Upregulation of miRNA-10a-5p promotes tumor progression in cervical cancer by suppressing UBE2I signaling

Abstract

Cervical cancer (CC) is a common malignant neoplasm in gynecology. There is increasing evidence to suggest that microRNAs (miRNAs) act as crucial regulators of CC. However, whether miR-10a-5p plays a role in CC is under investigation. The aim of this stuy was to assess the miR-10a-5p expression pattern in the development of CC and investigate its downstream target. MiR-10a-5p inhibition decreased CC cell proliferation and impaired CC cell invasion and migration but enhanced apoptosis. UBE2I was a direct target of miR-10a-5p. QRT-PCR results showed a down-regulation of UBE2I in CC cells, opposing miR-10a-5p. Besides, overexpression of miR-10a-5p down-regulated UBE2I. Functional rescue experiments further indicated the miR-10a-5p-UBE2I axis was linked to CC cell growth, apoptosis and metastasis. MiR-10a-5p upregulation promotes cervical cancer development by inhibiting UBE2I. These results also predict that miR-10a-5p may be a potential target for the clinical treatment of CC.

IMPACT STATEMENT

• What is already known on this subject? As a widely researched cancer-related miRNA, the overexpression of miR-10a-5p has been verified in various cancers. It has been described in a meta-analysis report that there were 42 miRNAs up-regulated and 21 miRNAs down-regulated in different stages of cervical cancer tissue versus healthy tissue.

• What do the results of this study add? We verified that miR-10a-5p initiates and promotes tumor cell development by decreasing UBE2I abundance. This miR-10a-5p-mediated post-transcriptional regulation of UBE2I is involved in the tumorigenesis, invasion and migration of human cervical cancer.

• What are the implications of these findings for clinical practice and/or further research? These findings provide mechanistic insights into how miR-10a-5p regulates cervical cancer hyper-proliferation and metastasis, as well as a new target for clinical treatment. Nevertheless, whether miR-10a-5p/UBE2I axis can be regulated by non-invasive methods need further exploration, which will be the focus of our future research.

Keywords:

• Cervical cancer
• metastasis
• microRNA (miRNA)
• epigenetic regulation
• UBE2I

1. Introduction

Cervical cancer (CC) is a prevalent carcinoma with high morbidity in women. It is also the fourth cause of death globally, following lung cancer, breast cancer and colon cancer (Torre et al. 2015). Persistent infection with oncogenic human papillomavirus (HPV) is a necessary and recognized causal factor. The distribution of cervical cancer shows geographical differences (Mezei et al. 2017, Cascardi et al. 2022, Dellino et al. 2022a, 2022b). According to the Human Papillomavirus And Related Diseases Report in 2018, 106,430 new cases and 47,739 deaths related to cervical cancer were reported in China, responsible for 18.7% of confirmed cases and 15.5% of deaths globally. Despite the governments putting increasing effort into promoting HPV vaccination and routine cancer screening, the rates in many countries remain low (Wang et al. 2020, Giannini et al. 2022). In addition, patients with cervical cancer suffer from a high disease burden and poor prognosis, especially those in the advanced stage (Waggoner 2003). Thereby, it is crucial to detect the underlying principles of the pathogenesis of cervical cancer and find promising therapeutic targets.

Previous reports have illustrated that tumor initiation and development are linked to a wide array of both genetic and epigenetic changes (Villanueva et al. 2020). In tumor cells, numerous epigenetic alterations, such as non-coding RNAs, RNA methylation and DNA methylation, have been found to contribute to tumorigenesis (Liu et al. 2017, Koch et al. 2018, Huang et al. 2021). MicroRNA (miRNA) is a highly conserved type of small non-protein-encoding single-strand RNA, playing an essential role in post-transcriptional regulation (Fabian et al. 2010, Alarcón et al. 2015). One miRNA is considered to be able to modulate the expression of hundreds of genes. What’s more, miRNA expression is specified in different tissues and diseases, thus providing a great diagnostic opportunity for various diseases (Thomou et al. 2017). It has been described in a meta-analysis report that there were 42 miRNAs up-regulated and 21 miRNAs down-regulated in different stages of cervical cancer tissue versus healthy tissue, supporting miRNA’s relevance to neoplasia (Marima et al. 2021).

As a widely researched cancer-related miRNA, the overexpression of miR-10a-5p has been verified in various cancers (Worst et al. 2019, Yang et al. 2021, Zhang et al. 2021). MiR-10a-5p abundance was markedly higher in dysplastic cervical tissues (Sommerova et al. 2019). Research has revealed that miR-10a-5p has the potential to enhance CC cell growth and metastasis (Long et al. 2012). Nevertheless, the underlying mechanism remains unclear. Ubiquitin-conjugating enzyme E2I (UBE2I) is an E2 enzyme family member that is thought to be a critical regulator of immunity (Ihara et al. 2008, Lu et al. 2021). UBE2I is found to impact human adipocyte differentiation (Cox et al. 2021). Furthermore, studies have concluded that UBE2I dysregulation was associated with a poor prognosis in hepatocellular carcinoma (HCC), glioblastoma and prostate cancer (Zhang et al. 2018, Yang et al. 2020, Li and Meng 2021). Though extensively studied in various diseases, its role in cervical cancer has not been well-established.

Thus, we investigate the impacts of miR-10a-5p dysfunction on the biological features of CC cells and found its downstream target UBE2I. The data verified that miR-10a-5p overexpression promoted tumorigenesis by inhibiting UBE2I signaling (Figure S1).

Figure 1. MiR-10a-5p targets UBE2I in CC cells. (A) Binding sites between UBE2I and miR-10a-5p. (B) Luciferase activity in UBE2I WT or Mut transfected Hela cells with or without miR-10a-5p upregulation. (C) UBE2I abundance in CC tissues. (D) The abundance of UBE2I in CC cells. (E) The UBE2I mRNA abundance in Hela cells transfecting the miR-10a-5p mimic. (F) The UBE2I protein abundance in Hela cells transfecting the miR-10a-5p mimic.

2. Methods and materials

2.1. Clinical tissues

All 30 CC tissues and the paired non‐tumorous tissues used in this study were gained from patients with the pathological and clinical diagnoses of CC who underwent surgery at the Affiliated Maternal and Child Health Hospital of Nantong University. Our work was supported by the Ethics Committee of the Affiliated Maternal and Child Health Hospital of Nantong University (Y2021040).

2.2. Cell culture and transfection

Immortalized CC cells (CaSki, SiHa, HeLa and C33A) and cervical epithelial cells (H8) were maintained in a DMEM medium, containing 10% FBS and 1% antibiotics. All the cells were tested negative for mycoplasma and maintained as monolayer cultures in the incubator with 5% CO₂ at 37 °C.

MiR-10a-5p inhibitor and mimics, and sh-UBE2I were constructed by Shanghai GenePharma Co. Transfection into CC cells was carried out by using Lipofectamine^® 3000 (Thermo, Waltham, MA, USA).

2.3. qRT-PCR

RNA of stored human CC tissues and immortalized human CC cells were isolated using TRIzol (Thermo, Waltham, MA, USA). The qPCR RT Kit (Invitrogen, Waltham, MA, USA) was then used to synthesize cDNA. Next, the cDNA abundance was detected on Applied Biosystems^® 7500 (Thermo, Waltham, MA, USA) using the SYBR Green kit (Roche, Basel, Switzerland).

2.4. Western blot analysis

For protein isolation, cells were mixed with RIPA (Beyotime, China) containing PMSF (Roche, Basel, Switzerland). Proteins were then electrophoresed and transferred to nitrocellulose membranes (Bio-Rad, USA). After 2-h blocking, the membranes were reacted with primary antibodies for 14 h at 4 °C, then hybridized with secondary antibodies for 50 min. The abundance of proteins was detected via the ECL solution (Beyotime, China).

2.5. CCK-8 analysis

The CCK-8 solution (MCE, China) was utilized for the measurement of cell viability. 2 × 10³ CC cells were seeded in each well of a 96-well plate. Subsequently, add CCK-8 mixture in each well at different times. Cells were protected from light and incubated for another 2 h at 37 °C. The OD values were detected at 450 nm.

2.6. EdU assay

The proliferative ability of CC cells was detected by EdU Proliferation Kit (Abcam, UK). Briefly, add EdU solution to the cells that need to be treated to obtain an EdU solution at 10 µM end concentration in the tube. Incubate cells with EdU for 2.5 h at 37 °C and then fixed with 4% PFA. Subsequently, cells were counterstained with Hoechst 33342 for 0.5 h at room temperature and protected from light. Finally, all marked cells were analyzed under a microscope.

2.7. TUNEL analysis

TUNEL kit (Abcam, UK) was utilized to calculate the apoptotic CC cells. Briefly, CC cells were reacted in the staining solution for 1 h at 37 °C. Resuspend the cells in PI/RNase A solution and reacted for 0.5 h in the dark. Finally, all marked cells were observed under fluorescence microscopy.

2.8. Wound healing analysis

In a well of the 6-well plate, 8 × 10⁵ CC cells were plated. When the monolayer cell confluence exceeded 80%, a scratch was made with the tip of a 200 µL pipette. Cells were then cultured for another 60 h. Wound healing was measured under the microscope.

2.9. Transwell assay

For migration measurement, 1 × 10⁵ cells were seeded on the upper layer. Then a solution containing DMEM and FBS was poured into the lower layer. After incubating for 1 day, the upper chamber was washed followed by fixed using 4% PFA for 10 min. Migrated cells were then stained and counted under the microscope.

For invasion detection, cells were seeded on the upper layer containing Matrigel. Then a solution containing DMEM and FBS was poured into the lower layer. A day after, cells that invaded the Matrigel were treated as described above.

2.10. Dual-luciferase reporter assay

Construct Mutant UBE2I 3′-UTR (UBE2I MUT) and wild-type UBE2I 3′-UTR (UBE2I WT) vectors. Then, co-transfect UBE2I MUT/WT vectors and NC/miR-10a-5p mimics into Hela cells by Lipo 3000 (Thermo, Waltham, MA, USA). Finally, the activity of luciferase was calculated by the Luciferase assay kit (Invitrogen, Waltham, MA, USA).

2.11. Statistical analysis

Results are depicted as mean ± SEM. Statistical significance was considered as p < 0.05. t-Test and One-way ANOVA were utilized to determine whether there were significant differences between groups.

3. Results

3.1. MiR-10a-5p is widely expressed in human cervical cancer

Firstly, we evaluated the abundance of miR-10a-5p in clinical CC samples and the adjacent normal tissues. As can be seen in Figure S2(A), miR-10a-5p abundance was relatively more in CC tissues than that in the paired healthy cervical samples (p < 0.001). MiR-10a-5p abundance was then examined in a normal cervical epithelial cell line (H8) and four CC cell lines (CaSki (p < 0.05), SiHa (p < 0.01), HeLa (p < 0.001) and C33A (p < 0.05)). In contrast to that in H8, the expressing pattern of miR-10a-5p showed a striking elevation in all CC cell lines (Figure S2(B)). The HeLa cells with the greatest abundance of miR-10a-5p were chosen for our ensuing studies. These data implied that miR-10a-5p was responsible for the tumorigenicity of human cervical cancer.

Figure 2. UBE2I suppression counteracts the influence of miR‑10a‑5p downregulation on CC cell growth. (A) The efficiency of UBE2I knockdown. (B) Cell viability was measured after co-transfecting cells with miR‑10a‑5p inhibitors and sh-UBE2I. (C) Hela cell proliferation after UBE2I knockdown was analyzed. (D) The proportion of apoptotic Hela cells after UBE2I knockdown. (E) The protein abundance of Bax, caspase-3 and -9, as well as Bcl-2 in Hela cells transfecting miR‑10a‑5p inhibitors and sh-UBE2I.

3.2. MiR-10a-5p downregulation hinders cervical cancer cell proliferation

We then focussed on the impact of miR-10a-5p on the growth of CC cell lines. Transfection of miR-10a-5p inhibitors was conducted in Hela cells. The efficacy of the inhibitors (p < 0.01) was confirmed (Figure S3(A)). CCK-8 results revealed that miR-10a-5p inhibitor rather than NC inhibitor restrained (p < 0.01) the viability of Hela cells (Figure S3(B)). The growth of Hela cells was then assessed using the EdU test. Hela cells expressing a small amount of miR-10a-5p exhibited reduced (p < 0.05) proliferative ability (Figure S3(C)). Subsequently, we examined the changes in cell apoptosis after transfection and data illustrated that the miR-10a-5p inhibitor remarkably increased (p < 0.01) the proportion of apoptotic Hela cells (Figure S3(D)). Additionally, the protein abundance of apoptotic activators (cleaved caspase-3 (p < 0.001) and Bax (p < 0.05)) was elevated. While the abundance of apoptotic inhibitor (Bcl-2) was reduced (p < 0.05) after miR-10a-5p inhibitor treatment (Figure S3(E)). Our findings indicated that downregulation of miR-10a-5p restrained Hela cell proliferation.

Figure 3. Inhibition of UBE2I counteracts the influence of miR‑10a‑5p downregulation on CC cell metastasis. (A) Hela cell migration after the knockdown of UBE2I was determined. (B) The metastasis compacity of Hela cells after UBE2I inhibition. (C) Western blot analyzed pro-invasive factors MMP2 and MMP9 abundance after co-transfecting cells with miR‑10a‑5p inhibitor and sh-UBE2I.

3.3. MiR-10a-5p downregulation prevents cervical cancer cells from migrating and invading

Following miR-10a-5p inhibitor transfection, CC cell metastasis was assessed to verify the involvement of miR-10a-5p in Hela cell metastasis. The reduced migratory (p < 0.01) and invasive capacity (p < 0.01) was found in cells transfected with miR-10a-5p inhibitor (Figure S4(A,B)). These findings suggest that miR-10a-5p inhibits the spreading of CC cells. These findings demonstrated the influence of miR-10a-5p on the dissemination of CC cells. MMP-2 along with MMP-9 can promote cancer cell invasion. Data depicted that miR-10a-5p suppression decreased MMP-2 and -9 protein abundance (p < 0.001) (Figure S4(C)). Together, our data showed that miR-10a-5p downregulation prevented cervical cancer cells from migrating and invading.

3.4. MiR-10a-5p targets UBE2I directly

Next, the StarBase (http://starbase.sysu.edu.cn/) database was utilized to investigate the targets of miR-10a-5p regulation, and UBE2I was discovered to be a potential target of miR-10a-5p (Figure 1(A)). The luciferase assay confirmed the connection between miR‑10a‑5p and UBE2I (Figure 1(B)). The result showed that when transfecting with wild-type 3′-UTR of UBE2I, the activity of luciferase was markedly reduced (p < 0.001) in the group transfected miR‑10a‑5p mimic. While there was no discernible variance between the NC and miR-10a-5p groups in CC cells transfected with the mutated 3′-UTR of UBE2I. This meant that miR-10a-5p was bound to the 3′-UTR of UBE2I directly. Subsequently, we detected the abundance of UBE2I in CC tissues and cells and demonstrated a decreased expression pattern of UBE2I in CC tissue (p < 0.001) as well as cell lines (p < 0.001) (Figure 1(C,D)). In addition, miR-10a-5p upregulation decreased the abundance of UBE2I mRNA (p < 0.01) as well as protein (p < 0.001) (Figure 1(E,F)). Our data indicated that miR‑10a‑5p negatively regulated UBE2I abundance by bound to UBE2I 3′-UTR.

3.5. MiR-10a-5p downregulation exerts anti-proliferative and anti-metastatic properties by targeting UBE2I

To determine whether the suppression of UBE2I may counteract the influence of miR‑10a‑5p downregulation on CC progression, a sh-UBE2I plasmid was constructed, which could decrease (p < 0.001) the expression of UBE2I effectively (Figure 2(A)). The effects of sh-UBE2I on the modification of cell survival and apoptosis caused by miR-10a-5p downregulation were then investigated. Inhibition of UBE2I in Hela cells partially overrode the miR-10a-5p inhibitor-induced decline in cell viability (p < 0.05) and proliferation (p < 0.01) (Figure 2(B,C)). The TUNEL assay additionally demonstrated that cells treated with miR-10a-5p inhibitor + sh-UBE2I had a lower (p < 0.01) rate of apoptosis than cells transfected with miR-10a-5p inhibitor (Figure 2(D)). Transfecting sh-UBE2I also counteracted the influence of decreased (p < 0.001) miR-10a-5p on Bax, cleaved caspase-3 and -9, as well as Bcl-2 (Figure 2(E)).

Subsequently, we measured whether UBE2I was also involved in mediating miR-10a-5p’s effect on Hela cell metastasis. The outcomes demonstrated that reduced UBE2I expression undermined the inhibitory effects (p < 0.001) on cell mobility driven by miR-10a-5p suppression (Figure 3(A–C)). Taken together, inhibition of UBE2I mediated by miR-10a-5p was critical for the growth and metastasis of CC cells.

4. Discussion

Cervical cancer has long been the greatest threat to women’s health and quality of life, frequently occurring among patients aged between 40 and 60 years old (Waggoner 2003). In recent years, however, statistics have shown that the age of disease onset tends to be younger (He and Li 2021). The mortality of cervical cancer has been reduced to some extent due to therapeutic advancement (D’Oria et al. 2022). However, tumor metastasis and recurrence are still the primary inducements affecting prognosis. Pre-operative conization before radical hysterectomy could reduce early cervical cancer recurrence rate (Bizzarri et al. 2021b; Casarin et al. 2021), whereas the postoperative complications may impair urinary bladder function (Kietpeerakool et al. 2019). Therefore, exploring the mechanisms related to CC initiation and progression is imperative for finding promising therapeutic regimens.

Infection with high-risk HPV16/18 is the primary causative factor of cervical cancer. The accumulation of genetic mutations and epigenetic alterations in HPV-infected cells has been proven to be crucial regulators in the development of cervical cancer (Szalmás and Kónya 2009). To date, there is some evidence supporting the possibility of using specific biomarkers to identify early cervical cancer and in this way providing a better prognosis for patients (Valenti et al. 2017, Bizzarri et al. 2021a). Emerging evidence shows that miRNA, whose transcription process is independent of other genes, is of great importance to the initiation and metastasis of various malignancies. MiRNAs were first reported to be linked to cancer in 2002 (Bhaskaran and Mohan 2014). The report found that miR-15a and miR-16-1 deletion led to the development of chronic lymphocytic leukemia (CLL), and later studies identified more tumor-associated miRNAs (Bhaskaran and Mohan 2014). MiRNAs are involved in gene expression regulation by binding to mRNA and affecting the transcription process. Most of the coding sequences of miRNAs appear in regions that are related to cancer, so dysfunction of miRNAs affects tumor-related biological processes (Saliminejad et al. 2019).

MiR-10a-5p was previously found overexpressed in colorectal cancer (CRC) as well as acute myeloid leukemia (AML) (Schee et al. 2013, Zhi et al. 2013). In recent years, it has been validated as a promising biomarker of cervical carcinogenesis as its highly aberrant expression in precancerous cervical lesions and peripheral blood of patients (Sommerova et al. 2019). According to the data, we proved the abnormal up-regulation of miR-10a-5p in CC samples, which was aligned with previous studies. Currently, research about the influence of miR-10a-5p on CC cells needs deeper insights. The latest report illustrated that miR-10a-5p encapsulated by extracellular vesicles promotes cell angiogenesis and tumorigenicity in cervical squamous cell carcinoma (Zhang et al. 2021).

For further research, HeLa cells with relatively high miR-10a-5p abundance were utilized for the subsequent experiments. Using miRNA inhibitors, we demonstrated that a lower abundance of miR-10a-5p was responsible for the prevention of cancer cell proliferation and dissemination, as well as the enhancement of cell apoptosis. The mechanisms underlying miR-10a-5p regulation in cervical carcinogenesis are still poorly understood, which impedes the development of targeted therapies. Only a limited number of studies reveal that miR-10a-5p influences human cervical cancer by modulating BDNF expression or activating Hedgehog signaling via targeting TBX5 (Zhai et al. 2017, Zhang et al. 2021). As miRNAs could regulate the expression of hundreds of genes, we carried out a dual-luciferase reporter assay in hopes of finding novel miRNA targets. Our data displayed that miR-10a-5p could bind to UBE2I mRNA, decreasing UBE2I protein abundance by nearly 60%.

UBE2I is a small ubiquitin-like modifier E2 enzyme reportedly expressed in tumors (Guo et al. 2012, Meng and Li 2021). It is reported to be linked to HCC progression and prognosis, but its role in cervical cancer is previously undescribed. Rescue experiments by transfection of miR-10a-5p mimics and sh-UBE2I were performed to investigate the role of the miR-10a-5p/UBE2I axis in HeLa cells. Results implied that miRNA-10a-5p up-regulation facilitated Hela cell proliferation and metastasis while inhibiting cell apoptosis by suppressing UBE2I signaling.

Although this study investigated the mechanism of action of miR-10a-5p in CC from the cellular level in vitro. However, this study still has some limitations. The effect of miR-10a-5p was not further verified from the in vivo level. Nevertheless, whether the miR-10a-5p/UBE2I axis can be regulated by non-invasive methods need further exploration, which will be the focus of our future research.

Overall, according to experiments utilizing miRNA mimics and inhibitors, we verified that miR-10a-5p initiates and promotes tumor cell development by decreasing UBE2I abundance. This miR-10a-5p-mediated post-transcriptional regulation of UBE2I is involved in the tumorigenesis, invasion and migration of human cervical cancer. These findings provide mechanistic insights into how miR-10a-5p regulates cervical cancer hyper-proliferation and metastasis, as well as a new target for clinical treatment.

Ethical approval

This study is in line with the Improving the Quality and Transparency of Health Research (Equator) Network guidelines. This study was approved by the Ethics Committee of the Maternal and Child Health Hospital of Nantong University (Y2021040), and all patients were informed of and signed the informed consent.

Author contributions

YF conceived and designed the study. YL and DL conducted most of the experiments. XF analyzed the data. YJ performed the literature search and data extraction. YG drafted the manuscript. LZ finalized the manuscript.

Disclosure statement

No potential conflict of interest was reported by the author(s). All authors read and approved the final manuscript.

Data availability statement

All data generated or analyzed during this study are included in this published article.

Additional information

Funding

This work was supported by the Scientific Research Project of Nantong Health Committee (QA2021046).