Cancer Research Infection and Cancer: Biology, Therapeutics, and Prevention
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
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Covello, K. L.
Right arrow Articles by Keith, B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Covello, K. L.
Right arrow Articles by Keith, B.
[Cancer Research 65, 2277-2286, March 15, 2005]
© 2005 American Association for Cancer Research

Cell and Tumor Biology

Targeted Replacement of Hypoxia-Inducible Factor-1{alpha} by a Hypoxia-Inducible Factor-2{alpha} Knock-in Allele Promotes Tumor Growth

Kelly L. Covello1, M. Celeste Simon1,2,3 and Brian Keith2

1 Department of Cell and Developmental Biology, 2 Abramson Family Cancer Research Institute, and 3 Howard Hughes Medical Institute, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania

Requests for reprints: M. Celeste Simon, Abramson Family Cancer Research Institute, University of Pennsylvania, BRB II/III, Room 450, Philadelphia, PA 19104. Phone: 215-746-5526; Fax: 215-746-5561; E-mail: celeste2{at}mail.med.upenn.edu.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Hypoxia-inducible factors (HIF) are essential transcriptional regulators that mediate adaptation to hypoxic stress in rapidly growing tissues such as tumors. HIF activity is regulated by hypoxic stabilization of the related HIF-1{alpha} and HIF-2{alpha} subunits, which are frequently overexpressed in cancer cells. To assess the relative tumor-promoting functions of HIF-1{alpha} and HIF-2{alpha} directly, we replaced HIF-1{alpha} expression with HIF-2{alpha} by creating a novel "knock-in" allele at the Hif-1{alpha} locus through homologous recombination in primary murine embryonic stem cells. Compared with controls, s.c. teratomas derived from knock-in embryonic stem cells were larger and more proliferative, had increased microvessel density, and exhibited increased expression of vascular endothelial growth factor, transforming growth factor-{alpha}, and cyclin D1. These and other data indicate that HIF-2{alpha} promotes tumor growth more effectively than HIF-1{alpha} in multiple contexts.

Key Words: HIF • Hypoxia • VEGF • angiogenesis


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Oxygen deprivation (hypoxia) can occur in rapidly growing tissues, or as a consequence of disrupted blood flow, and can result in depletion of intracellular ATP stores, compromised ion transport, and altered cellular homeostasis (1). Multicellular animals have evolved complex mechanisms that mediate adaptation to hypoxic stress during normal development, as well as in pathologic states such as ischemia and cancer. The master transcriptional regulators of both cellular and systemic hypoxic adaptation are the hypoxia-inducible factors (HIF). HIF heterodimers consist of an {alpha} subunit (HIF{alpha}) and ß subunit (HIFß, also known as aryl hydrocarbon receptor nuclear translocator; refs. 2, 3). HIFs regulate the expression of at least 150 genes involved in metabolism, cell survival, erythropoiesis, and vascular remodeling (1) by binding to cis-acting hypoxia response elements located in the enhancers and/or promoters of these genes (4, 5). The growing list of HIF target genes include those encoding important angiogenic factors such as vascular endothelial growth factor (VEGF) and platelet-derived growth factor-ß (6–10), as well as glycolytic enzymes, glucose transporters, and signaling molecules (1).

HIF transcriptional activity is regulated primarily at the level of HIF{alpha} subunit stability. Under atmospheric O2 conditions or "normoxia" (21% O2), HIF{alpha} subunits are constitutively transcribed and translated, but rapidly ubiquitinated and degraded through direct interaction with an E3 ubiquitin ligase complex containing the von Hippel-Lindau tumor suppressor protein (11–15). HIF-1{alpha} is expressed ubiquitously in human and mouse tissues, and has been proposed as the primary regulator of the hypoxic response (5, 16, 17). The more recent identification of HIF-2{alpha} (also known as endothelial PAS domain protein 1, HIF-1-like factor, and HIF-1-related factor) has raised important questions about the relative activity of these factors in mediating hypoxic adaptation (18–20). HIF-1{alpha} and HIF-2{alpha} share a high degree of sequence identity, which is underscored by their shared ability to heterodimerize with aryl hydrocarbon receptor nuclear translocator and bind hypoxia response elements to activate transcription in various in vitro reporter assays (18, 20). In addition, both HIF-1{alpha} and HIF-2{alpha} are subject to posttranslational regulation mediated by the von Hippel-Lindau tumor suppressor protein (13, 21) . Interestingly, recent reports have indicated that HIF-1{alpha} and HIF-2{alpha} have partly, but not completely, overlapping sets of target genes. For example, whereas HIF-1{alpha} and HIF-2{alpha} both activate expression of VEGF, glucose transporter 1, and several other targets, only HIF-1{alpha} seems to regulate the hypoxic induction of genes encoding glycolytic enzymes (22, 23).

Whereas HIF-1{alpha} is expressed ubiquitously, HIF-2{alpha} is expressed in a more spatially restricted pattern. For example, HIF-2{alpha} mRNA is expressed predominantly in endothelial cells, the mesenchyme of the lung, and neural crest derivatives during embryonic development (18–20). Postnatally, HIF-2{alpha} protein has been detected specifically in bone marrow macrophages, kidney epithelial cells, liver parenchyma, cardiac myocytes, uterine decidual cells, and pancreatic parenchymal cells in hypoxic rats (24). Genetic ablation of HIF-1{alpha} or HIF-2{alpha} has revealed dramatically different phenotypes, consistent with their differential expression patterns. HIF-1{alpha} null embryos die between embryonic days 8.5 and 10.5 (E8.5-E10.5), and display striking structural anomalies, including massive apoptosis of the neural mesenchyme and aberrant vasculature (6, 10). In contrast, HIF-2{alpha} null embryos display a variety of phenotypes depending on genetic background; some embryos die at E13.5 due to bradycardia or hemorrhagic yolk sacs, whereas other HIF-2{alpha} null pups are born but die within days or weeks due to respiratory distress or general mitochondrial dysfunction (25–28). The degree to which these apparent functional differences between HIF-1{alpha} and HIF-2{alpha} are a consequence of disparate tissue-specific expression patterns, as opposed to differential target gene activation, is currently unknown.

Both HIF-1{alpha} and HIF-2{alpha} are expressed at high levels in a variety of human tumors and tumor cell lines (29, 30), although the relative contribution of each protein to tumor initiation and progression is not yet clear. Mutational studies have shown that ES cells and transformed mouse embryonic fibroblasts lacking HIF-1{alpha} protein show diminished ability to generate s.c. teratomas and fibrosarcomas, respectively (10, 31). Recent studies have also revealed an important role for HIF-2{alpha} in promoting the growth of xenografts derived from von Hippel-Lindau tumor suppressor protein–deficient renal clear cell carcinomas (RCC; refs. 32–34). Intriguingly, these experiments indicate that HIF-2{alpha}, but not HIF-1{alpha}, promoted the growth of these tumors, and that specific suppression of HIF-2{alpha} expression abrogated tumor growth (32).

To fully understand the role of hypoxic adaptation in cancer biology, it will be important to dissect the specific roles of HIF-1{alpha} and HIF-2{alpha} in tumor growth. Consequently, we have generated embryonic stem (ES) cells in which a cDNA encoding a c-Myc epitope-tagged HIF-2{alpha} was targeted into the first exon of the Hif-1{alpha} locus, generating a novel "knock-in" (KI) allele. This allele expresses HIF-2{alpha} under the control of the Hif-1{alpha} locus, but in a Hif-1{alpha} null background, thereby allowing a direct comparison of HIF-1{alpha} and HIF-2{alpha} when expressed under identical regulatory controls. In this report, we tested the ability of wild-type and homozygous KI ES cells to generate teratomas when injected s.c. into nude mice. Our experiments revealed that tumors derived from KI ES cells were more proliferative and more highly vascularized than wild-type tumors. Of note, tumor mass and microvessel density correlated with increased expression of the HIF-2{alpha}-regulated targets VEGF, cyclin D1 and transforming growth factor-{alpha} (TGF-{alpha}). Moreover, KI tumors displayed a distinctive spectrum of differentiated bone, cartilage, adipocyte, and neuronal (ganglia and Schwann cells) cell types, as well as an increased proportion of undifferentiated cells. Our study of the novel KI allele reveals inherent differential effects of HIF-1{alpha} and HIF-2{alpha} on tumor growth and cellular differentiation. Consistent with a previous report on renal cell carcinomas, our results further indicate that HIF-2{alpha} preferentially promotes tumor growth.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Generation of the HIF-1{alpha}2{alpha} KI Vector. A double-stranded oligonucleotide encoding a c-Myc epitope tag (MEQKLISEEDLN) was introduced at the carboxyl terminus of the murine HIF-2{alpha} cDNA (22). This modified cDNA was then fused to an XbaI-NcoI fragment comprising 1.7 kb of DNA upstream of the HIF-1{alpha} initiator ATG (located within the NcoI site CCATGG; ref. 35). This gene fusion was cloned into the pLNT targeting vector, which was created by introducing loxP recombination sites immediately surrounding the Neomycin resistance cassette in the pPNT plasmid (36). A 300 bp fragment containing the SV40 polyadenylation sequence and stop codons in all three reading frames were initially introduced immediately 3' to the HIF-2{alpha} cDNA. To complete the targeting vector, a DNA fragment extending ~7 kb downstream of the HIF-1{alpha} ATG was amplified (upstream primer, 5'-cgcctcgaggagggcgccggcggcgagaacgagaagaa-3'; downstream primer, 5'-gaacatcgattgagccctaggtcccaagaaggagaccc-3') and cloned as an XhoI-ClaI fragment into the pLNT plasmid already containing the 5' HIF-1{alpha} promoter::HIF-2{alpha} cDNA::SV40 polyA gene fusion. The final targeting vector represents an insertion of the HIF-2{alpha} cDNA::SV40 polyA::loxP-NeoR-loxP cassette at the ATG of the HIF-1{alpha} locus. No HIF-1{alpha} sequences were deleted. This targeting vector was linearized at the ClaI site and electroporated into 129SvEvTac-derived TL1 ES cells (gift from P. Labosky, University of Pennsylvania, Philadelphia, PA). ES cells were grown in the presence of 300 µg/mL G418 and 2 µmol/L gancyclovir for 10 days. G418/Gancyclovir-resistant clones were screened by Southern blot analysis of NcoI-digested DNA hybridized to a probe derived from sequences upstream of the 5' XbaI site.

ES cells heterozygous for the correctly targeted HIF-1{alpha}2{alpha} KI allele were grown in the presence of increased G418 concentrations to select for clones that had become homozygous for the HIF-1{alpha}2{alpha} allele. The NeoR cassette was removed from the locus by loxP-mediated recombination. Briefly, HIF-1{alpha}2{alpha} ES cells were electroporated with a plasmid encoding Cre recombinase and diluted to generate single colonies, which were again assayed by Southern blot analysis to identify the final KI allele.

Cell Lines. All ES cells were cultured in DMEM (Invitrogen) supplemented with 15% fetal bovine serum, 1% nonessential amino acids, 2% L-glutamine, 1% penicillin-streptomycin, leukemia inhibitory factor (ESGRO, Life Technologies, Temecula, CA), and 2-mercaptoethanol (Invitrogen, Carlsbad, CA). For Northern blot, electrophoretic mobility shift assay (EMSA), and Western blot analysis, ES cells were gelatin adapted on 10-cm plates and exposed to normoxia or hypoxia (1.5% O2). The Hif-1{alpha}–/– ES cell line (37) was a kind gift from P. Carmeliet (Molecular and Cardiovascular Medicine Group, University of Leuven, Leuven, Belgium).

Electrophoretic Mobility Shift Assay. EMSA analysis was done for HIF-1{alpha} on nuclear extracts using a radiolabeled probe for the hypoxia response element of the erythropoietin promoter as previously described (38). Briefly, 5 µg of nuclear extracts were incubated with binding buffer, 0.05 mg/mL bovine serum albumin, 0.025 µg/mL of cyclic AMP-responsive element binding protein oligo, and 106 dpm of labeled probe. HIF/DNA complexes were supershifted with HIF-1{alpha} antibody (Novus, Littleton, CO). Protein-DNA adducts were detected by phosphoimager analysis (Molecular Dynamics, Piscataway, NJ).

Immunoprecipitation and Western Blot Analysis. c-Myc epitope-tagged HIF-2{alpha} protein was immunoprecipitated using a c-Myc polyclonal antibody (Cell Signaling, Inc., Beverly, MA) according to the manufacturer's protocol. Western blot analysis was done as previously described using HIF-1{alpha} or HIF-2{alpha} antibodies (38). Primary antibodies were used at a dilution of 1:1,000. Secondary antibodies were horseradish peroxidase-conjugated, detected by enhanced chemiluminescence (Amersham, Piscataway, NJ), and exposed to autoradiographic film (Kodak, Rochester, NY).

Immunofluorescence. ES cells were grown on glass cover slips and exposed to hypoxia (1.5% O2) for 4 hours or maintained in normoxia (20% O2). Cells were fixed in 3% paraformaldehyde and permeabilized in 0.1% Triton in PBS. The Myc-epitope was detected with 9E10 monoclonal antibody (Abcam, Cambridge, MA). Cells were stained with FITC-conjugated immunoglobulin-G (Vector Labs, Burlingame, CA). DNA was detected with Hoechst dye (Sigma, St. Louis, MO).

Mouse Teratoma Assay. ES cells (5 x 106) in 100 µL of PBS were injected s.c. into the dorsal area of 4- to 6-week-old NIH-III immunodeficient mice (Charles River, Wilmington, MA). Tumors were measured every 2 to 3 days with calipers in the two greatest dimensions to calculate tumor growth over a 21-day period. At 21 days, tumors were harvested, photographed, weighed, fixed in 4% paraformaldehyde, and processed for histologic analysis.

Immunohistochemistry. Tumor samples were fixed in paraformaldehyde and paraffin-embedded by standard methods. Sections were stained by standard immunohistochemistry techniques with H&E or using antibodies generated against cleaved caspase-3 (Cell Signaling), Ki67 (Novocastra, United Kingdom), cyclin D1 (Calbiochem, San Diego, CA), CD34 (PharMingen, San Jose, CA), and VEGF (NeoMarkers, Fremont, CA). Secondary detection was done with biotinylated anti-mouse and anti-rat antibodies. Antibody staining was visualized using the avidin-biotin complex technique and 3,3'-diaminobenzidine (Vector Labs). Dilute hematoxylin was used for counterstain (Vector Labs). Morphometric analysis was done on three randomly chosen sections of each tumor (three tumors per genotype from each of the three independent experiments) using the analytical program Image Pro (Phase 3 Imaging) and by counting positive cells. For quantitative analysis of Ki67, cyclin D1, and cleaved caspase-3-staining, positive staining cells and microvessels were counted and their density expressed as the number of positive cells or capillaries per total section area, respectively. For quantitation of VEGF expression, the total positive staining area was calculated using the Image Pro software. Total number of cartilage, bone, Schwann cell bundles, ganglia, and areas of fat were counted per slide and six slides (representing three independent experiments) were averaged together to quantitate differentiated tissues.

Tomato Lectin-FITC Labeling. Flourescein-Lycopersicon esculentum (tomato) lectin (100 µL; Vector Labs) were injected into the mouse tail vein at a dose of 2 mg/mL. Tumors were harvested, fixed in 4% paraformaldehyde, and paraffin-embedded.

Statistical Analysis. Statistical analyses were done by Student's t test. Statistical significance was defined as P < 0.05.

Quantitative Reverse Transcription-PCR. Total RNA was purified from wild-type and KI ES cells or teratomas using Trizol reagent (Life Technologies). Mixed Oligo-d(T)15 and ribosomal 18S RNA-specific primers were used to generate single-stranded cDNAs, which were assayed for levels of VEGF-A, phosphoglycerate kinase (PGK), aldolase A (ALDA), TGF-{alpha}, adipose differentiation related protein (ADRP), cyclin D1 (CCND1), and 18S transcripts using an Applied Biosystems 7900HT Sequence Detection System. Mixed primer/probe sets for each transcript were obtained from Applied Biosystems and used according to the manufacturer's instructions. Expression levels of VEGF, PGK, ALDA, TGF-{alpha}, ADRP, and CCND1 transcripts were normalized to endogenous ribosomal 18S transcripts. Importantly, all gene expression data has been reproduced at least thrice. In Fig. 6D, RNA samples from two independent wild-type or KI teratomas were analyzed, pooled according to genotype, and comparisons made between genotypes by {Delta}{Delta}Ct comparative analysis using manufacturer's software.



View larger version (72K):
[in this window]
[in a new window]
  		Figure 6. Increased expression of VEGF and other HIF targets in KI teratomas. A and B, VEGF protein staining of wild-type (A) and KI (B) teratomas; C, columns, quantification of total area of positive staining using Image Pro Software. Bars, ±SE (P < 001), final magnification 400x; D, real-time PCR analysis of HIF target genes VEGF, PGK, TGF-{alpha}, ADRP, and CCND1. Values are expressed as the ratio of target gene expression in KI tumors to that in wild-type tumors. Target gene transcript levels were normalized in each case to internal 18S transcript levels.

 

   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Generation of Embryonic Stem Cells in which HIF-2{alpha} Expression Replaces HIF-1{alpha}. To generate a HIF-2{alpha} KI allele, a cDNA encoding a c-Myc epitope-tagged murine HIF-2{alpha} was introduced into the first coding exon of the murine Hif-1{alpha} gene through homologous recombination in murine ES cells. The specific HIF-2{alpha} cDNA used in these studies was previously shown to encode a functional protein capable of inducing HIF-2{alpha} target gene expression (22). The targeting vector was electroporated into ES cells and G418/gancyclovir-resistant colonies were screened by Southern blot (Fig. 1E). Following identification of a correctly targeted clone, ES cells homozygous for the Hif-1{alpha}2{alpha} allele were selected by treatment with increasing concentrations of G418 (Fig. 1E, lanes 5-7).



View larger version (28K):
[in this window]
[in a new window]
  		Figure 1. HIF-2{alpha} KI allele targeting scheme and generation of HIF-2{alpha} KI allele. A, targeting construct showing cDNA encoding c-Myc epitope-tagged HIF-2{alpha} fused to the SV40 polyadenylation sequence, the neomycin resistance gene cassette (NeoR), and loxP recombination sites (hatched triangles). Thick lines, 1.7 kb 5' homology arm and 6 kb 3' homology arm; B, wild-type HIF-1{alpha} allele, showing NcoI site at initiator ATG in coding exon 1; C, properly targeted KI Hif-1{alpha}2{alpha} allele, resulting from recombination between homology arms in targeting vector and endogenous Hif-1{alpha} gene; D, strategy for Cre recombinase-mediated deletion of NeoR cassette, producing final KI allele; E, Southern blot analysis of NcoI-digested DNA from ES cell clones heterozygous or homozygous for Hif-1{alpha}2{alpha} allele verifying correct targeting. Probe derived from sequences 5' to homology arm (hatched box) distinguishes the Hif-1{alpha}2{alpha} allele (4 kb) from wild-type (1.5 kb). Lane 1, wild-type cells; lanes 2-4, Hif-1{alpha}2{alpha}/+ heterozygous clones; lanes 5-7, homozygous Hif-1{alpha}2{alpha} clones; F, Southern blot analysis of heterozygous and homozygous ES cells following Cre-mediated deletion of the NeoR cassette. XbaI-digested genomic DNA was hybridized to an allele-specific probe derived from the SV40 polyadenylation sequence fused to the HIF-2{alpha} cDNA (solid box). Lane 1, wild-type; lane 2, heterozygous Hif-1{alpha}2{alpha}/+ cells; lanes 3 and 4, independent heterozygous KI/+ clones; lane 5, homozygous Hif-1{alpha}2{alpha} cells; lanes 6 and 7, homozygous KI clones. NeoR+, cells resistant to G418; NeoR–, cells sensitive to G418.

 
We expected the homozygous Hif-1{alpha}2{alpha} ES cells to be devoid of HIF-1{alpha} protein expression, as multiple stop codons exist in all three reading frames of the SV40 polyadenylation sequence placed immediately downstream of the cDNA in the targeting vector. However, initial characterization revealed that these cells expressed reduced levels of HIF-1{alpha} protein (Fig. 2A, lane 4) and failed to express the epitope-tagged HIF-2{alpha} protein (data not shown), raising the possibility that the PGK promoter of the NeoR cassette might interfere with proper expression of the HIF-2{alpha} cDNA transcript from the Hif-1{alpha} promoter. 5'-Rapid amplification of cDNA ends experiments confirmed this interpretation, demonstrating that a novel HIF-1{alpha} mRNA was transcribed in these cells, originating from the PGK promoter in the NeoR cassette (data not shown). This transcript contained an in-frame methionine codon 11 amino acids downstream of the initiator ATG, resulting in deletion of three lysine residues within the DNA binding domain. EMSA experiments also showed reduced HIF-1{alpha} DNA-binding activity in homozygous Hif-1{alpha}2{alpha} ES cells, consistent with the reduced protein levels observed in the Western blot analysis (Fig. 2B, lanes 5-8). Intriguingly, mice homozygous for this hypomorphic allele are viable and seem to be normal.4



View larger version (33K):
[in this window]
[in a new window]
  		Figure 2. Characterization of HIF-2{alpha} KI allele. A, Western blot analysis of HIF-1{alpha} protein in wild-type (lanes 1 and 2); homozygous Hif-1{alpha}2{alpha} (lanes 3 and 4); homozygous KI (lanes 5 and 6); or Hif-1{alpha}–/– (lanes 7 and 8) cells. N, normoxia; H, hypoxia; B, EMSA of HIF-1{alpha} DNA-binding complexes in nuclear extracts from hypoxic wild-type (WT), homozygous Hif-1{alpha}2{alpha}, and homozygous KI cells. Arrows, HIF/DNA complex induced under hypoxic conditions and supershifted with HIF-1ß antibody (*). Hif-1{alpha}2{alpha} cells contained reduced HIF-1{alpha} binding activity (lanes 5-8). No detectable HIF-1{alpha} DNA binding activity was observed in homozygous KI nuclear extracts (lanes 9-12); C, quantitative reverse transcription-PCR analysis of transcripts from the HIF-1{alpha} specific target gene ALDA in hypoxic wild-type, homozygous Hif-1{alpha}2{alpha}/Hif-1{alpha}2{alpha}, heterozygous KI/+, homozygous KI/KI, and Hif-1{alpha}–/– ES cells. ALDA transcript levels were normalized to 18S rRNA transcripts in each sample. Data are expressed as the ratio of ALDA expression in each sample relative to normoxic wild-type ES cells; D, immunoprecipitation with c-Myc epitope antibodies, followed by Western blot using HIF-2{alpha} antibodies, of extracts from wild-type (lanes 1 and 2), heterozygous (lanes 3 and 4) and homozygous (lanes 5-8) KI cells. N, normoxia; H, hypoxia; E, immunofluorescence of myc-tagged HIF-2{alpha} protein expression wild-type cells (top) and homozygous KI cells (bottom). Hoescht dye was used to stain DNA, and c-Myc antibodies to visualize myc-tagged HIF-2{alpha} protein. Merged images show nuclear myc-tagged HIF-2{alpha} expression in all homozygous KI cells, but not in wild-type cells.

 
To activate expression of the HIF-2{alpha}-myc cDNA in the KI allele, the NeoR cassette was deleted in both heterozygous and homozygous ES cells through transient expression of Cre recombinase (Fig. 1F, lanes 3, 4, 6, and 7). Deletion of the NeoR cassette removed the PGK promoter sequences driving expression of the novel HIF-1{alpha} transcript observed in the Hif-1{alpha}2{alpha} cells. Further experiments revealed that ES cells homozygous for this recombined allele, designated "KI", exhibited no detectable HIF-1{alpha} protein by Western blot (Fig. 2A, lane 6) and EMSA analyses (Fig. 2B, lanes 9-12). Hypoxic expression of the HIF-1{alpha}-specific target gene ALDA (22) was also reduced to levels observed in Hif-1{alpha}–/– cells, confirming the absence of any functional HIF-1{alpha} protein in KI ES cells (Fig. 2C). Similar results were obtained for PGK, another HIF-1{alpha}-specific target gene (data not shown; ref. 22). As HIF-2{alpha} protein is not readily detected in cell extracts using standard EMSA protocols,5 alternate methods were employed to detect epitope-tagged HIF-2{alpha} protein in homozygous KI cells. Immunoprecipitation of cell extracts with a c-Myc epitope-specific antibody, followed by immunoblot analysis with HIF-2{alpha} antiserum, revealed that c-Myc-tagged HIF-2{alpha} protein is expressed and subject to oxygen-dependent regulation (Fig. 2D, lanes 3-8). In addition, immunofluorescence experiments using the Myc antibody showed that the epitope-tagged HIF-2{alpha} protein is expressed in all KI cells and localized to the nucleus (Fig. 2E). Together, these data indicate that HIF-1{alpha} expression has been replaced by HIF-2{alpha} in homozygous KI cells, and that Myc-tagged HIF-2{alpha} protein accumulation is regulated by O2 levels.

Knock-in Allele Promotes Growth of Larger, More Hemorrhagic Tumors in Nude Mice. To investigate whether replacement of HIF-1{alpha} by HIF-2{alpha} could alter growth characteristics of ES-derived teratomas, wild-type and homozygous KI ES cells were injected s.c. into NIH-III immunodeficient mice in three separate experiments (total n = 18 for each genotype). Teratomas appeared over a period of 10 to 21 days, but those generated from two independently derived KI ES cell lines grew more rapidly, exhibited increased mass, and seemed more hemorrhagic than wild-type controls (Fig. 3A-C). The differences in mass were statistically significant (P = 0.012 by Student's t test). In contrast, KI and Hif-1{alpha}–/– ES cells showed essentially identical proliferation rates in vitro, whereas wild-type ES cells grew slightly faster (Fig. 3D). Thus, the growth advantage noted for KI cells was only observed in the context of tumor growth. The expression of the Myc-tagged HIF-2{alpha} transcripts in KI tumors was confirmed by reverse transcription-PCR using allele-specific primers (data not shown).



View larger version (22K):
[in this window]
[in a new window]
  		Figure 3. Increased growth of KI tumors. A, gross appearance of teratomas resected from nude mice 21 days after s.c. injection with wild-type or homozygous KI ES cells; B, final tumor masses of wild-type and homozygous KI tumor formed after 21 days of growth. Bars, ±SE (P = 0.012); C, tumor volume measurements over 21 days show more rapid growth of KI teratomas; D, growth curves for wild-type, homozygous KI/KI and Hif-1{alpha}–/– ES cells grown in vitro under hypoxic conditions (1.5% O2). Essentially identical results were obtained under normoxic conditions (data not shown).

 
To determine whether the difference in growth rates between wild-type and KI tumors was due to increased proliferation or decreased apoptosis, we analyzed teratomas by immunostaining for Ki67 and cleaved-caspase 3, respectively (Fig. 4). In three independent experiments, KI tumors showed a 2.5-fold increase in Ki67 staining (Fig. 4A-C), whereas no quantitative differences in cleaved-caspase 3 staining were observed between the two genotypes (Fig. 4D-F). As cyclin D1 expression is induced by hypoxia in RCC cells expressing HIF-2{alpha} (39), immunohistochemistry was done for cyclin D1 in wild-type and KI tumors. Overall, KI tumors showed increased nuclear cyclin D1 immunostaining compared with wild-type tumors (Fig. 4G-I). Together, our data indicate that expanded HIF-2{alpha} expression, under the control of the Hif-1{alpha} locus, correlates with increased proliferation and larger tumor size.



View larger version (68K):
[in this window]
[in a new window]
  		Figure 4. KI teratomas exhibit an increase in proliferation, but no change in apoptosis. A and B, Ki67 staining of wild-type (A) and KI (B) teratoma sections; C, quantitation of Ki67 staining. Results are the average of three independent teratoma experiments. Bars, ±SE (P = 0.05); D and E, cleaved caspase-3 staining of wild-type (D) and KI (E) teratomas; F, quantitation of activated caspase-3-positive cells; G and H, cyclin D1 staining of wild-type (G) and KI (H) tumors; I, quantitation of cyclin D1-positive nuclei. Final magnifications are 400x (A and B) and 200x (D-H).

 
Knock-in Tumors Exhibit Increased Microvessel Density. Gross morphologic analysis revealed increased hemorrhaging of KI tumors (Fig. 3A). Histologic analysis also revealed larger blood-filled cavities and hemangiomas in KI tumors compared with those derived from wild-type ES cells (compare Fig. 7A with B). These data suggest that the KI allele promotes tumor angiogenesis. To assess the microvessel density of wild-type and KI teratomas, tumor sections were immunostained with an endothelial cell-specific marker, CD34 (Fig. 5A and B). Quantitation of CD34 staining revealed an increased microvessel density in KI tumors (19.6 ± 2.3/20x field) compared with wild-type (8.8 ± 1.6/20x field; Fig. 5E). Results are the averages of three independent experiments (three random sections of three different tumors for each genotype per experiment) and are statistically significant (P = 0.01 by Student's t test). The pattern of increased microvessel density in KI tumors is consistent with the increased vasculature observed in tumors from mice injected with fluorescently labeled tomato lectin (Fig. 5C and D).



View larger version (114K):
[in this window]
[in a new window]
  		Figure 7. Increased differentiation in KI teratomas. A-H, H&E staining of wild-type (A) and KI (B-H) teratomas. A, primitive neuroectoderm in wild-type teratomas. Endodermal and mesodermal derivatives were also observed in different sections. KI teratomas (B-H) contained regions marked either highly undifferentiated cells (B) or highly differentiated tissues (C) such as cartilage (D), bone (E), Schwann cells, ganglia (F), and adipocytes (G). H&E staining also revealed increased hemangioma (B) and endothelioma (H). n, neuroectoderm; c, cartilage; b, bone; s, Schwann cells; g, ganglia; f, fat; h, hemangiomas (arrowheads). I, quantitation of differentiated tissues in wild-type and KI teratomas (see Materials and Methods).

 


View larger version (68K):
[in this window]
[in a new window]
  		Figure 5. Increased vascular density in KI tumors. A and B, CD-34 staining of wild-type (A) and KI (B) teratomas; C and D, FITC-lectin images of wild-type (C) and KI (D) teratomas; E, quantitation of tumor microvessel density. Columns, mean; bars, ±SE (P = 0.01). Final magnification 200x.

 
Knock-in Tumors Display Increased VEGF and TGF-{alpha} Expression. We reasoned that the increased microvessel density and aberrant vessels observed in the KI tumors could reflect increased expression of angiogenic HIF target genes by HIF-2{alpha}. This idea is consistent with a previous report demonstrating that HIF-2{alpha} induces VEGF expression more potently than HIF-1{alpha} in Chinese hamster ovary Ka13 cells (40). To test this hypothesis, we did immunohistochemical stains for VEGF (Fig. 6A and B). KI tumors exhibited increased VEGF staining compared with wild-type tumors (Fig. 6C). Furthermore, the VEGF expression pattern correlated with the observed differences in microvessel density, increased numbers of hemangiomas, gross hemorrhaging, and presence of endotheliomas (Fig. 7H). These results suggest that the greater vascularity observed in KI tumors results at least partly from increased VEGF expression.

Quantitative reverse transcription-PCR analysis of tumor RNAs confirmed increased levels of VEGF transcripts in KI tumors, although the difference compared with wild-type tumors was modest (1.3 ± 0.05-fold; Fig. 6D). Intriguingly, expression of the HIF-2{alpha} target gene TGF-{alpha} (41) showed a substantial increase (3.05 ± 0.06-fold) in the KI teratomas, raising the possibility that increased TGF-{alpha} expression in these tumors may be responsible, at least in part, for the phenotypes observed. In addition to VEGF and TGF-{alpha}, mRNA levels for the HIF-2{alpha} target gene ADRP (22) were also modestly increased (1.3 ± 0.07-fold change) in KI tumors. Surprisingly, cyclin D1 transcript levels were slightly lower in KI tumors (0.87 ± 0.05-fold), suggesting that the increase in cyclin D1 protein levels observed in KI tumors (Fig. 4) was caused by a posttranscriptional mechanism. Finally, the expression of PGK mRNA was significantly reduced (0.55 ± 0.03-fold) in KI tumors relative to wild-type, consistent with previous reports that genes encoding glycolytic enzymes are regulated only by HIF-1{alpha}, but not by HIF-2{alpha} (22, 23). The differences between VEGF and cyclin D1 transcript and protein levels in KI tumors may reflect the indirect effects of expanded HIF-2{alpha} expression, as discussed below.

Knock-in Tumors Exhibit a Wide Range of Differentiated Cell Types. To investigate the degree of differentiation and spectrum of cell types present, tumor sections were evaluated by standard histologic methods. As expected, wild-type tumors exhibited tissues derived from all three germ layers. In addition, wild-type tumors contained predominately neuroectodermal lineages characterized by oval-shaped nuclei arranged in radiating rosettes (Fig. 7A). KI tumors differed from wild-type tumors in two notable ways: (a) KI tumors exhibited an expansion of apparently undifferentiated cells (Fig. 7B), and (b) KI tumors showed a substantial increase in specific highly differentiated structures (Fig. 7C). Compared with wild-type controls, KI tumors displayed 5- to 10-fold more cartilage, bone, neuronal cell types (Schwann cell and ganglia), fat cells, and endotheliomas (Fig. 7D-I).


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
A great deal of work over the past 10 years has shown that hypoxia is a critical factor regulating angiogenesis and cellular metabolism in tumors (30). The hypoxically regulated transcription factors HIF-1{alpha} and HIF-2{alpha} are overexpressed in a large number of human tumor types, including brain, prostate, breast, RCC, pheochromocytomas, astrocytomas, and non–small cell lung cancers (21, 42–44), consistent with the idea that each contributes to tumor growth by activating the expression of critical target genes. Although both HIF-1{alpha} and HIF-2{alpha} are expressed in many of the same human tumors and tumor cell lines (29, 45), the relative contribution of each protein to tumor initiation and progression has been difficult to determine. Genetic ablation experiments have confirmed a role for both proteins in tumor growth; for example, s.c. tumors derived from Hif-1{alpha}–/– mouse embryonic fibroblasts and ES cells grew less rapidly than wild-type controls (10, 31) . More recently, Kondo et al. (33) and Maranchie et al. (34) used complementary experimental approaches to show that expression of HIF-2{alpha}, but not HIF-1{alpha}, promoted the growth of RCC xenografts. In particular, it was shown that inhibition of HIF-2{alpha} is sufficient to suppress tumor formation by a RCC line lacking the von Hippel-Lindau tumor suppressor protein (32). The apparently unique activity of HIF-2{alpha} in promoting the growth of RCC, and possibly other tumors, could represent differences in the cell type-specific expression patterns of HIF-1{alpha} and HIF-2{alpha}, and/or differences in the spectrum of target genes they regulate. Clarifying this issue will be critical for understanding the mechanisms by which hypoxia affects the progression of many cancers.

To better distinguish the unique biological activities conferred by HIF-1{alpha} and HIF-2{alpha} in tumor growth and embryonic development, we created a KI allele of the murine Hif-1{alpha} gene. In this allele, a HIF-2{alpha} cDNA clone was introduced into the Hif-1{alpha} genomic locus, thereby replacing HIF-1{alpha} expression with HIF-2{alpha}. This novel allele permits a direct comparison of how HIF-1{alpha} and HIF-2{alpha} function when expressed independently, but under the influence of the same cis-acting regulatory sequences, in primary cells. In this report, we used the KI allele to investigate the relative contribution of HIF-1{alpha} and HIF-2{alpha} to the growth of ES cell-derived teratomas. This approach avoids the use of transgene constructs, which may produce nonphysiologic expression levels or patterns, or transformed cells in which confounding oncogenic mutations can dysregulate or override HIF function. Although a full appreciation of HIF-1{alpha} and HIF-2{alpha} function in human cancers will ultimately require an understanding of their interaction with other oncogenic mutations, these experiments represent an important step toward dissecting their unique and shared activities in tumor biology.

The enhanced proliferative and vascular phenotypes observed in KI tumors probably reflect the expanded and augmented expression of specific HIF-2{alpha} transcriptional targets, relative to wild-type cells. As multiple analyses have shown that HIF-2{alpha} regulates VEGF, cyclin D1, and TGF-{alpha} in RCC carcinoma lines (39, 41) and ADRP in HEK293 cells (22), the simplest model to explain our data would invoke direct transcriptional activation of these and other target genes by HIF-2{alpha} in the KI tumors. Immunohistochemical analysis indeed revealed increased VEGF and cyclin D1 protein expression, as well as modestly increased VEGF and ADRP mRNA expression in KI tumors. Cyclin D1 mRNA levels were, paradoxically, slightly decreased in these tumors, suggesting that the increased nuclear expression of cyclin D1 results from posttranscriptional mechanisms such as increased protein stability, as previously described (46). Whereas HIF-1{alpha}-specific target genes ALDA and PGK were significantly down-regulated in KI tumors, transcripts from other target genes previously shown to be common targets of HIF-1{alpha} and HIF-2{alpha}, including glucose transporter 1 and adrenomedullin 1, were expressed at nearly identical levels in wild-type and KI teratomas (data not shown). In contrast, TGF-{alpha} transcript levels were dramatically increased in KI tumors, suggesting that this target gene may be preferentially induced by HIF-2{alpha}. Together, these data indicate that the phenotype of the KI cells was not simply due to loss of HIF-1{alpha} function, but resulted from expanded expression of HIF-2{alpha} under the control of the HIF-1{alpha} locus. Moreover, the results reported here indicate that the preferential tumor-promoting effects of HIF-2{alpha} previously reported for RCCs (32) are more generally applicable.

Our data show clear differences between KI and wild-type ES cells that correlate to expression of the HIF-2{alpha} cDNA from the Hif-1{alpha} locus. Although HIF activity is primarily controlled at the level of protein stabilization, it is possible that the expression of a targeted cDNA may modulate overall expression of the locus, thereby affecting the phenotype of the cells. To address this concern, we recently generated ES cells in which the HIF-1{alpha} cDNA has been targeted to the Hif-1{alpha} locus in a manner identical to the HIF-2{alpha} cDNA. Preliminary data indicate that ES cells homozygous for this HIF-1{alpha} cDNA KI allele express ALDA, VEGF, and TGF-{alpha} transcripts at levels similar to wild-type ES cells, in stark contrast to the HIF-2{alpha} KI cells. Furthermore, these HIF-1{alpha} cDNA KI cells behave essentially identically to wild-type ES cells in their ability to generate hematopoietic progenitors from embryoid bodies, an assay in which the HIF-2{alpha} KI cells are strikingly deficient (data not shown). These and other preliminary data support our interpretation that the phenotypes exhibited by the HIF-2{alpha} KI ES cells are due to functional differences between HIF-1{alpha} and HIF-2{alpha}, and are not simply a consequence of expressing a targeted cDNA from the Hif-1{alpha} locus. Definitive conclusions, however, will require further comparison of HIF-1{alpha} KI and HIF-2{alpha} KI ES cells.

It is possible that the phenotypes observed for the KI teratomas result, at least in part, from indirect effects of expanded HIF-2{alpha} expression. The increase in TGF-{alpha} mRNA levels in KI tumors is of particular interest in this regard, as overexpression of TGF-{alpha} has been shown to profoundly affect cellular proliferation and differentiation (47–49). TGF-{alpha} expression has been correlated to increased cyclin D1 expression in a variety of cellular contexts (50, 51), and also to increased tumor microvessel density (52, 53). It is therefore possible that TGF-{alpha} signaling through the epidermal growth factor receptor may be a critical downstream effector of HIF-2{alpha} with respect to VEGF and cyclin D1 protein accumulation. It is noteworthy that the hypoxia-induced expression of TGF-{alpha} in RCC cells that express HIF-2{alpha} (but not HIF-1{alpha}) is critical for the serum-independent growth of these cells (41). Further work will be required to thoroughly evaluate the effects of HIF-2{alpha}-mediated TGF-{alpha} expression on tumor initiation and growth.

The particular spectrum and abundance of bone, cartilage, and neural cells observed in the KI teratomas raises questions about the activity of HIF-2{alpha} in multiple developmental pathways. Genetic ablation of HIF-2{alpha} does not seem to disrupt the development of these tissues (25–27), perhaps because HIF-1{alpha} can compensate for the loss of HIF-2{alpha} in these cell lineages. In contrast, our results suggest that when HIF-2{alpha} is broadly expressed in lieu of HIF-1{alpha}, the balance of signals controlling cellular differentiation is altered. Whether this imbalance results from markedly increased expression of particular HIF-2{alpha} targets such as TGF-{alpha}, or from more subtle changes in a wide range of HIF target genes, will be an important question to address in future work. Irrespective of the mechanisms, it would seem that expanded HIF-2{alpha} expression from the KI allele promotes differentiation of bone, cartilage, adipocytes, and endothelial structures from ES cells. A thorough evaluation of HIF target gene expression in differentiated primary cells carrying the KI allele may help to further distinguish the unique and overlapping functions of HIF-1{alpha} and HIF-2{alpha} in development and tumor progression.


   Acknowledgments
 
Grant support: NIH Grants HL66130 (M.C. Simon and B. Keith), American Heart Association 0415362U (K.L. Covello), and the Abramson Family Cancer Research Institute. M.C. Simon is an investigator of the Howard Hughes Medical Institute.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank Q.C. Yu, Frank Winslow, Anja Runge, Fiona Mack, Mercy Gohil, and Michelle Mooney for reagents and technical assistance.


   Footnotes
 
4 K.L. Covello and M.C. Simon, in preparation. Back

5 B. Keith, unpublished observations. Back

Received 9/ 7/04. Revised 1/ 4/05. Accepted 1/13/05.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Semenza GL. HIF-1 and human disease: one highly involved factor. Genes Dev 2000;14:1983–91.[Free Full Text]
  2. Wang GL, Jiang B-H, Rue EA, Semenza GL. Hypoxia-inducible factor 1 is a basic-helix-loop-helix-PAS heterodimer regulated by cellular O2 tension. Proc Natl Acad Sci U S A 1995;92:5510–4.[Abstract/Free Full Text]
  3. Wenger RH, Gassmann M. Oxygen(es) and the hypoxia-inducible factor-1. Biol Chem 1997;378:609–16.[Medline]
  4. Pugh CW, Tan CC, Jones RW, Ratcliffe PJ. Functional analysis of an oxygen-regulated transcriptional enhancer lying 3' to the mouse erythropoietin gene. Proc Natl Acad Sci U S A 1991;88:10553–7.[Abstract/Free Full Text]
  5. Semenza GL, Nejfelt MK, Chi SM, Antonarakis SE. Hypoxia-inducible nuclear factors bind to an enhancer element located 3' to the human erythropoietin gene. Proc Natl Acad Sci U S A 1991;88:5680–4.[Abstract/Free Full Text]
  6. Iyer NV, Kotch LE, Agani F, et al. Cellular and developmental control of O2 homeostasis by hypoxia-inducible factor 1{alpha}. Genes Dev 1998;12:149–62.[Abstract/Free Full Text]
  7. Wood SM, Gleadle JM, Pugh CW, Hankinson O, Ratcliffe PJ. The role of aryl hydrocarbon receptor nuclear translocator (ARNT) in hypoxia induction of gene expression. J Biol Chem 1996;271:15117–23.[Abstract/Free Full Text]
  8. Wood SM, Wiesener MS, Yeates KM, et al. Selection and analysis of a mutant cell line defective in the hypoxia-inducible factor-1{alpha}-subunit (HIF-1{alpha}). Characterization of hif-1{alpha}-dependent and -independent hypoxia-inducible gene expression. J Biol Chem 1998;273:8360–8.[Abstract/Free Full Text]
  9. Gerber HP, Condorelli F, Park J, Ferrara N. Differential transcriptional regulation of the two vascular endothelial growth factor receptor genes. Flt-1, but not Flk-1/KDR, is up-regulated by hypoxia. J Biol Chem 1997;272:23659–67.[Abstract/Free Full Text]
  10. Ryan HE, Lo J, Johnson RS. HIF-1{alpha} is required for solid tumor formation and embryonic vascularization. EMBO J 1998;17:3005–15.[CrossRef][Medline]
  11. Maxwell PH, Wiesener MS, Chang GW, et al. The tumour suppressor protein VHL targets hypoxia-inducible factors for oxygen-dependent proteolysis. Nature 1999;399:271–5.[CrossRef][Medline]
  12. Cockman ME, Masson N, Mole DR, et al. Hypoxia inducible factor-{alpha} binding and ubiquitylation by the von Hippel-Lindau tumor suppressor protein. J Biol Chem 2000;275:25733–41.[Abstract/Free Full Text]
  13. Ohh M, Park CW, Ivan M, et al. Ubiquitination of hypoxia-inducible factor requires direct binding to the ß-domain of the von Hippel-Lindau protein. Nat Cell Biol 2000;2:423–7.[CrossRef][Medline]
  14. Kallio PJ, Wilson WJ, O'Brien S, Makino Y, Poellinger L. Regulation of the hypoxia-inducible transcription factor 1{alpha} by the ubiquitin-proteasome pathway. J Biol Chem 1999;274:6519–25.[Abstract/Free Full Text]
  15. Salceda S, Caro J. Hypoxia-inducible factor 1{alpha} (HIF-1{alpha}) protein is rapidly degraded by the ubiquitin-proteasome system under normoxic conditions. Its stabilization by hypoxia depends on redox-induced changes. J Biol Chem 1997;272:22642–7.[Abstract/Free Full Text]
  16. Sowter HM, Raval R, Moore J, Ratcliffe PJ, Harris AL. Predominant role of hypoxia-inducible transcription factor (Hif)-1{alpha} versus Hif-2{alpha} in regulation of the transcriptional response to hypoxia. Cancer Res 2003;63:6130–4.[Abstract/Free Full Text]
  17. Park SK, Dadak AM, Haase VH, Fontana L, Giaccia AJ, Johnson RS. Hypoxia-induced gene expression occurs solely through the action of hypoxia-inducible factor 1{alpha} (HIF-1{alpha}): role of cytoplasmic trapping of HIF-2{alpha}. Mol Cell Biol 2003;23:4959–71.[Abstract/Free Full Text]
  18. Ema M, Taya S, Yokotani N, Sogawa K, Matsuda Y, Fujii-Kuriyama Y. A novel bHLH-PAS factor with close sequence similarity to hypoxia-inducible factor 1{alpha} regulates the VEGF expression and is potentially involved in lung and vascular development. Proc. Natl Acad Sci U S A 1997;94:4273–8.[Abstract/Free Full Text]
  19. Flamme I, Frohlich T, von Reutern M, Kappel A, Damert A, Risau W. HRF, a putative basic helix-loop-helix-PAS-domain transcription factor is closely related to hypoxia-inducible factor-1{alpha} and developmentally expressed in blood vessels. Mech Dev 1997;63:51–60.[CrossRef][Medline]
  20. Tian H, McKnight SL, Russell DW. Endothelial PAS domain protein 1 (EPAS1), a transcription factor selectively expressed in endothelial cells. Genes Dev 1997;11:72–82.[Abstract/Free Full Text]
  21. Jaakkola P, Mole DR, Tian YM, et al. Targeting of HIF-{alpha} to the von Hippel-Lindau ubiquitylation complex by O2-regulated prolyl hydroxylation. Science 2001;292:468–72.[Abstract/Free Full Text]
  22. Hu CJ, Wang LY, Chodosh LA, Keith B, Simon MC. Differential roles of hypoxia-inducible factor 1{alpha} (HIF-1{alpha}) and HIF-2{alpha} in hypoxic gene regulation. Mol Cell Biol 2003;23:9361–74.[Abstract/Free Full Text]
  23. Seagroves TN, Ryan HE, Lu H, et al. Transcription factor HIF-1 is a necessary mediator of the Pasteur effect in mammalian cells. Mol Cell Biol 2001;21:3436–44.[Abstract/Free Full Text]
  24. Wiesener MS, Jurgensen JS, Rosenberger C, et al. Widespread hypoxia-inducible expression of HIF-2{alpha} in distinct cell populations of different organs. FASEB J 2003;17:271–3.[Abstract/Free Full Text]
  25. Peng J, Zhang L, Drysdale L, Fong GH. The transcription factor EPAS-1/hypoxia-inducible factor 2{alpha} plays an important role in vascular remodeling. Proc Natl Acad Sci U S A 2000;97:8386–91.[Abstract/Free Full Text]
  26. Compernolle V, Brusselmans K, Acker T, et al. Loss of HIF-2{alpha} and inhibition of VEGF impair fetal lung maturation, whereas treatment with VEGF prevents fatal respiratory distress in premature mice. Nat Med 2002;8:702–10.[Medline]
  27. Tian H, Hammer RE, Matsumoto AM, Russell DW, McKnight SL. The hypoxia-responsive transcription factor EPAS1 is essential for catecholamine homeostasis and protection against heart failure during embryonic development. Genes Dev 1998;12:3320–4.[Abstract/Free Full Text]
  28. Scortegagna M, Ding K, Oktay Y, et al. Multiple organ pathology, metabolic abnormalities and impaired homeostasis of reactive oxygen species in Epas1(–/–) mice. Nat Genet 2003.
  29. Talks KL, Turley H, Gatter KC, et al. The expression and distribution of the hypoxia-inducible factors HIF-1{alpha} and HIF-2{alpha} in normal human tissues, cancers, and tumor-associated macrophages. Am J Pathol 2000;157:411–21.[Abstract/Free Full Text]
  30. Hopfl G, Ogunshola O, Gassmann M. HIFs and tumors—causes and consequences. Am J Physiol Regul Integr Comp Physiol 2004;286:R608–23.[Abstract/Free Full Text]
  31. Ryan HE, Poloni M, McNulty W, et al. Hypoxia-inducible factor-1{alpha} is a positive factor in solid tumor growth. Cancer Res 2000;60:4010–5.[Abstract/Free Full Text]
  32. Kondo K, Kim WY, Lechpammer M, Kaelin WG Jr. Inhibition of HIF2{alpha} is sufficient to suppress pVHL-defective tumor growth. PLoS Biol 2003;1:E83.[Medline]
  33. Kondo K, Klco J, Nakamura E, Lechpammer M, Kaelin WG Jr. Inhibition of HIF is necessary for tumor suppression by the von Hippel-Lindau protein. Cancer Cell 2002;1:237–46.[CrossRef][Medline]
  34. Maranchie JK, Vasselli JR, Riss J, Bonifacino JS, Linehan WM, Klausner RD. The contribution of VHL substrate binding and HIF1-{alpha} to the phenotype of VHL loss in renal cell carcinoma. Cancer Cell 2002;1:247–55.[CrossRef][Medline]
  35. Wenger RH, Camenisch G, Desbaillets I, Chilov D, Gassmann M. Up-regulation of hypoxia-inducible factor-1{alpha} is not sufficient for hypoxic/anoxic p53 induction. Cancer Res 1998;58:5678–80.[Abstract/Free Full Text]
  36. Tybulewicz VL, Crawford CE, Jackson PK, Bronson RT, Mulligan RC. Neonatal lethality and lymphopenia in mice with a homozygous disruption of the c-abl proto-oncogene. Cell 1991;65:1153–63.[CrossRef][Medline]
  37. Carmeliet P, Dor Y, Herbert JM, et al. Role of HIF-1{alpha} in hypoxia-mediated apoptosis, cell proliferation and tumour angiogenesis. Nature 1998;394:485–90.[CrossRef][Medline]
  38. Mack FA, Rathmell WK, Arsham AM, Gnarra J, Keith B, Simon MC. Loss of pVHL is sufficient to cause HIF dysregulation in primary cells but does not promote tumor growth. Cancer Cell 2003;3:75–88.[CrossRef][Medline]
  39. Wykoff CC, Sotiriou C, Cockman ME, et al. Gene array of VHL mutation and hypoxia shows novel hypoxia-induced genes and that cyclin D1 is a VHL target gene. Br J Cancer 2004;90:1235–43.[CrossRef][Medline]
  40. Wiesener MS, Turley H, Allen WE, et al. Induction of endothelial PAS domain protein-1 by hypoxia: characterization and comparison with hypoxia-inducible factor-1{alpha}. Blood 1998;92:2260–8.[Abstract/Free Full Text]
  41. Gunaratnam L, Morley M, Franovic A, et al. Hypoxia inducible factor activates the transforming growth factor-{alpha}/epidermal growth factor receptor growth stimulatory pathway in VHL(–/–) renal cell carcinoma cells. J Biol Chem 2003;278:44966–74.[Abstract/Free Full Text]
  42. Favier J, Kempf H, Corvol P, Gasc JM. Coexpression of endothelial PAS protein 1 with essential angiogenic factors suggests its involvement in human vascular development. Dev Dyn 2001;222:377–88.[CrossRef][Medline]
  43. Khatua S, Peterson KM, Brown KM, et al. Overexpression of the EGFR/FKBP12/HIF-2{alpha} pathway identified in childhood astrocytomas by angiogenesis gene profiling. Cancer Res 2003;63:1865–70.[Abstract/Free Full Text]
  44. Giatromanolaki A, Koukourakis MI, Sivridis E, et al. Relation of hypoxia inducible factor 1{alpha} and 2{alpha} in operable non-small cell lung cancer to angiogenic/molecular profile of tumours and survival. Br J Cancer 2001;85:881–90.[CrossRef][Medline]
  45. Blancher C, Moore JW, Talks KL, Houlbrook S, Harris AL. Relationship of hypoxia-inducible factor (HIF)-1{alpha} and HIF-2{alpha} expression to vascular endothelial growth factor induction and hypoxia survival in human breast cancer cell lines. Cancer Res 2000;60:7106–13.[Abstract/Free Full Text]
  46. Diehl JA, Cheng M, Roussel MF, Sherr CJ. Glycogen synthase kinase-3ß regulates cyclin D1 proteolysis and subcellular localization. Genes Dev 1998;12:3499–511.[Abstract/Free Full Text]
  47. Sandgren EP, Luetteke NC, Palmiter RD, Brinster RL, Lee DC. Overexpression of TGF {alpha} in transgenic mice: induction of epithelial hyperplasia, pancreatic metaplasia, and carcinoma of the breast. Cell 1990;61:1121–35.[CrossRef][Medline]
  48. Matsui Y, Halter SA, Holt JT, Hogan BL, Coffey RJ. Development of mammary hyperplasia and neoplasia in MMTV-TGF{alpha} transgenic mice. Cell 1990;61:1147–55.[CrossRef][Medline]
  49. Jhappan C, Stahle C, Harkins RN, Fausto N, Smith GH, Merlino GT. TGF {alpha} overexpression in transgenic mice induces liver neoplasia and abnormal development of the mammary gland and pancreas. Cell 1990;61:1137–46.[CrossRef][Medline]
  50. Yan YX, Nakagawa H, Lee MH, Rustgi AK. Transforming growth factor-{alpha} enhances cyclin D1 transcription through the binding of early growth response protein to a cis-regulatory element in the cyclin D1 promoter. J Biol Chem 1997;272:33181–90.[Abstract/Free Full Text]
  51. Getchell TV, Narla RK, Little S, Hyde JF, Getchell ML. Horizontal basal cell proliferation in the olfactory epithelium of transforming growth factor-{alpha} transgenic mice. Cell Tissue Res 2000;299:185–92.[Medline]
  52. Soares R, Pereira MB, Silva C, et al. Expression of TGF-{alpha} and EGFR in breast cancer and its relation to angiogenesis. Breast J 2000;6:171–7.[CrossRef][Medline]
  53. Li Z, Shimada Y, Uchida S, et al. TGF-{alpha} as well as VEGF, PD-ECGF and bFGF contribute to angiogenesis of esophageal squamous cell carcinoma. Int J Oncol 2000;17:453–60.[Medline]



This article has been cited by other articles:


Home page
 CarcinogenesisHome page
S. Mahajan, V. Dammai, T. Hsu, and A.S. Kraft
Hypoxia-inducible factor-2{alpha} regulates the expression of TRAIL receptor DR5 in renal cancer cells
Carcinogenesis, September 1, 2008; 29(9): 1734 - 1741.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
A. Taguchi, K. Yanagisawa, M. Tanaka, K. Cao, Y. Matsuyama, H. Goto, and T. Takahashi
Identification of Hypoxia-Inducible Factor-1{alpha} as a Novel Target for miR-17-92 MicroRNA Cluster
Cancer Res., July 15, 2008; 68(14): 5540 - 5545.
[Abstract] [Full Text] [PDF]