| The functional involvement of histone post-translational modifications in gene expression is well documented. As a major regulatory component, chromatin modification status along the hGH/CS locus was previously assessed in either human GH-secreting pituitary adenoma samples or transgenic mouse pituitaries with hyper-stimulation via GH releasing hormone. It was hypothesized that hyper-stimulation would affect the post-translational modification of histones along the GH/CS locus. In this thesis, histone modifications along the hGH/CS locus, including histone H4 hyperacetylation as well as H3 lysine (K) 4 methylation, were assessed for the first time in human pituitary tissues taken post mortem. Using the chromatin immunoprecipitation (ChIP) assay to assess protein/DNA interactions in situ, the LCR appears to be widely modified by both histone H4 hyperacetylation and H3K4 methylation, indicating their functional involvement in regulating pituitary gene expression. These modifications are also detected at the GH-N but not the CS promoter region, which is consistent with the distinct expression pattern for GH-N versus CS genes in the pituitary.Despite the extensive sequence homology and the presence of Pit-1 DNA elements in the promoter regions, the placental CS/GH-V genes and GH-N are expressed in a mutually exclusive way in the pituitary, to the extent that not even a basal level of placental hormone expression is detected. An active mechanism of repression was hypothesized to explain the lack of placental CS/GH-V expression in the pituitary, based on the observations: (i) all five hGH/CS genes share a common structure in pituitary chromatin, and (ii) the CS-A promoter is active and responsive to Pit-1 binding in transfected rat pituitary GC cells. A conserved P sequence, located upstream of each placental expressed gene but absent from the GH-N gene, was considered an excellent candidate to be involved in pituitary repression of the placental CS/GH-V genes. The repressor activity of P sequences (263P) was suggested by the ability to repress CS-A gene promoter activity in transfected rat pituitary GC cells. Two sub-fragments of 263P, PSE-A and PSE-B, were identified as protein-binding regions with repressor activity. However, when a 103P fragment, containing complete PSE-A and PSE-B sequences, was assessed for repressor activity in transfected rat pituitary GC cells, it was unable to repress the CS-A promoter activity. Thus, it was hypothesized that sequence information outside of PSE-A and PSE-B but contained within the remaining 263P was required for pituitary repression. An additional P sequence element that contributes to the repressor activity of 263P in transfected rat pituitary GC cells, designated as PSE-C, was identified downstream of PSE-A. The hepatocyte nuclear factor (HNF)-3alpha was identified as the PSE-C associated factor, and shown to participate in the repressor complex formed on 263P. In addition, the functional involvement of Pit-1 in the pituitary-repression activity of P sequences was also explored. A possible mechanism is proposed for pituitary repression with HNF-3 acting as a "bridge" stabilizing the repressor complex on 263P, and interfering with Pit-1 function. Alternatively, changes in chromatin structure during the activation process of GH-N promoter through Pit-1 binding to LCR, as a consequence, may also interfere with necessary modifications required for expression on the CS promoters, and eventually, resulting in the lack of gene expression. (Abstract shortened by UMI.)Pituitary-specific expression of GH-N is largely under the control of a remote LCR encompassing five hypersensitive sites (HS I/II-HS V). Binding of the pituitary-specific transcription factor Pit-1 to HS I/II is required for efficient pituitary-specific GH-N expression in vivo. However, the process by which Pit-1 gains access to HS I/II remains unclear. It was hypothesized that Pit-1 could induce chromatin remodeling in the LCR, resulting in activation of HS I/II and eventually, contribute to GH-N gene activation. Using human embryonic kidney (HEK) 293 cells overexpressing wild type and mutant Pit-1 proteins as a model system, the consequences after Pit-1 appearance in these cells were assessed. Addition of Pit-1 resulted in increased DNA accessibility at HS III in a POU-homeodomain dependent manner, as reflected by effects on histone H4 hyperacetylation and increased RNA polymerase II activity in the nearby region. Assessment of protein/DNA interactions in vitro revealed that direct Pit-1 binding to HS III is not supported by EMSA analysis. However, Pit-1 associates with HS III through indirect interaction via Elk-1, a member of the ETS family of transcription factors, as suggested through co-immunoprecipitation studies using human pituitary proteins. Furthermore, Elk-1 was identified as a co-factor which partners with Pit-1 to stimulate endogenous GH-N RNA levels in the HEK293 cells. |
PDF Full Text Request