HIF-1 (hypoxia-inducible factor-1) has been shown to essentially control the cellular response to hypoxia. Hypoxia stabilizes the inducible α-subunit, preventing post-translational hydroxylation and subsequent degradation via the proteasome. In recent years, clear evidence has emerged that HIF-1α is also responsive to many stimuli under normoxic conditions, including thrombin, growth factors, vasoactive peptides, insulin, lipopolysaccharide and cytokines such as TNF-α (tumour necrosis factor-α), and in many cases reactive oxygen species are involved. One important mechanism underlying these responses is the transcriptional regulation of HIF-1α by the redox-sensitive transcription factor NF-κB (nuclear factor κB), which binds at a distinct element in the proximal promoter of the HIF-1α gene. More recently, NF-κB binding to this site in the HIF-1α promoter has been shown also under hypoxic conditions. Thus these two major pathways regulating the responses to inflammation and oxidative stress on the one hand, and hypoxia on the other hand, appear to be intimately linked. In this issue of the Biochemical Journal, a study by van Uden et al. has supported these findings further, in which they have confirmed the binding of several proteins of the NF-κB family at the previously identified consensus site in the HIF-1α promoter and shown that TNF-α can also transcriptionally induce HIF-1α by this previously described pathway. The identification of HIF-1α as a target gene of NF-κB will have important implications for a variety of disorders related to hypoxia–ischaemia and/or inflammation and oxidative stress.
- hypoxia-inducible factor 1α (HIF-1α)
- nuclear factor κB (NF-κB)
- reactive oxygen species
- transcription factor
- tumour necrosis factor α (TNF-α)
The family of HIFs (hypoxia-inducible factors) has been widely acknowledged to be critically involved in the cellular response to low oxygen tension. The best-described family member to date, HIF-1, regulates more than 100 genes coding for metabolic enzymes, growth factors and many other factors ensuring an adequate response to hypoxia [1–3].
HIF-1 is composed of an inducible α-subunit (HIF-1α) and a constitutive β-subunit [also termed ARNT (aryl hydrocarbon nuclear translocator)]  (Figure 1). HIF-1α contains an ODD (oxygen-dependent degradation domain) which, when hydroxylated by specific PHDs (prolyl hydroxylases), binds pVHL (the von Hippel–Lindau protein) leading to HIF-1α ubiquitination and degradation by the 26S proteasome. At low oxygen levels, the PHDs lose their activity, which prevents hydroxylation and subsequent pVHL binding [2,4]. This results in HIF-1α stabilization, nuclear translocation, dimerization with ARNT, recruitment of co-activators and binding to HREs (hypoxia-response elements) in the promoters of target genes. Recently, HIF-1α has also been shown to be up-regulated under normoxia in response to growth factors, thrombin, LPS (lipopolysaccharide), angiotensin II, cytokines such as interleukins or TNF-α (tumour necrosis factor-α), or insulin [5–11]. In addition, ROS (reactive oxygen species) have been shown to contribute to many of these responses, indicating that HIF-1 can serve as a redox-sensitive transcription factor. In fact, the NF-κB family of transcription factors has been identified as the prototype of redox-sensitive transcription factors, mediating in particular inflammatory responses in a variety of disorders .
A cross-talk between the NF-κB pathway and the HIF pathway has been documented extensively at the protein level in response to TNF-α, colchicine or HGF (hepatocyte growth factor) [13–16]. Importantly, phosphorylation of IκB (inhibitory κB) and subsequent activation of the NF-κB subunits p50 and p65 (RelA) has been reported to contribute to basal levels of HIF-1α mRNA and protein, and to mediate HIF-1α expression and promoter activity in response to thrombin, H2O2 and even short-term hypoxia [11,17].
Indeed, the NF-κB subunits p50 and p65 have been shown to directly interact with HIF-1α at an NF-κB consensus site in the HIF-1α promoter at −197/−188 bp under these conditions. Mutation of this site abrogated these responses, confirming that this site is functional and implicating an important link between the activation of an inflammatory pathway and the hypoxia-response pathway [11,17]. Indeed, this NF-κB binding element was found to be conserved across different species , suggesting that cross-talk between these two pathways is also highly prevalent among species.
This important interaction has now been confirmed in a study in this issue of the Biochemical Journal by van Uden et al. , suggesting that several members of the NF-κB family can bind to the HIF-1α promoter under basal conditions, as well as under long-term exposure to TNF-α.
This assumption was based on the use of cell lines with either enhanced or reduced levels of the NF-κB subunits RelA (p65), RelB, c-Rel, p50 or p52 and different experimental approaches, including reporter gene assays using the previously described HIF-1α promoter constructs [11,17,19], EMSA (electrophoretic mobility-shift assay), ChIP (chromatin immunoprecipitation), real-time PCR and Western blot analyses. However, none of the experiments was ultimately suitable for proving a direct interaction of any of the subunits with the HIF-1α promoter at the NF-κB-binding site. In fact, the data shown in the present paper  were remarkably variable with respect to the impact of the different NF-κB subunits on the regulation of HIF-1α, thus preventing a clear-cut picture of the relative importance of the different subunits for the regulation of HIF-1α under basal conditions or in response to TNF-α.
Additional experiments were performed to analyse the impact of the NFκB pathway on hypoxic induction of HIF-1α. van Uden et al.  showed that knockdown or genetic deficiency of IKKα (IκB kinase α) and IKKβ depleted HIF-1α protein levels under hypoxic conditions, consistent with a previous study  that demonstrated that IκB is phosphorylated under hypoxia and that a dominant-negative mutant of IκB prevented HIF-1α mRNA and protein induction by hypoxia. However, it remains unclear from which set of data the conclusion was drawn that hypoxia does not impact on HIF-1α mRNA levels under hypoxia, since no data were provided in this regard. Nevertheless, it was concluded that the insensitivity of HIF-1α mRNA levels towards hypoxia was the reason why no difference in NF-κB binding to the HIF-1α promoter was observed under normoxic or hypoxic conditions.
In sharp contrast with this conclusion, however, are several studies which provided evidence that HIF-1α mRNA is indeed up-regulated not only in vivo, but also in vitro, under hypoxic conditions [20–25]. In fact, a recent study clearly showed that within 30 min of exposure to hypoxia, NF-κB subunits are translocated to the nucleus, where they interact with the NF-κB-binding site in the HIF-1α promoter at −197/−188 bp, and thus increase HIF-1α promoter activity and HIF-1α expression . Since the exact experimental conditions used for hypoxic stimulation were not provided in the present paper, it may be that the exposure time to hypoxia was not adequate to detect NF-κB binding to the HIF-1α promoter or modulation of HIF-1α mRNA levels, which appear to be rapid and transient events. Thus a more thorough analysis with regard to the experimental conditions under normoxia and hypoxia, and also to the use of genetically modified cell lines expressing variable levels of NF-κB subunits together with mutational analysis of the HIF-1α promoter, would have been beneficial to provide more detailed insights into the relative interaction of the different NF-κB proteins with the HIF-1α promoter.
Nevertheless, the present study  confirms previous results that NFκB influences basal HIF-1α mRNA levels [17,19]. Given that the levels of HIF-1α mRNA, and therefore basal HIF-1α protein levels, are critical for a cell's readiness to respond to hypoxia, NF-κB could be critical in the degree and speed of HIF activation after a hypoxic insult. This, in turn, might explain why some cells are more sensitive to hypoxia than others. Thus NF-κB might not only be critically linked to the redox-sensitive induction of HIF-1, but may also be involved in the fine-tuning of the hypoxic response, adding further complexity to the relationship between these two important transcription factors.
A further step of complexity in the cross-talk between HIF-1 and NF-κB arises from studies demonstrating that HIF-1 itself contributes to the activation of the NF-κB pathway. It has been previously shown that activation of NF-κB and transcript abundance of p65 are diminished in HIF-1α-deficient murine neutrophils , whereas in mice with increased epithelial levels of HIF-1α, NF-κB activation was induced due to IκB hyperphosphorylation and phosphorylation of Ser276 on p65, thus enhancing p65 nuclear localization and transcriptional activity respectively .
Moreover, NF-κB and HIF-1α appear to share common regulatory pathways, since it has been suggested previously that IKKβ is also a target of PHDs, and hypoxia may release repression of NF-κB activity through decreased PHD-dependent hydroxylation of IKKβ . In addition, IκBα and the NF-κB precursor p105 have been identified as ARD (ankyrin repeat domain)-containing proteins, which are hydroxylated by the asparagine hydroxylase FIH (factor inhibiting HIF) at specific residues within their ARD, although the functional importance of this observation is not sufficiently clear to date .
Taken together, there is no doubt that the cross-talk between NF-κB and HIF-1 is of major importance for disease states associated with low-oxygen tension, such as ischaemia, pulmonary hypertension or cancer on the one hand, as well as for a variety of acute or chronic inflammatory diseases on the other. An improved understanding of the HIF-1–NF-κB cross-talk is challenging due to the engagement of different signalling pathways to HIF-1 from NF-κB, and vice versa. The finding that HIF-1α is a direct target gene of NF-κB under not only non-hypoxic but also certain hypoxic conditions implies a greater importance of NF-κB in HIF-related pathways than was previously thought. Given the indications that HIF-1 can also activate NF-κB, it will be a demanding task to dissect the different signalling states and the potential of feedback signalling loops which may be of great importance for therapeutically targeting a variety of hypoxia- and redox-related disorders.
- © The Authors Journal compilation © 2008 Biochemical Society