Potential role for a carbohydrate moiety in anti-Candida activity of human oral epithelial cells

Abstract

Candida albicans is both a commensal and a pathogen at the oral mucosa. Although an intricate network of host defense mechanisms are expected for protection against oropharyngeal candidiasis, anti-Candida host defense mechanisms at the oral mucosa are poorly understood. Our laboratory recently showed that primary epithelial cells from human oral mucosa, as well as an oral epithelial cell line, inhibit the growth of blastoconidia and/or hyphal phases of several Candida species in vitro with a requirement for cell contact and with no demonstrable role for soluble factors. In the present study, we show that oral epithelial cell-mediated anti-Candida activity is resistant to gamma-irradiation and is not mediated by phagocytosis, nitric oxide, hydrogen peroxide, and superoxide oxidative inhibitory pathways or by nonoxidative components such as soluble defensin and calprotectin peptides. In contrast, epithelial cell-mediated anti-Candida activity was sensitive to heat, paraformaldehyde fixation, and detergents, but these treatments were accompanied by a significant loss in epithelial cell viability. Treatments that removed existing membrane protein or lipid moieties in the presence or absence of protein synthesis inhibitors had no effect on epithelial cell inhibitory activity. In contrast, the epithelial cell-mediated anti-Candida activity was abrogated after treatment of the epithelial cells with periodic acid, suggesting a role for carbohydrates. Adherence of C. albicans to oral epithelial cells was unaffected, indicating that the carbohydrate moiety is exclusively associated with the growth inhibition activity. Subsequent studies that evaluated specific membrane carbohydrate moieties, however, showed no role for sulfated polysaccharides, sialic acid residues, or glucose- and mannose-containing carbohydrates. These results suggest that oral epithelial cell-mediated anti-Candida activity occurs exclusively with viable epithelial cells through contact with C. albicans by an as-yet-undefined carbohydrate moiety.

FIG. 1

No evidence of phagocytosis in oral epithelial cell-mediated anti-Candida activity. Whole unstimulated saliva was collected from healthy human volunteers (n = 3), and epithelial cell-enriched populations were isolated by nylon membrane retention. Enriched epithelial cells were then examined for in vitro growth inhibition of C. albicans. Aliquots from nonradioactive cultures were cytospun onto slides and stained with 1% diaethanol for analysis of phagocytosis. The figure illustrates a representative preparation viewed under light (A) and fluorescent (B) microscopy. Magnification, ×400.

FIG. 2

No evidence for nonoxidative soluble factors in oral epithelial cell-mediated anti-Candida activity. Whole unstimulated saliva was collected from healthy human volunteers (n = 3), and epithelial-enriched populations were isolated by nylon membrane retention. Enriched epithelial cells were then examined for in vitro growth inhibition of C. albicans in medium containing Ca²⁺ and Mg²⁺ (control), Ca²⁺- and Mg²⁺-free medium (for evaluation of defensins), and medium containing excess Zn²⁺ (for evaluation of calprotectin). The figure shows representative results from three separate experiments.

FIG. 3

Physical characteristics of oral epithelial cell-mediated anti-Candida activity. Whole unstimulated saliva was collected from healthy human volunteers (n = 3), and epithelial cell-enriched populations were isolated by nylon membrane retention. Enriched epithelial cells were pretreated with various doses of gamma-irradiation, 1% paraformaldehyde, or heat (65°C) and then washed and examined for in vitro growth inhibition of C. albicans by measuring [³H]glucose uptake. The figure shows cumulative results (mean % inhibition ± the standard error of the mean [SEM]). Asterisks represent significant differences compared to controls (P < 0.05).

FIG. 4

Protein moieties are not involved in oral epithelial cell-mediated anti-Candida activity. Whole unstimulated saliva was collected from healthy human volunteers (n = 3), and epithelial cell-enriched populations were isolated by nylon membrane retention. Enriched epithelial cells were pretreated with SDS (A) and NP-40 (B). Thereafter, cells were washed and examined for in vitro growth inhibition of C. albicans by measuring [³H]glucose uptake. The figure shows show cumulative results (mean % inhibition ± the SEM). Asterisks represent significant differences compared to controls (P < 0.05).

FIG. 5

No role of membrane protein moieties in oral epithelial cell-mediated anti-Candida activity. Whole unstimulated saliva was collected from healthy human volunteers (n = 3), and epithelial cell-enriched populations were isolated by nylon membrane retention. Enriched epithelial cells were pretreated with various concentrations of proteinase K in the presence or absence of cycloheximide. Thereafter, cells were washed and examined for in vitro growth inhibition of C. albicans by [³H]glucose uptake. The figure shows representative results for three separate concentrations of proteinase K, proteinase K (250 μg/ml) in the presence of cycloheximide (CHX; 100 μg/ml), and cumulative results of controls for each concentration of proteinase K employed (mean % inhibition ± the SEM).

FIG. 6

Role of membrane carbohydrate moieties in oral epithelial cell-mediated anti-Candida activity. Whole unstimulated saliva was collected from healthy human volunteers (n = 3), and epithelial cell-enriched populations were isolated by nylon membrane retention. Enriched epithelial cells were pretreated with periodic acid and thereafter washed and examined for in vitro growth inhibition of C. albicans by quantitative plate counts. The figure shows cumulative results (mean % inhibition ± the SEM) at an E:T ratio of 5:1. Asterisks represent significant differences compared to controls (P < 0.05).

FIG. 7

Periodic acid treatment of epithelial cells does not affect adherence to C. albicans. Whole unstimulated saliva was collected from healthy human volunteers (n = 3), and epithelial cell-enriched populations were isolated by nylon membrane retention. Enriched epithelial cells were pretreated with periodic acid and then washed and examined for adherence to oral epithelial cells by using a standard adherence assay. The figure shows a representative preparation of PBS-treated cells (A) and periodic acid-treated cells (B) viewed under light microscopy. Magnification, ×400.