Abstract
Adherence to the endothelial cell lining of the vasculature is probably a critical step in the egress of Candida albicans from the intravascular compartment. To identify potential adhesins that mediate the attachment of this organism to endothelial cells, a genomic library from C. albicans was used to transform a nonadherent strain of Saccharomyces cerevisiae. The population of transformed yeasts was enriched for highly adherent clones by repeated passages over endothelial cells. One clone which exhibited a fivefold increase in endothelial cell adherence, compared with S. cerevisiae transformed with vector alone, was identified. This organism also flocculated. The candidal DNA fragment within this adherent/flocculent organism was found to contain a single 1.8-kb open reading frame, which was designated CAD1. It was found to be identical to AAF1. The predicted protein encoded by CAD1/AAF1 contained features suggestive of a regulatory factor. Consistent with this finding, immunoelectron microscopy revealed that CAD1/AAF1 localized to the cytoplasm and nucleus but not the cell wall or plasma membrane of the transformed yeasts. Because yeasts transformed with CAD1/AAF1 both flocculated and exhibited increased endothelial cell adherence, the relationship between adherence and flocculation was examined. S. cerevisiae expressing either of two flocculation phenotypes, Flo1 or NewFlo, adhered to endothelial cells as avidly as did yeasts expressing CAD1/AAF1. Inhibition studies revealed that the flocculation phenotype induced by CAD1/AAF1 was similar to Flo1. Thus, CAD1/AAF1 probably encodes a regulatory protein that stimulates endothelial cell adherence in S. cerevisiae by inducing a flocculation phenotype. Whether CAD1/AAF1 contributes to the adherence of C. albicans to endothelial cells remains to be determined.
Adherence to intravascular structures is considered to be a critical step in the egress of blood-borne Candida albicans from the intravascular compartment during hematogenous dissemination. For this reason, there have been numerous investigations into the mechanism(s) of adherence to endothelial cells (reviewed in references 12 and 24). A variety of candidal cell wall components, such as mannoproteins, glucans, lipids, and chitin, have been investigated as possible candidal adhesins. Additionally, C. albicans expresses an integrin-like molecule on its surface that is antigenically and functionally similar to the alpha chain of the human complement receptor CR3 (CD11b) (6). Other integrin-like molecules considered to be involved in the adherence of this organism to host cells and components of the basement membrane include the fibronectin, laminin, and entactin receptors (10, 16, 17).
It is highly probable that C. albicans expresses more than one adhesin. Furthermore, different adhesins probably mediate the attachment of the organism to different host cells (12). Also, blastospores may use different adhesins to bind to host cells from those used by germinated organisms (7). Although considerable effort has been expended to characterize the molecules that mediate the adherence of C. albicans to host cells, few adhesins of this medically important fungus have been characterized at the genomic level.
In this study, we used complementation cloning to identify a putative candidal adhesin. Saccharomyces cerevisiae were transformed with genomic DNA from C. albicans, and clones of S. cerevisiae that exhibited significantly enhanced adherence to endothelial cells, as well as self-aggregation, were identified. Analysis of the candidal gene expressed by one of these clones showed that it probably encodes a regulatory factor that induces an adherent phenotype.
MATERIALS AND METHODS
Organisms, plasmids, and culture conditions.
The fungal strains used in this study are listed in Table 1. C. albicans was grown in yeast nitrogen base (YNB) broth (Difco Laboratories, Detroit, Mich.) supplemented with 0.5% glucose. Prior to transformation, S. cerevisiae was grown in YPD medium (1% yeast extract, 2% peptone [Difco], 2% glucose). After transformation, the different strains of S. cerevisiae were grown in minimal medium (YNB broth without amino acids plus 2% glucose) supplemented with appropriate amino acids. All the yeasts were grown at 30°C. Escherichia coli was grown at 37°C in Luria-Bertani medium (1% tryptone, 0.5% yeast extract, 0.9% NaCl [pH 7.0]) with or without 80 μg of ampicillin per ml.
TABLE 1.
Organisms used in this study
| Organism | Description | Source (reference) |
|---|---|---|
| C. albicans ATCC 36082 | Clinical isolate | American Type Culture Collection, Rockville, Md. (8) |
| S. cerevisiae LL-20 | MATα leu2-3,112 his3-11,15 can1 | William Fonzi (22) |
| S. cerevisiae ATCC 48868 | FLO1/FLO1 trp1/+ ade1/+ (Flo1 phenotype) | American Type Culture Collection (20) |
| S. cerevisiae NCYC 1364 | NewFlo phenotype | National Collection of Yeast Cultures, Norwich, United Kingdom (31) |
| S. cerevisiae X2180A | Nonflocculent | Lee McAlister-Henn, Department of Biochemistry, University of Texas Health Science Center at San Antonio |
| S. cerevisiae YM21 | Nonflocculent | Lee McAlister-Henn |
Endothelial cells.
Endothelial cells were obtained from human umbilical cord veins by our prior modification of the method of Jaffe et al. (13). The cells were grown in M-199 medium (Gibco, Grand Island, N.Y.) containing 10% fetal bovine serum (Intergen, Purchase, N.Y.), 10% defined bovine calf serum (Hyclone, Logan, Utah), 2 mM l-glutamine, and penicillin and streptomycin as previously described (9). For use in the adherence assay (see below), third- or fourth-passage cells were grown to confluence in six-well tissue culture plates (Becton Dickinson, Franklin Lakes, N.J.) coated with gelatin.
Adherence assay.
The adherence of C. albicans and S. cerevisiae to endothelial cells was determined by our previously described method (9). Briefly, late-exponential-phase S. cerevisiae and C. albicans organisms were harvested by centrifugation, washed twice in 0.9% NaCl, and resuspended in Hanks balanced salt solution (HBSS; Irvine Scientific, Santa Ana, Calif.). After the organisms were sonicated for 3 to 5 s, the singlet blastospores were counted with a hemacytometer and adjusted to the desired concentration in HBSS. Next, either 102 C. albicans or 104 S. cerevisiae organisms suspended in HBSS were added to confluent endothelial cells in six-well tissue culture plates. The inocula were confirmed by quantitative culture on Sabouraud dextrose agar (Difco) supplemented with 25 μg of l-leucine per ml. The higher inoculum was used for experiments involving S. cerevisiae because of the lower adherence of this organism. Following incubation for 30 min at 37°C, the nonadherent organisms were aspirated and each well was rinsed twice with 10 ml of HBSS. Next, the wells were overlaid with Sabouraud dextrose agar plus leucine, and the number of adherent organisms was quantified by colony counting. Adherence was expressed as a percentage of the original inoculum. All assays were performed in triplicate and were repeated at least three times with endothelial cells from different umbilical cords.
Genomic library construction.
Total DNA was extracted from stationary-phase C. albicans ATCC 36082 by the method of Scherer and Stevens (28). The DNA was partially digested with Sau3A1 and fractionated by agarose gel electrophoresis. Fragments of 9 to 15 kb were extracted from the gel with GeneCleanII (Bio 101, Vista, Calif.). These fragments were cloned into the BamHI site of pE20-H, a shuttle vector kindly provided by Susan Sandmexer, University of California, Irvine.
The genomic library was transformed into E. coli XL1-Blue (Stratagene, San Diego, Calif.), generating 8,000 transformants. Analysis of random transformants indicated that 75% of the plasmids contained inserts of 9 to 15 kb. The library was amplified in E. coli and subsequently transformed into S. cerevisiae LL-20 with lithium acetate as described by Schiestl and Gietz (29). Approximately 24,000 transformants were generated.
Selecting for adherent clones of S. cerevisiae.
S. cerevisiae LL-20 transformed with the genomic library was grown for 2 h in 10 ml of YPD medium at 30°C, washed once, and then resuspended in HBSS. The organisms were sonication for 3 s to produce singlet yeasts. Next, 10 ml of this suspension, containing 108 organisms, was added to each of three 100-mm-diameter petri dishes containing confluent monolayers of endothelial cells. Each dish was incubated for 45 min at 37°C, after which the nonadherent organisms were aspirated and the dishes were rinsed three times with 10 ml of HBSS. The adherent organisms and endothelial cells were scraped from each dish with a cell scraper and pooled. The suspension was sonicated for 3 s to lyse the endothelial cells and then washed twice with HBSS. The yeasts were resuspended in 10 ml of YPD medium and grown on a rotating drum at 30°C for 1 h.
At the end of the incubation, the organisms were rinsed and allowed to adhere to endothelial cells in one 100-mm-diameter dish. The adherent organisms were collected and processed as above and then allowed to adhere to endothelial cells in a six-well tissue culture plate. After the nonadherent organisms were removed by rinsing, the wells of the six-well plate were overlaid with agar containing minimal medium supplemented with 100 μg of l-leucine per ml. Organisms were isolated from the individual colonies, and the endothelial cell adherence of each of these clones was quantified individually by the assay described above.
DNA manipulation.
Southern blotting, Northern blotting, restriction mapping, and subcloning were performed by standard methods (26). Sequencing was performed by the dideoxynucleotide chain termination method (27). The open reading frame that was identified was named CAD1.
Epitope tagging.
To produce a fusion protein containing the CAD1 protein plus a known epitope, CAD1 was cloned into an expression vector either upstream or downstream of a fragment of the influenza virus hemagglutinin (HA) gene. This HA gene fragment encoded the epitope YPYDVPDYA. The two yeast expression vectors, pYef1H and pYef2H, containing this HA gene fragment were kindly provided by L. Minvielle-Sebastia, Biozentrum University, Basel, Switzerland (4) (Table 1). PCR was used to add appropriate restriction enzyme sites at either end of the CAD1 open reading frame so that the gene could be inserted into the two vectors. PCR was performed with the PCR core kit (Boehringer Mannheim, Indianapolis, Ind.) as specified by the manufacturer. The primers used to add appropriate restriction enzyme sites to the ends of CAD1 for insertion into pYeF1H were 5′-ATAAGAATGCGGCCGCAATGCTGCTATCGGTACCA-3′ and 5′-CGGAATTCTTAGTAAAACTGTTTATTATAC-3′. The DNA fragment produced with these primers was ligated into the NotI and EcoRI sites of pYeF1H to generate pYCADN. The fusion protein encoded by pYCADN contained the HA epitope linked to the N terminus of the CAD1 protein. The primers used to modify CAD1 for ligation into pYef2H were 5′-CGGGATCCCATCAAATGCTGCTATC-3′ and 5′-CGGGATCCGAGTAAAACTGTTTATTATACAAC-3′. The PCR product obtained with these primers was ligated into the BamHI site of pYef2H to generate pYCADC, which encoded the HA epitope fused to the C terminus of the CAD1 protein. Immunoprecipitation of lysates of organisms labeled with Tran 35S-label (ICN, Irvine, Calif.) was used to obtain the CAD1-HA protein and to confirm that the fusion proteins actually contained both the CAD1 gene product and the HA epitope (11).
Electron microscopy.
Transmission electron microscopy was used to determine the cellular location of the epitope-tagged CAD1 protein. S. cerevisiae LL-20 transformed with pYCADN and pYeF1H were grown to log phase in minimal medium containing 2% raffinose and then incubated for 5 h in minimal medium containing 2% galactose. They were fixed first in periodate-lysine-paraformaldehyde (19) for 18 h at 4°C and then in 4% paraformaldehyde in 50 mM phosphate buffer (pH 7.2) for 2 h at 4°C. The cells were incubated in increasing concentrations of sucrose and then frozen in liquid nitrogen. Ultrathin sections (0.1 μm) were cut with a Reichert Ultracut R equipped with a Reichert FCR cryosectioning system (Leica, Vienna, Austria). The sections were placed in blocking buffer (phosphate-buffered saline containing 3% bovine serum albumin) and then incubated sequentially with anti-HA (clone 12CA5; Boehringer Mannheim) and 30-nm-diameter colloidal gold-labeled goat anti-mouse antibodies (Amersham Life Science, Arlington Heights, Ill.). Next, the samples were washed in phosphate-buffered saline containing 1% bovine serum albumin, fixed with 1% glutaraldehyde for 30 min, and stained with 1% uranyl acetate for 10 min. They were viewed with a 1200 EX electron microscope (JEOL Ltd., Tokyo, Japan) at 100 kV.
Flocculation assay.
To explore the possibility that endothelial cell adherence was related to autoaggregation and flocculation in S. cerevisiae, we defined the flocculation phenotype expressed by the transformed organisms. These organisms were grown in YNB broth at 30°C to a density of approximately 107 cells per ml. The cells were collected by centrifugation and washed twice with 10 mM Tris (pH 8.0) containing 2 mM EDTA. The washed cells were suspended in the same buffer, with or without potential flocculation inhibitors, to a density of approximately 7.5 × 107 cells per ml. The inhibitors used included glucose, galactose, maltose, mannose, NaCl, and proteinase K (31). Flocculation was initiated by the addition of 5 mM CaCl2. The critical cell density required for flocculation was determined spectrophotometrically as described by Miki et al. (20, 21).
Data analysis.
Statistical analyses were performed by analysis of variance with the Bonferroni correction for multiple comparisons. P ≤ 0.05 was considered significant.
GenBank accession number.
The sequence of the cloned gene was deposited into the GenBank database and assigned accession no. U18983.
RESULTS
Selecting clones of S. cerevisiae that exhibited increased adherence to endothelial cells.
The rationale for using complementation cloning to identify genes encoding potential candidal adhesins was based on our observation that S. cerevisiae is approximately 100-fold less adherent to endothelial cells than is C. albicans (25). Thus, the expression of a candidal gene in S. cerevisiae could result in increased endothelial cell adherence.
A panning procedure was used to select for adherent clones of S. cerevisiae. The initial selection for adherent organisms was performed by adding 3 × 108 cells of S. cerevisiae LL-20 transformed with the genomic library from C. albicans ATCC 36082 to endothelial cell monolayers. After three rounds of selection, 20 colonies of adherent S. cerevisiae were identified. Two clones (clones 13 and 68) were identified that appeared to be more adherent than was S. cerevisiae LL-20 transformed with the empty plasmid. On further testing, we found that the adherence of clone 13 was 3.6% ± 1.3% and the adherence of 68 was 1.3% ± 0.9%. In contrast, the adherence of the clone transformed with vector pE20-H alone was 0.7% ± 0.3% (P < 0.001 and P = 0.12 for clones 13 and 68, respectively, compared to the control). Both clones flocculated when grown in liquid medium, although clone 13 appeared to be more flocculent than clone 68. The inserts contained within clones 13 and 68 appeared to be different because they did not cross-hybridize in Southern blots. Because clone 13 was significantly more adherent than clone 68, the plasmid contained in clone 13 became the focus of further studies.
To ensure that the enhanced adherence of clone 13 was due to the plasmid it contained and not to a mutation in the S. cerevisiae LL-20 genome, the plasmid (designated p13) was isolated and amplified in E. coli XL1-Blue. This plasmid was then used to transform S. cerevisiae LL-20, after which the adherence of the newly transformed organisms was quantified. All secondary transformants adhered to the same extent as did the initial isolate (data not shown).
Locating the open reading frame within p13.
We determined that p13 contained a 9.1-kb insert. To determine the location of the open reading frame containing the adherence-conferring gene, this insert was cut into 4.9- and 4.2-kb fragments with BamHI. These two fragments were subcloned into pE20-H to create plasmids p13-1 and p13-2, respectively (Fig. 1). These plasmids were used to transform S. cerevisiae LL-20, and the adherence of these clones was determined. The p13-1 transformant showed no increase in adherence to endothelial cells compared to the control organism transformed with the empty plasmid (P > 0.5) (Fig. 1). However, the clone transformed with p13-2 was over eightfold more adherent than was the control organism (P < 0.0001). Thus, the adherence-promoting activity was localized to the 4.2-kb fragment contained in p13-2. To generate a smaller subclone, the 2.9-kb XhoI fragment within p13-2 was ligated into pE20-H, to produce p13-2x. The organisms transformed with p13-2x adhered significantly more than the control organisms did (P = 0.003) (Fig. 1). The adherence of S. cerevisiae transformed with each of the subclones of p13 was lower than that of organisms transformed with the full-size plasmid (P < 0.05).
FIG. 1.
Effects of p13 and its subclones on the adherence of S. cerevisiae to endothelial cells. The relative sizes of p13 and its subclones are displayed on the left. The graph on the right illustrates the percent adherence to endothelial cells of S. cerevisiae LL-20 transformed with the corresponding plasmids. Results are the mean ± standard deviation for at least three experiments. The hatched rectangle illustrates the location and orientation of the open reading frame of CAD1/AAF1. ∗, P ≤ 0.003 compared to organisms transformed with pE20-H.
By Southern blotting, we determined that the 2.9-kb XhoI fragment could hybridize with genomic DNA from C. albicans ATCC 36082 (Fig. 2). This fragment did not hybridize with DNA from S. cerevisiae LL-20 (data not shown). These results indicate that the cloned fragment was derived from C. albicans.
FIG. 2.
Southern hybridization analysis of genomic DNA from C. albicans ATCC 36082 probed with the CAD1/AAF1-containing XhoI fragment from p13-2x. DNA was digested with XhoI (lane 1), KpnI (lane 2), and BamHI (lane 3). The 2.9-kb XhoI and 2.2-kb KpnI bands were seen during restriction mapping of p13, indicating that the probe bound specifically to CAD1/AAF1. The 6-kb BamHI fragment was not seen during restriction mapping, indicating that a BamHI site exists downstream of the SmaI site in the genome.
Sequence analysis of the adherence-promoting gene.
Sequence analysis of the adherence-inducing region contained within p13 revealed a single 1.8-kb open reading frame. This open reading frame was located entirely within the insert contained within p13-2x. We have named this gene CAD1. The predicted protein contained 612 amino acids and had a molecular mass of 66 kDa. The only homolog present in the database was AAF1, a C. albicans gene that was identified by a complementation strategy similar to the one used in the present study (2). CAD1 was found to be identical to AAF1.
The sequence of the predicted CAD1/AAF1 protein was examined with the expectation that it might be a cell surface protein responsible for adherence. Although the protein contained nine potential N-glycosylation sites, it lacked a discernible secretory leader sequence and appeared to have no potential transmembrane domains. It also lacked serine/threonine-rich regions often found in cell surface proteins. However, the predicted protein did contain both singular and bipartite nuclear localization signals, as well as numerous potential phosphorylation sites, suggesting that it might be a nuclear phosphoprotein. A notable feature of the predicted CAD1/AAF1 protein was that both the amino and carboxyl termini were rich in glutamine. These glutamine-rich motifs are characteristic of multiple regulatory factors, including the SSN6 nuclear phosphoprotein and the protein encoded by the MCM1 transcriptional regulator in S. cerevisiae (23, 30). These glutamine-rich motifs have also been found in several surface proteins such as merozoite surface antigen 2 of Plasmodium falciparum (18). These results suggested that CAD1/AAF1 is either a regulatory factor or a surface adhesin.
CAD1/AAF1 is expressed in both blastospores and germinated C. albicans organisms.
Northern blotting with the 2.9-kb XhoI fragment contained within p13-2x was performed to determine the conditions under which CAD1/AAF1 mRNA accumulates. CAD1/AAF1 mRNA was detected in organisms grown under all conditions tested (data not shown). These conditions included blastospores grown in YNB broth, YPD medium, and Lee’s medium at pH 4.5 and 25°C. In addition, CAD1/AAF1 mRNA expression was present in organisms that were germinated by growth in Lee’s medium at pH 6.8 and 37°C. Thus, CAD1/AAF1 mRNA expression appeared to be independent of the morphology of the organism.
Immunoprecipitation of the HA-CAD1/AAF1 fusion protein from S. cerevisiae.
Epitope tagging was used to estimate the size and cellular location of the protein encoded by CAD1/AAF1. By expressing pYCADN and pYCADC (Table 1) in S. cerevisiae, we were able to synthesize proteins in which the HA epitope was fused to either the N or C terminus, respectively, of the CAD1/AAF1 protein. Immunoprecipitation of either fusion protein yielded a major band of approximately 66 kDa and at least one larger band when analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (Fig. 3). These bands were absent in S. cerevisiae that was transformed with either vector containing the HA epitope without CAD1/AAF1. The predicted size of the HA-CAD1/AAF1 fusion protein was also 66 kDa. Therefore, the finding that both fusion proteins were identical to their predicted size suggested that neither end of the CAD1/AAF1 protein is removed during posttranslational processing. Also, adding the HA epitope to the CAD1/AAF1 protein did not appear to alter its function, since S. cerevisiae LL-20 transformed with either pYCADN or pYCADC still flocculated when expression of the gene was induced by galactose. On the basis of these results, we concluded that the epitope-tagged fusion protein was suitable for determining the cellular location of the CAD1/AAF1 gene product.
FIG. 3.
Immunoprecipitation of the epitope-tagged CAD1/AAF1 protein. An anti-HA antibody was used to immunoprecipitate proteins from S. cerevisiae LL-20 transformed with pYef1H (lane 1), pYCADN (lane 2), pYef2H (lane 3), and pYCADC (lane 4). Organisms transformed with pYCADN synthesized CAD1/AAF1 with the HA epitope fused to the N terminus, whereas yeasts transformed with pYCADC synthesized CAD1/AAF1 with the HA epitope attached to the C terminus. Proteins were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis on an 8% gel and then identified by autoradiography. Molecular weight markers (in thousands) are shown on the right.
Determining the cellular location of the CAD1/AAF1 protein.
As noted above, the sequence analysis did not clearly indicate the nature of the CAD1/AAF1 protein. We therefore used immunoelectron microscopy of S. cerevisiae LL-20 transformed with epitope-tagged CAD1/AAF1 (pYCADN) to determine the cellular localization of this protein. In all electron micrographs, we observed consistent binding of the immunogold to the cytoplasm and nuclei of organisms transformed with pYCADN (Fig. 4). No accumulation of gold particles in either the cell wall or plasma membrane of these organisms was seen. Furthermore, in organisms transformed with the control plasmid pYeF1H, no specific accumulation of immunogold was observed. This localization of the HA-CAD1/AAF1 fusion protein to the cytoplasm and nucleus is consistent with the conclusion that CAD1/AAF1 probably encodes a regulatory protein rather than a surface adhesin.
FIG. 4.
Immunogold labeling demonstrated that the HA-CAD1/AAF1 fusion protein localizes to the cytoplasm and nucleus in S. cerevisiae. S. cerevisiae transformed with either vector alone (pYeF1H) (A) or the vector encoding the HA-CAD1/AAF1 fusion protein (pYCADN) (B) is shown. After being fixed, the organisms were sectioned and then the HA-CAD1/AAF1 fusion protein was labeled by exposure to murine anti-HA followed by gold-labeled goat anti-mouse antibodies. The organisms and gold particles were viewed by transmission electron microscopy. Arrows indicate the cell walls of the organisms.
Flocculation and endothelial cell adherence of S. cerevisiae appear to be related.
Because CAD1/AAF1 induced both flocculation and enhanced endothelial cell adherence, we examined the endothelial cell adherence of other strains of S. cerevisiae that expressed genes known to confer flocculation by two different mechanisms. These two different flocculation phenotypes were Flo1, which was expressed by S. cerevisiae ATCC 48868 (21), and NewFlo, which was expressed by S. cerevisiae NCYC 1364 (31). These organisms strongly flocculated, and their adherence to endothelial cells was similar to that of S. cerevisiae LL-20 containing p13 (Table 2). These data indicate that in S. cerevisiae, flocculation and adherence to endothelial cells appear to be related.
TABLE 2.
Comparison of adherence and flocculation of different strains of S. cerevisiae
| S. cerevisiae strain | % Adherence to endothelial cells | Critical cell density of flocculation (no. of cells/ml) |
|---|---|---|
| LL-20/pE20-H | 0.1 ± 0.1 | >108 |
| LL-20/p13 (CAD1/AAF1) | 1.1 ± 0.5 | (1.2 ± 0.2) × 107 |
| ATCC 48868 (Flo1) | 1.0 ± 0.6 | (0.61 ± 0.01) × 107 |
| NCYC 1364 (NewFlo) | 0.9 ± 0.7 | (0.11 ± 0.03) × 107 |
To further examine the relationship between flocculation and endothelial cell adherence, we tested the ability of two nonflocculating strains of S. cerevisiae to adhere to endothelial cells. The adherence of these two strains, X2180A and YM21, was 0.1% ± 0.1% and 0.2% ± 0.2%, respectively. This adherence was significantly lower than that of the flocculating organisms (P < 0.001) and was not significantly different from that of strain LL-20 transformed with the empty plasmid.
To determine whether the mechanism of flocculation in CAD1/AAF1-transformed organisms was similar to the mechanisms of flocculation in either the Flo1 or NewFlo phenotypes, we compared the effects of potential flocculation inhibitors on organisms expressing different flocculation phenotypes (Table 3). Strains LL-20/p13 and ATCC 48868 (Flo1 phenotype) exhibited the same pattern of inhibition, suggesting that CAD1/AAF1 may induce the Flo1 phenotype.
TABLE 3.
Effects of potential inhibitors on the relative critical cell density of flocculation
| Inhibitor (concn) | % Critical cell density of flocculation (% of control) inb:
|
||
|---|---|---|---|
| LL-20/p13 (CAD1/AAF1) | ATCC 48868 (Flo1) | NCYC 1364 (NewFlo) | |
| Nonea | 100 ± 17 | 100 ± 2 | 100 ± 27 |
| Glucose (1 M) | 90 ± 10 | 70 ± 40 | CIc |
| Galactose (1 M) | 130 ± 10 | 60 ± 20 | 460 ± 120 |
| Maltose (1 M) | 160 ± 10 | 110 ± 60 | CI |
| Mannose (1 M) | CI | CI | CI |
| NaCl (5 M) | 270 ± 50 | 260 ± 50 | CI |
| Proteinase K (10 μg/ml)d | CI | CI | CI |
Control values in the absence of inhibitors were as follows: LL-20/p13, 1.2 × 107 ± 0.2 × 107; ATCC 48868, 0.61 × 107 ± 0.01 × 107; NCYC 1364, 0.11 × 107 ± 0.03 × 107.
Cell densities are expressed as percentages of the value obtained with no inhibitor.
CI, complete inhibition.
Cells were incubated in proteinase K at 30°C for 30 min.
DISCUSSION
Using a complementation strategy, we have identified a 1.8-kb gene (CAD1/AAF1) from C. albicans which, when expressed in S. cerevisiae, causes these cells to flocculate and exhibit enhanced adherence to endothelial cells. Expression of CAD1/AAF1 in S. cerevisiae produced a phenotype similar to Flo1 in several respects. As with the Flo1 strain, flocculation of organisms expressing CAD1/AAF1 could be inhibited completely by mannose but not maltose or glucose. In addition, the endothelial cell adherence of S. cerevisiae expressing CAD1/AAF1 was similar to that of the organism with the Flo1 phenotype.
While performing these investigations, we found that flocculating strains of S. cerevisiae exhibited increased adherence to endothelial cells compared to nonflocculating strains. These results suggest that flocculation and adherence in S. cerevisiae may be mediated by the same receptor. This close relationship between flocculation and adherence is absent in C. albicans, because this organism flocculates significantly less than do flocculating strains of S. cerevisiae but adheres to endothelial cells with much greater affinity.
By immunoelectron microscopy, it was observed that the epitope-tagged CAD1/AAF1 protein localized to the cytoplasm and nucleus but not to the cell wall of S. cerevisiae. We have obtained similar results by indirect immunofluorescence (unpublished data). Consistent with these findings that CAD1/AAF1 encodes a regulatory protein, the predicted CAD1/AAF1 protein contains both singular and bipartite nuclear localization signals. However, it does not contain a consensus signal peptide or a definite transmembrane sequence, both of which are frequently present in surface proteins. Thus, it is likely that CAD1/AAF1 induces adherence and aggregation in S. cerevisiae by an indirect mechanism.
This function is consistent with the presence of the glutamine-rich motifs in CAD1/AAF1, which are also present in numerous transcription factors such as the SSN6 nuclear phosphoprotein and the protein encoded by the MCM1 transcriptional regulatory gene in S. cerevisiae (23, 30). It is known that mutations that inhibit the function of SSN6 in S. cerevisiae induce the Flo1 phenotype (15, 32). Thus, it is tempting to speculate that the expression of CAD1/AAF1 in S. cerevisiae interferes with the SSN6 regulatory cascade, which results in either the activation or derepression of genes responsible for producing the Flo1 phenotype.
The Flo1 phenotype is thought to be mediated by the expression of FLO1. This gene encodes a cell surface lectin that binds to mannoproteins (3, 21). Thus, it is likely that organisms expressing FLO1 can bind to mannose-containing ligands on the surface of the endothelial cells. Interestingly, both Streptococcus pneumoniae and Staphylococcus aureus are known to bind to mannose on the endothelial cell surface (5, 14). However, we have shown previously that neither mannose nor mannoproteins inhibit the adherence of C. albicans to endothelial cells (25). Furthermore, to our knowledge, a homolog of FLO1 has not been found in C. albicans. Therefore, whether CAD1/AAF1 regulates the endothelial cell adherence of C. albicans remains uncertain.
Barki et al. (2) used a complementation strategy similar to the one described in this paper to identify genes of C. albicans encoding potential adhesins. By this strategy, they cloned a fragment of DNA from C. albicans that induces S. cerevisiae to flocculate and adhere to polystyrene and buccal epithelial cells. The sequence of the gene contained within this fragment (AAF1) has not been published, although it has been submitted to GenBank. On the basis of its sequence, AAF1 is the same gene as CAD1.
Similar to our results, these investigators found that the epithelial cell adherence of S. cerevisiae transformed with AAF1 was significantly lower than that of C. albicans (2). It is possible that CAD1/AAF1 does not induce adherence in C. albicans. However, if it does stimulate the expression of an adhesin, there are several possible explanation for the difference in adherence between C. albicans and S. cerevisiae. First, the level of CAD1/AAF1 expression may be different in S. cerevisiae and in C. albicans. Second, it is possible that CAD1/AAF1 activates a different set of genes when it is expressed in S. cerevisiae. A final possibility is that the mechanism of adherence induced by CAD1/AAF1 is not the dominant mechanism that is operative in C. albicans.
When subclones of p13 that contained smaller fragments of candidal DNA were expressed in S. cerevisiae, the endothelial cell adherence of these organisms was reduced, although it still remained greater than that of yeast transformed with vector alone. Similar results were observed by Barki et al. (2). The most likely explanation for these findings is that the subclones contained successive deletions of the CAD1/AAF1 promoter. Alternatively, there may have been a reduction in the plasmid copy number in the subclones.
Barki et al. also raised polyclonal antibodies against intact S. cerevisiae organisms that were transformed with the fragment of candidal DNA containing AAF1 (1). When tested in immunoblots of cell surface extracts from C. albicans, these antibodies detected a protein of approximately 30 kDa. However, this protein was not seen in similar extracts from S. cerevisiae transformed with the candidal DNA. Our immunoprecipitation data indicate that the molecular mass of the CAD1/AAF1 gene product is about 66 kDa, which is similar to the size of the predicted protein based on the sequence data. Because the same size of protein was precipitated irrespective of the end of the gene to which the epitope tag was attached, it is likely that the CAD1/AAF1 protein does not undergo posttranslation shortening. These results, combined with our finding that the CAD1/AAF1 protein did not localize to the cell wall, suggest that the 30-kDa protein detected by Barki et al. is not encoded by CAD1/AAF1.
In conclusion, CAD1/AAF1 encodes a glutamine-rich protein that is probably a regulatory factor. When expressed in S. cerevisiae, CAD1/AAF1 causes these cells to flocculate and adhere to endothelial cells. Flocculation is mediated by a lectin-like interaction with mannose and appears to be closely related to endothelial cell adherence. Whether CAD1/AAF1 also stimulates the expression of one or more adhesins in C. albicans remains to be determined. To answer this question, construction of mutants of C. albicans in which both alleles of CAD1/AAF1 have been deleted is in progress.
ACKNOWLEDGMENTS
We thank the perinatal nurses at Harbor-UCLA and Torrance Memorial Medical Centers for collecting umbilical cords, Alison Orozco and Toshiko Lamkin for helping with tissue culture, and Toyota USA for donating the Olympus phase-contrast microscope.
This work was supported in part by Public Health Service grants R01 AI-19990, P01 AI-37194, R29 AI040636, and MO1 RR00425 from the National Institutes of Health.
REFERENCES
- 1.Barki M, Koltin Y, van Wetter M, Rosenberg M. A Candida albicans surface antigen mediating adhesion and autoaggregation in Saccharomyces cerevisiae. Infect Immun. 1994;62:4107–4111. doi: 10.1128/iai.62.10.4107-4111.1994. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Barki M, Koltin Y, Yanko M, Tamarkin A, Rosenberg M. Isolation of a Candida albicans DNA sequence conferring adhesion and aggregation on Saccharomyces cerevisiae. J Bacteriol. 1993;175:5683–5689. doi: 10.1128/jb.175.17.5683-5689.1993. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Bidard F, Bony M, Blondin B, Dequin S, Barre P. The Saccharomyces cerevisiae FLO1 flocculation gene encodes for a cell surface protein. Yeast. 1995;11:809–822. doi: 10.1002/yea.320110903. [DOI] [PubMed] [Google Scholar]
- 4.Christophe C, Minvielle-Sebastia L. Multipurpose vectors designed for the fast generation of N- or C-terminal epitope-tagged proteins. Yeast. 1994;10:105–112. doi: 10.1002/yea.320100110. [DOI] [PubMed] [Google Scholar]
- 5.Cundell D R, Tuomanen E I. Receptor specificity of adherence of Streptococcus pneumoniae to human type-II pneumocytes and vascular endothelial cells in vitro. Microb Pathog. 1994;17:361–374. doi: 10.1006/mpat.1994.1082. [DOI] [PubMed] [Google Scholar]
- 6.Edwards J E, Jr, Gaither T A, O’Shea J J, Rotrosen D, Lawley T J, Wright S A, Frank M M, Green I. Expression of specific binding sites on Candida with functional and antigenic characteristics of human complement receptors. J Immunol. 1986;137:3577–3583. [PubMed] [Google Scholar]
- 7.Edwards J E, Jr, Mayer C L, Filler S G, Wadsworth E, Calderone R A. Cell extracts of Candida albicans block adherence of the organisms to endothelial cells. Infect Immun. 1992;60:3087–3091. doi: 10.1128/iai.60.8.3087-3091.1992. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Edwards J E, Jr, Montgomerie J Z, Foos R Y, Shaw V K, Guze L B. Experimental hematogenous endophthalmitis caused by Candida albicans. J Infect Dis. 1975;131:649–657. doi: 10.1093/infdis/131.6.649. [DOI] [PubMed] [Google Scholar]
- 9.Ghannoum M A, Filler S G, Ibrahim A S, Fu Y, Edwards J E., Jr Modulations of interactions of Candida albicans and endothelial cells by fluconazole and amphotericin B. Antimicrob Agents Chemother. 1992;36:2239–2244. doi: 10.1128/aac.36.10.2239. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Glee P M, Masuoka J, Ozier W T, Hazen K C. Presence of multiple laminin- and fibronectin-binding proteins in cell wall extract of Candida albicans: influence of dialysis. J Med Vet Mycol. 1996;34:57–61. doi: 10.1080/02681219680000091. [DOI] [PubMed] [Google Scholar]
- 11.Harlow E, Lane D. Antibodies: a laboratory manual. Cold Spring Harbor, N.Y: Cold Spring Harbor Laboratory Press; 1988. [Google Scholar]
- 12.Hostetter M K. Adhesins and ligands involved in the interaction of Candida spp. with epithelial and endothelial surfaces. Clin Microbiol Rev. 1994;7:29–42. doi: 10.1128/cmr.7.1.29. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Jaffe E A, Nachman R L, Becker C G, Ninick C R. Culture of human endothelial cells derived from umbilical veins: identification by morphologic and immunologic criteria. J Clin Invest. 1973;52:2745–2756. doi: 10.1172/JCI107470. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Johnson C M. Staphylococcus aureus binding to cardiac endothelial cells is partly mediated by a 130 kilodalton glycoprotein. J Lab Clin Med. 1993;121:675–682. [PubMed] [Google Scholar]
- 15.Keleher C A, Redd M J, Schultz J, Carlson M, Johnson A D. Ssn6-Tup1 is a general repressor of transcription in yeast. Cell. 1992;68:709–719. doi: 10.1016/0092-8674(92)90146-4. [DOI] [PubMed] [Google Scholar]
- 16.Klotz S A, Rutten M J, Smith R L, Babcock S R, Cunningham M D. Adherence of Candida albicans to immobilized extracellular matrix proteins is mediated by calcium-dependent surface glycoproteins. Microb Pathog. 1993;14:133–147. doi: 10.1006/mpat.1993.1014. [DOI] [PubMed] [Google Scholar]
- 17.Lopez-Ribot J L, Chaffin W L. Binding of the extracellular matrix component entactin to Candida albicans. Infect Immun. 1994;62:4564–4571. doi: 10.1128/iai.62.10.4564-4571.1994. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Marshall V M, Coppel R L, Anders R F, Kemp D J. Two novel alleles within subfamilies of the merozoite surface antigen 2 (MSA-2) of Plasmodium falciparum. Mol Biochem Parasitol. 1992;50:181–184. doi: 10.1016/0166-6851(92)90255-i. [DOI] [PubMed] [Google Scholar]
- 19.McLean I W, Nakane P K. Periodate-lysine-paraformaldehyde fixative. A new fixation for immunoelectron microscopy. J Histochem Cytochem. 1974;22:1077–1083. doi: 10.1177/22.12.1077. [DOI] [PubMed] [Google Scholar]
- 20.Miki B L, Poon N H, James A P, Seligy V L. Possible mechanism for flocculation interactions governed by gene FLO1 in Saccharomyces cerevisiae. J Bacteriol. 1982;150:878–889. doi: 10.1128/jb.150.2.878-889.1982. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Miki B L, Poon N H, Seligy V L. Repression and induction of flocculation interactions in Saccharomyces cerevisiae. J Bacteriol. 1982;150:890–899. doi: 10.1128/jb.150.2.890-899.1982. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Orr-Weaver T L, Szostak J W, Rothstein R J. Yeast transformation: a model system for the study of recombination. Proc Natl Acad Sci USA. 1981;78:6354–6358. doi: 10.1073/pnas.78.10.6354. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Passmore S, Maine G T, Elble R, Christ C, Tye B K. Saccharomyces cerevisiae protein involved in plasmid maintenance is necessary for mating of MAT alpha cells. J Mol Biol. 1989;204:593–606. doi: 10.1016/0022-2836(88)90358-0. [DOI] [PubMed] [Google Scholar]
- 24.Pendrak M L, Klotz S A. Adherence of Candida albicans to host cells. FEMS Microbiol Lett. 1995;129:103–113. doi: 10.1111/j.1574-6968.1995.tb07566.x. [DOI] [PubMed] [Google Scholar]
- 25.Rotrosen D, Edwards J E, Jr, Gibson T R, Moore J D, Cohen A H, Green I. Adherence of Candida to cultured vascular endothelial cells: mechanisms of attachment and endothelial cell penetration. J Infect Dis. 1985;152:1264–1273. doi: 10.1093/infdis/152.6.1264. [DOI] [PubMed] [Google Scholar]
- 26.Sambrook J, Fritsch E F, Maniatis T. Molecular cloning: a laboratory manual. 2nd ed. Cold Spring Harbor, N.Y: Cold Spring Harbor Laboratory Press; 1989. [Google Scholar]
- 27.Sanger F, Nicklen S, Coulson A R. DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci USA. 1977;74:5463–5467. doi: 10.1073/pnas.74.12.5463. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Scherer S, Stevens D S. A Candida albicans dispersed, repeated gene family and its epidemiological application. Proc Natl Acad Sci USA. 1988;85:1452–1456. doi: 10.1073/pnas.85.5.1452. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Schiestl R H, Gietz R D. High efficiency transformation of intact yeast cells using single stranded nucleic acids as a carrier. Curr Genet. 1989;16:339–346. doi: 10.1007/BF00340712. [DOI] [PubMed] [Google Scholar]
- 30.Schultz J, Carlson M. Molecular analysis of SSN6, a gene functionally related to the SNF1 protein kinase of Saccharomyces cerevisiae. Mol Cell Biol. 1987;7:3637–3645. doi: 10.1128/mcb.7.10.3637. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Stratford M, Assinder S. Yeast flocculation: Flo1 and NewFlo phenotypes and receptor structure. Yeast. 1991;7:559–574. doi: 10.1002/yea.320070604. [DOI] [PubMed] [Google Scholar]
- 32.Teunissen A W, van den Berg J A, Steensma H Y. Transcriptional regulation of flocculation genes in Saccharomyces cerevisiae. Yeast. 1995;11:436–446. doi: 10.1002/yea.320110506. [DOI] [PubMed] [Google Scholar]