SEARCH

SEARCH BY CITATION

Keywords:

  • immunogenic peptides;
  • medium proteins;
  • proteomics;
  • proteotypic peptides;
  • secretory proteins;
  • wall proteins

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Concluding remarks
  7. Acknowledgements
  8. Supporting information on the internet
  9. References
  10. Supporting Information

The pathogenic fungus Candida albicans secretes a considerable number of hydrolases and other proteins. In-depth studies of the C. albicans secretome could thus provide new candidates for diagnostic markers and vaccine development. We compared various growth conditions differing in pH, temperature and the presence of the hyphal inducer N-acetylglucosamine. The polypeptide content of the growth media was ca. 0.1–0.2% of the total biomass. Using LC–tandem mass spectrometry, we identified 44 secretory proteins, the transmembrane protein Msb2, six secretory pathway-associated proteins and 28 predicted cytosolic proteins. Many secretory proteins are wall-related, suggesting that their presence in the growth medium is at least partially due to accidental release from the walls. Als3, Csa2, Rbt4, Sap4 and Sap6 were enriched in the medium of hyphal cultures; Bgl2, Cht3, Dag7, Eng1, Pir1, Rbe1, Scw11, Sim1/Sun42, Xog1 and Ywp1 in the medium of yeast cells; and Plb4.5 in pH 4 medium. Seven proteins (Cht3, Mp65, Orf19.5063/Coi1, Scw11, Sim1, Sun41 and Tos1) were consistently present under all conditions tested. These observations indicate that C. albicans tightly regulates its secretome. Mp65, Sun41, and Tos1 were each predicted to contain at least one highly immunogenic peptide. In total, we identified 29 highly immunogenic peptides originating from 18 proteins, including two members of the family of secreted aspartyl proteases. Fifty-six peptides were identified as proteotypic and will be useful for quantification purposes. In summary, the number of identified secretory proteins in the growth medium has been substantially extended, and growth conditions strongly affect the composition of the secretome. Copyright © 2010 John Wiley & Sons, Ltd.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Concluding remarks
  7. Acknowledgements
  8. Supporting information on the internet
  9. References
  10. Supporting Information

Candida albicans is commonly found as a benign commensal in many warm-blooded animals. In humans it mainly resides on the skin and on mucosal surfaces without causing significant harm to the host. But under certain circumstances, such as a weakened host immune system, epithelial damage, or disturbance of the microbial gut flora, C. albicans can cause superficial to deep-seated mucosal infections. When the immune system of the host is severely compromised, e.g. during HIV infection, by chemotherapy or after organ transplantation, the fungus can penetrate the tissue even further and gain access to the bloodstream, which causes life-threatening systemic infections.

Several factors contribute to the pathogenic nature of Candida albicans (Calderone and Fonzi, 2001). A major virulence factor is the ability of C. albicans to switch from yeast to hyphal growth. Loss of this ability leads to complete avirulence in a mouse model of systemic infection (Lo et al., 1997). The wall and especially the covalently anchored wall proteins are important contributors to pathogenicity as well. They mediate adherence and invasion of host cells, promote biofilm formation and protect against the immune system (Klis et al., 2009). Secreted hydrolytic enzymes, like lipases and proteases, play a role in tissue degradation. In addition, they facilitate nutrient acquisition and invasion (Schaller et al., 2005). Identification of other secreted proteins could determine novel virulence factors. The identification of secreted proteins could furthermore be of importance with regard to the development of clinical markers or vaccine candidates.

Proteins in the growth medium that follow the classical secretory pathway are marked for secretion by an N-terminal signal sequence. However, some proteins have been identified in the medium of C. albicans that do not possess this typical signal for secretion, raising the question of whether they reach the extracellular space via an alternative route (Albuquerque et al., 2008; Monteoliva et al., 2002). A genome-wide computational study, based on predicting the presence of a signal peptide, has identified 283 potentially secreted proteins. Using mass spectrometric analysis of the secretome of C. albicans grown under four different growth conditions, we identified 44 secretory proteins, a soluble form of the transmembrane protein Msb2, six proteins predicted to be associated with compartments of the secretory pathway and 28 cytosolic proteins in the growth medium of Candida albicans, thereby confirming and considerably extending earlier studies (Hiller et al., 2007; Maddi et al., 2009; Thomas et al., 2006). Additionally, we present evidence that many covalently anchored wall proteins are partially released into the growth medium. We show further that the protein composition of the secretome changes considerably in response to environmental conditions. Interestingly, we identified a core set of seven proteins that are found under all conditions tested. Most of the genes belonging to this core set are highly conserved among other fungi, underlining their functional importance. Finally, using the immunoinformatics algorithm POPI, we identified a number of peptides that are predicted to elicit a strong response in one or both arms of the immune response and thus might be valuable targets for vaccine development.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Concluding remarks
  7. Acknowledgements
  8. Supporting information on the internet
  9. References
  10. Supporting Information

Strains and growth conditions

Chemicals were obtained from Sigma-Aldrich unless otherwise stated. C. albicans SC5314 (Gillum et al., 1984) was used throughout this study. C. albicans was pre-cultured in liquid YPD medium (10 g/l yeast extract, 20 g/l peptone and 20 g/l glucose) in a rotary shaker at 200 rpm and 30 °C overnight. The overnight culture was used to inoculate flasks containing 50 ml YNBS (6.7 g/l YNB, 20 g/l sucrose and 75 mM MOPSO [3-(N-morpholino)-2-hydroxypropanesulphonic acid] set at pH 7.4 or 75 mM tartaric acid set at pH 4 at an initial OD600 = 0.05. These cultures were incubated for 18 h at 30 °C or 37 °C and shaken at 200 rpm. For hyphal induction, the culture medium was supplemented with 5 mMN-acetylglucosamine (GlcNAc).

Analysis of overnight cultures

For morphological analysis, the cells were visualized by light microscopy and photographed. For determination of the biomass, the cultures were spun down after 18 h of growth and the pellet was dried at 60 °C and weighed. To isolate the proteins from the growth solution, the cultures were spun down and the supernatant was centrifuged again to remove remaining cells. The medium was concentrated using 10 kDa cut-off filters (Amicon Ultra-15 Centrifugal filter units, Millipore). The whole procedure was performed at 4 °C. After concentration, the medium proteins were quantified using the BCA assay with bovine serum albumen (BSA) as a standard (Smith et al., 1985). The amount of protein was normalized to biomass dry weight.

Mass spectrometry of growth solution proteins

The concentrated proteins from the growth medium were reduced with 10 mM dithiothreitol in 100 mM NH4HCO3 (1 h at 55 °C). After cooling to room temperature, the protein solutions were transferred to 10 kDa cut-off spin filter (Millipore) tubes and centrifuged. The reduced proteins were alkylated with 65 mM iodoacetamide in 100 mM NH4HCO3 for 45 min at room temperature in the dark. The samples were quenched with 55 mM dithiothreitol in 100 mM NH4HCO3 for 5 min. Subsequently, samples were washed six times with 50 mM NH4HCO3 and either frozen in liquid nitrogen and stored at − 80 °C or directly digested using 2 µg Trypsin Gold (Promega, Madison, WI, USA) from a 1 µg/µl stock solution for 18 h at 37 °C. The tryptic digests were desalted using a C18 tip column (Varian, Palo Alto, CA, USA) according to the manufacturer's instructions and the peptide concentration was determined at 205 nm using a NanoDrop ND-1000 (Isogen Life Science, IJsselstein, The Netherlands) (Desjardins et al., 2009). Each sample was diluted with 0.1% trifluoroacetic acid to a final concentration of 75 ng/µl and 10 µl per run were injected onto an Ultimate 2000 nano-HPLC system (LC Packings, Amsterdam, The Netherlands) equipped with a PepMap100 C18 reverse-phase column (25 cm × 75 µm i.d.; Dionex, Sunnyvale, CA, USA). We used an elution flow rate of 0.3 µl/min along a linear gradient with increasing acetonitrile concentration over 45 min. The eluting peptides were directly ionized by electrospray in a Q-TOF (Micromass, Whyttenshawe, UK). Survey scans were acquired from m/z 350–1200. For low-energy collision-induced dissociation (MS/MS), the most intense ions were selected in a data-dependent mode.

Analysis of mass spectrometric data

The generated spectra were processed using the MaxEnt3 algorithm included in the Masslynx Proteinlynx software. The resulting pkl (peak list) files were submitted to an internally licensed version of MASCOT (Matrix Science, UK) using both a complete C. albicans proteome database derived from the complete ORF translation from the Candida Genome Database (Arnaud et al., 2007) and a dedicated database of GPI anchored and signal peptide-containing proteins with 143 entries. To generate this database, a preselected list of potential proteins was subjected to signal peptide prediction, using SignalP3.0 (Bendtsen et al., 2004). In addition, all potential proteins were analysed for the presence of a GPI anchor sequence using the BIG-PI fungal predictor (Eisenhaber et al., 2004). Subsequently, predicted N-terminal signal peptides and amino acids C-terminal of the predicted equation image-amino acid involved in GPI attachment were removed from the database. In MASCOT two miscleavages and a tolerance of 0.6 Da for peptides and MS/MS were allowed. Probabilistic MASCOT scoring was used to evaluate the identified peptides and proteins. p < 0.05 was considered significant for peptide identification. Four to six independently obtained biological samples were analysed for each condition (biological replicates). Each biological sample was subjected to three runs with MS/MS selection switching times of 2, 1.5 and 1.25 s, respectively. For a semi-quantitative analysis of our data, we calculated for each growth condition the mean of the total number of peptide identifications per biological replicate.

Peptides identified in at least 50% of all runs in a single condition were considered to be proteotypic (Mallick et al., 2007) (for details, see also Supporting information, Tables S5, S6). The immunogenicity of peptides was predicted using the POPI 2.0 web server (Tung and Ho, 2007).

Results and discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Concluding remarks
  7. Acknowledgements
  8. Supporting information on the internet
  9. References
  10. Supporting Information

Selection of growth conditions for secretome analysis

As C. albicans is known to acidify its environment (Tsuboi et al., 1989), a stable pH requires buffering of the culture medium. For cultures grown at pH 4, we selected tartaric acid as a buffer. It is not metabolized by C. albicans (data not shown) and has two pKa values, one at 3.01 and the other at 4.37, which makes it suitable for buffering at pH 4. To buffer cultures at pH 7.4, we used MOPSO, which has a pKa of 6.90 at 37 °C. It is a hydrophilic molecule that minimizes its uptake by the cell in its uncharged form. We used buffer concentrations of 75 mM, which kept the pH stable throughout the overnight culture period used (data not shown).

Elevated glucose concentrations inhibit hyphal growth and activate the glucose repression pathway leading to the repression of many genes, including several genes involved in the utilization of alternative carbon sources (Maidan et al., 2005; Sabina and Brown, 2009). To avoid glucose repression, we used sucrose as a carbon source. Sucrose does not affect hyphal growth and allows the expression of glucose-repressed genes such as HEX1, which encodes a secretory N-acetylglucosaminidase (Niimi et al., 1987), and MAL2, which encodes an α-glucosidase (maltase) (Geber et al., 1992). Importantly, in contrast to S. cerevisiae, C. albicans does not secrete an invertase to hydrolyse sucrose extracellularly but uses a H+-sucrose symporter to transfer it over the plasma membrane, and hydrolyses it intracellularly using Mal2 (Williamson et al., 1993).

To establish whether the composition of the C. albicans secretome depends on growth conditions, we selected conditions that were expected to result in a distinct protein expression and morphology (Table 1) (Bensen et al., 2004; Gow, 2009; Simonetti et al., 1974; Sudbery et al., 2004; Whiteway and Bachewich, 2007). When grown overnight at pH 7.4 and 37 °C, the culture consisted of a mixture of yeast and hyphal cells. In the presence of GlcNAc, hyphal growth predominated (data not shown). When the temperature was decreased to 30 °C or the medium pH lowered to 4, the cultures consisted exclusively of yeast cells.

Table 1. Morphology, biomass production, and medium protein concentration of overnight cultures grown under various conditions
Growth conditionaMorphology of overnight culturesBiomass (mg/ml)Medium protein (µg/mg biomass)
  • a

    YNB containing 2% sucrose as carbon source.

  • Averages of at least three cultures are shown (coefficient of variation ⩽25%).

pH 7.4, 37 °CYeast and hyphae2.20.83
pH 7.4, 37 °C + 5 mM GlcNAcMainly hyphae2.02.32
pH 7.4, 30 °CExclusively yeast cells2.80.98
pH 4.0, 37 °CExclusively yeast cells3.30.87

After concentration of the C. albicans culture supernatants, their protein contents were determined and normalized to the biomass of the cultures. Table 1 shows that growth was optimal at pH 4 and 37 °C and that hyphal growth resulted in diminished biomass formation. Except for the GlcNAc-supplemented culture, the protein content of the culture supernatants was similar [0.8–1.0 µg protein/mg biomass (dry weight), corresponding to about 0.1% of the total biomass]. The growth medium of GlcNAc-induced cultures contained about 2.3 µg protein/mg biomass, suggesting that hyphae either secrete more proteins or are more sensitive to shearing forces occurring in shaken cultures.

Composition of the Candida albicans secretome

The number of predicted secretory proteins (proteins with an N-terminal secretory signal sequence but without an internal transmembrane sequence) for C. albicans is 283 (Lee et al., 2003), including 104 predicted GPI proteins, which are believed to become tail-anchored to the plasma membrane or the cell wall (De Groot et al., 2003). Under our growth conditions, we identified a total of 44 secretory medium proteins. Interestingly, 29 of them have a cell wall-related function or location, including 18 GPI proteins that have been described as being covalently anchored to the cell wall (Klis et al., 2009). This is consistent with earlier studies of C. albicans and Candida glabrata, in which also known GPI wall proteins were found in the culture solution (Hiller et al., 2007; Stead et al., 2009). As our cultures were shaken at 200 rpm, it seems possible that non-covalently bound wall-associated proteins, such as Bgl2, Cht1, Cht3, Eng1 and Xog1 are partially washed out of the walls. Wall proteins known to be covalently bound to the wall polysaccharides, such as GPI proteins, Pir1, Scw11, Sim1/Sun42 and Tos1, might also be partially washed out, viz. after their release from the plasma membrane but before their covalent attachment to skeletal polysaccharides. Release might also take place during cell wall remodelling, for example, during isotropic growth. This hypothesis is supported by the identification of β-1,3-glucan-associated forms of the wall proteins Als3 and Hyr1 in the supernatant of C. albicans cultures (Torosantucci et al., 2009). In addition to the wall-related proteins, 15 secretory proteins were identified without a known wall-related function or location, including five members of the Sap (secreted aspartyl protease) family, a glucoamylase, a hexosaminidase, and a phospholipase. A fragment of Msb2, a transmembrane protein located in the plasma membrane (for further discussion, see below), and six proteins predicted to be associated with intracellular compartments of the secretory pathway were also found. Finally, 28 predicted cytosolic proteins were identified; many of which are known to be highly abundant intracellularly (see Supporting information, Tables S1–4) (Kusch et al., 2008).

Proteotypic peptides are defined as peptides that are observed in at least 50% of all corresponding protein identifications in mass spectrometry-based proteomics; they are responsible for the majority of protein identifications (Mallick et al., 2007). Since we performed 12–18 LC–MS/MS runs per condition, we have a sound basis to recognize proteotypic peptides. We identified 56 proteotypic peptides in 23 proteins, with up to six proteotypic peptides per protein (see Supporting information, Tables S5, S6). An important use of proteotypic peptides is absolute quantification of proteins (Mallick et al., 2007).

Dynamics of the Candida albicans secretome

Because the total number of peptide identifications of a particular protein is correlated with the concentration of that protein (Liu et al., 2004), we calculated for each protein the average number of peptide identifications/biological replicate under the four growth conditions studied (Table 2). This semi-quantitative approach shows that different modes of growth, such as yeast or hyphal growth, or certain conditions, such as low pH, lead to clear differences in the protein composition of the growth solution (Tables 2, 3). Seven proteins (Cht3, Mp65, Orf19.5063/Coi1, Scw11, Sim1/Sun42, Sun41 and Tos1) were found under all four growth conditions when ‘the mean ≥ 1 in all four conditions’ was used as the threshold value. Notice, however, that the concentrations of Cht3, Scw11 and Sim1/Sun42 were considerably decreased in hyphal cultures. Importantly, we identified almost all predicted tryptic peptides of Mp65, Sim1/Sun42 and Sun41 that were in the detection range of our instrument, suggesting that they are abundant medium proteins. This is also consistent with their high Codon Adaptation Index values of 0.65, 0.70 and 0.61, respectively (Candida Genome Database: http://www.candidagenome.org/). Table 3 further shows that four or five proteins are enriched in the growth medium of a hyphal culture (Als3, Rbt4, Sap4, Sap6 and presumably also Csa2). The transcript levels of PIR1, RBE1 and YWP1 rapidly and strongly decline and remain low when YPD-grown yeast cells are transferred to a hyphal induction medium (Sohn et al., 2003). Consistent with this, peptides of Pir1, Rbe1 and Ywp1 are absent in the growth medium of hyphal cultures (GlcNAc-induced hyphal growth) but present in the media of yeast(-containing) cultures. In total, 10 proteins are enriched in the media of yeast-containing cultures (Bgl2, Cht3, Dag7, Eng1, Pir1, Rbe1, Scw11, Sim1/Sun42, Xog1 and Ywp1). The medium of pH 4-grown cells is clearly enriched in Plb4.5 and probably also, but to a lesser extent, in Fgr41/Pga35, the opaque phase-specific proteins Op4 and Pry1, and the aspartyl proteases Sap8 and Sap10. These observations illustrate the dynamic nature of the secretome.

Table 2. Mass spectrometric analysis of the proteins released into the growth medium under various environmental conditions
   Mean of the peptide identifications in the biological replicatesc
ProteinProperties and functionaProteo-typic peptidesb (No.)pH 7.4 37 °C Y + HpH 7.4 37 °C GlcNAc HpH 7.4 30 °C YpH 4 37 °C Y
  • a

    CGD, Candida Genome Database; GPI-WP, GPI-anchored wall protein.

  • b

    For sequences of the proteotypic peptides see Supporting information, Table S6.

  • c

    Mean of the total number of peptide identifications in the biological replicates (n = 4–6); see also Supporting information, Tables S1–S4.

  • d

    Predicted to be ER- or Golgi-associated proteins.

  • e

    Identical peptides for both proteins.

  • f

    Transmembrane protein associated with the plasma membrane.

  • g

    Proposed name: Coi1.

  • Y, exclusively yeast cells; H, predominantly hyphal growth; Y + H, both growth forms present in considerable amounts.

Wall-related proteins
Als3GPI-WP, adhesin31.06.500
Als4GPI-WP, adhesin0000.20
Bgl2Transglucosylase23.00.34.27.4
Cht1Chitinase1002.83.0
Cht2GPI-WP, chitinase10.83.72.81.6
Cht3Chitinase38.21.310.28.4
Crh11GPI-WP, transglycosylase00.4000
Ecm33GPI-WP, wall integrity02.000.81.6
Eng1Endo-1,3-β-glucanase33.808.05.6
Fgr41Pga35, GPI protein00.200.21.4
Mp65Transglycosylase613.21014.515.4
Pga4GPI-WP, transglucosidase10.801.02.0
Pga45GPI-WP, unknown00.40.30.20.4
Phr1GPI-WP, transglucosidase00.2000
Pir1Wall cross-linking protein11.201.53.6
Rbt1GPI-protein, Flo11 domain00.2000
Rbt5GPI-WP, acquisition of haemoglobin10.60.32.51.8
Rhd3Pga29, GPI-WP00000.4
Sap10GPI-WP, protease00000.8
Scw111,3-β-Glucanase510.61.310.28.4
Sim1Sun42, wall maintenance37.01.89.211.6
Sod4GPI-WP, superoxide dismutase00.2000
Sod5GPI-WP, superoxide dismutase000.700
Ssr1GPI-WP, wall structure11.2002.8
Sun41Wall maintenance49.26.01012.2
Tos1Unknown610.48.711.813.6
Utr2GPI-WP, glycosidase33.40.508.0
Xog1Exo-1,3-β-glucanase2100.8107.6
Ywp1GPI-WP03.602.22.6
Other proteins
Ape2dMetallo-aminopeptidase000.200
Csa2CFEM domain00.61.700
Cyp5dPeptidyl-prolyl cis-trans-isomerase00.20.70.20
Dag7Unknown11.802.83.8
Gca1/2eGlucosyl hydrolases00.400.21.2
Hex1Hexosaminidase00.2000
Kar2dChaperone-like protein000.700
Mnt1dα1,2-Mannosyl transferase00.4000
Msb2fPutative sensor protein02.00.71.20.2
Op4Opaque-specific00000.8
Orf19.5063gCiclopirox olamine-induced37.03.37.21.6
Pdi1dProtein disulphide-isomerase01.0000
Plb4.5GPI-protein, phospholipase10006.0
Pra1Immune evasion protein00.2000
Pry1Opaque-specific, SCP_PRY-like domain00000.4
Rbe1SCP_PRY-like domain01.603.51.2
Rbt4SCP_PRY-like domain30.86.82.00.2
Sap4Secreted aspartyl protease106.200
Sap5Secreted aspartyl protease00.20.300
Sap6Secreted aspartyl protease109.200
Sap8Secreted aspartyl protease00000.6
Ubi3dRibosomal protein00.61.700
Table 3. Relative enrichment of proteins in the growth solution under various environmental conditions
ConditionsaCharacteristic featuresSecretory proteinsb
  • a

    Growth conditions: (1) pH 7.4, 37 °C; (2) pH 7.4, 37 °C, + 5 mM GlcNAc; (3) pH 7.4, 30 °C; (4) pH 4, 37 °C.

  • b

    Proteins with an average total peptide score in the biological replicates ≥ 1.

All Mp65, Cht3, Orf19.5063/Coi1, Scw11, Sim1, Sun41, Tos1
2Predominantly hyphaeAls3, Csa2, Rbt4, Sap4, Sap6
1 + 3 + 4Mixed culture (1) or exclusively yeast cells (3,4)Bgl2, Cht3, Dag7, Eng1, Pir1, Rbe1, Scw11, Sim1, Xog1, Ywp1
4pH 4Plb4.5

Discussion of individual secretory proteins in the growth solution

Csa2 is a small secretory protein without a GPI-anchor signal sequence. It has a CFEM domain but its function is unknown. Interestingly, it seems to be enriched in hyphal cultures (Table 3).

Fgr41 (Pga35) is a predicted GPI protein; the encoding gene has been annotated as a possibly spurious ORF (CGD). We have identified Fgr41/Pga35 in the growth medium in three of the four conditions tested, showing that the corresponding gene is a genuine ORF.

Gca1/2: the tryptic digest of medium proteins contained a peptide (in three growth conditions) that is shared by two closely related predicted secretory proteins, Gca1 and Gca2. Both are predicted glucoamylases. This is consistent with the observation that C. albicans can grow on soluble starch as a carbon source (Nickerson and Mankowski, 1953). Conceivably, it might also allow C. albicans to use host glycogen as a carbon source.

Hex1 is an N-acetylglucosaminidase. When GlcNAc is used as a carbon source, Hex1 activity is strongly induced (Niimi et al., 1997). As Hex1 is only found in the growth medium under a single growth condition (pH 7.4, 37 °C) and not in GlcNAc-supplemented medium, it seems likely that the presence of sucrose prevents the induction of Hex1 and that GlcNAc primarily acts a hyphal-inducing compound and does not need to be metabolized for this effect to occur.

Mp65 is found in all four growth media and its level does not seem to change significantly. This is in agreement with earlier observations indicating that the gene MP65 is constitutively expressed during yeast and hyphal growth (Sohn et al., 2003).

Msb2 lacks an N-terminal signal peptide. It is described as a sensor of cell wall damage and is predicted to be a transmembrane protein associated with the plasma membrane (Roman et al., 2009). We identified five different peptides that are unique for Msb2 in tryptic digests of medium proteins, indicating that Msb2 or a processed form of Msb2 is also present in the culture medium. This is consistent with the observations by Vadaie et al. (2008), who found in S. cerevisiae that Msb2 is processed into medium and cell-associated forms. As a result, a large fragment of ScMsb2 ends up in the growth medium (Vadaie et al., 2008). Interestingly, the peptides identified by us in CaMsb2 are located closely to each other (1154–1177, 1212–1219, 1223–1237, 1238–1253, 1254–1290) in a region that is homologous to the so-called cleavage domain of ScMsb2, consistent with the notion that CaMsb2 and ScMsb2 are functionally related (Vadaie et al., 2008).

Op4 and Pry1 are found only in the medium of pH 4-grown cells. Both have been described as specific for opaque-phase cells, but in view of our results it cannot be excluded that they are also synthesized by white-phase cells at low pH.

Orf19.5063 is a predicted protein in Assemblies 19, 20, and 21 of the Candida genome. We found it in the culture medium under all four growth conditions. Interestingly, Orf19.5063 is upregulated in the presence of the antifungal compound ciclopirox olamine, which causes a transcriptional profile resembling that of iron restriction (Sigle et al., 2005). Therefore we propose the name ciclopirox olamine induced 1 (Coi1) for Orf19.5063. Intriguingly, Coi1 is conserved only in Candida albicans, C. dubliniensis, C. tropicalis and Lodderomyces elongisporus, which have all been implicated in causing bloodstream infections in humans.

Plb4.5 is a putative phospholipase B and is predicted to be GPI-anchored. It is only found in the medium of pH 4-grown cells. Under these conditions, nine different peptides were identified, suggesting that it might be an abundant protein.

Pry1, Rbe1 and Rbt4 share a conserved domain (SCP_PRY1-like domain), but the function of this domain is unknown.

Rhd3 is a GPI wall protein that is also known as Pga29. We detected it only in the medium of pH 4-grown cells. It does not seem to be strictly correlated with yeast growth because it was absent in a yeast culture growing at pH 7.4 and 30 °C (Table 2).

The Sap family, which has been extensively investigated, consists of 10 members, two of which are wall-bound GPI proteins (Sap9 and Sap10), whereas the others are secreted into the growth medium (Naglik et al., 2003). The subfamily SAP4 to SAP6 is known to be coordinately expressed with hyphal growth and in accordance with this we found Sap4 and Sap6 (and to a much lesser extent Sap5) in the medium of hyphal cultures, but not in the media of yeast cultures (Tables 2, 3). Interestingly, Sap 8 and the GPI-modified Sap10 are only found in pH 4-grown cell cultures. The Sap family thus represents a clear example of the adaptability of the secretome.

Sim1 (Sun42) and Sun 41 both belong to the Sun family and share the so-called SUN domain. The transcript levels of SIM1/SUN42 strongly decrease during hyphal growth (Firon et al., 2007; Hube et al., 1994), consistent with our observation that the level of Sim1 in the medium of a predominantly hyphal culture is much lower than in (predominantly) yeast cultures.

Finally, Ape2, Cyp5, Kar2, Mnt1 and Pdi1 each possess a predicted N-terminal signal peptide or a transmembrane domain close to the N-terminus and are believed to be associated with the endoplasmic reticulum or the Golgi apparatus (Buurman et al., 1998; Skrzypek et al., 2010). Their presence in the medium is consistent with the observation that the culture medium of C. albicans contains membrane vesicles (Albuquerque et al., 2008). Table 2 shows that they are mainly found in the medium of (partially) hyphal cultures, and are almost completely absent in yeast cultures, suggesting that their presence in the medium is due to hyphal breakage. Intriguingly, Ubi3 also possesses a predicted N-terminal signal peptide or N-terminal transmembrane domain, raising the question of whether this predicted ribosomal protein might be associated with the rough endoplasmic reticulum.

Identification of peptides with high immunogenic potential

All identified tryptic peptides were subjected to a prediction of their potential to induce the proliferation of cytotoxic (CTL) and helper T lymphocytes (HTL; Table 4) (Tung and Ho, 2007). In our complete set of 228 peptides, 29 (∼13%) show a high immunogenic potential, while 77 (∼34%) show no immunogenic potential at all, whereas the remainder had an intermediate potential (Table 4; see also Supporting information, Table S6). Mp65, which seems to be an abundant medium protein under all growth conditions tested, also contains a predicted highly immunogenic peptide (LYGVDCDQVSAVLK). This result was consistent with the predicted antigenicity of this peptide by another algorithm (http://bio.dfci.harvard.edu/Tools/antigenic.html). Based on mouse vaccination studies, Mp65 seems indeed a promising vaccine candidate (Pietrella et al., 2008). Sim1/Sun42 and Sun 41 are functionally redundant and the double deletant is non-viable (Firon et al., 2007). As the SUN domain of both proteins contains peptides with a high predicted immunogenicity (Table 4), this domain also seems an attractive vaccine candidate.

Table 4. Predicted immunogenicity of secretory protein peptides
  Immunogenicitya
ProteinSequence (position)CTLbHTLc
  • a

    POPI algorithm (Tung and Ho, 2007); 3, PD50 < 1 nM; 2, PD50 = 1–100 nM; 1, PD50 = 100 nM–10 mM; 0, PD50 > 10 mM, in which PD50 = protective dose that protects 50% of the animals challenged.

  • b

    Cytotoxic T lymphocytes.

  • c

    Helper T lymphocytes.

Als3/5APFTLR (306–311)30
Cht2LSSAIEEIK (259–267)31
Eng1ELAANIAATVK (797–807)23
 DASNPSADDTYFPVSR (932–947)31
Gca1/2LNVHIEPTDLTDVFVLPEELVVKPK (108–132)30
 GHSITGLGESIHGSLNEPGVVK (193–214)03
 YFDNPVHPPFEVGYSGSDYPLGFDK (479–503)30
Mp65LYGVDCDQVSAVLK (161–174)13
Msb2SALNYPFVVENSISSAQIFQYLPR (1154–1177)33
 ALGSFITTPGSAIYR (1223–1237)32
Pga4AGIYVILDVNTPHSSITR (99–116)31
Pga45DANTEQTIEGILK (86–98)03
Rbt4LGCAYK (312–317)22
 SYMAENVLRPQ (348–358)22
Sap4/6TLSVGLR (269–275, 270–276)03
Sap4LSVIVDTGSSDLWVPDSNAVCIPK (102–125)03
 YADGSVAQGNLYQDTVGIGGVSVR (159–182)03
Sap6GNLYQDTVGIGGASVK (168–183)03
Sim1TDYPGSENMNIPTLLSAGGK (233–252)33
 TSTQYYVNNAGVSVEDGCIWGTEGSGVGNWAPVVLGSGTTGGK (271–313)03
Sod5HGNIMGESYK (118–127)22
 TEYDDSYISLNEK (128–140)23
Sun41IVGESGSTVSGSCSYANGK (376–394)03
Tos1DSYYTPGSTDNCVFLNYHGGSGSGVWSAK (231–259)03
Utr2EIYATAYDIPNDVK (297–310)22
Xog1QISNLGLNFVR (120–130)32
 DSYNFQNGDNTQVTLNVLNTIFK (189–211)22
 QFFLDGYNSLR (243–253)31
Ywp1VINVPAR (87–93)30

Concluding remarks

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Concluding remarks
  7. Acknowledgements
  8. Supporting information on the internet
  9. References
  10. Supporting Information

We have analysed the proteins of C. albicans that are released into the growth medium under various environmental conditions. This allowed us to compare yeast and hyphal cultures and the effect of pH. The polypeptide content of the growth medium was ⩽0.1% of the total biomass for three of the four conditions tested, but was > two-fold higher in the hyphal culture (Table 1). In total we identified 44 secretory proteins, which is considerably more than in earlier studies. Twenty-nine of them had a wall-related location or function, whereas 15 possessed a function unrelated to the cell wall. We also detected a soluble form of the plasma membrane sensor protein Msb2 and six proteins predicted to be associated with early compartments of the secretory pathway. Conceivably, some of the wall-related proteins found in the culture solution could be involved in biofilm formation and, in particular, in the formation of extracellular polysaccharides (Hiller et al., 2007). In total, we detected 28 predicted cytosolic proteins in the various culture solutions (see Supporting information, Tables S1–S4). Interestingly, in pH 4-grown yeast cultures we did not identify any cytosolic protein, whereas in the hyphal culture we identified considerably more cytosolic proteins than in the other cultures, suggesting that, in comparison to yeast cells, hyphae are more liable to breakage, due to the shearing forces inherent to shaking flask cultures. We cannot exclude that medium proteins are degraded over time. However, we found many more tryptic peptides than semitryptic peptides (peptides that match trypsin specificity only at one terminus), suggesting that proteolytic degradation was limited (data not shown). Albuquerque et al. (2008) have described membrane-bound vesicles in the culture medium of Histoplasma capsulatum and other ascomycetous fungi, including C. albicans. Interestingly, when the culture medium is filtered using a 200 nm filter, the number of cytosolic proteins in the filtrate is negligible, whereas the number of secretory proteins does not seem to be affected (unpublished data). This suggests that the cytosolic proteins found in the medium are contained in membranous vesicles.

Compared to the predicted secretomes (proteins with an N-terminal secretory signal peptide, including GPI proteins but excluding proteins with an internal transmembrane sequence) of other yeasts, such as: Kluyveromyces lactis (178 proteins) (Swaim et al., 2008); Saccharomyces cerevisiae (163 proteins) (Yang et al., 2006); Pichia pastoris (88 proteins) (Mattanovich et al., 2009); Schizosaccharomyces pombe (66 proteins) (Liu et al., 2007); C. albicans has a relatively large predicted secretome (283 proteins). Possibly, the secretomes of non-pathogenic yeasts tend to be smaller, as has also been observed for plant pathogenic mycelial fungi (Choi et al., 2010). Alternatively, the higher number of predicted secretory proteins might be related to the pleomorphic nature of C. albicans. Importantly, of the 44 secretory proteins we were able to identify in the growth medium, only Mp65, Sun41 and Tos1 seemed to be consistently and abundantly present under all four conditions. This observation is supported by previous studies (Hiller et al., 2007; Maddi et al., 2009). The corresponding genes are highly conserved in many fungi, underlining their functional importance. These proteins could therefore be useful for the development of clinical markers or as vaccine candidates. Interestingly, a recently developed algorithm for the identification of immunogenic peptides predicts that Mp65, Sun41 and Tos1 each possess a highly immunogenic peptide (Tung and Ho, 2007).

The composition of the secretome differed considerably between pH 4- and pH 7.4-grown cells and between yeast and hyphal cultures (Tables 2, 3). This clearly demonstrates that the composition of the secretome can vary widely. If our assumption that some of the medium proteins are actually washed-out wall proteins is correct, the dynamic nature of the wall proteome of C. albicans (Hoyer et al., 1998; Sosinska et al., 2008) could partially explain the observed differences. However, we also observed relative enrichment of specific hydrolytic enzymes in the medium, depending on growth conditions and morphology (Table 3). This is illustrated by the increased levels of Sap4 and Sap6 in hyphal culture medium and of Plb4.5 at pH 4 (to a lesser extent the levels of Gca1/2 and Sap8 also seem to increase at pH 4; see Table 2). The capability of adapting the composition of the secretome probably helps C. albicans to cope better with the different growth conditions that it encounters on the various sites of infection and during the subsequent infection process. The proposed tight regulation of the composition of the secretome by C. albicans probably also explains, at least to some extent, why only a limited number of the predicted secretory proteins have been identified so far. For example, we did not identify any member of the extended family of secretory lipases (Hube et al., 2000). The most likely explanation is that our growth conditions and in particular the use of sucrose instead of lipids as a carbon source did not favour the expression of the lipase genes (Hube et al., 2000).

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Concluding remarks
  7. Acknowledgements
  8. Supporting information on the internet
  9. References
  10. Supporting Information

We thank Dr Leo de Koning for his stimulating support and advice. F.M.K. acknowledges financial support by the EU Programme FP7-214004-2 FINSysB.

Supporting information on the internet

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Concluding remarks
  7. Acknowledgements
  8. Supporting information on the internet
  9. References
  10. Supporting Information

The following supporting information may be found in the online version of this article:

Table S1. All secretory medium proteins identified at pH 7.4 and 37 °C after 18 h

Table S2. All secretory medium proteins identified at pH 7.4, 37 °C and 5 mM GlcNAc after 18 h

Table S3. All secretory medium proteins identified at pH 7.4 and 30 °C after 18 h

Table S4. All secretory medium proteins identified at pH 4 and 37 °C after 18 h

Table S5. Proteotypic peptides of the core set of secretory proteins and their predicted immunogenicity

Table S6. All identified peptides and their predicted immunogenicity

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Concluding remarks
  7. Acknowledgements
  8. Supporting information on the internet
  9. References
  10. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Concluding remarks
  7. Acknowledgements
  8. Supporting information on the internet
  9. References
  10. Supporting Information
FilenameFormatSizeDescription
yea_1775_supportinginforTs1.pdf12KSupporting Information
yea_1775_supportinginforTs2.pdf12KSupporting Information
yea_1775_supportinginforTs3.pdf11KSupporting Information
yea_1775_supportinginforTs4.pdf12KSupporting Information
yea_1775_supportinginforTs5.pdf7KSupporting Information
yea_1775_supportinginforTs6.pdf15KSupporting Information

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.