Interferon-stimulated gene 15 expression negatively regulates resistance to Candida albicans infection in vulvovaginal epithelial cells

Abstract

Vulvovaginal candidiasis (VVC) is one of the most common fungal infections, affecting most women during their reproductive years. A core characteristic of VVC is the interplay between Candida albicans pathogenicity and dysregulated innate immune responses. While these inflammatory responses generally underlie VVC severity, epithelial type I interferon (IFN) responses have been associated with improved epithelial resistance to infection and dampened downstream neutrophil activation. Yet, the role of interferon-stimulated genes (ISGs) in epithelial resistance to C. albicans infection remains unclear. Here, we investigated the threshold for mounting type I IFN responses in A-431 vulvovaginal epithelial cells (VECs), and the role of interferon-stimulated gene 15 (ISG15) in epithelial resistance and inflammation. We found that increasing C. albicans burdens corresponded to increasing epithelial cytotoxicity and the release of neutrophil chemoattractant IL-8. Further, expression of IFNB1 and ISGs, specifically ISG15 and MX2, was induced at fungal burdens, before the induction of significant cytotoxicity at later time points. We observed intracellular ISG15 accumulation following C. albicans infection, yet detected reduced unconjugated intracellular ISG15, suggesting its conjugation to other proteins. Mechanistically, we show that ISG15 silencing reduced epithelial cytotoxicity and IL-8 responses to C. albicans infection. Concurrently, a C. albicans infection-specific downregulation of the gene encoding the cell death regulator Z-DNA binding protein 1 (ZBP1) was observed. Accordingly, ZBP1 silencing, similar to ISG15, exhibited reduced tissue damage and IL-8 responses. Collectively, our data suggest that ISG15, expressed by VECs upon C. albicans infection, may exert negative feedback on epithelial resistance to infection induced by type I IFNs.

The study reveals how early type I interferon defences are induced during epithelial Candida albicans infection, and how interferon-stimulated gene 15 (ISG15) regulates epithelial responses to infection.

Introduction

Candida albicans is a commensal fungus colonising the human gastrointestinal and vaginal mucosa (Achkar and Fries 2010, Cauchie et al. 2017, Nash et al. 2017, Pappas et al. 2018). While Candida species can cause life-threatening invasive candidiasis in immunocompromised patients (Brown et al. 2012, Armstrong-James et al. 2017, Pappas et al. 2018), vulvovaginal candidiasis (VVC) is more common, affecting three in four women during their reproductive years (Mårdh et al. 2002, Gonçalves et al. 2016). VVC can also present as a recurrent infection that significantly impacts quality of life (Zhu et al. 2016, Lietz et al. 2023). Unlike invasive candidiasis, a compromised immune system does not predispose to VVC (Fidel et al. 1993, Barousse et al. 2001). On the contrary, VVC is characterised by vulvovaginal inflammation, causing symptoms such as irritation, discharge, and pain and itching of the vulva and vagina (Sobel 2007).

In healthy women, the vaginal mucosa tolerates C. albicans without eliciting inflammatory responses (Barousse et al. 2001, Barousse et al. 2005, Moyes and Naglik 2011, Naglik et al. 2014). However, when C. albicans overgrows and expresses its virulence factors, it can overcome this threshold and elicit release of damage-associated molecular patterns (DAMPs) and alarmins. These catalyse pro-inflammatory responses that drive the recruitment and activation of neutrophils (Villar et al. 2005, Cheng et al. 2024). Infiltrating neutrophils in the vaginal mucosa fail to clear the fungal infection and rather cause tissue damage (Fidel et al. 2004, Yano et al. 2018, Pekmezovic et al. 2019). While evidence keeps growing that inflammatory responses during VVC contribute more to disease severity than resolving infection, type I interferon (IFN) responses are increasingly recognised for exerting beneficial roles in resistance to VVC (Kolben et al. 2017, Li et al. 2017, 2019, Pekmezovic et al. 2021, Sala et al. 2023, Rodriguez et al. 2025).

Type I IFNs include thirteen IFNα isoforms, IFNβ, and other less defined interferons (IFNɛ, IFNτ, IFNκ, and IFNω) (Pestka et al. 2004), which activate the IFNα/β receptor (IFNAR) to induce expression of interferon-stimulated genes (ISGs) (Ivashkiv and Donlin 2014). As such, they are best characterised for antiviral mechanisms, yet their roles in antifungal defence are becoming increasingly clear (Pekmezovic et al. 2022). In the context of VVC, type I IFNs seem to contribute to the maintenance of vaginal homeostasis, enhancing epithelial cell resistance to C. albicans infection and dampening neutrophil responses early during infection (Kolben et al. 2017, Li et al. 2017, 2019, Pekmezovic et al. 2021, Sala et al. 2023, Rodriguez et al. 2025). However, the ISGs responsible for these effects remain unknown (Pekmezovic et al. 2021). One ISG that is often among the top induced in response to C. albicans is interferon-stimulated gene 15 (ISG15) (Smeekens et al. 2013, Tucey et al. 2018, Bruno et al. 2020, Bruno et al. 2021, Pekmezovic et al. 2021, Sala et al. 2023). ISG15 encodes a ubiquitin-like protein that can bind (ISGylation) other intracellular proteins, but was shown to also exert signalling functions (Loeb and Haas 1992, Zhang and Zhang 2011).

We hypothesised that ISG15, as a key C. albicans-induced ISG, may play modulatory roles during C. albicans infections of the vulvovaginal mucosa. Therefore, the present work investigated the epithelial threshold to mount type I IFN responses, and how ISG15 shapes epithelial responses to C. albicans.

Results

ISG expression in VECs follows an epithelial activation threshold

Type I IFN responses are induced early during infection of VECs independent of tissue damage (Pekmezovic et al. 2021). In line with this, we observe ISG expression in A-431 cells as early as 3 h post-infection (hpi) (Fig. 1A). To better understand the relationship between infection dynamics and host responses, we monitored cell damage and ISG15 expression over time. Detectable cell damage emerged only from 18 hpi onwards at a multiplicity of infection (MOI) of 1 (Fig. 1B). Blocking of IFN Alpha and Beta Receptor Subunit 2 (IFNAR2) led to a trend towards increased tissue damage. Measuring ISG15 expression over time revealed transient ISG15 expression during early stages of infection dependent on IFNAR2 (Fig. 1C). As type I IFN responses have been attributed beneficial roles in the vaginal mucosa, a role in commensalism and maintenance of the microbiota-host cell interactions seems likely (Kotredes et al. 2017, Pekmezovic et al. 2022). Therefore, we identified the epithelial threshold for eliciting type I IFN responses to C. albicans. A-431 cells infected with a two-fold serial dilution of C. albicans showed significant expression of ISG15 at an MOI ≥ 0.25 at 3 hpi (Fig 1D). This matches the expression of IFN-beta (IFNB1), which was induced from an MOI ≥ 0.0313 (Fig 1E). In the case of MX Dynamin Like GTPase 2 (MX2), its expression was only significantly induced at an MOI of 1 (Fig 1F). In line with increasing C. albicans infection burdens, a dose-dependent increase in epithelial cytotoxicity was observed at 24 hpi (Fig 1G). However, significant release of the neutrophil chemoattractant interleukin-8 (IL-8) was only observed from a MOI ≥ 0.125 (Fig 1H). Epithelial cell supernatants were collected after 24 h of infection and assessed for their capacity to activate neutrophils by stimulation of freshly isolated neutrophils for 24 h. Neutrophil activation, assessed by IL-8 release, was significantly induced upon stimulation with supernatants from A-431 cells that underwent infection with MOI ≥ 0.0625 (Fig. 1I).

Figure 1.

ISG expression in vulvovaginal epithelial cells follows the epithelial activation threshold. (A) mRNA expression of ISG15 and the ISGs Dynamin Like GTPase 2 (MX2), Interferon Regulatory Factor 9 (IRF9), Interferon Induced Protein 44 (IFI44) and Z-DNA-binding protein 1 (ZBP1) in A-431 vulvovaginal epithelial cells 3 hpi with C. albicans (MOI 1). Colour scale represents the mean log₂ fold change in expression compared to uninfected cells (n = 3); normalisation was done to ACTB (β-actin). (B) Quantification of cytotoxicity dynamics in A-431 cells infected with C. albicans MOI 1 in the presence or absence of IFNAR2 blocking (α-IFNAR2) antibody over the course of 24 h via determination of lactate dehydrogenase (LDH) activity from cell-free supernatants (n = 3). (C) Expression of ISG15 by A-431 cells infected with C. albicans MOI 1 at multiple time points over an infection course of 24 h by determination of mRNA expression by RT-qPCR (n = 3) with six technical replicates in total. (D–F) ISG15 (n = 6) (D) and IFNB1 (n = 5) (E) and MX2 (n = 6) (F) mRNA expression at 3 hpi measured by RT-qPCR, relative to ACTB. Samples harvested from A-431 cells infected with a two-fold serial dilution of C. albicans MOI 1. (G) Cytotoxicity quantified by activity of the cytoplasmic enzyme LDH, and (H) IL-8 release detected by ELISA in the cell culture supernatants of A-431 cells infected with a two-fold serial dilution of C. albicans from MOI 1 at 24 hpi (n = 7, LDH and n = 4, IL-8 biological replicates). (I) IL-8 release by human neutrophils stimulated with the epithelial supernatants for 24 h, quantified via ELISA. Bars represent the mean ± SEM of four independent experiments (n = 4 biological replicates) with nine independent neutrophil donors in total. Dotted lines serve as a reference to the mean of the uninfected control condition. To take the variation in the control condition into account, mRNA expression data have been normalised to the mean of the uninfected condition. Statistical differences were assessed using a one-way ANOVA with a Holm–Šídák post-hoc correction (B–I), (* = P ≤ 0.05, ** = P ≤ 0.01, *** = P ≤ 0.001 or **** = P ≤ 0.0001).

ISG15 accumulates intracellularly in response to C. albicans and may conjugate to target proteins

ISG15 is one of the most strongly upregulated ISGs in response to C. albicans in both epithelial cells and peripheral blood mononuclear cells (Smeekens et al. 2013, Tucey et al. 2018, Bruno et al. 2020, 2021, Pekmezovic et al. 2021, Sala et al. 2023). We, therefore, further explored its role during epithelial C. albicans infection. Immunofluorescence microscopy revealed ISG15 accumulation in the cytoplasm of A-431 infected cells (Fig. 2A). This was even further increased by IFNβ stimulation, serving as a positive control for ISG induction (Fig. 2A). As ISG15 can conjugate to a variety of proteins to regulate their function, we assessed ISG15 protein levels using an ISG15 ELISA that exclusively detects unconjugated ISG15. This revealed a reduction in unconjugated ISG15 levels during early infections (3 hpi) (Fig. 2B), suggesting that a substantial fraction of ISG15 was conjugated to intracellular proteins. At later stages of infection (24 h), no differences in unconjugated intracellular ISG15 were observed upon C. albicans infection (Fig. 2B). C. albicans infection alone significantly reduced levels of unconjugated intracellular ISG15 compared to untreated cells at 3 hpi. Stimulation with IFNβ only resulted in a tendency towards reduction of unconjugated intracellular ISG15 when paired with C. albicans infection at 3 hpi. No such effects were observed at the later time point of 24 h (Fig. 2C). Despite significant cytotoxicity induced by C. albicans infection (Fig 1B), no significant release of unconjugated ISG15 was detected at both 3 and 24 hpi (Fig. 2D).

Figure 2.

ISG15 expression in vulvovaginal epithelial cells during C. albicans infection. (A) Fluorescence images of A-431 cells infected with C. albicans (green) for 3 h with or without IFNβ, immunostained for ISG15 (white), F-actin (red), and DAPI (cyan). From the images, the mean fluorescence intensity (MFI) of the ISG15 staining was measured (n = 3). Scale bar 10 µm. (B) Intracellular levels of unconjugated ISG15 quantified via ELISA from cell lysates of A-431 cells infected with C. albicans SC5314 at 3 hpi (n = 4) and 24 hpi (n = 3). (C) Intracellular levels of unconjugated ISG15 measured via ELISA from cell lysates of A-431 cells infected with C. albicans in the presence or absence of IFNβ 3 hpi (n = 3) and 24 hpi (n = 3). (D) Detection of released extracellular ISG15 after infection with C. albicans MOI 1, measured via ELISA at 3 hpi (n = 6) and 24 hpi (n = 3) from culture supernatants. Bars represent the mean ± SEM, with dotted lines in the graphs serving as a reference to the mean of the uninfected control condition. Statistical differences were assessed using two-way ANOVA (A, C) test or a paired t-test (B, D) (* = P ≤ 0.05 or **** = P ≤ 0.0001).

ISG15 differentially regulates epithelial cell death depending on C. albicans infection

To study how intracellular ISG15 expression shapes epithelial C. albicans infection, ISG15 expression was silenced in A-431 cells. Quantitative reverse transcription polymerase chain reaction (RT-qPCR) confirmed ISG15 silencing, as its expression was reduced in uninfected cells, but also upon C. albicans infection and stimulation with IFNβ (Fig. 3A). Interestingly, ISG15 silencing was associated with reduced epithelial cytotoxicity (∼20%) upon infection, compared to cells with intact ISG15 expression (Fig. 3B). Moreover, ISG15 silencing was associated with significantly reduced IL-8 release upon C. albicans infection (Fig. 3C). As ISG15 is known to negatively regulate IFNAR signalling through ISGylation of IFNAR2 (Zhang et al. 2015, Arimoto et al. 2017), expression of other ISGs following ISG15 silencing was quantified. In uninfected cells, only a few changes in ISG expression were detected (Fig. 3D). However, IFNβ-induced gene expression of ZBP1 (Z-DNA binding protein 1), a driver of PANoptosis (combined Pyroptosis, Apoptosis, and Necrosis) in response to fungal pathogens (Banoth et al. 2020), was strongly enhanced upon ISG15 silencing (Fig. 3D). In sharp contrast, ZBP1 expression in cells infected with C. albicans was reduced upon ISG15 silencing (Fig. 3D). We thus investigated whether the effects of ISG15 on cytotoxicity may be attributable to changed expression of ZBP1. Silencing of ZBP1, confirmed by RT- qPCR (Fig. 3E), significantly reduced cytotoxicity induced by C. albicans infection at 24 hpi (Fig. 3F). Further, IL-8 release was also significantly reduced in C. albicans-infected A-431 cells upon ZBP1 silencing (Fig. 3G). While ZBP1 silencing phenocopied the effects of ISG15 silencing in terms of cytotoxicity and IL-8 release, no effects on the expression of other ISGs like MX2, ISG15, and Interferon regulatory factor 9 (IRF9) were observed (Fig. 3H).

Figure 3.

ISG15 silencing dampens C. albicans-induced epithelial cytotoxicity and IL-8 responses. (A) ISG15 mRNA expression after transfection with ISG15-targeting siRNA pools (ISG15 siRNA) or non-targeting control siRNA (Control siRNA) after C. albicans infection or treatment with or without IFNβ, measured at 24 hpi by RT-qPCR (n = 6). (B) LDH release measured at 24 hpi from cell-free supernatants of A-431 cells infected with C. albicans MOI 1 (n = 5) or with IFNβ (n = 5) after ISG15 siRNA transfections. (C) IL-8 release by A-431 cells after ISG15 siRNA transfection and infection with C. albicans 24 hpi, quantified via ELISA from cell-free supernatants (n = 5). (D) mRNA expression of MX2, ZBP1, USP18, and IRF9 after ISG15 siRNA transfection during infection with C. albicans or following stimulation with IFNβ, measured at 24 hpi by qPCR (n = 3). (E) ZBP1 mRNA expression after transfection with ZBP1-targeting siRNA pools (ZBP1 siRNA) after C. albicans infection or treatment with or without IFNβ, measured at 24 hpi by qPCR (n = 5). (F) LDH release measured at 24 hpi from cell-free supernatants of A-431 cells infected with C. albicans MOI 1 (n = 5) or with IFNβ (n = 5) after ZBP1 siRNA transfections. (G) IL-8 release by A-431 cells after ZBP1 siRNA transfection and infection with C. albicans 24 hpi, quantified via ELISA from cell-free supernatants (n = 5). (H) mRNA expression of MX2, ISG15, and IRF9 after ZBP1 siRNA transfection during infection with C. albicans or following stimulation with IFNβ, measured at 24 hpi by qPCR (n = 4). Relative mRNA expression was normalised to ACTB (β-actin), and data were normalised to the mean of untreated controls (uninfected A-431 cells that were neither infected nor transfected with siRNA). Bars represent the mean, with dotted lines in the graphs serving as a reference to the mean of the uninfected control condition. Statistical differences were assessed using paired two-way ANOVA with a Holm–Šídák post hoc correction. (* = P ≤ 0.05, ** = P ≤ 0.01 *** = P ≤ 0.001, or **** = P ≤ 0.0001).

Discussion

Here, we investigated the induction of epithelial type I IFN responses to C. albicans infection and the role of ISG15. Previously, type I IFN signalling had been associated with epithelial resistance to C. albicans infection and downstream immunomodulatory effects in the context of VVC models (Pekmezovic et al. 2021). C. albicans induced IFNB1 and ISG15 expression in VECs at fungal loads that caused cytotoxicity and IL-8 release at later stages of infection. However, ISG expression was induced at timepoints before cell damage was detectable, showing unique dynamics of the type I IFN response. While ISG15 accumulated intracellularly upon C. albicans infection, no release of unconjugated ISG15 was observed. Silencing of ISG15 revealed a negative feedback of ISG15 on protective type I IFN responses. Expression analysis of other ISGs after silencing of ISG15 showed reduced expression of ZBP1, a driver of PANoptosis (combined Pyroptosis, Apoptosis, and Necrosis) in response to fungal pathogens (Banoth et al. 2020). Accordingly, ZBP1 silencing mimicked the effects of ISG15 silencing by reducing C. albicans-induced cytotoxicity and IL-8 release from A-431 cells (Fig. 4).

Figure 4.

Schematic summary of type I interferon and ISG15 in epithelial response to C. albicans infection. (A) C. albicans burden that exceeds epithelial thresholds triggers epithelial cytotoxicity, IL-8 release, neutrophil activation, and IFNB1 expression. (B) Activation of the IFNα/β receptor by type I IFNs induces the expression of interferon-stimulated genes (ISGs). ISG15 accumulates in the cell in response to infection and conjugates to target proteins. As an intracellular protein, ISG15 may stabilise USP18 and negatively regulate IFNAR signalling to prevent overactivation of the pathway (Zhang et al. 2015, Arimoto et al. 2017). (C) Silencing of ISG15 gene is associated with reduced IL-8 release and cytotoxicity. The latter may be due to effects of ISG15 on expression of other ISGs during infection. Absence of ISG15 during C. albicans infection was associated with decreased expression of ZBP1, a driver of PANoptosis in response to fungal pathogens (Banoth et al. 2020).

While C. albicans infections in vitro increased epithelial cytotoxicity in an MOI-dependent manner, this was not the case for inflammatory responses. In line with the previously proposed epithelial activation threshold (Yano et al. 2012, Ardizzoni et al. 2021), we observed that VECs tolerate a certain amount of cytotoxicity before mounting significant IL-8 responses. This was similarly true for the ISG expression. The fungal burdens that induced IFNB1 and ISG15 expression at an early time point matched with the burdens that caused strong cytotoxicity at 24 hpi. This suggests that epithelial type I IFN responses are mounted only in response to fungal burdens that can inflict tissue damage, without being dependent on the damage itself. In line with this, non-pathogenic Candida species and an avirulent ece1Δ/Δ mutant were found to induce epithelial type I IFN responses (Pekmezovic et al. 2021). In addition, C. albicans isolates that asymptomatically colonised the vagina or caused VVC also differentially induced type I IFN responses (Sala et al. 2023), suggesting that some strains circumvent type I IFN responses.

Type I IFN responses to C. albicans are characterised by ISG15 expression (Smeekens et al. 2013, Tucey et al. 2018, Bruno et al. 2020, 2021, Pekmezovic et al. 2021, Sala et al. 2023). ISG15 can be released and act as an immunomodulatory mediator, for example via the lymphocyte function-associated antigen 1 (LFA-1) receptor on various leukocytes (Swaim et al. 2017). Nevertheless, in our model we did not detect increased release of unconjugated ISG15 from C. albicans-infected VECs. This was counter-intuitive since cytoplasmic enzymes like LDH leak through the compromised cell membrane upon cell death induced by C. albicans. Potentially, ISG15 conjugation to target proteins may have prevented release of unconjugated ISG15.

Intracellularly, ISG15 can either be found in its unconjugated form or bound to other proteins (Loeb and Haas 1992, D’Cunha et al. 1996). Immunofluorescence staining with an antibody binding both unconjugated and conjugated ISG15 demonstrated that IFNβ stimulation and/or C. albicans infection led to ISG15 accumulation inside epithelial cells. Yet, ELISAs showed that unconjugated intracellular ISG15 levels were reduced upon C. albicans infection, suggesting that ISG15 was conjugated (ISGylation) to target proteins (Loeb and Haas 1992). ISG15 can target a variety of proteins, including IFNα/β-induced antiviral proteins, which are relevant in viral infections (Perng and Lenschow 2018, Freitas et al. 2020). Most targets of ISGylation are constitutively expressed proteins involved in diverse cellular pathways such as polymerase II transcription, RNA splicing and translation, chromatin remodelling, and cytoskeleton reorganisation (Perng and Lenschow 2018), all restricting viral replication and particle formation. Therefore, ISGylation plays diverse roles in autophagy, metabolic reprogramming, response to DNA damage, and stress responses.

Importantly, ISG15 can negatively regulate IFNAR signalling by stabilising USP18 (Kang et al. 2022) and preventing excessive IFNα/β signalling (Zhang et al. 2015, Arimoto et al. 2017, Kang et al. 2022). Consequently, many cells exhibit increased ISG expression when carrying a loss-of-function mutation in ISG15, which could be attributed to the ability of ISG15 to stabilise USP18 and negatively regulate IFNAR signalling (Zhang et al. 2015, Basters et al. 2018). In line with the negative feedback of ISG15 on IFNAR2 signalling, we found that IFNβ-induced ZPB1 expression increased upon ISG15 silencing. In contrast, C. albicans infection downregulated ZPB1 expression, which was further decreased upon ISG15 silencing. This underscores that in VECs, type I IFN responses induced by C. albicans diverge from those induced by IFNβ. Further supporting this, C. albicans infection only transiently induced ISG15 expression, in contrast to IFNβ stimulation, which showed persistent expression of ISG15 after 24 h. This suggests that C. albicans infection gives type I IFN responses a unique signature, with ZBP1 showing a divergent response.

ISG15 silencing reduced C. albicans-induced cytotoxicity of epithelial cells. This is in contrast with a study demonstrating increased corneal susceptibility to C. albicans infection in mice treated with ISG15 siRNA (Dong et al. 2017). In addition, studies in Chlamydia trachomatis infection models demonstrated that ISG15 exerts a dual function, contributing to bacterial clearance while fine-tuning immune responses through modulation of IFNγ production and dampening excessive inflammatory signalling, thereby acting as an “immune brake” (Wu et al. 2024, Guo et al. 2025).

ISG15 distinguishes itself from other ISGs in its negative effect on epithelial resistance, as interferon alpha-inducible protein 6 (IFI6) and Cytidine/Uridine Monophosphate Kinase 2 (CMPK2) silencing impaired epithelial resistance to C. albicans infection (Pekmezovic et al. 2021). We could, however, attribute the reduced cytotoxicity of epithelial cells to downregulation of ZBP1 expression upon ISG15 silencing. ZBP1 can engage receptor-interacting serine/threonine-protein kinase 3 (RIPK3) to trigger cell death (Koerner et al. 2024), and has been linked with induction of PANoptosis in response to C. albicans (Banoth et al. 2020). We found that silencing ZBP1 phenocopied the reduction in C. albicans-induced damage and IL-8 release observed after silencing of ISG15. This suggests that the reduction in tissue damage following ISG15 silencing might be partially attributable to reduced ZBP1 expression. The reduced IL-8 release following silencing of ZBP1 may be explained by the link between cell damage and IL- 8 responses in A-431 cells (Valentine et al. 2024).

Collectively, epithelial type I IFN responses occur prior to induction of cytotoxicity at an activation threshold, similar to proinflammatory IL-8 responses. As a key interferon-stimulated gene, ISG15 is dynamically expressed upon epithelial C. albicans infection. ISG15 negatively regulates resistance of vaginal epithelial cells to infection and associated IL-8 responses, in part via regulation of ZBP1 expression.

Methods

Ethics statement

Human peripheral blood was collected from healthy volunteers with ethical approval after obtaining written informed consent. The Jena institutional ethics committee approved the procedure (Ethik-Kommission des Universitätsklinikums Jena, Permission No 2207–01/08). Fresh blood was collected in 5.5 ml S-Monovette tubes (Sarstedt) containing ethylenediaminetetraacetic acid (EDTA) to prevent blood from coagulating.

Fungal strains and culture conditions

This study used the C. albicans wild-type reference strain SC5314 (Gillum et al. 1984). Candida albicans was grown from glycerol stocks on yeast extract peptone dextrose (YPD) agar plates and stored at 4°C for up to 2 weeks. Overnight cultures were grown in YPD media from a single picked colony at 30°C in an orbital shaker at 180 rpm. For collection, yeast cells were centrifuged, washed three times with phosphate-buffered saline (PBS, pH 7.4), counted using a Neubauer chamber (Merck) and adjusted to the desired MOI in Roswell Park Memorial Institute (RPMI)-1640 medium (Thermo Scientific).

In vitro epithelial infection model

Human A-431 epithelial cells were used to mimic the vulvovaginal epithelium. This cell line was derived from a solid vulvar epidermoid carcinoma and is routinely used to model the vaginal mucosa (Schaller et al. 2006, Hernandez and Rupp 2009). Cells were cultured in RPMI-1640 (Thermo Fisher Scientific) supplemented with 10% heat-inactivated fetal bovine serum (FBS) (Bio&Sell GmbH) and incubated at 37°C and 5% CO₂. All infection experiments were done with RPMI-1640 without FBS supplementation. For incubation steps, cells were kept inside a humidified incubator at 37°C and 5% CO₂, unless stated otherwise. Cells were infected with MOI 1 unless otherwise indicated.

To assess ISG15 expression over the course of infection (24 h), A-431 cells were seeded in 6-well plates, infected with C. albicans SC5314 at MOI 1, and RNA samples as well as supernatants were harvested at the indicated time points. To block IFNAR2 signalling, anti-human IFNAR2 antibody (4 ng/ml; PBL InterferonSource) was added to the epithelial cells 3 h before infection.

Supernatants for cytotoxicity and enzyme-linked immunosorbent assay (ELISA) measurements were collected after centrifuging the cells at 250 × g for 10 min. Collected supernatants were directly used for cytotoxicity quantification and stored at –20°C for ELISA measurements. Cell lysates were created by removing the remaining media from all wells and adding radioimmunoprecipitation assay (RIPA) (10x; Millipore) lysis buffer + 1 × protease inhibitor (Roche) mix.

Alternatively, cells were lysed in RLT buffer (Qiagen) containing 1% β-mercaptoethanol (Carl Roth) and the cells detached using a cell scraper for RNA extraction. Samples were stored at -80°C until RNA extraction.

Epithelial cytotoxicity quantification assay

To assess the cytotoxicity inflicted by C. albicans, the release of the cytoplasmic enzyme LDH was measured as a marker for necrotic epithelial cell death from the supernatants of infected cells (Chan et al. 2013). Supernatants were diluted 1:10 in PBS, and the amount of LDH was measured using a cytotoxicity detection kit (Roche) according to the manufacturer’s protocol. Absorbance was measured at 490 nm (reference 630 nm) using a microplate reader (Tecan; M-plex).

ELISA

The release of human IL-8 (R&D Systems) and ISG15 (AssayGenie; biomol) was measured using commercially available human ELISA kits according to the manufacturer’s instructions. Supernatants that had been stored at -80°C were thawed on ice before measurements. Washing steps were done using a microplate washer (Hydrospeed; Tecan), and absorbance was measured with a microplate reader (Tecan; M-Plex) at 450 nm (reference 520 nm).

RNA isolation and cDNA preparation

Cell lysates in RLT buffer were thawed on ice and centrifuged for 10 min (20 000 × g at 4°C) before transferring the supernatant into new tubes containing 300 µl ethanol. After mixing, the samples were added onto spinning columns and human RNA was isolated using a kit (RNeasy Mini Kit; Qiagen) according to the manufacturer’s protocol. RNA concentrations were quantified using a NanoDrop (NanoDrop One; Thermo Fisher Scientific). Isolated RNA was treated with DNase I (Qiagen) according to the kit’s instructions and transcribed into complementary DNA (cDNA) by diluting the RNA (500 ng or 250 ng) with nuclease-free water to 20 µl. After the addition of 1 µl Oligo-dT primers (Invitrogen), samples were incubated at 70°C for 10 min inside a thermocycler (lid set to 95°C). After incubation, samples were put on ice for 2 min. For cDNA synthesis, 12 µl of cDNA synthesis mix (Table 1) was added to each sample before incubating inside a thermocycler for 2 h at 42°C, then 15 min at 70°C.

Table 1.

cDNA synthesis master mix.

cDNA synthesis mix (1x)	Volume (µl)
5x First strand buffer (Invitrogen)	7
DTT (Invitrogen)	2
dNTPS (New England BioLabs Inc.)	1
RNase OUT (Invitrogen)	1
Reverse Transcriptase Superscript III (Invitrogen)	1
Total volume	12

RT-qPCR

To measure mRNA expression of target genes, 1 µl of generated cDNA was added to a 96-well microtiter qPCR plate. After adding 19 µl of Master Mix (Table 2), containing the corresponding primer pairs (Table 3) for the respective gene to the wells, the plate was centrifuged for 3 min at 250 × g before measuring the gene expression over time in a CFX96 thermocycler (Bio-Rad). The mRNA expression of selected ISGs (ISG15, MX2, IFNB1, IRF9, IFI44, and ZBP1) were measured against ACTB. To take the variation in the control condition into account, mRNA expression data were normalised to the mean of the uninfected condition.

Table 2.

qPCR master mix.

RT- qPCR master mix (1x)	Volume (µl)
Forward primer	0,8
Reverse primer	0,8
Nuclease-free water (Qiagen)	7,4
GoTaq qPCR master mix (2x; Promega)	10
Total volume	19

Table 3.

Primers.

Gene	Forward primer sequence (5’ → 3’)	Reverse primer sequence (5’ → 3’)
ISG15	CGCAGATCACCCAGAAGATCG	TTCGTCGCATTTGTCCACCA
MX2	AAAAGCAGCCCTGTGAGGCATG	GTGATCTCCAGGCTGATGAGCT
ACTB	AGAGCTACGAGCTGCCTGAC	AAGGGACTTCCTGTAACAACG
IFNB1	AAACTCATGAGCAGTCTGCA	AGGAGATCTTCAGTTTCGGAGG
IRF9	GATACAGCTAAGACCATGTTCCG	TGATACACCTTGTAGGGCTCA
IFI44	ATGGCAGTGACAACTCGTTTG	GCAGGATCTTTCGTGCAACC
ZBP1	CCGACTCCTTGCAGCTGCTGTC	CCAGCCACCCCTGGGCTGATAG
USP18	TGGACAGACCTGCTGCCTAAC	CTGTCCTGCATCTTCTCCAGCA

ISG15 immunostaining

At indicated time points post-infection, A-431 epithelial cells were infected with wild-type C. albicans in the presence or absence of IFNβ (10 ng/ml). At 3 hpi, cells were fixed in Histofix (Carl Roth) for 20 min at room temperature (RT), permeabilised with 0.05% Saponin (Carl Roth) and blocked with 1% bovine serum albumin (BSA; Serva) for 60 min. C. albicans was stained using rabbit anti-C. albicans antibody (1:600 in PBS with 0.1% BSA and 0.05% saponin, BP1006, Acris Antibodies) overnight, together with ISG15 using mouse anti-ISG15 IgG (1:100 in PBS with 0.1% BSA and 0.05% saponin, sc-166755, Santa Cruz Biotechnology). Afterwards, samples were washed with PBS and incubated with a secondary goat anti-rabbit AlexaFluor488 and a secondary goat anti-mouse AlexaFluor647 (1:1000 in PBS with 0.1% BSA and 0.05% saponin, Invitrogen A32731 and Invitrogen A21235, respectively) for 1 h in the dark at RT. Nuclei were stained with DAPI (Sigma D9542) and F- actin was stained with Phalloidin conjugated to Alexa Fluor™ 594 (Invitrogen A12381). Finally, samples were mounted with Mowiol 4–88 (Carl Roth). All samples were imaged using an Axio Observer fluorescence microscope (Carl Zeiss, Germany). Images were taken using a 63×/1.40 oil immersion objective.

Mean fluorescence intensity of ISG15 was determined by segmenting individual cells using the DAPI channel to identify single-cell nuclei and the Phalloidin channel to identify cell outlines. Subsequently, a threshold was applied to each cell. The regions of interest (ROI) were analysed using the “Analyse Particles” plugin from Fiji ImageJ. Puncta in clusters were separated using the watershed function. The background was subtracted from the total fluorescence intensity of each image. At least 10 different fields were imaged for each experimental condition, and the images were acquired with the same settings.

RNA interference assay

Cells were transfected using small interfering RNA (siRNA) pools targeting either ISG15 or ZBP1. Each transfection was carried out by preparing a solution of 100 nM ISG15 siRNA pool consisting of three target-specific siRNAs (sc-43869; Santa Cruz Biotechnology) with transfection reagent (sc-29528, Santa Cruz Biotechnology) in transfection medium (sc-36868, Santa Cruz Biotechnology) or 25 nM ON- TARGETplus Human ZBP1 siRNA SMARTPool (L-014650–00-0010; Dharmacon) with DharmaFECT 1 Transfection Reagent (T-2001–02, Dharmacon) and serum-free RPMI-1640 according to the manufacturer’s instructions. As controls, additional transfection solutions were prepared with either untargeted control siRNA-A (sc-37007; Santa Cruz Biotechnology) or ON- TARGETplus Non-targeting Control Pool (D-001810–10-05, Dharmacon). The cells were then transfected with the prepared transfection solutions according to the supplier’s instructions for a total of 24 h before infection.

After incubation, the media in all wells were replaced with RPMI without FBS and cells were infected with C. albicans in the presence or absence of IFNβ (10 ng/ml). After 24 h of infection, cell culture supernatants were collected, and cells were lysed for RNA extraction as described above. Silencing of the selected genes ISG15 and ZBP1 was confirmed by performing RT-qPCR.

Isolation of polymorphonuclear neutrophils (PMNs) from fresh blood

Neutrophils were isolated from fresh blood from healthy donors according to a previously published protocol (Gresnigt et al. 2012). After separation by density gradient centrifugation with Lymphocyte separation media (Histopaque 1077; Capricon Scientific), the Histopaque 1077 and plasma layers were removed from the 50 ml tube without disturbing the erythrocyte layer containing the neutrophils. The tubes were filled up to 50 ml with ice-cold hypotonic lysis buffer (155 mM NH₄Cl, 10 mM KHCO₃; Sigma-Aldrich), gently mixed, kept on ice for 15 min, and centrifuged at 300 × g for 10 min at 4°C. After centrifugation, the supernatant was discarded, and the lysis step was repeated with a shortened 10 min incubation time. After removing the supernatant, the isolated neutrophils were washed twice with ice-cold PBS by centrifugation (300 × g, 15 min, 4°C). The washed cell pellet was resuspended in 3 ml RPMI-1640 with phenol red. Cells were counted inside a Neubauer counting chamber, and the cell concentration was adjusted according to the experiment’s demands.

Neutrophil cytokine release

Neutrophils were seeded in 96-well plates at a concentration of 5 × 10⁵ cells/ml in RPMI. One hundred microliters of supernatants from VEC infected with C. albicans for 24 h were added to the neutrophils and incubated for 24 h at 37°C and 5% CO₂. Subsequently, supernatants were taken and measured for IL-8 release via ELISA as described above.

Data and statistical analysis

Experiments were done in biological triplicate (n = 3) unless indicated differently in the figure legend. Graphs show the mean values of biological replicates, including the standard error of the mean (SEM). GraphPad Prism 10 was used for statistical analyses and plotting. Used statistical tests are specified in the figure legends. Statistical significances always refer to the controls depicted in the figures: * = P ≤ 0.05, ** = P ≤ 0.01, *** = P ≤ 0.001 or **** = P ≤ 0.0001. Image handling, quantification, and analysis of fluorescence images were performed using ImageJ (Fiji) software.

Author contributions

Jördis V. Schuchardt (Investigation, Writing–review & editing, Writing–original draft, Visualisation, Validation, Formal analysis), Alexis Urtecho Valverde (Investigation, Writing–review & editing, Writing–original draft, Visualisation, Validation, Formal analysis), Beatriz Cristóvão (Writing-review & editing, Investigation, Supervision, Validation, Formal analysis), Marina Pekmezović (Methodology, Supervision), Raquel Alonso-Roman (Writing–review & editing, Writing–original draft), and Mark S. Gresnigt (Writing–review & editing, Visualisation, Supervision, Project administration, Funding acquisition, Conceptualisation, Formal analysis, Methodology, Resources)

Conflicts of interest

The authors declare no competing interests.

Funding

J.V.S., A.U.V., M.P., B.C., and M.S.G. were supported by the DFG within the Collaborative Research Centre (CRC)/Transregio (TRR) 124 “FungiNet” projects C1 (DFG project number 210879364). M.S.G. was funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) Emmy Noether Program (project no. 434385622/GR 5617/1-1). We acknowledge support by the Jena School for Microbial Communication (JSMC) that is third-party funded by the Thuringian Ministry of Education, Science and Culture (TMBWK).