Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Cervical cancer subtypes harbouring integrated and/or episomal HPV16 portray distinct molecular phenotypes based on transcriptome profiling of mRNAs and miRNAs

Cervical cancer subtypes harbouring integrated and/or episomal HPV16 portray distinct molecular... Heterogeneity in cervical cancers (CaCx) in terms of HPV16 physical status prompted us to investigate the mRNA and miRNA signatures among the different categories of CaCx samples. We performed microarray-based mRNA expression profiling and quantitative real-time PCR-based expression analysis of some prioritised miRNAs implicated in cancer- related pathways among various categories of cervical samples. Such samples included HPV16-positive CaCx cases that harboured either purely integrated HPV16 genomes (integrated) and those that harboured episomal viral genomes, either pure or concomitant with integrated viral genomes (episomal), which were compared with normal cervical samples that were either HPV negative or positive for HPV16. The mRNA expression profile differed characteristically between integrated and episomal CaCx cases for enriched biological pathways. miRNA expression profiles also differed among CaCx cases compared with controls (upregulation—miR-21, miR-16, miR-205, miR-323; downregulation—miR-143, miR-196b, miR-203, miR-34a; progressive upregulation—miR-21 and progressive downregulation—miR-143, miR-34a, miR-196b and miR-203) in the order of HPV-negative controls, HPV16-positive non-malignant samples and HPV16-positive CaCx cases. miR-200a was upregulated in HPV16-positive cervical tissues irrespective of histopathological status. Expression of majority of the predicted target genes was negatively correlated with their corresponding miRNAs, irrespective of the CaCx subtypes. E7 mRNA expression correlated positively with miR-323 expression among episomal cases and miR-203, among integrated cases. miR-181c expression was downregulated only among the episomal CaCx cases and negatively correlated with protein coding transcript of the proliferative target gene, CKS1B of the significantly enriched “G2/M DNA Damage Checkpoint Regulation” pathway among CaCx cases. Thus, the two CaCx subtypes are distinct entities at the molecular level, which could be differentially targeted for therapy. In fact, availability of a small molecule inhibitor of CKS1B, suggests that drugging CKS1B could be a potential avenue of treating the large majority of CaCx cases harbouring episomal HPV16. Introduction Correspondence: Sharmila Sengupta (ssg1@nibmg.ac.in)(sharmilasg@gmail. Cervical cancer (CaCx) appears to be the most common com) National Institute of Biomedical Genomics, Kalyani, West Bengal, India malignancy in Indian women, characterised by high Crystallography & Molecular Biology Division, Saha Institute of Nuclear Physics, incidence and mortality rates that are attributable to late 1/AF Bidhannagar, Kolkata 700064, India detection and lack of access to affordable health care. Full list of author information is available at the end of the article. Edited by E. Sayan © The Author(s) 2019 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to theCreativeCommons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. Official journal of the Cell Death Differentiation Association 1234567890():,; 1234567890():,; 1234567890():,; 1234567890():,; Mandal et al. Cell Death Discovery (2019) 5:81 Page 2 of 13 Oncogenic human papillomavirus (HPV) infections are respect to such gene expression profiles and corre- the major aetiological factors, with a predominance of sponding enriched pathways. We further explored whe- HPV16 that accounts for over 50% of such cancers . Viral ther a prioritised set of miRNAs, implicated in various oncoprotein E6 is known to interact with the tumour cancer-related pathways, show altered expression and suppressor p53 and E7, with the PDZ domain of cellular play a role in CaCx pathogenesis under the impact proteins and Rb, thereby facilitating neoplastic progres- of HPV16 infection. Subsequently, we determined whether the miRNAs showing altered expression in sion through loss of cellular homoeostasis . Loss of E2 repressor activity, as a consequence of integration of the CaCx cases, correlate with (i) the physical status of virus into the host genome, is known to impart pro- HPV16 genomes (episomal and integrated), (ii) expression liferative advantage to the host cells . Besides this, our of the oncoprotein E7, (iii) the expression of cellular earlier studies have highlighted mechanisms that act to target genes of specific biological pathways and checked block the E2-mediated repression of the P97 promoter in the relationship, if any, between such target genes and CaCx cases harbouring episomal HPV16 and revealed that viral oncoprotein E7. such cancers harbour higher viral load and E7 expression in contrast to the ones that portray integrated HPV16. Results Cervical carcinogenesis, under the impact of oncogenic Microarray-based gene expression profiling reveals HPV infections, could be influenced by several epigenetic characteristic differences among CaCx cases harbouring mechanisms like DNA methylation, histone modifica- episomal and integrated HPV16 genomes tions, by microRNAs, etc., through perturbations of To identify the differences between CaCx cases har- gene expression profiles. In this communication, we bouring episomal and integrated HPV16 genomes, we chose to focus on the expression of host mRNAs and first attempted to identify global gene expression level miRNAs in various categories of cervical samples changes between the two categories of CaCx cases. Our spanning the discrete stages of CaCx development. microarray-based analysis revealed that 334 genes were These small non-coding RNAs, through recognition of differentially upregulated and 253 were differentially sequence-complementary target elements, can either downregulated among episomal CaCx cases, whereas translationally suppress or catalytically degrade both cel- 1373 genes were differentially upregulated and 1240 were 5,6 lular and viral RNAs . Both DNA and RNA viruses have differentially downregulated among integrated CaCx evolved mechanisms to degrade, boost, or hijack cellular cases, compared with the HPV-negative control samples miRNAs to benefit the viral life cycle (Figure S1). Ingenuity pathway analysis (IPA) revealed . Therefore, a glimpse of the deregulated expression of mRNAs and that biological processes such as cell proliferation, retro- miRNAs in CaCx cases might offer some insights into the viridae infection, and viral Infections were the most sig- mechanisms of CaCx pathogenesis among the two sub- nificantly enriched among the episomal CaCx cases. On types that bear integrated HPV16 and episomal HPV16 in the other hand, biological processes such as transcription, presence or absence of integration, subsequently referred proliferation of tumour cells, and protein metabolism to as integrated and episomal CaCx cases, respectively. were most significantly enriched among integrated CaCx About 50% of the known human miRNAs are located cases (Table S1). IPA also revealed that role of CTLA4- at cancer-associated regions of the genome and a num- signalling in cytotoxic T lymphocytes and T-cell receptor ber of miRNAs are located close to HPV integration sites. signalling were the top significantly altered pathways Host miRNAs are also capable of influencing viral life among the episomal CaCx cases, whereas EIF2 signalling, cycles, viral tropism, and the pathogenesis of viral dis- regulation of eIF4 and p70S6K signalling, mitochondrial eases . Therefore, miRNA-mediated epigenetic regulation dysfunction, and mTOR signalling were the top sig- of gene expression might play a role in virus associated nificantly altered pathways among integrated CaCx cases cancers, like CaCx. (Table S2). Moreover, the upstream regulators involved Some studies have strongly suggested that miRNA in alteration of gene expression, differed distinctly among profiling is more robust than mRNA profiling in deter- the two CaCx subtypes (Table S3). Such observations mining the heterogeneity among cancers . The studies further strengthen our hypothesis that the episomal and 11,12 on CaCx have also demonstrated the association of integrated CaCx cases represent two distinct molecular miRNA deregulation with tumour heterogeneity. The subtypes. heterogeneity among CaCx in terms of physical status of HPV16 genomes by our group , prompted us to explore miRNAs reveal altered expression among CaCx cases the mRNA and miRNA signatures among the different compared with HPV-negative controls and HPV16-positive categories of CaCx samples. non-malignant samples Hence, we tested the hypothesis that CaCx cases with The 34 miRNAs selected for this study belonged to integrated and episomal HPV16 genomes differ with eight different categories by function, as depicted in Official journal of the Cell Death Differentiation Association Mandal et al. Cell Death Discovery (2019) 5:81 Page 3 of 13 Table S4. Of these, nine miRNAs revealed significant samples (fold change = −15.89 fold; p < 0.001). Similar alterations in expression (p value < 0.05) among CaCx analysis considering CaCx cases with integrated HPV16, cases compared with HPV-negative controls, as well as failed to show differential expression of miR-181c among HPV16-positive non-malignant samples as depicted in such CaCx cases. Thus, HPV16 physical status within Table 1. Among these, four miRNAs (miR-21, miR-16, CaCx cases failed to portray any profound effect on miR-205, and miR-323) were significantly overexpressed, majority of the miRNAs as depicted in Table 1. while significant decreased expression was recorded in case of four miRNAs (miR-143, miR-196b, miR-203, and HPV16 E7 mRNA expression levels activate expression of miR-34a). miR-200a, however, showed a statistically sig- miR-323 resulting in upregulation among CaCx cases nificant increased expression by 37.53-fold, among CaCx Based on our clinical sample data we identified that only cases compared with HPV-negative controls but not when two miRNAs (miR-203 and miR-323) were significantly compared with HPV16-positive non-malignant samples. and positively correlated with E7 mRNA expression, The expression status of all the altered miRNAs are among the CaCx cases. Interestingly, the correlation depicted in Table 1, Figure S2, Fig. 1, Fig. 2 and Fig. 3. between E7 and miR-323 (Fig. 5) was obvious among Likewise, seven miRNAs revealed significant alterations CaCx cases harbouring episomal HPV16, whereas the in expression among HPV16-positive non-malignant correlation between E7 and miR-203 (Figure S4) was samples, compared with HPV-negative controls. Of these, evident among CaCx cases with integrated HPV16. As the two miRNAs (miR-21, miR-200a) revealed significant correlation between miR-203 and E7 has already been increased expression and five miRNAs (miR-143, miR-16, documented in earlier studies , we used an in vitro miR-196b, miR-203, and miR-34a) showed significantly approach to confirm the correlation between miR-323 decreased expression. The expression status of all such expression and E7 expression in CaCx cases. altered miRNAs among the HPV16-positive non-malig- We transfected the HPV-negative CaCx cell line, C33A nant sample group is depicted in Table S5, Figure S2, with HPV16 E7 cloned into mammalian expression vector Fig. 1, Fig. 2, and Fig. 3. pcDNA3.1 (+), as explained in our previous study . Progressive increase of miR-21 expression and pro- This resulted in a significant increase in miR-323 gressively decreased expression of miR-143, miR-34a, expression (8.17-fold; p = 0.032) concomitant with miR-196b, and miR-203 was recorded with increase in HPV16 E7 expression (Fig. 6) in such cells. We also severity of cervical status through HPV-negative control determined miR-323 expression in HPV16-positive CaCx samples, HPV16-positive non-malignant samples and cell lines, SiHa, and Caski. Expression of miR-323 was CaCx cases. Such trends were found to be statistically found to be higher among SiHa cells (2.82-fold; p = 0.016) significant (p < 0.001). On the contrary, miR-16 revealed that portray relatively higher E7 expression , as com- a deviation of the trend with decreased expression pared WITH Caski cells (Fig. 6). Thus, both the obser- among HPV16-positive non-malignant samples, followed vations confirmed that E7 could probably activate miR- by increased expression among CaCx cases compared 323 expression in CaCx cases. with HPV-negative controls, as depicted in Figure S3. Such observations are suggestive of the fact that miRNA Target genes of majority of miRNAs that differ in deregulations could potentially serve as risk markers of expression between CaCx cases harbouring episomal and CaCx development. integrated HPV16, also portrayed similar expression level difference between such cases Lack of profound impact of physical status of HPV16 miRNAs are known to target multiple transcripts reg- genome (episomal and integrated) on expression of ulating important cellular processes such as cell pro- majority of miRNAs in CaCx cases, excepting for miR-181c liferation, apoptosis, cellular metastasis, and many more . In view of the biological relevance of episomal HPV16 To check the impact of deregulated expression of the among CaCx cases , expression data of all 34 miRNAs nine significantly altered miRNAs including miR-181c, on considered in this study was further analysed to determine their target genes, we identified the expression profiles whether they showed differentially altered expression of the predicted targets from the microarray-based gene among the CaCx cases with episomal and integrated expression analysis (Table S6). We selected the target HPV16, when compared with HPV-negative controls and genes of the above mentioned 10 miRNAs, considering HPV16-positive non-malignant samples. The sole miRNA only those targets that portrayed ≥ or ≤ 2 fold change in that revealed significantly altered expression only among expression as compared with the controls. Majority of episomal CaCx cases was miR-181c, as depicted in Fig. 4. such target genes of the corresponding miRNAs seemed miR-181c expression was significantly increased in such to be experimentally non-validated, as per the databases cases (fold change = −2.34-fold, p < 0.001), compared considered for target identification as detailed under the with both HPV-negative and HPV16-positive control “Methods” section. For miR-181c, only the expression Official journal of the Cell Death Differentiation Association Mandal et al. Cell Death Discovery (2019) 5:81 Page 4 of 13 Official journal of the Cell Death Differentiation Association Table 1 Expression status of different categories of miRNAs among CaCx cases compared with HPV-negative controls and HPV16-positive non-malignant samples Category microRNAs Comparisons Episomal CaCx cases vs HPV- Integrated CaCx cases vs HPV- Episomal CaCx cases vs HPV16- Integrated CaCx cases vs HPV16- negative controls negative controls positive non-malignant samples positive non-malignant samples Mann–Whitney Up FDR of Mann–Whitney Up FDR of Mann–Whitney Up FDR of Mann–Whitney Up FDR of value (fold change) 0.05 value (fold change) 0.05 value (fold change) 0.05 value (fold change) 0.05 Enhance tumour growth miR-21 <0.001 (25.46); 0.0015 <0.001 (15.01); 0.0015 <0.001 (6.54); 0.0015 <0.001 (3.86); 0.0015 and proliferation upregulation upregulation upregulation upregulation miR-200a <0.001 (48.5); 0.01 <0.001 (33.93); 0.01 0.172 (5.11) 0.02 0.062 (2.536) 0.02 upregulation upregulation Differentiation regulatory miR-143 <0.001 (−41.07); 0.003 <0.001 (−76.24); 0.003 <0.001 (−13.36); 0.003 <0.001 (−24.08); 0.003 miRNAs downregulation downregulation downregulation downregulation Target tumour suppressors miR-16 <0.001 (3.58); 0.006 <0.001 (4.32); upregulation 0.006 <0.001 (5.96); 0.006 <0.001 (15.78); 0.006 and tumour suppressor upregulation upregulation upregulation miRNAs miR-205 <0.001 (116.07); 0.012 <0.001 (156.9); 0.012 <0.001 (94.18); 0.01 <0.001 (124.56); 0.01 upregulation upregulation upregulation upregulation miR-323 0.003 (9.87); upregulation 0.014 0.006 (15.77); upregulation 0.014 <0.001 (37.6); 0.012 <0.001 (58.23); 0.012 upregulation upregulation miR-34a <0.001 (−28.6); 0.009 <0.001 (−20); 0.009 <0.001 (−15.09); 0.009 <0.001 (−9.33); 0.009 downregulation downregulation downregulation downregulation Metastatic suppressor miR-196b <0.001 (−39.34); 0.004 <0.001 (−47.21); 0.004 <0.001 (−12.65); 0.004 <0.001 (−20.88); 0.004 miRNAs downregulation downregulation downregulation downregulation Mesenchymal to epithelial miR-203 <0.001 (−4.77); 0.007 <0.001 (−5.79); 0.007 <0.001 (−1.33); 0.007 <0.001 (−1.76); 0.007 transition miRNAs downregulation downregulation downregulation downregulation Metastatic miRNAs None None None None None None None None None Bold indicates significant p value Mandal et al. Cell Death Discovery (2019) 5:81 Page 5 of 13 Fig. 1 Box plots representing distribution of upregulated miRNA expressions (miR-21, miR-200a, and miR-16, normalised with endogenous control RNU6b and miR-127) among different categories of cervical samples. Lower ΔCT means higher expression. A= HPV- negative control samples (n = 25), B = HPV16-positive non-malignant samples (n = 25), C= HPV16-positive episomal CaCx cases (n = 43), D = HPV16-positive integrated CaCx cases (n = 19) status of target genes in case of episomal CaCx cases were gene that was captured in a significantly enriched pathway considered for expression changes, as this miRNA was among CaCx cases, upon employing IPA (Table S2). differentially expressed, compared with controls, only The microarray-based gene expression profile result of among the episomal cases but remained unaltered in CKS1B is depicted in Table S7. Real-time quantitative expression among the HPV16-integrated CaCx cases. PCR-based validation of the expression of CKS1B on a larger set of cervical samples, revealed differentially Overexpression of miR-181c target gene, CKS1B, among increased expression of CKS1B by 21.16-fold (p < 0.001) episomal CaCx cases among CaCx cases (episomal HPV16; n = 43) compared According to microarray-based gene expression analy- with HPV16-positive non-malignant samples (n = 25) sis, 44 miR-181c target genes showed differential and by 18.5-fold (p < 0.001) in comparison with HPV- expression at more than equal to twofolds among the negative controls (n = 25), as depicted in Figure S5. CaCx episomal CaCx cases. These genes were also found to be cases with integrated HPV16 failed to show significant overexpressed among the episomal CaCx cases compared alteration of the expression of this gene, compared with with integrated CaCx cases (Table S6). Out of these 44 HPV16-positive non-malignant samples and HPV- differentially overexpressed genes, CKS1B was the only negative controls. Official journal of the Cell Death Differentiation Association Mandal et al. Cell Death Discovery (2019) 5:81 Page 6 of 13 Fig. 2 Box plots representing distribution of upregulated and downregulated miRNA expressions (miR-205, miR-323, and miR-143, normalised with endogenous control RNU6b and miR-127) among different categories of cervical samples. Lower ΔCT means higher expression. A = HPV-negative control samples (n = 25), B = HPV16-positive non-malignant samples (n = 25), C = HPV16-positive episomal CaCx cases (n = 43), D = HPV16-positive integrated CaCx cases (n = 19) The expression of CKS1B was inversely and significantly CaCx subtypes (Table S5). Such observations also reflect correlated with miR-181c expression (p = 0.03) based on that the two CaCx subtypes with episomal and integrated pair-wise correlation analysis of various categories of HPV16 are distinct, molecularly. cervical samples, as depicted in Fig. 4. The episomal CaCx cases appeared to cluster separately, whereas the inte- miR-181c target gene, CKS1B, correlates negatively with grated CaCx cases clustered together with HPV16- HPV16 E7 expression among the HPV16-integrated CaCx positive non-malignant samples and HPV-negative con- cases trols. Such analysis further confirmed that CKS1B was In an earlier study , we recorded that the two cate- differentially overexpressed among episomal CaCx cases gories of CaCx cases differ with respect to levels of viral possibly as a consequence of differentially decreased oncogene E7 expression. Hence, it became necessary to expression of miR-181c among such cases. On the other explore if the viral oncogene expression had any influence hand, CKS1B is transcriptionally regulated by MYCN, on the expression of the cellular gene, CKS1B, which is which has been predicted to be inhibited among the CaCx targeted by miR-181c. Among all the CaCx cases, irre- cases harbouring integrated HPV16, based on IPA in this spective of the viral physical status and among the epi- study (Table S2). This observation was therefore, in somal CaCx cases, no significant correlation was recorded concordance with the fact that CKS1B failed to show any between the expression of these cellular genes and viral alteration in expression, compared with controls, in such oncogene E7. But strikingly, only among integrated CaCx Official journal of the Cell Death Differentiation Association Mandal et al. Cell Death Discovery (2019) 5:81 Page 7 of 13 Fig. 3 Box plots representing distribution of downregulated miRNA expressions (miR-34a, miR-196b, and miR-203, normalised with endogenous control RNU6b and miR-127) among different categories of cervical samples. Lower ΔCT means higher expression. A= HPV- negative control samples (n = 25), B = HPV16-positive non-malignant samples (n = 25), C= HPV16-positive episomal CaCx cases (n = 43), D = HPV16-positive integrated CaCx cases (n = 19) cases, a strong negative correlation was identified between Regulation”, “Cell Cycle Control of Chromosomal Repli- E7 and CKS1B expression (R = 0.516; p = 0.03) as shown cation, Mitotic Roles of Polo-Like Kinase”, and “Role of in Fig. 7. BRCA1 in DNA Damage Response”. Altered expression of proteins belonging to these pathways exert direct Discussion influence on the efficacy of antitumor agents as well as 20–23 Global gene expression profiling has been used widely tumour outcomes . Therefore, a clinical correlation to identify genes that are aberrantly expressed in cervical of genes captured in the enriched pathways would help 17,18 19 tumours as compared with normal cervix , and . in determining the translational efficacy of the findings Considering our earlier study depicting the relevance of recorded in this study. differences between CaCx cases harbouring episomal and The most significant pathways identified in case of integrated viral genomes , in the current study, we chose episomal CaCx cases was “CTLA4 signalling in cytotoxic to identify differentially expressed host genes and relevant T lymphocytes” and “T-cell receptor signalling”, portray- pathways for CaCx pathogenesis in general or specifically ing overexpression of most of the genes captured in such distinct among episomal and integrated CaCx cases. pathways. Such exclusively enriched pathways could Besides confirming this, the most important pathways probably be attributable to higher viral load and E7 identified for CaCx pathogenesis, irrespective of HPV expression, in such cancers harbouring episomal HPV16 physical status were the cell cycle regulatory pathways and reflective of the role of the tumour immune micro- such as “Cell Cycle: G2/M DNA Damage Checkpoint environment or the role of tumour infiltrating Official journal of the Cell Death Differentiation Association Mandal et al. Cell Death Discovery (2019) 5:81 Page 8 of 13 Fig. 4 Expression status of miR-181c and its correlation with CKS1B expression. a Box plots representing distribution of miR-181c expression among different categories of cervical samples. b Linear regression analysis of the correlation between CKS1B mRNA expression and miR-181c expression in all cervical samples (p = 0.03) Lower ΔC means higher expression. A= HPV-negative control samples (n = 25), B = HPV16-positive non- malignant samples (n = 25), C = HPV16-positive episomal CaCx cases (n = 43), D = HPV16-positive integrated CaCx cases (n = 19) 26 27 lymphocytes in such CaCx subtypes. Therefore, genes of cancers and ovarian cancers have been recorded and such pathways could potentially serve as targets for attributed as contributing to the epithelial phenotype combating such cancers, efficiently. On the other hand, and cell proliferation maintaining the tumour bulk, the most significantly enriched pathways in CaCx cases respectively. Although such phenomenon might explain harbouring integrated HPV16 were signalling pathways our findings and be prognostically relevant, it could also such as “eIF2 Signalling”, “Regulation of eIF4 and p70S6K be worthwhile to hypothesise that this miRNA may Signalling”, and “mTOR Signalling”, together with serve as an early marker of HPV16 infection, specifically “Mitochondrial Dysfunction” pathway. Such signalling for singling out HPV16-positive women who might be pathways associate with processes like oxidative stress at heightened risk for CaCx development. and DNA repair, autophagy, and cell migration. These Unlike miR-200a, miR-181c showed differentially observations could be suggestive of the potential role of downregulated expression, exclusively among episomal the host virus fusion proteins that are common among CaCx cases but not among integrated CaCx cases. miR- HPV16-integrated CaCx cases, where heterogeneous 181c has already been reported as a p53-miR, which tar- levels of viral oncoprotein E7 do not correlate with viral gets a number of components of the miRNA processing 13 28 29 copy numbers . complex . A study suggested that expression level of Among the miRNAs with altered expression levels miR-181c in the serum can be used as a biomarker for recorded in our study, majority followed expression pat- therapeutic efficacy in cervical cancers. Thus, miR-181c terns that have already been recorded in case of CaCx expression status could potentially serve as a novel risk cases, with the exception of miR-200a and miR-181c. The marker for episomal CaCx cases and calls for functional overexpression of miR-200a revealed a pattern that dif- characterisation of the role of this miRNA in episomal fered from the other miRNAs analysed, suggesting that CaCx pathogenesis. We tested this possibility by focussing this miRNA could be associated with HPV16 positivity. our attention on the miR-181c target gene, CKS1B, which A recent study identified overexpression of miR-200a belonged to the significantly enriched pathway in cervical in HPV induced CaCx and tonsillar cancers, which seems cancers. We identified that CKS1B was differentially to be consistent with our observation. Another recent overexpressed only among episomal CaCx cases, whose study by the Cancer Genome Atlas Research Net- expression was significantly and inversely correlated with work, 2017 identified that expression of miR-200 family miR-181c expression in such CaCx subtype. was negatively correlated with those of the Enhanced expression of CKS1B in a number of cancers epithelial–mesenchymal transition-related transcription such as myeloma, breast cancer, lymphoma, renal carci- factors. Overexpression of miR-200a in pancreatic noma, ovarian cancer, salivary, and oesophageal cancers Official journal of the Cell Death Differentiation Association Mandal et al. Cell Death Discovery (2019) 5:81 Page 9 of 13 Fig. 5 Correlation between miR-323 and E7 expression among cervical cancers. Linear regression analysis of the correlation between miR-323 expression and E7 mRNA expression in (a) CaCx cases (episomal and integrated; p = 0.471), (b) episomal CaCx cases (p = 0.011), and (c) integrated CaCx cases (p= 0.711). Lower ΔC means higher expression etc. has been associated with poor prognosis. This has Hence, availability of a small molecule inhibitor of CKS1B, been attributed to amplification of CKS1B gene leading to fluoxetine, which is used as an antidepressant in clinics , lymph node metastases . Besides influencing cell growth suggests that drugging CKS1B could be a potential avenue Kip1 and survival through regulation of p27 , silencing of of treating such CaCx subtypes harbouring episomal CKS1B is also known to induce cell death and inhibit HPV16, as opposed to those harbouring purely integrated growth of tumour cells through mechanisms that are HPV16 and this could be subject to clinical trials. 31,32 independent of p27Kip1and SKP2 . Taken together, Our previous study also demonstrated that there exists our study highlights a novel mechanism of CKS1B over- heterogeneity among integrated CaCx cases, based on expression in HPV16 episomal CaCx subtypes, which viral load and E7 expression level . The negative corre- could be attributable to loss of epigenetic regulation of lation between E7 expression and expression of CKS1B CKS1B by miR-181c, which is downregulated in such among such cases potentially suggests the complementary CaCx subtypes. Thus, this could serve as one of the role of viral and host proteins, in the pathogenesis of such mechanisms of proliferation activation in this CaCx sub- CaCx cases harbouring integrated HPV16 irrespective type that could be of translational relevance. of miR-181c (Fig. 7). Thus, our study substantiates that There exist at least two transcript variants of CKS1B CaCx cases harbouring integrated and episomal HPV16 gene and it appears that only transcript variant 1 encodes are distinct at the molecular level. the protein . In our study, we employed primers designed Besides miR-181c, our study also revealed a strong to capture the expression of this coding variant of CKS1B. positive correlation between miR-323 expression and Official journal of the Cell Death Differentiation Association Mandal et al. Cell Death Discovery (2019) 5:81 Page 10 of 13 Fig. 6 miR-323 expression among various categories of CaCx cell lines. Lower ΔC means higher expression In summary, employing both microarray-based gene HPV16 E7 expression among episomal CaCx cases. There are reports showing miR-323 upregulation in metastatic expression profiling and expression analysis of a priori- squamous cell CaCx cases and several studies on pros- tised set of miRNAs, we have been instrumental in 36,37 tate cancers , which are suggestive of the potential of establishing that CaCx cases harbouring both integrated this microRNA in imparting aggressiveness to such can- and episomal HPV16 genomes are distinct molecular cers. We also validated this through cell line based in vitro subtypes. Although the TCGA study on cervical cancers analysis as indicated in Fig. 4. This was indicative of has been instrumental in highlighting distinct molecular the fact that the viral oncogene E7 could regulate host phenotypes of HPV-active and HPV-inactive cancers, we miRNA expression, which calls for further studies on have succeeded in revealing distinct molecular subtypes of detailed molecular mechanisms. The influence of HPV16 HPV16-positive cervical cancers. Our study further oncoproteins E6 and E7 on miR-203 expression during demonstrated deregulation of miRNA expression in CaCx proliferation and differentiation has already been estab- cases compared with controls, progressive deregulation lished . Therefore, the significant positive correlation of miRNA expression with increase in severity of cervical between the expression of miR-203 and E7 in CaCx status based on histopathology and HPV infections, and subtype harbouring integrated HPV16 also merits atten- differential expression of miRNAs in CaCx cases har- tion and further exploration. bouring episomal and integrated viral genomes. There- In this study, through the correlation analysis between fore, in consideration with our previous observation of the the prioritised set of miRNAs and their target genes loss of miRNA-binding sites in viral regulatory regions (experimentally validated and predicted), we have been facilitating viral gene expressions in CaCx cases and instrumental in identifying functional miRNA-mRNA gain of miRNA binding sites in host regulatory regions relationships as detailed in Table S6. Infact, a large pro- we conclude that host cellular miRNAs have a role in portion of the non-validated target genes revealed nega- CaCx pathogenesis. Thus, our findings highlight that tive correlations with the respective miRNAs at the level CaCx cases with episomal and integrated HPV16 are of expression, suggesting that these could be further distinct entities at the molecular level and such findings analysed for exploring the regulation of miRNAs and are of immense translational value, specifically for mRNAs in CaCx pathogenesis. designing targeted therapy of CaCx cases. Official journal of the Cell Death Differentiation Association Mandal et al. Cell Death Discovery (2019) 5:81 Page 11 of 13 Fig. 7 Correlation between CKS1B and E7 expression among cervical cancers. Linear regression analysis of the correlation between CKS1B mRNA expression and E7 mRNA expression in (a) CaCx cases (episomal and integrated; p = 0.839), (b) episomal CaCx cases (p = 0.61), and (c) integrated CaCx cases (p = 0.03). Lower ΔC means higher expression Materials and methods hysterectomy attending a hospital (Calcutta Medical Samples and subjects College Hospital, Kolkata, West Bengal, India). All such The samples used for this study were nested to an samples were collected during the period of 2010–2013, ongoing natural cohort study. The malignant samples with informed consent approved by the institutional selected for the study were histopathologically confirmed ethical committee for human experimentation. Details invasive squamous cell carcinomas and clinically diag- regarding DNA isolation, HPV screening and determi- nosed as tumour stage III and above as per FIGO classi- nation of HPV16, physical status of HPV16 genomes, and 4,41–44 fication. Samples were derived from married subjects viral load are described earlier from our laboratory . attending a cancer referral hospital (Saroj Gupta Centre and Research Institute, South 24 Parganas, West Bengal, Microarray-based gene expression analysis India) collected within the period of 1998–2013. Global gene expression profilingdatareportedbySharma The non-malignant and control samples were derived et al. (GEO database, NCBI Accession No: GSE67522) was from married and non-pregnant women with no previous used following reclassification of the CaCx samples based history of cervical dysplasia/malignancy undergone for on the physical status of HPV16 genome, to identify Official journal of the Cell Death Differentiation Association Mandal et al. Cell Death Discovery (2019) 5:81 Page 12 of 13 mechanistic differences between cervical cancers harbour- pcDNA3.1-HPV16 E7 vector generated in our labora- ing episomal and integrated HPV16. Thus, differential tory . The cells were harvested and washed with 1× PBS expression analysis was performed using different categories (pH 7.4), trypsinized, and collected by centrifugation of cervical tissues, i.e., HPV-negative control samples at 300 g for 10 min The transfected cells were used (n= 11), HPV positive non-malignant samples (n= 11), further for RNA isolation. The transfection experiments episomal CaCx cases (n= 10), and integrated CaCx cases were carried out in three sets, each in triplicates and (n=10)(TableS8).The analysis pipelinewas thesameas selected miRNAs expression was carried out in selected reported earlier . cell lines according to the protocol described earlier. Expression analysis of some candidate microRNAs Acknowledgements We thank Saroj Gupta Cancer Centre and Research Institute (SGCC & RI, The expression of some prioritised miRNAs (34 Thakurpukur, South 24 Parganas, West Bengal, India) and Calcutta Medical miRNAs selected on the basis of frequent alterations College Hospital (Kolkata, West Bengal, India) for their support in sample in many cancers, identified through literature survey) collection; Dr. Samsiddhi Bhattacharjee of the National Institute of Biomedical Genomics (NIBMG), Kalyani, India. Dr. Saurabh Ghosh and Dr. Indranil were determined by TaqMan miRNA assays as well as Mukhopadhyay of Human Genetics Unit, Indian Statistical Institute, Kolkata, Power SYBR Green based assays (Supplementary India for providing guidelines for statistical analysis of the data; Department of methods 1 and 2 and Figures S7 and S8, respectively) Biotechnology, Government of India (Grant No.: BT/PR2012/MED/29/312/2011) and NIBMG, Kalyani, India for funding and NIBMG for technical support and employing 25 HPV-negative normal samples, 25 special thanks also to Council of Scientific and Industrial Research, India for HPV16-positive non-malignant samples, and 62 providing Ms. Paramita Mandal and Ms. Sweta Sharma Saha with a Fellowship (JRF and SRF) to work on this project. HPV16-positive CaCx cases (43 episomal and 19 inte- grated HPV16 genomes). The various categories of Author details miRNAs analysed, and the assays employed are provided National Institute of Biomedical Genomics, Kalyani, West Bengal, India. in Supplementary methods 1–3and Tables S4,Table S9, Crystallography & Molecular Biology Division, Saha Institute of Nuclear Physics, 1/AF Bidhannagar, Kolkata 700064, India. Department of Gynecology, Saroj and Table S10. Gupta Cancer Centre and Research Institute, Kolkata, India. Sri Aurobindo Seva Kendra, 1H, Gariahat Road (S) Jodhpur Park, Kolkata 700068 West Bengal, India. Target prediction of miRNA Health Science Library System, University of Pittsburgh, Pittsburgh, PA, USA. Present address: Department of Zoology, The University of Burdwan, The target human genes corresponding to a subset of Burdwan, West Bengal, India. Present address: Section of Haematology/ the miRNAs analysed in the study, were predicted based Oncology, Department of Medicine, university of Chicago, 5841 S Maryland on three prediction softwares like miRanda, PICTAR, Ave MC 2115, Chicago, IL 60637, USA. Present address: Molecular Biology and Genetics Unit, Jawaharlal Nehru Centre for Advanced Scientific Research, and TargetScan, considering validated miRNA binding Jakkur, Bengaluru, Karnataka 560064, India. Present address: Comprehensive sites as well. Wound Center, Center for Regenerative Medicine and Cell Based Therapies, The Ohio State University, Columbus, OH, USA Validation of miRNA target gene CKS1B by quantitative Conflict of interest real-time PCR The authors declare that they have no conflict of interest. Validation was done considering only CKS1B gene that appeared as target of the significantly altered miR-181c Publisher’s note recorded among episomal CaCx cases only. This gene was Springer Nature remains neutral with regard to jurisdictional claims in selected based on the fact it belonged to the significantly published maps and institutional affiliations. enriched pathway corresponding to the CaCx cases. Samples used for validation by quantitative real-time The online version of this article (https://doi.org/10.1038/s41420-019-0154-x) PCR included episomal CaCx cases (n = 43), integrated contains supplementary material, which is available to authorised users. CaCx cases (n = 19), HPV-negative controls (n = 25), and HPV16-positive non-malignant samples (n = 25). The Received: 6 December 2018 Revised: 3 February 2019 Accepted: 13 February 2019 details of primer sequences of CKS1B and GAPDH, amplicon size and PCR conditions employed is depicted in Table S10. References Cell culture and transfection 1. Das, D., Bhattacharjee, B.,Sen,S., Mukhopadhyay,I.& Sengupta,S.Association of viral load with HPV16 positive cervical cancer pathogenesis: Causal HPV-negative cell line C33A and HPV16-positive cell relevance in isolates harboring intact viral E2 gene. Virology 402,197–202 lines Caski and SiHa were cultured in Dulbecco's (2010). Modified Eagle Medium supplemented with 10% fetal 2. Thomas, M., Massimi, P., Navarro, C., Borg, J. P. & Banks, L. The hScrib/Dlg apico- basal control complex is differentially targeted by HPV-16 and HPV-18 E6 bovine serum, 50 Units/ml of penicillin and 50 µg/ml proteins. Oncogene 24, 6222–6230 (2005). of streptomycin at 37 °C and 5% CO . C33A cells were 3. Romanczuk, H. & Howley, P. M. Disruption of either the E1 or the E2 regulatory transfected using Lipofectamine 2000 reagent according gene of human papillomavirus type 16 increases viral immortalization capa- to manufacturer’s protocol, using 1 µg of plasmid city. Proc. Natl. Acad. Sci. USA 89,3159–3163 (1992). Official journal of the Cell Death Differentiation Association Mandal et al. Cell Death Discovery (2019) 5:81 Page 13 of 13 4. Bhattacharjee, B. & Sengupta, S. CpG methylation of HPV 16 LCR at E2 binding 27. Liu, N. et al. Upregulation of microRNA-200a associates with tumor prolifera- site proximal to P97 is associated with cervical cancer in presence of intact E2. tion, CSCs phenotype and chemosensitivity in ovarian cancer. Neoplasma 62, Virology 354, 280–285 (2006). 550–568 (2015). 5. Umbach, J. L. et al. MiRNAs expressed by herpes simplex virus1 during latent 28. Boominathan, L. The tumor suppressors p53, p63, and p73 are regulators infection regulate viral mRNAs. Nature 454, 780–783 (2008). of MicroRNA processing complex. PLoS. ONE 5, e10615 (2010). 6. Skalsky, R. L. & Cullen, B. R. Viruses, microRNAs, and Host Interactions. Annu. 29. Wang, W. T. et al. Differentially expressed microRNAs in the serum of cervical Rev. Microbiol. 64, 123–141 (2010). squamous cell carcinoma patients before and after surgery. J. Hematol. Oncol. 7. Guo, Y. E. & Steitz, J. A. Virus meets host MicroRNA: the destroyer, the booster, 7, 6 (2014). the hijacker. Mol. Cell Biol. 34, 3780–3787 (2014). 30. Zhang, Y. CKS1B (CDC28 protein kinase regulatory subunit 1B). Atlas Genet. 8. Calin, G. A. et al. Human microRNA genes are frequently located at fragile sites Cytogenet. Oncol. Haematol. 14,676–678 (2010). and genomic regions involved in cancers. Proc. Natl. Acad. Sci. USA 101, 31. Liberal, V. et al. Cyclin-dependent kinase subunit (Cks) 1 or Cks2 over- 2999–3004 (2004). expression overrides the DNA damage response barrier triggered by activated 9. Farazi, T.A., Hoell, J. I.,Morozov,P.&Tuschl, T.microRNAs in humancancer. oncoproteins. Proc.Natl. Acad.Sci.USA 109,2754–2759 (2012). Adv. Exp. Med. Biol. 774,1–20 (2013). 32. Lee, S. W. et al. Overexpression of CDC28 protein kinase regulatory subunit1B 10. Calin, G. A. & Croce, C. M. MicroRNA-cancer connection: the beginning of a confers an independent prognostic factor in nasopharyngeal carcinoma. new tale. Cancer Res. 66,7390–7394 (2006). APMIS 122, 206–214 (2014). 11. Sharma, G., Dua, P. & Agarwal, S. M. A comprehensive review of dysregulated 33. Fagerberg, L. et al. Analysis of the human tissue-specificexpressionby miRNAs involved in cervical cancer. Curr. Genomics 15, 310–323 (2014). genome-wide integration of transcriptomics and antibody-based proteomics. 12. Jiménez-Wences, H. et al. Methylation and expression of miRNAs in pre- Mol. Cell Proteom. 13,397–406 (2014). cancerous lesions and cervical cancer with HPV16 infection. Oncol. Rep. 35, 34. Chayka, O., D'Acunto, C. W., Middleton, O., Arab, M. & Sala, A. Identification 2297–2305 (2016). and pharmacological inactivation of the MYCN gene network as a 13. Das Ghosh, D. et al. Some novel insights on HPV16 related cervical cancer therapeutic strategy for neuroblastic tumor cells. J. Biol. Chem. 290, pathogenesis based on analyses of LCRmethylation, viral load, E7 and E2/E4 2198–2212 (2015). expressions. PLoS. ONE 7, e44678 (2012). 35. Ding, H.,Wu, W. L.,Wang, Y. X. &Zhu,F.F. Characterizationofthe microRNA 14. McKenna,D.J., McDade,S.S., Patel, D. &McCance,D.J.MicroRNA203 expression profile of cervical squamous cell carcinoma metastases. Asian Pac. expression in keratinocytes is dependent on regulation of p53 levels by E6. J. J. Cancer Prev. 15, 1675–1679 (2014). Virol. 84, 10644–10652 (2010). 36. Gao, Q., Yao, X. & Zheng, J. MiR-323 inhibits prostate cancer 15. Sharma, S. et al. Bridging links between long noncoding RNA HOTAIR and vascularization through adiponectin receptor. Cell Physiol. Biochem. 36, HPV oncoprotein E7 in cervical cancer pathogenesis. Sci. Rep. 5, 11724 (2015). 1491–1498 (2015). 16. Bartel, D. P. MicroRNAs: target recognition and regulatory functions. Cell 136, 37. Gao, Q. & Zheng, J. microRNA-323 upregulation promotes prostate cancer 215–233 (2009). growth and docetaxel resistance by repressing p73. Biomed. Pharmacother. 97, 17. Ahn, W. S. et al. Searching for pathogenic functions to cervical cancer. Gynecol. 528–534 (2018). Oncol. 93,41–48 (2004). 38. Banister, C. E., Liu, C., Pirisi, L., Creek, K. E. & Buckhaults, P. J. Identification 18. Medina-Martinez, I. et al. Impact of gene dosage on gene expression, biolo- and characterization of HPV-independent cervical cancers. Oncotarget 8, gicalprocesses andsurvivalincervicalcancer: agenome-wide follow-up 13375–13386 (2017). study. PLoS. ONE 9, e97842 (2014). 39. Mandal, P. et al. Differential expression of HPV16 L2 gene in cervical cancers 19. Lee, H. S. et al. Identification of differentially-expressed genes by DNA harboring episomal HPV16 genomes: influence of synonymous and non- methylation in cervical cancer. Oncol. Lett. 9,1691–1698 (2015). coding region variations. PLoS. ONE 8, e65647 (2013). 20. Takai,N., Hamanaka,R., Yoshimatsu,J.&Miyakawa, I.Polo-like kinases(Plks) 40. Sharma Saha, S. et al. Identification of genetic variation in the lncRNA HOTAIR and cancer. Oncogene 24,287–291 (2005). associated with HPV16-related cervical cancer pathogenesis. Cell. Oncol. 39, 21. Kauffmann, A. et al. High expression of DNA repair pathways is associated with 559–572 (2016). metastasis in melanoma patients. Oncogene 27, 565–573 (2008). 41. Bhattacharjee, B. & Sengupta, S. HPV16 E2 gene disruption and polymorph- 22. Chen, R. H. & Tian, Y. J. Enhanced anti-tumor efficacy of aspirin combined with isms of E2 and LCR: some significant associations with cervical cancer in triptolide in cervical cancer cells. AsianPac.J.Cancer Prev. 14, 3041–3044 Indian women. Gynecol. Oncol. 100,372–378 (2006). (2013). 42. Bhattacharya, P. & Sengupta, S. Predisposition to HPV16/18-related cervical 23. Ujhelyi, Z. et al. The enhanced inhibitory effect of different antitumor agents in cancer because of proline homozygosity at codon 72 of p53 among Indian self-microemulsifying drug delivery systems on human cervical cancer HeLa womenisinfluenced by HLA-B*07 and homozygosity of HLA-DQB1*03. Tissue cells. Molecules 20, 13226–13239 (2015). Antigens 70,283–293 (2007). 24. Vojtechova, Z. et al. Comparison of the miRNA profiles in HPV-positive and 43. Laikangbam, P. et al. A comparative profile of the prevalence and age dis- HPV-negative tonsillar tumors and a model system of human keratinocyte tribution of human papillomavirus type 16/18 infections among three states clones. BMC Cancer 16, 382 (2016). of India with focus on northeast India. Int. J. Gynecol. Cancer 17,107–117 25. Cancer Genome Atlas Research Network. Integrated genomic and molecular (2007). characterization of cervical cancer. Nature 543,378–384 (2017). 44. Bhattacharjee, B., Mandal, N. R., Roy, S. & Sengupta, S. Characterisation 26. Li, A. et al. Pancreatic cancers epigenetically silence SIP1 and hypomethylate of sequence variations within HPV16 isolates among Indian women: and overexpress miR-200a/200b in association with elevated circulating prediction of causal role of rare non-synonymous variations within intact miR-200a and miR-200b levels. Cancer Res. 70,5226–5237 (2010). isolates in cervical cancer pathogenesis. Virology 377,143–150 (2008). Official journal of the Cell Death Differentiation Association http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Cell Death Discovery Springer Journals

Cervical cancer subtypes harbouring integrated and/or episomal HPV16 portray distinct molecular phenotypes based on transcriptome profiling of mRNAs and miRNAs

Download PDF
13 pages

Loading next page...
 
/lp/springer-journals/cervical-cancer-subtypes-harbouring-integrated-and-or-episomal-hpv16-0Tw4FMrxeD

References (63)

Publisher
Springer Journals
Copyright
Copyright © 2019 by The Author(s)
Subject
Life Sciences; Life Sciences, general; Biochemistry, general; Cell Biology; Stem Cells; Apoptosis; Cell Cycle Analysis
eISSN
2058-7716
DOI
10.1038/s41420-019-0154-x
Publisher site
See Article on Publisher Site

Abstract

Heterogeneity in cervical cancers (CaCx) in terms of HPV16 physical status prompted us to investigate the mRNA and miRNA signatures among the different categories of CaCx samples. We performed microarray-based mRNA expression profiling and quantitative real-time PCR-based expression analysis of some prioritised miRNAs implicated in cancer- related pathways among various categories of cervical samples. Such samples included HPV16-positive CaCx cases that harboured either purely integrated HPV16 genomes (integrated) and those that harboured episomal viral genomes, either pure or concomitant with integrated viral genomes (episomal), which were compared with normal cervical samples that were either HPV negative or positive for HPV16. The mRNA expression profile differed characteristically between integrated and episomal CaCx cases for enriched biological pathways. miRNA expression profiles also differed among CaCx cases compared with controls (upregulation—miR-21, miR-16, miR-205, miR-323; downregulation—miR-143, miR-196b, miR-203, miR-34a; progressive upregulation—miR-21 and progressive downregulation—miR-143, miR-34a, miR-196b and miR-203) in the order of HPV-negative controls, HPV16-positive non-malignant samples and HPV16-positive CaCx cases. miR-200a was upregulated in HPV16-positive cervical tissues irrespective of histopathological status. Expression of majority of the predicted target genes was negatively correlated with their corresponding miRNAs, irrespective of the CaCx subtypes. E7 mRNA expression correlated positively with miR-323 expression among episomal cases and miR-203, among integrated cases. miR-181c expression was downregulated only among the episomal CaCx cases and negatively correlated with protein coding transcript of the proliferative target gene, CKS1B of the significantly enriched “G2/M DNA Damage Checkpoint Regulation” pathway among CaCx cases. Thus, the two CaCx subtypes are distinct entities at the molecular level, which could be differentially targeted for therapy. In fact, availability of a small molecule inhibitor of CKS1B, suggests that drugging CKS1B could be a potential avenue of treating the large majority of CaCx cases harbouring episomal HPV16. Introduction Correspondence: Sharmila Sengupta (ssg1@nibmg.ac.in)(sharmilasg@gmail. Cervical cancer (CaCx) appears to be the most common com) National Institute of Biomedical Genomics, Kalyani, West Bengal, India malignancy in Indian women, characterised by high Crystallography & Molecular Biology Division, Saha Institute of Nuclear Physics, incidence and mortality rates that are attributable to late 1/AF Bidhannagar, Kolkata 700064, India detection and lack of access to affordable health care. Full list of author information is available at the end of the article. Edited by E. Sayan © The Author(s) 2019 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to theCreativeCommons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. Official journal of the Cell Death Differentiation Association 1234567890():,; 1234567890():,; 1234567890():,; 1234567890():,; Mandal et al. Cell Death Discovery (2019) 5:81 Page 2 of 13 Oncogenic human papillomavirus (HPV) infections are respect to such gene expression profiles and corre- the major aetiological factors, with a predominance of sponding enriched pathways. We further explored whe- HPV16 that accounts for over 50% of such cancers . Viral ther a prioritised set of miRNAs, implicated in various oncoprotein E6 is known to interact with the tumour cancer-related pathways, show altered expression and suppressor p53 and E7, with the PDZ domain of cellular play a role in CaCx pathogenesis under the impact proteins and Rb, thereby facilitating neoplastic progres- of HPV16 infection. Subsequently, we determined whether the miRNAs showing altered expression in sion through loss of cellular homoeostasis . Loss of E2 repressor activity, as a consequence of integration of the CaCx cases, correlate with (i) the physical status of virus into the host genome, is known to impart pro- HPV16 genomes (episomal and integrated), (ii) expression liferative advantage to the host cells . Besides this, our of the oncoprotein E7, (iii) the expression of cellular earlier studies have highlighted mechanisms that act to target genes of specific biological pathways and checked block the E2-mediated repression of the P97 promoter in the relationship, if any, between such target genes and CaCx cases harbouring episomal HPV16 and revealed that viral oncoprotein E7. such cancers harbour higher viral load and E7 expression in contrast to the ones that portray integrated HPV16. Results Cervical carcinogenesis, under the impact of oncogenic Microarray-based gene expression profiling reveals HPV infections, could be influenced by several epigenetic characteristic differences among CaCx cases harbouring mechanisms like DNA methylation, histone modifica- episomal and integrated HPV16 genomes tions, by microRNAs, etc., through perturbations of To identify the differences between CaCx cases har- gene expression profiles. In this communication, we bouring episomal and integrated HPV16 genomes, we chose to focus on the expression of host mRNAs and first attempted to identify global gene expression level miRNAs in various categories of cervical samples changes between the two categories of CaCx cases. Our spanning the discrete stages of CaCx development. microarray-based analysis revealed that 334 genes were These small non-coding RNAs, through recognition of differentially upregulated and 253 were differentially sequence-complementary target elements, can either downregulated among episomal CaCx cases, whereas translationally suppress or catalytically degrade both cel- 1373 genes were differentially upregulated and 1240 were 5,6 lular and viral RNAs . Both DNA and RNA viruses have differentially downregulated among integrated CaCx evolved mechanisms to degrade, boost, or hijack cellular cases, compared with the HPV-negative control samples miRNAs to benefit the viral life cycle (Figure S1). Ingenuity pathway analysis (IPA) revealed . Therefore, a glimpse of the deregulated expression of mRNAs and that biological processes such as cell proliferation, retro- miRNAs in CaCx cases might offer some insights into the viridae infection, and viral Infections were the most sig- mechanisms of CaCx pathogenesis among the two sub- nificantly enriched among the episomal CaCx cases. On types that bear integrated HPV16 and episomal HPV16 in the other hand, biological processes such as transcription, presence or absence of integration, subsequently referred proliferation of tumour cells, and protein metabolism to as integrated and episomal CaCx cases, respectively. were most significantly enriched among integrated CaCx About 50% of the known human miRNAs are located cases (Table S1). IPA also revealed that role of CTLA4- at cancer-associated regions of the genome and a num- signalling in cytotoxic T lymphocytes and T-cell receptor ber of miRNAs are located close to HPV integration sites. signalling were the top significantly altered pathways Host miRNAs are also capable of influencing viral life among the episomal CaCx cases, whereas EIF2 signalling, cycles, viral tropism, and the pathogenesis of viral dis- regulation of eIF4 and p70S6K signalling, mitochondrial eases . Therefore, miRNA-mediated epigenetic regulation dysfunction, and mTOR signalling were the top sig- of gene expression might play a role in virus associated nificantly altered pathways among integrated CaCx cases cancers, like CaCx. (Table S2). Moreover, the upstream regulators involved Some studies have strongly suggested that miRNA in alteration of gene expression, differed distinctly among profiling is more robust than mRNA profiling in deter- the two CaCx subtypes (Table S3). Such observations mining the heterogeneity among cancers . The studies further strengthen our hypothesis that the episomal and 11,12 on CaCx have also demonstrated the association of integrated CaCx cases represent two distinct molecular miRNA deregulation with tumour heterogeneity. The subtypes. heterogeneity among CaCx in terms of physical status of HPV16 genomes by our group , prompted us to explore miRNAs reveal altered expression among CaCx cases the mRNA and miRNA signatures among the different compared with HPV-negative controls and HPV16-positive categories of CaCx samples. non-malignant samples Hence, we tested the hypothesis that CaCx cases with The 34 miRNAs selected for this study belonged to integrated and episomal HPV16 genomes differ with eight different categories by function, as depicted in Official journal of the Cell Death Differentiation Association Mandal et al. Cell Death Discovery (2019) 5:81 Page 3 of 13 Table S4. Of these, nine miRNAs revealed significant samples (fold change = −15.89 fold; p < 0.001). Similar alterations in expression (p value < 0.05) among CaCx analysis considering CaCx cases with integrated HPV16, cases compared with HPV-negative controls, as well as failed to show differential expression of miR-181c among HPV16-positive non-malignant samples as depicted in such CaCx cases. Thus, HPV16 physical status within Table 1. Among these, four miRNAs (miR-21, miR-16, CaCx cases failed to portray any profound effect on miR-205, and miR-323) were significantly overexpressed, majority of the miRNAs as depicted in Table 1. while significant decreased expression was recorded in case of four miRNAs (miR-143, miR-196b, miR-203, and HPV16 E7 mRNA expression levels activate expression of miR-34a). miR-200a, however, showed a statistically sig- miR-323 resulting in upregulation among CaCx cases nificant increased expression by 37.53-fold, among CaCx Based on our clinical sample data we identified that only cases compared with HPV-negative controls but not when two miRNAs (miR-203 and miR-323) were significantly compared with HPV16-positive non-malignant samples. and positively correlated with E7 mRNA expression, The expression status of all the altered miRNAs are among the CaCx cases. Interestingly, the correlation depicted in Table 1, Figure S2, Fig. 1, Fig. 2 and Fig. 3. between E7 and miR-323 (Fig. 5) was obvious among Likewise, seven miRNAs revealed significant alterations CaCx cases harbouring episomal HPV16, whereas the in expression among HPV16-positive non-malignant correlation between E7 and miR-203 (Figure S4) was samples, compared with HPV-negative controls. Of these, evident among CaCx cases with integrated HPV16. As the two miRNAs (miR-21, miR-200a) revealed significant correlation between miR-203 and E7 has already been increased expression and five miRNAs (miR-143, miR-16, documented in earlier studies , we used an in vitro miR-196b, miR-203, and miR-34a) showed significantly approach to confirm the correlation between miR-323 decreased expression. The expression status of all such expression and E7 expression in CaCx cases. altered miRNAs among the HPV16-positive non-malig- We transfected the HPV-negative CaCx cell line, C33A nant sample group is depicted in Table S5, Figure S2, with HPV16 E7 cloned into mammalian expression vector Fig. 1, Fig. 2, and Fig. 3. pcDNA3.1 (+), as explained in our previous study . Progressive increase of miR-21 expression and pro- This resulted in a significant increase in miR-323 gressively decreased expression of miR-143, miR-34a, expression (8.17-fold; p = 0.032) concomitant with miR-196b, and miR-203 was recorded with increase in HPV16 E7 expression (Fig. 6) in such cells. We also severity of cervical status through HPV-negative control determined miR-323 expression in HPV16-positive CaCx samples, HPV16-positive non-malignant samples and cell lines, SiHa, and Caski. Expression of miR-323 was CaCx cases. Such trends were found to be statistically found to be higher among SiHa cells (2.82-fold; p = 0.016) significant (p < 0.001). On the contrary, miR-16 revealed that portray relatively higher E7 expression , as com- a deviation of the trend with decreased expression pared WITH Caski cells (Fig. 6). Thus, both the obser- among HPV16-positive non-malignant samples, followed vations confirmed that E7 could probably activate miR- by increased expression among CaCx cases compared 323 expression in CaCx cases. with HPV-negative controls, as depicted in Figure S3. Such observations are suggestive of the fact that miRNA Target genes of majority of miRNAs that differ in deregulations could potentially serve as risk markers of expression between CaCx cases harbouring episomal and CaCx development. integrated HPV16, also portrayed similar expression level difference between such cases Lack of profound impact of physical status of HPV16 miRNAs are known to target multiple transcripts reg- genome (episomal and integrated) on expression of ulating important cellular processes such as cell pro- majority of miRNAs in CaCx cases, excepting for miR-181c liferation, apoptosis, cellular metastasis, and many more . In view of the biological relevance of episomal HPV16 To check the impact of deregulated expression of the among CaCx cases , expression data of all 34 miRNAs nine significantly altered miRNAs including miR-181c, on considered in this study was further analysed to determine their target genes, we identified the expression profiles whether they showed differentially altered expression of the predicted targets from the microarray-based gene among the CaCx cases with episomal and integrated expression analysis (Table S6). We selected the target HPV16, when compared with HPV-negative controls and genes of the above mentioned 10 miRNAs, considering HPV16-positive non-malignant samples. The sole miRNA only those targets that portrayed ≥ or ≤ 2 fold change in that revealed significantly altered expression only among expression as compared with the controls. Majority of episomal CaCx cases was miR-181c, as depicted in Fig. 4. such target genes of the corresponding miRNAs seemed miR-181c expression was significantly increased in such to be experimentally non-validated, as per the databases cases (fold change = −2.34-fold, p < 0.001), compared considered for target identification as detailed under the with both HPV-negative and HPV16-positive control “Methods” section. For miR-181c, only the expression Official journal of the Cell Death Differentiation Association Mandal et al. Cell Death Discovery (2019) 5:81 Page 4 of 13 Official journal of the Cell Death Differentiation Association Table 1 Expression status of different categories of miRNAs among CaCx cases compared with HPV-negative controls and HPV16-positive non-malignant samples Category microRNAs Comparisons Episomal CaCx cases vs HPV- Integrated CaCx cases vs HPV- Episomal CaCx cases vs HPV16- Integrated CaCx cases vs HPV16- negative controls negative controls positive non-malignant samples positive non-malignant samples Mann–Whitney Up FDR of Mann–Whitney Up FDR of Mann–Whitney Up FDR of Mann–Whitney Up FDR of value (fold change) 0.05 value (fold change) 0.05 value (fold change) 0.05 value (fold change) 0.05 Enhance tumour growth miR-21 <0.001 (25.46); 0.0015 <0.001 (15.01); 0.0015 <0.001 (6.54); 0.0015 <0.001 (3.86); 0.0015 and proliferation upregulation upregulation upregulation upregulation miR-200a <0.001 (48.5); 0.01 <0.001 (33.93); 0.01 0.172 (5.11) 0.02 0.062 (2.536) 0.02 upregulation upregulation Differentiation regulatory miR-143 <0.001 (−41.07); 0.003 <0.001 (−76.24); 0.003 <0.001 (−13.36); 0.003 <0.001 (−24.08); 0.003 miRNAs downregulation downregulation downregulation downregulation Target tumour suppressors miR-16 <0.001 (3.58); 0.006 <0.001 (4.32); upregulation 0.006 <0.001 (5.96); 0.006 <0.001 (15.78); 0.006 and tumour suppressor upregulation upregulation upregulation miRNAs miR-205 <0.001 (116.07); 0.012 <0.001 (156.9); 0.012 <0.001 (94.18); 0.01 <0.001 (124.56); 0.01 upregulation upregulation upregulation upregulation miR-323 0.003 (9.87); upregulation 0.014 0.006 (15.77); upregulation 0.014 <0.001 (37.6); 0.012 <0.001 (58.23); 0.012 upregulation upregulation miR-34a <0.001 (−28.6); 0.009 <0.001 (−20); 0.009 <0.001 (−15.09); 0.009 <0.001 (−9.33); 0.009 downregulation downregulation downregulation downregulation Metastatic suppressor miR-196b <0.001 (−39.34); 0.004 <0.001 (−47.21); 0.004 <0.001 (−12.65); 0.004 <0.001 (−20.88); 0.004 miRNAs downregulation downregulation downregulation downregulation Mesenchymal to epithelial miR-203 <0.001 (−4.77); 0.007 <0.001 (−5.79); 0.007 <0.001 (−1.33); 0.007 <0.001 (−1.76); 0.007 transition miRNAs downregulation downregulation downregulation downregulation Metastatic miRNAs None None None None None None None None None Bold indicates significant p value Mandal et al. Cell Death Discovery (2019) 5:81 Page 5 of 13 Fig. 1 Box plots representing distribution of upregulated miRNA expressions (miR-21, miR-200a, and miR-16, normalised with endogenous control RNU6b and miR-127) among different categories of cervical samples. Lower ΔCT means higher expression. A= HPV- negative control samples (n = 25), B = HPV16-positive non-malignant samples (n = 25), C= HPV16-positive episomal CaCx cases (n = 43), D = HPV16-positive integrated CaCx cases (n = 19) status of target genes in case of episomal CaCx cases were gene that was captured in a significantly enriched pathway considered for expression changes, as this miRNA was among CaCx cases, upon employing IPA (Table S2). differentially expressed, compared with controls, only The microarray-based gene expression profile result of among the episomal cases but remained unaltered in CKS1B is depicted in Table S7. Real-time quantitative expression among the HPV16-integrated CaCx cases. PCR-based validation of the expression of CKS1B on a larger set of cervical samples, revealed differentially Overexpression of miR-181c target gene, CKS1B, among increased expression of CKS1B by 21.16-fold (p < 0.001) episomal CaCx cases among CaCx cases (episomal HPV16; n = 43) compared According to microarray-based gene expression analy- with HPV16-positive non-malignant samples (n = 25) sis, 44 miR-181c target genes showed differential and by 18.5-fold (p < 0.001) in comparison with HPV- expression at more than equal to twofolds among the negative controls (n = 25), as depicted in Figure S5. CaCx episomal CaCx cases. These genes were also found to be cases with integrated HPV16 failed to show significant overexpressed among the episomal CaCx cases compared alteration of the expression of this gene, compared with with integrated CaCx cases (Table S6). Out of these 44 HPV16-positive non-malignant samples and HPV- differentially overexpressed genes, CKS1B was the only negative controls. Official journal of the Cell Death Differentiation Association Mandal et al. Cell Death Discovery (2019) 5:81 Page 6 of 13 Fig. 2 Box plots representing distribution of upregulated and downregulated miRNA expressions (miR-205, miR-323, and miR-143, normalised with endogenous control RNU6b and miR-127) among different categories of cervical samples. Lower ΔCT means higher expression. A = HPV-negative control samples (n = 25), B = HPV16-positive non-malignant samples (n = 25), C = HPV16-positive episomal CaCx cases (n = 43), D = HPV16-positive integrated CaCx cases (n = 19) The expression of CKS1B was inversely and significantly CaCx subtypes (Table S5). Such observations also reflect correlated with miR-181c expression (p = 0.03) based on that the two CaCx subtypes with episomal and integrated pair-wise correlation analysis of various categories of HPV16 are distinct, molecularly. cervical samples, as depicted in Fig. 4. The episomal CaCx cases appeared to cluster separately, whereas the inte- miR-181c target gene, CKS1B, correlates negatively with grated CaCx cases clustered together with HPV16- HPV16 E7 expression among the HPV16-integrated CaCx positive non-malignant samples and HPV-negative con- cases trols. Such analysis further confirmed that CKS1B was In an earlier study , we recorded that the two cate- differentially overexpressed among episomal CaCx cases gories of CaCx cases differ with respect to levels of viral possibly as a consequence of differentially decreased oncogene E7 expression. Hence, it became necessary to expression of miR-181c among such cases. On the other explore if the viral oncogene expression had any influence hand, CKS1B is transcriptionally regulated by MYCN, on the expression of the cellular gene, CKS1B, which is which has been predicted to be inhibited among the CaCx targeted by miR-181c. Among all the CaCx cases, irre- cases harbouring integrated HPV16, based on IPA in this spective of the viral physical status and among the epi- study (Table S2). This observation was therefore, in somal CaCx cases, no significant correlation was recorded concordance with the fact that CKS1B failed to show any between the expression of these cellular genes and viral alteration in expression, compared with controls, in such oncogene E7. But strikingly, only among integrated CaCx Official journal of the Cell Death Differentiation Association Mandal et al. Cell Death Discovery (2019) 5:81 Page 7 of 13 Fig. 3 Box plots representing distribution of downregulated miRNA expressions (miR-34a, miR-196b, and miR-203, normalised with endogenous control RNU6b and miR-127) among different categories of cervical samples. Lower ΔCT means higher expression. A= HPV- negative control samples (n = 25), B = HPV16-positive non-malignant samples (n = 25), C= HPV16-positive episomal CaCx cases (n = 43), D = HPV16-positive integrated CaCx cases (n = 19) cases, a strong negative correlation was identified between Regulation”, “Cell Cycle Control of Chromosomal Repli- E7 and CKS1B expression (R = 0.516; p = 0.03) as shown cation, Mitotic Roles of Polo-Like Kinase”, and “Role of in Fig. 7. BRCA1 in DNA Damage Response”. Altered expression of proteins belonging to these pathways exert direct Discussion influence on the efficacy of antitumor agents as well as 20–23 Global gene expression profiling has been used widely tumour outcomes . Therefore, a clinical correlation to identify genes that are aberrantly expressed in cervical of genes captured in the enriched pathways would help 17,18 19 tumours as compared with normal cervix , and . in determining the translational efficacy of the findings Considering our earlier study depicting the relevance of recorded in this study. differences between CaCx cases harbouring episomal and The most significant pathways identified in case of integrated viral genomes , in the current study, we chose episomal CaCx cases was “CTLA4 signalling in cytotoxic to identify differentially expressed host genes and relevant T lymphocytes” and “T-cell receptor signalling”, portray- pathways for CaCx pathogenesis in general or specifically ing overexpression of most of the genes captured in such distinct among episomal and integrated CaCx cases. pathways. Such exclusively enriched pathways could Besides confirming this, the most important pathways probably be attributable to higher viral load and E7 identified for CaCx pathogenesis, irrespective of HPV expression, in such cancers harbouring episomal HPV16 physical status were the cell cycle regulatory pathways and reflective of the role of the tumour immune micro- such as “Cell Cycle: G2/M DNA Damage Checkpoint environment or the role of tumour infiltrating Official journal of the Cell Death Differentiation Association Mandal et al. Cell Death Discovery (2019) 5:81 Page 8 of 13 Fig. 4 Expression status of miR-181c and its correlation with CKS1B expression. a Box plots representing distribution of miR-181c expression among different categories of cervical samples. b Linear regression analysis of the correlation between CKS1B mRNA expression and miR-181c expression in all cervical samples (p = 0.03) Lower ΔC means higher expression. A= HPV-negative control samples (n = 25), B = HPV16-positive non- malignant samples (n = 25), C = HPV16-positive episomal CaCx cases (n = 43), D = HPV16-positive integrated CaCx cases (n = 19) 26 27 lymphocytes in such CaCx subtypes. Therefore, genes of cancers and ovarian cancers have been recorded and such pathways could potentially serve as targets for attributed as contributing to the epithelial phenotype combating such cancers, efficiently. On the other hand, and cell proliferation maintaining the tumour bulk, the most significantly enriched pathways in CaCx cases respectively. Although such phenomenon might explain harbouring integrated HPV16 were signalling pathways our findings and be prognostically relevant, it could also such as “eIF2 Signalling”, “Regulation of eIF4 and p70S6K be worthwhile to hypothesise that this miRNA may Signalling”, and “mTOR Signalling”, together with serve as an early marker of HPV16 infection, specifically “Mitochondrial Dysfunction” pathway. Such signalling for singling out HPV16-positive women who might be pathways associate with processes like oxidative stress at heightened risk for CaCx development. and DNA repair, autophagy, and cell migration. These Unlike miR-200a, miR-181c showed differentially observations could be suggestive of the potential role of downregulated expression, exclusively among episomal the host virus fusion proteins that are common among CaCx cases but not among integrated CaCx cases. miR- HPV16-integrated CaCx cases, where heterogeneous 181c has already been reported as a p53-miR, which tar- levels of viral oncoprotein E7 do not correlate with viral gets a number of components of the miRNA processing 13 28 29 copy numbers . complex . A study suggested that expression level of Among the miRNAs with altered expression levels miR-181c in the serum can be used as a biomarker for recorded in our study, majority followed expression pat- therapeutic efficacy in cervical cancers. Thus, miR-181c terns that have already been recorded in case of CaCx expression status could potentially serve as a novel risk cases, with the exception of miR-200a and miR-181c. The marker for episomal CaCx cases and calls for functional overexpression of miR-200a revealed a pattern that dif- characterisation of the role of this miRNA in episomal fered from the other miRNAs analysed, suggesting that CaCx pathogenesis. We tested this possibility by focussing this miRNA could be associated with HPV16 positivity. our attention on the miR-181c target gene, CKS1B, which A recent study identified overexpression of miR-200a belonged to the significantly enriched pathway in cervical in HPV induced CaCx and tonsillar cancers, which seems cancers. We identified that CKS1B was differentially to be consistent with our observation. Another recent overexpressed only among episomal CaCx cases, whose study by the Cancer Genome Atlas Research Net- expression was significantly and inversely correlated with work, 2017 identified that expression of miR-200 family miR-181c expression in such CaCx subtype. was negatively correlated with those of the Enhanced expression of CKS1B in a number of cancers epithelial–mesenchymal transition-related transcription such as myeloma, breast cancer, lymphoma, renal carci- factors. Overexpression of miR-200a in pancreatic noma, ovarian cancer, salivary, and oesophageal cancers Official journal of the Cell Death Differentiation Association Mandal et al. Cell Death Discovery (2019) 5:81 Page 9 of 13 Fig. 5 Correlation between miR-323 and E7 expression among cervical cancers. Linear regression analysis of the correlation between miR-323 expression and E7 mRNA expression in (a) CaCx cases (episomal and integrated; p = 0.471), (b) episomal CaCx cases (p = 0.011), and (c) integrated CaCx cases (p= 0.711). Lower ΔC means higher expression etc. has been associated with poor prognosis. This has Hence, availability of a small molecule inhibitor of CKS1B, been attributed to amplification of CKS1B gene leading to fluoxetine, which is used as an antidepressant in clinics , lymph node metastases . Besides influencing cell growth suggests that drugging CKS1B could be a potential avenue Kip1 and survival through regulation of p27 , silencing of of treating such CaCx subtypes harbouring episomal CKS1B is also known to induce cell death and inhibit HPV16, as opposed to those harbouring purely integrated growth of tumour cells through mechanisms that are HPV16 and this could be subject to clinical trials. 31,32 independent of p27Kip1and SKP2 . Taken together, Our previous study also demonstrated that there exists our study highlights a novel mechanism of CKS1B over- heterogeneity among integrated CaCx cases, based on expression in HPV16 episomal CaCx subtypes, which viral load and E7 expression level . The negative corre- could be attributable to loss of epigenetic regulation of lation between E7 expression and expression of CKS1B CKS1B by miR-181c, which is downregulated in such among such cases potentially suggests the complementary CaCx subtypes. Thus, this could serve as one of the role of viral and host proteins, in the pathogenesis of such mechanisms of proliferation activation in this CaCx sub- CaCx cases harbouring integrated HPV16 irrespective type that could be of translational relevance. of miR-181c (Fig. 7). Thus, our study substantiates that There exist at least two transcript variants of CKS1B CaCx cases harbouring integrated and episomal HPV16 gene and it appears that only transcript variant 1 encodes are distinct at the molecular level. the protein . In our study, we employed primers designed Besides miR-181c, our study also revealed a strong to capture the expression of this coding variant of CKS1B. positive correlation between miR-323 expression and Official journal of the Cell Death Differentiation Association Mandal et al. Cell Death Discovery (2019) 5:81 Page 10 of 13 Fig. 6 miR-323 expression among various categories of CaCx cell lines. Lower ΔC means higher expression In summary, employing both microarray-based gene HPV16 E7 expression among episomal CaCx cases. There are reports showing miR-323 upregulation in metastatic expression profiling and expression analysis of a priori- squamous cell CaCx cases and several studies on pros- tised set of miRNAs, we have been instrumental in 36,37 tate cancers , which are suggestive of the potential of establishing that CaCx cases harbouring both integrated this microRNA in imparting aggressiveness to such can- and episomal HPV16 genomes are distinct molecular cers. We also validated this through cell line based in vitro subtypes. Although the TCGA study on cervical cancers analysis as indicated in Fig. 4. This was indicative of has been instrumental in highlighting distinct molecular the fact that the viral oncogene E7 could regulate host phenotypes of HPV-active and HPV-inactive cancers, we miRNA expression, which calls for further studies on have succeeded in revealing distinct molecular subtypes of detailed molecular mechanisms. The influence of HPV16 HPV16-positive cervical cancers. Our study further oncoproteins E6 and E7 on miR-203 expression during demonstrated deregulation of miRNA expression in CaCx proliferation and differentiation has already been estab- cases compared with controls, progressive deregulation lished . Therefore, the significant positive correlation of miRNA expression with increase in severity of cervical between the expression of miR-203 and E7 in CaCx status based on histopathology and HPV infections, and subtype harbouring integrated HPV16 also merits atten- differential expression of miRNAs in CaCx cases har- tion and further exploration. bouring episomal and integrated viral genomes. There- In this study, through the correlation analysis between fore, in consideration with our previous observation of the the prioritised set of miRNAs and their target genes loss of miRNA-binding sites in viral regulatory regions (experimentally validated and predicted), we have been facilitating viral gene expressions in CaCx cases and instrumental in identifying functional miRNA-mRNA gain of miRNA binding sites in host regulatory regions relationships as detailed in Table S6. Infact, a large pro- we conclude that host cellular miRNAs have a role in portion of the non-validated target genes revealed nega- CaCx pathogenesis. Thus, our findings highlight that tive correlations with the respective miRNAs at the level CaCx cases with episomal and integrated HPV16 are of expression, suggesting that these could be further distinct entities at the molecular level and such findings analysed for exploring the regulation of miRNAs and are of immense translational value, specifically for mRNAs in CaCx pathogenesis. designing targeted therapy of CaCx cases. Official journal of the Cell Death Differentiation Association Mandal et al. Cell Death Discovery (2019) 5:81 Page 11 of 13 Fig. 7 Correlation between CKS1B and E7 expression among cervical cancers. Linear regression analysis of the correlation between CKS1B mRNA expression and E7 mRNA expression in (a) CaCx cases (episomal and integrated; p = 0.839), (b) episomal CaCx cases (p = 0.61), and (c) integrated CaCx cases (p = 0.03). Lower ΔC means higher expression Materials and methods hysterectomy attending a hospital (Calcutta Medical Samples and subjects College Hospital, Kolkata, West Bengal, India). All such The samples used for this study were nested to an samples were collected during the period of 2010–2013, ongoing natural cohort study. The malignant samples with informed consent approved by the institutional selected for the study were histopathologically confirmed ethical committee for human experimentation. Details invasive squamous cell carcinomas and clinically diag- regarding DNA isolation, HPV screening and determi- nosed as tumour stage III and above as per FIGO classi- nation of HPV16, physical status of HPV16 genomes, and 4,41–44 fication. Samples were derived from married subjects viral load are described earlier from our laboratory . attending a cancer referral hospital (Saroj Gupta Centre and Research Institute, South 24 Parganas, West Bengal, Microarray-based gene expression analysis India) collected within the period of 1998–2013. Global gene expression profilingdatareportedbySharma The non-malignant and control samples were derived et al. (GEO database, NCBI Accession No: GSE67522) was from married and non-pregnant women with no previous used following reclassification of the CaCx samples based history of cervical dysplasia/malignancy undergone for on the physical status of HPV16 genome, to identify Official journal of the Cell Death Differentiation Association Mandal et al. Cell Death Discovery (2019) 5:81 Page 12 of 13 mechanistic differences between cervical cancers harbour- pcDNA3.1-HPV16 E7 vector generated in our labora- ing episomal and integrated HPV16. Thus, differential tory . The cells were harvested and washed with 1× PBS expression analysis was performed using different categories (pH 7.4), trypsinized, and collected by centrifugation of cervical tissues, i.e., HPV-negative control samples at 300 g for 10 min The transfected cells were used (n= 11), HPV positive non-malignant samples (n= 11), further for RNA isolation. The transfection experiments episomal CaCx cases (n= 10), and integrated CaCx cases were carried out in three sets, each in triplicates and (n=10)(TableS8).The analysis pipelinewas thesameas selected miRNAs expression was carried out in selected reported earlier . cell lines according to the protocol described earlier. Expression analysis of some candidate microRNAs Acknowledgements We thank Saroj Gupta Cancer Centre and Research Institute (SGCC & RI, The expression of some prioritised miRNAs (34 Thakurpukur, South 24 Parganas, West Bengal, India) and Calcutta Medical miRNAs selected on the basis of frequent alterations College Hospital (Kolkata, West Bengal, India) for their support in sample in many cancers, identified through literature survey) collection; Dr. Samsiddhi Bhattacharjee of the National Institute of Biomedical Genomics (NIBMG), Kalyani, India. Dr. Saurabh Ghosh and Dr. Indranil were determined by TaqMan miRNA assays as well as Mukhopadhyay of Human Genetics Unit, Indian Statistical Institute, Kolkata, Power SYBR Green based assays (Supplementary India for providing guidelines for statistical analysis of the data; Department of methods 1 and 2 and Figures S7 and S8, respectively) Biotechnology, Government of India (Grant No.: BT/PR2012/MED/29/312/2011) and NIBMG, Kalyani, India for funding and NIBMG for technical support and employing 25 HPV-negative normal samples, 25 special thanks also to Council of Scientific and Industrial Research, India for HPV16-positive non-malignant samples, and 62 providing Ms. Paramita Mandal and Ms. Sweta Sharma Saha with a Fellowship (JRF and SRF) to work on this project. HPV16-positive CaCx cases (43 episomal and 19 inte- grated HPV16 genomes). The various categories of Author details miRNAs analysed, and the assays employed are provided National Institute of Biomedical Genomics, Kalyani, West Bengal, India. in Supplementary methods 1–3and Tables S4,Table S9, Crystallography & Molecular Biology Division, Saha Institute of Nuclear Physics, 1/AF Bidhannagar, Kolkata 700064, India. Department of Gynecology, Saroj and Table S10. Gupta Cancer Centre and Research Institute, Kolkata, India. Sri Aurobindo Seva Kendra, 1H, Gariahat Road (S) Jodhpur Park, Kolkata 700068 West Bengal, India. Target prediction of miRNA Health Science Library System, University of Pittsburgh, Pittsburgh, PA, USA. Present address: Department of Zoology, The University of Burdwan, The target human genes corresponding to a subset of Burdwan, West Bengal, India. Present address: Section of Haematology/ the miRNAs analysed in the study, were predicted based Oncology, Department of Medicine, university of Chicago, 5841 S Maryland on three prediction softwares like miRanda, PICTAR, Ave MC 2115, Chicago, IL 60637, USA. Present address: Molecular Biology and Genetics Unit, Jawaharlal Nehru Centre for Advanced Scientific Research, and TargetScan, considering validated miRNA binding Jakkur, Bengaluru, Karnataka 560064, India. Present address: Comprehensive sites as well. Wound Center, Center for Regenerative Medicine and Cell Based Therapies, The Ohio State University, Columbus, OH, USA Validation of miRNA target gene CKS1B by quantitative Conflict of interest real-time PCR The authors declare that they have no conflict of interest. Validation was done considering only CKS1B gene that appeared as target of the significantly altered miR-181c Publisher’s note recorded among episomal CaCx cases only. This gene was Springer Nature remains neutral with regard to jurisdictional claims in selected based on the fact it belonged to the significantly published maps and institutional affiliations. enriched pathway corresponding to the CaCx cases. Samples used for validation by quantitative real-time The online version of this article (https://doi.org/10.1038/s41420-019-0154-x) PCR included episomal CaCx cases (n = 43), integrated contains supplementary material, which is available to authorised users. CaCx cases (n = 19), HPV-negative controls (n = 25), and HPV16-positive non-malignant samples (n = 25). The Received: 6 December 2018 Revised: 3 February 2019 Accepted: 13 February 2019 details of primer sequences of CKS1B and GAPDH, amplicon size and PCR conditions employed is depicted in Table S10. References Cell culture and transfection 1. Das, D., Bhattacharjee, B.,Sen,S., Mukhopadhyay,I.& Sengupta,S.Association of viral load with HPV16 positive cervical cancer pathogenesis: Causal HPV-negative cell line C33A and HPV16-positive cell relevance in isolates harboring intact viral E2 gene. Virology 402,197–202 lines Caski and SiHa were cultured in Dulbecco's (2010). Modified Eagle Medium supplemented with 10% fetal 2. Thomas, M., Massimi, P., Navarro, C., Borg, J. P. & Banks, L. The hScrib/Dlg apico- basal control complex is differentially targeted by HPV-16 and HPV-18 E6 bovine serum, 50 Units/ml of penicillin and 50 µg/ml proteins. Oncogene 24, 6222–6230 (2005). of streptomycin at 37 °C and 5% CO . C33A cells were 3. Romanczuk, H. & Howley, P. M. Disruption of either the E1 or the E2 regulatory transfected using Lipofectamine 2000 reagent according gene of human papillomavirus type 16 increases viral immortalization capa- to manufacturer’s protocol, using 1 µg of plasmid city. Proc. Natl. Acad. Sci. USA 89,3159–3163 (1992). Official journal of the Cell Death Differentiation Association Mandal et al. Cell Death Discovery (2019) 5:81 Page 13 of 13 4. Bhattacharjee, B. & Sengupta, S. CpG methylation of HPV 16 LCR at E2 binding 27. Liu, N. et al. Upregulation of microRNA-200a associates with tumor prolifera- site proximal to P97 is associated with cervical cancer in presence of intact E2. tion, CSCs phenotype and chemosensitivity in ovarian cancer. Neoplasma 62, Virology 354, 280–285 (2006). 550–568 (2015). 5. Umbach, J. L. et al. MiRNAs expressed by herpes simplex virus1 during latent 28. Boominathan, L. The tumor suppressors p53, p63, and p73 are regulators infection regulate viral mRNAs. Nature 454, 780–783 (2008). of MicroRNA processing complex. PLoS. ONE 5, e10615 (2010). 6. Skalsky, R. L. & Cullen, B. R. Viruses, microRNAs, and Host Interactions. Annu. 29. Wang, W. T. et al. Differentially expressed microRNAs in the serum of cervical Rev. Microbiol. 64, 123–141 (2010). squamous cell carcinoma patients before and after surgery. J. Hematol. Oncol. 7. Guo, Y. E. & Steitz, J. A. Virus meets host MicroRNA: the destroyer, the booster, 7, 6 (2014). the hijacker. Mol. Cell Biol. 34, 3780–3787 (2014). 30. Zhang, Y. CKS1B (CDC28 protein kinase regulatory subunit 1B). Atlas Genet. 8. Calin, G. A. et al. Human microRNA genes are frequently located at fragile sites Cytogenet. Oncol. Haematol. 14,676–678 (2010). and genomic regions involved in cancers. Proc. Natl. Acad. Sci. USA 101, 31. Liberal, V. et al. Cyclin-dependent kinase subunit (Cks) 1 or Cks2 over- 2999–3004 (2004). expression overrides the DNA damage response barrier triggered by activated 9. Farazi, T.A., Hoell, J. I.,Morozov,P.&Tuschl, T.microRNAs in humancancer. oncoproteins. Proc.Natl. Acad.Sci.USA 109,2754–2759 (2012). Adv. Exp. Med. Biol. 774,1–20 (2013). 32. Lee, S. W. et al. Overexpression of CDC28 protein kinase regulatory subunit1B 10. Calin, G. A. & Croce, C. M. MicroRNA-cancer connection: the beginning of a confers an independent prognostic factor in nasopharyngeal carcinoma. new tale. Cancer Res. 66,7390–7394 (2006). APMIS 122, 206–214 (2014). 11. Sharma, G., Dua, P. & Agarwal, S. M. A comprehensive review of dysregulated 33. Fagerberg, L. et al. Analysis of the human tissue-specificexpressionby miRNAs involved in cervical cancer. Curr. Genomics 15, 310–323 (2014). genome-wide integration of transcriptomics and antibody-based proteomics. 12. Jiménez-Wences, H. et al. Methylation and expression of miRNAs in pre- Mol. Cell Proteom. 13,397–406 (2014). cancerous lesions and cervical cancer with HPV16 infection. Oncol. Rep. 35, 34. Chayka, O., D'Acunto, C. W., Middleton, O., Arab, M. & Sala, A. Identification 2297–2305 (2016). and pharmacological inactivation of the MYCN gene network as a 13. Das Ghosh, D. et al. Some novel insights on HPV16 related cervical cancer therapeutic strategy for neuroblastic tumor cells. J. Biol. Chem. 290, pathogenesis based on analyses of LCRmethylation, viral load, E7 and E2/E4 2198–2212 (2015). expressions. PLoS. ONE 7, e44678 (2012). 35. Ding, H.,Wu, W. L.,Wang, Y. X. &Zhu,F.F. Characterizationofthe microRNA 14. McKenna,D.J., McDade,S.S., Patel, D. &McCance,D.J.MicroRNA203 expression profile of cervical squamous cell carcinoma metastases. Asian Pac. expression in keratinocytes is dependent on regulation of p53 levels by E6. J. J. Cancer Prev. 15, 1675–1679 (2014). Virol. 84, 10644–10652 (2010). 36. Gao, Q., Yao, X. & Zheng, J. MiR-323 inhibits prostate cancer 15. Sharma, S. et al. Bridging links between long noncoding RNA HOTAIR and vascularization through adiponectin receptor. Cell Physiol. Biochem. 36, HPV oncoprotein E7 in cervical cancer pathogenesis. Sci. Rep. 5, 11724 (2015). 1491–1498 (2015). 16. Bartel, D. P. MicroRNAs: target recognition and regulatory functions. Cell 136, 37. Gao, Q. & Zheng, J. microRNA-323 upregulation promotes prostate cancer 215–233 (2009). growth and docetaxel resistance by repressing p73. Biomed. Pharmacother. 97, 17. Ahn, W. S. et al. Searching for pathogenic functions to cervical cancer. Gynecol. 528–534 (2018). Oncol. 93,41–48 (2004). 38. Banister, C. E., Liu, C., Pirisi, L., Creek, K. E. & Buckhaults, P. J. Identification 18. Medina-Martinez, I. et al. Impact of gene dosage on gene expression, biolo- and characterization of HPV-independent cervical cancers. Oncotarget 8, gicalprocesses andsurvivalincervicalcancer: agenome-wide follow-up 13375–13386 (2017). study. PLoS. ONE 9, e97842 (2014). 39. Mandal, P. et al. Differential expression of HPV16 L2 gene in cervical cancers 19. Lee, H. S. et al. Identification of differentially-expressed genes by DNA harboring episomal HPV16 genomes: influence of synonymous and non- methylation in cervical cancer. Oncol. Lett. 9,1691–1698 (2015). coding region variations. PLoS. ONE 8, e65647 (2013). 20. Takai,N., Hamanaka,R., Yoshimatsu,J.&Miyakawa, I.Polo-like kinases(Plks) 40. Sharma Saha, S. et al. Identification of genetic variation in the lncRNA HOTAIR and cancer. Oncogene 24,287–291 (2005). associated with HPV16-related cervical cancer pathogenesis. Cell. Oncol. 39, 21. Kauffmann, A. et al. High expression of DNA repair pathways is associated with 559–572 (2016). metastasis in melanoma patients. Oncogene 27, 565–573 (2008). 41. Bhattacharjee, B. & Sengupta, S. HPV16 E2 gene disruption and polymorph- 22. Chen, R. H. & Tian, Y. J. Enhanced anti-tumor efficacy of aspirin combined with isms of E2 and LCR: some significant associations with cervical cancer in triptolide in cervical cancer cells. AsianPac.J.Cancer Prev. 14, 3041–3044 Indian women. Gynecol. Oncol. 100,372–378 (2006). (2013). 42. Bhattacharya, P. & Sengupta, S. Predisposition to HPV16/18-related cervical 23. Ujhelyi, Z. et al. The enhanced inhibitory effect of different antitumor agents in cancer because of proline homozygosity at codon 72 of p53 among Indian self-microemulsifying drug delivery systems on human cervical cancer HeLa womenisinfluenced by HLA-B*07 and homozygosity of HLA-DQB1*03. Tissue cells. Molecules 20, 13226–13239 (2015). Antigens 70,283–293 (2007). 24. Vojtechova, Z. et al. Comparison of the miRNA profiles in HPV-positive and 43. Laikangbam, P. et al. A comparative profile of the prevalence and age dis- HPV-negative tonsillar tumors and a model system of human keratinocyte tribution of human papillomavirus type 16/18 infections among three states clones. BMC Cancer 16, 382 (2016). of India with focus on northeast India. Int. J. Gynecol. Cancer 17,107–117 25. Cancer Genome Atlas Research Network. Integrated genomic and molecular (2007). characterization of cervical cancer. Nature 543,378–384 (2017). 44. Bhattacharjee, B., Mandal, N. R., Roy, S. & Sengupta, S. Characterisation 26. Li, A. et al. Pancreatic cancers epigenetically silence SIP1 and hypomethylate of sequence variations within HPV16 isolates among Indian women: and overexpress miR-200a/200b in association with elevated circulating prediction of causal role of rare non-synonymous variations within intact miR-200a and miR-200b levels. Cancer Res. 70,5226–5237 (2010). isolates in cervical cancer pathogenesis. Virology 377,143–150 (2008). Official journal of the Cell Death Differentiation Association

Journal

Cell Death Discovery – Springer Journals

Published: Mar 25, 2019

There are no references for this article.