isolate.f Spots present exclusively in the ETB-induced petite mutant (M).www.bjournal.com.brBraz J Med Biol Res 43(12) 2010

Table 2 .
Spots identified by MS/MS that were higher or lower in the ethidium bromide (ETB)-induced petite mutant compared to parent cells.
a Protein abundance in the ETB-induced petite mutant: ↑ represents higher and ↓ represents lower than in the parent cells.www.bjournal.com.brBraz J Med Biol Res 43(12) 2010

Table 3 .
Summary of spots detected only in the parent isolate or its derived ethidium bromide (ETB)-induced petite mutant.

Table 4 .
Transketolase, aconitase and catalase activities in cytosolic preparations obtained from the parent isolate and the ethidium bromide (ETB)-induced petite mutant.