Differential sensitivity of hypoxia inducible factor hydroxylation sites to hypoxia and hydroxylase inhibitors

Abstract

Hypoxia inducible factor (HIF) is regulated by dual pathways involving oxygen-dependent prolyl and asparaginyl hydroxylation of its α-subunits. Prolyl hydroxylation at two sites within a central degradation domain promotes association of HIF-α with the von Hippel-Lindau ubiquitin E3 ligase and destruction by the ubiquitin-proteasome pathways. Asparaginyl hydroxylation blocks the recruitment of p300/CBP co-activators to a C-terminal activation domain in HIF-α. These hydroxylations are catalyzed by members of the Fe(II) and 2-oxoglutarate (2-OG) oxygenase family. Activity of the enzymes is suppressed by hypoxia, increasing both the abundance and activity of the HIF transcriptional complex. We have used hydroxy residue-specific antibodies to compare and contrast the regulation of each site of prolyl hydroxylation (Pro(402), Pro(564)) with that of asparaginyl hydroxylation (Asn(803)) in human HIF-1α. Our findings reveal striking differences in the sensitivity of these hydroxylations to hypoxia and to different inhibitor types of 2-OG oxygenases. Hydroxylation at the three sites in endogenous human HIF-1α proteins was suppressed by hypoxia in the order Pro(402) > Pro(564) > Asn(803). In contrast to some predictions from in vitro studies, prolyl hydroxylation was substantially more sensitive than asparaginyl hydroxylation to inhibition by iron chelators and transition metal ions; studies of a range of different small molecule 2-OG analogues demonstrated the feasibility of selectively inhibiting either prolyl or asparaginyl hydroxylation within cells.

FIGURE 1.

Reactivity of hydroxy residue-specific anti-HIF1α antibodies. Immunoblots demonstrating the specificity of anti-hydroxy Pro⁴⁰² (Hyp402), anti-hydroxy Pro⁵⁶⁴ (Hyp564), and anti-hydroxy Asn⁸⁰³ (HyAsn803) antibodies for these hydroxylated residues in human HIF-1α. HIF-1α (wild-type and mutants P402A, P564G, or N803A) and HIF-2α polypeptides were produced as GAL4 fusions in IVTT reactions in the presence of DFO (Df) or ferrous chloride (Fe). Anti-GAL4 signals provide an internal control for normalization. Note a low level of cross-activity with HIF-2α observed for anti-hydroxy Pro⁵⁶⁴ and anti-hydroxy Asn⁸⁰³ in this immunoblot in lane 3.

FIGURE 2.

Validation of hydroxy residue-specific antibody performance by MS. A–C show data for sites in HIF-1α NODD, CODD, and C-terminal activation domain (CAD) sequences, respectively. Cells expressing doxycycline-inducible PK-tagged wild-type or mutant (M561A + M568A) HIF-1α were induced with doxycycline (1 μg/ml) in the presence of 1 mm DMOG for 24 h, or 25 μm MG132 for the last 4 h. For MS, PK immunoprecipitation was followed by in-gel digestion. The resulting peptides were analyzed by MS. The [M+2H]²⁺ precursor ions corresponding to the unhydroxylated and hydroxylated peptides are given together with chromatographic output (extracted ion counts) at these masses. Percentage hydroxylation was calculated from peak areas. * indicates confounding ions with a similar mass to the hydroxylated CODD peptide. Aliquots of the immunoprecipitation samples were immunoblotted with the hydroxy residue-specific or pan-HIF-1α antibodies as indicated. Note these hydroxyl residue-specific immunoblots were differentially exposed to allow secure ascertainment of the absence of signal in samples in which hydroxylation was demonstrated by MS to be suppressed.

FIGURE 3.

Levels of HIF asparaginyl hydroxylation in hypoxic cells and in rat and human tissues. A and B illustrate HIF-1α protein levels and Asn⁸⁰³ hydroxylation status in different cell lines under nomoxia (Nx), hypoxia (Hx, 1% oxygen), or exposed to 1 mm DMOG (Dm) for 5 h. HSF, primary human skin fibroblasts. C illustrates densitometric analyses of asparaginyl hydroxylation in two hypoxic cell extracts from A, with calibration against a standard assayed by MS. D illustrates the effect of more severe hypoxia (5 h) on HIF-1α asparaginyl hydroxylation in HT1080 and MCF7 cells. E, immunoblots of rat kidney and human cancer tissues. Lanes 1 and 2, MS standard as in C loaded at different volumes (v). Lanes 3 and 4, kidney extracts of rats breathing air (Nx) or treated with carbon monoxide (CO, 0.1%) for 6 h. Lanes 5 and 6, extracts from normal kidney and papillary kidney tumor from the same patient. Lanes 7–9, tumor extracts from different patients.

FIGURE 4.

Effect of graded hypoxia and chronic hypoxia on HIF-1α hydroxylation. A and B, graded hypoxia. RCC4 cells grown on Lumox dishes were incubated at the indicated oxygen concentrations for 4 h. The experiment was performed in triplicate. A, representative immunoblots; B, semi-quantitative analysis of relative band intensities (mean ± S.E., n = 3). *, significantly different from normoxia (21%), p < 0.05. C and D, chronic hypoxia. RCC4 cells, seeded to achieve similar density at time of analysis, were incubated in normoxia (Nx) or hypoxia (Hx, 1% oxygen) for 8 h to 7 days. Two independent experiments were performed, each in duplicate. C, representative immunoblots. D, semi-quantitative analysis of relative band intensities (mean ± S.E., n = 4). In panels B and D, the relative intensity is the ratio of the hydroxy residue-specific signal to total anti-HIF-1α in that sample. Based on preliminary data indicating that hydroxylation at all sites were essentially 100% in normoxic RCC4 cells, these values were plotted with the ratio in normoxic cells set to one (equivalent to 100% hydroxylation). In B, normoxic S.E. values have been omitted for clarity.

FIGURE 5.

Differential regulation of HIF-1α hydroxylation sites during hypoxic induction and during re-oxygenation. A, reoxygenation. RCC4 cells, incubated in Lumox dishes in normoxia (Nx) or hypoxia (Hx; 0.1% oxygen) for 4 h, were then exposed to normoxia for the indicated times. Hydroxylation at Hyp564 and HyAsn⁸⁰³ is complete within 1 min, but delayed at Hyp402. B and C, induction of HIF-1α by hypoxia. HeLa or HT1080 cells were incubated in normoxia or hypoxia (1 or 3% oxygen) for the indicated times (h). Immunoblots were probed with the indicated antibodies. During the accumulation of HIF-1α in hypoxic cells, substantial hydroxylation at Hyp564 but not Hyp402 is observed, particularly in HeLa cells.

FIGURE 6.

Differential inhibition of HIF prolyl and asparaginyl hydroxylation by HIF hydroxylase inhibitors. A–C, VHL-defective (RCC4) cells. D–F, VHL-competent (RCC4/VHL) cells. Cells were treated with inhibitors as indicated (see supplemental Fig. S5 for structures). Immunoblots were probed with the indicated antibodies. Controls were exposed to 0–2% DMSO in accordance with the diluent used for the test substance.

FIGURE 7.

Selective chemical inhibition of HIF asparaginyl hydroxylation. A, VHL-defective (RCC4) cells were treated with inhibitors as indicated for 5 h. Immunoblots show inhibition of Asn⁸⁰³ hydroxylation by compounds E and F. B, compound E (5 h) did not up-regulate HIF-1α in RCC4/VHL cells. C, RCC4/VHL cells were exposed to compound E (5 h) in the presence or absence of hypoxia (1% oxygen) for the last 2 h. Compound E inhibited Asn⁸⁰³ hydroxylation in the HIF-1α induced by hypoxia.