Divergent EGFR/MAPK-Mediated Immune Responses to Clinical Candida Pathogens in Vulvovaginal Candidiasis

Abstract

Vulvovaginal candidiasis (VVC) is characterized by symptomatic inflammatory responses in the vagina caused by Candida albicans and non-albicans Candida (NAC) species. The epidermal growth factor receptor (EGFR) -mitogen-activated protein kinase (MAPK) signaling pathway has been linked to immune responses of oral mucosa after C. albicans exposure, but whether this pathway plays a similar response in vaginal epithelial cells is not known. Here, we observed that phosphorylation of EGFR and p38 was continuously activated in vaginal epithelial cells by C. albicans strain SC5314. This differs markedly from oral epithelial cells, which respond in a biphasic manner in order to properly discriminate the morphology of C. albicans. When compared with SC5314, a highly azole-resistant C. albicans isolate 1052 can induce a stronger phosphorylated signal of EGFR and p38, while clinically-isolated NAC strains including C. tropicalis, C. glabrata, C. parapsilosis and C. auris trigger higher levels of phosphorylated ERK1/2 and c-Fos than C. albicans. Inhibition of EGFR significantly reduces inflammatory response and epithelial damage induced by C. albicans both in vitro and in vivo, while inhibition of p38 leads to significant repair of epithelial damage triggered by both C. albicans and NAC species. These results confirm the importance of the EGFR-MAPK signaling in VVC pathogenesis and highlight the remarkable immunogenic differences between C. albicans and NAC species in host-microbe interactions.

Keywords: Candida albicans; EGFR; MAPK; non-albicans Candida species; vaginal epithelial cells; vulvovaginal candidiasis (VVC).

Figure 1

C. albicans activates EGFR, NF-κB, MAPK and c-Fos signaling in vaginal epithelial cells. Pathway enrichment analysis and heatmaps of VK2/E6E7 vaginal epithelial cells stimulating with SC5314 and PBS. Using the phyper function in the R software to perform enrichment analysis and calculate the P value. Then performing FDR correction on the P value to obtain the Q value. A Q value < 0.05 is regarded as a significant enrichment. Red underlines highlight the upregulated pathways related to EGFR and MAPK. Heat map displaying the upregulated(red) or downregulated(blue) genes. Each column represents an individual sample(n=3) (A). VK2/E6E7 cells were stimulated with SC5314 (MOI=5) for the indicated time. Cell lysates were analyzed by immunoblotting for the indicated proteins. The phosphorylated forms of proteins are preceded by “p” (B). SC5314 elicited significantly stronger phosphorylation of p38 in EGFR-Ad-VK2, VK2/E6E7 cells overexpressing EGFR, when compared to VK2/E6E7 cells. EGFR inhibitor (AG1478) suppressed SC5314-induced phosphorylation of ERK1/2 and p38 in EGFR-Ad-VK2 and VK2/E6E7 cells (C). Female ICR mice (n=3 or 2/each group) were infected with SC5314 or clinical isolate1052 after pre-treatment with DMSO or AG1478 for 2d and vaginal tissues were harvested 1 d after infection for detection of EGFR phosphorylation by immunoblotting analysis (D). Data are representative of three independent experiments and bands are shown relative to β-tubulin loading control.

Figure 2

Functional role of EGFR and MAPK signaling in regulating vaginal epithelial inflammatory responses. Quantification of cytokines were assessed by multiplex microbead assay (luminex), secreted from VK2/E6E7 vaginal epithelial cells stimulating with SC5314(MOI=0.01) for 24 h. EGFR inhibitor AG1478, p38 inhibitor SB203580 or ERK1/2 inhibitor SD5978 was added to the cells 2 h prior to fungal stimulation. All these inhibitors suppressed SC5314-induced secretion of cytokines (CCL20, G-CSF, GM-CSF, IL-1β, IL-6, IL-17A) (A). EGFR-Ad-VK2 released higher levels of cytokines (G-CSF, GM-CSF, IL-1β, IL-6, IL-17A) induced by SC5314 when compared to VK2/E6E7 cells, which can be suppressed by AG1478 (B). VVC mice were treated as previously ( Figure 1D ). Vaginal sections from C. albicans–infected mice were assayed for immunofluorescence staining of Ly-6G (neutrophil marker, green) and nuclei were counterstained with DAPI (blue) (C). Numbers in graph (D) provide the mean neutrophils count per group (n=3/each group). Fungal load in vaginal tissues was assessed by periodic acid-Schiff (PAS). Insets show regions of vaginal fungal burden and higher-magnification images of the organisms indicated by the arrows (E). Numbers in graph (F) provide the mean C. albicans count per group (n=3/each group). Data are representative of three independent experiments. Error bars represent SD, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Figure 3

Differential activation of EGFR and MAPK and cytokine induction by clinical isolate 1052 and WT strain SC5314. VK2/E6E7 cells were stimulated with SC5314 and clinical isolate 1052 (MOI=5) for the indicated time respectively. Phosphorylation of EGFR, MAPK proteins, NF-κB (p65) and c-Fos was detected by immunoblotting of cell lysates (A). 1052 induced higher levels of cytokines secretion (CCL20, G-CSF, GM-CSF, IL-1β, IL-6, IL-17A) than SC5314. All inhibitors (EGFR, p38 and ERK1/2 inhibitor) suppressed 1052-induced secretion of CCL20, G-CSF, GM-CSF, IL-1β, IL-6 and IL-17A (B). Infection with 1052 in the vaginal mucosa of mice induced significantly increased neutrophil recruitment at the infection foci when compared with SC5314 and pre-treatment of AG1478 resulted in significantly decreased neutrophil recruitment (C). Numbers in graph (D) provide the mean neutrophils count per group (n=3/each group). Morphology of SC5314 and clinical isolate 1052 adhering to VK2/E6E7 cells at 0, 1, 2, 4, 6, 8, 10 h post infection (E), gently washed with PBS for at least 3 times. Data are representative of three independent experiments. Error bars represent SD, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Figure 4

Divergent activation of EGFR and MAPK and cytokine induction by C. albicans and non-albicans Candida species. VK2/E6E7 cells were stimulated with C. albicans (SC5314, 1052, ATCC90028 and C1-14) and non-albicans Candida (NAC) isolates C. tropicalis, C. glabrata, C. parapsilosis, CBS10913 and CBS14918) (MOI=5) respectively. Protein lysates were taken at 6 h post-infection for western blotting analysis (A). Cytokines induced by C. albicans and NAC isolates (MOI=0.01) were assessed at 24h post-infection via Luminex (B). All inhibitors suppressed NAC isolates-induced cytokines production (C). Morphology of C. tropicalis, C. glabrata, C. parapsilosis and CBS14918 strains on VK2/E6E7 cells at 6 and 24 h post infection (D). Data are representative of three independent experiments. Error bars represent SD, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Figure 5

EGFR and MAPK signaling regulate the viability of vaginal epithelial cells after infection with C. albicans. Cell damage was monitored continuously by xCELLigence Real-Time Cellular Analysis (RTCA) system. Normalized Cell Index (NCI) demonstrated cell viability. 1052-infected VK2/E6E7 cells exhibited a significant decrease in viability when compared to SC5314 infections and pre-treatment of the EGFR and p38 inhibitors protected cells against SC5314 or 1052-killing (A, C). SC5314-infected EGFR-Ad cells exhibited increased damage when compared to the infected VK2/E6E7 cells (B). NCI at 24 h post infection in graph (D–F) provides the mean cell viability. Green arrow represents the time points on which the medium was changed. Blue and red arrows represent the time point at which inhibitors and Candida strains were added into the well respectively. VVC mice were treated as previously ( Figure 1D ). Immunofluorescent staining of E-cadherin (green), caspase-3(red) and DAPI (blue) demonstrated apoptotic vaginal epithelial cells in vaginal tissues sections (G). Numbers in graph (H) provide the mean apoptotic vaginal epithelial cells count per group (n=3/each group). Data are representative of three independent experiments. Error bars represent SD, *p < 0.05, Error bars represent SD, *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 6

EGFR and MAPK signaling regulate the viability of vaginal epithelial cells after infection with non-albicans Candida species. Cells damage caused by SC5314 and non-albicans Candida (NAC) isolates were monitored continuously for 24 h (A). Only pre-treatment of p38 inhibitor (SB203580) exhibited a significant increased protective capacity against NAC isolates compared to EGFR and ERK1/2 inhibitor (B–E). NCI at 24 h post infection in graph (F, G) provides the mean cell viability. Green arrow represents the time points at which the medium was changed. Blue and red arrows represent the time point on which inhibitors and Candida strains were added into the well respectively. Data are representative of three independent experiments. Error bars represent SD, *p < 0.05.

Figure 7

Activation of the EGFR-MAPK signaling pathway in vaginal epithelial cells by Candida pathogens. C. albicans infection on vaginal epithelial cells manifests by a modest yeast-to-hyphal transition within the first 2 hours, when the EGFR-ERK1/2 signaling pathway (small red arrow) is moderately activated. At a later stage of stimulation (4 to 10 hours), after the formation of a greater number of hyphae, the EGFR-p38 signaling pathway (thick red arrow) is strongly activated. Activated EGFR-p38 and-ERK1/2 signaling pathways lead to an increased phosphorylation of c-Fos and p65 and an increased secretion of cytokines (IL-1, IL-6, G-CSF, GM-CSF, CCL20). In particular, the EGFR-p38 signaling pathway also moderates epithelial damage induced by C. albicans. While non-albicans Candida (NAC) spp. infections of vaginal epithelial cells rarely produce hyphae, they activate a strong phosphorylation of ERK1/2, c-fos and p65 and a subsequent secretion of cytokines. Although not strongly activated, the p38 pathway still has some effect in the moderation of epithelial damage.