| Summary: | Thesis (Ph.D.)--Memorial University of Newfoundland, 1990. Medicine Bibliography: leaves 288-326 The Regulation of eucaryotic gene expression and cellular transformation have been a central focus of research in the last two decades. Studies on DNA tumor viruses have provided information not only on viral pathogenicity but also on general eucaryotic gene regulation, as they depend on host cell machinery for their transcription. I have examined some aspects of transcriptional control in six DNA tumor viruses belonging to the papc’’aviridae family and the results are discussed in Section 1. Additionally, I have studied cellular transformation induced by one of these viruses the results of which are presented in Section 2. -- The cis-acting regulatory elements of a gene required for accurate, efficient and cell type specific expression are classified into promoters and enhancers. To study the role of enhancers in cell type specific expression, I have performed comparative analysis of the 3V4 0, BK and JC enhancers in differentiated and undifferentiated embryonal carcinoma (EC) cells. In transient transfection assays, transcription of a reporter gene was not activated by any of the viral regulatory elements in undifferentiated EC cells. However, transcriptional activation of the reporter gene was observed in retinoic acid (RA) differentiated neuronal cell types by all three regulatory elements. Moreover, SV40 and BK but not JC regulatory elements demonstrated activity in dimethylsulfoxide (DMSO)-differentiated muscle cell types. To correlate in vivo activity with the binding of transcription factors, I performed DNasel footprinting experiments. For SV4 0, the region containing the transcription factor Spl binding motif was protected in undifferentiated and differentiated cell types. However, in both RA- and DMSO-differentiated cells, additional protection at regions corresponding to P, SphI and Sphll motifs was detected. The JCV enhancer demonstrated three retinoic acid differentiated cell type specific footprints, each containing sequences with homology to the nuclear factor-1 (NF-1) binding motif. With the BKV enhancer, a GC rich region was protected in all three cell types. Additional protection in four regions, each containing sequences with homology to the NF-1 motif, was observed in the two differentiated cell types. Further experiments suggest that the factor(s) interacting with NF-1 motif containing regions are different in the two differentiating cell types. Overall, the results demonstrate clear correlation of in vivo activity of the regulatory elements of the three viruses with in vitro DNA-protein interaction. -- To analyze epitheliotropism of human papillomavirus types 11, 16 and 18, I performed similar studies in four epithelial cervical carcinoma (C33A, HeLa, SiHa and CaSki) and one fibroblast (143B) cell lines. All three viral enhancers demonstrated varying degrees of activity in C33A, HeLa and SiHa but not in CaSki and 143B. By DNasel footprinting, I have identified seven, nine and five footprints on the HPV 11 , 16 and 18 enhancers, respectively. NF-1 motifs were present in five, six and one protected regions of the HPV 11, 16 and 18 enhancers, respectively. Sequences homologous to LVc, and 0AP3 (HPV 11); API (HPV 16 and HPV 18) and EFII (HPV 18) were also observed in a few protected regions. In vitro transcription- oligonucleotide competition and deletion analyses the of HPV 11 enhancer revealed that one of the NF-1 motif containing regions acts as a negative control element. One sequence motif of HPV 16, for which UV cross-linking studies revealed interaction with four protein molecules, is a strong modulator of HPV 16 enhancer function in vivo and shares 100% homology to a sequence motif, GTTTTAA in the tissue-specific enhancer of the c-mos oncogene. One sequence motif of the HPV 18 enhancer has three repeats of a TTTTA sequence contained within the c-mos sequence motif and interacts with at least four different polypeptides, as judged by UV cross-linking experiments. -- Previous studies have indicated that tumor (T)-antigens of BK and SV40, although structurally related, possess different transformation potentials. To understand these differences, I have examined the role of BKV tumor antigens in the maintenance of transformation and have identified the domain of the T-antigen gene essential for the transformed phenotype. BKV DNA-transformed BHK 21 and NIH 3T3 cells expressing antisense BK T-antigen RNA lose their ability to grow in soft agar, indicating the need for continued expression of T-antigen for the maintenance of the transformed phenotype. Experiments using translation termination linker insertion and deletion mutagenesis of BKV T-antigen demonstrate that amino acids 356 to 384 are essential for transformation. Although BKV T- antigen shares 100, 85, and 80% amino acid homology with the SV40 T-antigen for the nuclear localization signal, DNA-binding domain and p53-binding domain, respectively, the transformation domains of BKV and SV40 T-antigens share only 54% homology. In addition, the BKV T-antigen lacks a substantial portion of the ATPase domain present in the SV40 T-antigen. Further results indicate the dispensability of the remaining portion for transformation by this protein. I suggest that the differences in the amino acids in the identified transformation domains together with the differences in the ATPase domains may account for the differences in the transformation potentials of the two proteins
|
|---|