ABSTRACT
TMPOP2 was previously suggested to be an oncogenic long noncoding RNA which is excessively expressed in cervical cancer cells and inhibits E-cadherin gene expression by recruiting transcription repressor EZH2 to the gene promoter. So far, the function and regulation of TMPOP2 in cervical cancer remain largely unknown. Herein, we found that TMPOP2 expression was correlated with human papillomavirus 16/18 (HPV16/18) E6 and E7 in cervical cancer cell lines CaSki and HeLa. Tumor suppressor p53, which is targeted for degradation by HPV16/18, was demonstrated to associate with two p53 response elements in the TMPOP2 promoter to repress the transcription of the TMPOP2 gene. Reciprocally, ectopic expression of TMPOP2 was demonstrated to sequester tumor repressor microRNAs (miRNAs) miR-375 and miR-139 which target HPV16/18 E6/E7 mRNA and resulted in an upregulation of HPV16/18 E6/E7 genes. Thereby, HPV16/18 E6/E7 and the long noncoding RNA (lncRNA) TMPOP2 form a positive feedback loop to mutually derepress gene expression in cervical cancer cells. Moreover, results of RNA sequencing and cell cycle analysis showed that knockdown of TMPOP2 impaired the expression of cell cycle genes, induced cell cycle arrest, and inhibited HeLa cell proliferation. Together, our results indicate that TMPOP2 and HPV16/18 E6/E7 mutually strengthen their expression in cervical cancer cells to enhance tumorigenic activities.
IMPORTANCE Human papillomaviruses 16 and 18 (HPV16/18) are the main causative agents of cervical cancer. Viral proteins HPV16/18 E6 and E7 are constitutively expressed in cancer cells to maintain oncogenic phenotypes. Accumulating evidences suggest that HPVs are correlated with the deregulation of long noncoding RNAs (lncRNAs) in cervical cancer, although the mechanism was unexplored in most cases. TMPOP2 is a newly identified lncRNA excessively expressed in cervical cancer. However, the mechanism for the upregulation of TMPOP2 in cervical cancer cells remains largely unknown and its relationship with HPVs is still elusive. The significance of our research is in revealing the mutual upregulation of HPV16/18 E6/E7 and TMPOP2 with the molecular mechanisms explored. This study will expand our understandings of the oncogenic activities of human papillomaviruses and lncRNAs.
INTRODUCTION
High-risk human papillomaviruses (HPVs), primarily including HPV16 and HPV18, are causative agents of cervical cancer, which is one of the leading lethal female cancers worldwide. Integration of HPV DNA into the host cell genome results in the constitutive high expression of two viral oncogenes, HPV E6 and E7, under the control of cellular transcription factors. Viral protein E6 promotes p53 protein degradation through E6AP-mediated ubiquitination, while E7 interacts with the PDZ domain of cellular proteins, such as retinoblastoma protein pRb, to induce cellular transformation. In addition to p53 and pRb, high-risk HPV E6 and E7 were identified to associate with multiple cellular proteins and contribute to neoplastic progression (1–6). Due to the critical roles of HPV E6 and E7 in tumorigenesis, these viral proteins were potential targets for the therapy of HPV-related cancers (7).
Long noncoding RNAs (lncRNAs) are broadly described as RNAs more than 200 nucleotides (nt) in length, which possess a lot of structural features of the mRNAs but have no conservative open reading domain (8). lncRNAs are crucial regulators in various biological processes such as cell proliferation, differentiation, metastasis, and death. Quite a few lncRNAs have been reported to be aberrantly expressed in cervical cancer cells, and they were supposed to play important roles in the progression of cervical cancer and be potential diagnostic biomarkers (9–14).
As reported, there was a close relevance between HPV16 E7 and lncRNA HOTAIR expression in cervical cancer cells (15, 16). Additionally, HPV16 E6 was recently proposed to increase lncRNA CCEPR expression in cervical cancer cells (17). The role of HPV proteins E6 and E7 in the regulation of lncRNA genes remains largely obscure, although HPV E6 and E7 have been demonstrated to alter the expression of many protein-coding or miRNA-coding genes directly or indirectly in cervical cancer cells. As an important causative factor of cervical cancer, it is highly possible that high-risk HPVs influence the expression of more cancer-related lncRNAs.
Thymopoietin pseudogene 2 (TMPOP2), also named lncRNA-EBIC, is a pseudogene which was recently reported to be highly expressed in human cervical cancer tissues and cell lines (18). This lncRNA was shown to interact with EZH2 (enhancer of zeste homolog 2) to repress the expression of E-cadherin in cervical cancer cells and promote cell invasion, suggesting an oncogenic role of TMPOP2. However, the relationship between TMPOP2 and HPV viral proteins was not reported (18). Previously, we introduced HPV18 E6- or E7-encoding plasmids into HeLa or HaCaT cells and found that both HPV18 E6 and E7 stimulated TMPOP2 expression, suggesting the involvement of HPV18 viral proteins in the regulation of TMPOP2 gene in cervical cancer cells (19).
In the present study, we further investigated the relationship between HPV16/18 E6/E7 and TMPOP2 in cervical cancer cells, and the effects of TMPOP2 on the proliferation of cervical cancer cells were also determined. Results of this study suggest a mechanism by which TMPOP2 and HPV16/18 E6 mutually regulate gene expression and reveal a novel function of TMPOP2 in cervical cancer cell proliferation.
RESULTS
HPV16/18 proteins E6 and E7 promoted the expression of lncRNA TMPOP2.TMPOP2 was previously reported to be highly expressed in human cervical cancer tissues and cell lines (18). We also observed a higher RNA level of TMPOP2 in HeLa cervical cancer cells than in nonmalignant HaCaT cells (Fig. 1A). Overexpression of HPV18 E6 or E7 enhanced the expression of TMPOP2 in HeLa cells (Fig. 1B), which was consistent with our previous observation that both HPV18 E6 and HPV18 E7 possessed the capability to induce TMPOP2 expression in HaCaT cells (19). To confirm the involvement of HPV18 E6 and E7 in the expression of TMPOP2, small interfering RNAs (siRNAs) specific to the HPV18 E6/E7 transcript were transfected into HeLa cells. The efficiency of HPV18 E6/E7 depletion is shown in Fig. 1C. In these HPV-deficient cells, the p53 protein accumulated (Fig. 1C, row 3, lane 2). Meanwhile, the expression of TMPOP2 was significantly downregulated (Fig. 1D), supporting that HPV18 E6 and E7 facilitate the TMPOP2 gene upregulation in HeLa cells.
Human papillomavirus proteins E6 and E7 promoted the expression of LncRNA TMPOP2. (A) Expression of TMPOP2 in HeLa cervical cancer cells was higher that than in nonmalignant HaCaT cells. Total RNA was extracted from HeLa and HaCaT cells. RNA levels of TMPOP2 were detected by real-time qPCR. (B) Overexpression of HPV18 E6 or E7 enhanced the expression of TMPOP2 in HeLa cells. HPV18 E6- or E7-encoding plasmids were transfected into HeLa cells for 48 h before extraction of total RNA. (C) The efficiency of HPV18 E6/E7 depletion and p53 accumulation in HeLa cells. Western blotting was performed with whole-cell extracts of HeLa cells transfected with siHPV18 E6/E7. (D) Depletion of HPV18 E6/E7 reduced the expression of TMPOP2 in HeLa cells. (E) The efficiency of HPV16 E6/E7 depletion and p53 accumulation in CaSki cells. Western blotting was performed with whole-cell extracts of CaSki cells transfected with siHPV16 E6/E7. (F) Depletion of HPV16 E6/E7 reduced the expression of TMPOP2 in CaSki cells. n = 3. *, P < 0.05; **, P < 0.01.
To test whether the regulatory role of E6 and E7 in TMPOP2 expression is limited to HPV18, HPV16-positive CaSki cells were transfected with siRNA to knockdown HPV16 E6/E7 (Fig. 1E). Results of reverse transcriptase quantitative PCR (RT-qPCR) showed that depletion of HPV16 E6/E7 significantly diminished the expression of TMPOP2 (Fig. 1F), which was similar to the phenomenon observed with HeLa cells.
Together, the above results suggest that HPV16/18 E6 and E7 promote the excessive expression of TMPOP2 in cervical cancer cells.
The regulation of lncRNA gene TMPOP2 by HPV16/18 was mediated by p53.HPV E6 and E7 themselves are not DNA-binding transcription factors. Thus, the effect of HPV E6/E7 on TMPOP2 expression might be mediated by cellular transcription factors. It was previously shown that high-risk HPV E6 cooperates with E6AP to promote p53 degradation and that HPV E7 impairs the repressive activity of the p53-p21-DREAM pathway (20), which consequently alter the expression of p53 target genes. As shown in Fig. 1C and E, knockdown of HPV16/18 E6/E7 led to an accumulation of p53 protein. Thus, we anticipated that p53 might be involved in the regulation of the TMPOP2 gene.
To test whether p53 really affects the expression of TMPOP2, p53-encoding plasmids were cloned and transfected into HeLa cells. The efficacy of p53 overexpression was determined with Western blotting, and the increased expression of p21, a well-defined target of p53, was used as an indicator of the p53 transactivity (Fig. 2A). In these p53-overexpressed HeLa cells, the RNA level of TMPOP2 was dramatically decreased, and in contrast to that of TMPOP2, the expression of p21 was significantly upregulated (Fig. 2B), suggesting a specific repressive role of p53 in the regulation of the TMPOP2 gene. Similar results were observed with CaSki cells (Fig. 2C and D). In our previous data, TMPOP2 was found to be highly expressed in MCF-7 breast cancer cells as well (19). To examine the role of p53 in the regulation of TMPOP2 in HPV-negative cancer cells, p53-encoding plasmids were introduced into MCF-7 cells. Results of RT-qPCR showed a suppressive effect on the expression of TMPOP2 (Fig. 2E), supporting that p53 repressed the transcription of TMPOP2 in cancer cells.
The regulation of lncRNA TMPOP2 by HPV16/18 was mediated by p53. (A) The efficacy of p53 overexpression was determined by Western blotting with HeLa cells. The increased expression of p21 indicated the enhanced activity of p53 in HeLa cells. (B) The RNA level of TMPOP2 was dramatically decreased opposing the upregulation of p21 in p53-overexpressing HeLa cells. (C) The efficacy of p53 overexpression in CaSki cells. The increased expression of p21 indicated the enhanced activity of p53 in CaSki cells. (D) The RNA level of TMPOP2 was dramatically decreased while the expression of p21 was upregulated in p53-overexpressing CaSki cells. (E) The overexpression of p53 significantly suppressed the expression of TMPOP2 in MCF-7 cells. (F) The efficacy of p53 depletion with specific siRNA in HeLa cells. The reduced expression of p21 indicated impairment in p53 activity. (G) TMPOP2 gene was upregulated whereas the expression of p21 was reduced in the p53-depleted HeLa cells. (H) Knockdown of p53 increased the expression of TMPOP2 but diminished the expression of p21 in MCF-7 cells. (I) The RNA level of TMPOP2 in HeLa cells co-overexpressing HPV18 E6 and p53. (J) The RNA level of TMPOP2 in HeLa cells treated with siHPV18 and pifithrin-α. HeLa cells were transfected with siHPV18 E6/E7 and cotreated with 40 µM p53 inhibitor pifithrin-α for 24 h. n = 3. *, P < 0.05; **, P < 0.01.
To further confirm the repressive function of p53 on the TMPOP2 gene, p53 was depleted with p53-specific siRNA. The efficacy of p53 knockdown was demonstrated with results of Western blotting (Fig. 2F). In these p53-depleted HeLa cells, the expression of TMPOP2 was elevated, whereas the expression of p21 was reduced (Fig. 2F and G), suggesting that the TMPOP2 gene was repressed by p53 while the p21 gene was activated by p53. In MCF-7 cells, similar results were observed (Fig. 2H), supporting the repressor role of p53 in the regulation of TMPOP2.
To test whether p53 negatively mediates the HPV protein-induced upregulation of TMPOP2, HPV18 E6 and p53 were coexpressed in HeLa cells. As shown in Fig. 2I, the expression of TMPOP2 was upregulated by HPV18 E6 overexpression but inhibited by p53 overexpression. Furthermore, the co-overexpression of p53 abolished the HPV18 E6-induced upregulation of TMPOP2 (Fig. 2I, compare the 4th bar with the 2nd bar), confirming the dominant negative role of p53 in TMPOP2 regulation. In HeLa cells depleted of HPV18 E6/E7, TMPOP2 was downregulated, and inhibition of p53 with pifithrin-α, an inhibitor of p53 transcriptional activity, increased TMPOP2 expression (Fig. 2J). Inhibition of p53 reversed the downregulation of TMPOP2 resulting from HPV18 depletion (Fig. 2J, compare the 4th bar with the 2nd bar), demonstrating that the activity of p53 was required for the repression of the TMPOP2 gene in HPV18-deficient HeLa cells.
p53 associated with the promoter of TMPOP2 to repress its activation.It is controversial whether the transcriptional repressor function of p53 is direct or indirect (21–24). To determine whether p53 plays a direct role in TMPOP2 regulation, the promoter sequence of TMPOP2 was analyzed with bioinformatics programs UCSC Genome Browser and ALGGEN to predict the potential regulatory factors. As shown in Fig. 3A, there are two mirror symmetric nonconsensus p53 response elements (p53 REs), GGGCAGG and CCTGCCC, separated by a spacer of 48 nt in the proximal promoter region of TMPOP2.
p53 associated with the promoter of TMPOP2 to repress its transcription. (A) Schematic description of the TMPOP2-promoter-luciferase construction. The TMPOP2 promoter sequence from −1704 to +6 was cloned into pGAL3-luciferase reporter vector. Positions of two p53 REs are indicated and the sequences for the wild-type and mutated p53 RE are depicted. (B) TMPOP2 promoter activity was significantly suppressed by p53 overexpression in a p53 RE-dependent manner. p53-encoding plasmids were introduced into HeLa cells for 24 h, and subsequently, wild-type or mutated TMPOP2 promoter luciferase plasmids were transfected for another 24 h before the measurement of luciferase activity. (C) Schematic of the TMPOP2 promoter with the locations of two ChIP-qPCR fragments depicted. Fragment I is a control fragment which is at 1.5 kb upstream of the p53 binding sites, fragment II covers the p53-binding sequences. (D) p53 physically interacted with the p53 RE in the TMPOP2 promoter. ChIP assays were performed with antibodies against p53 or normal mouse IgG. n.s., not significant.
To test the impact of p53 on the activity of the TMPOP2 promoter, luciferase reporter plasmids were constructed. The wild-type luciferase reporter plasmid pGL3-TMPOP2 contained a 1,710-bp region of the TMPOP2 promoter, including p53 REs (Fig. 3A). In addition, three mutated plasmids were constructed with site-directed mutagenesis as indicated in Fig. 3A. Luciferase assays were performed with HeLa cells, and the results showed that the activity of the TMPOP2 promoter was significantly decreased by p53 overexpression (Fig. 3B). However, mutations in either one or both p53 REs nearly abolished the response of reporter plasmids to p53 overexpression (Fig. 3B), suggesting a cis-repressive function of p53 REs in the regulation of the TMPOP2 gene.
To confirm that p53 plays a direct role in TMPOP2 regulation, the association of p53 with the TMPOP2 promoter was examined with chromatin immunoprecipitation (ChIP) assays. PCR fragments in ChIP assays are described in Fig. 3C; fragment I is a control fragment which is at 1.5 kb upstream of the p53 binding site, fragment II covers p53 binding sites. Results of ChIP assays with antibodies against p53 demonstrated that p53 physically interacted with the p53 REs in the TMPOP2 promoter suggesting that p53 plays a direct role in the repression of the TMPOP2 gene (Fig. 3D).
TMPOP2 reciprocally regulated the expression of HPV16/18 E6/E7.To further understand the oncologic significance of increased TMPOP2 expression in cervical cancer, TMPOP2 was depleted with specific siRNA in HeLa cells (Fig. 4A). As shown in Fig. 4B, the protein levels of HPV18 E6/E7 were also decreased following TMPOP2 knockdown. Meanwhile, the mRNA levels of HPV18 E6/E7 were abrogated (Fig. 4C). In contrast, the expression of GAPDH (glyceraldehyde-3-phosphate dehydrogenase) was basically unchanged under the same condition (Fig. 4C). Similar effects were observed in CaSki cells; the expression of HPV16 E6/E7 was impaired by TMPOP2 depletion (Fig. 4D and E). These results suggest that lncRNA TMPOP2 reciprocally regulates the expression of HPV genes in cervical cancer cells.
TMPOP2 reciprocally regulated the expression of HPV E6/E7 in cervical cancer cells. (A) The efficacy of TMPOP2 knockdown in HeLa cells. (B) The protein levels of HPV18 E6/E7 were decreased following TMPOP2 knockdown. GAPDH was the protein loading control. (C) The mRNA levels of HPV18 E6/E7 were attenuated in TMPOP2-depleted HeLa cells with the mRNA level of GAPDH unaffected. (D) TMPOP2-specific siRNA reduced the mRNA levels of HPV16 E6/E7 but not that of GAPDH in CaSki cells. (E) The protein levels of HPV16 E6/E7 were decreased following TMPOP2 knockdown. (F) Overexpression of TMPOP2 increased the mRNA levels of HPV18 E6/E7. (G) Overexpression of TMPOP2 increased the protein levels of HPV18 E6/E7. In panels F and G, pcDNA3.1-TMPOP2-wt was transfected into HeLa cells for 48 h to ectopically express wild-type TMPOP2. (H) The activity of HPV18 promoter was not affected by TMPOP2 depletion in HeLa cells. siTMPOP2 was introduced into HeLa cells 24 h before the transfection of HPV18-LCR-luciferase plasmids for promoter activity assays. (I) TMPOP2 depletion diminished the expression of HPV18 E6/E7 from a CMV promoter in MCF-7 cells. Plasmids pCMV-Tag2B-HPV18 E6/E7 and siTMPOP2 number 2 were cotransfected into breast cancer cell MCF-7 for 24 h. Expression of TMPOP2, HPV18 E6, and HPV18 E7 was measured by RT-qPCR. n = 3. *, P < 0.05; **, P < 0.01.
To confirm the effect of TMPOP2 on HPV gene expression, a TMPOP2-encoding plasmid was constructed and transfected into HeLa cells. Results showed that the overexpression of TMPOP2 augmented HPV18 E6/E7 mRNA levels (Fig. 4F) and protein levels (Fig. 4G). These results support that TMPOP2 facilitates HPV gene expression.
To explore the mechanism by which TMPOP2 promotes the expression of HPV E6/E7, HPV18 promoter activity was examined with luciferase assays. The results showed that HPV18 promoter activity was not affected by TMPOP2 depletion in HeLa cells (Fig. 4H), indicating that TMPOP2 might regulate HPV genes posttranscriptionally.
To verify this assumption, HPV18 E6/E7-encoding plasmids, in which the open reading frames of HPV18 E6 and E7 were driven by the cytomegalovirus (CMV) promoter in the vector backbone sequence, and TMPOP2-interfering siRNA were cotransfected into MCF-7 cells. Results showed that HPV18 E6 and E7 were successfully expressed in MCF-7 cells, while with the depletion of TMPOP2, the mRNA levels of HPV18 E6 and E7 were dramatically decreased (Fig. 4I), confirming that TMPOP2 regulated the expression of HPV18 E6/E7 from a heterologous promoter. Hence, the regulation of HPV genes by TMPOP2 is a posttranscriptional event.
miR-375 and miR-139 targeted HPV E6/E7 mRNA in cervical cancer cells.As reported, the mRNA of the HPV E6/E7 gene was targeted by multiple microRNAs, including miR-375 (25, 26) and miR-139 (27), in cervical cancer cells. To confirm the targeting effect, mimics of miR-375 were transfected into HeLa cells. The results of RT-qPCR showed that the mRNA levels of HPV18 E6 and E7 was reduced significantly (Fig. 5A). Consequently, the protein expression of HPV18 E6 and E7 was also reduced by miR-375 mimics (Fig. 5B). Similarly, in CaSki cells, mimics of miR-375 suppressed the expression of HPV16 genes at both the mRNA level (Fig. 5C) and the protein level (Fig. 5D). In addition, mimics of miR-139-3p inhibited HPV18 E6 and E7 expression in HeLa cells (Fig. 5E). These results support that miR-375 and miR-139 are able to target HPV16/18 genes in cervical cancer cells.
miR-375 and miR-139-3p targeted HPV16/18 E6/E8 mRNA in cervical cancer cells. (A) Mimics of miR-375 reduced the mRNA of HPV18 E6 and E7 in HeLa cells. (B) Mimics of miR-375 reduced the protein expression of HPV18 E6 and E7 in HeLa cells. (C) Mimics of miR-375 reduced the mRNA of HPV16 E6 and E7 in CaSki cells. (D) Mimics of miR-375 reduced the protein of HPV16 E7 in CaSki cells. (E) Mimics of miR-139-3p reduced the mRNA of HPV16/18 E6 and E7 in CaSki or HeLa cells as indicated. (F) Inhibitors of miR-375 increased the mRNA of HPV16/18 E6 and E7 in CaSki or HeLa cells as indicated. (G) Inhibitors of miR-139-3p increased the mRNA of HPV16/18 E6 and E7 in CaSki or HeLa cells as indicated. n = 3. *, P < 0.05; **, P < 0.01.
To further confirm the targeting effects of miR-375 on HPV16/18 genes, inhibitors of miR-375 were transfected into HeLa or CaSki cells. Results of RT-qPCR showed that HPV18/16 E6 and E7 mRNA levels were upregulated (Fig. 5F). Inhibitors of miR-139-3p increased the expression of HPV18/16 E6 and E7 mRNA as well (Fig. 5G). Taken together, these results confirm that miR-375 and miR-139 inhibit HPV gene expression.
miR-375 and miR-139 mediated the effect of TMPOP2 on HPV gene expression.With a nucleotide blast program, the sequences of miRNA-375 and miR-139 were identified to be partially complementary to the sequence of TMPOP2 (Fig. 6A), suggesting interactions between TMPOP2 and miR-375 and miR-139. In fact, mimics of miR-375 and miR-139-3p depleted the expression of TMPOP2 (Fig. 6B) and inhibitors of miR-375 and miR-139-3p increased the expression of TMPOP2 (Fig. 6C) in HeLa cells, supporting interactions between miR-375, miR-139-3p, and lncRNA TMPOP2.
miR-375 and miR-139-39 were decoyed by TMPOP2 in cervical cancer cells. (A) Sequence alignments between HPV E6/E7-targeting miRNAs miR-375, miR-139-3p, and lncRNA TMPOP2. There are three putative miR-375 binding sites and five putative miR-139-3p binding sites in the sequence of TMPOP2. (B) Mimics of miR-375 and miR-139-3p decreased the RNA level of TMPOP2 in HeLa cells. (C) Inhibitors of miR-375 and miR-139-3p increased the RNA level of TMPOP2 in HeLa cells. (D and E) The levels of miRNA-375 and miRNA-139 were augmented in TMPOP2-depleted HeLa and CaSki cells. (F) miR-375 and miR-139-3p were reduced in cervical cancer cells overexpressing wild-type TMPOP2. (G) miR-375 and miR-139-3p were not affected in cervical cancer cells overexpressing mutated TMPOP2. (H) The mRNA levels of HPV18 E6 and E7 were not changed by the TMPOP2 mutant. (I) The protein levels of HPV18 E6 and E7 were not changed by the TMPOP2 mutant. n = 3. *, P < 0.05; **, P < 0.01.
It was previously reported that certain lncRNAs work in vivo as a sponge of miRNAs (28). So, it is possible that TMPOP2 functions as a competitive endogenous RNA (ceRNA) to decoy miR-375 and miR-139. To test this possibility, the expression of miRNA-375 and miRNA-139 was first measured in TMPOP2-depleted HeLa and CaSki cells. The results showed that the levels of miRNA-375 and miRNA-139 were indeed significantly augmented following TMPOP2 knockdown (Fig. 6D and E), indicating that the levels of miRNA-375 and miRNA-139 are negatively correlated with the expression of TMPOP2.
To further confirm the role of TMPOP2 in miRNA-375 and miRNA-139 expression, TMPOP2-encoding plasmids were introduced into HeLa or CaSki cells. The results showed that the expression of miR-375 and miR-139-3p was significantly attenuated (Fig. 6F). In contrast to the wild-type TMPOP2, the TMPOP2 mutant with miR-375 and miR-139 matching sequences altered did not affect the expression of miR-375 and miR-139 (Fig. 6G), confirming the role of wild type TMPOP2 as a ceRNA for miR-375 and miR-139.
Given that miR-375 and miR-139 target both HPV E6/E7 mRNA (Fig. 5) and lncRNA TMPOP2 (Fig. 6B and C), and TMPOP2 sponges miR-375 and miR-139 (Fig. 6D to G), it is reasonable that miR-375 and miR-139 mediate the regulatory effect of TMPOP2 on HPV gene expression. To further verify this hypothesis, the influence of the TMPOP2 mutant on HPV gene expression was examined. The results showed that neither the mRNA (Fig. 6H) nor the protein (Fig. 6I) levels of HPV18 E6 and E7 were changed by the TMPOP2 mutant, suggesting that TMPOP2 indirectly stabilizes HPV18 E6/E7 mRNA by sponging miRNA-375 and miRNA-139.
TMPOP2 was required for the proliferation of cervical cancer cells.Since TMPOP2 increased the expression of HPV E6 and E7 viral oncoproteins, it should be an oncogene contributing to the development of cervical cancer. To further dissect the function of TMPOP2, transcriptome sequencing was performed to analyze the effects of TMPOP2 depletion on the mRNA levels of all protein-coding genes. Figure 7A shows a scatter plot of all differentially expressed genes, including 146 that were upregulated and 248 that were downregulated, in TMPOP2-depleted HeLa cells. Interestingly, expression of several cell cycle-related and p53 signaling genes was changed, such as that of p21, which was upregulated, and cyclin E and CDK2, which were downregulated, indicating that TMPOP2 might be involved in the regulation of the cell cycle and p53 pathway.
TMPOP2 affected the expression of cell cycle genes. (A) Depletion of TMPOP2 altered the expression of approximate 400 protein-coding genes in HeLa cells. Analysis of transcriptome sequencing data showed that there were 146 genes upregulated (yellow dots) and 248 genes downregulated (blue dots) in TMPOP2-depleted HeLa cells. (B) mRNA levels of cell cycle genes p21, cyclin E, and CDK2 in TMPOP2-depleted HeLa cells. (C) Protein levels of cell cycle genes p21, cyclin E, and CDK2 in TMPOP2-depleted HeLa cells. (D and E) The expression of p21, cyclin E, and CDK2 in TMPOP2-depleted CaSki cells. (F) mRNA levels of cell cycle genes p21, cyclin E, and CDK2 in TMPOP2-depleted MCF-7 cells. n = 3. *, P < 0.05; **, P < 0.01.
To confirm the results of mRNA sequencing, RT-qPCR and Western blotting were used to measure the expression of p21, cyclin E, and CDK2. As shown in Fig. 7B and C, in HeLa cells with TMPOP2 depletion, expression of p21 was increased whereas expression of cyclin E and CDK2 was reduced, which is in agreement with results of mRNA sequencing. Similar changes were observed in CaSki cells (Fig. 7D and E). These results support a role of TMPOP2 in the cell cycle progression of cervical cancer cells.
Previously, TMPOP2 was observed to be highly expressed in multiple cell lines, including HPV-positive cervical cancer cell lines and HPV-negative breast cancer cell lines (19). To test whether TMPOP2 regulates cell cycle-related genes in other cancer cell lines, breast cancer MCF-7 cells were transfected with siTMPOP2 and then the expression of cell cycle-related genes was measured. As shown in Fig. 7F, depletion of TMPOP2 resulted in the upregulation of p21 and downregulation of CDK2 and cyclin E, suggesting the HPV proteins are dispensable for TMPOP2 to affect cell cycle.
To confirm the importance of TMPOP2 in cervical cancer cell proliferation, CCK8 assays were performed and, as shown in Fig. 8A, the proliferation of HeLa cells was inhibited when TMPOP2 was knocked down. To determine the effects of TMPOP2 knockdown on cell cycle progression, HeLa cells were stained with propidium iodide and analyzed with fluorescence-activated cell sorting (FACS). The results showed that there were obviously less cells in the S phase and more cells in the G1 phase (Fig. 8B), indicating that the downregulation of TMPOP2 induced G1/S arrest. To verify that TMPOP2 was required for the S phase entry of cell cycle, EdU assays were used to visualize cells in the S phase. As shown in Fig. 8C, with TMPOP2 knockdown, there were significantly fewer cells replicating DNA, demonstrating that TMPOP2 is critical for the G1/S transition of the cell cycle.
TMPOP2 was required for the proliferation of HeLa cells. (A) The proliferation of HeLa cells was inhibited following TMPOP2 knockdown. CCK8 assays were performed at 24-h intervals as indicated. (B) Cell cycle analysis of the TMPOP2-depleted HeLa cells. Propidium iodide (PI)-stained HeLa cells were subjected to fluorescence-activated cell sorting (FACS). (C) The fraction of S-phase HeLa cells was reduced upon TMPOP2 knockdown. EdU assays were applied to visualize cells in the S phase of the cell cycle.
Together, these results indicate that TMPOP2 could promote the proliferation of cervical cancer cells. Targeting TMPOP2 might be a potent therapy of cervical cancer.
DISCUSSION
In this study, HPV16/18 E6 and E7 were demonstrated to be critical for the high expression of lncRNA TMPOP2 in cervical cancer cells. Reciprocally, TMPOP2 enhances the expression of HPV16/18 E6 and E7. Hence, HPV16/18 viral proteins and lncRNA TMPOP2 form a positive feedback to synergistically promote the maintenance of cervical cancer cells. The molecular mechanism involves the tumor suppressor p53, which represses the transcription of lncRNA TMPOP2 through direct association with the p53 REs in the TMPOP2 promoter. The constitutive expression of HPV E6 and E7 in cervical cancer cells leads to the degradation of p53 to derepress TMPOP2. Highly expressed TMPOP2 in turn sequesters HPV E6/E7-targeting microRNAs miR-375 (26) and miR-139 (27), resulting in the elevated expression of HPV E6/E7 proteins. HPV16/18 E6 and E7, together with lncRNA TMPOP2, indirectly regulate the expression of cell cycle genes and consequently augment the proliferation of cervical cancer cells.
Aberrantly expressed lncRNAs were believed to be correlated with the development of cancer by various mechanisms, including recruiting chromatin modifiers to tumor suppressor genes (29), sponging of tumor repressive miRNAs, regulating modifications of transcription factors (30), etc. TMPOP2 was previously demonstrated to recruit EZH2 to the promoter of CDH1 to repress E-cadherin expression (18). Herein, lncRNA TMPOP2 was bioinformatically predicated to possess multiple binding sites for miR-375 and miR-139, suggesting the possibility that TMPOP2, as a ceRNA, sequesters miR-375 and miR-139. Mimics and inhibitors of these miRNAs regulated the expression of TMPOP2, and in reverse, the wild type but not the binding sites-mutated TMPOP2 attenuated the expression of miR-375 and miR-139. Thus, a novel function of TMPOP2 was identified in this study.
TMPOP2 was previously identified as an oncogenic lncRNA facilitating the invasion of cervical cancer (18). With RNA sequencing, TMPOP2 was herein suggested to affect multiple processes, including cell cycle regulation, DNA repair, and cell metabolism. Since TMPOP2 augments the expression of HPV E6/E7, it is reasonable that TMPOP2 plays multiple roles in cervical cancer development. Like many lncRNAs (31), TMPOP2 is regulated by p53, which is deficient or mutated in a variety of cancers, indicating a correlation between TMPOP2 and other cancers. The role of TMPOP2 in the development of other cancers is worth further study. TMPO (thymopoietin), also named LAP2 (lamina-associated polypeptide 2), is a conserved nuclear protein important for nuclear assembly and is overexpressed in various tumors (32–34). TMPOP2 is a pseudogene of TMPO/LAP2. It is possible that, like other pseudogenes (28, 35), TMPOP2 functions as a ceRNA to enhance the expression of TMPO, which needs further validation.
In conclusion, this study explored the mechanisms for the mutual upregulation between HPV16/18 and lncRNA TMPOP2. A novel function of lncRNA TMPOP2 in cervical cancer cell proliferation was also revealed. Results of this study suggested TMPOP2 to be a potential diagnostic maker and therapeutic target of cervical cancer.
MATERIALS AND METHODS
Cell culture and transfection.Cervical cancer cell lines HeLa (HPV18 positive) and CaSki (HPV16 positive) and breast cancer cell line MCF-7 were cultured in Dulbecco’s modified Eagle’s medium (DMEM)-low glucose (Gibco, USA) containing 10% fetal bovine serum (FBS; Kangyuan, China), 100 µg/ml streptomycin, and 100 U/ml penicillin (Solarbio, China). Human immortalized keratinocyte cell line HaCaT was grown in MEM supplemented with 15% FBS. Cells were incubated at 37°C with 5% CO2. The protein-coding plasmids pCMV-Tag2B-HPV18 E6, pCMV-HA-HPV18 E7, and pCMV-Tag2B-HPV18 E6/E7 were described previously (19). p53-encoding plasmid pLV07-p53 was constructed in this study. Sequences of wild-type or mutated TMPOP2 were chemically synthesized and inserted into pcDNA3.1 by Saier Biotechnology, Tianjin, China, to construct plasmids pcDNA3.1-TMPOP2-wt and pcDNA3.1-TMPOP2-mut. The sequences of wild-type and mutated TMPOP2 are available upon request. Plasmids were transiently transfected into cells using Lipofectamine 2000 as instructed by the manufacture (Invitrogen). Mimics and inhibitors of miR-375 and miR-139-3p were purchased from RiboBio, Guangzhou, China. The HPV16 E6/E7, HPV18 E6/E7, p53, and TMPOP2 interfering siRNA duplexes, designed and synthesized by RiboBio, Guangzhou, China, were introduced into cells with TurboFect (Thermo, USA).
Extraction of total RNA and RT-qPCR.Total RNA was extracted with Trizol (Invitrogen). cDNA was synthesized with Moloney murine leukemia virus (M-MLV) reverse transcriptase (Promega) and quantified by real-time qPCR using Biosystems StepOne real-time PCR system and Fast SYBR green Master Mix (Applied Biosystems). 18S rRNA or GAPDH was used as endogenous control in qPCR analyses of mRNA and lncRNA. snRNA U6 was the endogenous control for the analysis of miRNA. PCR primers were designed with NCBI online software Primer-BLAST and synthesized by Invitrogen. The sequences of primers used in this study are available upon request.
Western blotting.Antibodies against HPV18/16 E6/E7, p53, p21, CDK2, and cyclin E were from Santa Cruz Biotechnology, USA. Antibodies against GAPDH were from Beyotime Biotechnology, China. Secondary antibodies were IRDye-conjugated donkey anti-mouse, anti-goat, or anti-rabbit IgG (LI-COR Biosciences, USA). Protein samples were prepared with SDS lysis buffer. Whole-cell lysates were separated by SDS-PAGE and transferred to a nitrocellulose membrane. Protein bands were visualized using an Odyssey imaging system with GAPDH serving as the loading control.
Luciferase reporter assay.HPV18-LCR-luciferase (Addgene plasmid 22859) was previously described (36, 37). The pGL3-TMPOP2-luciferase reporter plasmid carrying the wild-type TMPOP2 promoter sequence was constructed with pGL3-basic firefly luciferase using NheI and XhoI sites. Then, pGL3-TMPOP2-mut-luciferase plasmids were generated with a site-directed mutagenesis kit (TransGen, China). HeLa cells were transfected with pLV07 or pLV07-p53 24 h before the transfection of luciferase reporter plasmids. Then, luciferase activity was measured with the Luciferase assay system (Promega) in a Synergy 4 luminometer (BioTek, USA).
EdU staining assay.For EdU staining, exponentially growing cells were seeded at approximately 1,000 cells per well in 96-well plates and incubated overnight. Cells were fixed in 4% paraformaldehyde for 15 min and permeabilized with 0.25% Triton X-100 in phosphate-buffered saline (PBS) for 20 min. The EdU assay was performed with the EdU Alexa Fluor 488 imaging kit according to the manufacturer’s instructions (Beyotime Biotechnology, China).
Cell cycle analysis with FACS.HeLa cells treated with 10 nM siTMPOP2 for 48 h were harvested, fixed with 75% cold ethanol, and conserved at −20°C for 12 h. Then, cells were washed with PBS, placed in 100 µl of propidium iodide (100 µg/ml, Sigma) for 20 min in the dark, and analyzed by flow cytometry (BD).
Chromatin immunoprecipitation.HeLa cells were transfected with 6 µg pLV07-p53 plasmids for 24 h before being harvested for ChIP. HeLa cells were cross-linked with 1% of formaldehyde (Sigma) for 15 min at room temperature. Chromatin was fragmentized to an average of 300 bp in length with a sonicator. ChIP with anti-p53 (Santa Cruz) was performed as previously described (38). After protein A bead (17-295; Millipore) binding, washing, and eluting, ChIP products were purified and measured by real-time qPCR. The sequences of primers used for ChIP qPCR are available upon request.
RNA sequencing.HeLa cells were transfected with siTMPOP2 or scramble siRNA for 48 h. Then, the total RNA was extracted with Trizol. RNA samples were purified, sequenced, and analyzed by BGI, China.
Statistical analysis.The data are representative of three to six independent experiments, and GraphPad Prism 6 was used for analysis. The data are presented as means ± standard deviations. Two tailed P values of <0.05 or <0.01 are indicated in the figures.
ACKNOWLEDGMENTS
This work was supported by National Natural Science Foundation of China (31301073), the Natural Science Foundation of Tianjin (numbers 18JCYBJC91500 and 17JCZDJC33600), and the Innovative Research Team of Tianjin Municipal Education Commission (TD13-5015).
We declare no conflict of interests.
FOOTNOTES
- Received 14 October 2018.
- Accepted 28 January 2019.
- Accepted manuscript posted online 6 February 2019.
- Copyright © 2019 American Society for Microbiology.
