Differential expression of secreted aspartyl proteinases in a model of human oral candidosis and in patient samples from the oral cavity

Authors

  • Martin Schaller,

    1. Institute for General Botany, Applied Molecular Biology III, University of Hamburg, Ohnhorststr. 18, D-22609 Hamburg, Germany.,
    2. Department of Dermatology, Ludwig-Maximilians-University of Munich, Munich, Germany.
    Search for more papers by this author
  • Wilhelm Schäfer,

    1. Institute for General Botany, Applied Molecular Biology III, University of Hamburg, Ohnhorststr. 18, D-22609 Hamburg, Germany.,
    Search for more papers by this author
  • Hans C. Korting,

    1. Department of Dermatology, Ludwig-Maximilians-University of Munich, Munich, Germany.
    Search for more papers by this author
  • Bernhard Hube

    1. Institute for General Botany, Applied Molecular Biology III, University of Hamburg, Ohnhorststr. 18, D-22609 Hamburg, Germany.,
    Search for more papers by this author

Bernhard Hube Tel. (40) 82 282 393; Fax (40) 82 282 513.

Abstract

Candida albicans, an opportunistic pathogen in humans, secretes secretory aspartyl proteinases (Saps), which have been correlated with virulence. We examined the temporal regulation of the mRNA expression of seven known members of the SAP gene family by reverse transcription polymerase chain reaction (RT–PCR) in (i) an in vitro model of oral candidosis based on reconstituted human epithelium (RHE); and (ii) clinical samples from patients with oral candidosis. SAP1 and SAP3 transcripts were first detected 42 h after inoculation of RHE, while at the same time, slight morphological alterations in the epithelium were documented by light microscopy. SAP6 expression occurred 6 h later concomitantly with germ tube formation of some infecting Candida cells and severe lesions of the epithelial tissue. SAP2 and SAP8 RT–PCR products were first detected 60 h after infection, while SAP4 and SAP5 transcripts were never discovered. Thus, a temporal progression of SAP expression in the order SAP1 and SAP3 > SAP6 > SAP2 and SAP8 was observed at the same time as increasing RHE damage occurred. At the protein level, Sap antigen was found within the C. albicans yeast cells and the epithelial cells by immunoelectron microscopy using an anti-Sap murine monoclonal antibody directed against the gene products Sap1–3. Expression of SAP1–3 and 6 was also detected by RT–PCR in samples from patients suffering from oral candidosis. Our results suggest that the pathogenesis of experimental and clinical oral candidosis is associated with the differential and temporal regulation of SAP gene expression.

Introduction

Candida albicans is an opportunistic pathogen that colonizes human mucosal surfaces, such as the oral cavity, as a commensal. Under certain circumstances, usually linked to a compromised host immune system, C. albicans causes infections that may be restricted to the mucosa or, in severe cases of immunodepression, progress to systemic invasion. Several factors, e.g. hyphae formation, phenotypic switching and secretion of hydrolytic enzymes, contribute to its virulence (Odds, 1988). Among these, the secretion of aspartyl proteinases (Saps), encoded by a gene family with at least nine members (SAP 1–9 ), is of key importance (Hube, 1996; Hube et al., 1997a; Sanglard et al., 1997). The in vitro expression of the different secreted aspartyl proteinase (SAP ) genes has been extensively investigated by Northern analysis. Different expression of various SAP mRNA has been detected according to growth state, medium composition and cell morphology (Morrow et al., 1992; Hube et al., 1994; White and Agabian, 1995; Hube et al., 1997b). In vitro experiments have also been performed at the protein level (Sap) by Western analysis. These studies have shown that Sap enzyme production was predominantly regulated at the level of transcription (White and Agabian, 1995). Prophylactic treatment with the Sap inhibitor pepstatin A has been shown to prevent adhesion to epithelial cells or to protect infected animals, suggesting that Sap isoenzymes are important for pathogenesis (Borg and Rüchel, 1988; Ray and Payne, 1988; Ollert et al., 1993; Fallon et al., 1997). In addition, antibodies against Saps have been shown to be produced in patients with systemic candidosis (Rüchel, 1983).

In vitro data, therefore, suggest that the temporal and specific regulation of SAP expression is important for the progression of individual infections. However, direct examination of the expression of individual members of the SAP gene family during infection has yet to be carefully addressed. The expression of SAP1 and SAP2 has been demonstrated by Northern analysis in an experimental rat vaginitis model (De Bernardis et al., 1995). However, the pattern of SAP expression in experimental and clinical oral candidosis awaits investigation. In vitro models of cutaneous candidosis based on reconstructed human epidermis have been used successfully for investigation of the pathogenesis and treatment of candidosis (Korting et al., 1998).

In the present study, we have established an in vitro model of oral candidosis in order to investigate the expression of secreted aspartyl proteinases at the mRNA and protein level and histological alterations during the course of infection. These results were compared with the in vivo situation as represented by biopsies of oral candidosis.

Results

Reproducibility and verification of expression studies

We used reverse transcription polymerase chain reaction (RT–PCR) as a highly sensitive method for studying the expression of a few infecting Candida cells in a small amount of tissue. The complete absence of genomic DNA contamination in RT–PCR analysis was controlled by using primers for intron-containing genes. Furthermore, we chose house-keeping genes, such as the elongation factor gene EFB1 (Maneu et al., 1996), the inosine monophosphate dehydrogenase gene (Köhler et al., 1997) and the actin gene (Losberger and Ernst, 1989) of C. albicans, as internal mRNA controls for the evaluation of the sensitivity and efficiency of the RT–PCR analysis. EFB1 was found to be the best internal control with high levels of transcripts detected at all tested time points. Specific primers were used to amplify a 526 bp RT–PCR fragment of the EFB1 transcript, which does not contain an intron of 365 bp in size (Maneu et al., 1996). A 891 bp PCR fragment was amplified when genomic DNA was added. The amplification of this short mRNA sequence was used to verify the absence of contaminating genomic DNA and allowed the unambiguous detection of SAP gene transcripts.

Further controls for the absence of genomic DNA involved an additional PCR reaction in the absence of reverse transcriptase with selected RNA samples. No amplification product was detected in these control reactions (data not shown).

Finally, experiments were all carried out to examine the efficiency and the reproducibility of the RNA isolation and the RT–PCR method. Five independent samples of epithelial tissue infected at the same time with C. albicans were investigated after 48 h. Identical patterns of SAP gene expression were obtained (data not shown), indicating that the method was highly reproducible.

In order to examine the reproducibility of the model RHE infection, histological evaluations and RT–PCR investigations of the infected epithelia were carried out after 0, 12, 24, 36, 42, 48, 60, 72, 84 and 96 h in an initial experiment and at 12, 24, 36, 42 and 48 h in replicate experiments. Two of these experiments showed identical results with respect to histological alterations accompanying the progression of infection and the corresponding patterns of SAP gene expression. In a third experiment, the onset of histological alterations and the corresponding detection of SAP gene expression started 30 h earlier than above. Nonetheless, the histological alterations and the pattern of SAP gene expression during the course were equivalent to that observed before.

Morphology of uninfected reconstituted human epithelium (RHE)

The uninfected epithelial tissue consisted of well-stratified keratinocytes devoid of stratum corneum (Fig. 1). Dyskeratotic cells were seen only rarely. Incubation of uninfected reconstituted human epithelium (RHE) with phosphate-buffered saline (PBS) for up to 96 h led to an increased number of cell layers (not shown). This RHE can be readily maintained and is then strikingly similar to human oral mucosa in its structural organization, with a pH value (pH 7.0–7.4) corresponding to the normally neutral milieu of the oral cavity. Although no bacterial colonization of the RHE model is present, it can be seen as a reasonable model for the human oral mucosa.

Figure 1.

. Light micrographs of untreated reconstituted human epithelium (RHE) before infection with C. albicans. Stratified keratinocytes without stratum corneum (× 400).

Morphology of RHE after infection with C. albicans

We infected RHE and investigated the morphological alterations of the epithelial tissue by light microscopy. Compared with the uninfected control, there was no alteration of the epithelial in vitro model 0–36 h after infection with Candida albicans (SC5314), although several yeast cells adhered to superficial keratinocytes (Fig. 2). The uppermost keratinocytes showed slight intracellular oedema 42 h after infection (not shown). The structural alterations had increased in the following 6 h. By this time, clusters of yeast cells were detected on the superficial keratinocytes, which showed severe oedema and detachment (acantholysis) of the upper cell layers. In addition, vacuolization of the keratinocytes and margination of the nuclear chromatin were seen (Fig. 3). Furthermore, a pseudomembranous layer covering the superficial epithelium layer was detectable at this stage. The majority of this pseudomembrane was often separated from the mucosal equivalent during the fixation and embedding process. Histological evaluation of the pseudomembrane revealed necrotic and vacuolated keratinocytes and the presence of multiple yeast and, for the first time, hyphal cells (Fig. 4). Pseudomembranes were seen in all samples taken 48–96 h after inoculation. Severe acantholysis and epithelial atrophy with loss of several cell layers and invasion of Candida cells into the intercellular space was seen at the later stages of infection (60–96 h) (Fig. 5). Penetration of the epithelium cells by C. albicans cells was not observed. As the RHE was incubated under non-sterile conditions, a few airborne bacterial cells could be detected in a few samples.

Figure 2.

. RHE, 36 h after inoculation with C. albicans cells. Adhesion of some Candida cells to the superficial keratinocytes is seen. No marked morphological alterations are visible (× 400).

Figure 3.

. RHE, 48 h after infection. Clusters of Candida cells adhered to the superficial keratinocytes are seen, and severe oedema and detachment (acantholysis) of the upper cell layers has taken place. Keratinocytes show vacuolation (× 400).

Figure 4.

. Horizontal sectional view of the pseudomembranous layer after separation from the underlying RHE 48 h after infection with C. albicans cells. Multiple yeast and hyphal cells are seen with several severely damaged vacuolated necrotic keratinocytes (× 400).

Figure 5.

. RHE 60 h after infection showing mucosal erosion with severe acantholysis and oedema of the keratinocytes. Candida cells are adhering to the superficial keratinocytes and are found within the enlarged intercellular spaces. Intracytoplasmic lipid inclusions are visible in the lower epithelial cells (× 400).

Thus, while the presence of inflammatory cells cannot be expected in this in vitro model, several characteristic histological features of human oral candidosis could be demonstrated (Odds, 1988; Luna and Tortoledo, 1993).

SAP expression by C. albicans during the course of infection

As structural alterations of the RHE were observed during infection with C. albicans, we investigated the temporal expression of genes encoding secreted aspartyl proteinases during the course of infection. Total RNA was isolated from uninfected RHE and from samples at different times during the infection process. No mRNA transcript was amplified from the RNA of the uninfected RHE by RT–PCR for either the SAP- or EFB1-specific primers (not shown). In all infected samples investigated, RT–PCR fragments of 526 bp in size were amplified with EFB1-specific primers, indicating the cDNA origin of the templates (Figs 6A–C). After 12, 24 and 36 h, no amplification of SAP gene-specific fragments was detected, but controls indicated the presence of intact C. albicans mRNA, suggesting that SAP expression was truly silent (Fig. 6A). SAP1 and SAP3 expression were first detectable after 42 h. SAP6 expression was observed 6 h later (Fig. 6B), and SAP2 and SAP8 expression occurred after 60 h. This expression pattern was stable for the remaining period of investigation up to 96 h (Fig. 6C). RT–PCR performed in the absence of reverse transcriptase showed no SAP signals, verifying the absence of genomic DNA contamination (not shown). Reproducibility of the RT–PCR results was demonstrated in three independent experiments. In all cases, the expression of SAP genes was first detected at the same time as damage to the keratinocytes was first observed.

Figure 6.

. Analysis of RT–PCR products of RNA (lanes 2–8) from samples taken after 36 h (A), 48 h (B) and 96 h (C). A 526 bp fragment size obtained by amplification with EFB1 primers demonstrates cDNA origin (lane 9). Note the shift in 891 bp fragment size after PCR amplification of genomic DNA with the same primer pair (lane 10). Molecular mass marker pBR322 DNA/MvaI (M) (MBI Fermentas) with fragments of 1857, 1058, 929 and 383 bp (lane 1).

Immunoelectron microscopy

As the expression of SAP genes correlated with lesions of the in vitro model, we used immunoelectron microscopy to demonstrate the presence of Sap antigens and to study the distribution of the proteinases in the RHE.

In most of the infected tissue samples, yeast elements attached to keratinocytes were observed. Hyphal penetration of keratinocytes and Candida cells within keratinocytes were not found. For the detection of Sap immunoreactivity within C. albicans cells or keratinocytes, post-embedding labelling with Sap antibody conjugated to 10 nm gold particles was carried out. Gold labelling was observed to have a regular distribution within the cytoplasm of the Candida cells and within the damaged epithelial cells nearby after 48 h (Fig. 7). Control experiments without the addition of monoclonal antibody showed no non-specific gold labelling.

Figure 7.

. Post-embedding immunogold labelling with 10 nm gold particles in a sample of RHE taken 48 h after infection. Deposition of Sap antigen within the cytoplasm of a C. albicans cell and the damaged epithelial cell are seen. Magnification × 33 000. Bar = 1 μm.

SAP expression in clinical samples

We compared this pattern of gene expression observed in our in vitro model of oral candidosis with that of two patients with oral candidosis.

The first patient was a 29-year-old HIV-negative female outpatient who had shown an untreated whitish plaque on the oral mucosa for a few days. The second patient was a 44-year-old HIV-positive male outpatient who had suffered from oral candidosis for at least 1 year. In spite of continuous treatment with systemic antimycotic drugs at the time the sample was obtained, no improvement in the oral candidosis had been achieved.

A direct smear of the lesions was shock frozen for RNA isolation. Another smear was cultured, and C. albicans was identified microbiologically in both patients.

RT–PCR of RNA isolated from the first patient sample demonstrated expression of SAP1, SAP3 and SAP6. Using EFB1-specific primers, the amplification of a 526 bp product proved that only cDNA was present in DNase-treated extracts (Fig. 8A).

Figure 8.

. Analysis of RT–PCR products of RNA (lanes 2–8) from an HIV-negative patient with oral candidosis at an early stage (A) and an HIV-positive patient with chronic oral candidosis (B). Molecular mass marker pBR322 DNA/MvaI (M) (MBI Fermentas) with fragments of 1857, 1058, 929 and 383 bp (lane 1).

Therefore, the SAP gene expression pattern detected in the in vitro model 48 h after inoculation with C. albicans was identical to that in the in vivo situation at an early stage of oral candidosis.

RT–PCR analysis of the second patient showed the expression of SAP1, SAP2, SAP3 and SAP6 (Fig. 8B). Such an expression pattern resembles the later stages in our RHE model.

Discussion

Reconstructed human epidermis has been demonstrated to be useful for skin pharmacology experiments (Schaller et al., 1997) and for in vitro models of cutaneous candidosis in man (Korting et al., 1998; M. Schaller, unpublished). This type of human epidermis has provided useful insights into the pathogenesis and treatment of cutaneous candidosis. Experimental oral candidosis has been studied in several animal species (McMillan and Cowell, 1985; Odds, 1988; Anaissie et al., 1993; Allen et al., 1994), but the commonly used animal models incompletely mimic the situation in humans. From the ethical point of view, experiments involving animals are also more and more critically discussed. In this study, we investigated the utility of an in vitro model of oral candidosis based on reconstituted epithelium. Sufficient material can be obtained from this system for the molecular analysis of gene expression. The system resembles morphologically normal human oral mucosal epithelium, and the pathology of infection is also remarkably similar. In human oral candidosis, parakeratosis or complete absence of keratinization, epithelial atrophy, oedema, mucosal erosions with pseudomembrane formation and intraepithelial neutrophilic leukocyte infiltration are usually observed. The pseudomembrane consists of desquamated epithelium, leukocytes, necrotic tissue, bacteria and fungal cells (Odds, 1988; Luna and Tortoledo, 1993) and is formed at the base of the partially destroyed surface epithelium. Many of these elements were present in the RHE model. The invasion of keratinocytes by C. albicans is not known to occur during in vivo infections and was also not observed in the RHE model (Odds, 1988; Luna and Tortoledo, 1993). A few airborne bacterial cells were seen in the samples of the first experiment taken at 60–96 h after inoculation. In some experimental infections, tissue lesions occurred without bacterial contamination, indicating that the development of the morphological alterations was not caused by the bacteria.

The presence of inflammatory cells cannot be expected in the model, but several prominent histological features of acute pseudomembranous oral candidosis were found. The usefulness of our model is, of course, limited by the absence of a normal humoral and cell-mediated immune response. However, altered or depressed immunity is often a significant predisposing factor in human Candida infections and an important component of animal experimental studies (Odds, 1988; Anaissie et al., 1993).

Another benefit of the RHE in vitro model is the ability to sample the tissue at various stages during an infection for histological, immunoelectron microscopical examination and for gene expression studies.

During the first 36 h of experimental candidosis, SAP expression was not detected, although several C. albicans cells were observed adhering to superficial keratinocytes. This finding implies that Saps may not be involved in the initial stages of adhesion in our in vitro model. The possible absence of SAP signals resulting from the degradation of C. albicans mRNA was ruled out by the amplification of a 528 bp amplification product with EFB1-specific primers.

The development of the first visible lesions on the in vitro model epithelium was accompanied by the expression of SAP1 and SAP3. This suggests that Sap1 and Sap3 may play a role at this early stage of experimental oral candidosis. Extracellular Saps were detectable only 4 min after translation, as shown by pulse-chase experiments (Homma et al., 1993). Therefore, Saps seem to be regulated primarily at the transcriptional level and secreted directly after translation. Of course, we cannot exclude the possibility that other virulence factors may also aid in the development of lesions.

The increasing appearance of tissue lesions 48 h after infection coincided with the additional expression of SAP6. Hube et al. (1994) observed transcripts of SAP4–6 during germ tube formation at neutral pH. This is confirmed by our results, as additional SAP6 expression occurred at the same time as the first observation of yeast-to-hyphae transition. We found no expression of SAP4 or SAP5, which corresponds with the observation of White and Agabian (1995), who demonstrated that SAP6 mRNA is the predominant SAP transcript in vitro within the closely related SAP4–6 gene group.

Interestingly, no SAP2 transcript was detected in the early stages of infection, although SAP2 mRNA is the dominant transcript under several in vitro conditions (Hube et al., 1994; White and Agabian, 1995). SAP2 expression is known to be inhibited at pH values above 6.0 (Hube et al., 1994). The pH value of our oral candidosis model ranged between 7.0 and 7.4, corresponding to the normally neutral milieu of the oral cavity, and SAP2 transcripts were not seen at the early stage of infection. However, SAP2 expression was observed after 60 h, when some Candida cells were located within the enlarged intercellular spaces of the epithelium layers. Production of the Sap2 enzyme may occur here in acidic microniches between the keratinocytes, as proposed by Rüchel et al. (1991). This hypothesis is further supported by our observation that PHR2, a gene preferentially expressed at acidic pH values (Mühlschlegel and Fonzi, 1997), is expressed in the late stages of RHE infections (M. Schaller, W. Schäfer, H. C. Korting and B. Hube, unpublished data).

SAP8 mRNA transcripts were detected after 60 h. No concomitant specific morphological alterations in the C. albicans cells or the epithelial tissue were noted at this time. The reproducibility of these results was confirmed in two independent experiments. In a third experiment, lesion development and SAP gene expression started 30 h earlier, but showed an identical pattern of temporal progression of SAP mRNA synthesis. The earlier onset of the infection may have been caused by slight differences in culture conditions during the incubation process of C. albicans cells or by alterations in the RHE. Nevertheless, the patterns of SAP gene expression and its correlation with the lesion development of the epithelial tissue during the course of infection was confirmed.

Expression of the more distantly related members of the SAP gene family (SAP7 and SAP9 ) were not investigated in this study, as the function and regulation of these genes may be different from other members of the SAP gene family (Monod et al., 1994; Hube et al., 1994; White and Agabian, 1995; M. Monod, B. Hube, D. Hess and D. Sanglard, unpublished data).

The expression of secreted aspartyl proteinases during infection of the in vitro model by C. albicans was also confirmed by immunoelectron microscopy. Using a monoclonal antibody directed against Sap1–3, proteinase antigens were detected within C. albicans cells and in the epithelial cells. These antigens were detected 6 h after the initial amplification of SAP1 and SAP3 mRNA.

In vitro studies of C. albicans cultures by Hube et al. (1994); White and Agabian (1995) and Hube (1996) have demonstrated the differential regulation of SAP1–6 and 8 according to growth state, medium composition, temperature and cell morphology. As mentioned above, these studies demonstrated that a pH range between 3.0 and 6.0 was necessary for the expression of SAP2. The expression of SAP4–6 occurred only under conditions inducing hyphal growth, while the expression of SAP1–3 and 8 occurred during phenotypic switching of the C. albicans strain WO-1 in the opaque form at 25°C. This suggested that expression of the various SAP genes may play different roles in the survival of C. albicans in its natural environment and, thus, in the invasion of the host. Recently, animal experiments with SAP null mutants have shown indirectly that SAP1–3 and at least one of the closely related SAP4–6 genes are expressed during disseminated infections, and presumably contribute to the overall virulence of C. albicans (Hube et al., 1997a; Sanglard et al., 1997). SAP1 and SAP2 were shown to be expressed during experimental vaginal infection of rats with C. albicans and may be essential for this type of infection (De Bernardis et al., 1995). In our in vitro model of oral infection, the pattern of SAP gene expression differed in some respects from the results obtained by cell culture studies and in experimental vaginal infections. Our results showed a temporal pattern of expression beginning with the expression of SAP1 and SAP3 42 h after inoculation, followed by SAP6, SAP8 and, finally, SAP2 transcripts as the infection processes. This programme of SAP gene expression in our in vitro model is presumably related to progressive changes in environmental conditions, such as the availability of Sap-inducing substrates, the pH in microniches and the development of hyphae of C. albicans during the infection process.

The usefulness of different in vitro models can be evaluated by the extent to which each can mimic the in vivo situation. We therefore have performed RT–PCR analysis of two different patient samples. The first clinical sample from an untreated oral candidosis at an early stage showed expression of SAP1, SAP3 and SAP6 corresponding to the initial expression pattern of RHE infection. The analysis of a second clinical sample from a chronic oral infection revealed the additional expression of SAP2, which resembles the later stages of our experimental candidosis. This is the first direct demonstration of SAP gene expression in human oral candidosis in vivo, suggesting that secreted aspartyl proteinases do play a role in oral Candida infections. The specific pattern of gene expression found in oral candidosis further supports the relevance of our in vitro model of oral candidosis.

In summary, these results illustrate not only a correlation between distinct SAP expression patterns and epithelial lesion in an in vitro model of mucosal infections, but also provide evidence for a role of Sap isoenzymes in human oral candidosis.

Experimental procedures

Candida strain

The clinical Candida albicans isolate designated SC5314 was used for this study (Gillum et al., 1984).

Culture media and growth conditions

Inocula were prepared by culturing yeast cells for 24 h at 37°C on Sabouraud dextrose agar (Difco). A sample of the culture was washed three times in 0.9% NaCl. A sample of approximately 2 × 105 cells was then suspended in 10 ml of modified Sabouraud dextrose medium [1.0 g of mycological peptone (Oxoid) and 0.2 g of dextrose in 1.0 l of dH2O]. The suspension was cultured for 16 h at 25°C with orbital shaking. A suspension of 4 × 106 cells was incubated with shaking in fresh medium for 24 h at 37°C. After washing three times with phosphate-buffered saline (PBS), the final inoculum was then adjusted to the desired density with PBS solution.

Reconstituted human epithelium (RHE)

The human epithelium for the in vitro model of oral candidosis was supplied by Skinethic Laboratory. It was obtained by culturing transformed human keratinocytes derived from squamous cell carcinoma of the buccal mucosa of the cell line TR146 on an inert supporting membrane (Rupniak et al., 1985). Cultures were incubated in serum-free conditions in a defined medium based on the MCDB-153 medium (Clonetics) containing 5 μg ml−1 insulin on a 0.5 cm2 microporous polycarbonate filter for 7 days. The in vitro model and all culture media were prepared without antibiotics and antimycotics.

Three replicate infection experiments were performed. RHE was infected with 2 × 106C. albicans yeast cells in 50 μl of PBS. Controls contained 50 μl of PBS alone. Inoculated and uninoculated cultures were incubated at 37°C with 5% CO2 at 100% humidity for 12, 24, 36, 42, 48, 60, 72, 84 and 96 h. The medium was changed every 24 h.

Light microscopy

A part of each specimen was fixed with 2.5% glutaraldehyde and 2.0% formaldehyde in a 0.05 M cacodylate-buffered solution at pH 7.3 (Karnovsky, 1965). Specimens were post-fixed in 1% osmium tetroxide in 0.1 M cacodylate buffer at pH 7.3 at room temperature (Dalton, 1955). After several washing and dehydration procedures for 1 h at room temperature, the specimens were embedded in glycide ether. The small blocks of tissue were cut using an ultramicrotome (Ultracut; Reichert). Semi-thin sections (1 μm) were studied with a light microscope after staining with 1% toluidine blue and 1% pyronine G (Merck). The sections were viewed at a magnification of × 400.

Immunoelectron microscopy

Post-embedding immunogold labelling was carried out for the intracellular detection of Sap antigen in an infected sample taken after 48 h. A part of each specimen was fixed in Karnovsky solution for 30 min at room temperature and embedded in glycide ether. Sections (80–100 nm thick) were mounted on nickel grids. The grids were rinsed on drops of dH2O for 10 min and floated on drops of PBS containing 5% (v/v) normal goat serum for 2 × 10 min. Grids were then incubated with the anti-Sap monoclonal mouse IgG antibody FX 7-10 (Ollert et al., 1995) diluted 1:50 in PBS supplemented with 0.1% BSA and 0.05% Tween 20 (PBS-BT) for 14 h at 4°C. FX 7-10 was shown to react strongly with Sap1, Sap2 and Sap3 but not with Sap4, Sap5 and Sap6 (M. Borg-von Zepelin, unpublished data).

After washing the grids repeatedly with PBS-BT, they were incubated with 10 nm gold-conjugated goat anti-mouse IgG (Auroprobe EM Immunogold reagents; Amersham) diluted 1:25 in TBS-BT (0.02 M Tris-HCl buffer, 0.15 M NaCl, 0.015 M sodium azide, 0.1% BSA, 0.05% Tween 20, adjusted to pH 8.2) for 1 h at room temperature. In control samples, the anti-Sap1–3 monoclonal antibody was omitted. After several washing steps with TBS-BT, grids were fixed with 2% glutaraldehyde and washed again in dH2O. They were stained with 0.5% uranyl acetate for 10 min and 2.7% lead citrate for 5 min (Ultrastainer; LKB) at 20°C. Grids were examined using a Zeiss EM 902 transmission electron microscope (Zeiss) operating at 80 kV, at magnifications between × 3000 and × 85 000.

In vivo biopsies

Samples of pseudomembrane were removed from the oral mucosa of a volunteer HIV-negative female patient and an HIV-positive male patient. The first patient was 29 years old and had been suffering from an untreated pseudomembraneous oral candidosis for a few (2–3) days. The HIV-infected patient was 44 years old and had been exhibiting clinical signs characteristic of an oral candidosis for at least 1 year. Despite systemic antimycotic treatment, no improvement in the infection was achieved. Parts of the clinical material from both patients were used for microbiological culture and biochemical characterization. For RT–PCR investigation, material was also immediately shock frozen in liquid nitrogen and stored at −20°C.

Mycological examination

The specimens obtained from the patients were inoculated on Kimoleg's agar (Merck) and incubated for 72 h at 37°C. Biochemical identification of C. albicans was based on the use of the ready-made system ATB 32 C (API System; Bio Mérieux) (Caniaux et al., 1985).

RNA isolation

Total RNA from the infected and uninfected epithelia and from the clinical samples was prepared as described by Hube et al. (1994). Frozen cells were resuspended in 0.5 ml of extraction buffer (0.1 M Tris-HCl, pH 7.5, 0.1 M LiCl, 0.01 M dithiothreitol) containing 0.5 g of glass beads (425–600 microns), 17 μl of 20% SDS, 0.5 ml of phenol/chloroform (50:50, saturated in 1 M Tris-HCl, pH 7.5) and vortexed for 10 min to break the cells. After being spun for 3 min at 3000 g, the aqueous phase was removed and extracted twice in phenol/chloroform. The RNA was precipitated with two volumes of 96% ethanol and stored at −20°C. Precipitated RNA was pelleted and resuspended in 4 μl of 10 mM Tris-HCl, pH 7.5, 1 mM EDTA and 1 μl of RNase inhibitor (Gibco BRL; Life Technologies).

cDNA synthesis (RT)

After DNase I (Gibco BRL; Life Technologies) treatment according to the manufacturer's instructions, annealing was performed with oligo (dT) primer (Gibco BRL; Life Technologies) at 70°C for 10 min, and the product was chilled on ice. The reaction mixture was made up to a final volume of 20 μl, containing RT buffer (50 mM Tris, pH 8.3, 75 mM KCl, 3 mM MgCl2), 10 mM dithiothreitol, 10 μM desoxynucleoside triphosphate (dNTP) and 200 U of Superscript II reverse transcriptase (Gibco BRL; Life Technologies). cDNA synthesis was carried out at 42°C for 1 h. The reaction was then stopped by heating for 15 min at 70°C. Finally RNase H (Gibco BRL; Life Technologies) treatment was carried out for 20 min at 37°C. To examine for the absence of genomic DNA contamination, cDNA synthesis was also performed without the addition of reverse transcriptase.

Primers

For the detection of gene expression, the following pairs of primers were used: 5′-AGGGAAAGGTATTTACACT-3′ and 5′-GATTTGCTTACATAGTAAGTAC-3′ for SAP1; 5′-TGATTGTCAAGTCACTTATAGT-3′ or 5′-CCTAAAGCATTCCCAGGTTAC-3′ and 5′-CTTAGGTCAAGGCAGAAATACTG-3′ for SAP2; 5′-TGGATTGGAACATTTCTAATTC-3′ and 5′-CAATCTCCAGAGGAGT ACTTCC-3′ for SAP3; 5′-TTTTCATTAACAACCAACCATTC-3′ and 5′-GTCCTGGTGGCTTCGTTGC-3′ for SAP4; 5′-ATAATTAATCTAAAGTCAAA-GTTC-3′ and 5′-CAATCTCCAGAGGAGTACTTCC-3′ for SAP5; 5′-TTCTTCAAACGTTTTAATTCTCT-3′ and 5′-CATAAATGACTTCAAAATATAAAT-3′ for SAP6; and 5′-CTCTATAAAGTAGAAATACTTGA-3′ and 5′-GTTGACACAGG-TTCTTCTG-3′ for SAP8. For an internal mRNA control and for the detection of genomic DNA, we designed a primer pair to identify and amplify a 891 bp fragment of the EFB1 gene (elongation factor), which encodes EF-1β and contains a 365 bp intron (Maneu et al., 1996). The presence of the intron allowed the detection of amplified signals arising from genomic and cDNA origins. Primers used for EFB1 expression were 5′-ATTGAACGAATTCTTGGCTGAC-3′ and 5′-CATCTTCTTCAACAGCAGCTTG-3′.

PCR

A single PCR protocol was used with all nine primer sets. The samples were subjected to 40 cycles (denaturation for 3 min at 94°C, annealing for 3 min at 50°C and extension for 10 min at 72°C).

Acknowledgements

The authors thank E. Januschke (Ludwig-Maximilians-University, Munich, Germany) for excellent technical assistance and M. Borg-von Zepelin (University of Göttingen, Germany) for providing the monoclonal Sap antibody; N.A.R. Gow (University of Aberdeen, UK) for critical reading of the manuscript. This work was supported by grants from the Theodor-Nasemann-Stipendium and the Manfred-Plempel-Stipendium to M.S.

Ancillary