By the evolutionarily conserved Hypoxia Inducible Factor family of basic helix-loop-helix transcription factors. HIFs are heterodimers of a beta subunit, and an alpha subunit. While ARNT levels are not sensitive to oxygen, both HIFa stability and its transcriptional activity are regulated by oxygen-dependent hydroxylation. Under oxygen restriction, HIFa subunits escape 
proteasomal degradation, heterodimerize with HIFb subunits and translocate to the cell nucleus, where they bind the RCGTG consensus sequence within regulatory regions of target genes, leading to their transcriptional activation in hypoxia. Mammals present three isoforms of HIFa that differ in their tissue distribution, HIF1a being the more ubiquitous and best characterized. A large number of studies focusing on single genes have identified individual HIF targets that, collectively, account for the functional responses to hypoxia, mainly metabolic adaptation and induction of angiogenesis. More recently, works employing HIF1a and HIF2a chromatin immunoprecipitation coupled to genomic microarrays or high-throughput sequencing have addressed the genome-wide identification of HIF binding locations, thereby improving the existing knowledge on the HIF-modulated transcriptome and largely confirming the RCGTG HIF binding consensus. Additionally, these studies have provided important insights into the global properties of HIF1 binding and transactivation. First, these works reported a significant association between the presence of a HIF binding site and hypoxic induction of the neighboring genes. The same trend was not found for genes repressed by hypoxia, Folinic acid calcium salt pentahydrate suggesting that hypoxia-mediated repression is largely indirect or HIF independent. Furthermore, they have clearly shown that only a small subset of about a hundred of all RCGTG-containing genes is robustly regulated by hypoxia. Hence, and in agreement with work on other transcription factors, HIFs bind a small proportion of potential binding sites, albeit the basis of their binding and target selectivity are incompletely understood. Understanding the mechanisms that explain HIFs transactivation selectivity is of paramount importance to expand our knowledge on transcriptional regulation and to improve the sensitivity and specificity of genome-wide efforts to characterize the HIF transcriptional response. DNA accessibility of transcription factor binding sites can clearly contribute to binding selectivity. For HIFs, recent evidence includes enhanced HIF1 and HIF2 binding to normoxic DNAse hypersensitivity sites and enrichment of HIF1 binding in the proximity of genes with a “permissive” transcriptional state in normoxia, as evidenced by significant basal expression. Additionally, DNA methylation has been also shown to modulate HIF1 binding, as originally demonstrated for the 39 enhancer of the Lomitapide Mesylate erythropoietin gene. A further mechanism that can impact target selectivity is direct or indirect cooperativity between transcription factors. Models of direct cooperativity have been mainly derived from developmental enhancers, and include the strict enhanceosome model, where cooperative occupancy occurs through extensive protein-protein interactions between TFs or common cofactors, and the more flexible billboard model, which suggests that enhancers contain submodules that interact independently or redundantly with promoters. Conversely, indirect cooperativity is based on the equilibrium competition between nucleosomes and DNA-binding proteins, thereby not requiring protein-protein interactions. In the case of HIFmediated transcription, the binding of cooperating transcription factors has been demonstrated for several target genes. In particular, HIF-mediated expression of the erythropoietin gene requires an adjacent HNF4 binding site, and PAI-1 induction by hypoxia has been linked to cooperative promoter activation by CEBPa.