SEARCH

SEARCH BY CITATION

Keywords:

  • Fungal cell surface proteins;
  • PGA10;
  • RBT5 and CSA1/WAP1 genes;
  • biofilms;
  • atomic force microscopy;
  • cell surface hydrophobicity

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Several biological features of Candida albicans genes (PGA10, RBT5 and CSA1) coding for putative polypeptide species belonging to a subset of fungal proteins containing an eight-cysteine domain referred as common in several fungal extracellular membrane (CFEM) are described. The deletion of these genes resulted in a cascade of pleiotropic effects. Thus, mutant strains exhibited higher cell surface hydrophobicity levels and an increased ability to bind to inert or biological substrates. Confocal scanning laser microscopy using concanavalin A-Alexafluor 488 (which binds to mannose and glucose residues) and FUN-1 (a cytoplasmic fluorescent probe for cell viability) dyes showed that mutant strains formed thinner and more fragile biofilms. These apparently contained lower quantities of extracellular matrix material and less metabolically active cells than their parental strain counterpart, although the relative percentage of mycelial forms was similar in all cases. The cell surface of C. albicans strains harbouring deletions for genes coding CFEM-domain proteins appeared to be severely altered according to atomic force microscopy observations. Assessment of the relative gene expression within individual C. albicans cells revealed that CFEM-coding genes were upregulated in mycelium, although these genes were shown not to affect virulence in animal models. Overall, this study has demonstrated that CFEM domain protein-encoding genes are pleiotropic, influencing cell surface characteristics and biofilm formation.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Candida albicans is unique among fungal pathogens in terms of the diversity of infections it can cause. The fungus is a normal commensal on the mucosal surfaces of the gastrointestinal and urogenital tract without clinical symptoms in the majority of humans, but by contrast to numerous other commensals, C. albicans has the ability to colonize and invade host tissues when the host immune system is weak or when the competing flora is eliminated, usually causing superficial infection of mucosal epithelium (Calderone, 2002). However, in immunocompromised individuals, infections can progress to a severe systemic invasion, leading to a life-threatening situation. In this context, C. albicans is the major fungal pathogen in humans (Calderone, 2002), and recent surveys in the United States have shown that Candida species are the third to fourth most commonly isolated bloodstream pathogen, having surpassed gram-negative rods in frequency as causative agents of septicaemia in humans (Edmond et al., 1999; Diekema et al., 2002).

The virulence factors expressed or required by C. albicans to invade the host may vary depending on the type of clinical manifestation (i.e. mucosal or systemic), the site and stage of infection, and the nature of the host response. It seems apparent that virulence in C. albicans is multifactorial and although many virulence traits have been suggested, the production of extracellular hydrolytic enzymes, hyphae formation (morphologic transition or dimorphism), phenotypic switching, the presence of surface recognition molecules (adhesins and receptors for ligands in host tissues) and the ability to form biofilms have been the most widely studied (Calderone & Fonzi, 2001; Soll, 2002; Sudbery et al., 2004; Hube, 2006; Ramage et al., 2009).

Adherence of C. albicans to host cells is seen as an essential early step in the establishment of disease. Attachment of microorganisms to tissues is a complex process, involving both specific receptor molecules and nonspecific physical and chemical cell surface properties. The most important factors mediating adhesiveness are considered to be hyphal morphogenesis, cell surface hydrophobicity (CSH) and the presence of cell surface adhesins and receptors for host ligands (Hazen & Glee, 1995; Sundstrom, 1999, 2002; Calderone et al., 2000; Calderone & Fonzi, 2001; Sudbery et al., 2004; Chaffin, 2008).

The yeast-to-mycelium transition in C. albicans is linked to its pathogenic nature. This is due in part to the fact that newly formed filaments (germ tubes) are more adherent to mammalian cells than are yeast cells (Odds, 1988; Cutler, 1991) and adherence certainly is a fundamental stage before host tissue penetration. Genes that control hyphal morphology are coregulated with genes that encode more standard virulence factors such as proteases and adhesins (Pitarch et al., 2006; Chaffin, 2008). Besides, hyphae are essential elements for providing the structural integrity and multilayered architecture characteristic of mature/fully developed biofilms, although yeast-only biofilms have also been described (Baillie & Douglas, 1998, 1999). Candida albicans strains with mutations in genes governing morphogenesis and that are defective in filamentation also display defects in their biofilm-forming abilities (Ramage et al., 2002; García-Sánchez et al., 2004; Kelly et al., 2004; Krueger et al., 2004).

On the other hand, CSH contributes to the pathogenesis of the opportunistic fungal pathogen C. albicans. Hydrophobic C. albicans cells are more adherent than hydrophilic cells to a variety of host tissues, and the pattern of adherence is more widespread (Hazen & Glee, 1995).

Finally, studies aimed at the identification of cell surface components of C. albicans involved in the interaction of fungal cells with host tissues have revealed the existence of a large assortment of cell wall-bound carbohydrates such as mannan, which has been shown to play an important role in adhesion, host recognition and virulence (Calderone & Fonzi, 2001), and proteins displaying adhesin characteristics, such as the glycosylphosphatidylinositol (GPI)-anchored species (GPI-CWP) including the ALS gene family (Hoyer et al., 2008), the CSA1, HYR1, HWP1 and EAP1 gene products (Bailey et al., 1996; Staab et al., 1999; Lamarre et al., 2000; Sundstrom et al., 2002; Li & Palecek, 2008), and a family of surface-bound proteins containing an eight-cysteine domain referred to as common in several fungal extracellular membrane (CFEM), which may function as cell-surface receptors or signal transducers, or as adhesion molecules in host–pathogen interactions (Kulkarni et al., 2003).

In a previous report (Pérez et al., 2006), we described several characteristics and functions of PGA10 (for predicted glycosylphosphatidylinositol-anchored), which is the standard designation given by de Groot et al. (2003) to genes coding for fungal glycosylphosphatidylinositol-anchored proteins without a specific function. PGA10 gene (also designated as RBT51; Weissman & Kornitzer, 2004) codes for a putative member of the CFEM family, whose deletion resulted in a cascade of pleiotropic effects, mostly affecting cell surface-related properties (Pérez et al., 2006). We also examined the biofilm-forming ability, a feature that appears to play a key role in virulence and pathogenesis in C. albicans (Douglas, 2003; d'Enfert, 2006; Nett & Andes, 2006), of C. albicans homozygous mutant strains harbouring single, double and triple deletions for PGA10, as well as RBT5 and CSA1 genes that also code for other CFEM proteins (Braun et al., 2000; Lamarre et al., 2000), and found that these gene products could be involved in the biogenesis and/or the maintenance of biofilm structure and integrity in C. albicans (Pérez et al., 2006). In this paper, we have performed further functional characterization of these mutant strains by examining CSH, relative gene expression profiling, cell surface structure by atomic force microscopy (AFM), additional structural features of biofilms and virulence in animal models. The results reported in this work support the contention for a role of the different proteins belonging to the CFEM family present in C. albicans in the interaction of fungal cells with the external environment (including biofilm biogenesis), although they do not appear to be directly involved in virulence. Consequently, further work is necessary to fully elucidate all possible aspects of the biological and functional role (for instance, to determine whether these genes may represent potential biological targets for new anti-Candida therapies) of this intriguing family of proteins in C. albicans.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Strains and growth conditions

The C. albicans strains used are listed in Table 1. Cells were routinely grown in YPD [2% glucose, 1% yeast extract, 2% Bacto peptone (Difco)] or YNB (0.67% yeast nitrogen base without amino acids, 2% glucose) media at 28 °C with shaking (100 r.p.m.). Media were supplemented with uridine (25 μg mL−1) when required.

Table 1. Candida albicans strains used in this work
StrainGenotypeParental strainReference
  • *

    The CAI4-URA3 strain was kindly provided by Dr Gwyneth Bertram, School of Medical Science, University of Aberdeen.

SC5314 Wild-typeGillum et al. (1984)
CAI4ura3Δ∷λimm434/ura3Δ∷λimm434SC5314Fonzi & Irwin (1993)
CAI4-URA3ura3Δ∷λimm434/ura3Δ∷λimm434, RP10URA3CAI4This work*
CAN1ura3Δ∷λimm434/ura3Δ∷λimm434, Pga10Δ∷hisG/pga10hisG RP10URA3CA3This work
BCa18-2ura3Δ∷λimm434/ura3Δ∷λimm434, rbt5Δ∷hisG/rbt5Δ∷hisG-URA3-hisGCAI4Braun et al. (2000)
BCa17-4ura3Δ∷λimm434/ura3Δ∷λimm434, wap1Δ∷hisG/wap1Δ∷hisG-URA3-hisGCAI4Braun et al. (2000)
KC100ura3Δ∷λimm434/ura3Δ∷λimm434, rbt5Δ∷hisG/rbt5Δ∷hisG pga10Δ∷hisG/pga10Δ∷hisG-URA3-hisGCAI4Weissman & Kornitzer (2004)
KCU1ura3Δ∷λimm434/ura3Δ∷λimm434, rbt5Δ∷hisG/rbt5Δ∷hisG pga10Δ∷hisG/pga10Δ∷hisG RP10URA3 This work
KC171ura3Δ∷λimm434/ura3Δ∷λimm434, rbt5Δ∷hisG/rbt5Δ∷hisG pga10Δ∷hisG/pga10Δ∷hisG ccc2Δ∷hisG/ccc2Δ∷hisG wap1Δ∷hisG/wap1Δ∷hisG ade2Δ∷CaCCC2CAI4Weissman & Kornitzer (2004)
GAP3CAI4CAI4 derivative, pPGA10-GFPThis work

Germ tube formation in C. albicans was induced using the starvation method (Casanova et al., 1989). The formation of biofilms by the different C. albicans strains was assessed using the procedure described elsewhere (Ramage et al., 2001). Briefly, cells were grown overnight in an orbital shaker in YPD medium, harvested and washed in sterile 10 mM phosphate-buffered saline (PBS), pH 7.4. Cells were suspended in Roswell Park Memorial Institute (RPMI)-1640 medium supplemented with l-glutamine and buffered with 4-2-hydroxyethyl-1-piperazineethanesulfonic acid (Sigma Chemical Co., St. Louis, MO) to a final concentration of 1 × 106 cells mL−1. Biofilms were formed by pipetting appropriate volumes of the standardized cell suspensions into wells of commercially available presterilized, polystyrene, flat-bottomed, 96-well microtitre plates or cell tissue flasks (Nalge Nunc International, Denmark) and incubated at 37 °C.

Detection of PGA10 promoter activity by flow cytometry of green fluorescent protein (GFP) reporter expression

Plasmid construction was based on the procedure reported by Barelle et al. (2004). The PGA10 promoter region was amplified by PCR using the primers 5′-GCGCCTCGAGTGTTGAAGAGCAGGCTATGG-3′ and 5′-GCGCAAGCTTGCGGATT GACTTGAAGAAAATC-3′ (XhoI and HindIII sites underlined) and cloned between the XhoI and the HindIII sites of pGFP. The plasmid was linearized by digestion with BglII and used to transform C. albicans, generating strain GAP3 (see Table 1).

The relative GFP fluorescence of GAP3 cells was visualized using confocal scanning laser microscopy (CSLM) analysis with an LSM 510 META laser scanning microscope (Zeiss, Germany) attached to an Axioplan II microscope (Zeiss).

The relative gene expression in C. albicans cells was determined by flow cytometry. Strain GAP3 was grown overnight in YPD medium at 28 °C with shaking. Cells were collected by centrifugation and grown under yeast- and hyphae-inducing conditions until an OD600 nm of 0.2 was reached. Aliquots of 0.5 mL from each culture were inoculated in 20 mL of fresh YPD and RPMI media and incubated at 28 and 37 °C, respectively. Samples (10 μL) were taken at different time intervals (0, 0.5, 1, 1.5 and 2 h) and examined by optical microscopy to ensure that only yeast forms and early hyphal filaments were present in each of the cultures. Cells were collected by centrifugation, washed twice in PBS and analysed in a Modular Flow Cytometer (MoFlo, Beckman Coulter).

Gene expression analysis by quantitative reverse transcriptase (RT)-PCR

PGA10 gene expression in both morphological forms was assessed by quantitative RT-PCR and compared with the expression of the 18S housekeeping gene.

For this purpose, total RNA from yeast and hyphal forms was isolated from C. albicans with TRizol reagent (Invitrogen, Paisley, UK) following the manufacturer's protocol, after homogenizing the cells for three periods of time (30 s each) using a mini-beadbeater that intensely agitates the sealed microcentrifuge vial containing cells and 0.5 mL of glass beads.

For cDNA synthesis in a first step, a known concentration (from 200 ng to 2 μg) of total RNA was mixed with 1 μL of random hexamer pd(N)6 (2 mM) and water up to 7 μL. The solution was denatured at 65 °C for 10 min and allowed to cool on ice for 5 min. Subsequently, 0.5 μL of dNTPs (0.2 μM), 2 μL of Moloney Murine Leukemia Virus (MMLV) 5 × reaction buffer and 0.5 μL of MMLV reverse transcriptase (Invitrogen) were added to each of the tubes (10 μL final volume) and incubated at 37 °C for 1 h and subsequently at 75 °C for 10 min. The cDNA was stored at −70 °C until required.

For each qPCR reaction, 0.25 μL of cDNA (1500 ng μL−1) was mixed with 12.5 μL SYBR® Green (Invitrogen), 0.5 μL Rox reference dye (Invitrogen), 0.5 μL each pair of primers generating unique cDNA amplifications for PGA10 (5′-CCACTACCTCCGACACCACT-3′ and 5′-TTCTGCTGCGGAGGACTT-3′) and 18s (5′-GATTAGATACCCTGGTAG-3′ and 5′-ATTGTAGCACGTGTGTAG-3′), 1.5 μL of DNA Taq polymerase and distilled water to a final volume of 25 μL.

All samples were prepared and processed in triplicate and the mean Ct value was calculated. The cDNA levels during the linear phase of amplification were normalized against the 18S ribosomal housekeeping gene and used to calculate the relative levels of expression using the inline image method.

All quantitative RT-PCR reactions were performed using the DNA Engine Opticon 2qPCR machine (MJ Research, Waltham, MA), and all optical reading data and Ct values were stored and analysed using Opticon Monitor version 2.01 (MJ Research).

Evaluation of relative CSH by flow cytometry

Relative yeast CSH was determined following a method developed by Colling et al. (2005), which is a variant of the latex-polystyrene microsphere assay described by Hazen & Hazen (1987) and López-Ribot et al. (1991). Briefly, yeast cells were suspended in PBS and diluted to an OD600 nm of 1.0 ± 0.1. Polystyrene beads (0.834 μm, average diameter), deep blue-dyed, were purchased from Sigma Chemical Co. A beads suspension was provided in 1 mL of suspension (10% solids). For bead adherence tests, a microsphere stock suspension was vigorously mixed and 10 μL of this suspension was added to a 1 : 100 dilution of a yeast cells suspension in PBS. Beads and yeasts were allowed to interact at room temperature for 30 min on a rotator at 14 cycles min−1. Subsequently, each sample was vortexed vigorously and analysed by flow cytometry in a Beckman Coulter MoFlo cytometer.

AFM

For AFM examination, cells were grown overnight at 30 °C in YPD medium, collected by centrifugation (1–5 mL of cell culture), washed once in PBS and resuspended in about 100 μL of distilled water. A drop of the resulting cell suspension was placed on a coverslip and allowed to dry completely. Coverslips were placed in a Nanoscope IV/Dimension 3100 SPM from Veeco Metrology (Santa Barbara, CA), and the experiments were conducted in the tapping mode for the imagery, using a Phosphorous-doped Si cantilever (RTESP Model, Veeco Metrology). The parameters for measurement were as follows: Spring constant, ∼80 N m−1; resonance frequency, ∼260 kHz; scan speed, 0.5 Hz; and scan range varying from 500 nm to 10 μm. Images were processed using the nanoscope iv version 5.30 software.

Determination of biofilm structure by CSLM

Biofilms formed by different mutant strains studied in this work were further characterized by CSLM.

Cells were grown under biofilm-forming conditions as described by Ramage et al. (2001) on Thermanox coverslips placed in sterile 12-well cell culture plates. After incubation for 48 h, coverslips were transferred to new cell culture plates, gently washed with PBS and incubated in the dark for 30 min in 1 mL of PBS containing FUN-1 (10 μM) alone or a mixture of FUN-1 (10 μM) and Concanavalin A (ConA)-Alexa Fluor 488 (5 μM) (Molecular Probes, Eugene, OR). FUN-1 is a cytoplasmic fluorescent probe to assess cell viability, whereas the ConA-Alexa Fluor 488 Conjugate selectively binds to mannose and glucose residues present in polysaccharides, which are major constituents of the cell wall and the extracellular matrix (EM) of biofilms in C. albicans.

The CSLM analysis was performed using an LSM 510 META laser scanning microscope (Zeiss) attached to an Axioplan II microscope (Zeiss). Biofilms were observed using × 40 and × 100 oil immersion objectives. The excitation wavelengths were 488 nm (Argon laser) and 543 nm (He-Ne laser), and the emission wavelengths were 505 and 560 nm for Alexa Fluor 488 and FUN-1, respectively. To determine biofilm structure, a series of horizontal (x–y axes) optical sections were taken throughout the depth of the biofilm (z axis). Three-dimensional representations showing the relative fluorescence for each fluorophore in biofilms formed by the different C. albicans strains were made using the built-in software provided with the equipment.

Virulence assays

Cultures of the different strains were obtained by incubation at 28 °C for 14–16 h in YPD medium. The average OD600 nm determined using a spectrophotometer was found to be in the range of 1.1–1.3 for all the cultures after the incubation period (according to the OD600 nm values measured, all the cultures were close to the late exponential growth phase), which indicated a similar growth rate for all the different strains tested. Cells were subsequently harvested by centrifugation, washed three times in a sterile pyrogen-free saline solution and counted using a haemocytometer chamber. Appropriate suspensions from the cultures to reach a final concentration of 5.0 × 105 cells mL−1 were prepared and 200 μL aliquots from these suspensions containing a total infecting dosage of 105 cells were immediately injected into the lateral tail veins of 6–8-week-old female BALB/c mice (obtained from the National Cancer Institute).

Groups of six mice were used for each of the strains tested, and were observed during a total of 28 days postinfection. For statistical analysis, survival data and differences between groups were analysed using the Kaplan–Meier and log-rank tests. All experiments were performed in accordance with Institutional regulations in an Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC)-certified facility at the University of Texas Health Sciences Center at San Antonio (UTHSCSA). Mice were allowed a 1-week acclimatization period before the experiments were started.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Relative gene expression analysis using a reporter strain

The relative gene expression within individual C. albicans cells was assessed as described in Materials and methods. The PGA10 promoter was inserted into the plasmid pGFP, which contains the codon-optimized yeast enhanced green fluorescent protein (yEGFP; Cormack et al., 1997), generating a promoter–GFP fusion (Fig. 1). Strain CAI4 of C. albicans was transformed with plasmid pPGA10-yEGFP and a new reporter strain, GAP3, was generated.

image

Figure 1.  Structure of the pPGA10-yEGFP reporter plasmid. The vector was linearized with BglII before the integration into the Candida albicans CAI4 strain.

Download figure to PowerPoint

The activity of this GFP fusion was measured in individual cells using fluorescence microscopy (Fig. 2a). GFP fluorescence must increase relatively quickly following gene activation, and decline relatively quickly once the gene is repressed. Therefore, these reporter cells can also be used to determine the relative gene expression by flow cytometry. Cells corresponding to strains CAI4 and GAP3 were grown under yeast- and hyphae-inducing conditions as described in Materials and methods and analysed by flow cytometry. As shown in Fig. 2b, there was an increase in fluorescence (fluorescence channel 1) as the cells began to form hyphae, which means that PGA10 is a gene whose expression is induced under mycelium-forming conditions. These results were confirmed by qRT-PCR analysis, which revealed that the expression of the PGA10 gene increased up to 42 times in mycelia cells compared with yeasts (Fig. 3). A similar overexpression pattern under hyphae-inducing conditions was reported earlier for some genes of the CFEM family (RBT5 and CSA1/WAP1) by Braun et al. (2000) and Lamarre et al. (2000) and more recently by Sosinska et al. (2011), which demonstrated that the protein level of Rbt5 was 10-fold higher in the cell walls of filamentous cultures growing at pH 7.0 compared with yeast cultures growing at pH 4.0.

image

Figure 2. PGA10 expression analysis using a reporter strain. CSLM analysis revealed higher fluorescence levels in biofilms formed by GAP3 strain (see Table 1) (mostly hyphal filaments) compared with planktonic (free-floating forms) yeast cells (a). Studies of PGA10 expression by flow cytometry (b) revealed that fluorescence corresponding to the GAP3 strain is increased under hyphae-inducing conditions (green line), whereas it remains stable under yeast-forming conditions (pink line). Strain CAI4 was used as a negative fluorescence control.

Download figure to PowerPoint

image

Figure 3.  Evaluation of PGA10 gene expression by qRT-PCR. The relative expression of PGA10 was analysed in yeast and hyphae using the endogenous 18S ribosomal gene transcript as a housekeeping (HK) reference. Based on the levels of relative expression, yeast cells expressed 42.52 times less PGA10 (0.00044) than hyphae (0.0187) under the conditions tested. The amplification efficiencies of PGA10 were shown to be similar under both conditions.

Download figure to PowerPoint

Determination of CSH

CSH was determined as described above. Before quantifying CSH in cells, the optimal working conditions for the assay were established. Fig. 4a shows that fluorescence channel 4 (FL4) detected the blue-dyed microspheres alone, whereas the control yeast cells did not show any fluorescence (Fig. 4b). After incubating the cells with the microspheres, the fluorescence level that had a direct correlation with the binding of the microspheres to the cells and therefore with the CSH was measured using flow cytometry. Cells of all mutant strains manifested similar CSH levels that were, in all cases, higher than the CSH exhibited by the control CAI4-URA3 parental strain (Fig. 4c–h). These higher CSH values were in correlation with an increased ability of adhesion to different serum and animal tissues proteins (fibrinogen, laminin, fibronectin and EM) displayed by the mutant strain (results not shown).

image

Figure 4.  Measure of relative CSH. The Candida albicans parental CAI4-URA3 strain did not show fluorescence on the channel FL4 (a), whereas the blue-dyed polystyrene beads showed a very strong fluorescence (b). After incubating the cells with the microspheres, the fluorescence level was measured using a flow cytometer. The control strain cells CAI4-URA3 (c) presented a lower binding degree of microspheres to the cell surface compared with the mutant strains csa1Δ (d), pga10Δ (e), pga10Δ/rbt5Δ (f), rbt5Δ (g), y pga10Δ/rbt5Δ/csa1Δ (h).

Download figure to PowerPoint

Cell surface examination by AFM

The phenotype observed, i.e. higher CSH values and an increased ability of adhesion to different serum and animal tissue proteins, were indicative of a defective cell wall structure and/or composition in the mutant strains. In addition, we have also reported that the null pga10Δ mutant displayed an increased sensitivity to cell wall-perturbing agents such as calcofluor white, Congo red and sodium dodecyl sulphate (Pérez et al., 2006). Consequently, we considered it of interest to perform a visual examination of the cell surface of C. albicans strains bearing single, double and triple deletions for PGA10, RBT5 and CSA1 genes using AFM, a novel and powerful tool that provides a topographical image of cell surface at a high magnification and allows the determination of cell wall nanomechanical and functional (including CSH) properties of yeasts (Dague et al., 2010; Dufrêne, 2010).

Images of the cells were taken in order to obtain information regarding the morphological characteristics of the cell surface for each strain examined. As can be observed in Fig. 5a–d, the parental CAI4-URA3 strain showed a smooth and homogeneous cell surface, whereas the single mutants such as pga10Δ, rbt5Δ and csa1Δ showed a highly heterogeneous and rougher cell surface that was clearly noticeable in the case of cells from strain csa1Δ. Alterations of the cell surface were dramatically apparent in the case of the strains bearing double and triple gene knockouts such as pga10Δ/rbt5Δ and pga10Δ/rbt5Δ/csa1Δ (Fig. 5e and f).

image

Figure 5.  AFM observations at two scan ranges of the surface of Candida albicans yeast cells from CAI4-URA3, pga10Δ, rbt5Δ, csa1Δ, pga10Δ/rbt5Δ and pga10Δ/rbt5Δ/csa1Δ strains. All single mutant strains presented a highly heterogeneous and rougher cell surface when compared with the control CAI4-URA3 strain. These alterations of the cell surface were dramatically apparent in the case of the double and triple mutant strains.

Download figure to PowerPoint

Biofilm structure analysis by CSLM

The cell surface plays a key role in the initial interaction of C. albicans cells with an inert or a biological substrate in the process of biofilm formation. We have reported previously that C. albicans mutants with single, double and triple deletions for genes encoding proteins species bearing the CFEM domain (Pga10p, Rbt5p and Csa1p) exhibited an abnormal ability to form biofilms (Pérez et al., 2006), and in this communication, we reported that the surface of mutant strains displayed severe morphological alterations. Consequently, we considered it of interest to further inspect the biofilm three-dimensional structure and development by means of CSLM. This technique was preferred to scanning electron microscopy because it is a nondestructive technique that maintains the biofilm under native conditions with minimal structural alterations, allowing for the in situ visualization of the mature biofilms.

For this purpose, a combination of FUN-1, a fluorescent dye taken up by fungal cells that, in the presence of metabolic viability, is converted from a diffuse yellow cytoplasmic stain to red, rod-like collections, and ConA conjugated with the dye Alexa Fluor 488, which binds specifically to polysaccharides including α-mannopyranosyl and á-glucopyranosyl residues and yields green fluorescence (Kuhn et al., 2002), was used.

Examination by CSLM of mature (24 h) biofilms stained with FUN-1 revealed that biofilms formed by mutant strains had only a 20–30% depth compared with the biofilm belonging to the control strain CAI4-URA3. Thus, the depth of biofilms formed by the parental strain was 200 ± 20 μm, whereas in the case of the different mutants examined, biofilm depths ranged from 40 ± 11 to 60 ± 17 μm (figures were the mean values of five independent measures of biofilm depth in three different areas for each of the strains examined). These results indicated that biofilms produced by the mutants contained a lower proportion of metabolically active cells, which in turn may account for a lower thickness of mature biofilms, a contention that was additionally supported by subsequent CSLM observations of biofilms double stained with FUN-1 (red fluorescence) and ConA Alexa Fluor 488 (green fluorescence). Thus, a comparison of the relative fluorescence levels due to both fluorophores revealed that biofilms formed by mutant strains contained not only a lower proportion of metabolically active cells but also a lower quantity of exopolymeric material, compared with the parental CAI4-URA3 strain (Fig. 6). In any case, the mutant strains retained the ability to form hyphae in the biofilms, because similar relative proportions of mycelial forms were consistently observed (when compared with the parental CAI4-URA3 strain).

image

Figure 6.  Relative fluorescence associated with biofilms formed by strains CAI4-URA3 (a), pga10Δ (b), rbt5Δ (c), csa1Δ (d), pga10Δ/rbt5Δ (e) y pga10Δ/rbt5Δ/csa1Δ (f) on Thermanox coverslips, double stained with FUN-1® (red fluorescence) and Con-A Alexa fluor 488 (green fluorescence). Results from CSLM visualization are shown as a three-dimensional reconstruction plot representing the intensity of fluorescence for each fluorophore. Colour images were produced by pasting each of the greyscale images (the scale values for apparent brightness in x, y and z axes range from 0 to 250) from the confocal microscope using the built-in software of the equipment.

Download figure to PowerPoint

Virulence assessment

The abilities to grow as mycelia filaments and to form biofilms are believed to be important virulence traits in C. albicans. Besides, hyphae are the most abundant cellular elements present in candidal biofilms. We have found that proteins belonging to the CFEM family are overexpressed under hyphae-inducing conditions and mutants for genes coding for such proteins are defective in biofilm formation. Consequently, a possible role for CFEM proteins in C. albicans virulence could be expected. In order to assess such a contention, virulence studies were performed in animal models. Female BALB/c mice were infected with 200 μL of cell suspensions containing 5.0 × 105 cells mL−1 (the total infecting dosage administered was 105 cells). The number and viability of cells present in the infecting inocula were further assessed by a plate count in YPD agar. In all cases, the plate count yielded lower values [CAI4-URA3 (3.8 × 105 cells mL−1); pga10Δ (3.3 × 105 cells mL−1); pga10Δ/rbt5Δ (4.1 × 105 cells mL−1); and pga10Δ/rbt5Δ/csa1Δ (2.4 × 105 cells mL−1)] when compared with the results obtained by direct counting with the haemocytometer chamber. This lack of correlation seemed to be most likely due to the clumping tendency of C. albicans cells (clumps are extremely difficult to disintegrate when cell suspensions are mixed with the melted YPD agar before pouring into the Petri dishes), which was found to be particularly high in the case of the triple homozygous mutant (pga10Δ/rbt5Δ/csa1Δ) (Pérez et al., 2006), rather than to a reduction in the number of viable cells since cultures were harvested before they reach the late exponential growth phase (see Materials and methods), a phase in which most, if not all, cells are viable and metabolically active.

Figure 7 shows that the virulence of pga10Δ and pga10Δ/rbt5Δ mutants was similar to that displayed by the control CAI4-URA3 strain. In the case of the triple mutant pga10Δ/rbt5Δ/csa1Δ strain, the lack of virulence could be a consequence of the wrong localization of the URA3 gene. The right orientation and location of the gene URA3 has been shown to affect C. albicans virulence (Cheng et al., 2003; Brand et al., 2004). Under our experimental conditions, we have found that the effect of URA3 seems to be variable. Thus, in the case of the pga10Δ mutant, the lack of URA3 resulted in loss of virulence (CAI4-URA3 vs. pga10Δ-URA3; results not shown), whereas reintegration of URA3 restores its virulent nature (Fig. 7). However, we found that a csa1Δ null mutant strain in which the URA3 was not reintegrated exhibited a virulence very similar to the parental CAI4-URA3 strain (P=not significant for CAI4-URA3 vs. csa1Δ-URA3; results not shown). In any case, we replaced the URA3 gene in the RPS10 locus using the integration vector CIp10 (Murad et al., 2000) in both the single (pga10Δ) and the double (pga10Δ/rbt5Δ) homozygous mutant strains before testing virulence. However, several attempts to achieve the reintegration of the URA3 gene in its right locus in the triple mutant were unsuccessful and, therefore, virulence assays ought to be carried out with the original strain (Weissman & Kornitzer, 2004). Consequently, it seems that further investigation is required to elucidate not only the molecular basis of the avirulent nature displayed by the triple mutant but also the actual role of URA3 in C. albicans virulence.

image

Figure 7.  Survival/death graph of mice infected in vitro with the control strain CAI4-URA3 and with the single (pga10Δ), double (pga10Δ/rbt5Δ) and triple (pga10Δ/rbt5Δ/csa1Δ) null mutants. Statistically significant differences (P-values) were found to be as follows: CAI4-URA3 vs. pga10Δ, P=0.0018; CAI4-URA3 vs. pga10Δ/rbt5Δ, P=0.0070; CAI4-URA3 vs. pga10Δ/rbt5Δ/csa1Δ, P=0.0008.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Most of pathogenic fungal species known carry a large number of species belonging to the family of the CFEM domain (Kulkarni et al., 2003), which suggests a potential role of these proteins in pathogenesis. Six CFEM-containing proteins that are the products of RBT5, CSA1 (also designated as WAP1), CSA2, PGA7, SSR1 and PGA10 genes have been currently identified in C. albicans (Braun et al., 2000; Lamarre et al., 2000; de Groot et al., 2003; Weissman & Kornitzer, 2004; Garceráet al., 2005; Castillo et al., 2008; Weissman et al., 2008).

We have previously examined some biological features of strains carrying single, double and triple deletions for PGA10, RBT5 and CSA1. The single null mutant for the PGA10 gene was generated in our laboratory (Pérez et al., 2006). The rest of the mutants examined were kindly provided by other groups (Braun et al., 2000; Weissman & Kornitzer, 2004). All these mutants formed fragile biofilms in vitro, with a low adherence to the substrate (Pérez et al., 2006). This suggests that CFEM-containing proteins are involved in the biogenesis of biofilms in C. albicans. These results are in agreement with reports from other laboratories indicating that several GPI–CWP species including Als1p, Als3p, Hwp1p and Eap1p appear to be required for biofilm formation in C. albicans (Firon et al., 2007; Hiller et al., 2007; Li et al., 2007; Nobile et al., 2008). To some extent, biofilms formed by strains bearing deletions in genes coding for members of the CFEM family phenocopy those formed by als1Δ/als3Δ and hwp1Δ mutants (Nobile et al., 2008).

In this work, we have further examined biofilms formed by mutant strains for the CFEM domain by means of CSLM using a combination of the fluorescent dyes FUN-1 and ConA-Alexa Fluor 488 Conjugate (see Materials and methods), and found that such biofilms displayed a thin three-dimensional structure with 40–60 μm depth, which represented only 20–30% thickness of the biofilm formed by the control strain CAI4-URA3, with lower quantities of the EM component and also a decrease in the quantity of metabolically active cells. These observations strongly suggest that the deletion of genes coding proteins belonging to the CFEM family is associated with a defect in the general structure of biofilms. The biofilm-deficient phenotype could be a nonspecific effect of the absence of one or more CFEM proteins or, alternatively, due to the fact that these species are putative cell surface-bound components involved in the interaction of fungal cells with the environment. In any case, the abnormal biofilm-forming abilities observed do not appear to be related to defects in filamentation because the mutant strains retained unaltered their ability to form hyphae and similar relative proportions of mycelial forms were consistently observed in all biofilms examined in this work. In this context, it has been clearly established that hyphae are essential elements for providing the structural integrity and multilayered architecture characteristic of mature/fully developed biofilms (Baillie & Douglas, 1999), because C. albicans strains with mutations in genes governing morphogenesis and that are defective in filamentation also display defects in their biofilm-forming abilities (Ramage et al., 2002; García-Sánchez et al., 2004; Kelly et al., 2004; Krueger et al., 2004).

On the other hand, it has been suggested that CFEM proteins may have a function as adhesion molecules in host–pathogen interactions (Kulkarni et al., 2003). Therefore, strains carrying deletions in any of the genes coding for these proteins would be expected to display altered cell surface properties (i.e. CSH and/or adhesiveness to inert or biological surfaces and substrates) related to interactions of fungal cells with the external environment. Thus, mutant strains exhibited higher CSH levels and increased binding ability to all different substrates tested. Because hydrophobic C. albicans cells are more adherent than their hydrophilic counterparts to a variety of proteins from host tissues (Masuoka et al., 1999; de Repentigny et al., 2000), the observations reported here suggest that the loss of functionality of genes coding for CFEM domain-containing proteins may result in a cascade of pleiotropic effects including (1) overexpression of genes coding for other proteins having adhesive properties or that confer CSH to compensate the absence of CFEM species or (2) structural cell surface alterations [CFEM mutants displayed higher sensitivity to cell wall-disturbing agents, which suggests possible alterations in the fungal cell wall structure and/or function of these mutants (Pérez et al., 2006)]. AFM observations reported in this work strongly support this latter contention because the cell wall surface of mutant strains was found to display a very rough and disrupted appearance when compared with the parental strain. Although only yeast cells were examined by AFM, one can speculate on the possibility that the phenotype observed under AFM should be much more pronounced in hyphal cells because, as above stated, some genes of the CFEM family are overexpressed under hyphae-inducing conditions (Braun et al., 2000; Lamarre et al., 2000; Sosinska et al., 2011) and, consequently, knockout of such genes could be associated with stronger cell surface alterations, taking into account that the cell wall in hyphal filaments appears to be thinner than that present in yeast cells (Rico et al., 1991).

Another interesting observation reported here is that the PGA10 gene was found to be upregulated under hyphae-inducing conditions, which has also been reported for other genes (RBT5 and CSA1) of the CFEM family (Braun et al., 2000; Lamarre et al., 2000). Because the ability to grow as mycelia filaments is believed to be an important virulence trait in C. albicans and hyphae are the most abundant cellular elements present in candidal biofilms, proteins belonging to the CFEM family may be expected to play an important role in the virulence of this fungal species. However, we found that strains bearing mutations in genes coding for proteins belonging to the CFEM family were not defective in animal models, thus suggesting that individually, neither of these genes were important to successfully infect the host. Moreover, mutant strains exhibited a pattern of susceptibility/resistance to antifungals similar to that displayed by the parental strain (data not shown).

Other roles, for example haem–iron utilization and haemin-binding capacity, have been suggested for PGA10 and RBT5 (Weissman & Kornitzer, 2004; Weissman et al., 2008). Although none of the other members of the CFEM family in C. albicans or in other fungal species have been reported to share this function, the possibility that the different phenotypes displayed by the CFEM mutant strains examined in this and previous work from our group (Pérez et al., 2006) could be due to a defective iron uptake cannot be completely dismissed. Hence, the question is whether Pga10p and Rbt5p are the first proteins belonging to the CFEM family to be assigned to a different specific function or, in fact, they act as multifunctional proteins (Nombela et al., 2006) that can display several functions depending on their localization in the cell. Although CFEM-containing proteins are believed to play a role in pathogenesis, acting as cell-surface receptors or signal transducers, as adhesion molecules in host–pathogen interactions in C. albicans or in biofilm formation, further work is necessary to fully elucidate all possible aspects of the biological and functional role (including their potential as targets for new therapeutic approaches for Candida infections) of this intriguing family of proteins in C. albicans.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

This work was supported by grants BFU2005-02572, Ministerio de Educación y Ciencia, Spain, and ACOMP06/103, Generalitat Valenciana, Valencia, Spain (to J.P.M.). A.P. was the recipient of a predoctoral grant from Ministerio de Educación y Ciencia, Spain. We acknowledge D. Kornitzer (Haifa, Israel) and A.D. Johnson (San Francisco, CA) for the kind gift of the single, double and triple C. albicans mutant strains for the RBT51, RBT5 and CSA1 genes. We thank John Graham and Mahesh Uttamlal (Glasgow, UK) for their advice with the AFM experiments, and Anna Lazzell (San Antonio, TX) for her assistance with the virulence assays in mice.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References