|Home | About | Journals | Submit | Contact Us | Français|
Monoclonal antibody 3D9.3 (MAb 3D9.3) reacts with the surface of Candida albicans germ tubes and recognizes a protein epitope. We used a two-step chromatography procedure to purify and identify the antigen (3D9) from C. albicans strain 66396 germ tubes. MAb 3D9.3 recognized two intense protein bands at 140 and 180 kDa. A comparative analysis between theoretical and experimental mass spectrum peaks showed that both bands corresponded to Als3. This conclusion was supported by lack of reactivity between MAb 3D9.3 and an als3Δ/als3Δ mutant strain, and the fact that an immunoglobulin preparation enriched for Als3 specificity recognized the purified 3D9 antigen. PCR demonstrated that C. albicans strain 66396 has two different-sized ALS3 alleles that correspond to the two purified protein bands. Strain- and species-specificity of the 3D9 epitope were studied with various C. albicans strains and Candida species, such as closely related C. dubliniensis. The 3D9 epitope was detected only in C. albicans, demonstrating the utility of MAb 3D9.3 for differentiation between C. albicans and C. dubliniensis. Adhesion assays demonstrated that MAb 3D9.3 blocks adhesion of C. albicans germ tubes to human buccal epithelial cells and vascular endothelial cells.
The opportunistic fungal pathogen, Candida albicans, is a commensal of the oral cavity and digestive tract that can cause severe disease, especially in immunocompromised patients. A hypothetical set of virulence-associated genes has been proposed and supported by various studies. These fungal attributes include the production of secreted hydrolytic enzymes, morphological and functional modifications resulting from transition between yeast and hypha forms, adhesion to inert and biological substrates, immunomodulation of host defense mechanisms, and antigenic variability (Calderone & Fonzi, 2001).
Antigenic changes in cell wall mannoproteins resulting from the transition between yeast and hypha growth forms, have been investigated with polyclonal and monoclonal antibodies (MAbs). These antibodies were most frequently directed toward carbohydrates carried by cell wall proteins. Among these antibodies, three MAbs have been demonstrated by indirect immunofluorescence assay (IFA) to only recognize C. albicans germ tubes and hyphae (Ollert & Calderone; 1990, Marot-Leblond et al., 1993; Ponton, et al., 1993; Marot-Leblond et al., 1995; Marot-Leblond et al., 2000). Molecular genetic approaches have identified genes such as ALS3, HWP1 and HYR1 that encode hypha-specific surface proteins that may contribute to differences in cell wall structure and functions (Bailey et al., 1996; Hoyer et al., 1998; Staab & Sundstrom, 1998).
ALS3 transcription is associated with germ tubes and hyphae (Hoyer et al., 1998). ALS3 is part of the ALS (agglutinin-like sequence) gene family which encodes cell-surface glycoproteins of C. albicans, some of which function in adhesion to host surfaces (reviewed in Hoyer et al., 2008). The eight ALS genes share a common three-domain structure. The 5’ domain is relatively identical across the family. The central domain of each ALS gene is composed entirely of tandemly repeated copies of a 108-bp motif. The 3’ domain is relatively variable in length and sequence, but in all genes, encodes a Ser/Thr-rich sequence that, in the mature Als protein, is heavily glycosylated. Although allelic sequence variability has been observed in each region of the ALS genes (reviewed in Hoyer et al., 2008), when considering the entire ALS family, the central tandem repeat domain contributes to the greatest differences between ALS alleles as a result of variability in the number of copies of the 108-bp sequence. For example, the two ALS3 alleles often differ in the number of copies of the tandem repeat sequence within the central domain of the coding region (Oh et al., 2005). In strain SC5314, 9 tandem repeat copies are present in the smaller allele and 12 copies in the larger allele. Disruption of ALS3 results in cells with reduced adherence to vascular endothelial and buccal epithelial cells (Zhao et al., 2004; Phan et al., 2007) and defects in C. albicans biofilm formation (Nobile et al., 2006; Zhao et al., 2006).
In previous work (Marot-Leblond et al., 1993; Marot-Leblond et al., 1995), a C. albicans germ-tube-specific antigen (3D9 antigen) was identified by the use of a MAb (MAb 3D9.3). By indirect immunofluorescence, MAb 3D9.3 was specific for the surface of C. albicans germ tubes and hyphae and was not able to label the cell wall of any other Candida species (Marot-Leblond et al., 1993). These morphological shape and species specificities were confirmed by immunoblotting regardless of the extraction procedure used (Marot-Leblond et al., 1995). After a two-step purification by chromatography, antigen was recovered but not identified. The relative molecular mass ranged from 120 to 220 kDa and the epitope was destroyed by proteinase digestion. In this paper we describe the identification of 3D9 antigen as Als3 and we present evidence that this protein carries a germ-tube-specific epitope of C. albicans.
Fungal strains used in this study are shown in Table 1. Yeast strains were grown on Sabouraud dextrose agar (SDA) slants (Merck, Darmstadt, Germany) at 37°C for 48 h. To prepare germ tubes or hyphae, yeast forms were inoculated into Medium 199, pH 6.7 (Gibco Laboratories, Grand Island, NY) at 37°C and 200 rpm shaking for 3 h or 24 h, respectively. Cultures were washed with distilled water, harvested by centrifugation (2000 × g, 10 min) and used immediately or stored at −20°C. For antigen extraction, cells were freeze dried.
Cell components were extracted by enzymatic digestion. Briefly, 320 mg of freeze-dried yeasts, germ tubes or mycelium were digested with 10 ml of Zymolyase 20T™ 2 mg per ml (Arthrobacter luteus; Seikagaku, Kogyu Co., Tokyo, Japan) containing 1 mM phenylmethylsulfonyl fluoride for 1 h and 30 min at 37°C with shaking at 1,500 rpm on a titer plate shaker (Heidolph; Bioblock Scientific, Strasbourg, France). Solubilized antigenic components were recovered by centrifugation at 10,000 × g for 10 min and stored at −20°C.
A two-step procedure was used to obtain a purified preparation of the 3D9 antigen and was performed as previously described (Marot-Leblond et al., 1995). Briefly, antigenic extract from C. albicans ATCC 66396 was applied to a Superdex™ 200 HR 10/30 column (Amersham Pharmacia Biotech, Uppsala, Sweden) previously equilibrated with HCl-glycine 0.15 M pH 3 buffer. Each recovered fraction was assayed for 3D9 antigenic activity by ELISA. Reactive fractions containing 3D9 antigen were pooled.
The pre-purified extract from the gel filtration column was solubilized in 2 M ammonium sulfate by slowly adding phosphate (100 mM)-ammonium sulfate (4 M) buffer, pH 7.2. Insoluble components were removed by centrifugation at 10,000 × g for 10 min. The supernatant was then applied to a Phenyl Sepharose® 6 Fast Flow (Low sub) column (Amersham Pharmacia Biotech, Uppsala, Sweden), equilibrated with phosphate (50 mM)-ammonium sulfate (2 M) buffer. The column was washed until no absorbance at 280 nm was detected in the effluent. Elution was carried out at a flow rate of 1 ml min−1 by a stepwise decrease in the concentration of ammonium sulfate while maintaining the concentration of phosphate at 50 mM throughout, until an ammonium sulfate concentration of 0.1 M was achieved. The remaining unwanted material was eluted using 50 mM phosphate buffer, distilled water and ethanol. Fractions of 5 ml were collected and checked for 3D9 antigenic activity by ELISA. Positive fractions were pooled and dialysed to obtain 3D9 purified antigen.
The method for preparation of serum Ig enriched for Als3 specificity was described previously (Zhao et al., 2006). Briefly, a polyclonal antiserum raised against the purified Als5 N-terminal domain was treated to remove lipoproteins, and the gamma-globulin fraction was precipitated with ammonium sulfate. After resuspension in, and exhaustive dialysis against, Dulbecco’s phosphate buffered saline (DPBS), the Ig preparation was enriched for reactivity against Als3 by absorption against C. albicans strain 1843 (als3Δ/als3Δ; Table 1) in yeast and germ tube forms. Absorption was continued until the resulting Ig preparation did not recognize strain 1843 by indirect immunofluorescence.
ELISA was performed by coating the wells of microtitration plates for 90 min at 37°C with antigenic extracts or chromatographic fractions that were diluted in carbonate-bicarbonate 0.05 M buffer pH 9.6 and then blocked overnight at 4°C with PBS containing 10% (wt/vol) non fat dry milk (Régilait, Saint-Martin-Belle-Roche, France). After rinsing with 0.05% (vol/vol) Tween 20 in PBS (PBST), purified MAb 3D9.3 or rabbit anti-Als3 Ig was added and incubated for 1 h at 37°C and then washed again with PBST. Goat anti-mouse immunoglobulin M (IgM) (μ specific) or goat anti-rabbit IgG (H+L) coupled to horseradish peroxidase (Caltag Laboratories, San Francisco, CA), diluted in PBST containing 1% (wt/vol) non-fat dry milk was added and incubated for 1 h at 37°C. For color development, the OPD peroxidase substrate (Sigma, St. Louis, MO) was added, developed for 20 min at room temperature, and read at a wavelength of 490 nm on an automated microplate reader ELX800UV (Bio-tek Instruments, Winooski, VT).
Electrophoresis was carried out as described previously on a gel electrophoresis apparatus (Hoefer SE 600 Ruby, Amersham Pharmacia Biotech, Uppsala, Sweden) on isotropic 5 to 15% (wt/vol) acrylamide slab gels by using the discontinuous buffer system of Laemmli (Laemmli, 1970). All electrophoresis was performed using constant amounts of proteins. Subsequently, gels were stained for proteins by either Coomassie Brillant R250 (0.01% wt/vol) in methanol-acetic acid or MS-compatible silver staining, or were electrophoretically transferred (Towbin et al., 1979) to Hybond-P polyvinylidene difluoride sheet (Amersham Pharmacia Biotech, Buckinghamshire, UK) at 200 mA for 1 h and 15 min. The membranes were blocked overnight with 10% (wt/vol) non-fat dry milk in PBS at 4°C and probed with MAb 3D9.3 or anti-Als3 Ig and goat anti-mouse IgM (μ specific) or goat anti-rabbit IgG (γ specific) respectively coupled to horseradish peroxidase (Caltag Laboratories, San Francisco, CA). Bound antibodies were revealed by submersing the sheets in 0.1 M Tris buffer (pH 7.6) containing 0.5 mg 3-3’diaminobenzidine (Sigma-Aldrich, St. Louis, MO) and 0.1% (vol/vol) hydrogen peroxide. The color reaction was arrested by rinsing in 5% acetic acid (vol/vol). In some experiments, the blots were developed using enhanced chemiluminescence (ECL) Advance™ Western Blotting Detection Kit (Amersham Biosciences, Buckinghamshire, UK).
Indirect IFA was performed as previously described by Marot et al. (2000). Briefly, the assay was carried out on yeast, germ tubes or hyphae. Cells were mounted on a glass microscope slide (PolyLabo Paul Block, Strasbourg, France) and dried at 4°C overnight. After rinsing with PBS, MAb 3D9.3 was transferred to slides and incubated for 1 h at 37°C. The slides were then washed in PBS and incubated with fluorescein isothiocyanate-conjugated goat anti-mouse IgG (H+L) (Caltag Laboratories, San Francisco, CA) diluted 1:100 in PBS. The slides were incubated at 37°C for 1 h and then washed. The slides were examined with a Nikon microscope equipped with reflected light fluorescence.
Selected spots were manually excised from preparative 1-dimensional electrophoresis gels and were in-gel reduced, alkylated, and digested with trypsin. After digestion, the supernatant was collected and 0.5 µL of each peptide mixture was crystallized with 0.5 µL of freshly prepared α-cyano-4-hydroxycinnamic acid in 50% acetonitrile and 0.1% TFA onto a MALDI target plate. Mass spectra were obtained on a Voyager-DE PRO MALDI-TOF mass spectrometer (Perseptive Biosystems, Inc., Framingham, MA), equipped with a pulsed nitrogen laser (337 nm, 3 ns pulse) and a delayed extraction, and operated in positive ion reflector mode. All mass spectra were calibrated externally using close standard peptide mixtures. The monoisotopic peptide mass fingerprinting (PMF) data obtained from MALDI-TOF were used to search for protein candidates in the complete C. albicans genomic database, CandidaDB (www.genolist.Pasteur.fr/CandidaDB), using Mascot software program (www.matrixscience.com).
The size of the tandem repeat domain in each ALS3 allele of C. albicans 66396 was determined by PCR. Cells of different C. albicans strains were grown for 18 h in 5 ml of YPD (per liter: 10 g yeast extract, 20 g peptone, 20 g glucose) and recovered by centrifugation. Pellets were resuspended in 400 µl of lysing buffer (2% Triton X-100, 1% SDS, 100 mM NaCl, 10 mM Tris buffer, pH 8.0, 1 mM EDTA), and 400 µl of phenol/chloroform (1:1), and 0.2 g of glass beads were added. This mixture was vortexed. After centrifugation, the aqueous phase was transferred to a new tube, and 1 ml of ethanol was added to precipitate the DNA. After centrifugation, the pellet was resuspended in 200 µl of TE buffer (10 mM Tris buffer, 1 mM EDTA, pH 8.0) and treated with 10 µg ml−1 of RNase A. DNA was precipitated with 20 µl of ammonium acetate 7.5 M buffer and 400 µl of ethanol. After centrifugation, the DNA was resuspended in 100 µl of TE buffer.
The primer pair ALS3GenoF (5’-ACC TTA CCA TTC GAT CCT AAC C-3’) and ALS3GenoR (5’-GAT GGG GAT TGT GAA GTG G-3’) was used as described by Oh et al. (2005). Amplification reactions were performed in 30 µl aliquots of a solution containing 3 µl of 10x polymerase buffer, 3 mM MgCl2, 0.1 mM each dNTP, 0.4 µM each primer, 1.5 U of SurePrime™ DNA polymerase (Q-biogene, Irvine, CA) and 1.5 µl extracted DNA. Reactions were heated for 15 min at 94°C followed by 35 cycles of 94 °C (1 min), 57 °C (1 min), 72 °C (3 min). A final 72 °C (7 min) extension completed the reaction. PCR products were separated on 0.8 % agarose-Tris/borate/EDTA gels stained with ethidium bromide. Amplicon sizes were determined by comparison with Felix™ marker (Q-biogene, Irvine, CA) and with ALS3 amplification products from genomic DNA of C. albicans SC5314.
The HUVEC adhesion assay was modified from the assay described previously (Zhao et al., 2004). IgM concentration for purified preparations of MAb 3D9.3 and a control IgM (BioLegend catalog #401602) was measured using a quantitative mouse IgM ELISA kit (ZeptoMetrix, catalog # 0801181). C. albicans cells of strain CAI12 (Porta et al., 1999) were grown in YPD using the published method and induced to form germ tubes by inoculating 7 × 103 cells into 7 ml of pre-warmed RPMI 1640 in a 50 ml Falcon tube. Following 30 min incubation at 37°C and 5% CO2, 10 µg/ml of the MAb 3D9.3 or the control IgM was added to the Falcon tube and the tube moved to a 37°C shaker bath (200 rpm) for an additional 30 min. Another control tube, to which only DPBS was added (no IgM), was also incubated. Microscopic examination showed that MAb 3D9.3 did not cause aggregation of germ tubes. The HUVEC monolayers in a 6-well plate were rinsed twice with pre-warmed (at 37°C) RPMI 1640. Then, 1 ml RPMI 1640, containing 103 C. albicans germ tubes was added into each of the 6 replicate wells. The plate was incubated for 30 min in the 37°C CO2 incubator to allow C. albicans adhesion to the HUVEC monolayer. Non-adherent cells were removed by washing as previously described (Zhao et al., 2004). Viability of C. albicans was verified by plating 100 µl of C. albicans inoculum on YPD plates in triplicate. Both the 6-well and C. albicans viability plates were incubated overnight at 37°C and CFU counted. The percentage adherence for each antibody or control treatment was calculated as (mean adherent CFU/mean total CFU)×100. Adhesion assays were conducted in six wells per day on four different days. Results were evaluated statistically using a mixed model analysis of variance (PROC MIXED in SAS). Separation of means was performed using the LSMEANS option.
BECs were collected from healthy humans and used in a C. albicans adhesion assay similar to that described previously (Zhao et al., 2004). Modifications are noted here. A total of 2 × 106 YPD-grown, DPBS-washed yeast cells were inoculated into 4 ml RPMI 1640 in a 25 ml Erlenmeyer flask. Cultures were incubated at 37°C and 200 rpm shaking to induce germ tube formation. MAb 3D9.3 or the control IgM (BioLegend catalog #401602) was then added to the culture at a final concentration of 10 µg/ml. A third flask had only DPBS added. Microscopic examination showed that MAb 3D9.3 did not cause aggregation of germ tubes. Flasks were incubated for 30 min and then 2 × 104 BEC added to each flask. Following incubation for 30 min at 37°C and 200 rpm shaking, BEC were collected onto 12 µm pore size filters, washed, transferred to glass microscope slides, and stained as described previously (Zhao et al., 2004). The total number of germ tubes adherent to 50 BECs was counted and expressed as the mean number of germ tubes per BEC. The assay was conducted on three separate days. A mixed model analysis of variance was used to study the differences between treatments. Data were analyzed using PROC MIXED in SAS. Separation of means was performed with the LSMEANS option.
The 3D9 antigen was purified from Zymolyase™ extracts of C. albicans 66396 germ tubes using a column chromatography procedure that involved fractionation by gel filtration and hydrophobic interaction chromatography. At each step, 3D9-positive fractions were identified using SDS-PAGE with silver staining, Concanavalin A staining, and immunodetection with MAb 3D9.3. Silver staining of germ tube extract exhibited many molecular species ranging from 10 to 400 kDa (Fig. 1, lane A1). Concanavalin A staining and immunodetection using MAb 3D9.3, yielded a polydisperse staining pattern with an apparent molecular mass of 120 to 220 kDa (Fig. 1, lanes A2 and A3). Analysis of prepurified 3D9 antigen after gel filtration chromatography showed the same results compared with germ tube extract analysis, although the disperse staining pattern ranging from 120 to 220 kDa was more intense (Fig. 1, lanes B1, B2 and B3). Following the last fractionation by hydrophobic interaction chromatography, fractions of 1-0.5 M ammonium sulfate were pooled, dialyzed and further analyzed. This method yielded purified 3D9 antigen (Fig. 1, lane C1). Two diffuse bands centered at 140 and 180 kDa were visible after Concanavalin A staining and immunodetection using MAb 3D9.3 (Fig. 1, lanes C2 and C3).
The immunoreactive proteins were excised from a Coomassie blue or silver-stained one-dimensional electrophoresis gel and subjected to in-gel trypsin digestion. Peptide mass fingerprinting (PMF) of the two bands (140 and 180 kDa) was compared with the recently public C. albicans genomic database of strain SC5314 (CandidaDB), but did not allow significant identification of a single protein, although Als3 protein was one of the first candidates listed. Notably, though, 11 peptides from PMF of the 180 kDa band and 9 peptides from PMF of the 140 kDa band matched with peptides from theoretical PMF of Als3, a protein associated with C. albicans germ tubes and hyphae (Hoyer et al., 1998). The presence of two protein bands on SDS-PAGE (Fig. 1) is consistent with the finding that there are two different-sized ALS3 alleles many C. albicans isolates (Oh et al., 2005). These results suggested that Als3 is the 3D9 antigen.
Several experimental methods were used to verify our hypothesis that 3D9 antigen corresponded to Als3. The surface expression of 3D9 antigen on yeast cells and germ tubes was investigated by indirect IFA. Incubation with MAb 3D9.3 revealed that 3D9 antigen was not detectable on C. albicans yeast forms (Fig. 2). When filamentation occurred, the parent yeast cells were still not stained but the germ tubes exhibited a strong and homogeneous surface fluorescence (Fig. 2, a, e and f). Germ tubes of the C. albicans als3Δ/als3Δ mutant showed no fluorescence (Fig. 2, c and d). Reintegration of a wild-type copy of ALS3 into the als3Δ/als3Δ mutant restored the wild-type staining pattern (data not shown).
Presence of the 3D9 antigen on various C. albicans strains was studied by ELISA using a constant amount of crude extract protein from yeasts, germ tubes and hyphae (Table 2) and by Western blotting (Fig. 3a). No reaction with MAb 3D9.3 was detected for the C. albicans als3Δ/als3Δ mutant using either method, while SC5314, CAI12 and als1Δ/als1Δ mutant strains gave a positive signal only with extracts from germ tubes and hyphae. Immunodetection of the germ tube crude extracts of SC5314, CAI12 and als1Δ/als1Δ mutant strains using MAb 3D9.3 yielded an intensive polydisperse staining pattern with an apparent molecular mass of 120 to 220 kDa (Fig. 3a). Finally, an anti-Als3 Ig preparation was used to immunoblot purified 3D9 antigen (Fig. 3b, lane 3). This method revealed a polydisperse signal with a molecular mass ranging from 120 to 220 kDa. Together, these methods support the conclusion that antigen 3D9 is Als3.
The utility of hydrophobic interaction chromatography for purification of 3D9 antigen, particularly elution of 3D9 antigen using a 1 M to 0.5 M ammonium sulfate gradient, suggests that Als3 is quite hydrophobic. Hydrophobic properties for Als proteins were noted previously, when eluting an Als5 N-terminal domain fragment from phenyl sepharose (Hoyer & Hecht, 2001). The Als5 fragment bound to phenyl sepharose at 1 M ammonium sulfate and did not elute until ammonium sulfate was absent from the buffer. Previous work to identify the 3D9 antigen suggested that it was destroyed by protease treatment. We found that MAb 3D9.3 recognizes the soluble N-terminal domain of Als3 on a Western blot, suggesting that the 3D9 epitope is localized within this portion of Als3 (data not shown). Together, the observations suggest a hydrophobic character for the N-terminal domain of Als3.
C. albicans strain SC5314 has two different-sized ALS3 alleles: one encodes 9 copies of the tandemly repeated sequence in the central domain and the other encodes 12 copies (Oh et al., 2005). The nature of the ALS3 alleles in strain 66396 was examined by PCR using primers that amplify the tandem repeat domain (Fig. 4a). This analysis demonstrated t hat the smaller allele in both SC5314 and 66396 has 9 tandem repeat copies. The larger allele in strain 66396 has 15 copies of the tandemly repeated sequence suggesting that this allele should encode a protein of greater size than produced by the larger ALS3 allele in strain SC5314. This Als3 size relationship was demonstrated by immunoblotting of germ tube Zymolyase extracts with MAb 3D9.3 (Fig. 4b). Extracts from strain SC5314 showed a polydisperse staining pattern with an apparent molecular mass ranging from 120 to 170 kDa that had intense bands at 140 and 165 kDa (Fig. 4b, lane 2). Extracts from strain 1702, an SC5314 derivative in which the larger ALS3 allele was deleted, showed a band only at 140 kDa (Fig. 4b, lane 3). Similarly, strain 1704, an SC5314 derivative strain in which the smaller ALS3 allele was deleted, showed only the larger protein band at 165 kDa (Fig. 4b, lane 4). This protein band is smaller in size, consistent with the presence of fewer tandem repeat copies than found in the larger allele of strain 66396 (Fig. 4b, lane 1).
Although Als3 size estimates are consistent with predictions made from the varying numbers of tandem repeat sequence copies, these observations differ from data of Kapteyn et al. (2000) who showed a molecular mass of 440 kDa for Als3. Several factors may explain these differences. First, the difference in mass may be due to the extraction method used to release Als3 from the C. albicans cell wall. Zymolyase™-20T, a β-1,3 glucanase, was used in this study, while Kapteyn et al. (2000) used β-1,6 glucanase. According to the manufacturer’s specifications, Zymolyase™-20T contains mannanase activity that may partially deglycosylate the Als3 protein and explain the observed size differences between the two reports. Size differences could also be due to proteolytic activity associated with Zymolyase™-20T, suggesting that an Als3 fragment was purified in this study, rather than the full-size protein. Although PMSF was added to the Zymolyase extraction, it may not counteract all of the proteolytic activity. Finally, some of the observed differences in Als3 molecular mass may be attributable to differential glycosylation of the 440 kDa molecule and the smaller proteins isolated here. Overall, the data suggest that Als3 purified in this study minimally includes the N-terminal and tandem repeat domains.
Preincubation of C. albicans germ tubes with MAb 3D9.3 significantly decreased adhesion of C. albicans germ tubes to human epithelial (Fig. 5a) and vascular endothelial cells (Fig. 5b). Statistical significance was calculated by comparison to assays where an irrelevant control IgM was used. In the BEC adhesion assay, preincubation of C. albicans germ tubes with a control IgM significantly decreased C. albicans adhesion compared to adhesion in the absence of antibody (DPBS alone; Fig. 5a). Indirect IFA using the control IgM and C. albicans germ tubes did not produce visible fluorescence of the fungal cells (data not shown), consistent with the conclusion that addition of any irrelevant protein decreases C. albicans adhesion to BECs in this assay. Similar significant reductions in C. albicans adhesion to BECs were obtained for inclusion of an irrelevant control IgG in the assay (data not shown). However, pre-incubation with MAb 3D9.3 significantly decreased adhesion compared to the control IgM (p = 0.007). Assay of C. albicans germ tube adhesion to human vascular endothelial cells showed no difference between DPBS alone and the control IgM (p = 0.671; Fig. 5b). Pre-incubation with MAb 3D9.3 significantly decreased germ tube adhesion compared to the control IgM (p = 0.015).
Indirect IFA using MAb 3D9.3 showed a strong signal along the germ tube length (Fig. 2) consistent with the results of previous work that indicated this localization for Als3 (Zhao et al., 2006). In this study, the large size of an IgM molecule (900 kDa) bound along the germ tube length would surely block adhesion, if not due to a specific anti-Als3 effect, then due to steric obstruction of other germ tube proteins. Whether adhesion inhibition is due specifically to blocking Als3 or steric effects, targeting Als3 is an effective means of interrupting interaction between host and fungal cells.
In addition to ELISA testing of Zymolyase extracts from various C. albicans isolates for reaction to MAb 3D9.3, other Candida species were studied (Table 2). Cells of the yeast and hypha forms were grown on SDA at 22°C for 48 h and in medium 199 (pH 6.7) at 37°C for 3 h and 24 h. In this latter medium, many of strains tested formed germ tubes and true hyphae. MAb 3D9.3 did not bind to extracts from any of the non-albicans Candida species tested suggesting that the 3D9 epitope is unique to C. albicans.
Previous studies examined the specificity of MAb 3D9.3 and showed that it did not recognize other Candida species besides C. albicans (Marot-Leblond et al., 1993). The previous work was completed before C. dubliniensis was recognized as a different species (Sullivan & Coleman, 1997). C. dubliniensis is the most closely related species to C. albicans and was originally isolated from patients with human immunodeficiency virus infection and recurrent oral candidiasis (Sullivan & Coleman, 1997). The ALS family is present in C. dubliniensis, but has distinct differences from the C. albicans ALS family (Hoyer et al., 2001). MAb 3D9.3 was not reactive with C. dubliniensis strains regardless of the cellular morphology tested. BLAST searches with the Als N-terminal domain reveal numerous other similar sequences within the Saccharomycotina suggesting that Als3-like proteins may exist on the surface of these fungi (Linder & Gustafsson, 2007). It is possible that different growth conditions are required for existing epitopes to be detected, or it is possible that the epitopes are not found in the other fungal species. The specificity of MAb 3D9.3 for germ tubes and hyphae of C. albicans under the conditions described here could be used diagnostically to differentiate between the two closely related species, C. albicans and C. dubliniensis. Diagnostic use of MAb 3D9.3 further increases the utility of this reagent.
Bertrand Beucher and Agnès Marot-Leblond contributed equally to this work. We thank Pr. J. M. Camadro and J. J. Montagne for processing and analyzing mass spectrometry results. This work was supported by a grant from the Communauté d’Agglomération Angers Loire Métropole and by Public Health Service grant DE14158 from the National Institute of Dental and Craniofacial Research, National Institutes of Health. A portion of this investigation was conducted in a facility constructed with support from Research Facilities Improvement Program Grant Number C06 RR16515-01 from the National Center for Research Resources, National Institutes of Health.