Adhesion of Candida albicans to host cells is believed to represent a fungal virulence factor and a significant step in the development of candidiasis. As C. albicans strains may differ in their in vitro adhesion ability we initiated a study to investigate whether mutant strains differ in this respect from their parent wild-type. We assessed the in vitro adhesion of C. albicans CBS562 and two mutants obtained by mutagenesis with N′-nitrosoguanidine: a histidine auxotroph, SAG5, derived from CBS562, and a respiratory-deficient strain (a petite mutant), SAR1, derived from SAG5. The adhesion was tested in vitro using two target cell systems: (1) exfoliated human buccal epithelial cells (BEC); and (2) human keratinocyte tissue line cells (HaCaT cells). Adhesion to BEC was evaluated microscopically and that to HaCaT cells by a direct ELISA technique. The results indicated a 54% reduction in adhesion to BEC for SAG5 and 30% for SAR1 as compared to the wild-type, and a 25% reduction in adhesion to HaCaT cells for SAG5 and 20% for SAR1. To verify whether the prototrophy restores the adhesion ability, we complemented the his-negative auxotroph by transforming the strain with the HIS4 gene. Then we assayed the adhesion to BEC of the complemented his-negative mutant in comparison to that of the wild-type, the his-negative mutant (SAG5) and the plasmid-cured transformant. The adhesion values of the complemented his-negative strain were similar to those of the wild-type, whereas the values of the plasmid-cured strain were similar to those of SAG5.
Candida albicans is a human commensal. In recognition of the increased significance of C. albicans as a pathogen, an expansion of research aimed at defining factors associated with virulence, as well as towards understanding the basic biology of the organism, has occurred [1, 2]. Attachment of microorganisms to animal or human tissues is believed to be a significant step in the interaction between pathogen and host, initiating the evolution of infection . Candidal adhesion to the host, similarly to other microbial systems, is considered a significant step in the development of candidiasis and a fungal virulence factor [1, 4, 5]. The mechanisms underlying the process of adhesion have been widely investigated in an effort to understand the pathogenesis of evolution of infection.
In addition to virulence factors which are generally considered attributes of pathogenicity, such as adhesion, proteinase and phospholipase production [4, 6, 7], other factors may also influence the pathogenic potential of the fungus. Polak  reported that inhibitors of sterol biosynthesis significantly reduced the virulence of C. albicans strains. Other factors may include auxotrophy, as auxotrophic mutants exhibited a lowered level of pathogenicity in comparison to a control wild-type strain [8, 10–12]. However, this may be variable, e.g. Kirsch and Whitney  reported that while three out of four auxotrophs (adenine, uracil, heme) showed reduced virulence, the fourth mutant, a leucine auxotroph, was not less pathogenic than the wild-type. Although the molecular basis for this phenomenon is not fully established , these observations suggest that metabolic abnormalities may lead to a reduction in virulence.
In spite of the importance attributed to adhesion as a virulence factor, only few studies have focused on comparative evaluation of in vitro adhesion of mutants versus the respective parent strains. We report here on the adhesion of two mutants obtained by N′-nitrosoguanidine mutagenesis: a histidine auxotroph, SAG5, isolated from C. albicans CBS562 , and a respiratory-deficient strain, SAR1, derived from the histidine auxotroph (SAR1 has a double mutation), in comparison to the wild-type.
To further assess the relation between the auxotrophic mutation and the adhesion ability, we evaluated the adhesion of the complemented his− auxotroph and its plasmid-cured transformant. The adhesion was tested in vitro using two target cell systems and two evaluation methods: (1) exfoliated human buccal epithelial cells (BEC), adhesion was evaluated microscopically; and (2) a human keratinocyte tissue cell line (HaCaT cells), adhesion evaluated by a direct ELISA technique.
2Materials and methods
2.1C. albicans strains
The following strains were used throughout the study: (1) wild-type strain C. albicans CBS562 (ATCC 18804), which was obtained from the Centraalbureau Schimmelcultures, Delft, The Netherlands and is the type species strain; (2) SAG5, a histidine auxotroph, which was isolated following mutagenesis of CBS562 with N′-nitrosoguanidine (NTG) [13, 14]; (3) a complemented his− auxotroph which was transformed with the HIS4 gene ; (4) a plasmid-cured complemented his− auxotroph; (5) SAR1, a respiratory-deficient strain, a petite mutant, which was isolated following NTG mutagenesis of SAG5 .
The strains were maintained on yeast-peptone-dextrose (YPD) agar slants containing 1% yeast extract, 2% peptone, 2% dextrose and 2% bacto agar  or on SD agar slants, a minimal medium containing 0.67% yeast nitrogen base without amino acids, 2% dextrose and 2% bacto agar, supplemented with histidine if needed.
2.2Complementation of his-negative auxotroph with the HIS4 gene
Complementation of the his-negative strain was performed by transforming SAG5 with the HIS4 gene. Transformation was carried out by a modification of the procedure described by Varma et al.  using an electroporation technique (Gene Pulser, Bio-Rad). The HIS4 gene was constructed on the plasmid p4173 (unpublished data). To test whether the his+ colonies were plasmid-encoded we performed plasmid curing of the transformants as described by Goshorn et al. .
2.3Preparation of yeast cultures for adhesion assays
For each experiment fresh subcultures were grown in liquid YPD or in DS medium at 28°C under constant shaking. Cells were collected by centrifugation, washed three times in phosphate-buffered saline (PBS) and standardized to 1×108 yeasts ml−1 by spectrophotometry (Kllet, OD550), and microscopic counts (via hemocytometer).
The adhesion assays were performed with a pool of human BEC that were collected from healthy donors by gently rubbing the buccal epithelium with sterile cotton swabs. The swabs were stirred in PBS to release the cells. The cells were washed three times with PBS, counted microscopically (hemocytometer) and resuspended in PBS to a concentration of 1×106 cells ml−1.
2.5Growth of HaCaT cells
HaCaT cells (human keratinocyte cell line) were grown in 20 ml of Dulbecco's modified Eagle medium (DMEM) supplemented with 10% fetal calf serum (FCS). A tenth (0.1 ml) of the cell suspension (2.5×106 ml−1) in DMEM+10% FCS (1 ml of FCS+10 ml of DMEM) was inoculated into each well in a 96-well microtiter plate (flat bottom, Cell Cult). The cells were then incubated at 37°C and CO2 for at least 48 h or until a confluent cell layer was seen in each of the wells.
2.6Adhesion to human exfoliated BEC
The method used was as described previously for buccal and vaginal cells . A mixture of equal volumes (0.2 ml) of BEC (106 cells ml−1) and C. albicans (108 yeasts ml−1) was incubated on a rotator at 37°C for 2 h. Adhesion was assayed microscopically by counting the total number of adherent yeasts to 100 epithelial cells. In addition, adhesion was evaluated according to five categories of epithelial cells: category A, epithelial cells without adherent yeasts; categories B, C, D and E were epithelial cells with 1–5, 5–10, 10–20, or 20 and more adherent yeasts, respectively.
2.7Adhesion to keratinocytes cell line
The adherence assay was performed as described previously for different cells . Microtiter plates with HaCaT cells were washed twice with PBS, 0.1 ml of 0.2% glutaraldehyde was added to each well for 10 min and the plates were then washed again with PBS. To each well, 0.1 ml of 5% bovine serum albumin (BSA) in PBS was added and the plates were incubated at 37°C for 1 h. Following incubation, the plates were washed four times with PBS and 0.1 ml of the yeast culture (108 ml−1, see above) was added to each well. The plates were incubated for 1 h at 37°C. After incubation, the plates were washed four times with PBS and kept overnight at 4°C.
Detection of adhesion was performed by a direct ELISA technique: peroxidase-conjugated rabbit anti-C. albicans mannan (Dako, Denmark) was added (0.1 ml per well, diluted 1:250 in 5% BSA-PBS) for 1 h incubation at 37°C. After five washes, a substrate solution [o-phenylenediamine dihydrochloride (OPD)] was added; the color was allowed to develop and measured at OD405 using an ELISA reader. Controls consisted of wells with the cells without C. albicans.
The adhesion ability of: (1) a wild-type strain of C. albicans (CBS562), (2) a histidine auxotroph (SAG5), (3) a complemented his-negative auxotroph, (4) its plasmid-cured transformant and (5) a respiratory-deficient mutant (SAR1) was assessed to BEC and/or HaCaT cells using two techniques for evaluation: microscopy (adhesion to BEC) and a direct ELISA technique (adhesion to HaCaT cell line).
3.1Adhesion to human buccal epithelial cells
Fig. 1 presents the adhesion of C. albicans CBS562 and two mutants. Adhesion to BEC is expressed as total number of adherent C. albicans per 100 BEC. Values were compared to those of the wild-type and were analyzed for statistical significance by Student's t-test.
As noticed, while 1225 CBS562 yeasts adhered to 100 BEC, only 523 of SAG5 and 789 of SAR1 organisms adhered. The adhesion values of SAG5 and SAR1 in comparison to those of CBS562 are statistically significantly lower (P<0.01).
Microscopic evaluation of in vitro adhesion mixtures reveals that differences exist as to the number of adherent yeasts per individual epithelial cell. Based on this working hypothesis, our evaluation technique also included differentiation between several types of adhesion patterns: from BEC with no adherent yeasts to BEC covered with numerous yeasts (20 or more), as described earlier in Section 2. Fig. 2 depicts the distribution of adhesion to BEC of CBS562, SAG5 and SAR1 according to the five categories. In category E, which represents the highest binding epithelial cells (20 and more adherent C.A./BEC), the highest values are for CBS (38%) whereas the two mutants show significantly lower values, SAG5 7% and SAR1 16%. On the other hand, among categories A and B, groups consisting of BEC which do not adhere yeasts or have only 0–5 adherent yeasts/BEC, the higher values were for the mutants (P<0.01).
The data analysis using Student's t-test shows that there was no significant difference in the adhesion values of the two mutants SAG5 and SAR1.
To assess a possible correlation between mutation and the reduction in the adherence ability noticed in these experiments, we tested in the next step the adhesion of a complemented his-negative auxotroph in comparison to the wild-type, and on the other hand to its plasmid-cured transformant. The results are presented in Fig. 3, showing that the adhesion values of the complemented his-negative mutant are similar to those of the wild-type, and significantly higher than those of SAG5 (t-test, P<0.01). In addition, loss of the gene by plasmid curing again led to reduction of adherence ability, revealing adhesion values similar to those of SAG5, which were significantly lower (P<0.01) from those of the complemented his-negative mutant.
3.2Adhesion to HaCaT cells
Adhesion to HaCaT cells, evaluated by a direct ELISA technique, of CBS, SAG5 and SAR1 strains was assessed in four independent experiments. In each experiment, 10 wells were assayed for each strain. Fig. 4 summarizes mean data from the four experiments of adhesion to the HaCaT cells. In this system as well, data were compared to those of CBS562 and analyzed by Student's t-test. The mean values of the OD405 readings were 0.243 for CBS562, 0.183 for SAG5 and 0.194 for SAR1. According to these data, the adhesion ability of SAG5 and SAR1 was decreased significantly in comparison to the wild-type CBS562 (P<0.05).
The results of the adhesion experiments to the HaCaT cell line indicate that although different target cells and different evaluation methods were used, a similar trend can be noted as in adhesion to BEC. This trend can be demonstrated by calculating the ratio of adhesion ability for both mutants versus the wild-type (considered 100%) following evaluation by the two cell systems, shown in Table 1.
Table 1. Adhesion of the mutant strains
Ratio of adhesion values vs. wild-type (%)
aThe adhesion values relate to C.A. cells that adhere to 100 BEC.
The results obtained in this study indicate a marked reduction in the in vitro ability to adhere to two types of human cells, of the auxotroph and respiratory-deficient mutants, respectively. Reduced adhesion of both mutants is evident in the two evaluation systems (microscopy and ELISA) and the two types of cells (BEC and HaCaT) tested, although some differences between the SAG5 and SAR1 mutants can be noted. To the best of our knowledge, there are no reports concerning the in vitro adhesion ability of auxotrophic or respiratory-deficient strains in C. albicans.
Respiratory deficiency, petite mutation, in C. albicans is a metabolic defect on which only scant information has been published . Our SAR1 strain is a double mutant, that in addition to auxotrophy also has a respiratory deficiency.
To verify the relation between a phenotypic mutation, such as his-negative auxotrophy, and the adhesion ability we conducted experiments in which the mutation was complemented. These experiments revealed that complementing our his-negative mutant with the plasmid carrying the HIS4 gene led to the recovery of the adhesion ability while loss of the gene by plasmid curing resulted again in reduced adhesion ability. Thus, the his-negative auxotrophic mutation is, apparently, related to the reduction in adhesion.
Studies by several investigators exploring pathogenicity of auxotroph or respiratory-deficient mutants [9–13, 22] in experimental animals demonstrated a lower pathogenic potential of these strains. Thus, our in vitro studies on adhesion, a process considered a marker for virulence, are compatible with these in vivo data. Moreover, Lehrer et al.  showed with a cerulenine-resistant C. albicans mutant a correlation between in vitro adhesion ability and in vivo pathogenicity. The mutants, which adhered less in vitro to vaginal cells, were also less pathogenic in a murine experimental vaginitis model. We plan to investigate in vivo the virulence of our mutants. Gaining this information may contribute to a further understanding of the pathogenesis of candidiasis.