Studies into the receptor-mediated induction of hypoxia inducible factor 1α.

The transcription factor Hypoxia Inducible Factor (HIF) is pivotal to the cellular response upon exposure to low oxygen stress (hypoxia). Post-translational modification of the HIF a-subunit (HIFa) by hydroxylase enzymes has recently been identified as the key 'oxygen sensing' mechanism within the cell. Regulation of HIF is complex; during hypoxia, regulation primarily occurs through inhibition of the hydroxylase enzyrnes and facilitates dramatic up-regulation of the protein stability and transcriptional activation capabilities of the HIF a-subunits. HIF also appears to play a role in target gene transcription in a hypoxia-independent manner. This novel, hypoxia-independent function of HIF is poorly characterised but has been recently shown to involve HIFa induction by receptor-mediated factors and the phosphatidyl-inositol-3-kinase/Akt or mitogen activated protein kinase (MAPK) signalling cascades. Activation of HIFa through these pathways is more subtle than that which occurs during induction by hypoxia. Activation of Akt and MAPK is reported to increase HIFa protein synthesis and results in increased HIFa protein levels and transcriptional activity. In addition, in some cell types MAPK potentiates the transcriptional activity of HIFa via a separate mechanism that does not require protein stabilization, but instead effects the recruitment of transcriptional co-factors. Adding a further level of complexity to the mechanisms for HIFa regulation is the existence of alternatively spliced isoforms for the three characterised HIFa genes. Recent work has identified tissue-specific expression of HIF1a isoforms that have diminished or absent transcriptional activation capabilities, suggesting that the overall transcription of target genes by HIFla may be modulated by interactions between these isoforms. This thesis focuses on the role of growth factor signalling to HIFIa and the potential influence of insulin resistance and Type 2 Diabetes on this signalling. It aims to test the hypothesis that HIFla response to growth factor stimulus is downgraded in insulin resistance and hence Type 2 Diabetes. In order to address this, the development and validation of model cellular systems and the generation of antibody tools were performed to better characterise components of the pathway involved in hypoxia-independent, receptor-mediated activation of HIFla. Chapter 3 describes the generation of in vitro reporter gene models in growth factor responsive adipocyte (3T3Ll) and liver (HepG2) cell lines. Western analysis of HIFla, protein expression and stabilisation and parallel assessment of HIFla transcriptional activity during reporter-gene studies showed that in 3T3L1 adipocyte and HepG2 cell types stimulation by insulin, insulin-like growth factor-l or -2 (IGFI or 2) up-regulated HRE-mediated transcription, though not through increased HIFla protein stabilisation. It is possible that this occurred due to increased recruitment of transcription co-factors such as p300/CBP. Results presented in this thesis, which characterise ligand-activated auto-phosphorylation of insulin and IGF1 receptors (IR and IGFRI) clearly indicated that these cells possess functional insulin and IGF receptor signalling pathways. However, as non-physiological (100 nM) concentrations of insulin were required to achieve an induction of the HRE-mediated luciferase reporter gene, it is likely that activation of HIFIa in these cells is mediated instead through the IGFR1 and not the IR. This suggests complexity of control of receptor-mediated HIFla function, since the presence of functional insulin receptors was not sufficient to induce a successful outcome for HRE-mediated gene transcription. In order to test the hypothesis, an in vitro insulin resistant phenotype, to mimic that observed in vivo during Type 2 Diabetes, was successfully generated and validated within these model cell lines (discussed within Chapter 3). Development of this phenotype enabled investigation into the effect of defective growth factor signalling conditions upon HIFla. This was of interest because many of the genes with abnormal expression patterns during Type 2 Diabetes are also regulated by HIF. It was proposed that the absence of normal receptor-mediated HIFIa activation during the progression of the disease might result in the complications that affect chronic sufferers. However insulin resistance did not alter the induction of HIFla protein or transcriptional activity by hypoxic oxygen exposure or serum, insulin and IGFI treatment. As shown in the data presented in Chapter 3, the transcriptional activation of HRE-mediated luciferase expression during insulin resistance was found to be equal to that obtained from normal cells. Thus the hypothesis was disproved. This is likely to be due to the finding that IGF1 activation of the IR and IGFR1 was not impeded during insulin resistance. In support of these in vitro findings, analyses of protein stability were performed by Western blot of lysates from human adipose tissue (Chapter 5). Similar to the data obtained from the 3T3Ll adipocyte cell line, the amount of HIFIa protein expressed and subsequently stabilised was not altered in vivo between Type 2 Diabetic and non-diabetic human adipose samples. However, during these studies a putative HIF1cr isoform was identified on the basis of recognition by several anti-HlFla antibodies. This was important because, as mentioned above, expression of alternatively spliced isoforms of HIFa subunits are believed to intimately affect the overall HIF transcriptional output within specific tissue types. An extensive process of Western analysis, immuno-precipitation, and size-exclusion chromatography achieved isolation of the putative HIFIa isoform, and the protein was identified by mass-spectrophotometry as human albumin, a protein abundantly found in adipose. Given that other primary and secondary antibodies did not recognise the albumin protein during Western analysis, it is likely that the anti-HIFla antibodies possess cross-reactivity to albumin at high protein concentrations. It became clear during the analyses of HIFla protein expression performed in Chapters 3 and 5, that the available HIFla antibodies did not possess the sensitivity of detection required to monitor the low levels of HIFla protein stabilised by hypoxia-independent, receptor-mediated mechanisms. Additionally the availability of a high-affinity monoclonal would contribute to HIFIa research by enabling generation of affinity chromatography columns for purification of post-translationally modified HIFla or HIFla isoforms, so that they can be characterised by mass spectrophotometry. To address this deficit, generation of high-affinity monoclonal HIFIa antibodies was performed and these results are presented in Chapter 4. 1642 hybndoma clones were screened and 20 were identified as expressing antibody with specificity and affinity for the HIF1cr epitope. Of these, 5 hybridoma clones were selected for further study and shown to secrete antibody of the IgM isotype. These HIFla monoclonal antibodies had weak affinity to HIFIa and were difficult to purify due their IgM isotype properties. Previous studies of receptor-mediated induction of HIFIa indicated that up-regulation of protein stability and transcriptional activity of HIFIa occurred through increased HIFla protein synthesis after insulin or IGF1 treatment of human colon carcinoma or retinal epithelial cells. However, the findings presented here suggest that not all cell types with functional growth factor receptor signalling machinery are capable of activating HIFla through these pathways. In HepG2 and 3T3L1 adipocyte cells, activation of HIFIa regulated gene transcription appears to occur in the absence of increased protein stabilisation, possibly by increased co-factor recruitment. In these cell types, insulin and IGF1 up-regulate HIF1a via IGF1 receptors and insulin resistance did not impede IGFR1 signalling. Furthermore, insulin resistance in vitro in cell lines or in vivo in human adipose tissue did not influence HIFla protein levels.Thesis (Ph.D.) -- University of Adelaide, School of Molecular Bioscience, 200