HIF and Hemoglobin in the Himalayas

At high-altitude where hypoxia reigns, the expectation is that humans will develop a compensatory erythrocytosis so as to increase tissue oxygen delivery; however, such is not the case with inhabitants of the Tibetan plateau who maintain an hematocrit similar to that of individuals living at sea level. To understand the basis of this counterintuitive adaptive response, a review of the physiologic response to hypoxia is helpful. The master regulators of this process are the oxygen-sensing pathway’s HIF transcription factors, which consist of a constitutively expressed β subunit, and a labile α subunit with three distinct isoforms (HIF-1α, HIF-2α, and HIF-3α). HIF-2α (encoded by the EPAS1 gene) is the primary regulator of erythropoietin (EPO) production. Under normoxic conditions, HIF prolyl 4-hydroxylase 2 (PHD2), encoded by the gene EGLN1, hydroxylates HIFα — a process that tags the transcription factor for recognition by the von Hippel-Lindau (VHL) protein. VHL is a component of an E3 ubiquitin ligase complex that targets HIF-2α for proteasomal degradation. Because the prolyl hydroxylation step is oxygen-dependent, under hypoxic conditions, HIF-2α is not hydroxylated and therefore is not degraded by the VHL-ubiquitin-proteasome system. Consequently, the half-life of HIF-2α is prolonged under hypoxic conditions, thereby supporting enhanced expression of HIF-regulated genes such as EPO. Some cases of hereditary secondary erythrocytosis, typically associated with normal to high EPO levels, have been linked to variants in this complex oxygen-sensing pathway through identification of mutations in the genes EPAS1, EGLN1, and VHL. For example, homozygous germ-line mutations affecting VHL underlie an inherited form of secondary polycythemia that is endemic in the Chuvash region of Russia.

In a multicenter collaboration directed by Dr. Josef Prchal of the University of Utah School of Medicine and Dr. Peppi Koivunen from the University of Oulu in Finland, researchers uncovered the genetic mechanism that protects Tibetan highlanders from polycythemia. They identified a missense variant, 12C>G [Asp4Glu] in the EGLN1 gene, that exists in almost complete linkage disequilibrium with a variant, 380G>C [Cys127Ser], in the same gene that was previously shown to be associated with adaptation to high altitude. While the 380G>C variant was found in all 26 U.S. Tibetan DNA samples analyzed, it also occurred in non-Tibetan controls at a frequency of ≈20 percent. In contrast, the12C>G variant was found in 85 percent of the Tibetan cohort but in only 0.8 percent of 242 non-Tibetan individuals. Using SNP array data derived from unrelated U.S. Tibetans, investigators estimated that the 12C>G variant arose relatively recently, approximately 8,000 years ago, and that its rapid increase in allele frequency among Tibetans is a consequence of powerful natural selection pressure.

Functional assays revealed that the PHD2 double variant protein [Asp4Glu; Cys127Ser] more efficiently hydroxylated HIF subunits at lower oxygen tensions than wild-type PHD2. In addition, lower levels of HIF-2α protein were observed in lentivirally-infected Hep3B hepatoma cells expressing the double variant compared with wild-type PHD2 protein when exposed to hypoxic (1% O₂) conditions. These data indicate that double variant PHD2 down-regulates HIF more effectively than wild-type PHD2 under low oxygen tension. Invitro, the development of erythroid colonies (burst-forming unit-erythroid [BFU-E]) and sensitivity to EPO were subsequently compared between Tibetan and control peripheral blood samples under hypoxemic conditions of 1 percent and 5 percent O₂(Figure). Whereas increased proliferation, size, and hemoglobinization of BFU-E (as well as heightened sensitivity to EPO) were demonstrated with normal erythroid progenitors, these BFU-E readouts from Tibetans with double variant PHD2 were significantly decreased at 5 percent O₂, and growth was abrogated at 1 percent O₂. Surprisingly, however, erythroid progenitors from Tibetans homozygous for the double PHD2 variant were hypersensitive to EPO under normoxicconditions — a finding that is unexplained but that mirrors the erythrocytosis phenotypes associated with the aforementioned mutations of EGLN1, EPAS1, and VHL.