Upregulation of IFNE in cervical biopsies of patients with high‐risk human papillomavirus infections

Abstract Problem Interferon epsilon (IFN‐ε) is constitutively expressed in the epithelium of female reproductive tract and confers vital protection against sexually transmitted pathogens in mouse models. However, there is limited insight into the role of IFN‐ε in human sexually transmitted infections such as human papillomavirus (HPV). Method of Study Cervical biopsies were obtained from high‐risk (HR) HPV positive (n = 28) and HR‐HPV negative (n = 10) women. mRNA expression of IFN‐ε in cervical tissues was measured by qPCR. Expression of the IFN‐ε protein was determined by Western blot analysis, immunohistochemistry and immunofluorescence staining. Results mRNA expression of IFN‐ε was higher in the ectocervix than that of other IFNs, and was further upregulated in HR‐HPV positive women compared with HR‐HPV negative women. Expression of the IFN‐ε protein was comparable between HR‐HPV infected patients and healthy controls. Conclusions These results reveal differential expression of IFN‐ε mRNA between individuals with or without HR‐HPV infection, and imply direct or indirect regulatory mechanisms for IFN‐ε transcription by HPV. Expression of IFN‐ε protein in HPV infections would require further validation.


| INTRODUCTION
The recently characterized interferon (IFN)-ε is unique for its tissue-specific expression profile.Unlike other type I IFNs, IFN-ε is constitutively expressed in the epithelial cells of the female reproductive tract (FRT). 1,2This unique feature suggests its potential role in protecting against sexually transmitted infections (STIs) in the FRT, as demonstrated in murine models. 1 IFN-ε binds to IFNα/β receptor 1 and 2 (IFNAR1 and IFNAR2) and regulates expression of IFN-stimulated genes (ISGs) via the JAK-STAT signaling pathway. 1,3Thus, it appears to share the antiviral, antibacterial, antiproliferative and immunoregulatory properties with other type I IFNs.In vitro investigations into these biological functions mostly showed subdued potency when compared to other IFNs. 4,5For example, antiviral activity against Semliki Forest virus (SFV) of IFN-ε in vitro is 100 and 1000-fold weaker than that of IFN-α and IFN-β, respectively.Similarly, its antibacterial activity is diminished 1000-fold compared to IFN-β. 5 IFN-ε is reported to have comparable potency of HIV restriction factor induction as conventional type I IFNs. 6In vivo studies have shown that IFN-ε deficiency in mice is associated with increased susceptibility and worsened disease outcomes in herpes simplex virus type 2 (HSV-2) and Chlamydia infection. 1n contrast to most type I interferons (IFNs), which are highly inducible by pathogens, 7 Ifne expression remains unaltered in mice following infection with Mengo virus, 8 Chlamydia muridarum, and HSV-2. 1 However, a pilot study by Nickodem et al. reported downregulation of IFN-ε in HSV-infected women in pregnancy, 9 suggesting a potential regulation of IFN-ε expression by certain pathogens in human.Characterization of human type I IFN locus revealed that IFNE has a highly conserved 5′ flanking region sharing >70% identity with its counterpart in mice. 10Previous attempts at characterizing putative transcription factor binding sites in the 5′ flanking region of IFN-ε genes have yielded contradictory results regarding the presence of response elements to pattern recognition receptor (PRR) pathways.Recent analysis of the ~1 kb 5′ flanking region by Fung et al. 1 showed that Ifne lacks these response elements, which contradicts earlier reports. 10This observation is supported by in vitro evidence, such as the limited impact on IFN-ε expression in HeLa cells when stimulated with PRR ligands like poly(I:C).Additionally, IFN regulatory factor (IRF)-3, −5, and −7 expression vectors fail to induce IFN-ε expression in luciferase reporter assays. 1,11Subsequent research has consequently shifted its focus towards the hormonal aspect of IFN-ε regulation.A notable discovery in this regard was the lack of correlation between progesterone receptor and IFN-ε expression in the FRT epithelium.Sex hormones are dominant regulators of FRT physiology and exert a profound impact on immune responses against STIs.Unique expression profiles of IFN-ε add another layer of complexity to its host defense function.
While many fundamental questions regarding the function and regulation of IFN-ε remain to be fully understood, previous studies using cell lines and mouse models collectively suggest that IFN-ε may provide a mild but persistent level of protection against pathogens specifically in the FRT mucosa.Human papillomaviruses (HPVs) are among the most prevalent sexually transmitted pathogens affecting the FRT.HPVs are highly tissue-tropic viruses that infect stratified epithelia, especially that of the human cervix. 12HPVs have received great public attention due to the oncogenic effects of certain subtypes.Oncogenic or high-risk (HR) HPV types are responsible for approximately 4.5% of all cancers worldwide, including the majority of cervical cancer and a substantial portion of other anogenital and oropharyngeal cancers. 13While most HPV infections are cleared by the immune system, persistent infections of HR-HPV in the basal layers of epithelial cells can lead to the development of precancerous lesion or cancer.Multivalent prophylactic HPV vaccines have proven to be highly effective in preventing HPV-attributable cancers, but they do not eliminate pre-existing infections. 14Further investigations into the viral-host interactions are necessary to develop therapeutics against persistent HR-HPV infections.The benefits of type I IFNs in pathogen control have been extensively studied.Given its unique spatial expression patterns, IFN-ε holds the potential to be a pivotal component in immune responses during HPV infections.
Given the limited number of clinical studies, there is still much to uncover regarding the roles of IFN-ε in human.Particularly, IFN-ε expression in patients infected with sexually-transmitted pathogens, such as HPV, is poorly characterized.Marrero-Rodriguez et al. 15 reported that IFN-ε expression is elevated in cervical cancer, but it remains unclear whether IFN-ε expression is regulated by HPV infection.To answer this question, we investigated the expression of IFN-ε at both the mRNA and protein levels in women with or without HR-HPV infections.

| Patient samples
This study was approved by the ethics committee of Beijing Obstetrics and Gynecology Hospital, Capital Medical University (2019-KY-079-02).A total of 38 women (28 HR-HPV positive and 10 HR-HPV negative) were recruited from Beijing Obstetrics and Gynecology Hospital and Beijing Chaoyang Integrative Medicine Emergency Medical Center and signed informed consent before study participation.Samples were obtained during either routine cervical biopsy procedures for diagnosis of precancerous conditions and cervical cancer, or surgeries for nonmalignant gynecological diseases.Cervical lesions were detected with ThinPrep cytologic test before sample collection.Diagnosis was confirmed histologically in patients eligible for cervical biopsy.Cervical tissues (3 × 3 × 3 mm 3 ) obtained from HR-HPV negative individuals were divided into two aliquots.One aliquot was frozen in liquid nitrogen and stored in −80℃ before lysis using TRIzol (Invitrogen).The other was incubated in 4% paraformaldehyde or 10% formalin for 24 h at 4℃.Fixed samples were later transferred to 70% EtOH.Due to the limited volume of samples obtained from HR-HPV positive patients, only a subset of these samples (n = 10) were divided into two aliquots for both mRNA and histology analysis.The remaining samples were either immediately frozen in liquid nitrogen (n = 12) or fixed in 4% paraformaldehyde or 10% formalin (n = 6).

| Quantitative polymerase chain reaction (qPCR)
RNA was extracted from samples with TRIzol reagent (Invitrogen) and treated with TURBO DNase (Invitrogen).cDNA was synthesized using M-MLV reverse transcriptase (Takara) and oligo (dT) primers.qPCR was performed using PowerUp SYBR Green master mix (Applied Biosystems) on the StepOnePlus system (Applied Biosystems).Expression of target genes was normalized to housekeeping gene EEF1A1 with the comparative threshold cycle method.Primer sequences are listed in Table 1.

| Statistical analysis
Statistical analysis was performed with two-tailed Welch's t-test or other tests where indicated.p < .05 was taken as statistically significant.Statistical analysis was performed using GraphPad Prism 9.1.1.

| RESULTS
Cervical tissues were collected from a total of 38 enrolled individuals (10 negative for HR-HPV and 28 positive for HR-HPV).Due to the limited volume of each sample, only a subset of samples (10 h-HPV negative and 22 h-HPV positive) were subjected to RNA extraction and characterization of IFN expression at the mRNA level (demographic and clinical characteristics shown in Table 2).
To assess the expression of IFN-ε in relation to other IFNs, we measured mRNA expression of type I IFNs (IFN-α2, -β, -κ, -ω), type II IFN (IFN-γ), and type III (IFN IFN-λ1) alongside IFN-ε.Our analysis revealed considerable variability in the expression levels of all IFN genes.Notably, IFNE showed higher expression than genes which encode other IFNs (Figure 1A), as previously documented. 2 No significant difference were observed in IFNE expression among patients grouped by age (p = .43),gravidity (p = .10),parity (p = .51),or cervical squamous intraepithelial lesion (SIL) diagnosis (p = .15).The expression of IFNE and other genes encoding IFNs did not differ significantly among individuals at different stages of menstrual cycle (p = .12;Figure 1B), which was consistent with the observations of Bourke et al. 2 in their studies with an American cohort.
Next, we examined IFN mRNA levels in HR-HPV positive patients and healthy controls.HR-HPV positive patients showed higher expression of IFNE compare to healthy individuals (p = .03),while no statistically significant differences were detected for the expression of other IFN genes between these two groups (Figure 2A).
IFN-ε expression showed considerable variations at the protein level among individuals in both HR-HPV positive and HR-HPV negative groups.However, no substantial difference was observed between these two groups (Figure 2B).This prompted us to explore potential differences in the spatial expression patterns of IFN-ε.Consistent with previous reports, 2 immunofluorescence and immunohistochemical staining both demonstrated that IFN-ε is primarily expressed in the basal layer of cervical epithelium (Figures 2C,E).We also observed strong signals in the stromal endothelial cells, which accounted for the unexpectedly high expression level of IFN-ε in the stroma.Quantitative analysis of immunofluorescence staining demonstrated comparable expression between HR-HPV positive and negative individuals (Figure 2D).

| DISCUSSION
The critical role of IFNs in innate immune responses against viral infection has been well-established.As a novel type I IFN predominantly found in FRT epithelium, IFN-ε may confer vital protection against sexually transmitted pathogens.Bourke et al. 2 have provided a comprehensive analysis of IFN-ε expression at homeostasis, highlighting hormonal control in different anatomical locations of the FRT.However, IFN-ε expression during infection remains to be elucidated.This study is the first to compare the expression of IFNs in HR-HPV positive and HR-HPV negative individuals.IFNE showed highest basal level of transcription among genes encoding IFNs and was further induced in HR-HPV infection, suggesting a potential role in persistent protection.By contrast, other IFNs investigated in this study showed comparable mRNA levels in individuals with or without HR-HPV infection.Marrero-Rodriguez et al. 15 previously reported that IFN-ε expression was upregulated in cervical cancer, and suggested that IFN-ε could serve as a biomarker for cervical cancer.Here, we quantitatively assessed IFNE mRNA abundance in HR-HPV infection and further showed that IFN-ε is upregulated in HR-HPV infected normal or precancerous tissue, suggesting a potential role for IFN-ε in the host defense against HPV during early stages of the disease.
We did not observe significant differences in IFN-ε protein abundance between the HR-HPV positive and negative samples.Such discrepancies between the mRNA and protein abundance could be attributed to sample variability, particularly when working with a limited sample size, as in some cases the sample volume did not permit us to assess both protein and mRNA levels.Alternatively, this could suggest underlying regulatory mechanisms operating at the translational or posttranslational level.Future in-depth mechanistic studies are warranted to gain a comprehensive understanding of these discrepancies and to unravel the precise mechanisms regulating IFN-ε expression.
The specific mechanisms underlying the induction of IFN-ε upon HR-HPV infection remain elusive, but it is reasonable to speculate that both direct and indirect pathways may be at play.Experimental evidence has demonstrated that viral oncoproteins E5, E6 and E7 of certain HR-HPV subtypes contribute to the immune evasion mechanisms of HR-HPV by directly interfering with the IFN signaling cascade. 16These subtype-specific regulations have been predominantly explored in the context of the expression of other type I IFNs, including IFN-α, -β and -κ.8][19][20] Additionally, other mechanisms such as epigenetic silencing of IFN can also be involved. 21We were unable to examine subtypespecific regulations on IFN expression due to limited sample size.Whether similar subtype-specific modulatory pathways exist for IFN-ε requires further investigation.
Induction of IFN-α and -β in viral infections is dependent on pathogen sensing by cell-surface and intracellular PRRs. 22Previous studies indicate that IFNε is unlikely to be subject to such regulations. 3Yet, it remains uncertain whether IFN-ε could be upregulated as a secondary effect to viral entry by cytokines and other immunoregulatory molecules in the tissue microenvironment or via interactions with other immune or nonimmune cell types.Reports have suggested that treatment with TNF-α upregulates IFN-ε expression by modulating mRNA stability, 11 suggesting the potential involvement of multiple regulatory mechanisms operating at different levels.
This study also investigates the effects of other potential regulators, including sex hormones.Some of our observations were consistent with previous studies.IFNE expression did not demonstrate marked fluctuations in different stages of the menstrual cycle.There is a substantial amount of evidence indicating that mucosal immunity of the FRT is heavily regulated by sex hormones to avoid compromising conditions for fertilization and pregnancy. 23Bourke et al. 2 first demonstrated the absence of hormonal regulation in IFN-ε expression in the vagina and ectocervix and attributed this to the lack of progesterone receptors in the lower FRT.We further showed that there was no significant difference nor consistent trends in the expression of other IFNs at different stages of the menstrual cycle.
Intriguingly, we noticed strong IFN-ε signals from the endothelial cells of cervical stroma in both immunofluorescence and immunohistochemical staining, which had largely been overlooked in previous characterizations of IFN-ε expression. 2 It remains to be determined whether these IFN-ε high endothelial cells are vascular or lymphatic.Cytokine secretion is considered to be among the many innate immune functions that endothelial cells share with macrophages. 24,25High IFN-ε expression potentially provides new evidence for the role of endothelial cells in protection against STIs.
Due to the cross-sectional nature of this study, the clinical significance of IFN-ε expression awaits future investigations.IFN-ε has been proposed as a promising therapeutic target for mucosal viral infections, given its protective capacity and unique expression profile. 26arly clinical experiments with IFN-α and -β treatment against HPV infection and associated diseases yielded inconsistent results. 27Unfavorable clinical outcomes may, in part, be explained by the detrimental effects of conventional IFNs in persistent viral infections, which are mostly illustrated in human immunodeficiency virus (HIV) and lymphocytic choriomeningitis virus infections. 26Less is understood regarding the role of IFN-ε in chronic infections.Further follow-up studies are warranted to establish a relation between IFN-ε expression and clinical outcome.
expressed at high levels in cervical tissue regardless of hormonal status.(Α) Cervical biopsy samples (n = 32) were examined for mRNA levels of IFN-α2, IFN-β, IFN-ε, IFN-κ, IFN-ω, IFN-γ, and IFN-λ1.(B) IFN expression levels in cervical samples from patients in follicular (n = 12) and luteal (n = 13) stage and postmenopausal patients (n = 3).IFN expression of postpartum patients and patients undergoing GnRH-α therapy is not shown.F I G U R E 2 (See caption on next page).

F
I G U R E 2 IFN-ε expression is upregulated in HR-HPV positive patients at the mRNA but not protein level.(A) IFN expression levels in cervical samples from HR-HPV negative (n = 10) and HR-HPV positive (n = 22) patients (*p < .05).(B) IFN-ε expression determined in representative samples of HR-HPV negative (n = 10) and HR-HPV positive (n = 22) patients by immunoblotting.(C) Immunofluorescence staining for IFN-ε (red) and DAPI (blue) in cervical tissues of HR-HPV negative and HR-HPV positive patients.(D) Quantitative analysis of immunofluorescence staining for IFN-ε in cervical samples of HR-HPV negative (n = 4) and HR-HPV positive (n = 6) patients.(E) Immunohistochemical analysis of IFN-ε protein levels in cervical biopsies of HR-HPV positive and HR-HPV negative patients.
Statistical correlation between IFN-ε expression and clinical characteristics.
T A B L E 2 a Presented as average ± standard deviation or frequency (percent).b Data analyzed with the Brown-Forsythe test.c Statistical significance.