Inhibition and Cofactor Targeting of Hypoxia-Sensing Proteins

Hypoxia Inducible Factor (HIF) is a transcription activator considered to be the main regulator of O2 homeostasis in humans. The transcriptional ability of HIF is regulated by the Fe2+/αKG-dependent enzyme, Factor Inhibiting HIF (FIH). FIH uses molecular oxygen to catalyze hydroxylation of an asparagine residue (Asn803) in the C-terminal transactivation domain (CTAD) of the HIFα subunit, abrogating HIF target gene expression. The mechanism of FIH and other αKG-dependent oxygenases involves the ordered sequential binding of αKG, substrate, and O2, which becomes activated to form a reactive ferryl intermediate that hydroxylates the substrate. The key to understanding FIH’s function as an oxygen sensor, and the chemistry of the broader class of enzymes, is to elucidate steps coincident with O2 binding and activation. These steps are shared amongst this class of enzymes and are the least understood. In order to understand structural factors that influence O2 binding and reactivity, we solved the crystal structure of FIH in the presence of NO, an O2 mimic. Our data suggests that the sterics of the target asparagine residue modestly influences O2 binding, but plays a major role in stimulating O2 reactivity by orienting bound gas productively for decarboxylation chemistry. The increased O2 reactivity in the presence of CTAD would ensure tight coupling of O2 binding and activation to hydroxylation. The reorientation of bound O2 may be the mechanism utilized by the broader class of αKG oxygenases to activate molecular oxygen. The active site of the Fe2+/αKG-dependent enzymes is highly conserved. The Fe cofactor is typically ligated by 2 His residues and 1 carboxylate residue. A class of αKG-dependent enzymes that catalyze halogenation chemistry ligate the Fe cofactor with only 2 His residues, and bind a halide ion in place of the carboxylate ligand. In order to understand the structural factors that control the incorporation of alternative ligands such as halides, we mutated Asp201 in FIH to Glu (D201E), and Gly (D201G). The D201G variant activity was dependent on chloride, and the crystal structure suggested that chloride was bound at the active site. XAS data indicated that chloride was bound, implicating D201 in controlling ligand access. Other non-native ligands such as azide, and cyanate were able to access the active site Fe cofactor of D201G, consistent with relaxed ligand selectivity upon mutation of the facial triad carboxylate. The identification of exogenous ligands targeting the Fe cofactor raises the possibility that FIH can be engineered for alternative rebound chemistry. Recently, there have been many reports of “gasotransmitters” such as NO and H2S impacting HIF controlled gene expression in cells, suggesting that the HIF hydroxylases are potential physiological targets. We tested the effect of H2S on the activity of FIH. Our data suggests that H2S binds to the Fe cofactor of FIH to inhibit activity. H2S was also demonstrated to reduce inactive ferric FIH back to the ferrous state, restoring activity. The dual regulation of FIH activity provides a rationale for some of the observed physiological responses to H2S treatment. Taken together, our data suggests that the active site of FIH relies on several structural motifs to direct chemistry. The active site carboxylate controls the access of O2 and other ligands to the Fe cofactor. The target asparagine residue stimulates O2 reactivity by provided the proper steric environment to orient O2 productively for chemistry. However, certain ligands such as H2S are able to access the Fe cofactor to inhibit chemistry, implicating them in the hypoxia sensing pathway in cells