Comment on Licht et al, page 584

Expression of a dominant-negative form of hypoxia-inducible factor 2α (HIF-2α) in endothelial cells (ECs) disrupts cardiovascular development in mouse embryos, providing further evidence that HIFs exert both non–cell-autonomous and cell-autonomous control over ECs.

The developmental and physiologic processes that regulate O2 homeostasis in metazoans are directed by a family of hypoxia-inducible factors (HIFs), the prototype of which (HIF-1) was identified in 1992 as a mediator of erythropoietin gene transcription in hypoxic cells.1  Biochemical and molecular analyses revealed that HIF-1 was a heterodimer of constitutively expressed HIF-1β and O2-regulated HIF-1α subunits. Database searches identified HIF-2α, which dimerizes with HIF-1β and transactivates an overlapping but distinct set of genes from those regulated by HIF-1.2  In contrast to the ubiquitous expression of HIF-1α and HIF-1β, the expression of HIF-2α is tissue specific, most notably within vascular endothelial cells (ECs). HIF-3α is another family member that dimerizes with HIF-1β. HIF-1α and HIF-2α, but not HIF-3α, contain transactivation domains that interact with the coactivator proteins CBP and p300 (see figure). The half-life of HIF-1α, HIF-2α, and HIF-3α is regulated by O2-dependent hydroxylation of prolyl residues, which is necessary for binding of the von Hippel-Lindau protein, the recognition component of an E3-ubiquitin ligase that targets the proteins for proteasomal degradation.3  In addition, hydroxylation of an asparaginyl residue in HIF-1α and HIF-2α blocks their interaction with coactivators. Because O2 is a rate-limiting substrate for HIF asparaginyl and prolyl hydroxylases, changes in oxygenation are transduced into changes in transcription.FIG1 

Mammalian hypoxia-inducible factors. HIF-1α, HIF-2α, HIF-3α, and HIF-1β are each encoded by a distinct gene, with alternative mRNA splicing generating multiple isoforms (not shown) of HIF-1α, HIF-3α, and HIF-1β. HIF-1α, HIF-2α, and HIF-3α dimerize with HIF-1β to form the functional proteins HIF-1, HIF-2, and HIF-3, respectively. The basic-helix-loop-helix (bHLH)–PAS domains mediate dimerization and DNA binding. All known HIF-1 binding sites contain the core DNA sequence 5′-RCGTG-3′ (R, A, or G). Hydroxylation of prolyl residues (402 and 564 in human HIF-1α) is required for the binding of the von Hippel-Lindau protein (VHL), which targets the proteins for ubiquitination and proteasomal degradation. Hydroxylation of an asparagine residue (803 in human HIF-1α) blocks the binding of coactivators p300 and CBP. The HIF prolyl and asparaginyl hydroxylases use O2 and α-ketoglutarate and generate CO2 and succinate as by-products.

Mammalian hypoxia-inducible factors. HIF-1α, HIF-2α, HIF-3α, and HIF-1β are each encoded by a distinct gene, with alternative mRNA splicing generating multiple isoforms (not shown) of HIF-1α, HIF-3α, and HIF-1β. HIF-1α, HIF-2α, and HIF-3α dimerize with HIF-1β to form the functional proteins HIF-1, HIF-2, and HIF-3, respectively. The basic-helix-loop-helix (bHLH)–PAS domains mediate dimerization and DNA binding. All known HIF-1 binding sites contain the core DNA sequence 5′-RCGTG-3′ (R, A, or G). Hydroxylation of prolyl residues (402 and 564 in human HIF-1α) is required for the binding of the von Hippel-Lindau protein (VHL), which targets the proteins for ubiquitination and proteasomal degradation. Hydroxylation of an asparagine residue (803 in human HIF-1α) blocks the binding of coactivators p300 and CBP. The HIF prolyl and asparaginyl hydroxylases use O2 and α-ketoglutarate and generate CO2 and succinate as by-products.

HIF-1 regulates genes encoding angiogenic cytokines, including vascular endothelial growth factor (VEGF), placental growth factor, angiopoietins 1 and 2, and stromal-derived factor 1, providing a mechanism by which cells are assured of adequate perfusion, as hypoxia-induced VEGF stimulates angiogenesis. Thus, HIF-1 was viewed as an extrinsic (non–cell-autonomous) regulator of ECs and their progenitors, which express VEGFR2 (Flk-1), VEGFR1 (Flt-1), Tie2, and CXCR4, the cell-surface receptors for VEGF, PLGF, angiopoietins, and SDF-1, respectively. HIF-2 was thought to function as an intrinsic (cell-autonomous) regulator of ECs. However, HIF-1 regulates the expression of hundreds of genes in ECs,4  and tumor angiogenesis is significantly impaired in mice with EC-specific loss of HIF-1α.5  Mice completely lacking HIF-2α expression have normal cardiovascular development in certain genetic backgrounds,6  whereas complete HIF-1α deficiency results in major defects in cardiovascular development.1 

A logical conclusion is that HIF-1α and HIF-2α play critical and partially overlapping roles in ECs. To test this hypothesis, Licht and colleagues generated transgenic mice with EC-specific expression of a dominant-negative form of HIF-2α (HIFdn) directed by Flk1 gene promoter/enhancer elements. Overexpressed HIFdn competes with HIF-1α, HIF-2α, and HIF-3α for dimerization with HIF-1β, but the dimers cannot bind DNA or activate transcription. HIFdn transgenic embryos manifested cardiovascular defects similar to those of Tie2-deficient mice and the embryos completely lacked Tie2 expression. Thus, whereas loss of HIF-1 or HIF-2 activity in ECs is compatible with normal cardiovascular development, the combined loss of HIF-1 and HIF-2 activity is not. Overexpression of HIFdn may have confounding effects, such as competitive binding to proteins that interact with HIF-2α at residues not deleted in HIFdn, resulting in their cytoplasmic sequestration. It will be interesting to determine whether EC-specific HIF-1β deficiency phenocopies HIFdn and whether combined loss of HIF-1 and HIF-2 at earlier developmental stages (ie, prior to Flk1 expression) reveals essential roles for these proteins in hemangiopoiesis or vasculogenesis. Nevertheless, the results of Licht et al provide strong evidence that HIF-1 and HIF-2 play critical intrinsic roles in ECs. Because of their dual (extrinsic and intrinsic) effects, targeting these factors may represent a powerful approach to inhibiting angiogenesis in cancer and other disorders that are dependent upon neovascularization. ▪

1
Semenza GL. Hydroxylation of HIF-1: oxygen sensing at the molecular level.
Physiology (Bethesda).
2004
;
19
:
176
-182.
2
Raval RR, Lau KW, Tran MG, et al. Contrasting properties of hypoxia-inducible factor 1 (HIF-1) and HIF-2 in von Hippel-Lindau-associated renal cell carcinoma.
Mol Cell Biol.
2005
;
25
:
5675
-5686.
3
Kaelin WG Jr. The von Hippel-Lindau protein, HIF hydroxylation, and oxygen sensing.
Biochem Biophys Res Commun.
2005
;
338
:
627
-638.
4
Manalo DJ, Rowan A, Lavoie T, et al. Transcriptional regulation of vascular endothelial cell responses to hypoxia by HIF-1.
Blood.
2005
;
105
:
659
-669.
5
Tang N, Wang L, Esko J, et al. Loss of HIF-1α in endothelial cells disrupts a hypoxia-driven VEGF autocrine loop necessary for tumorigenesis.
Cancer Cell.
2004
;
6
:
485
-495.
6
Scortegagna M, Ding K, Oktay Y, et al. Multiple organ pathology, metabolic abnormalities and impaired homeostasis of reactive oxygen species in Epas1–/– mice.
Nat Genet.
2003
;
35
:
331
-340.