The cellular splicing regulator SRSF2 controls HPV16 E6 mRNA stability and contributes to the cervical tumour phenotype.
PhD thesis, University of Glasgow.
Full text available as:
HPV16 is a double stranded DNA virus which has a close association with cervical cancer development. HPV16 expresses two oncoproteins E6 and E7, which when overexpressed transform the virus-infected epithelial cells. E6 and E7-encoding RNAs have been shown to be alternatively spliced to give at least four mRNA isoforms (E6 full length, E6*I, E6*II, E6*X) however it is not known what functions any putative alternative oncoproteins or indeed, RNA isoforms may have. Alternative splicing is regulated by host cell factors, namely SR proteins and hnRNP proteins. SR proteins are the positive splicing regulators in the cell and generally promote splicing of both cellular and viral RNAs. It is not known which RNA splicing factors are required for E6/E7 RNA processing during infection and transformation. HPV16 viral RNAs are extensively alternatively spliced therefore identification of the host factors involved in the processing of viral RNAs could have therapeutic benefits because completion of the virus life cycle requires alternative splicing and if this could be prevented by targeting of the splicing factors involved, then the virus infection could be prevented. Similarly if splicing is altered upon transformation of the cervical epithelial cells, then prevention of this alteration in splicing could perhaps inhibit virus-induced transformation of the cervical epithelial cells. Recently it has been shown that overexpression of SR proteins can lead to cellular transformation and certain SR proteins have already been reported to be upregulated in some cancers. Therefore the focus of this PhD was to investigate the expression of the alternative E6/E7 RNA isoforms and identify the protein (s) responsible for their processing.
The first aim was to investigate the expression of the E6/E7 RNA isoforms in virus-infected cells and during cervical epithelial cell transformation and to try to assign any function to the individual isoforms. E6 and E7 are encoded by the viral early genes but are expressed in both undifferentiated and differentiated virus infected cells. Therefore I investigated whether the expression of the isoforms was altered during differentiation. RT-PCR experiments were carried out using RNA extracted from undifferentiated and differentiated W12E cells (HPV16-infected cervical epithelial cells). Results indicated that E6/E7 splicing is not altered upon epithelial differentiation however E6 and E7 mRNA abundance increased upon differentiation. Furthermore, experiments carried out in W12G cells, where HPV16 genomes are integrated into the host genome and no viral life cycle is taking place, suggested that the increase in E6 and E7 RNA expression was not due to cell differentiation, but due to a virus-induced increase in expression. Further RT-PCR experiments in HPV16 transformed cells lines demonstrated that E6/E7 RNA isoform expression is altered as cervical epithelial cells become tumourigenic. Small E6 isoforms, E6*II and E6*X are upregulated in virus transformed cells suggesting a tumour promoting function for the isoforms. To test this, E6/E7 isoform-expressing constructs were created and transiently transfected into HPV-negative C33a cervical cancer cells and the proliferation rate and ability to form colonies in soft agar investigated. Compared to the longer E6/E7 isoforms the two smallest isoforms promoted cellular proliferation as the cell growth rate increased. However anchorage independent growth assays were inconclusive suggesting there may be a combinatorial effect of the E6/E7 isoforms on transformation of the cells.
My next aim was to investigate the expression of SR proteins during transformation of HPV16-infected epithelial cells. Western blot and immunohistochemical analysis showed that SR proteins SRSF1-3 were specifically upregulated upon cervical epithelial transformation. For SRSF2 and 3, this was not due to gene amplification as qPCR analysis of gene copy number showed no significant difference in CT values between the W12 cell lines suggesting that upregulation of SRSF2 and 3 may be at a transcriptional level. However there was a significant difference in SRSF1 gene copy number that may account for its upregulation.
My final aim was to identify the SR protein (s) responsible for the alteration in E6/E7 isoform expression in HPV16-transformed cells. This was achieved by siRNA depletion of the overexpressed SR proteins and RT-PCR of E6/E7 RNA. Surprisingly, none of the SR protein knockdowns resulted in any detectable alteration of RNA isoforms. However, SRSF2 knockdown specifically resulted in a significant reduction in all E6/E7 encoding RNAs. Moreover, after SRSF2 knockdown, p53 levels were increased suggesting an impairment of E6 protein function. The reduction in E6/E7 RNA was not due to a decrease in transcription as demonstrated by transcription assays utilising an HPV16 LCR luciferase reporter. Interestingly, E6/E7 RNA stability assays showed that RNA half life is reduced when SRSF2 is knocked down. SRSF1 and 2 have previously been shown to be oncogenic in breast and ovarian cancers respectively. So the effects of SRSF2 knockdown on cell growth rate, colony formation, apoptosis entry and cell cycle were analysed in transformed cervical epithelial cells. The results from these experiments indicated that overexpression of SRSF2 in cervical epithelial cells is tumour promoting. My data clearly indicates that SRSF2 should be considered to be a proto-oncogene.
Actions (login required)