Mechanism of hypoxia-induced NF-kappaB

Abstract

NF-κB activation is a critical component in the transcriptional response to hypoxia. However, the underlying mechanisms that control its activity under these conditions are unknown. Here we report that under hypoxic conditions, IκB kinase (IKK) activity is induced through a calcium/calmodulin-dependent kinase 2 (CaMK2)-dependent pathway distinct from that for other common inducers of NF-κB. This process still requires IKK and the IKK kinase TAK1, like that for inflammatory inducers of NF-κB, but the TAK1-associated proteins TAB1 and TAB2 are not essential. IKK complex activation following hypoxia requires Ubc13 but not the recently identified LUBAC (linear ubiquitin chain assembly complex) ubiquitin conjugation system. In contrast to the action of other NF-κB inducers, IKK-mediated phosphorylation of IκBα does not result in its degradation. We show that this results from IκBα sumoylation by Sumo-2/3 on critical lysine residues, normally required for K-48-linked polyubiquitination. Furthermore, inhibition of specific Sumo proteases is sufficient to release RelA from IκBα and activate NF-κB target genes. These results define a novel pathway regulating NF-κB activation, important to its physiological role in human health and disease.

FIG. 1.

Hypoxia induces NF-κB. (A) U2OS, MDA-MB-231, and MCF10A cells were exposed to 1% O₂ for the indicated times or were treated with 10 ng/ml TNF-α for 30 min prior to lysis. Whole-cell lysates were analyzed by Western blotting with the indicated antibodies. m, minutes. (B) U2OS cells were exposed for 60 min to decreasing concentrations of O₂ prior to harvest, and cytoplasmic/nuclear extracts were prepared. Nuclear (a) and cytoplasmic (b) extracts were analyzed as for panel A. (C) U2OS cells were exposed to 1% O₂ for different times before nuclear extracts were prepared and analyzed for the DNA binding activities of the indicated transcription factors. OCT-1 was used as a loading control. (D) U2OS cells were exposed to 1% O₂ or TNF-α for 30 min prior to nuclear extraction of proteins. Nuclear extracts were analyzed for DNA binding activity by using a radiolabeled oligonucleotide with a mutated NF-κB binding site (HIV-mut κB). (E) Nuclear extracts obtained under the same experimental conditions as the nuclear extracts analyzed in panel B were used in EMSA analysis for NF-κB and OCT-1 DNA binding. (F) Control nuclear extracts (0 min) and nuclear extracts exposed to 1% O₂ for 30 min were incubated with the indicated antibodies prior to being analyzed for DNA binding activity. Results of two representative experiments (a and b) are shown. Arrows indicate the shifted bands.

FIG. 2.

Hypoxia modulates NF-κB to both activate and repress target genes. (A) U2OS cells were transfected with 1 μg of the indicated luciferase reporter constructs and were treated for 16 h with 1% O₂ prior to luciferase measurements. The graph depicts mean activation or repression relative to the luciferase activity of control samples, plus standard deviations. (B) U2OS cells were transfected with 1 μg of the 3×κB-ConA-luciferase or deleted-κB ConA-luciferase (delκB) construct. Cells either were cotransfected with 1 μg of an empty plasmid (CTL) or a HIF-1α expression plasmid or were treated with 1% O₂ (Hyp) or 100 μM DFX for 16 h prior to harvest and luciferase measurement. (C and D) U2OS, MDA-MB-231, and MCF10A cells were treated with 1% O₂ for the indicated times prior to harvest and mRNA extraction. IL-8 (C) and (D) PTEN mRNA levels were analyzed by quantitative PCR. Graphs depict IL-8 and PTEN mRNA levels compared to levels in untreated samples (0h). (E) Mouse endothelial cells were exposed to 1% O₂ for the indicated times prior to harvest and mRNA extraction. The indicated genes were analyzed by qPCR. *, P ≤ 0.050; **, P ≤ 0.010.

FIG. 3.

Induction of IL-8 and repression of PTEN by hypoxia are RelA dependent. (A) U2OS cells were transfected with specific siRNA oligonucleotides prior to hypoxic stress exposure for the indicated times. IL-8 and PTEN mRNAs were analyzed by qPCR. (B) U2OS cells were treated and processed as for panel A, but qPCR was performed for RelA and p100 levels. (C) U2OS cells were transfected with specific siRNA oligonucleotides prior to hypoxic stress exposure for the indicated times. Whole-cell lysates were analyzed by Western blotting. (D) MCF10A cells were transfected with specific siRNA oligonucleotides prior to hypoxia exposure for the indicated times. IL-8 and PTEN mRNAs were analyzed by qPCR. *, P ≤ 0.050; **, P ≤ 0.010.

FIG. 4.

NF-κB-mediated responses to hypoxia differ from those to TNF-α. (A) U2OS cells were transfected with HIF-1α siRNA oligonucleotides and were then exposed to hypoxia for 2 h prior to RNA extraction. HIF-1α, IL-8, and ADM mRNA levels were analyzed by qPCR. (B) U2OS cells were exposed to 1% O₂ for the indicated periods prior to RNA extraction. p100 and IAP2 mRNA levels were analyzed by qPCR. (C) U2OS cells were transfected with specific siRNA oligonucleotides prior to hypoxic stress exposure for the indicated times. Whole-cell lysates were analyzed by Western blotting. (D) U2OS cells were transfected with specific siRNA oligonucleotides prior to hypoxic stress exposure for the indicated times. IAP2 mRNA was analyzed by qPCR. (E) U2OS cells were treated with 10 ng/ml TNF-α for the indicated periods prior to RNA extraction. IL-8, p100, IAP2, and PTEN mRNA levels were analyzed by qPCR. *, P ≤ 0.050; **, P ≤ 0.010.

FIG. 5.

Hypoxia induces the dynamic recruitment of RelA to the promoters of IL-8 and PTEN with opposing transcriptional outcomes. (A and B) U2OS (A) and MDA-MB-231 (B) cells were exposed to 1% O₂ for the indicated periods prior to being cross-linked and harvested for ChIP analysis. Following immunoprecipitation (IP) with the indicated antibodies, DNA was purified and analyzed by qPCR using primers specific for the indicated promoters. (C) U2OS cells were treated as for panel A, and RNA polymerase II (Pol II) occupancy and the levels of acetylated H3 (AcH3) at the IL-8 and PTEN promoters were analyzed by qPCR. (D) U2OS cells were treated as for panel A, and the levels of RelA and p52 at the control regions of the IL-8 and PTEN genes were analyzed by qPCR. *, P ≤ 0.050; **, P ≤ 0.010.

FIG. 6.

Hypoxia-induced NF-κB requires the IKK complex. (A) U2OS cells either were treated with 1% O₂ for the indicated times or were treated with 10 ng/ml TNF-α, 50 μM H₂O₂, or 100 μM pervanadate (PVn) for 30 min prior to harvest, and cytoplasmic/nuclear extracts were prepared. Protein extracts were analyzed by Western blotting. (B) U2OS cells either were treated with 1% O₂ for the indicated times or were treated with 100 μM pervanadate for 30 min prior to harvest. Two hundred micrograms of the total-cell lysate was used to immunoprecipitate IκBα. Immunoprecipitated material was analyzed by Western blotting. (C) U2OS cells were pretreated with the IKK inhibitor (IKKi) 20 μM Bay 117082 30 min prior to treatment with 1% O₂ or 10 ng/ml TNF-α for an additional 30 min. Cells were then harvested, and nuclear (a) and cytoplasmic (b) extracts were obtained. These were analyzed by Western blotting. (D) Nuclear extracts obtained in the experiment described for panel C were analyzed by EMSA. (E) Wild-type IKK and IKKα/β-null MEFs were treated with 1% O₂ for the indicated times prior to harvest, and nuclear and cytoplasmic extracts were obtained. Nuclear extracts were analyzed by EMSA for NF-κB and OCT-1 DNA binding. (F) Cytoplasmic extracts obtained as described for panel E were analyzed by Western blotting. (G) Wild-type and Nemo/IKKγ-null MEFs were exposed to 1% O₂ for the indicated periods prior to lysis, and cytoplasmic/nuclear extracts were prepared. Cytoplasmic extracts were analyzed by Western blotting. (H) MEFs with wild-type IKK (IKKWT) and IKKα- and IKKβ-null MEFS were treated with 1% O₂ for the indicated times, and nuclear extracts were obtained. NF-κB DNA binding activity was analyzed by EMSA.

FIG. 7.

Hypoxia-induced IKK-NF-κB is TAK1 dependent. (A) MEFs with wild-type (WT) TAK1 and TAK1-null MEFS were exposed to 1% O₂ for the indicated periods prior to being harvested, and cytoplasmic/nuclear extracts were obtained. Nuclear extracts were analyzed by EMSA for the DNA binding activities of NF-κB and OCT-1. (B) Cytoplasmic extracts obtained as for panel A were analyzed by Western blotting. (C and D) U2OS (C) and HeLa (D) cells were transfected with the indicated siRNA oligonucleotides 48 h prior to harvest. Thirty minutes and 15 min prior to lysis, cells were exposed to 1% O₂. Whole-cell lysates were analyzed by Western blotting. (E) TAK1 inhibition reduces hypoxia-induced IKK activity. U2OS cells were pretreated with 1 μM 5Z-7-oxozeaenol for 45 min prior to treatment with 1% O₂ for an additional 30 min. Nuclear and cytoplasmic extracts were prepared and were analyzed by Western blotting. C, control; Ti, TAK inhibitor; H, hypoxia. (F) U2OS cells were transfected with control and TAK1 siRNA oligonucleotides prior to being exposed to 1% O₂ for the indicated times. Whole-cell lysates were analyzed by Western blotting. (G) MEFs with wild-type TAB1 and TAB1-null MEFs were exposed to 1% O₂ for the indicated times prior to being harvested. Cytoplasmic extracts were analyzed by Western blotting. (H) MEFs with wild-type TAB2 and TAB2-null MEFs were exposed to 1% O₂ for the indicated times prior to being harvested. Cytoplasmic extracts were analyzed by Western blotting. (I) U2OS cells were transfected with 1 μg of a control construct or a GST-TAK-TAB1 fusion construct for 48 h prior to harvest. Whole-cell lysates were analyzed by Western blotting.

FIG. 8.

Hypoxia-induced IKK is PHD1, PHD2, PHD3, and HIF-1α independent. U2OS cells were transiently (right) or stably (left) depleted of PHD1 (A) (arrow indicates specific band), PHD2 (B), or PHD3 (C and D) or were stably depleted of HIF-1α (E and F). Cells were exposed to 1% O₂ for the indicated times or to 10 ng/ml TNF-α (T) for 30 min prior to lysis. Whole-cell lysates were analyzed by Western blotting.

FIG. 9.

Hypoxia-induced TAK1-IKK-NF-κB does not involve TRAF2, TRAF6, PERK, or ATM. (A) MEFs with wild-type (WT) TRAF2 and TRAF2-null MEFs were exposed to 1% O₂ for the indicated periods prior to being harvested. Cytoplasmic extracts were analyzed by Western blotting. (B) MEFs with wild-type TRAF6 and TRAF6-null MEFs were treated and analyzed as for panel A. (C) ATM-null and reconstituted ATM cells were treated and analyzed as for panel A. (D) U2OS cells were exposed to 1% O₂ for the indicated times or to 30 min of 10 ng/ml TNF-α prior to lysis, and whole-cell lysates were analyzed by Western blotting. (E) MEFs with wild-type PERK and PERK-null MEFS were treated and analyzed as for panel A. These cells were also treated with 10 ng/ml TNF-α for 30 min prior to being analyzed.

FIG. 10.

Hypoxia-induced TAK1-IKK-NF-κB requires Ca²⁺ and CAMK2. (A) U2OS cells were treated with 2 μM KN-93 30 min prior to exposure to 1% O₂ for the indicated times. Whole-cell lysates were analyzed by Western blotting. (B) U2OS cells were transfected with a control or a CaMK2δ siRNA oligonucleotide, and the level of knockdown was assessed by RT-PCR. (C) U2OS cells were transfected as for panel B and were exposed to 1% O₂ for the indicate times prior to lysis. Whole-cell lysates were analyzed by Western blotting. (D) U2OS cells were transfected as for panel B prior to exposure to 1% O₂ for the indicated times and RNA extraction. IL-8 and PTEN mRNAs were analyzed by qPCR. *, P ≤ 0.050; **, P ≤ 0.010. (E) U2OS cells were treated with 1 μM thapsigargin for 1 h prior to exposure to 1% O₂ for the indicated periods. Whole-cell lysates were analyzed by Western blotting. (F) U2OS cells were transfected with 1 μg of a control or a CAMK2δ construct for 48 h prior to harvest. Whole-cell lysates were analyzed by Western blotting.

FIG. 11.

Hypoxia-induced TAK1-IKK-NF-κB requires Ubc13 but not UbcH5 or LUBAC. (A) U2OS cells were transfected with the indicated siRNA oligonucleotides and were exposed to 1% O₂ for the indicated times before lysis. Whole-cell lysates were analyzed by Western blotting. (B) U2OS cells were transfected with a control, HOIP, or HOIL-1L siRNA oligonucleotide prior to being treated and processed as for panel A. (C) U2OS cells were transfected with a control or a UbcH5a siRNA oligonucleotide prior being treated and processed as for panel A.

FIG. 12.

Hypoxia-promoted Sumo-2/3 conjugation to IκBα prevents its degradation. (A) U2OS cells were treated with the proteasome inhibitor MG132 3 h prior to a 30-min treatment with 1% O₂ or 10 ng/ml TNF-α. Cells were harvested directly into SDS loading buffer, and IκBα levels were determined by Western blotting (WB). (B) U2OS cells were treated with TNF-α for the indicated periods in the presence or absence of 1% O₂ prior to lysis. Whole-cell lysates were analyzed by Western blotting. (C) HEK293 cells were transfected with 1 μg of IκBα and 1 μg of His₆-ubiquitin, and where indicated, they were treated with TNF-α or MG132, or both. Ubiquitinated proteins were purified using Ni²⁺ agarose, and pulldown eluates were analyzed by Western blotting for IκBα. Arrows indicate modified versions of IκBα. (D) HEK293 cells were transfected with 1 μg of IκBα and 1 μg of a His-tagged version of ubiquitin or Nedd8 where indicated. Cells were treated or not with 1% O₂ for 15 min and were harvested 48 h posttransfection. Extracts were processed and analyzed as for panel C. Arrows indicate modified versions of IκBα. (E) HEK293 cells were transfected with 1 μg of IκBα and 1 μg of a His-tagged version of Sumo-2 (S2) or Sumo-3 (S3) where indicated. Cells were treated or not with 1% O₂ for 15 min and were harvested 48 h posttransfection. Extracts were processed and analyzed as for panel C. Arrows indicate modified versions of IκBα. (F) Sumo-2 is conjugated to lysine 21 of IκBα. HEK293 cells were treated with1 μg of wild-type (Wt) or K21A or Y42A mutant IκBα and were exposed or not to 1% O₂ for 15 min prior to lysis. Sumoylated proteins were purified as for panel C, and pulldown eluates were analyzed for IκBα by Western blotting. Arrows indicate modified versions of IκBα.

FIG. 13.

Hypoxia-promoted Sumo-2/3 conjugation to IκBα prevents its degradation but results in NF-κB activation. (A) U2OS cells were transfected with the indicated siRNA oligonucleotides prior to lysis. Whole-cell lysates were analyzed by Western blotting. Senp3 levels were analyzed by qPCR. *, P ≤ 0.050; **, P ≤ 0.010. NT, nontargeted. (B) U2OS cells were transfected with the indicated siRNA oligonucleotides and were either left untreated or treated with 10 ng/ml TNF-α for 30 min prior to lysis. Whole-cell lysates were analyzed by Western blotting. (C) U2OS cells were transfected with either 2 μg of an empty control vector or 1 μg of Senp1 and 1 μg of Senp2. Cells were either left untreated or treated with 1% O₂ for the indicated times and were harvested 48 h posttransfection. Whole-cell lysates were analyzed by Western blotting. (D) U2OS cells were transfected with the indicated siRNA oligonucleotides for 48 h prior to lysis. Whole-cell lysates (200 μg) were used to immunoprecipitate IκBα. Levels of RelA and IκBα present in the immunoprecipitates were analyzed by Western blotting and quantified using ImageJ software. (E) U2OS cells were transfected with a control siRNA or a Senp5 (S5a, S5b), Senp6 (S6a, S6b), or Senp7 (S7a, S7b) siRNA oligonucleotide, and total RNA was extracted. IL-8 and IAP2 mRNA levels were determined by qPCR. *, P ≤ 0.050; **, P ≤ 0.010. (F) U2OS cells were transfected with the indicated siRNA oligonucleotides for 48 h prior to lysis. Whole-cell lysates were analyzed by Western blotting. Levels of IAP2 were quantified using ImageJ software.