Hypoxia (low oxygen tension) is tightly associated with tumorigenesis and angiogenesis, and is also involved in cerebral and myocardial ischemia. Solid tumors contain hypoxic regions that often acquire diminished apoptotic potential and resistance to chemo- and radio-therapy. Cells respond to hypoxia, at least in part, by the transcriptional up-regulation of a variety of genes contributing to oxygen homeostasis, such as vascular endothelial growth factor, inducible nitric oxide synthase, glucose transporters, and glycolytic enzymes. Remarkably, the activation of all these genes depends primarily upon the hypoxia-inducible factor 1 (HIF-1), an ab heterodimeric transcription factor of the basic helix-loop-helix PAS family. Our research program comprises investigating the molecular mechanisms underlying HIF-1 activation and, in turn, providing molecular bases for drug targeting of tumor growth and progression. Previously, we demonstrated that HIF-1 activation is regulated primarily by its a subunit (HIF-1a), involving protein stabilization and transcriptional activation. The abundance of HIF-1a is regulated by oxygen-dependent proteolysis through the ubiquitin-proteasomal pathway, which targets the oxygen-dependent degradation domain (ODD) of HIF-1a. Recently, the von Hippel-Lindau (VHL) protein (a tumor suppressor gene) has been identified as a HIF-1a E3 ubiquitin ligase. The VHL protein targets the ODD by independent recognition of two hydroxylated prolines (Pro402 and Pro564). To gain further insights, we are investigating both the VHL-dependent and -independent degradation pathways that control HIF-1a protein abundance. On the other hand, hypoxia activates HIF-1a by stimulating HIF-1a transcriptional activity, which is mediated by the recruitment of a critical co-activator, p300/CBP. Under hypoxia, the C-terminal activation domain (CAD) of HIF-1a interacts with the CH1 domain of p300/CBP, whereas under normoxia this interaction is prevented due to the hydroxylation of Asn803 in the CAD. To systematically understand the molecular basis of this interaction, we developed a random mutagenesis screen in yeast (RAMSY) approach for efficient identification of residues that are functionally critical for protein interactions. As a result, residues involved in the HIF1a-p300 interaction were successfully identified and validated to be crucial for HIF-1 transcriptional activity in mammalian systems. Based on the extensive amino acid substitutions, we proposed that hypoxia-induced HIF1a-p300 interaction relies on a leucine-rich hydrophobic interface that is controlled by HIF-1a Cys800. Interestingly, all of the identified residues have been confirmed independently by the co-crystallization of CAD and CH1domains, providing an ultimate proof for the accuracy and reliability of the RAMSY method. Therefore, we are continuing to take advantage of this approach for further structure-function studies. To investigate the role of HIF-1 in angiogenesis and tumorigenesis, we collaborated with the Jeffrey Arbeit Laboratory at the University of California San Francisco. We showed that only a constitutively active HIF-1a transgene, but not the wild-type, resulted in hypervascularity in the mouse. The induced vasculatures were mature, and free of leak and inflammation, indicating the therapeutic potential of HIF-1a. We are also asking how HIF-1a is involved in cell proliferation and apoptosis by employing different gene transfer techniques.
