Relationship between vaginal microecological changes and oncogene E6/E7 and high-risk human papillomavirus infection

Abstract High-risk (HR)-human papillomavirus (HPV) is the leading cause of precancerous cervical lesions in patients with chronic untreated infection. We investigated the relationships among several vaginal microbiological alterations, oncogene E6/E7 expression, and HR-HPV. A total of 1327 women who underwent HPV screening, vaginal microecology determination, and fluid-based thin-layer cytological test were enrolled and classified into the HPV-negative group, the low-risk (LR)-HPV-positive group, and the HR-HPV-positive group. The status of cervical HPV infection, vaginal microecology, and E6/E7 mRNA expression were examined sequentially. The effect of HR-HPV infection on cervical cancer (CC) was meticulously assessed, and associations between HR-HPV infection and vaginal microecology and E6/E7 mRNA were identified. In total 548/1327 patients were HPV positive, including LR-HPV infection (N = 132) and HR-HPV infection (N = 416). Patients in the HR-HPV positive group revealed higher detection rates of bacterial vaginosis (BV), trichomonal vaginitis (TV), and vulvovaginal candidiasis (VVC) relative to the HPV negative group. A higher E6/E7 mRNA expression was identified in HR-HPV patients compared to LR-HPV patients. BV and E6/E7 mRNA were classified as independent risk factors for HR-HPV infection. Patients with HR-HPV infection were more susceptible to CC development. Overall, BV and E6/E7 mRNA expression were identified as independent risk factors for HR-HPV infection. IMPACT STATEMENT What is already known on this subject? Through literature review, we found that vaginal ecological changes increase the risk of HPV infection, and HPV persistent infection is an important risk factor for cervical precancerous lesions and cervical cancer. In addition, HPV gene E6/E7 is expressed in HPV-positive cervical cancer cells, which is related to cell malignant transformation and even tumorigenesis. What do the results of this study add? This study further revealed that bacterial vaginosis (BV) and E6/E7 mRNA were independently correlated with HR-HPV infection, and HR-HPV infection increased the risk of cervical cancer. What are the implications of these findings for clinical practice and/or further research? E6/E7 mRNA detection may be used as a new auxiliary diagnostic index for HR-HPV infection. In addition, this study provides a reference for whether the restoration of vaginal microecological balance in patients with BV undergoing clinical treatment is conducive to HR-HPV regression, and provides theoretical support for the prevention and control of cervical cancer microecological approach and the occurrence and development of cervical cancer.


Introduction
Cervical cancer (CC) is the second most prevalent female malignancy after breast cancer and is associated with the highest morbidity and mortality among all female genital tract tumours (Paskeh et al. 2021).With the gradual identification of human papillomavirus (HPV) infection as the primary cause of CC, it is only cancer with a preventable and treatable aetiology (Dorji et al. 2022).Currently, most countries recommend early CC screening with a combination of cervical cytology, HPV testing (Rajaram and Gupta 2022).
Accumulating evidence has cited that HPV vaccines can significantly reduce HPV infection rates, however, most developing countries have been slow to introduce vaccines (Liang et al. 2019).Therefore, timely and early identification of associated risk factors of HR-HPV infection and treatment of chronic HR-HPV infection is vital for the prevention of CC.
HPV is a small, circular, double-stranded DNA virus whose several subtypes are distinctively classified into low-risk (LR) and high-risk (HR) groups (Medeiros et al. 2020).Additionally, treatment of almost all LR-HPV infections and more than two-thirds of HR-HPV infections can eradicate the infection within 24 months, and only a small proportion of women infected with HPV develop CC (Gravitt 2011).Chronic infection with HR-HPV is a vital cause of CC and precancerous lesions (Rajendra and Sharma 2019).The pathogenesis of HPV infection principally involves overexpression of viral oncoprotein, which can inhibit various cellular proteins and affect biological processes including cell proliferation, cell cycle, and apoptosis (Balasubramaniam et al. 2019).The HPV genome comprises six early genes E1, E2, E4, E5, E6, and E7, and two late genes L1 and L2, encoding the capsid protein (Ruttkay-Nedecky et al. 2013).Among the early genes, E6 and E7 are terminally carcinogenic by inactivation of the tumour suppressor protein p53 and retinoblastoma protein (Rb), respectively (Liu et al. 2018).An existing study by Chai et al. documented a high expression of E6/E7 mRNA in HPVinfected tonsil squamous cell carcinoma (Chai et al. 2015).Additionally, rutin can essentially induce caspase-dependent apoptosis of CC HeLa cells through HPV-E6 and E7 downregulation to elicit an anticancer effect (Pandey et al. 2021).Therefore, further investigation into the relationship between E6/E7 and CC is of vital significance for the prevention and treatment of CC.
Vaginal microecology comprises vaginal microbiota, vaginal anatomy, immune prevention mechanism, and endocrine regulatory factors (Shen et al. 2022).A healthy vaginal microenvironment is the first line of defense against HR-HPV infection and subsequent cervical invasion (Liang et al. 2019), the vaginal microbiome forms a protective bacterial membrane on the mucosal surfaces of the cervix and vagina and further secretes copious amounts of lactic acid, bacteritin, and a biosurfactant to prevent adhesion and promote autophagy and clearance of pathogens (Mitra et al. 2016).Furthermore, a vaginal infection can radically disrupt the vaginal microecological (VM) balance and thus terminate its defense system, increasing the risk and susceptibility of HPV infection and weakening immunity against HPV (Wei et al. 2022).For instance, a study by Zhang et al. demonstrated a notable vaginal microbial imbalance in HPV-positive CC patients (Zhang et al. 2021).Currently, the studies on the relationship between VM alterations and oncogene E6/E7 expression and HR-HPV are limited, and more elementary studies and longitudinal clinical studies are needed for verification.The current study sought to investigate the underlying relationship between vaginal microbiological alterations and oncogene E6/E7 expression and HR-HPV infection.

Ethics statement
This study protocol was in compliance with the Enhancing the QUAlity and Transparency Of health Research (EQUATOR) network guidelines and the ethical standards of the World Medical Association Declaration of Helsinki and the Ethical Guidelines for Clinical Research.This study was conducted with the approval of the Academic Ethics Committee of Ma'anshan Maternal and Child Health Hospital (Approval number: 2020-005).All participants provided signed consent before sampling.

Study subjects
A total of 2598 women who sought treatment in the outpatient department of Ma'anshan Maternal and Child Health Hospital from November 2021 to August 2022 were invited to participate in this study.After the exclusion of 521 cases due to refusal to participate, 146 cases with incomplete clinical history, and 604 cases who did not participate in regular CC screening at any stage or didn't comply with monthly screening, 1327 women (median age 34 years, average age 35.81 ± 8.51 years) were finally included as the subjects of this study.A vaginal microecology test was performed on the vaginal secretions of all subjects, and HPV screening and thin layer cytology of base fluid were performed based on cervical exfoliated cells.The subjects were classified into the HPV negative group (N ¼ 779), the LR-HPV positive group (N ¼ 132), and the HR-HPV positive group (N ¼ 416) accordingly.

Inclusion and exclusion criteria
Inclusion criteria were as follows: (1) patients with a sexual history; (2) all samples were collected during non-menstrual periods; (3) no history of vaginal lavage and drug administration 1 week before sampling; (4) no sexual activity, vaginal irrigation, and surgical history within 24 h; (5) no radiation or chemotherapy was administered before sampling.
Exclusion criteria were as follows: (1) no or incomplete sexual life history; (2) women in pregnancy, lactation, or menopause; (3) patients with co-morbidities, such as diabetes, hypertension, and autoimmune diseases; (4) patients with administration history of antibiotics, sex hormones and immunosuppressants in the past 1 month; (5) patients with vaginal administration and vaginal lavage within 3 days.

HPV examination
Detection and typing of HPV were performed based on the flow fluorescence hybridisation typing protocol using the HPV typing detection kits (TEGEN, Shanghai, China) in strict accordance with the provided instructions.Briefly, HPV-DNA was extracted from the cervical exfoliated cell samples.The extracted DNA was amplified by multiplex polymerase chain reaction with corresponding primers, and fluorescent labelling was performed after hybridisation with nucleic acid.Nineteen HPV subtypes were detected using a flow fluorescence detector (San Jose Medical Products, Shanghai, China), including six LR types (HPV6,11,42,43,81,83) and 13 HR types (HPV16,18,31,33,35,39,45,56,59,66,68,73,82).

VM detection
Vaginal discharge was isolated from 1/3 of the lateral wall of the vagina using two sterile cotton swabs.The collected vaginal secretions were prepared into two smears.One of the smears containing 0.9% normal saline was added for the identification of trichomonas, fungal pseudohyphae, blastospores, spores, and other pathogenic microorganisms using a microscope (Olympus, Tokyo, Japan).Another smear was prepared for Gram staining to visualise vaginal microbiota under an oil microscope (Nugent et al. 1991).
Diagnostic criteria were as follows: (1) flora density: the number of bacteria in at least 10 visual fields was observed under the oil microscope, and observation results were documented as grades I (þ) to grade IV (þþþþ) according to the average number of bacteria in each visual field, which were 1-9, 10-99, more than 100, more than 1000 or covering all visual fields.(2) Flora diversity: the bacterial species in at least 10 visual fields were observed under the oil microscope, and the bacterial species in all visual fields were classified into grades I (þ) to grade IV (þþþþ), which were 1-3, 4-6, 7-9, and �10 species.(3) Pathogenic microorganisms: fungal mycelia, spores, trichomonas, and other pathogenic microorganisms were detected based on the microscope wet slide method.If any of the fungal spores and mycelia were identified, vulvovaginal candidiasis (VVC) could be subsequently diagnosed.Trichomonal vaginitis (TV) can be diagnosed if active trichomonas vaginalis is observed.The diagnosis of Aerobic vaginitis (AV) is determined using a wet mount microscopy, ideally by phase contrast (Donders et al. 2002).Based on parameters including inflammation presence, lactobacillary grade, toxic leucocyte proportion, microflora characteristics, and immature epithelial cell presence, the AV score is calculated (Donders et al. 2017).AV scoring criteria: 0-2 points for no AV, 3-4 points for mild AV, 5-6 points for moderate AV, and 7-10 points for severe AV (Vieira-Baptista et al. 2016).Bacterial vaginosis (BV): the BV was diagnosed by Gram staining Nugent score and was defined as normal (0-3 points), intermediate BV (4-6 points), and BV (�7 points).All results were independently evaluated and the Nugent and AV scores were given by two experienced cell counting experts, and the final results were estimated by mutual agreement (Donders 2010).

HPV E6/E7 mRNA detection
The Autocyteprep liquid-based thin-layer cell system-specific brush was inserted into the subject's uterus and rotated for 8-10 laps to isolate the cervical and cervical duct exfoliated epithelial cells.The brush head was then placed in a vial containing CytoRich preservation solution (Yi Bai Biotechnology, Zhejiang, China) for HPV E6/E7 mRNA detection.Briefly, the liquid-based cytology specimens were transferred in a centrifuge tube and horizontally centrifuged twice (300 g, 5 min) to homogenise the specimens.Next, 600 lL of cell lysate and 5 lL of proteinase K (Yi Bai Biotech) were added to the sample, blown and mixed, and then placed into a 65 � C incubator for 1 h.During incubation, the specimen tube was removed and shaken 2-3 times to completely lysate the cells.In strict accordance with the provided instructions in the cervical homeostasis kit, relevant solutions were prepared and subjected on plates, and the positive control solution and blank control solution were set with duplicated wells.After signal amplification, the substrate labelled with the fluorescent substance is supplemented, and the data were detected and analysed on a cold light metre (DiaCarta, CA, USA) (Sun et al. 2021).

Follow-up
HPV-infected patients were monthly screened for CC and followed up for 6 months by consultation with the hospital electronic medical record system and pathology.

Statistical analysis
The sample size required for the independent sample t-test and Chi-square test was estimated using G Power, respectively (Supplementary Figure 1).The SPSS21.0 statistical Software (IBM Corp. Armonk, NY, USA) and GraphPad Prism 6.0 Software (GraphPad Software Inc., San Diego, CA, USA) were adopted for data statistical analysis and plotting.Data comparison between groups was analysed by the t-test.The VM evaluation index was the count data and expressed by the rate, and the chi-square test was used for data comparison.Logistic regression analysis was employed to calculate the odds ratio (OR) value and 95% confidence interval (CI) to evaluate the strength of the relationship between the associated risk factors and HR-HPV infection.The test level was a ¼ 0.05.The p-value was obtained from the bilateral test.In all statistical references, a difference of p < 0.05 was considered to be statistically significant.

Baseline data characteristics of the enrolled population
The clinical data of 1327 women who underwent HPV screening for vaginal secretions were documented and analysed (Table 1).Among them, 779 cases were HPV negative, with an average age of 35.68 ± 7.06 years.There were 132 LR-HPVpositive patients, with an infection rate of 9.94% and an average age of 34.90 ± 11.23 years.A total of 416 cases were classified as positive for HR-HPV, with an infection rate of 31.34% and an average age of 36.37 ± 9.22 years.No significant differences were observed in parameters, such as age, body mass index (BMI), and the number of pregnancies among the groups (p > 0.05), while some significant differences were evident in education level and cleanliness (p < 0.05).Additionally, a total of 19 subtypes were detected, and the detection rate of each subtype is shown in Supplementary Table 1.Among them, 53 patients were infected with 2 or more types of HPV.

Relationship between VM abnormalities and HR-HPV
The vaginal secretions of the subjects were examined (Table 2).The results manifested that the proportion of BV-positive patients in the HR-HPV-positive group was 35.34% (147/416), which was notably higher than that of the HPV-negative group 20.41% (159/779) (p < 0.0001).A significantly smaller proportion of patients with BV-positive vaginal secretions were identified in the LR-HPV positive group (23.48%, 21/ 132) compared to the HR-HPV-positive group (p ¼ 0.0139).The proportion of AV-positive women was 2.70% (21/779) in the HPV-negative group and 4.09% (17/416) in the HR-HPVpositive group; no significant difference was evident between the two groups.The proportion of AV-positive patients in the LR-HPV positive group was 1.52% (2/132), which exhibited no considerable differences from the HR-HPV positive group (p ¼ 0.2716).The proportion of women with VM imbalance in the HPV negative group (14.75%, 115/779) was lower than that in the HR-HPV positive group (93.03%, 387/416) (p < 0.0001).The proportion of patients with VM imbalance in the LR-HPV positive group was 93.18% (123/132), and no significant difference was identified between the LR-HPV and HR-HPV positive groups (p > 0.9999).Our results revealed that the HPV-negative group had a lower proportion of TVpositive patients (2.95%, 23/779) relative to the HR-HPV-positive group (10.58%, 44/416) (p < 0.0001).The proportion of TV-positive patients in the LR-HPV positive group was 6.06% (8/132), which was not markedly different from the proportion in the HR-HPV positive group (p ¼ 0.1713).The proportion of VVC-positive patients in the HR-HPV positive group (8.17%, 34/416) was markedly higher than that in the HPVnegative group (3.72%, 29/779) (p ¼ 0.0016).The LR-HPV positive group was identified with a lower proportion of VVCpositive patients (2.27%, 3/132) than the HR-HPV group (p ¼ 0.0164).Altogether, the detection rates of VM indicators BV, AV, TV, and VVC in patients with HR-HPV infection were significantly increased, accompanied by an elevated proportion of VM imbalance.

Oncogene (E6/E7) is highly expressed in HR-HPV patients
Subsequently, the E6/E7 mRNA expression pattern was examined in liquid cytology samples from 548 patients, including 132 LR-HPV-positive patients and 416 HR-HPV-positive patients.The results were illustrative of a higher expression pattern of E6/E7 mRNA in patients infected with HR-HPV than in patients infected with LR-HPV (p < 0.001, Figure 1).

BV and E6/E7 were independently correlated with HR-HPV infection
To further explore a potential correlation between vaginal microecology and E6/E7 mRNA with patients infected with   LR-HPV and HR-HPV, we initially analysed the correlation between vaginal microecology indicators and E6/E7 mRNA and HR-HPV infection by logistic univariate regression analysis.Subsequently, as the results indicated, parameters, such as age, BV, TV, VVC, and E6/E7 mRNA (p < 0.15) were included in the multivariate logistic regression analysis as independent variables.The results revealed independent associations of BV (p ¼ 0.016, OR ¼ 2.328, 95% CI ¼ 1.172-4.622)and E6/E7 mRNA (p < 0.001, OR ¼ 1.03, 95% CI ¼ 1.025-1.036)with HR-HPV infection (Table 3).Collectively, our results elicited that patients with high BV and E6/E7 mRNA expression were at a higher risk of HR-HPV infection.

HR-HPV infection increased the risk of CC
Furthermore, the incidence of CC in the LR-HPV group and HR-HPV group were compared within 6 months.Kaplan-Meier analysis showed a left shift in the curve of the HR-HPV group (p ¼ 0.0474, Figure 2), indicating that patients with HR-HPV infection had a higher risk of CC during the same follow-up time.

Discussion
CC is the fourth most prevalent cancer among women worldwide, with 570,000 newly diagnosed cases and 311,000 deaths reported in 2018 (Bray et al. 2018).The primary causative factor of CC is chronic infection with HPV (Tonhajzerova et al. 2019).An existing study identified a close relationship between the expression of HPV oncogenes E6 and E7 and the occurrence and development of cervical tumours (Rho et al. 2020).Additionally, the integrity of the vaginal microbiome is also strongly associated with the ability of HPV to facilitate the development of CC (Liu et al. 2022).Our findings collectively identified BV and E6/E7 mRNA expression as independent risk factors of HR-HPV infection.
HPV is classified into low-risk and high-risk types based on nucleotide sequences and differences in carcinogenesis (Mariani et al. 2010).Additionally, the number of vaginal deliveries has been identified as a notable risk factor for HPV (Chen et al. 2019).However, a high level of education, rich sources of information about HPV infection received from physicians, and comprehensive knowledge about HPV are essential to prevent the contraction of HPV (Chen et al. 2013).In our study, the proportion of HR-HPV women with junior high school education or below was evidently higher than that of women with senior high school education or above; the proportion of HR-HPV patients with 1 and 2 or more pregnancies was markedly higher than those without pregnancy history.Similarly, numerous studies have elicited a correlation between women with less education and a higher number of deliveries and a higher HPV-positive rate (Rocha-Brischiliari et al. 2014, Lyu et al. 2019), thus speculating that the HR population may be highly susceptible to CC development.
Emerging research has highlighted the significance of vaginal microecology as a protective barrier in preventing HPV infection and expediting HPV clearance, and the vaginal environmental imbalance might be an exacerbating factor of HPV infection (Lee et al. 2013, Parkin et al. 2020).Currently, some types of vaginal infection, especially the conventional indicators of vaginal infection BV, VVC, and TV are essential for the analysis of the vaginal microenvironment (Teng and Hao 2020).Subsequent experimentation in our study revealed notably increased detection rates of BV, AV, TV, and VVC in patients with HR-HPV infection, with a marked elevation in the proportion of VM imbalance.Consistently, an existing study has documented a higher incidence of VM imbalance in women with HR-HPV infection for 6 months or longer than in healthy individuals (Zhang et al. 2022).Altogether, the aforementioned finding elicited a correlation between women with HR-HPV infection and a higher incidence of VM imbalance than uninfected women.
The aetiology of CC is closely associated with the underlying HR subtypes of HPV, with notable involvement of E6 and E7 oncoproteins in early dysplasia and alterations in epithelial cells (Rho et al. 2020).Moreover, the development of HPV infection from benign to transforming infection is evident  with an increase in HPV E6/E7 mRNA and protein expression (Doorbar 2007).Similarly, our results demonstrated a markedly elevated expression of E6/E7 mRNA in patients infected with HR-HPV compared to patients infected with LR-HPV.Previously, BV has been identified as a mixed infection characterised by lactobacillus reduction and Gardnerella vaginalis proliferation, accompanied by high vaginal pH (Lin et al. 2022).Gillet et al. revealed that BV patients were more susceptible to HR-HPV infection, while Guo et al. further determined that an active BV infection radically delayed the duration and resolution time of HR-HPV infection (Gillet et al. 2011, Guo et al. 2012).Essentially, HR-HPV can induce CC via modulation of E6 and E7 gene expressions (Sui et al. 2019).
Expectedly, subsequent experimentation in our study revealed an independent correlation between BV and E6/E7 mRNA with HR-HPV infection.An existing study has demonstrated a close relationship between BV infection and HR-HPV persistence (Lin et al. 2022).Coordinately, the presence of HPV E6/E7 mRNA is suggestive of an elevated risk of CC development, as it is indicative of an active or progressive HPV infection (Tiiti et al. 2022).Collectively, patients with high BV and E6/E7 mRNA expression are more susceptible to HR-HPV infection.Furthermore, we compared the incidence of CC between LR-HPV patients and HR-HPV patients within 6 months.Interestingly, our results were indicative of a higher risk of CC among patients with HR-HPV infection.In accordance with our findings, chronic infection with one or more HR-HPV types substantially increases the risk of intraepithelial neoplasia and CC (Flatley et al. 2014).Consistently, the presence of HPV in the abdominal cavity is a clinical marker of locally advanced CC and neoplasia (Bizzarri et al. 2021).The latest research has demonstrated that early CC can be identified using specific biomarkers (Valenti et al. 2017).For instance, hypermethylation of the hLHX6 and CCNA1 genes increases with the occurrence and development of CC (Liu et al. 2017).TNF-a polymorphism is a critical risk marker for the development of invasive CC, additionally to identify very early locally invasive cancers (Sousa et al. 2014).
In conclusion, this study investigated the relationship between vaginal microecology and HR-HPV infection from the morphological and functional aspects of vaginal microecology, further identifying BV and E6/E7 mRNA as independent risk factors for HR-HPV infection, where HR-HPV infection increased the risk of CC.E6/E7 mRNA detection could potentially be utilised as a new auxiliary diagnostic index for HR-HPV infection.Additionally, this study provides a comprehensive reference for whether the restoration of vaginal microecological balance in patients with BV undergoing clinical treatment is conducive to HR-HPV regression, and provides theoretical support for the prevention and regulation of CC microecological mechanism and the occurrence and development of CC.The findings of this study should be interpreted in light of the following limitations.Firstly, this is a single-center study, hence the possibility of potential selection bias cannot be ruled out.Secondly, a limited number of samples, especially for patients with LR-HPV infection and HR-HPV infection, with a brief and short follow-up period.Finally, this study simply identified the correlation between VM abnormality and E6/E7 mRNA with HR-HPV infection, however, the underlying mechanism remains elusive and warrants investigation.Our future research will be conducted with bigger sample sizes and extended follow-up periods to increase the reliability and validity of the results, and further, explore the fundamental mechanism between vaginal microecology and HR-HPV infection.

Ethical approval
This study protocol was in compliance with the Enhancing the QUAlity and Transparency Of health Research (EQUATOR) network guidelines and the ethical standards of the World Medical Association Declaration of Helsinki and the Ethical Guidelines for Clinical Research.This study was conducted with approval of the Academic Ethics Committee of Ma'anshan Maternal and Child Health Hospital (Approval number: 2020-005).All participants provided signed consent before sampling.
Note.Chi-square test was used to analyse the data.a Comparison between HPV negative group and LR-HPV positive group.b Comparison between HPV negative group and HR-HPV positive group.c Comparison between LR-HPV positive group and HR-HPV positive group.

Figure 2 .
Figure 2. Analysis of the effect of HR-HPV infection on the cumulative incidence of CC.Kaplan-Meier method was used to analyse the effect of HR-HPV infection on CC, and the curve of the HR-HPV group shifted to the left compared with the LR-HPV group.

Table 1 .
General information of the enrolled population.Note.Measurement data expressed as mean ± standard deviation were analysed by t-test.The count data represented by (%) were analysed by chi-square test.a Comparison between HPV negative group and LR-HPV positive group.b Comparison between HPV negative group and HR-HPV positive group.c Comparison between LR-HPV positive group and HR-HPV positive group.

Table 2 .
Comparison of an abnormal rate of vaginal microecology.