Zinc prevents vaginal candidiasis by inhibiting expression of an inflammatory fungal protein

Abstract

Candida causes an estimated half-billion cases of vulvovaginal candidiasis (VVC) every year. VVC is most commonly caused by Candida albicans, which, in this setting, triggers nonprotective neutrophil infiltration, aggressive local inflammation, and symptomatic disease. Despite its prevalence, little is known about the molecular mechanisms underpinning the immunopathology of this fungal infection. In this study, we describe the molecular determinant of VVC immunopathology and a potentially straightforward way to prevent disease. In response to zinc limitation, C. albicans releases a trace mineral binding molecule called Pra1 (pH-regulated antigen). Here, we show that the PRA1 gene is strongly up-regulated during vaginal infections and that its expression positively correlated with proinflammatory cytokine concentrations in women. Genetic deletion of PRA1 prevented vaginal inflammation in mice, and application of a zinc solution down-regulated expression of the gene and also blocked immunopathology. We also show that treatment of women suffering from recurrent VVC with a zinc gel prevented reinfections. We have therefore identified a key mediator of symptomatic VVC, giving us an opportunity to develop a range of preventative measures for combatting this disease.

Figure 1. PRA1 zinc scavenger gene expression correlates with inflammatory cytokine concentrations during Candida albicans vaginal colonisation and infection in humans.

(A to D) Vaginal swabs from uninfected (n = 12) and C. albicans infected or colonised (n = 17) women were assessed for pH (A), calprotectin (B), IL-1β (C), and IL-8 (D). (E) Pearson correlation matrix of the C. albicans-infected samples showing the relationships between host parameters (displayed in A-D) and expression of the indicated fungal genes as determined by qRT-PCR (2^-ΔΔCt); the mean RNA (pg/ml) of two housekeeping genes (HKG), ACT1 and CEF3, was used as an indication of fungal burden. (F & G) Correlations between PRA1 expression in women and cytokines IL-1β (F) and IL-8 (G) (r values from E and p values are displayed on graph). (H) PRA1 expression in response to low zinc during vaginal epithelial tissue culture infection at both neutral and acidic pH. C. albicans (clinical isolate SC5314) was incubated on A-431 vaginal epithelial cells for 16h in RPMI at pH 7 or adjusted to pH 5, with or without 25 μM ZnSO₄. Whiskers show minimum to maximum of all points. (I) Model showing the transcriptional control of PRA1 in C. albicans. Statistical tests. A-D, Mann-Whitney test; E, Pearson r correlation displaying r values. Positive correlations are highlighted in magenta and negative in cyan. Significant (p < 0.05) r values are underlined in the matrix. H, t-test between vehicle and zinc treated samples. * p < 0.05; ** p < 0.01; **** p <0.0001.

Figure 2. PRA1 is essential for neutrophil attraction.

(A) Schematic overview of the neutrophil chemotaxis assay. Candida culture filtrate was added to the lower compartment of the chemotaxis plate. Fluorescently labelled neutrophils were added to the upper compartment and chemotaxis assessed by measuring fluorescence in the lower compartment following migration of neutrophils through the membrane. (B) C. albicans wild type (Wt), pralΔ and pra1Δ+PRA1 strains were incubated for 5 days in RPMI tissue culture media after which PRA1 transcription was assessed by qRT-PCR (n = 3). (C) PRA1 involvement in neutrophil chemotaxis. After the five-day incubation period, culture filtrate, unconditioned media (RPMI) or media containing neutrophil chemotactic factor IL-8 (100 ng/ml) were added to the lower compartment of the chemotaxis plate. Freshly isolated human neutrophils were fluorescently labelled with Calcein and added to the upper compartment. After 2 h incubation, neutrophil chemotaxis was determined by measuring the fluorescence intensity (at 485/530 nm) in the lower compartment (n=3). (D) As an alternative model, individual C. albicans yeast cells (~20 cells per well) were incubated in ibidi-VIII well microscopy slides in RPMI culture media for 10 h to induce microcolony development. Human neutrophils were added and incubated for a further 3 h. Samples were fixed and fungi stained with Calcofluor white (blue) and neutrophils with Sytox Green. Forty individual microcolonies per strain (from two independent experiments) were imaged and the number of neutrophils per field enumerated. Representative micrographs show the greater number of neutrophils in proximity to PRA1⁺ C. albicans microcolonies. Scale bar 150 μm. Statistical tests. B-C, one way ANOVA with Tukey post hoc test (excluding positive and negative control in C). D, Kruskal-Wallis test. Error bars show SEM. * p < 0.05; ** p < 0.01; *** p < 0.001; **** p <0.0001.

Figure 3. Candida PRA1 genetic status influences neutrophil attraction.

(A) The PRA1 gene has been lost by certain Candida species including C. glabrata. A phylogenetic tree was drawn based on (Munoz 2018 Nat Comm) (17) and the presence or absence of PRA1 orthologues determined by BLASTp. (B) Vaginal swabs from uninfected, C. albicans-infected, and C. glabrata-infected women were evaluated for neutrophil infiltrate. n=12 samples per group. (C) PRA1 neutrophil chemotaxis assay. C. albicans PRA1 from was codon-optimised for expression in C. glabrata, placed downstream of the C. glabrata GPD2 promoter and transformed into C. glabrata BG2. Conditioned culture filtrate of C. albicans and C. glabrata strains with and without PRA1 were used in a chemotaxis assay as outlined in Figure 2. Unconditioned media (RPMI), media containing neutrophil chemotactic factor IL-8 (100 ng/ml), or indicted culture filtrates were added to the lower compartment of the chemotaxis plate. Freshly isolated human neutrophils were fluorescently labelled with Calcein and added to the upper compartment. After 2 h incubation, neutrophil chemotaxis was determined by measuring fluorescence intensity (at 485/530 nm) in the lower compartment. n = 3. Statistical tests. B and C, One-way ANOVA with Tukey post hoc test. ** p < 0.01; *** p < 0.001; **** p < 0.0001.

Figure 4. Deletion of PRA1 prevents inflammation in a mouse model of vulvovaginal candidiasis.

(A-D) Mice were vaginally infected with C. albicans wild type (Wt) or PRA1-deletion mutant (pra1Δ). 24 h post-infection, vaginal lavage fluid was collected and assessed for the presence of the proinflammatory cytokines IL-1β (A) and CxCL-2 (which has similarities to human IL-8; B) by ELISA. The presence of neutrophil (PMN) infiltration by flow cytometry (C) and fungal burden by plating colony forming units (CFU, D) were also assessed. Statistical tests. A to C, One way ANOVA with Tukey post hoc test. D, unpaired t-test. Error bars = SEM. * <0.05; ** p < 0.01 ns = not significant.

Fig. 5. Zinc down-regulates PRA1 and prevents vaginal candidiasis.

Mice were vaginally infected with C. albicans wild type and administered with vehicle control or with zinc sulphate (+Zn²⁺, 10 μl at 25 μM) at the time of infection and 8 h post-infection. After 24 h, vaginal lavage fluid (red circles, A, D-F) or organs (grey circles, B-C, G-K) were collected. (A to C) Fungal burden was determined by measuring colony forming units (A) or expression of housekeeping genes ACT1 and CEF3 (B); PRA1 expression was evaluated by qRT-PCR (C). (D to F) Luminal cytokines (IL-1β and CxCL-2, D-E) and neutrophils (PMN, F) were determined by ELISA and flow cytometry, respectively. (G to K) Vaginal tissue-associated expression of proinflammatory cytokines IL-1β (G), CxCL-2 (H), CxCL-1 (I), CxCL-5 (J), and IL-6 (K) were measured by qRT-PCR. (L) C. albicans clinical isolate SC5314 was incubated for five days in tissue culture medium with or without zinc supplementation and PRA1 gene expression measured by qRT-PCR. (M) Culture filtrates were added to the lower well and calcein-labelled human neutrophils to the upper well of the chemotaxis system and neutrophil migration assessed by measuring fluorescence at 485/530 nm in the lower compartment after 2 h incubation. (N) In a retrospective study, eight women who had been suffering from recurrent vaginal infections used a zinc (20 μM) containing gel. Rates of reinfection for recurrent bacterial vaginitis (RBV n=2) and for recurrent vulvovaginal candidiasis (RVVC) were assessed following the three month study. Statistical tests. A-K, Mann-Whitney test. L and M, unpaired t-test; N, Paired t-test. Error bars = SEM. * p < 0.05; ** p < 0.005, *** p < 0.0005, **** p < 0.0001.