| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
with the von Hippel-Lindau Tumor Suppressor Protein1
Department of Pathology and Laboratory Medicine, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania 19104
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
is a transcription factor that plays a critical role in regulating genes involved in erythropoiesis and angiogenesis. Recent evidence indicates that the von Hippel-Lindau tumor suppressor protein (VHL) is part of a ubiquitin ligase complex that promotes the degradation of HIF-1 under normoxic conditions. Under hypoxic conditions, HIF-1 is markedly stabilized. A critical issue in understanding the hypoxic response is the identification of hypoxia-regulated steps. We show here that hypoxia and cobalt treatment modulate the capacity of a HIF-1 fragment comprising residues 531–652 to coimmunoprecipitate with VHL. Hypoxia and cobalt both significantly diminish the interaction, and furthermore, normoxia treatment after hypoxia rapidly normalizes it. This HIF-1 fragment confers hypoxia and cobalt inducibility on a heterologous protein. Significantly, contained within this fragment is a short 27-residue sequence that behaves identically in all respects noted above. Finally, evidence is provided to show that cobalt and hypoxia both induce a posttranslational modification (or loss of one) in HIF-1 that affects its binding to VHL. We propose that dynamic, site-specific interaction of HIF-1 with VHL provides one mechanism by which HIF-1 can be regulated. |
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The enhancers of the above-mentioned genes bind to HIF (1)
. HIF is a heterodimeric protein composed of and ß subunits. The ß subunit is a member of the ARNT family of proteins and includes ARNT (HIF-1ß), ARNT2, and ARNT3. The ß subunit is expressed constitutively, and its protein level is not significantly affected by hypoxia. The subunit is a member of a family of proteins that includes HIF-1, HIF-2 (EPAS-1, HLF, and MOP2), and HIF-3 (3, 4, 5, 6, 7)
. These proteins contain, at their NH2-terminal halves, basic helix-loop-helix DNA-binding and PAS domains that mediate dimerization. At their COOH-terminal halves, they contain transactivation domains. In the case of HIF-1, the most extensively studied subunit, they reside between amino acids 531–575 and 775–826 (3
, 8
, 9)
. In marked contrast to the protein levels of the ß subunits, the protein levels of HIF-1 and other subunits rise markedly on hypoxic stimulation. Thus, the activation of HIF-1 is critically dependent on hypoxia-induced stabilization of HIF-1, which is otherwise degraded rapidly under normoxic conditions. Interestingly, cobaltous ions can mimic hypoxia in inducing HIF-1 (10)
. The mechanism by which cobalt mimics hypoxia is not entirely clear; possibilities include its displacement of iron in an unidentified heme-containing oxygen sensor or its ability to generate free radicals (10, 11, 12)
.
Under normoxic conditions (21% O2), HIF-1 is degraded rapidly by the ubiquitin-proteasome pathway, and an oxygen-dependent degradation domain in HIF-1 comprising amino acids 401–603 confers hypoxia inducibility on a heterologous protein (13, 14, 15)
. The ubiquitin-proteasome pathway involves a cascade of three enzymes, E1, E2, and E3 (16)
. E1 initially transfers ubiquitin to E2 through a thioester bond. E3, in conjunction with E2, then directs the transfer of ubiquitin to target proteins.
Recent evidence indicates that VHL is a component of an E3 ubiquitin ligase complex that ubiquitinates HIF-1. VHL is the product of the gene mutated in von Hippel-Lindau syndrome, a disease characterized by a genetic predisposition to renal cell carcinomas, pheochromocytomas, and hemangioblastomas (17
, 18)
. In certain VHL-deficient cell lines, HIF-1 is constitutively stabilized, and reintroduction of VHL restores hypoxia inducibility of HIF-1 (19)
. HIF-1 coimmunoprecipitates with VHL (19)
. VHL-deficient renal cell carcinoma cell lines contain elevated levels of VEGF mRNA, which has been proposed to account for the prominent neovascularization observed in the VHL-associated tumors (20
, 21)
. Importantly, VHL mutations have also been identified in the majority of sporadic renal cell carcinomas (17)
.
Biochemical studies have independently provided evidence that VHL is a component of a functional E3 complex that also includes elongin B, elongin C, Cul 2, and Rbx 1 (22, 23, 24, 25, 26)
. Indeed, VHL has recently been shown to mediate ubiquitination of HIF-1 in vitro and to bind to a small region of HIF-1 contained within its oxygen-dependent degradation domain (27, 28, 29, 30)
.
These recent findings have provided a framework for understanding HIF-1 under normoxic conditions. What is unclear is the nature of the hypoxia-regulated steps. We show here that VHL interacts with the aforementioned small region of HIF-1 in a hypoxia- and cobalt-sensitive manner. We furthermore provide evidence that a posttranslational alteration of HIF-1 (or loss of one) is responsible for the hypoxia and cobalt sensitivity of this interaction. Thus, stimulus-sensitive interactions between HIF-1 and VHL provide one mechanism by which HIF-1 can be regulated. Our experiments also suggest the existence of alternative mechanisms, as evidenced by a different hypoxia-inducible region of HIF-1 that does not show a regulated interaction with VHL.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Plasmids.
A mammalian expression vector for HIF-1, pEBB-HIF-1, was obtained from Novus Biologicals. A mammalian expression vector for Flag-tagged VHL, pSX/g7F, was a gift of Dr. Richard Klausner (National Cancer Institute, Bethesda, MD) and has been described previously (31)
. pcDNA3-FlagVHL was prepared by subcloning a 0.7-kb PvuII/BamHI fragment of pSX/g7F containing the VHL coding sequence into the Asp718 (blunt)/BamHI site of pcDNA3. The pcDNA3-FlagVHL (P86H) mutant was prepared using a QuikChange mutagenesis kit (Stratagene). pcDNA3-Flag-HIF-1 was constructed by first subcloning the 3.2-kb NcoI/XbaI fragment of pEBB-HIF-1 into the NcoI/XbaI site of pBS-Flag-MEKK1
to yield pBS-Flag-HIF-1. The 3.3-kb BamHI/NotI fragment of the latter was then subcloned into the BamHI/NotI site of pcDNA3. pcDNA3-HA-HIF-1 was constructed by first preparing a pcDNA3 plasmid containing the coding sequence for a HA tag fused to that of HIF-1 residues 2–21 by PCR. The resulting plasmid was then digested with XhoI and XbaI and ligated with the 3.1-kb XhoI/XbaI fragment of pcDNA3-Flag-HIF-1.
pcDNA3-HA-HIF-1-(1–329) was constructed by digesting pcDNA3-HA-HIF-1 with EcoRI, followed by self-ligation. Expression vectors for GAL4-HIF-1-(330–530), GAL4-HIF-1-(531–652), and GAL4-HIF-1-(653–826) were prepared by subcloning the 0.6-kb EcoRI (blunt), 0.4-kb EcoRI/SpeI, and 1.1-kb SpeI fragments of pEBB-HIF-1 into the SmaI, EcoRI/SpeI, and SpeI sites, respectively, of pBXG1.
An expression vector for GAL4-HIF-1-(549–575) was constructed by first amplifying the coding sequence for HIF-1-(549–575) by PCR using pBXG1-HIF-1-(531–652) as a template and primers 5'-CCGGAATTCCCATTTTCTACTCAGGACAC-3' and 5'-CTAGTCTAGACTCGAGCTAACGTAACTGGAAGTCATCATC-3'. The PCR product was then digested with EcoRI and XbaI and subcloned into the EcoRI/XbaI site of pBXG1. An expression vector for GAL4-HIF-1-(775–826) was constructed by first introducing an EcoRI site into the coding sequence of HIF-1 at residue 775 using a QuikChange mutagenesis kit and primers 5'-GGAGCAAAAGACAATTATTGAATTCCCCTCTGATTTAGCATGTAG-3' and 5'-CTACATGCTAAATCAGAGGGGAATTCAATAATTGTCTTTTGCTCC-3'. Subsequently, a 0.8-kb EcoRI/XbaI fragment containing the relevant coding sequence was subcloned into the EcoRI/XbaI site of pBXG1.
(eHRE)3-Luc contains three copies of the HRE from the EPO gene enhancer (32) and was prepared by first subcloning into the HindIII/XbaI site of -40 IFNß-CAT (33) a duplex comprised of the following two oligonucleotides: (a) 5'-AGCTGCCCTACGTGCTGTCTCACACAGCCTGTCTGAGCCCTACGTGCTGTCTCACACAGCCTGTCTGAGCCCTACGTGCTGTCTCACACAGCCTGTCTGA-3'; and (b) 5'-CTAGTCAGACAGGCTGTGTGAGACAGCACGTAGGGCTCAGACAGGCTGTGTGAGACAGCACGTAGGGCTCAGACAGGCTGTGTGAGACAGCACGTAGGGC-3'. The resulting product, (eHRE)3-CAT, was subsequently digested with PvuII/BamHI, and the resulting 106-bp fragment was subcloned into the SmaI/BglII site of pGL3 (Promega) to yield (eHRE)3-Luc. Where appropriate, sequences of recombinant plasmids were confirmed by automated sequencing using a Big Dye Terminator kit (PE Applied Biosystems).
Transient Transfection and Whole Cell Extract Preparation.
The day before transfection, cells were seeded in 12-well or 6-well plates so as to achieve 60–80% confluence. The cells were then transfected with plasmids using Fugene 6 (Roche) according to the manufacturer’s instructions. Cells were briefly washed with PBS/1 mM EDTA, and then lysed by the addition of Buffer A [20 mM Tris (pH 7.6), 150 mM NaCl, 25 mM ß-glycerophosphate, 2 mM EDTA, 2 mM PPi, 10% glycerol, 1% Triton X-100, 1 mM DTT, and 1 mM orthovanadate] containing 10 µg/ml leupeptin and 1 mM phenylmethylsulfonyl fluoride. Whole cell extracts were obtained after clarification of the lysate at 13,000 x g for 10 min at 4°C.
Immunoprecipitation and Immunoblotting.
For immunoprecipitations, whole cell extract was incubated with 10 µl of anti-Flag (M2) agarose (Sigma Chemical Co.) for 2 h at 4°C. The resin was washed four times with Buffer A, eluted with 20 µl of 2x SDS sample buffer, and then heated at 100°C for 3 min. For immunoblotting, samples were subjected to SDS-PAGE and transferred to polyvinylidene difluoride membranes (Immobilon-P; Millipore), and the membranes were then blocked in PBS/0.05% Tween 20/5% nonfat milk. Primary antibodies, used at a 1:1000 dilution, included anti-HIF-1 monoclonal antibody (H167; Novus Biologicals), anti-GAL4 monoclonal antibody (RK5C1; Santa Cruz Biotechnology), anti-GAL4 polyclonal antibody (sc-577; Santa Cruz Biotechnology), and anti-Flag polyclonal antibody (D-8; Santa Cruz Biotechnology). Secondary antibodies included horseradish peroxidase-conjugated antimouse IgG or antirabbit IgG (Santa Cruz Biotechnology). SuperSignal West Pico substrate (Pierce) was used.
Far Western Blotting.
Samples were subjected to SDS-PAGE, transferred to polyvinylidene difluoride membrane, and then washed sequentially in Buffer B [10 mM HEPES (pH 7.6), 60 mM KCl, 1 mM EDTA, and 1 mM ß-mercaptoethanol] containing 8, 4, 2, 1, 0.5, 0.25, or 0 M urea. The membrane was then incubated with 35S-labeled VHL for 18 h at 4°C in Buffer B containing 1% BSA and 0.05% NP40. The membrane was then washed, dried, and subjected to autoradiography. The 35S-labeled VHL was prepared using a TnT in vitro transcription and translation kit (Promega) and pcDNA3-FlagVHL as a template.
Luciferase Assay.
COS-1 cells were cotransfected with 1 µg of (eHRE)3-Luc, 0.5 µg of pRL-TK (Promega), and expression plasmids for GAL4 or GAL4 fusion proteins. After 42 h, the cells were harvested. Luciferase activities were measured using a Dual-Luciferase Reporter Assay System (Promega) and a Wallac LB9507 luminometer.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, endogenous HIF-1 protein levels are inducible by hypoxia or cobalt in both cell lines (compare Lanes 3 and 2 with Lane 1 and compare Lanes 6 and 5 with Lane 4), thus providing evidence for a functional HIF-1 activation pathway in these cells. Note that HIF-1 appears as multiple bands, likely reflecting differentially phosphorylated species (27)
. To determine protein sequences in HIF-1 that might convey hypoxia or cobalt inducibility, we prepared expression constructs for four fragments of HIF-1 (residues 1–329, 330–530, 531–652, and 653–826) that collectively comprise its entire coding sequence (Fig. 1B
; constructs containing residues 549–575 and 775–826 will be discussed later). These fragments were fused to either a HA tag or the DNA-binding domain of the GAL4 protein. After transfection with these expression constructs, some cells were exposed to hypoxia or cobalt and then analyzed by immunoblotting.
|
, protein levels of both GAL4-HIF-1-(330–530) and GAL4-HIF-1-(531–652) are induced by hypoxia or cobalt (Lanes 2 and 4, second and third panels from the top). GAL4-HIF-1-(531–652), it might be noted, migrates on SDS-PAGE as a more rapidly migrating species plus multiple slower migrating species (indicated by the asterisk). Treatment of GAL4-HIF-1-(531–652) with calf intestinal phosphatase abolishes the more slowly migrating species, which is consistent with them representing phosphorylated forms of this protein (data not shown). The protein levels of the other fragments examined, including HA-HIF-1-(1–329), GAL4-HIF-1-(653–826), and GAL4, are not inducible by either stimuli. Thus, two regions of HIF-1 (residues 330–530 and 531–652) can independently confer hypoxia and cobalt inducibility on a heterologous protein. These findings are consistent with previous findings that defined an oxygen-dependent degradation domain (residues 401–603; Ref. 14
), portions of which are included in each of the two regions just indicated.
HIF-1 is degraded under normoxic conditions by the ubiquitin-proteasome pathway, and VHL has been implicated in this process. Recent studies have shown that VHL is part of a complex that can function as an E3 ubiquitin ligase and that VHL can coimmunoprecipitate with HIF-1 (19
, 27, 28, 29, 30)
. To examine whether VHL can coimmunoprecipitate with fragments of HIF-1, COS-1 cells were transfected with expression vectors for the various HIF-1 fragments in the absence or presence of cotransfected Flag-VHL expression vector. Cells were then subjected to hypoxia or cobalt in the presence of the proteasome inhibitor Cbz-LLL. Whole cell extracts were then prepared, Flag-VHL was immunoprecipitated with anti-Flag antibodies, and the immunoprecipitates were examined for the presence of HIF-1 fragments with either anti-HA or anti-GAL4 antibodies.
As shown in Fig. 2
, full-length HIF-1 coimmunoprecipitates with VHL, as expected (Lane 2, top panel). Similar behavior is observed with all four of the HIF-1 fragments examined, namely, HA-HIF-1-(1–329) (Lane 6, top panel), GAL4-HIF-1-(330–530) (Lane 10, top panel), GAL4-HIF-1-(531–652) (Lane 14, top panel), and GAL4-HIF-1-(653–826) (Lane 18, top panel). As negative controls, the various HIF-1 fragments do not appear in mock immunoprecipitations lacking Flag-VHL (Lanes 1, 5, 9, 13, and 17, top panel), nor does GAL4 coimmunoprecipitate with Flag-VHL (Lane 22, top panel).
|
fragments to coimmunoprecipitate with VHL was examined. As shown in Fig. 2
, the coimmunoprecipitation of full-length HIF-1 is weakened by both hypoxia and cobalt (compare Lanes 4 and 3 with Lane 2, top panel). When the various HIF-1 fragments are examined in an analogous manner, only GAL4-HIF-1-(531–652) behaves in a similar way: its coimmunoprecipitation with VHL is weakened by hypoxia and is in fact abolished by cobalt (compare Lanes 16 and 15 with Lane 14, top panel). Similar results were obtained in HeLa cells (data not shown). In reciprocal experiments, Flag-VHL coimmunoprecipitates with GAL4-HIF-1-(531–652), and this coimmunoprecipitation is similarly abolished by cobalt treatment (data not shown). The coimmunoprecipitation of the other HIF-1 fragments with VHL is not significantly altered by either hypoxia or cobalt (Lanes 6–8, 10–12, and 18–20).
The coimmunoprecipitation data raise the possibility of interactions between VHL and multiple HIF-1 fragments. The specificity of these interactions was therefore further examined with the use of a tumor-derived VHL mutant (P86H; Ref. 34
). Accordingly, COS-1 cells were cotransfected with expression vectors for the various HIF-1 fragments along with one for Flag-VHL (P86H). Cells were exposed to Cbz-LLL, whole cell extracts were prepared, and then the Flag-VHL (P86H) was immunoprecipitated with anti-Flag antibodies. The immunoprecipitates were examined for the presence of HIF-1 fragments with either anti-HA or anti-GAL4 antibodies. As shown in Fig. 3
, all of the HIF-1 fragments except for HIF-1-(531–652) (Lane 2, top panel) coimmunoprecipitate with the VHL mutant. Thus, the coimmunoprecipitation of GAL4-HIF-1-(531–652) with VHL is sensitive not only to cobalt and hypoxia but also to a tumor-derived missense mutation in VHL. This mutation has been shown to abolish the interaction of VHL with a similar HIF-1 fragment in vitro (27)
.
|
-VHL interactions might have functional significance by testing which fragments can behave as dominant negative inhibitors of the HIF-1 degradation pathway. Thus, we cotransfected COS-1 cells with expression vectors for HIF-1 fragments and a luciferase reporter construct driven by three copies of the HRE from the human EPO enhancer to measure endogenous HIF-1 activation. As shown in Fig. 4A
, both GAL4-HIF-1-(330–530) and GAL4-HIF-1-(531–652) induce activation of the HRE reporter gene (second and third columns), whereas GAL4-HIF-1-(653–826) does not (fourth column). These data suggest that both GAL4-HIF-1-(330–530) and GAL4-HIF-1-(531–652) can act as dominant negative inhibitors of the HIF-1 degradation pathway. Consistent with this, immunoblotting with a HIF-1 monoclonal antibody demonstrates that overexpression of GAL4-HIF-1-(531–652) increases the endogenous HIF-1 protein level, whereas overexpression of GAL4-HIF-1-(653–826) and GAL4 do not (Fig. 4C
, compare Lane 2 with Lanes 1 and 3). Combined with the data of Fig. 2
, these results suggest that this region of HIF-1-(531–652) can compete with endogenous HIF-1 for the functional site(s) of VHL. Overexpression of HIF-1-(540–580) has similarly been shown to augment activation of a HRE reporter gene (35)
.
|
-(531–652) interacts with VHL in a stimulus-sensitive manner, suggesting that its regulated association with VHL is an important mechanism for its inducibility. The functional involvement of VHL with this fragment was further examined as follows. We have observed that COS-1 cells occasionally display diminished hypoxia and cobalt inducibility of overexpressed GAL4-HIF-1-(531–652) on subculturing. We cotransfected such COS-1 cells with expression constructs for GAL4-HIF-1-(531–652) in the absence or presence of VHL expression vector, and some cells were then exposed to cobalt. Whole cell extracts were prepared and examined by immunoblotting. As shown in Fig. 4D
(compare Lanes 1 and 2 with Lanes 3 and 4), coexpression with VHL markedly enhances the cobalt inducibility of this fusion protein, thus providing evidence of the functional involvement of VHL in the inducibility of this particular HIF-1 fragment.
Only one HIF-1 fragment, encompassing residues 531–652, displays activity in all of the assays detailed above, namely, inducibility by hypoxia and cobalt, coimmunoprecipitation with VHL in a hypoxia- and cobalt-sensitive manner, and the ability to behave in a dominant negative manner to activate a HRE reporter gene. Contained within this fragment is a sequence that is highly conserved not only between different HIF- isoforms but also between HIF-1 from different species (Fig. 5A)
. In fact, this sequence is by far the most highly conserved sequence in the COOH-terminal halves of different HIF- isoforms. Because strongly conserved sequences are often functionally important, we examined the behavior of GAL4 fused to a 27-residue sequence encompassing this region (residues 549–575; Fig. 1B
, second to last construct).
|
, Lane 3, top panel). Equally notable, this coimmunoprecipitation is diminished by hypoxia and abolished by cobalt (compare Lanes 4 and 2 with Lane 3, top panel), identical in behavior to the larger HIF-1 fragment (531–652) encompassing it (Fig. 2)
. The 27-amino acid sequence is contained within the first transcriptional activation domain of HIF-1, raising the possibility that the behavior of this sequence may be a general one of HIF-1 transcriptional activation domains. We therefore also examined the behavior of the second such domain of HIF-1, which encompasses residues 775–826 (Fig. 1B
, last construct). In contrast to the first domain, however, the second does not coimmunoprecipitate with VHL (Fig. 5B
, Lane 7, top panel). Therefore, the stimulus sensitivity of VHL coimmunoprecipitation is a distinctive property of the first transcriptional activation domain. In addition, protein levels of GAL4-HIF-1-(549–575) but not those of GAL4-HIF-1-(775–826) are inducible by hypoxia and cobalt (Fig. 5C)
.
Under normoxic conditions, the HIF-1 protein has a half-life of less than 5 min (36)
. Accordingly, when hypoxic cells are returned to normoxic conditions, HIF-1 protein levels are virtually undetectable after 10 min of normoxic reexposure (36
, 37)
. We examined the binding of GAL4-HIF-1-(549–575) to VHL under these conditions. COS-1 cells were cotransfected with expression vectors for GAL4-HIF-1-(549–575) and Flag-VHL. Subsequently, some cells were maintained under normoxic or hypoxic conditions, whereas others were initially subjected to hypoxic conditions and then returned to normoxic conditions for either 10 or 20 min. As shown in Fig. 6A
, GAL4-HIF-1-(549–575) coimmunoprecipitates with VHL, and this coimmunoprecipitation is weakened by hypoxia, as expected (compare Lanes 1 and 2, top panel). The notable result is that after hypoxia followed by 10 or 20 min of normoxia (Lanes 3 and 4, respectively, top panel), the capacity of GAL4-HIF-1-(549–575) to coimmunoprecipitate with VHL is restored to levels virtually identical to that seen under normoxic conditions. Thus, on normoxia reexposure, the interaction of this segment of HIF-1 with VHL is rapidly normalized. Similar results were observed with GAL4-HIF-1-(531–652) (Fig. 6B)
.
|
-(549–575) and HIF-1-(531–652) coimmunoprecipitate with VHL in a hypoxia- and cobalt-sensitive manner. In principle, this sensitivity could be due to a posttranslational alteration of HIF-1, a posttranslational alteration of VHL, or both. To distinguish between these possibilities, COS-1 cells were transfected separately with expression vectors for either GAL4-HIF-1-(531–652) or Flag-VHL, and the cells were then treated with either cobalt or hypoxia or mock-treated in the presence of Cbz-LLL. Whole cell extracts were prepared, and the extracts containing GAL4-HIF-1-(531–652) from cobalt-, hypoxia-, or mock-treated cells were mixed with extracts containing Flag-VHL from cobalt-, hypoxia-, or mock-treated cells, as indicated (Fig. 7A)
. Flag-VHL was then immunoprecipitated and examined for the presence of GAL4-HIF-1-(531–652) by immunoblotting.
|
, GAL4-HIF-1-(531–652) from mock-treated cells coimmunoprecipitates with Flag-VHL from mock-treated cells (Lane 1), whereas GAL4-HIF-1-(531–652) from cobalt-treated cells does not coimmunoprecipitate with Flag-VHL from cobalt-treated cells (Lane 4). Both of these results are not unexpected, given the results of Fig. 2
. The striking finding is that GAL4-HIF-1-(531–652) from cobalt-treated cells does not coimmunoprecipitate with Flag-VHL from mock-treated cells (Lane 2). In marked contrast, GAL4-HIF-1-(531–652) from mock-treated cells does coimmunoprecipitate with Flag-VHL from cobalt-treated cells (Lane 3). Thus, the ability of GAL4-HIF-1-(531–652) to coimmunoprecipitate with Flag-VHL in this assay is determined by whether or not the cells expressing GAL4-HIF-1-(531–652) had been exposed to cobalt.
Experiments performed with hypoxia treatment reveal similar findings. Thus, GAL4-HIF-1-(531–652) obtained from hypoxic cells coimmunoprecipitates more weakly with Flag-VHL than that obtained from normoxic cells, regardless of the source of the VHL (compare Lane 6 with Lane 5 and compare Lane 8 with Lane 7). Thus, as with cobalt, the degree to which GAL4-HIF-1-(531–652) coimmunoprecipitates with VHL is determined by whether or not the cells expressing GAL4-HIF-1-(531–652) were made hypoxic. Finally, analogous experiments with GAL4-HIF-1-(549–575) reveal essentially identical findings (Fig. 7B)
. In particular, GAL4-HIF-1-(549–575) obtained from hypoxia-treated cells is weakened in its ability to coimmunoprecipitate with Flag-VHL (compare Lane 6 with Lane 5), whereas that obtained from cobalt-treated cells is virtually inactive in this regard (compare Lane 2 with Lane 1). The data collectively suggest that a posttranslational alteration of HIF-1-(549–575), but not one of VHL, is responsible for this altered interaction.
The posttranslational alteration of HIF-1-(549–575) could, in turn, be due to either a posttranslational modification (or loss of one) of HIF-1-(549–575) or the association/dissociation of a third factor that mediates the binding of HIF-1-(549–575) to VHL. To distinguish between these possibilities, the following far Western analysis was performed. COS-1 cells were transfected with an expression vector for GAL4-HIF-1-(549–575), and some cells were then exposed to either cobalt or hypoxia. Cells were lysed, and GAL4-HIF-1-(549–575) was immunoprecipitated with anti-GAL4 antibodies. The immunoprecipitates were subjected to SDS-PAGE, transferred to membranes, renatured, and then incubated with in vitro translated, 35S-labeled VHL. As shown in Fig. 8
, GAL4-HIF-1-(549–575) from normoxic cells shows readily detectable VHL binding (top panel, Lanes 2 and 5). Strikingly, that from either cobalt-treated or hypoxic cells displays significantly weakened binding (top panel, Lanes 3 and 6, respectively). This therefore provides evidence that cobalt or hypoxia treatment induces a posttranslational modification (or loss of one) of HIF-1-(549–575).
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
regulation is determining the step at which its VHL-mediated degradation is affected by hypoxia. Here, we identify a 27-residue segment of HIF-1 that interacts with VHL (Fig. 5)
. This 27-residue segment is contained within the larger 213-residue oxygen-dependent degradation domain (14)
of HIF-1 that confers hypoxia inducibility on a heterologous protein. The importance of this region was highlighted by other recent studies, which indicate that block mutations within this region result in defective interaction with VHL, decreased ubiquitination under normoxic conditions, and normoxic stabilization of HIF-1 (30
, 35)
. Importantly, our data now indicate that not only does this short segment of HIF-1 display an altered interaction with VHL on hypoxia but that subsequent reexposure to normoxia rapidly normalizes this interaction. We propose that dynamic, stimulus-modulated interaction of HIF-1 with VHL provides a mechanism by which HIF-1 degradation can be regulated.
The results provide evidence that hypoxia and cobalt induce a posttranslational modification in HIF-1 and raise the possibility that this modification might be the same with both stimuli. Thus, whereas there is evidence that hypoxia and cobalt are not completely identical in the manner in which they activate HIF-1 (1
, 19)
, these data indicate that an alteration in the interaction of HIF-1-(549–575) with VHL is a common end result for both of these stimuli. That being said, the hypoxia-modulated interaction revealed in the coimmunoprecipitation experiments does not necessarily imply that HIF-1 dissociates from VHL in vivo. For example, HIF-1 might remain associated with VHL in vivo after hypoxia treatment but partially dissociate from VHL during the washing steps of the coimmunoprecipitation assay. Indeed, there is evidence that VHL remains associated with HIF-1 after hypoxia activation: VHL can be detected in HIF-1 DNA-binding complexes obtained from hypoxia-treated cells (19)
. In the context of full-length HIF-1, the weakening of the interaction of HIF-1-(531–652) with VHL might be compensated for by interactions between other portions of HIF-1 (e.g., 1–329, 330–530, or 653–826) and sites on VHL distinct from that contacted by HIF-1-(531–652). It remains to be determined which of these contacts are physiologically relevant and which might be due to overexpression conditions. The latter possibility is highlighted by the observation that the HIF-1 fragments encompassing residues 1–329, 330–530, and 653–826 all coimmunoprecipitate with a tumor-derived VHL mutant (P86H; Fig. 3
).
Whereas the present data provide evidence for a stimulus-sensitive interaction between HIF-1 and VHL, there is evidence that other mechanisms may be operative as well. For example, HIF-1-(330–530) is inducible by hypoxia or cobalt, yet its association with VHL is not modulated by these agents (Fig. 2)
. Certainly VHL ubiquitination activity might also be altered, a possibility that is currently being explored. It also is possible that VHL-independent mechanisms of HIF-1 degradation exist. For example, Mdm2, an E3 distinct from VHL, can mediate HIF-1 degradation by the ubiquitin-proteasome pathway (38)
.
The posttranslational modification of HIF-1 that occurs (or is diminished) under hypoxia remains to be determined. HIF-1 is a phosphoprotein, and both protein kinase and phosphatase inhibitors can inhibit the activation of HIF-1 (39)
. This therefore raises the possibility that phosphorylation of HIF-1 might be this alteration. Intriguingly, mutation of two potential phosphoacceptor residues, Ser551 and Thr552, results in constitutive expression of certain HIF-1 constructs (15)
. However, protein levels of other HIF-1 constructs with this mutation remain hypoxia inducible (15)
, and mutation of these as well as the two other potential phosphoacceptor residues in the amino acid segment 549–575 (Thr555 and Tyr565) does not affect hypoxia or cobalt inducibility of the transcriptional activity of a GAL4/HIF-1-(530–652)/ARNT transactivation domain fusion protein (9)
. In addition, a point mutation of T544A and a block mutation of the PEST domain (S585/586/587/589/592/594A), which are residues in the vicinity of this region, have no effect on the cobalt-sensitive coimmunoprecipitation of GAL4-HIF-1-(531–652) with VHL (data not shown). Thus, the precise role of phosphorylation in the hypoxic regulation of HIF-1 remains to be seen.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 Supported by Research Grant C9901 from the W. W. Smith Charitable Trust, a Pilot Project Grant from the University of Pennsylvania Cancer Center, an award from the University of Pennsylvania Research Foundation, and NIH Grant DK55672. 
2 To whom requests for reprints should be addressed, at the Department of Pathology and Laboratory Medicine, University of Pennsylvania School of Medicine, 605 Stellar-Chance Building, 422 Curie Boulevard, Philadelphia, PA 19104. Phone: (215) 898-4701; Fax: (215) 573-2272; E-mail: franklee{at}mail.med.upenn.edu
3 The abbreviations used are: EPO, erythropoietin; Cbz-LLL, N-Cbz-L-Leu-L-Leu-L-norvalinal; HA, hemagglutinin; HIF, hypoxia-inducible factor; VEGF, vascular endothelial growth factor; VHL, von Hippel-Lindau tumor suppressor protein; ARNT, aryl hydrocarbon nuclear translocator; HRE, hypoxia response element.
Received 8/16/00. Accepted 3/19/01.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
in response to hypoxia: their stabilization and redox signal-induced interaction with CBP/p300. EMBO J., 18: 1905-1914, 1999.[Medline]
-class hypoxia inducible factor subunit, HIF3. Gene Expr., 7: 205-213, 1998.[Medline]
. Modulation of transcriptional activity by oxygen tension. J. Biol. Chem., 272: 19253-19260, 1997. subunit. J. Biol. Chem., 272: 11205-11214, 1997. during hypoxia: a mechanism of O2 sensing. J. Biol. Chem., 275: 25130-25138, 2000. (HIF-1) protein is rapidly degraded by the ubiquitin-proteasome system under normoxic conditions. Its stabilization by hypoxia depends on redox-induced changes. J. Biol. Chem., 272: 22642-22647, 1997. is mediated by an O2-dependent degradation domain via the ubiquitin-proteasome pathway. Proc. Natl. Acad. Sci. USA, 95: 7987-7992, 1998. protein expression is controlled by oxygen-regulated ubiquitination that is disrupted by deletions and missense mutations. Proc. Natl. Acad. Sci. USA, 97: 4748-4753, 2000. binding and ubiquitylation by the von Hippel-Lindau tumor suppressor protein. J. Biol. Chem., 275: 25733-25741, 2000. ubiquitination by a reconstituted von Hippel-Lindau (VHL) tumor suppressor complex. Proc. Natl. Acad. Sci. USA, 97: 10430-10435, 2000. by the von Hippel-Lindau tumor suppressor protein. EMBO J., 19: 4298-4309, 2000.[Medline]
(HIF-) proteins. Biochem. Biophys. Res. Commun., 260: 557-561, 1999.[Medline]
subunit. J. Biol. Chem., 271: 32253-32259, 1996.. Genes Dev., 14: 34-44, 2000.This article has been cited by other articles:
|
|
 |
|
  P. W. Furlow, M. J. Percy, S. Sutherland, C. Bierl, M. F. McMullin, S. R. Master, T. R. J. Lappin, and F. S. Lee Erythrocytosis-associated HIF-2{alpha} Mutations Demonstrate a Critical Role for Residues C-terminal to the Hydroxylacceptor Proline J. Biol. Chem., April 3, 2009; 284(14): 9050 - 9058. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Song and F. S. Lee A Role for IOP1 in Mammalian Cytosolic Iron-Sulfur Protein Biogenesis J. Biol. Chem., April 4, 2008; 283(14): 9231 - 9238. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. R. Walmsley, N. N. McGovern, M. K. B. Whyte, and E. R. Chilvers The HIF/VHL Pathway: From Oxygen Sensing to Innate Immunity Am. J. Respir. Cell Mol. Biol., March 1, 2008; 38(3): 251 - 255. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. J. Percy, P. W. Furlow, P. A. Beer, T. R. J. Lappin, M. F. McMullin, and F. S. Lee A novel erythrocytosis-associated PHD2 mutation suggests the location of a HIF binding groove Blood, September 15, 2007; 110(6): 2193 - 2196. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Chen, K. Uchida, A. Endler, and F. Shibasaki Mammalian Tumor Suppressor Int6 Specifically Targets Hypoxia Inducible Factor 2{alpha} for Degradation by Hypoxia- and pVHL-independent Regulation J. Biol. Chem., April 27, 2007; 282(17): 12707 - 12716. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. J. Percy, Q. Zhao, A. Flores, C. Harrison, T. R. J. Lappin, P. H. Maxwell, M. F. McMullin, and F. S. Lee From the Cover: A family with erythrocytosis establishes a role for prolyl hydroxylase domain protein 2 in oxygen homeostasis PNAS, January 17, 2006; 103(3): 654 - 659. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Siddiq, I. A. Ayoub, J. C. Chavez, L. Aminova, S. Shah, J. C. LaManna, S. M. Patton, J. R. Connor, R. A. Cherny, I. Volitakis, et al. Hypoxia-inducible Factor Prolyl 4-Hydroxylase Inhibition: A TARGET FOR NEUROPROTECTION IN THE CENTRAL NERVOUS SYSTEM J. Biol. Chem., December 16, 2005; 280(50): 41732 - 41743. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. M. Li, B. P. Zhou, J. Deng, Y. Pan, N. Hay, and M.-C. Hung A Hypoxia-Independent Hypoxia-Inducible Factor-1 Activation Pathway Induced by Phosphatidylinositol-3 Kinase/Akt in HER2 Overexpressing Cells Cancer Res., April 15, 2005; 65(8): 3257 - 3263. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. R. Walmsley, C. Print, N. Farahi, C. Peyssonnaux, R. S. Johnson, T. Cramer, A. Sobolewski, A. M. Condliffe, A. S. Cowburn, N. Johnson, et al. Hypoxia-induced neutrophil survival is mediated by HIF-1{alpha}-dependent NF-{kappa}B activity J. Exp. Med., January 3, 2005; 201(1): 105 - 115. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Hopfl, O. Ogunshola, and M. Gassmann HIFs and tumors--causes and consequences Am J Physiol Regulatory Integrative Comp Physiol, April 1, 2004; 286(4): R608 - R623. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C.-J. Hu, L.-Y. Wang, L. A. Chodosh, B. Keith, and M. C. Simon Differential Roles of Hypoxia-Inducible Factor 1{alpha} (HIF-1{alpha}) and HIF-2{alpha} in Hypoxic Gene Regulation Mol. Cell. Biol., December 15, 2003; 23(24): 9361 - 9374. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. G. Kaelin Jr. The von Hippel-Lindau Gene, Kidney Cancer, and Oxygen Sensing J. Am. Soc. Nephrol., November 1, 2003; 14(11): 2703 - 2711. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Pereira, X. Zheng, J. L. Ruas, K. Tanimoto, and L. Poellinger Identification of Residues Critical for Regulation of Protein Stability and the Transactivation Function of the Hypoxia-inducible Factor-1alpha by the von Hippel-Lindau Tumor Suppressor Gene Product J. Biol. Chem., February 21, 2003; 278(9): 6816 - 6823. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. H. Liu, A. Kirschenbaum, M. Lu, S. Yao, A. Dosoretz, J. F. Holland, and A. C. Levine Prostaglandin E2 Induces Hypoxia-inducible Factor-1alpha Stabilization and Nuclear Localization in a Human Prostate Cancer Cell Line J. Biol. Chem., December 13, 2002; 277(51): 50081 - 50086. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Huang, Q. Zhao, S. M. Mooney, and F. S. Lee Sequence Determinants in Hypoxia-inducible Factor-1alpha for Hydroxylation by the Prolyl Hydroxylases PHD1, PHD2, and PHD3 J. Biol. Chem., October 11, 2002; 277(42): 39792 - 39800. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. A. Seta, Z. Spicer, Y. Yuan, G. Lu, and D. E. Millhorn Responding to Hypoxia: Lessons From a Model Cell Line Sci. Signal., August 20, 2002; 2002(146): re11 - re11. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Willam, N. Masson, Y.-M. Tian, S. A. Mahmood, M. I. Wilson, R. Bicknell, K.-U. Eckardt, P. H. Maxwell, P. J. Ratcliffe, and C. W. Pugh Peptide blockade of HIFalpha degradation modulates cellular metabolism and angiogenesis PNAS, August 6, 2002; 99(16): 10423 - 10428. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. H. WENGER Cellular adaptation to hypoxia: O2-sensing protein hydroxylases, hypoxia-inducible transcription factors, and O2-regulated gene expression FASEB J, August 1, 2002; 16(10): 1151 - 1162. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. C. Mahon, K. Hirota, and G. L. Semenza FIH-1: a novel protein that interacts with HIF-1alpha and VHL to mediate repression of HIF-1 transcriptional activity Genes & Dev., October 15, 2001; 15(20): 2675 - 2686. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Yu, S. B. White, Q. Zhao, and F. S. Lee HIF-1alpha binding to VHL is regulated by stimulus-sensitive proline hydroxylation PNAS, August 14, 2001; 98(17): 9630 - 9635. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. J. Klasa, A. F. List, and B. D. Cheson Rational Approaches to Design of Therapeutics Targeting Molecular Markers Hematology, January 1, 2001; 2001(1): 443 - 462. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
