A yeast three-hybrid system that reconstitutes mammalian hypoxia inducible factor regulatory machinery

Background: Several human pathologies, including neoplasia and ischemic cardiovascular diseases, course with an unbalance between oxygen supply and demand ( hypoxia). Cells within hypoxic regions respond with the induction of a specific genetic program, under the control of the Hypoxia Inducible Factor (HIF), that mediates their adaptation to the lack of oxygen. The activity of HIF is mainly regulated by the EGL-nine homolog (EGLN) enzymes that hydroxylate the alpha subunit of this transcription factor in an oxygen-dependent reaction. Hydroxylated HIF is then recognized and ubiquitinilated by the product of the tumor suppressor gene, pVHL, leading to its proteosomal degradation. Under hypoxia, the hydroxylation of HIF by the EGLNs is compromised due to the lack of oxygen, which is a reaction cosubstrate. Thus, HIF escapes degradation and drives the transcription of its target genes. Since the progression of the aforementioned pathologies might be influenced by activation of HIF-target genes, development of small molecules with the ability to interfere with the HIF-regulatory machinery is of great interest. Results: Herein we describe a yeast three-hybrid system that reconstitutes mammalian HIF regulation by the EGLNs and VHL. In this system, yeast growth, under specific nutrient restrictions, is driven by the interaction between the beta domain of VHL and a hydroxyproline-containing HIF alpha peptide. In turn, this interaction is strictly dependent on EGLN activity that hydroxylates the HIFa peptide. Importantly, this system accurately preserves the specificity of the hydroxylation reaction toward specific substrates. We propose that this system, in combination with a matched control, can be used as a simple and inexpensive assay to identify molecules that specifically modulate EGLN activity. As a proof of principle we show that two known EGLN inhibitors, dimethyloxaloylglycine (DMOG) and 6-chlor-3-hydroxychinolin-2-carbonic acid-N-carboxymethylamide (S956711), have a profound and specific effect on the yeast HIF/EGLN/VHL system. Conclusion: The system described in this work accurately reconstitutes HIF regulation while preserving EGLN substrate specificity. Thus, it is a valuable tool to study HIF regulation, and particularly EGLN biochemistry, in a cellular context. In addition, we demonstrate that this system can be used to identify specific inhibitors of the EGLN enzymes.