This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Zheng, X. Articles by Pereira, T. Search for Related Content PubMed PubMed Citation Articles by Zheng, X. Articles by Pereira, T.

Cell-Type-Specific Regulation of Degradation of Hypoxia-Inducible Factor 1: Role of Subcellular Compartmentalization

Department of Cell and Molecular Biology,¹ Microbiology and Tumor Biology Center, Karolinska Institutet, S-171 77 Stockholm, Sweden,² Department of Cardiology, First Affiliated Hospital, China Medical University, Shenyang 110001, People's Republic of China³

Received 19 November 2005/ Returned for modification 15 December 2005/ Accepted 24 March 2006

	ABSTRACT

Top Abstract Introduction Materials and Methods Results Discussion References

 
The hypoxia-inducible factor-1 (HIF-1) is a transcription factor that mediates adaptive cellular responses to decreased oxygen availability (hypoxia). At normoxia, HIF-1 is targeted by the von Hippel-Lindau tumor suppressor protein (pVHL) for degradation by the ubiquitin-proteasome pathway. In the present study we have observed distinct cell-type-specific differences in the ability of various tested pVHL-interacting subfragments to stabilize HIF-1 and unmask its function at normoxia. These properties correlated with differences in subcellular compartmentalization and degradation of HIF-1. We observed that the absence or presence of nuclear localization or export signals differently affected the ability of a minimal HIF-1 peptide spanning residues 559 to 573 of mouse HIF-1 to stabilize endogenous HIF and induce HIF-driven reporter gene activity in two different cell types (primary mouse endothelial and HepG2 hepatoma cells). Degradation of HIF-1 occurred mainly in the cytoplasm of HepG2 cells, whereas it occurs with equal efficiency in nuclear and cytoplasmic compartments of primary endothelial cells. Consistent with these observations, green fluorescent protein-HIF-1 is differently distributed during hypoxia and reoxygenation in hepatoma and endothelial cells. Consequently, we propose that differential compartmentalization of degradation of HIF-1 and the subcellular distribution of HIF-1 may account for cell-type-specific differences in stabilizing HIF-1 protein levels under hypoxic conditions.

	INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

 
The transcription factor hypoxia-inducible factor 1 (HIF-1) plays a key role in acute and long-term adaptive physiological cellular responses to decreased oxygen availability (hypoxia) by activating the transcription of genes modulating energy metabolism, iron transport, vasomotor control, and erythropoiesis, as well as genes encoding angiogenic factors such as vascular endothelial growth factor (51). This gene activation response contributes to the survival of ischemic or hypoxic cells. HIF-1 is also critical for embryonal vascular development (7, 49). HIF-1 is a heterodimer of two basic-helix-loop-helix/PAS factors: aryl hydrocarbon receptor nuclear translocator (ARNT) and HIF-1. ARNT is expressed constitutively, whereas oxygen levels regulate HIF-1 expression. At normoxia (21% O₂) HIF-1 is recognized as a substrate of the ubiquitin-proteasome pathway (22, 28, 29, 50) by the von-Hippel Lindau tumor suppressor gene product (pVHL), a key component of a complex with E3-ubiquitin ligase activity (10, 31, 41, 54). Interaction between HIF-1 and pVHL has been proposed to occur in the nucleus, where the protein is ubiquitylated before it is exported to the cytoplasm in order to be degraded by the proteasome (19).

Hydroxylation of HIF-1 residues Pro⁴⁰² and Pro⁵⁶⁴ by soluble prolyl hydroxylases mediates interaction of HIF-1 with pVHL (5, 15, 24, 25, 38, 43, 62). These enzymes are 2-oxoglutarate-dependent dioxygenases that require molecular oxygen as a cosubstrate. It has been proposed that the prolyl hydroxylases are inactivated at hypoxia (1% O₂), resulting in abrogation of pVHL-HIF-1 interaction and subsequent stabilization of HIF-1 protein (5, 15). After stabilization HIF-1 translocates to the nucleus where it heterodimerizes with ARNT to activate transcription of target genes (27).

In the present study we have investigated various strategies to stabilize HIF and activate its function under normoxic conditions. Among other effects, this mode of activation may result in inducing angiogenic responses. Consistent with this model, it has been reported that exposure of cells to pVHL-interacting HIF-1 peptides containing a nuclear localization signal (NLS) increases HIF function at normoxia (58). Here we have observed distinct cell-type-specific differences in the ability of pVHL-interacting subfragments of HIF-1 to unmask HIF function at normoxia. Consistent with these observations, we detected striking cell-type-specific differences in subcellular compartmentalization and degradation of HIF-1. The absence or presence of an NLS or a nuclear export signal (NES) motif differently affected the ability of a minimal HIF-1 peptide spanning residues 559 to 573 of mouse HIF-1 (mHIF-1) to stabilize endogenous HIF and induce hypoxia response element (HRE)-driven reporter gene activity in two different cell types (hepatoma and primary endothelial cells). Degradation of HIF-1 occurred predominantly in one compartment in HepG2 cells, whereas it occurred with equal efficiency in nuclear and cytosolic compartments in primary endothelial cells. In agreement with these data, green fluorescent protein (GFP)-HIF-1 is differently distributed during hypoxia and reoxygenation in hepatoma and endothelial cells.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

 
Plasmids and fusion proteins. Plasmid pFLAG-GAL4/mHIF-1(546-574) spanning residues 546 to 574 of mHIF-1 and plasmids spanning residues 556 to 573, 557 to 573, 558 to 573, 559 to 573, 560 to 573, 561 to 573, 556 to 571, and 560 to 571 of mHIF-1 were constructed by generating PCR fragments from mHIF-1 using primer pairs carrying restriction sites for EcoRI in the 5' end and stop codon/BamHI in the 3' end. The PCR fragments were inserted in frame into EcoRI-BamHI-digested pFLAG-GAL4. pFLAG/mHIF-1(559-573) was generated by removing a HindIII fragment corresponding to the GAL4 DNA-binding domain from pFLAG-GAL4/mHIF-1. The pFLAG-GFP/mHIF-1(559-573) construct was generated by inserting a HindIII-EcoRV GFP fragment from pCMX-SAH/Y145F (27) into HindIII-NotI-digested pFLAG/mHIF-1(559-573), with the NotI restriction site being previously filled in with the Klenow fragment of DNA polymerase I. Plasmid pFLAG-GFP-NLS/mHIF-1(559-573) was constructed by generating PCR fragments containing the NLS sequence from pCMX/NLS (a generous gift from Keiji Tanimoto, Hiroshima University, Japan) using primer pairs carrying restriction sites for SalI in the 5' end and EcoRI in the 3' end. The PCR fragments were inserted in frame into SalI-EcoRI-digested pFLAG-GFP/mHIF-1(559-573). pFLAG-GFP-NES/mHIF-1(559-573) or pFLAG-GFP-NES(L37A/L40A)/mHIF-1(559-573) was generated by inserting annealed oligonucleotides carrying the NES or NES(L37A/L40A) sequence into SalI-EcoRI-digested pFLAG-GFP/mHIF-1(559-573). pFLAG-GST-GFP/NLS[NES, NLS(K128T/K129T)]/mHIF-1(559-573) or pFLAG-GST-NES(L37A/L40A)/mHIF-1(559-573) was constructed by inserting the glutathione transferase (GST) cDNA from pCMX/GST-GFP (a kind gift from Arunas Kazlauskas) into HindIII-Acc65I-digested pFLAG-GFP-NLS[NES, NLS(K128/K129T)]/mHIF-1(559-573) or pFLAG-GFP-NES(L37A/L40A)/mHIF-1(559-573). The inserts generated by PCR were completely sequenced using a DYEnamic sequencing kit (Amersham Biosciences). Amino acid mutations were introduced using a QuikChange site-directed mutagenesis kit (Stratagene) according to the instructions of the manufacturer, and positive mutants were screened by sequencing. The proteasome sensor vector pZsProSensor-1 was obtained from BD Biosciences Clontech. pFLAG-mHIF-2 was a kind gift from Sallyann O'Brien. pFLAG-CMV-2, pFLAG-GAL4, pT81/HRE-luc, pCMX-VHL, pFLAG-mHIF-1, pGEMT/FLAG-mHIF1, pGFP-HIF-1(1-74), pGFP-HIF-1(178390), and pGFP-HIF-1(576-826) have been described previously (27, 48, 54).

Cell culture and transient transfections. HepG2 and HeLa cells were routinely maintained in Dulbecco's minimal essential medium containing 10% fetal calf serum, 50 IU/ml penicillin, 50 µg/ml streptomycin sulfate, 2 mM L-glutamine and nonessential amino acids. F-12 (HAM) medium supplemented with 10% fetal calf serum, 50 IU/ml penicillin, and 50 µg/ml streptomycin sulfate was used to maintain mouse brain endothelial cells (MBECs). MBECs were used up to passage 10 (positive for endothelial markers). Bovine capillary endothelial cells immortalized by stable expression of human telomerase catalytic subunit (hTERT⁺-BCE) cells were cultured as previously described (55). Medium and other products for cell culture were purchased from Invitrogen. Cells were cultured at 37°C. Oxygen tensions in the incubator were either 140 mm Hg (21% O₂ [vol/vol]; normoxia) or 7 mm Hg (1% O₂ [vol/vol]; hypoxia). Transient transfections were performed using Fugene6 (Roche Molecular Biochemicals) according to the manufacturer's instructions. After transfection, cells were cultured at normoxia or hypoxia for different times as indicated in the figure legends. Empty expression vector was used to make the total amount of DNA the same in all samples. In HRE-driven luciferase reporter gene assays, 0.5 µg of pT81/HRE-luc encoding the HRE-driven luciferase reporter gene was transfected together with 0.125, 0.25, 0.5, or 0.75 µg of pFLAG-GAL4/mHIF-1(546-574) or corresponding mutants. In reporter gene assays analyzing the activity of GST-GFP/mHIF-1(559-573) or mutants, 0.5 µg of pT81/HRE-luc and 1.0, 2.0, or 3.0 µg of expression plasmids encoding GST-GFP fusions were cotransfected into the cells. Cells were cultured under normoxic conditions for 44 h. Luciferase activity was measured and normalized by protein concentration. In the analysis of the subcellular distribution of GFP-HIF-1, GST-GFP/mHIF-1(559-573), and corresponding mutants, cells were transfected with 1.0 µg of expression plasmid. For analysis of ZsProSensor-1 expression, cells were transfected with 1.0 µg of pZsProSensor-1 and 2.5 µg of pFLAG-CMV-2 or pFLAG-mHIF-1.

Cellular extracts. To prepare whole-cell extracts, cells were washed with phosphate-buffered saline (PBS), collected by centrifugation, and lysed in high-salt buffer (50 mM Tris-HCl, pH 7.4, 500 mM NaCl, 1% NP-40, and 20% glycerol) supplemented with 0.5 mM phenylmethylsulfonyl fluoride (PMSF), 5 mM ß-mercaptoethanol, and protease inhibitor mix (Complete-Mini; Roche Biochemicals). Lysates were cleared by centrifugation for 30 min at 14,000 rpm at 4°C. To prepare nuclear and cytosolic extracts, cells were collected, washed, and resuspended in hypotonic buffer (10 mM KCl, 1.5 mM MgCl₂, 0.2 mM PMSF, 0.5 mM dithiothreitol, and protease inhibitor mix). The swollen cells were homogenized, and nuclei were pelleted. The cytosolic fraction was collected and centrifuged, and the supernatant was saved. The nuclei were resuspended in a low-salt buffer (20 mM Tris, pH 7.4, 25% glycerol, 1.5 mM MgCl₂, 0.2 mM EDTA, 0.02 M KCl), and gentle drop-wise addition of a high-salt buffer (20 mM Tris, pH 7.4, 25% glycerol, 1.5 mM MgCl₂, 0.2 mM EDTA, 0.6 M KCl) to a final concentration of 0.42 M KCl allowed the release of soluble proteins from the nuclei. Following extraction and centrifugation, the nuclear extracts were collected. Both nuclear and cytosolic extracts were dialyzed in dialysis buffer (20 mM Tris, pH 7.4, 20% glycerol, 0.2 mM EDTA, 100 mM KCl, 0.2 mM PMSF, and 0.5 mM dithiothreitol). Protein concentration was determined by a colorimetric method (Bio-Rad).

Immunoprecipitation assays. In vitro immunoprecipitation assays were performed using proteins translated in rabbit reticulocyte lysate (Promega) in the absence or presence of [³⁵S]methionine. Translated FLAG-tagged proteins were incubated for 1.5 h under rotation with 50 µl of a 50% protein G-Sepharose bead slurry (Amersham Biosciences) preincubated with anti-FLAG M2 monoclonal antibody (F3165; Sigma) at room temperature. After three washes with TBS (50 mM Tris, pH 7.4, 150 mM NaCl) supplemented with 0.5% Triton X-100, the beads were incubated with [³⁵S]methionine-labeled pVHL for 1 h at room temperature under rotation. After incubation the beads were washed, and precipitated proteins were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), followed by autoradiography. Endogenous HIF-1 was immunoprecipitated from cytosolic extracts that were incubated for 1.5 h with protein G-Sepharose beads preincubated with anti-HIF-1 monoclonal antibodies (NB100-123 and NB 100-296; Novus-Biologicals) at room temperature. After three washes with TBS containing 0.5% Triton X-100, precipitated proteins were analyzed by SDS-PAGE, followed by immunoblotting.

Immunoblotting assays. Immunoprecipitated proteins, proteins from whole-cell extracts or from nuclear extracts, were separated by SDS-PAGE and blotted onto nitrocellulose filters. Blocking was performed at room temperature for 2 h in TBS with 5% nonfat milk, followed by incubation for 1 h at room temperature with anti-FLAG M2, anti-HIF-1, anti-Nrf2 (sc-13032; Santa Cruz Biotechnology), anti-YY1 (sc-7341; Santa Cruz Biotechnology), antipaxillin (05-417; Upstate Biotechnology), or anti-HIF-2 (NB100-132; Novus-Biologicals) antibodies in TBS with 1% nonfat milk. After several washes with TBS containing 0.5% Tween 20, the filters were incubated with anti-mouse immunoglobulin G-horseradish peroxidase conjugate (Amersham Biosciences) in TBS with 1% nonfat milk. Following several washes, proteins were visualized using enhanced chemiluminescence (Amersham Biosciences) according to the manufacturer's recommendations.

Visualization of intracellular localization of GFP fusions and ZsProSensor-1 protein. Cells were grown on coverslips in culture dishes. After 36 h of transfection, cells were exposed to either normoxia or hypoxia for 8 h and fixed in 4% paraformaldehyde in PBS. Coverslips were then mounted on glass slides for observation. Samples were scanned with a Zeiss LSM 510 laser scanning confocal device coupled to an Axiovert 100 M microscope (Carl Zeiss). Quantitative evaluations were performed as described before (27, 61). Briefly, green fluorescent cells were classified into four categories according to their intracellular localization: N, exclusively nuclear localization; N>C, predominantly nuclear localization; N=C, equal distribution in both cytoplasm and nucleus; and N<C, predominantly cytoplasmic localization. On average, 250 cells were evaluated on each coverslip.

Fluorescence-activated cell sorter (FACS) analysis. Single-cell suspensions were fixed with 4% paraformaldehyde in PBS, washed twice with PBS, and analyzed using a FACSCalibur flow cytometer with CellQuest software (Becton Dickinson). Data on 10,000 cells were collected. ZsProSensor-1 fluorescence was detected in the FL1 emission channel of the flow cytometer.

Fluorescence recovery after photobleaching (FRAP) analysis. MBECs and HepG2 cells were transfected with 1.0 µg pCMX-GFP or pZsProSensor-1. The cells were treated with the proteasome inhibitor MG132 for 8 h and analyzed immediately or 4 h after medium was changed to remove MG132. The fluorescence was detected using a Zeiss LSM 510 M microscope. An image was acquired, followed by 50 or 30 s of photobleaching (100% laser power; 488 nm) in either the nucleus or the cytoplasm of one cell. After the bleaching, images were taken every minute during a period of 10 min. This was repeated three times. Measurement of the intensity of the fluorescent signal was performed using ImageJ, version 1.33u (National Institutes of Health, Bethesda, Md.).

Mouse corneal micropocket angiogenesis assay. The mouse corneal angiogenesis assay was performed as previously described (6). Micropellets (0.35 by 0.35 mm) of sucrose aluminum sulfate (Bukh Meditec) coated with hydron polymer type NCC containing the same amounts of the synthesized peptides were surgically implanted into each micropocket of eyes of male 5- to 6-week-old C57BL/6 mice. Six mice were used in each group. Eyes were examined by a slit-lamp biomicroscope on day 5 after pellet implantation. Neovascularization areas were measured and statistically evaluated by a Student's t test. All animal studies were reviewed and approved by the animal care and use committee of the Stockholm Animal Board.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

 
Activation of HIF function by expression of peptide fragments interacting with pVHL. The region of HIF-1 spanning amino acids 546 to 574 interacts with pVHL and is located within the N-terminal transactivation domain (schematically represented in Fig. 1A) (43). In the present study we have investigated in primary endothelial cells the ability of pVHL-interacting subdomains of this region of HIF-1 to mediate induction of HRE-driven luciferase reporter gene expression following stabilization of the endogenous HIF protein at normoxia. To address this question, we used previously characterized constructs (43) encoding wild-type or mutant forms of mHIF-1(546-574), fused to the GAL4 DNA-binding domain and tagged with FLAG (Fig. 1A). Overexpression of wild-type HIF and mutants (L556A/L558A, D568A, D569A, and Q572A), which have all been previously shown to interact with pVHL, increased luciferase activity in a dose-dependent manner in MBECs (Fig. 1B). As expected, HIF mutants (P563A, P563D/Y564D/Y565D, D570A, and F571A/F573A) that are unable to bind pVHL had no effect on luciferase activity (43). Since binding of pVHL to HIF mediates the degradation of HIF protein at normoxia, these results indicate, as expected (58), that overexpression of a pVHL-interacting domain of HIF-1 can interfere with the degradation pathway of endogenous HIF and increase its activity at normoxia.

View larger version (45K): [in this window] [in a new window]   FIG. 1. Activation of HIF function at normoxia. (A) Schematic representation of mHIF-1(546-574) and deletion mutants. Asterisks indicate mutated residues. (B and D) Relative reporter gene activity obtained following expression of mHIF-1(546-574) mutants. pT81/HRE-luc and increasing amounts of pFLAG-GAL4-mHIF-1 fusion constructs indicated in panel A were cotransfected into MBECs. Cells were cultured under normoxic conditions or treated with hypoxia (H), 100 µM 2,2'-dipyridyl (DP), or 100 µM CoCl2 as indicated. Data are presented as luciferase activity relative to cells transfected with pFLAG-CMV-2 (FLAG). Values represent means ± standard deviations of three independent experiments performed in duplicate. *, P < 0.05, compared to the luciferase activity in cells overexpressing pFLAG-GAL4 (GAL4) (two-way analysis of variance test); ^, P < 0.05, compared to the cells overexpressing pFLAG-CMV-2 cultured at normoxia (Student's t test). (C and E) Expression levels of mHIF-1-derived peptides detected by immunoblotting using an anti-FLAG antibody. (F) Schematic representation of chemically synthesized peptides spanning mHIF-1 residues 559 to 573 fused to the HIV-1 TAT protein transduction domain (PTD). (G) Peptide competition assay. In vitro translated FLAG-tagged full-length mHIF-1 was immunoprecipitated (IP) with anti-FLAG antibody and incubated with in vitro translated [35S]methionine-labeled pVHL in the presence of either vehicle (lane 2); 10 µM TAT peptide (lane 3); 0.5, 1.0, 2.0, 5.0, or 10.0 µM THIF15 peptide (lanes 4 to 8); or 0.05, 0.1, 0.2, 0.5, or 1.0 µM THIF15OH peptide (lanes 9 to 13). Peptides were preincubated with either lysis buffer (–WCE) or whole-cell extracts from HepG2 cells (+WCE) for 1 h at room temperature. Precipitated pVHL was analyzed by 12.5% SDS-PAGE, followed by autoradiography. (H) Mouse corneal angiogenesis assay. Micropellets of the peptides were implanted into mouse cornea pockets (arrows), and 5 days later corneal angiogenesis was monitored. Photographs represent x20 magnification of the mouse eye. Quantification (graph at right) of corneal neovascularization, presented as maximal areas of neovascularization on day 5. Graph represents mean values ± standard deviation. ***, P < 0.001 in comparison to the neovascularization area induced with TAT peptide alone (Student's t test).

A chemically synthesized HIF-1 peptide spanning residues 559 to 573 fused to the human immunodeficiency virus type 1 (HIV-1) TAT protein transduction domain induces angiogenesis in vivo. To evaluate the angiogenic potential of the mHIF-1 peptide spanning residues 559 to 573, we tested the effect of this peptide in an in vivo angiogenesis assay. mHIF-1(559-573) was synthesized in a hydroxylated (Pro⁵⁶³) (THIF15OH) or unhydroxylated (THIF15) form and fused to the human HIV-1 TAT protein transduction domain to allow intracellular delivery (11, 46) (schematically represented in Fig. 1F). We first investigated if the peptides could compete with the binding of full-length mHIF-1 to pVHL (Fig. 1G). In these studies, in vitro translated FLAG-tagged full-length mHIF-1 and [³⁵S]methionine-labeled pVHL were used. Full-length mHIF-1 was first immunoprecipitated using anti-FLAG antibody bound to protein G-Sepharose, and after the beads were washed, [³⁵S]methionine-labeled pVHL was incubated with mHIF-1 in the presence of the synthetic peptides previously preincubated with either lysis buffer or whole-cell extracts from HepG2 cells. The HIV-1 TAT protein transduction domain had no effect on the interaction of mHIF-1 with pVHL (Fig. 1G, lane 3). In contrast, the hydroxylated mHIF-1 peptide spanning residues 559 to 573 (THIF15OH) was able to compete with the binding of full-length mHIF-1 to pVHL in a dose-responsive manner (0.05 to 1 µM) (Fig. 1G, lanes 9 to 13) in the absence or presence of whole-cell extract. The unhydroxylated peptide (THIF15) inhibited the binding of full-length mHIF-1 to pVHL only in the presence of whole-cell extract, and this effect was observed at a concentration 10-fold higher (0.5 to 10 µM) (Fig. 1G, lanes 4 to 8) than the concentration required for the hydroxylated peptide to produce the same effect. These results are consistent with the requirement of Pro⁵⁶³ hydroxylation (corresponding to Pro⁵⁶⁴ in human HIF-1) for the interaction of pVHL with the degradation box of HIF-1 (24, 25, 62).

We next performed an angiogenesis assay in mouse cornea (6) in order to investigate the angiogenic potential of the generated fusion peptides. To this end, the peptides were adsorbed onto slow-release micropellets and implanted into micropockets in mouse cornea. The stimulation of blood vessel growth from the limbus region was recorded after 5 days without any reduction of atmospheric oxygen levels. Notably, this analysis revealed significantly (P < 0.001) induced neovascularization in the corneas of mice treated with THIF15OH (Fig. 1H). In contrast, THIF15 did not induce any neovascularization in the mouse cornea, consistent with the finding that this peptide inhibited the interaction of full-length HIF-1 with pVHL with a much lower efficiency than the hydroxylated peptide. In conclusion, these data indicate that a synthetic 15-mer peptide of mHIF-1 spanning residues 559 to 573 with hydroxylated Pro⁵⁶³ can be used as a proangiogenic agent when fused to the HIV-1 TAT protein transduction domain.

Intracellular localization of mHIF-1(559-573) peptide affects induction of HIF activity at normoxia in a cell-type-specific manner. The angiogenic potential of a HIF-1-derived peptide has previously been investigated in the context of a fusion containing an NLS motif (58). In the present study we have investigated the relevance of the subcellular localization of the peptide for activation of HIF function at normoxia. To this end we generated GST-GFP fusions of mHIF-1(559-573) containing a wild-type simian virus 40 large-T antigen NLS (TPPKKKRKVED) or an NES (ALQKKLEELELDE) motif from Xenopus mitogen-activated protein kinase kinase and corresponding mutants (NLS mutant K128T/K129T and NES mutant L37A/L40A) (18, 26) (schematically represented in Fig. 2A).

View larger version (29K): [in this window] [in a new window]   FIG. 2. Subcellular localization of the mHIF-1(559-573) peptide affects activation of HIF function in a cell-type-specific manner. (A) Schematic representation of the GST-GFP-HIF-1 fusion peptides. (B) Subcellular localization of the mHIF-1 fusion peptides. Plasmids encoding wild-type or mutant FLAG-GST-GFP/mHIF-1(559-573) were transiently transfected into MBECs and HepG2 cells cultured under normoxic conditions. The subcellular localization of fusion peptides was detected by confocal microscopy. Representative images are shown. Categorization and quantitative evaluation of the subcellular localization pattern of the GST-GFP fusion peptides are presented as percentages of cells belonging to the categories N, N>C, N=C, and N<C as described in Materials and Methods. (C) Expression levels of the GST-GFP fusion peptides detected by immunoblotting using an anti-FLAG antibody. (D to F) Relative HRE-driven luciferase activity obtained upon transient expression of wild-type or mutant GST-GFP-mHIF-1(559-573). The plasmid encoding the HRE-driven luciferase reporter gene and increasing amounts of the GST-GFP-fused constructs were cotransfected into MBECs and HepG2 (D), HeLa (E), and hTERT+-BCE (F) cells. Cells were cultured under normoxic conditions. Data are presented as luciferase activity relative to cells transfected with pFLAG-CMV-2 (FLAG) alone. Values represent means ± standard deviations of three independent experiments performed in duplicate.

Differential proteasomal activity in the nuclear and cytoplasmic compartments of MBECs and HepG2 cells. Since our data showed that the intracellular localization of the overexpressed HIF-1 peptide (residues 559 to 573) affected HRE-dependent luciferase activity in a cell-specific fashion, we next characterized the proteasome activity in the nuclear and cytoplasmic compartments of MBECs and HepG2 cells. In these experiments we used as a "sensor" of proteasome activity the construct pZsProSensor-1 in which a specific degradation motif of ornithine decarboxylase (35) has been fused to the fluorescent ZsGreen protein, generating a fusion protein that is rapidly degraded by the proteasome without requiring prior ubiquitylation. As shown in Fig. 3A, treatment of MBECs or HepG2 cells expressing the ZsProSensor-1 protein with the proteasome inhibitor MG132 increased the number of fluorescent cells, demonstrating that inhibition of proteasome activity leads to stabilization of ZsProSensor-1. The different levels of ZsProSensor-1 protein observed in the two cell types in the absence of MG132 suggest different levels of proteasome activity, with the activity in transformed HepG2 cells being higher than in the primary MBECs (Fig. 3A). This observation is in agreement with previous observations showing that, in highly proliferative and transformed cell lines, proteasome levels and activity are higher than in quiescent and nontransformed cells (33).

View larger version (29K): [in this window] [in a new window]   FIG. 3. Compartmentalization of proteasome activity in MBECs and HepG2 cells. (A) Treatment of cells with the proteasome inhibitor MG132 leads to stabilization of the ZsProSensor-1 protein. pZsProSensor-1 was transfected into MBECs or HepG2 cells cultured at normoxia. The cells were treated with dimethyl sulfoxide (–MG132) or 10 µM MG132 (+MG132) for 8 h before fixation. Cells were observed with lower magnification. Representative images are shown. (B) Differential proteasome-mediated degradation of ZsProSensor-1 protein occurs in MBECs and HepG2 subcellular compartments. Cells were transfected with pZsProSensor-1 and cultured at normoxia. After treatment with 10 µM MG132 for 8 h, the cells were fixed (0 h) or washed three times with culture medium to remove MG132 and then fixed at 4 h, 6 h, 8 h, or 10 h of incubation. Cells were analyzed by confocal microscopy, and representative images are presented. Quantitative evaluation and categorization was performed as described in the legend of Fig. 2B. (C) FRAP analysis of cells transfected with pZsProSensor-1. Fluorescence before bleaching was considered 100%. Data are shown as the ratio between the fluorescence observed within the nuclear and cytoplasmic compartments or between cytoplasmic and nuclear compartments following bleaching of the nucleus or cytoplasm, respectively. The results presented are the average of the analysis of MBECs or HepG2 cells bleached in the nuclear compartment (MBEC_BN or HepG2_BN) or in the cytoplasm (MBEC_BC or HepG2_BC) after 4 h following MG132 removal. (D) Overexpressed HIF-1 stabilizes endogenous Nrf2. MBECs and HepG2 cells were cultured at normoxia (–), exposed to hypoxia or treated with 10 µM MG132 for 8 h, or transfected with pFLAG or pFLAG-mHIF-1. Endogenous Nrf2 was detected by immunoblotting. (E) Overexpressed HIF-1 increased the expression of ZsProSensor-1 protein. MBECs and HepG2 cells were transfected with expression vector encoding ZsProSensor-1 alone (–) or treated with 10 µM MG132 (MG132) for 8 h or cotransfected with pFLAG or pFLAG-mHIF-1 as indicated. Levels of ZsProSensor-1 protein were determined by FACS analysis. The percentage of fluorescent cells is indicated. (F) Stabilization of the ZsProSensor-1 protein by expression of HIF-1. MBECs or HepG2 cells transfected with pZsProSensor-1 and pFLAG-mHIF-1 or pCMV2-FLAG was observed at lower magnification.

In order to investigate if the rate of diffusion affected the distribution of ZsProSensor-1 activity, we performed FRAP analysis of cells treated with MG132 or 4 h after MG132 removal and transfected with ZsProSensor-1 or GFP expression plasmids. In cells expressing GFP, 80 to 100% of fluorescence was recovered 4 to 5 min after bleaching (data not shown). In contrast, in MBECs the ZsProSensor-1 diffusion rate was reduced to a greater extent, with recovery of only 40 to 50% of fluorescence after 10 min (Fig. 3C). In HepG2 cells, the ZsProSensor-1 diffusion rate from cytoplasm to the nucleus was similar to GFP, whereas the mobility of the protein into the cytoplasm was reduced more (Fig. 3C). The same pattern of diffusion rates was observed in cells treated with MG132 and 4 h after MG132 removal. Considering that in MG132-treated cells ZsProSensor-1 was equally distributed in the cytoplasm and nucleus (N=C) in 80 to 90% of the cells, the results presented in Fig. 3B were interpreted as measurement of proteasome activity in the different compartments and not diffusion of the ZsProSensor-1 protein. Cellular viability after MG132 treatment was investigated by trypan blue exclusion, and the large majority of the cells proved to be viable (data not shown). In conclusion, these data indicate that the proteasome is more active in the cytoplasm of HepG2 cells, whereas in MBECs its activity is predominantly localized in the nuclear compartment.

We next investigated if overexpression of HIF-1 was able to interfere with proteasome-dependent degradation of proteins in the two cell types. We transfected cells with pFLAG-CMV-2 or pFLAG-mHIF-1 and investigated the levels of the endogenous protein levels of the transcription factor Nrf2 by Western blot analysis. Nrf2 is a short-lived protein that is degraded by the ubiquitin-proteasome pathway (32). As a control, cells were treated with MG132. Overexpression of HIF-1 slightly increased the levels of endogenous Nrf2 (Fig. 3D), but this mode of stabilization was much less pronounced than following MG132 treatment. The decrease in the levels of Nrf2 protein at hypoxia may reflect the global inhibition of translation occurring at hypoxia (1) and exclude Nrf2 as a downstream target of HIF-1. Since only a fraction of the cells are efficiently transfected with expression plasmids encoding HIF-1 or FLAG, it is difficult to assess the effect of overexpression of HIF-1 on endogenous proteins. To overcome this problem we used FACS analysis to score for transfected cells expressing the ZsProSensor-1 protein under different conditions (Fig. 3E). The results showed that treatment of MBECs with MG132 led to an increase in the number of fluorescent cells (from 0.05% to 2.78%). MBECs transfected with an HIF-1 expression plasmid produced fluorescence in 2.19% of the cells, whereas transfection with a FLAG expression plasmid resulted in 0.88% of the cells being fluorescent. HepG2 cells showed a higher degree of responsiveness to MG132 treatment (with the number of fluorescent cells increasing from 0.15% to 10.34%), whereas expression of HIF-1 resulted in a lower effect on the number of fluorescent cells compared to MBECs (Fig. 3B). As shown in Fig. 3F, overexpression of HIF-1 increased the number of fluorescent cells in both MBECs and HepG2 cells. These results show that in a small fraction of cells, probably corresponding to efficiently transfected cells, overexpression of HIF-1 is able to stabilize a protein that is degraded in a proteasome-dependent manner. Possibly, this effect was achieved because the HIF-1 protein has one of the shortest-known half-lives (22, 29), and its overexpression may decrease the efficiency of degradation of other unstable proteins by direct competition for proteasome-dependent degradation.

Proteasome-mediated degradation of HIF-1 occurs predominantly in different compartments of primary endothelial and hepatoma cells. The mechanism that regulates the compartmentalization of HIF-1 degradation is poorly understood. It has been proposed that HIF-1 binds to pVHL in the nucleus and is degraded in the cytoplasm after translocation (19). In contrast, it has been reported that trapping of HIF-1 either in the nucleus or in the cytoplasm does not prevent HIF-1 degradation (2). We therefore investigated HIF-1 degradation in both the cytosolic and nuclear compartments of MBECs and HepG2 cells. We transfected cells with HIF-1 and ZsProSensor-1 expression plasmids and investigated the distribution of the ZsProSensor-1 protein at normoxia, hypoxia, and following reoxygenation of previously hypoxic cells (Fig. 4A). In MBECs we did not observe any shift in the distribution of the ZsProSensor-1 protein at normoxia and following reoxygenation at time points when degradation of HIF-1 was detected. In contrast, in HepG2 cells there was a decrease in the number of cells with predominantly cytoplasmic (N<C) localization of ZsProSensor-1 from 59% in normoxic cells and 47% in reoxygenated cells to 23% in hypoxic cells. These results imply that HIF-1 was degraded mainly in the cytoplasm, in which the ZsProSensor-1 was accumulated at both normoxia and during reoxygenation. Immunocytochemical analysis confirmed the expression of FLAG-HIF-1 (data not shown). These assays showed that both at normoxia and during reoxygenation, HIF-1 was not equally degraded in both compartments in these two different cell types. Whereas in HepG2 cells HIF-1 was mainly degraded in the cytoplasm, in MBECs degradation seemed to occur efficiently in both the nuclear and cytoplasmic compartments.

View larger version (30K): [in this window] [in a new window]   FIG. 4. Compartmentalized proteasome-dependent degradation of HIF-1. (A) Competition of HIF-1 and ZsProSensor-1 protein for proteasomal degradation at normoxia and during reoxygenation. Cells were cotransfected with pZsProSensor-1 and pFLAG-mHIF-1 and cultured at normoxia (Nor) or hypoxia (H) for 8 h or exposed to reoxygenation for 30 min (R30) before fixation. The intracellular localization of ZsProSensor-1 protein was detected and evaluated using confocal microscopy. Quantitative evaluation and categorization of the cells were performed as described in the legend of Fig. 2B. (B and C) Degradation rate of endogenous pools of HIF-1 in MBECs and HepG2 cells in the nuclear and cytoplasmic compartments. Cells were treated with either dimethyl sulfoxide (-MG) or 10 µM MG132 (+MG) and grown under hypoxic conditions for 8 h, followed by reoxygenation for 0.1, 2, 4, 6, 8, or 10 min before harvest. Nuclear (Nuc) and cytosolic (Cyt) extracts were prepared, and immunoblot analysis using anti-HIF-1, antipaxillin, and anti-YY1 antibodies was performed. To facilitate detection in the cytosolic extracts, HIF-1 was immunoprecipitated using an anti-HIF-1 antibody before immunoblotting. (C) Quantification of endogenous HIF-1 protein levels was normalized by YY1 or paxillin expression levels (control) in the nucleus or cytoplasm, respectively. Relative HIF-1 protein levels at hypoxia were considered 100%.

Subcellular localization of HIF-1 at hypoxia is cell type specific. Since our results indicate that degradation of hypoxia-inducible factors did not occur in the same compartment in the hepatoma and endothelial cells, we investigated whether the intracellular distribution of HIF-1 also showed cell type specificity. To this end we studied by confocal microscopy the intracellular distribution of GFP-HIF-1 following transient expression in both MBECs and HepG2 cells. The cells were cultured either at normoxia or hypoxia for 8 h and then underwent reoxygenation for different periods of time as indicated in Fig. 5A. In MBECs the percentage of cells demonstrating exclusively nuclear localization (N) was significantly reduced compared to HepG2 cells. In the majority of MBECs, GFP-HIF-1 had a predominantly nuclear localization (N>C) at normoxia, both upon hypoxic treatment and during the reoxygenation process. In contrast to these results, GFP-HIF-1 had a predominantly nuclear distribution (N>C) in 60% of HepG2 cells at normoxia, whereas upon exposure to hypoxia, it was exclusively localized in the nucleus (N) in 80% of cells. These results indicate that GFP-HIF-1 was either imported to the nucleus of HepG2 cells or that the export to the cytoplasm was inhibited. During the reoxygenation process, GFP-HIF-1 showed again a predominantly nuclear localization (N>C) in 60 to 70% of cells, suggesting that GFP-HIF-1 was exported from the nucleus to the cytoplasm. In order to confirm these results we performed the same experiment using lower amounts (500 ng and 250 ng) of GFP-HIF-1 expression plasmid or treating the cells with hypoxia-mimicking agents such as 2,2'-dipyridyl or CoCl₂, and similar results were obtained (data not shown). We also tested HIF-1 deletion mutants, GFP-HIF-1(1-74), GFP-HIF-1(178390), and GFP-HIF-1(526-826) that show exclusively nuclear localization (N) in COS7 cells (27). In MBECs the majority of cells presented a predominantly nuclear distribution (N>C) both at normoxia and hypoxia (Fig. 5B). In HepG2 cells all these mutants were expressed in a fraction of cells where an exclusively nuclear distribution (N) was observed at normoxia and was increased upon exposure to hypoxia. In conclusion, these data show that HIF-1 is retained in the nucleus of hypoxic HepG2 cells and returns to the cytoplasmic compartment during reoxygenation. In contrast, in MBECs the subcellular distribution of HIF-1 did not show any major differences at hypoxia or during the reoxygenation process.

View larger version (28K): [in this window] [in a new window]   FIG. 5. Subcellular distribution of GFP-fused HIF-1 in MBECs and HepG2 cells. (A) Intracellular localization of GFP-HIF-1. GFP-HIF-1 expression vector was transiently transfected into MBECs or HepG2 cells. Cells exposed to hypoxia were fixed immediately or after 10, 20, or 30 min of reoxygenation (R10, R20, or R30, respectively). The intracellular localization of GFP-HIF-1 was detected and categorized by confocal microscopy. Representative images are shown. (B) Intracellular localization of GFP-HIF-1 deletion mutants. MBECs or HepG2 cells were transfected with GFP expression vectors of HIF-1 deletion mutants (shown schematically) and cultured at normoxia (Nor) or hypoxia (H). Quantitative evaluation and categorization of the subcellular distribution were done as described in Materials and Methods. (C) Protein levels of endogenous HIF-1 in MBECs and HepG2 cells. Cells were cultured at normoxia (N) or hypoxia (H) or treated with 100 µM 2,2'-dipyridyl (DP) in the presence or absence of 10 µM MG132 for 8 h as indicated. Endogenous HIF-1 levels were analyzed by immunoblot analysis using anti-HIF-1 antibodies. (D) Protein levels of endogenous HIF-2 in MBECs. MBECs were cultured at normoxia (N) or hypoxia (H) or treated with 100 µM 2,2'-dipyridyl (DP) for 8 h as indicated. Endogenous HIF-2 levels in whole-cell extracts were analyzed by immunoblotting using anti-HIF-2 antibodies. Whole-cell extracts from HEK293 cells transfected with pFLAG-HIF-2 (F-HIF-2) were used as a positive control.

In primary endothelial cells endogenous HIF-1 is still degraded by the ubiquitin-proteasome pathway under hypoxic conditions.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

 
In the current study we have identified a minimal pVHL-interacting subfragment of HIF-1 that stabilizes HIF against degradation and activates HIF function in primary endothelial cells. We observed that the activity of this peptide is significantly affected by its subcellular distribution in a cell-type-specific manner, correlating with the subcellular compartmentalization of proteasome-dependent degradation of HIF-1.

It has previously been shown that HIF-1 is located in the nuclear compartment in hypoxic cells and is exported to the cytoplasm during reoxygenation in many cell lines (9, 19, 27). In earlier studies on pVHL-interacting structures of HIF-1, angiogenic peptide fragments have been identified. All these experiments were performed with peptides fused to an NLS motif (58). Interestingly, our results show that fusion of an NLS motif to a pVHL-interacting HIF-1 peptide can impair the HIF-1-activating potential of the peptide in some cells and may abolish its activity in others. It remains to be determined whether there are actually cell types in which fusion with an NLS increases the efficiency of the peptide.

Within eukaryotic cells, the 26S proteasome is ubiquitously present in both the nucleus and the cytoplasm (59). In the case of nucleocytoplasmic shuttling proteins, e.g., p53, IkB, and p27^kit1, the efficiency of proteasome-mediated degradation has been linked to the nuclear export rate of the protein (17, 30, 47). In the case of HIF-1, it has been reported that, following ubiquitylation in the nucleus, HIF-1 is exported to the cytoplasm during reoxygenation, where it is degraded by the proteasome (19). However, there are a large number of studies demonstrating that protein degradation can also occur in the nucleus, consistent with the detection of proteasome activity in the nuclear compartment. Proteins such as C/EBP (13) and MyoD (16, 36) have been shown to be degraded in the nuclear compartment, and, in the case of most nuclear receptors and other transcription factors, activation of transcription and degradation by the proteasome have been shown to be interdependent events occurring in the nucleus (4, 14, 44, 56). In a similar fashion, it has been demonstrated that in the presence of high levels of Mdm2, p53 is degraded in the nuclear compartment (34, 37). In the present study we show that proteasome-mediated degradation of HIF-1 occurs mainly in the cytoplasm in the HepG2 hepatoma cell line, whereas in primary endothelial cells it occurs with equal efficiency in both compartments. In a similar fashion, it has been observed that HIF-1 retained either in the nucleus or in the cytoplasm shows the same degradation rate as the endogenous protein (2). However, in this study we present evidence that this observation is not common to all cell types.

As schematically summarized in Fig. 6, our results show that in hypoxic HepG2 cells, HIF-1 is retained in the nucleus and that, upon reoxygenation, it is exported to the cytoplasm where it is degraded. These results are in agreement with previous work from our laboratory showing that localization of HIF-1 in the nucleus of hypoxic cells is mediated by an NLS motif present in the C terminus of the protein (27) and that nuclear retention of the protein can protect HIF-1 from degradation (54). In this type of cells, pVHL-interacting peptides can only inhibit pVHL-mediated degradation of HIF-1 when targeted to the cytoplasmic compartment (Fig. 6). A different scenario is observed in primary endothelial cells. In these cells HIF-1 can be degraded equally efficiently in the nucleus or cytoplasm, and we do not observe any significant inhibition of nucleocytoplasmic shuttling of the protein at hypoxia. Consequently, following reoxygenation, the protein is degraded in the compartment where it localizes without requiring any specific translocation.

View larger version (39K): [in this window] [in a new window]   FIG. 6. Cell-type-specific regulation of subcellular compartmentalization and degradation of HIF-1. In primary endothelial cells (MBECs) (A and B), HIF-1 is present in both nuclear and cytoplasmic compartments under conditions of normoxia, hypoxia, and reoxygenation, and proteasome-dependent degradation occurs at a similar rate in both compartments. In this context pVHL-interacting peptides are able to inhibit pVHL-mediated degradation of HIF-1 in both compartments. In contrast, in highly proliferating or transformed cell types (e.g., HepG2 cells) (C and D), proteasome activity dominates in the cytoplasmic compartment. HIF-1 translocates from the cytoplasm to the nucleus in hypoxia. Upon reoxygenation, the protein needs to be transported back to the cytoplasm to be efficiently degraded by the proteasome. In this type of cells, pVHL-interacting peptides can only inhibit pVHL-mediated degradation of HIF-1 when targeted to the cytoplasmic compartment.

The mechanism of HIF-1 import and export from the nucleus to the cytoplasm is poorly understood. It has been proposed that pVHL through a non-CRM1-dependent mechanism is responsible for the export of HIF-1 during normoxia or reoxygenation. According to this model, HIF-1 becomes a completely nuclear protein at hypoxia because it does not interact with pVHL and therefore cannot be exported to the cytoplasm (19). In the present study this model is challenged by the results obtained with primary endothelial cells since we observe that at hypoxia HIF-1 does not become exclusively nuclear but is still present in the cytoplasm. It remains to be investigated if, in analogy to other proteins (20, 39, 60), export of HIF-1 is dependent on phosphorylation events that could be differentially regulated in various cell types.

In the current study we provide evidence that competition between HIF-1-derived peptide fragments and endogenous HIF for interaction with pVHL becomes efficient when the peptide is localized in the same compartment in which proteasome-dependent degradation of HIF-1 occurs. Our results suggest that in HepG2 cells, degradation of HIF-1 occurs predominantly in the cytoplasm, and in these cells the HIF-stabilizing and activating potential of the peptide was abolished following retention of the peptide in the nucleus by fusion to an NLS motif. In MBECs, in which we observe equal degradation of HIF-1 in both compartments, insertion of either an NLS or an NES motif into the peptide only partially reduces its ability to compete with endogenous HIF for protection against pVHL and its HIF-activating potential. It has been proposed that HIF-2 is degraded by a mechanism similar to HIF-1 destruction, and HIF-2 has been shown to be expressed in many cell lines (3), tissues (57), and tumors (53). However, the relevance of HIF-2 for adaptive responses to hypoxia has been questioned in some studies (42, 52), namely because in mouse embryo fibroblasts, HIF-2 is localized in the cytoplasm and is unable to activate target genes (42).

Levels of hydroxylase activities and pVHL availability have been proposed to regulate degradation efficiency of HIF-1 proteins (5, 10, 15, 31, 41, 54). In the present study we indicate another level of regulation depending on compartmentalization of the proteasome-dependent degradation activity and subcellular distribution of the protein. At hypoxia HIF-1 becomes stable due to the impairment of prolyl hydroxylase activities, resulting in inhibition of pVHL interaction. However, a previous study has shown that some residual hydroxylation of HIF-1 can be observed in hypoxic cells (8), which leads us to speculate that this fraction of HIF-1 protein may undergo degradation at hypoxia. In this context, compartmentalization of HIF-1 degradation by the proteasome together with subcellular distribution may regulate protein levels in hypoxic cells. In HepG2 cells, proteasome-dependent degradation of HIF-1 is more efficient in the cytoplasm, and the protein is retained in the nucleus at hypoxia, possibly generating high levels of protein stabilization. In contrast, in primary endothelial cells nucleo-cytoplasmic shuttling of HIF-1 is not inhibited, and the protein can be degraded with the same efficiency in both compartments, contributing to relatively low levels of stabilized HIF-1. Taken together our results indicate new levels of complexity in the regulation of HIF-1 protein levels by cell-type-specific parameters such as subcellular distribution of the proteins and compartmentalized degradation of HIF-1. These differences may ultimately be exploited in efforts to either enhance HIF-1 activity (e.g., in proangiogenic therapy) or inhibit HIF-1 function (e.g., in tumor therapy) in a tissue-specific manner.

	ACKNOWLEDGMENTS

 
We thank Kensaku Okamoto and Nico Dantuma for stimulating discussions and helpful advice.

X.Z. was supported by a fellowship by the World Health Organization, and J.L.R. was supported by a Ph.D. Student Fellowship, PRAXIS XXI/BD/19994/99, from the Ministry of Science/Fundação para a Ciência e a Tecnologia of Portugal. F.A.S. was supported by a grant of the Nordic Center of Excellence "Neurodegeneration" to Nico Dantuma. These studies were supported by grants from the Swedish Research Council, Swedish Cancer Society, Swedish Heart and Lung foundation, and the European Union (Euroxy).

	FOOTNOTES

 
* Corresponding author. Mailing address: Department of Cell and Molecular Biology, Karolinska Institutet, S-171 77 Stockholm, Sweden. Phone: 46 8 5248 7330. Fax: 46 8 34 88 19. E-mail: lorenz.poellinger{at}cmb.ki.se.

	REFERENCES

Top Abstract Introduction Materials and Methods Results Discussion References

 
1. Arsham, A. M., J. J. Howell, and M. C. Simon. 2003. A novel hypoxia-inducible factor-independent hypoxic response regulating mammalian target of rapamycin and its targets. J. Biol. Chem. 278:29655-29660.[Abstract/Free Full Text]

2. Berra, E., D. Roux, D. E. Richard, and J. Pouyssegur. 2001. Hypoxia-inducible factor-1 alpha (HIF-1 alpha) escapes O(2)-driven proteasomal degradation irrespective of its subcellular localization: nucleus or cytoplasm. EMBO Rep. 2:615-620.[CrossRef][Medline]

3. Blancher, C., J. W. Moore, K. L. Talks, S. Houlbrook, and A. L. Harris. 2000. Relationship of hypoxia-inducible factor (HIF)-1 and HIF-2 expression to vascular endothelial growth factor induction and hypoxia survival in human breast cancer cell lines. Cancer Res. 60:7106-7113.[Abstract/Free Full Text]

4. Blanquart, C., O. Barbier, J. C. Fruchart, B. Staels, and C. Glineur. 2002. Peroxisome proliferator-activated receptor (PPAR) turnover by the ubiquitin-proteasome system controls the ligand-induced expression level of its target genes. J. Biol. Chem. 277:37254-37259.[Abstract/Free Full Text]

5. Bruick, R. K., and S. L. McKnight. 2001. A conserved family of prolyl-4-hydroxylases that modify HIF. Science 294:1337-1340.[Abstract/Free Full Text]

6. Cao, Y., P. Linden, J. Farnebo, R. Cao, A. Eriksson, V. Kumar, J. H. Qi, L. Claesson-Welsh, and K. Alitalo. 1998. Vascular endothelial growth factor C induces angiogenesis in vivo. Proc. Natl. Acad. Sci. USA 95:14389-14394.[Abstract/Free Full Text]

7. Carmeliet, P., Y. Dor, J. M. Herbert, D. Fukumura, K. Brusselmans, M. Dewerchin, M. Neeman, F. Bono, R. Abramovitch, P. Maxwell, C. J. Koch, P. Ratcliffe, L. Moons, R. K. Jain, D. Collen, and E. Keshert. 1998. Role of HIF-1 in hypoxia-mediated apoptosis, cell proliferation and tumour angiogenesis. Nature 394:485-490.[CrossRef][Medline]

8. Chan, D. A., P. D. Sutphin, N. C. Denko, and A. J. Giaccia. 2002. Role of prolyl hydroxylation in oncogenically stabilized hypoxia-inducible factor-1alpha. J. Biol. Chem. 277:40112-40117.[Abstract/Free Full Text]

9. Chilov, D., G. Camenisch, I. Kvietikova, U. Ziegler, M. Gassmann, and R. H. Wenger. 1999. Induction and nuclear translocation of hypoxia-inducible factor-1 (HIF-1): heterodimerization with ARNT is not necessary for nuclear accumulation of HIF-1. J. Cell Sci. 112:1203-1212.[Abstract]

10. Cockman, M. E., N. Masson, D. R. Mole, P. Jaakkola, G. W. Chang, S. C. Clifford, E. R. Maher, C. W. Pugh, P. J. Ratcliffe, and P. H. Maxwell. 2000. Hypoxia inducible factor-alpha binding and ubiquitylation by the von Hippel-Lindau tumor suppressor protein. J. Biol. Chem. 275:25733-25741.[Abstract/Free Full Text]

11. Console, S., C. Marty, C. Garcia-Echeverria, R. Schwendener, and K. Ballmer-Hofer. 2003. Antennapedia and HIV transactivator of transcription (TAT) "protein transduction domains" promote endocytosis of high molecular weight cargo upon binding to cell surface glycosaminoglycans. J. Biol. Chem. 278:35109-35114.[Abstract/Free Full Text]

12. Dantuma, N. P., K. Lindsten, R. Glas, M. Jellne, and M. G. Masucci. 2000. Short-lived green fluorescent proteins for quantifying ubiquitin/proteasome-dependent proteolysis in living cells. Nat. Biotechnol. 18:538-543.[CrossRef][Medline]

13. Dearth, L. R., and J. DeWille. 2003. Posttranscriptional and posttranslational regulation of C/EBP delta in G0 growth-arrested mammary epithelial cells. J. Biol. Chem. 278:11246-11255.[Abstract/Free Full Text]

14. DeFranco, D. B. 2002. Navigating steroid hormone receptors through the nuclear compartment. Mol. Endocrinol. 16:1449-1455.[Abstract/Free Full Text]

15. Epstein, A. C., J. M. Gleadle, L. A. McNeill, K. S. Hewitson, J. O'Rourke, D. R. Mole, M. Mukherji, E. Metzen, M. I. Wilson, A. Dhanda, Y. M. Tian, N. Masson, D. L. Hamilton, P. Jaakkola, R. Barstead, J. Hodgkin, P. H. Maxwell, C. W. Pugh, C. J. Schofield, and P. J. Ratcliffe. 2001. C. elegans EGL-9 and mammalian homologs define a family of dioxygenases that regulate HIF by prolyl hydroxylation. Cell 107:43-54.[CrossRef][Medline]

16. Floyd, Z. E., J. S. Trausch-Azar, E. Reinstein, A. Ciechanover, and A. L. Schwartz. 2001. The nuclear ubiquitin-proteasome system degrades MyoD. J. Biol. Chem. 276:22468-22475.[Abstract/Free Full Text]

17. Freedman, D. A., and A. J. Levine. 1998. Nuclear export is required for degradation of endogenous p53 by MDM2 and human papillomavirus E6. Mol. Cell. Biol. 18:7288-7293.[Abstract/Free Full Text]

18. Fukuda, M., I. Gotoh, Y. Gotoh, and E. Nishida. 1996. Cytoplasmic localization of mitogen-activated protein kinase kinase directed by its NH2-terminal, leucine-rich short amino acid sequence, which acts as a nuclear export signal. J. Biol. Chem. 271:20024-20028.[Abstract/Free Full Text]

19. Groulx, I., and S. Lee. 2002. Oxygen-dependent ubiquitination and degradation of hypoxia-inducible factor requires nuclear-cytoplasmic trafficking of the von Hippel-Lindau tumor suppressor protein. Mol. Cell. Biol. 22:5319-5336.[Abstract/Free Full Text]

20. Haendeler, J., J. Hoffmann, R. P. Brandes, A. M. Zeiher, and S. Dimmeler. 2003. Hydrogen peroxide triggers nuclear export of telomerase reverse transcriptase via Src kinase family dependent phosphorylation of tyrosine 707. Mol. Cell. Biol. 23:4598-4610.[Abstract/Free Full Text]

21. Hergovich, A., J. Lisztwan, R. Barry, P. Ballschmieter, and W. Krek. 2003. Regulation of microtubule stability by the von Hippel-Lindau tumour suppressor protein pVHL. Nat. Cell Biol. 5:64-70.[CrossRef][Medline]

22. Huang, L. E., J. Gu, M. Schau, and H. F. Bunn. 1998. Regulation of hypoxia-inducible factor 1 is mediated by an O2-dependent degradation domain via the ubiquitin-proteasome pathway. Proc. Natl. Acad. Sci. USA 95:7987-7992.[Abstract/Free Full Text]

23. Iliopoulos, O., M. Ohh, and W. G. Kaelin, Jr. 1998. pVHL19 is a biologically active product of the von Hippel-Lindau gene arising from internal translation initiation. Proc. Natl. Acad. Sci. USA 95:11661-11666.[Abstract/Free Full Text]

24. Ivan, M., K. Kondo, H. Yang, W. Kim, J. Valiando, M. Ohh, A. Salic, J. M. Asara, W. S. Lane, and W. G. Kaelin, Jr. 2001. HIF targeted for VHL-mediated destruction by proline hydroxylation: implications for O2 sensing. Science 292:464-468.[Abstract/Free Full Text]

25. Jaakkola, P., D. R. Mole, Y. M. Tian, M. I. Wilson, J. Gielbert, S. J. Gaskell, A. Kriegsheim, H. F. Hebestreit, M. Mukherji, C. J. Schofield, P. H. Maxwell, C. W. Pugh, and P. J. Ratcliffe. 2001. Targeting of HIF- to the von Hippel-Lindau ubiquitylation complex by O2-regulated prolyl hydroxylation. Science 292:468-472.