SGI-1027 br Tissue Specific Roles for HIF PHD Isoforms

Tissue-Specific Roles for HIF/PHD Isoforms Hypoxia signaling pathways are active in almost all mammalian cells, including immune SGI-1027 and tissues such as the epithelium and endothelium. This is of crucial importance when considering the influence of environmental hypoxia in inflammatory conditions and is particularly relevant in the specialized tissues of the intestine, lung, and joint, all of which may be the site of infection, sterile inflammation (owing to physical, chemical, or metabolic stressors [21]), or autoimmune disease, and are also subject to local hypoxia. This hypoxia may be due to pathology but can also occur in normal physiology, as in the intestine.
Specific Roles of HIF/PHDs in Immune Cells
HIFs and PHDs in Adaptive Immunity
Therapeutic Targeting of the HIF/PHD Pathway Specialized fibroblastic cells of the kidney are the main source of the HIF target erythropoietin [83], and EPO production is often impaired in chronic renal disease, leading to renal anemia. HIF2α and PHD2 are essential for EPO production 84, 85. Genetic silencing of two or all three isoforms of PHD was sufficient to enhance erythropoietin production 86, 87, and Phase III trials of PHD inhibitors are ongoing for the treatment of renal anemia (reviewed in [88]). Both the PHDs and FIH are dependent on ferrous iron and the metabolite 2-oxaloglutarate (2-OG). Early studies identified that metal chelators, including the iron chelator desferrioxamine, could induce hypoxia signaling in the form of EPO production [89]. Of the compounds currently under investigation in clinical trials, the majority are thought to act through competitive inhibition of the essential PHD co-substrate 2-OG (reviewed in detail by Chan et al. [90]). Given the pleiotropic effects of HIF (as demonstrated by the targets identified in Figure 1), it is interesting to consider utilizing these compounds in the treatment of other disease processes, including inflammation. These strategies are less advanced than those targeting anemia, but several mouse models have been investigated. The FibroGen compound FG4497 was found to be protective in a model of acute lung injury, an effect attributed to HIF2α stabilization in the vascular endothelial barrier [91]. FG4497 was also shown to induce HIF1α in IECs in a murine colitis model resulting in improved clinical symptoms [33]. The Akebia compound AKB-4924 was shown to be therapeutic in the murine colitis model when administered intravenously or orally 31, 32. This protection was lost in IEC-specific Hif1a knockout mice, confirming the importance of epithelial HIF1α in these cells for protection. Although both intravenous and oral administration of FG4497 also led to induction of EPO in vivo, oral administration of AKB-4924 did not, suggesting a potential advantage in terms of side effects. Further evidence that targeted drug delivery approaches may beneficial was provided by Tambuwala et al. [92] who used a liquid emulsion formulation of the hydroxylase inhibitor dimethyloxalylglycine (DMOG) that allowed improved delivery to the colon with reduced systemic exposure. Inflammation associated with infection presents an additional challenge: preserving immunity without excessive inflammation. Murine models of both skin [93] and urinary tract infections [94] showed potential benefits of PHD inhibition with AKB-4924 owing to enhanced antibacterial capacity of immune cells. Importantly, in both models the compound was delivered locally rather than systemically, reducing the risk of systemic effects and this would need to be considered in any clinical trial.
Concluding Remarks The studies described above highlight the therapeutic potential of targeting hypoxia signaling pathways in infection and inflammation. However, the pleiotropic effects of HIF, through direct transcriptional regulation and other mechanisms, mean that the risk of unwanted or unexpected side effects is considerable: evidence from multiple studies suggests that activation of HIF1α may be protective in intestinal inflammation 22, 28, 31, 32, 33. However, constitutively active HIF1α is known to cause lymphoproliferative disorders in mouse models [95], a concern when considering long-term systemic treatment with compounds that activate HIF1α. The divergent roles of PHD isoforms in the intestine and in immune cells such as neutrophils suggest a role for developing isoform-specific inhibitors. Equally, a more complete understanding of which PHD isoform is dominant in a particular context may help to predict the effect of pan-hydroxylase inhibition. In addition, in targeting hypoxia signaling pathways there is likely to be a trade-off between anti-inflammatory effects and maintaining effective immunity, as illustrated by studies on T cells [4] and macrophages [51]. The use of tissue- and isoform-specific transgenic mice has provided valuable insights into the specific functions of HIF and PHD family members, but these cannot always predict the effect on the organism as a whole. Nor are transgenic models necessarily equivalent to pharmacological inhibition of HIF/PHD pathway members. These differences are likely to relate to both off-target effects of inhibitors and ‘promiscuous’ functions of PHDs, independently of their hydroxylase activity. What is clear is that it is the outcome of the inflammatory response is dependent upon multiple inputs: the cell type(s) in question, the inflammatory and metabolic milieu, the duration of the insult, and the location of the inflammation (Figure 3, Key Figure).