Human papillomavirus type 16 E7 promotes cell viability and migration in cervical cancer by regulating the miR-23a/HOXC8 axis

Abstract

Background

Human papillomavirus (HPV) is a risk factor for the occurrence of cervical cancer (CC). Here, we aimed to explore the role of HPV16 in CC and identify the underlying mechanism.

Methods

The expression of miR-23a, HPV16 E6/E7 and homeobox C8 (HOXC8) was measured by quantitative real-time PCR or western blot. Cell viability and migration were evaluated using cell counting kit-8, Transwell and wound healing assays. The targeting relationship between miR-23a and HOXC8 was revealed by dual-luciferase reporter assay.

Results

miR-23a was downregulated in HPV16-positive (HPV16+) CC tissues and HPV16+ and HPV18+ cells. Additionally, E6/E7 expression was increased in CC cells. Then, we found that E7, rather than E6, positively regulated miR-23a expression. miR-23a suppressed cell viability and migration, whereas E7 overexpression abrogated this suppression. miR-23a targeted HOXC8, which reversed miR-23a-mediated cell viability and migration.

Conclusions

HPV16 E7-mediated miR-23a suppressed CC cell viability and migration by targeting HOXC8, suggesting a novel mechanism of HPV-induced CC.

Plain Language Summary

Cervical cancer (CC) is a common gynaecological malignancy, and persistent human papillomavirus (HPV) infection, especially HPV16, is a main cause of CC. In this study, we explored the role of HPV16 in CC and the molecular mechanism. We used in vitro study to measure CC cell biological behaviours mediated by HPV16 E7, miR-23a and homeobox C8 (HOXC8). We found that HPV16 E7 promotes CC cell viability and migration. miR-23a expression is decreased in CC cells and inhibits cell viability and migration. HOXC8 is a target of miR-23a that reversed the effects on cellular processes caused by miR-23a. These results showed that miR-23a and HOXC8 may be the therapeutic targets of HPV16 E7-infected CC. What is more, our findings provide new insights into the progression of CC.

Keywords:

• Cervical cancer
• human papillomavirus type 16 E7
• HOXC8
• miRNA-23a

Introduction

Cervical cancer (CC) is a common fatal malignancy of women worldwide. There are about 600,000 new cases of CC worldwide each year, and about 300,000 deaths (Castle et al. 2021). CC is preventable, and early screening significantly reduces the incidence and mortality of the disease. Especially with the development of emerging biospectroscopy, FTIR spectroscopy can be used to screen for cervical precancerous lesions due to its ability to separate different grades of dysplasia (Purandare et al. 2013). So, more than 85% of case deaths occur in developing countries (Bedell et al. 2020). In addition, tumour metastasis and recurrence lead to incurable CC and poor outcomes (Li et al. 2016, Wendel and Leath 2020). Hence, it is essential to clarify the pathology of CC and explore novel therapeutic strategies.

Human papillomavirus (HPV) infection is a direct high-risk factor for CC, and in particular HPV16, the most carcinogenic subtype, is associated with over 50% of CC cases (de Martel et al. 2017). HPVs are epitheliotropic DNA viruses that can infect the skin, mouth and reproductive tract (Chelimo et al. 2013). Therefore, HPV infection can not only induce CC but also be associated with other tumours, such as other anogenital cancers and oropharyngeal cancers (de Martel et al. 2017). Early HPV infection is brief and asymptomatic, which can be cleared by the host immune system, making it difficult to drive cancer occurrence (Hu and Ma 2018). Persistent HPV infection may take more than a decade to develop cervical lesions (Hu and Ma 2018, Johnson et al. 2019). Viruses bind their own genome to host DNA and replicate at consistently low levels to prevent detection by the immune system. At the later stage, the virus enters the vegetative expansion phase, assembling into complete virus particles (Shanmugasundaram and You 2017). HPV mainly infects poorly differentiated, basal keratinocytes within stratified squamous epithelia, interfering with multiple signalling pathways to avoid immune surveillance (Yuan et al. 2021). Indeed, studies have reported that HPV-infected cells conspire with normal resident cells to induce the formation of a supportive and immunosuppressive post-infection microenvironment. In this environment, the HPV can continue to multiply and promote the occurrence of CC (Cao et al. 2020, Cigno et al. 2020). Moreover, HPV infection causes immunosuppression and immune escape, accelerating CC development (Shamseddine et al. 2021). Therefore, exploring the interaction pattern of HPV and CC and identifying possible targets may provide new ideas for the prevention and early treatment of CC.

microRNAs (miRNAs) are small non-coding RNAs with 19–25 nucleotides that regulate gene expression post-transcription. It has been reported that miRNAs are associated with the pathogenesis of human diseases, including CC (Shen et al. 2020, Mitra and Elangovan 2021). Currently, targeted therapeutics against miRNAs have been gradually used in clinical trials, showing good effects in the treatment of cancer and viral infections (Saliminejad et al. 2019). Large amounts of miRNAs are differentially expressed in the tumour tissues of most HPV-positive CC patients (Sadri et al. 2020). Among them, miR-23a is lowly expressed in CC tissues (Sadri et al. 2020). However, whether HPV can influence CC progress by regulating miR-23a has not been clarified.

Homeobox C8 (HOXC8) is a member of the HOX family which contains 39 transcription factors. Aberrant expression of HOXC8 is involved in the progression of several cancers, including breast (Shah et al. 2017), gastric (Gu et al. 2021), lung (Su et al. 2022) and CC (Alami et al. 1999). HOXC8 is upregulated in CC and is a prognostic biomarker (Huang et al. 2018). Thus, it is interesting to focus on the role of HOXC8 in CC.

In this study, we investigated the effects of HPV16 on the biological functions of CC cells and molecular mechanisms. The effects of HPV16 E7, miR-23a and HOXC8 on CC cell viability and migration were measured. This will provide a new pathologic mechanism for HPV-mediated carcinogenesis.

Materials and methods

Clinical specimens

Cervical cancer tissues and adjacent non-tumour tissues were obtained from patients with CC (n = 31) who were diagnosed in Heilongjiang Provincial Hospital. Written informed consent was provided by each patient. This study was approved by the Ethics Committee of Heilongjiang Provincial Hospital (approval no. SYYLLBA202140). The tumour tissues were divided into HPV16-positive (HPV16+) tissues (n = 19) and HPV16-negative (HPV16–) tissues (n = 12). All tissues were stored at −80 °C until use. HPV16 infection was evaluated using an HPV DNA detection kit (Acon, Hangzhou, China).

Cell culture

Cervical cancer cell lines (C33A(HPV16–), HeLa (HPV18+), SiHa (HPV16+) and CaSki (HPV16+)) were purchased from the ATCC (Manassas, VA). All these cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM; Gibco, Thermo Fisher Scientific, Waltham, MA) supplemented with 10% foetal bovine serum (FBS; Gibco, Carlsbad, CA). The cells were maintained at 37 °C with 5% CO₂.

Cell transfection

E6 overexpression vector, E7 overexpression vector, E6/E7 overexpression vector, HOXC8 overexpression vector and empty vector were synthesised by GenePharma (Shanghai, China). Small interference RNAs (si-E6#1, si-E6#2, si-E7#1, si-E7#2, si-E6/E7 and negative mismatched control si-NC), miR-23a mimic and negative control mimic (miR-NC) were obtained from Ribobio (Guangzhou, China). Transient transfection was performed using Lipofectamine 2000 (Invitrogen, Thermo Fisher Scientific, Waltham, MA). Following 48 h, cells were harvested. The siRNA sequences are listed in Table 1.

Table 1. siRNA sequences used for cell transfection.

Name	Sequences (5′–3′)
Si-E6#1	CAGTTATGCACAGAGCTGCAAACAA
Si-E6#2	CACAGAGCTGCAAACAACTATACAT
Si-E7#1	CAACCAGAGACAACTGATCTCTACT
Si-E7#2	CAGAACCGGACAGAGCCCATTACAA
Si-E6/E7	GGUUGUGCGUACAAAGCAC
Si-NC	TTCTCCGAACGTGTCACGT

Reverse transcription-quantitative PCR (qPCR)

TRIzol^® reagent (Invitrogen; Thermo Fisher Scientific, Waltham, MA) was used to extract total RNA from clinical tissues per the manufacturer’s instructions. Reverse transcription was performed using iScript™ cDNA Synthesis Kit (Bio-Rad, Hercules, CA). Subsequently, qPCR was conducted by a PowerTrack™ SYBR Green Master Mix (Applied Biosystems, Thermo Fisher Scientific, Waltham, MA) on a 7300 real-time PCR system (Applied Biosystems, Foster City, CA). Relative expression was calculated using the 2^−ΔΔCT method. miR-23a expression was normalised to U6, and mRNA expression was normalised to GAPDH. The primer sequences are shown in Table 2.

Table 2. Primer sequences used for qPCR.

Name	Sequences (5′–3′)
HPV16 E6 (forward)	AAGCAACAGTTACTGCGACGTGAGG
HPV16 E6 (reverse)	GGTCGGTGGACCGGTCGATGT
HPV16 E7 (forward)	CCAGCTGGACAAGCAGAACCGGA
HPV16 E7 (reverse)	GTGCCCCATCTGTTCTCAGAAACCA
HPV16 E6/E7 (forward)	CCGATTTCGGTTACGCCCTTAG
HPV16 E6/E7 (reverse)	CCATCTGTTCTCAGAAACC
HPV18 E6/E7 (forward)	AGCGCGCCATAGTATTGTGGT
HPV18 E6/E7 (reverse)	TTGCAGGAGGCAGCACAGAA
miR-23a (forward)	CGGATCACATTGCCAGGG
miR-23a (reverse)	CAGTGCGTGTCGTGGAGT
U6 (forward)	AAAGCAAATCATCGGACGACC
U6 (reverse)	GTACAACACATTGTTTCCTCGGA
HOXC8 (forward)	GAACCCGTGCTCGCTTAG
HOXC8 (reverse)	ACGCTCGCCTCTTGCTGA
GAPDH (forward)	GCAAGTTCAACGGCACAG
GAPDH (reverse)	CGCCAGTAGACTCCACGAC

The expression of E6/E7 in CC cells was detected using the HPV E6/E7 mRNA kit (Kodia, Zhengzhou, China) according to the manufacturer’s protocol.

Western blotting

C33A, SiHa and CaSki cells were lysed using RIPA lysis buffer (Beyotime, Shanghai, China). The isolated proteins were separated using SDS-PAGE and transferred onto PVDF membranes. Following blocking in 5% skim milk for 1 h, the membranes were incubated with anti-HPV16 E6 (bs-49102R, 1:1000, Bioss, Beijing, China), anti-HPV16 E7 (bs-10202R, 1:1000, Bioss, Beijing, China), anti-HOXC8 (bs-0394R, 1:500, Bioss, Beijing, China) and anti-GAPDH (bs-41373R, 1:5000, Bioss, Beijing, China) at 4 °C overnight. The next day, the membranes were incubated with HRP labelled mouse anti-rabbit IgG (bsm-0295M-HRP, 1:1000, Bioss, Beijing, China) at room temperature for 1 h. The bands were visualised using BeyoECL Plus (Beyotime, Shanghai, China).

Determination of cell viability

Cell viability was evaluated using a Cell Count Kit-8 (CCK-8; Dojindo, Kumamoto, Japan) according to the manufacturer’s instructions. The transfected cells were seeded into 96-well plates (2000 cells/well) and incubated for 48 h. Then, the cells were treated with 10 μL CCK-8 reagent for 1 h. The absorbance per well was measured by a microplate reader at 450 nm.

Determination of cell migration

Transwell assay and wound healing assay were conducted to assess cell migration. For Transwell assay, SiHa and CaSki cells (5 × 10⁴ cells/chamber) were added into the upper layer of 24-well Transwell chambers (8-μm pore, Corning, Corning, NY). Serum-contained DMEM filled the bottom chambers. After incubation for 24 h at 37 °C, cells that cross the membrane were fixed with 4% formaldehyde and stained with 0.1% crystal violet. The number of invasive cells was counted under an inverted light microplate.

For wound healing assay, SiHa and CaSki cells (6 × 10⁵) were seeded into six-well plates and grown until 100% confluence. The wound was made using a sterile pipette tip. Cell debris was removed using PBS. The images were taken at 0 and 24 h after incubation.

Dual-luciferase reporter assay

The binding sites between miR-23a and HOXC8 were predicted using the TargetScan database. Wild-type (wt) and mutant (mut) sequences at 3′UTR of HOXC8 were amplified using PCR and cloned into the pGL3 vector (Promega, Madison, WI). SiHa and CaSki cells were co-transfected with wt or mut recombined reporter plasmids together with miR-23a mimic or miR-nc using Lipofectamine 2000. A dual-luciferase reporter assay kit (Promega, Madison, WI) was used to detect luciferase activity. Renilla luciferase activity was the internal control of firefly luciferase activity.

Statistical analysis

All data are shown as the mean ± standard deviation from three independent experiments. One-way analysis of variance was used to compare the differences among groups. p < .05 indicated a statistically significant difference.

Results

E7 regulates the expression of miR-23a

First, the expression of miR-23a was detected in patients with CC. As shown in Figure 1(a), miR-23a was downregulated in tumour tissues, especially in HPV16+ tissues. Then, we examined the expression of miR-23a in different cell lines. miR-23a was down-regulated in HPV-positive CC cell lines (HeLa, SiHa, and CaSki), compared to HPV– C33A cells (Figure 1(b)). E6/E7 was not expressed in C33A cells. Compared with HeLa cells, E6/E7 expression in HPV16+ cells including SiHa and CaSki was markedly elevated (Figure 1(c)). To explore which HPV16 regulated miR-23a, we overexpressed or knocked down E6 or E7. The results showed that both E6 and E7 mRNA and protein levels were increased in C33A cells after transfection of their overexpression vectors (Figure 1(d,e)). Additionally, E6 and E7 mRNA expression and protein levels were decreased in SiHa and Hela cells after transfection of corresponding siRNAs, especially si-E6#2 and si-E7#2 transfection (Figure 1(f,g)). Overexpression of E7, but not E6, reduced miR-23a expression in C33A cells (Figure 1(h)). Knockdown of E7, rather than E6, elevated miR-23a expression in SiHa and CaSki cells (Figure 1(i)). The results showed that miR-23a expression was regulated by E7 in CC cells.

Figure 1. E7 regulates the expression of miR-23a. (a) Expression of miR-23a in adjacent normal tissues, and HPV16+ and HPV16– tumour tissues. (b) Expression of miR-23a in CC cell lines (C33A, Hela, SiHa and CaSki). (c) Expression of E6/E7 in CC cell lines (C33A, Hela, SiHa and CaSki). (d) Transfection efficiency of E6 overexpression vector in C33A cells. (e) Transfection efficiency of E7 overexpression vector in C33A cells. (f) E6 mRNA and protein levels in SiHa and CaSki cells after si-E6#1 and si-E6#2 transfection. (g) E7 mRNA and protein levels in SiHa and CaSki cells after si-E7#1 and si-E7#2 transfection. (h) Expression of miR-23a in C33A cells after E6/E7, E6 and E7 overexpression. (i) Expression of miR-23a in SiHa and CaSki after E6/E7, E6 and E7 knockdown. **p < .01. ***p < .001.

E7 promotes the viability and migration of SiHa and CaSki cell lines by downregulating miR-23a expression

To verify the effects of E7 and miR-23a on the viability and migration of CC cells, miR-23a mimic and E7 overexpression vectors were transfected into SiHa and CaSki cells. miR-23a expression was elevated in SiHa and CaSki cells following mimic transfection (Figure 2(a)). Overexpression of miR-23a inhibited cell viability, whereas E7 reversed the inhibition of cell viability (Figure 2(b)). The results of Transwell assay and wound healing assay showed that compared with miR-NC, miR-23a suppressed SiHa and CaSki cell migration, which was abrogated by E7 overexpression (Figure 2(c,d)). Taken together, E7 downregulated miR-23a to promote CC cell viability and migration.

Figure 2. E7 promotes the viability and migration of SiHa and CaSki cell lines by downregulating miR-23a expression. (a) The expression of miR-23a in SiHa and Hela cells after miR-23a mimic transfection. (b) Effects of miR-23a and E7 on CC cell viability were determined by CCK-8. Effects of miR-23a and E7 on CC cell migration were assessed using (c) Transwell assay and (d) wound healing assay. **p < .01 vs. the miR-nc + vector group. ^##p < .01 vs. the miR-23a + vector group.

miR-23a targets HOXC8

Subsequently, we used bioinformatic analysis to explore the targets of miR-23a. HOXC8 was predicted to have the binding sites in miR-23a (Figure 3(a)). To confirm the targeting relationship, we designed the mut-HOXC8 sequences (Figure 3(a)). Overexpression of miR-23a decreased the luciferase activity when-co-transfected with wt-HOXC8 into SiHa and CaSki cells, compared with miR-NC, whereas miR-23a or miR-NC did not affect the luciferase activity when co-transfected with mut-HOXC8 (Figure 3(b)). Overexpression of miR-23a decreased HOXC8 expression, while E7 increased its expression (Figure 3(c)). HOXC8 expression was significantly increased in CC tissues, especially in HPV16+ tumour tissues (Figure 3(d)). In addition, HOXC8 was highly expressed in Hela, SiHa and CaSki cells, compared with that in the C33A cells (Figure 3(e)). To sum up, HOXC8 was confirmed as a target of miR-23a.

Figure 3. miR-23a targets HOXC8. The targeting relationship (a) was predicted using the TargetScan database and (b) was confirmed by dual-luciferase reporter assay. (c) Effects of miR-23a and E7 on HOXC8 expression in SiHa and CaSki cells were detected using qPCR. (d) HOXC8 expression in normal and HPV16+ and HPV16– tumour tissues. (e) HOXC8 expression in CC cell lines (C33A, Hela, SiHa and CaSki). **p < .01. ^##p < .01 the miR-23a + vector group.

miR-23a suppresses cell viability and migration by targeting HOXC8

Following HOXC8 overexpression vector transfection, the mRNA and protein levels of HOXC8 were elevated (Figure S1a). Then, cell viability and migration were evaluated. Overexpression of HOXC8 promoted cell viability of miR-23a overexpressed SiHa and CaSki cells (Figure S1b). Additionally, miR-23a suppressed cell migration, whereas HOXC8 overexpression rescued the suppression induced by miR-23a (Figures S1c and d). The results indicated that miR-23a targeted HOXC8 to suppress CC cell viability and migration.

Discussion

In the present study, we found that miR-23a was downregulated in HPV16+ CC tissues and cell lines. The expression of miR-23a was negatively regulated by E7 and targeted to HOXC8. Moreover, E7 mediated-miR-23a inhibited CC cell viability and migration via targeting HOXC8, suggesting that the miR-23a/HOXC8 was involved in the progression of HPV16-induced CC.

The role of miR-23a in cancers has been largely investigated in recent years. The expression of miR-23a is commonly reduced in several tumours, which is mediated by the alteration of epigenetics and biological pathways (Roufayel and Kadry 2017). Moreover, miR-23a regulates cellular biological behaviours in malignancy, including cell growth, death and metastasis. For example, miR-23a reduces the progression of pancreatic cancer through inhibiting cell proliferation, and metastasis, and inducing apoptosis (Chen et al. 2018). In addition, miR-23a is lowly expressed in glioma and inhibits tumour cell proliferation (Gao et al. 2021). Interestingly, miR-23a acts as an oncogene to promote tumour progression, such as colon cancer and ovarian cancer (Zhang et al. 2023, Zuo et al. 2023). The expression and role of miR-23a are uncertain. Shen et al. (2011) have revealed that there is no significant difference in miR-23a expression in normal, HPV+ and cervical intraepithelial neoplasia tissues. However, Fullár et al. (2020) have shown that miR-23a inactivates TFPI2 to be involved in CC progress. Zhou et al. (2022) have reported that miR-23a downregulation facilitates the proliferation, invasion and migration of CC cells. Here, we found that miR-23a expression was decreased in CC tissues and cells. Overexpression of miR-23a suppressed CC cell viability and migration. The results indicated that miR-23a functioned as a tumour promoter in CC, consistent with Zhou et al. (2022) study.

Persistent infection of high-risk HPV, genotypes 16 and 18, is the main cause of CC. HPV16 E6 and E7 drive the initial changes of cervical epithelial cells, thereby leading to carcinoma in situ or invasive CC (Small et al. 2017, Balasubramaniam et al. 2019). Moreover, HPV infection causes changes in the expression of multiple miRNAs in CC, such as miR-34a (Singh et al. 2022), miR-106a (Cui et al. 2020) and miR-485-5p (Dai et al. 2020). A previous study has revealed that silencing of E6/E7 upregulates miR-23a to regulate the growth, apoptosis and senescence of HPV + CC cells (Honegger et al. 2015). Similarly, we found that miR-23a was negatively regulated by E6/E7. In more detail, miR-23a expression was regulated by E7 rather than E6. Overexpression of E7 reversed the inhibition of cell migration and viability induced by miR-23a, suggesting that E7 induced CC progress in a miR-23a-dependent manner.

HOXC8 was confirmed to be a target of miR-23a in CC cells. HOXC8 is an oncogene that is associated with the prognosis and drug resistance in malignancy (Xu et al. 2015, Huang et al. 2018). HOXC8 promotes tumour cell biological functions, including proliferation, migration and epithelial–mesenchymal transition (EMT), thereby facilitating tumour development (Liu et al. 2018, Fang and Yan 2019, Gong et al. 2019). However, the role of HOXC8 in CC remains not yet expounded. A previous study has identified that HOXC8 expression is increased in patients with CC, especially dysregulated in HPV16+ CC. E7 elevates the expression of HOX gene and regulates the EMT of CC cells (Saha et al. 2017). Herein, we found that E7 regulated CC cell viability and migration via the miR-23a/HOXC8 axis, demonstrating a novel underlying mechanism of carcinogenicity of E7 in CC.

This study provides a theoretical basis for miR-23a and HOXC8 to be potential therapeutic targets for HPV16 E7-infected patients with CC. This indicates the direction for future animal experiments and clinical trials.

In conclusion, miR-23a was lowly expressed in CC tissues and cells, accompanied by the upregulation of E6/E7 and HOXC8. Overexpression of miR-23a, which was regulated by silencing of E7, inhibited the viability and migration of CC cells via targeting HOXC8. The findings suggested an important role of the HPV16 E7/miR-23a/HOXC8 axis in CC.

Author contributions

YC and LS conceived the study; YC and LL conducted the experiments; YC, LS and LL analysed the data; YC was a major contributor in writing the manuscript. All authors read and approved the final manuscript.

Ethical approval

This study was approved by the Ethics Committee of Heilongjiang Provincial Hospital.

Consent form

Written informed consent was provided by each patient.

Data availability statement

All data included in this study are available upon request by contact with the corresponding author.

Additional information

Funding

This study was supported by Science and Technology Program of Heilongjiang Provincial Health Commission (20210505010139).