Genetic and Functional Studies Implicate Hif1aas a 14Q Kidney Cancer Suppressor Gene

Published OnlineFirst June 7, 2011; DOI: 10.1158/2159-8290.CD-11-0098

RESEARCH ARTICLE Genetic and Functional Studies Implicate HIF1a as a 14q Kidney Cancer Suppressor Gene

Chuan Shen1 Rameen Beroukhim1,2,3,5 Steven E. Schumacher2,5 Jing Zhou3 Michelle Chang4 Sabina Signoretti1,4 William G. Kaelin, Jr1,3,5,6

The BATTLE Trial: Personalizing Therapy for Lung Cancer RESEARCH ARTICLE

VHL A B S T R A C T Kidney cancers often delete chromosome 3p, spanning the tumor suppressor gene, and chromosome 14q, which presumably harbors ≥1 tumor suppressor genes. pVHL inhibits the hypoxia-inducible transcription factor (HIF), and HIF2α is a kidney cancer oncoprotein. In this article, we identify focal, homozygous deletions of the HIF1 α locus on 14q in clear cell renal carcinoma cell lines. Wild-type HIF1α suppresses renal carcinoma growth, but the products of these altered loci do not. Conversely, downregulation of HIF1α in HIF1α- proficient lines promotes tumor growth. HIF1α activity is diminished in 14q-deleted kidney cancers, and all somatic HIF1α mutations identified in kidney cancers tested to date are loss of function. Therefore, HIF1α has the credentials of a kidney cancer suppressor gene.

SIGNIFICANCE: Deletion of 14q is a frequent event in clear cell renal carcinoma and portends a poor prognosis. In this study, we provide genetic and functional evidence that HIF1α is a target of 14q loss in kidney cancer. Cancer Discovery; 1(3); 222–35. ©2011 AACR.

INTRODUCTION number of drugs that inhibit VEGF, or its receptor KDR, Kidney cancer causes >10,000 deaths each year in the have demonstrated significant activity in the treatment of United States ( 1 ). Although surgery is potentially curative for metastatic kidney cancer (7 ). α kidney cancers that are detected at an early stage, recurrences Multiple lines of evidence suggest that HIF2 , and not α after surgery remain common, and late-stage, inoperable kid- its more intensively studied paralog HIF1 , acts as a driver ney cancer is usually fatal. in pVHL-defective renal carcinomas. For example, pVHL- Clear cell renal carcinoma is the most common form of defective renal carcinoma cell lines and tumors produce both α α α kidney cancer and is usually linked to biallelic inactivation of HIF1 and HIF2 or HIF2 alone (6 , 8 ), and the appearance α the von Hippel–Lindau VHL tumor suppressor gene, which is of HIF2 in preneoplastic lesions in the kidneys of patients located on chromosome 3p25. Individuals who carry a mu- with von Hippel–Lindau disease correlates with increased tant VHL allele in the germline (von Hippel–Lindau disease) histologic evidence of impending malignancy ( 9 ). Moreover, α α are at increased risk of clear cell renal carcinoma, in addition HIF2 , but not HIF1 , can override the tumor suppressor α to pheochromocytomas and central nervous system heman- activity of pVHL ( 10–12 ), whereas eliminating HIF2 is suf- gioblastomas. Somatic mutation, or hypermethylation, of the ficient to suppress tumor formation by pVHL-defective re- VHL locus is also common in sporadic clear cell renal carci- nal carcinoma cells in preclinical models ( 13 , 14 ). A recent nomas (2 ). genome-wide association study linked the risk of renal carci- HIF2α α The VHL gene product, pVHL, has multiple functions, noma to polymorphisms (15 ). Finally, HIF2 , rather α including serving as the substrate recognition subunit of than HIF1 , appears to be responsible for much of the dis- an ubiquitin ligase complex that targets the alpha subunits ease that develops following pVHL inactivation in the mouse of hypoxia-inducible factor (HIF), a heterodimeric tran- (16 , 17 ). α α scription factor, for polyubiquitination and proteasomal Although HIF1 and HIF2 are similar, they can clearly degradation when oxygen is present ( 3 ). Accordingly, dereg- antagonize one another in certain settings. For example, in α α ulation of HIF target genes, such as VEGF , is a signature ab- some models HIF1 antagonizes, whereas HIF2 potenti- α α normality in pVHL-defective neoplasms, and the degree of ates, c-Myc activity ( 8 , 18 , 19 ). In addition, HIF1 and HIF2 HIF deregulation correlates well with renal carcinoma reciprocally regulate each other’s protein levels in some con- α risk linked to different VHL alleles (4–6 ). Notably, a texts, so that, for example, loss of HIF1 leads to induction of HIF2α and vice versa ( 10 ). In keeping with these observations, overproduction of wild-type HIF1α in pVHL-defective renal carcinoma cells suppresses tumor formation (10 ), Authors' Affiliations: Departments of 1Medical Oncology and 2 Cancer whereas overproduction of HIF2α promotes tumor growth Biology, Dana-Farber Cancer Institute; Departments of 3Medicine and (10 , 11 ). In contrast, HIF1α is believed to promote, rather 4Pathology, Brigham and Women's Hospital, Harvard Medical School, 5 6 than inhibit, many other tumor types of nonrenal origin ( 20 ). Boston; Broad Institute, Cambridge, Massachusetts; Howard Hughes A number of chromosomal abnormalities, in addition to Medical Institute, Chevy Chase, Maryland chromosome 3p loss, have been described in clear cell re- Note: Supplementary data for this article are available at Cancer Discovery Online (http://www.cancerdiscovery.aacrjournals.org). nal carcinoma, including, most commonly, amplification of 5q and loss of chromosome 14q. In numerous studies, loss Corresponding Author: William Kaelin, Dana-Farber Cancer Institute, 450 Brookline Avenue, Mayer Building, Room 457, Boston, MA 02215. of 14q has been associated with poorer outcomes in renal Phone: 617-632-3975; Fax: 617-632-4760; E-mail: william_kaelin@ carcinoma (21–24 ). The knowledge that HIF1 α is located at dfci.harvard.edu chromosome 14q, together with the considerations outlined doi: 10.1158/2159-8290.CD-11-0098 above, led us to explore further whether HIF1 α might be a ©2011 American Association for Cancer Research. clear cell carcinoma tumor suppressor gene.

AUGUST 2011 CANCER DISCOVERY | 223

RESEARCH ARTICLE Shen et al.

RESULTS which harbor p53 and RB1, respectively, did not occur more frequently in kidney cancer than in other cancers (data not Loss of Chromosome 14q Spanning shown). a –/– the HIF1 Locus Is a Common Feature HIF1α expression is lost in many VHL renal carcinoma –/– of Human Kidney Cancer lines, can suppress tumor formation by VHL renal carci- Kidney cancers frequently undergo deletions affecting noma cells when overexpressed (10), and maps to 14q23. In chromosome 14q. To ascertain whether this abnormality oc- contrast, previous studies, including our own, pinpointed curs more often in kidney cancers than in other forms of 14q31-ter as the most likely region to harbor a kidney cancer cancer, we examined a recently published collection of copy tumor suppressor gene (23, 27, 28). Nonetheless, the 14q number data generated with high-density single-nucleotide deletions in kidney cancer are typically very large, with the lo- polymorphism (SNP) arrays from 3,131 cancers representing calization to 14q31-ter based on relatively rare kidney cancers 26 different tumor types (25). The frequency of large dele- with smaller deletions. For example, in our recent analysis of tions affecting most of chromosome 14q was highest in kid- 90 clear cell renal carcinomas, 39 tumors (43%) had sustained ney cancer, followed by melanoma, gastrointestinal stromal 14q deletions (27). Of these, 36 (93%) were large deletions tumor, and esophageal cancer (Fig. 1A). As expected, loss of that also encompassed the HIF1α locus (27). This observa- chromosome 3p, which harbors the VHL tumor suppressor tion suggests the existence of multiple tumor suppressor gene and other tumor suppressor genes such as PBRM1 (26), genes on 14q, including, perhaps, HIF1α. Consistent with as well as amplification of 5q, was also extremely common in this hypothesis, deletions affecting HIF1α are more common kidney cancer relative to other tumor types (Fig. 1B and C). in kidney cancer than in the other 16 tumor types for which These data do not, however, reflect a general proclivity for we have ≥40 samples (Fig. 1D). This strong bias toward kid- copy number alterations in kidney cancer because other copy ney cancer is not apparent, however, if one includes deletions number changes, such as loss of chromosomes 17p and 13q, elsewhere on 14q (Supplementary Fig. S1).

Figure 1. Frequencies of chromosomal abnormalities across different cancers. A, large deletions affecting most of 14q arm. B, large deletions affecting most of 3p arm. C, amplification of any region of 5q. D, deletions affecting HIF1α locus. ALL, acute lymphoblastic leukemia; GIST, gastrointestinal stromal tumor; MPD, myeloproliferative disorder; NSC, non-small cell; SC, small cell, See also Supplementary Fig. S1.

224 | CANCER DISCOVERY AUGUST 2011 www.aacrjournals.org

HIF1α Is a 14q Kidney Cancer Suppressor Gene RESEARCH ARTICLE

Frequent Homozygous Deletions of the HIF1a also been noted by other investigators (30). SKRC-20, A498, Locus in Kidney Cancer Cell Lines and 786-O also produce aberrantly migrating HIF1α proteins We next surveyed a panel of 16 clear cell renal carcinoma (Fig. 2A). These findings raised the possibility that the HIF1α lines, most of which had undergone biallelic VHL inactiva- locus, in addition to undergoing copy number loss, is rear- tion, for HIF1α protein and mRNA production, along with ranged in a subset of renal tumors. HK-2 immortalized, diploid, human renal epithelial cells. In To explore this concept further, we isolated genomic keeping with earlier reports (6, 8, 29), we found that many DNA from the 16 renal carcinoma lines and performed –/– VHL renal carcinoma lines produce no detectable wild-type multiplex ligation-dependent probe amplification (MLPA) –/– HIF1α mRNA or protein, whereas all VHL lines produce analysis to look for copy number changes affecting specific HIF2α (Fig. 2; Supplementary Fig. S2; data not shown). Of HIF1α exons (31). As controls, we also interrogated ran- interest, some lines, such as RCC4, SKRC-20, A498, and domly chosen exons on chromosomes 1, 10, and 17. Two 786-O cells, generate mRNAs with increased electrophoretic cell lines (SLR20 and SLR21) appeared to be diploid across mobility (Fig. 2B). The truncated HIF1α mRNA in 786-O has the HIF1α locus (Fig. 2C; Supplementary Fig. S3). Both of

Figure 2. HIF1α deletions and altered HIF1α gene products in renal carcinoma cells. Immunoblot (A) and Northern blot (B) analysis of the indicated cell lines. The difference in HIF1α electrophoretic mobility between HK-2 immortalized renal epithelial cells and the HIF1α-positive lines, such as A704, Caki-2, RCC4, and UMRC-2, might reflect differential phosphorylation. See also Supplementary Fig. S2. C, MLPA data for the indicated cell lines, normalized to HK-2 immortalized renal epithelial cells (diploid = 1). Black bars, selected control exons on chromosomes 1, 10, and 17; gray bars, HIF1α exons. See also Supplementary Figs. S3 and S4 and Supplementary Table S1.

AUGUST 2011 CANCER DISCOVERY | 225

RESEARCH ARTICLE Shen et al.

+/+ these cell lines, which are phenotypically VHL , were tested (Supplementary Fig. S4B and C). We also did not find ho- by us previously, using high-density SNP arrays, and did not mozygous deletions of SAV1 by MLPA, with the exception exhibit 14q loss (27). Three cell lines (RCC4, Caki-2, and of the previously reported deletion in 786-O cells (ref. 33; A704) appeared to have lost 1 HIF1α allele in its entirety Supplementary Fig. S4A). The discovery of focal, homozy- and to have retained the other (Fig. 2C; Supplementary Fig. gous HIF1α deletions provides genetic evidence that HIF1α S3). Of interest, 7 of 16 cell lines (SKRC-20, A498, 769-P, has a tumor suppressor role in clear cell renal carcinoma. 786-O, UOK101, SLR24, and SLR26) had clearly sustained a homozygous deletions, which in some cases were very fo- HIF1 Suppresses Kidney Cancer Proliferation cal and involved only a subset of contiguous HIF1α exons In Vitro and In Vivo (Fig. 2C; Supplementary Fig. S3). The remaining 4 cell lines To address the suppression issue further, we made ret- (UMRC2, UMRC6, SLR23, SLR25) displayed more complex roviral HIF1α expression vectors in which the amount of MLPA patterns that were intermediate between haploid and HIF1α produced can be regulated by the addition of doxy- –/– diploid across the HIF1α locus, with preferential loss of par- cycline. VHL renal carcinoma cells were then infected with ticular exons (Supplementary Fig. S3). the viruses and maintained in pools. Immunoblot analysis Several other putative tumor suppressor genes reside of these cells grown in the presence or absence of 1 μg/mL of on chromosome 14, including the Hippo pathway genes doxycycline confirmed that HIF1α expression was induced SAV1 (14q22) and FRMD6 (14q22; ref. 32). In contrast to by doxycycline and that the HIF1α levels achieved were HIF1α, we did not detect altered transcripts for these genes similar to those observed after treating HK2 immortalized

A B

Figure 3. Suppression of VHL–/– renal carcinoma proliferation by HIF1α. Immunoblot (A) and proliferation (B) assays of indicated cell lines infected with a doxycycline-inducible retrovirus encoding HIF1α and propagated in 5% serum in the presence or absence of doxycycline. HK-2 cells treated with vehicle, DMOG (1 mM), or MG132 (10 μM) were included as a control in A. See also Supplementary Fig. S5.

226 | CANCER DISCOVERY AUGUST 2011 www.aacrjournals.org

HIF1α Is a 14q Kidney Cancer Suppressor Gene RESEARCH ARTICLE renal epithelial cells with the prolyl hydroxylase inhibitor modest inhibition of cell proliferation observed with HIF2α dimethyloxalylglycine (DMOG) or the proteasomal inhibitor shRNA is consistent with earlier studies using cells grown MG132 (Fig. 3A; Supplementary Fig. S5). Re-expression of under standard serum conditions. –/– HIF1α in the VHL renal carcinoma cell lines A498, 769-P, In addition to affecting proliferation in vitro, downregula- UOK101, and SLR24—all of which produce HIF2α, but not tion of HIF1α promoted the growth of RCC4 renal carci- wild-type HIF1α—impaired proliferation in vitro (Fig. 3B). noma cells that had been implanted in the kidneys of nude This effect was specific because induction of HIF1α did not mice (Fig. 5A–C). Similar results were obtained with Caki-2 –/– diminish the proliferation of RCC4 and UMRC-2 VHL re- cells grown s.c. in nude mice (Supplementary Fig. S6F and G), nal carcinoma cells, which express both HIF1α and HIF2α whereas SLR25 cells did not form tumors in nude mice, irre- (Fig. 3B; Supplementary Fig. S5). Therefore, HIF1α can sup- spective of HIF1α status (data not shown). Downregulation –/– press the proliferation of VHL renal carcinoma cells when of HIF1α in UMRC2 renal carcinoma cells also dramatically expressed at levels approximating those achieved after VHL enhanced tumor growth (Fig. 5D–G), despite having incon- inactivation. sistent effects in vitro (Supplementary Fig. S6D). Therefore, –/– In a reciprocal set of experiments, we downregulated HIF1α suppresses tumor formation by VHL renal carci- –/– HIF1α or HIF2α in 3 VHL renal carcinoma cell lines that noma cells. α α express both HIF1 and HIF2 (Caki-2, RCC4, and SLR25; a a Fig. 4A). In contrast to a recent report, we did not observe HIF1 Variants Resulting from HIF1 Genomic an increase in HIF2α in the cells treated with HIF1α short Deletions Are Defective as Tumor Suppressors hairpin RNA (shRNA; ref. 10; Fig. 4A). The significance of Next we attempted to recover the aberrant mRNAs that we this discrepancy is unclear. In all 3 cases, downregulation of had detected by Northern blot analysis in a subset of renal HIF1α with 2 independent shRNAs enhanced proliferation carcinoma lines. mRNA was harvested from these cell lines, in vitro, compared with cells expressing a scrambled control converted to cDNA, and amplified by PCR. As predicted by shRNA or HIF2α shRNA (Fig. 4B; Supplementary Fig. S6), MLPA analysis, the HIF1α transcript in SKRC-20 cells spe- in keeping with a recent study using RCC4 cells (8). The very cifically lacked exons 3 and 4 (∆3–4), the HIF1α transcript

Figure 4. Downregulation of HIF1α in VHL–/– renal carcinoma cells enhances cell proliferation. Immunoblot (A) and proliferation (B) assays of the indicated cell lines after infection with lentiviruses encoding HIF1α shRNA, HIF2α shRNA, or scrambled control shRNA and grown in the presence of 5% serum. See also Supplementary Fig. S6.

AUGUST 2011 CANCER DISCOVERY | 227

RESEARCH ARTICLE Shen et al.

AB C Figure 5. Downregulation of HIF1α promotes tumor cell growth in vivo. A, representative bioluminescent images of nude mice orthotopically injected with RCC4 renal carcinoma cells that stably express firefly luciferase and an shRNA directed against HIF1α (right kidney) or a scrambled shRNA control (left kidney). Shown are images 1 week and 9 weeks after tumor cell implantation. B, bioluminescence ratio as a function of time for mice, as in A. For each mouse, the ratio was normalized to the week 1 ratio for that mouse. C, mean fold change in normalized bioluminescence ratio for DEFG mice analyzed in B at week 9. Error bars = 1 SEM. D and F, representative images of nude mice 4 weeks after s.c. injection of UMRC-2 renal carcinoma cells that stably express an shRNA directed against HIF1α (right side) or a scrambled shRNA control (left side). E and G, mean tumor weights at necropsy of mice as in D and F. Error bars = 1 SEM. See also Supplementary Fig. S6.

in A498 cells lacked exons 2 to 6 (∆2–6), and the HIF1α tran- variant migrated in accordance with its predicted molecular script in SLR26 cells lacked exons 5 to 10 (∆5–10; Fig. 6A). weight and, with the exception of the ∆2–6 variant, was pro- These transcripts presumably reflect the homozygous dele- duced at levels that were similar to those of wild-type HIF1α tion detected by MLPA together with alternative splicing. In (Fig. 6B). addition, PCR primers based on the 5′ and 3′ HIF1α un- Notably, wild-type HIF1α suppressed the proliferation of translated regions detected a transcript lacking exons 2 to 12 769-P cells to a greater extent than did any of the 3 variants in RCC4 and A498 cells, perhaps responsible for the faster tested (Fig. 6C). In the next set of experiments, the 769-P cells migrating Northern blot band detected in these cells (Fig. 2B; that had been engineered to inducibly express wild-type or data not shown). This variant (∆2–12), in contrast to the mutant HIF1α were pooled and propagated in vitro in the other mRNA variants, was also detected in some normal kid- presence or absence of doxycycline (Fig. 6D). Growth in the ney mRNA samples (Supplementary Table S2). presence of doxycycline led to progressive loss of cells ex- In 786-O cells, we recovered a transcript lacking exons pressing wild-type HIF1α relative to those expressing mu- 13 to 15, using 3′ rapid amplification of cDNA ends, in tants, suggesting that cells expressing wild-type HIF1α are at keeping with the MLPA data (Supplementary Fig. S3; data a growth disadvantage in such competition assays (Fig. 6E). not shown) and with Western blot data indicating that Similar results were obtained when the cells were injected these cells produce a HIF1α variant that reacts with a poly- into the kidneys of NOD/SCID mice and propagated in vivo clonal antibody, but not a C-terminal monoclonal antibody (Fig. 6F), indicating that the HIF1α variants tested in this (Fig. 2A). This transcript is predicted to encode a fusion pro- study are enfeebled as tumor suppressors relative to wild-type tein, as it contains several in-frame exons from a neighbor- HIF1α. ing gene (data not shown). For unclear reasons, however, this transcript greatly exceeds the apparent molecular weight of HIF1a Activity Is Diminished in Human Kidney the mRNA detected in these cells by Northern blot analysis Cancers Harboring 14q Deletions (Fig. 2B). To determine if our cell line data were relevant to human The ∆3–4, ∆2–6, and ∆5–10 HIF1α variants, as well as kidney cancers, we next asked whether HIF1α activity is di- –/– wild-type HIF1α, were introduced into 769-P VHL renal minished in human kidney cancers that have sustained 14q carcinoma cells, using the doxycycline-inducible retrovi- deletions encompassing the HIF1α locus. Toward this end, ral expression vector described above. Notably, each vari- we performed gene expression profiling on the cell lines de- ant, except for the ∆5–10 one, preserves the proper reading scribed above that either expressed both HIF1α and HIF2α frame. An N-terminal hemagglutinin (HA) epitope tag was (either naturally or by virtue of induced expression of HIF1α) introduced to facilitate the detection of wild-type and mu- or expressed HIF2α alone (either naturally or by virtue of a tant HIF1α proteins after induction with doxycycline. Each HIF1α shRNA), followed by supervised clustering to arrive at

228 | CANCER DISCOVERY AUGUST 2011 www.aacrjournals.org

HIF1α Is a 14q Kidney Cancer Suppressor Gene RESEARCH ARTICLE

A Figure 6. Differential effects of cell line–derived HIF1α variants on proliferation and tumor cell fitness. A, HIF1α variants identified in renal carcinoma lines. B and C, immunoblot (B) and proliferation (C) assays of 769-P cells infected with doxycycline- inducible retroviruses encoding the indicated HIF1α variants and propagated in the presence of doxycycline. 769-P cells infected with an inducible virus encoding wild-type HIF1α but grown in the absence of B C doxycycline were included as a control in B. D, in vitro and in vivo competition assays. E and F, PCR-based analysis indicating relative abundance of cells, as in B and C, after growth in vitro for 9 days (E) or orthotopically in vivo for 6 weeks (F) in the presence or absence of doxycycline. Neg., PCR reaction with no input DNA. WT, wild-type. See also Supplementary Table S2.

D E

a “HIF1α transcriptional signature,” which included known Somatic HIF1a Mutations in Human Clear Cell HIF1α-specific targets, such as BNIP3, PGK1, HK1, and TPI1 Carcinomas Are Loss of Function (Fig. 7A; refs. 10, 34). Gene set enrichment analysis (GSEA) We have not yet been able to definitively identify focal, ho- using this HIF1α signature and gene expression data from mozygous deletions in renal tumor DNA by MLPA (data not –/– 52 VHL clear cell renal carcinomas, including 32 without a shown). This inability might be related to technical issues HIF1α deletion and 20 with a HIF1α deletion, confirmed that because MLPA is very sensitive to contamination by host HIF1α activity is diminished in tumors that have sustained DNA. In 3 of 23 primary clear cell renal carcinomas, how- a 14q deletion spanning the HIF1α locus (P < 0.01; Fig. 7B). ever, we did detect presumptively pathogenic HIF1α mRNA

AUGUST 2011 CANCER DISCOVERY | 229

RESEARCH ARTICLE Shen et al.

A B Figure 7. HIF1α activity is impaired by copy number changes and mutations in pVHL- defective kidney cancers. A, heat map depicting genes that are differentially expressed between VHL–/– renal carcinoma cell lines that do (H1H2) or do not (H2) express high levels of HIF1α. Genes (right) are ordered from top to bottom according to P values showing the degree to which they are significantly altered in H1H2 cells compared with H2 cells. A t-test and

hierarchical clustering were SLR24+HIF1a SLR25+Scr-shRNA 769P+HIF1a Caki-2+Scr-shRNA A498+HIF1a SLR24+Vector SLR25+HIF1a-shRNA A498+Vector 769P+Vector Caki-2+HIF1a-shRNA performed using software BNIP3 GENE-E at the Broad Institute HK2 (57). B, GSEA plot showing ERO1L location of the enrichment score ACAD11 of 71 HIF1α upregulated genes PDK1 in a set of ccRCC tumors with P4HA1 C VHL INSIG2 biallelic inactivation. The RLF HIF1α positively regulated genes RCTPI1 are significantly enriched (P ≤ TPI1 0.01) in tumors that have not DARS sustained 14q deletions BNIP3L HIF1α HIF1α ANKZF1 encompassing . C, RNF24 mutations (mu) identified in renal SPAG4 carcinoma patients. Immunoblot PRKAR1A (D) and proliferation (E) assays ANKRD37 of 769-P cells infected with AK3L1 doxycycline-inducible CCNG2 LOC401152 D retroviruses encoding wild- PGK1 type HIF1α or the indicated GOLGA9P HIF1α variants and propagated PFKFB4 in the presence of doxycycline. FBXO42 Cells grown in the absence of PPP1R3C GBE1 doxycycline were included as CRYBB2P1 controls in D. TMCC1 DENND1A WSB1 ACADVL JMJD2B PDK3 HAGH FER1L4 FUT11 FER1L4 NCKIPSD GOLGA8E RORA DDIT4 SLC25A36 ARRDC4 STBD1 E LOC728780 UPRT ENO2 VKORC1 JMJD1A L1CAM YEATS2 HERC3 RAPGEF4 TPD52 TXNIP IMPAD1 FAM162A PPFIA4 HK1 IFNA21 APAF1 ALDOC CLK3 ARRDC2 KCTD11 PPP1R16A PLAGL1 RIOK3 GPI relative PLXNA3 FAM57A LPIN3 -3 0 3

230 | CANCER DISCOVERY AUGUST 2011 www.aacrjournals.org

HIF1α Is a 14q Kidney Cancer Suppressor Gene RESEARCH ARTICLE splice variants (absent in normal kidney RNA samples these genetic and functional data credential HIF1α as a clear and absent in publicly available databases), analogous to cell renal carcinoma suppressor gene. Hence loss of pVHL those present in renal carcinoma cell lines (Supplementary simultaneously leads to activation of an oncoprotein (HIF2α) Table S2). and a tumor suppressor protein (HIF1α). This finding would Somatic HIF1α mutations have been described at low fre- explain the frequent loss of chromosome 14q in kidney can- quency in human kidney cancer (refs. 29, 35; H. Greulich, cer and is consistent with the observation that loss of HIF1α M. Meyerson, and W.G. Kaelin, unpublished observations). protein in preneoplastic lesions in the kidneys of patients We therefore made doxycycline-inducible retroviral vectors with von Hippel–Lindau disease heralds further malignant corresponding to 3 of these mutations (c.2120delA, c.2180 transformation (9). C→A, and V116E), as well to a presumed benign SNP (A475S; In a recent study the percentage of clear cell renal car- Fig. 7C and D). All 3 mutations suspected of being patho- cinomas with low HIF1α expression approximated the genic compromised the ability of HIF1α to suppress renal frequency of 14q loss for this tumor type (8), and 14q carcinoma growth, whereas the presumptive SNP did not loss was enriched among the HIF1α-negative tumors (K. (Fig. 7E). A 4th mutant recently reported (35), exon 4 –5 del Nathanson, personal communication). Moreover, we con- (tacagTTTGAACTAACTGGA), was not tested because it was firmed that HIF1α transcriptional activity is indeed de- –/– predicted to induce a frameshift after exon 3. Collectively, creased in VHL kidney tumors that have sustained 14q these results support that HIF1α activity is diminished in deletions encompassing HIF1α, compared with those that –/– a subset of VHL kidney cancers because of a reduction in have not. Although homozygous HIF1α deletions appear gene dosage and, in some cases, as a result of intragenic, loss- to be common in clear cell renal carcinoma cell lines, we of-function mutations. have not documented a similarly high frequency in primary renal tumors. This discrepancy might be due to technical factors, but it does raise the possibility that HIF1α haploin- DISCUSSION sufficiency is sufficient to promote primary tumor growth in vivo, and that reduction to nullizgyosity is selected for Loss of the region of chromosome 3p spanning the VHL during tumor progression in vivo or the propagation of clear gene is the most frequent genomic change in clear cell renal cell carcinoma lines in vitro. carcinoma, the most common form of kidney cancer. The HIF1α is usually thought to promote tumor growth, but next 2 most common genomic abnormalities in kidney can- precedent exists for its function as a tumor suppressor. For cer are chromosome 5q amplification and chromosome 14q example, loss of HIF1α enhances tumor formation by em- loss. Moreover, in this article we show that kidney cancer has bryonic stem cell–derived teratoma cells and by murine as- the highest rate of chromosome 3p loss, chromosome 5q am- trocytes (37–39). In the context of renal carcinoma, HIF1α plification, and 14q loss among a wide variety of tumor types. might act as a tumor suppressor specifically by antagoniz- VHL loss leads to increased abundance of HIF1α and ing HIF2α. For example, transactivation by 1 of the 2 HIFα HIF2α, and deregulation of HIF-dependent transcription is transactivation domains [the C-terminal transactivation do- a signature abnormality in kidney cancer. Mounting evidence main (CTAD)] is inhibited by the asparaginyl hydroxylase suggests that HIF2α, rather than HIF1α, promotes pVHL-de- factor inhibiting HIF1 (FIH1). HIF2α is less sensitive than fective renal carcinogenesis. Indeed, many pVHL-defective re- HIF1α to FIH1-mediated inhibition (40, 41). Competitive nal carcinomas produce low, or undetectable, levels of HIF1α displacement of HIF2α by HIF1α from HIF-responsive –/– (6, 8, 36), and restoring HIF1α expression in a VHL renal promoters that depend upon the CTAD for full activation carcinoma line was previously shown to suppress tumorigen- would therefore potentially decrease promoter activity. esis (10). We confirmed the latter finding and revealed that Moreover, in some contexts HIF1α can suppress HIF2α lev- HIF1α, at levels comparable to those seen when pVHL func- els via mechanisms that are as yet unclear (10). In short, tion is impaired, suppresses renal carcinoma proliferation loss of HIF1α might, paradoxically, increase the activity of and tumor growth. certain HIF-responsive promoters in pVHL-defective tumor Prompted by this knowledge, we asked whether HIF1α, cells. Experimental evidence exists to support this conten- which resides on chromosome 14q23.2, might be one of the tion (ref. 10; C. Shen and W.G. Kaelin, unpublished observa- genes targeted by 14q deletions in kidney cancer. Indeed, we tions and data not shown). found that the vast majority of 14q deletions detected in re- In addition to quantitative differences on shared HIF- nal carcinoma encompass HIF1α. Moreover, we documented responsive promoters, a number of qualitative differences that HIF1α, but not neighboring genes on chromosome 14q, between HIF1α and HIF2α might relate to HIF1α scoring as is often subject to focal deletions in kidney cancer cell lines. a tumor suppressor protein. For example, some genes regu- Some of these deletions led to the production of aberrant lated by HIF1α are not regulated by HIF2α and vice versa. mRNAs and proteins that compromised the ability of HIF1α Conceivably, some genes that are preferentially activated by –/– to suppress VHL renal carcinoma proliferation and tumori- HIF1α decrease renal carcinoma cell fitness. In this regard, genesis. We also discovered that downregulation of wild-type 3 genes found in our HIF1α signature—TXNIP, KCTD11, and HIF1α promotes growth of renal carcinoma in vivo. Finally, PLAGL1—have been implicated as tumor suppressors in other we showed that somatic, presumably pathogenic, HIF1α mu- contexts (42–45). Further, HIF1α and HIF2α differ in terms tations in human clear cell carcinomas enfeeble HIF1α as of their ability to engage collateral signaling pathways, such a tumor suppressor in cell proliferation assays. Collectively, as those involving c-Myc and Notch. For example, HIF1α, via

AUGUST 2011 CANCER DISCOVERY | 231

RESEARCH ARTICLE Shen et al. a variety of mechanisms, can inhibit c-Myc activity in certain introduced a MluI site. The products were digested with BamHI and settings, whereas HIF2α does not (8, 18, 19). MluI and cloned into pRetroX-Tight-Pur vector (Clontech) cut with The 14q deletions in kidney cancer are typically very large these 2 enzymes. α and usually span HIF1α, as noted above. Nonetheless, rare The tumor-derived HIF1 mutations in Fig. 7 were generated by the site-directed mutagenesis of the pRetroX-Tight-HIF1α-WT tumors with small deletions had pinpointed 14q31-ter as the by using QuikChange II XL Site-Directed Mutagenesis Kit (Agilent likely location for a kidney cancer tumor suppressor gene (23, Technologies) and confirmed by DNA sequencing. The forward 27, 28). The simplest reconciliation of these findings would primers for mutagenesis of HIF1α mu1-mu4 are as follows: be the existence of multiple kidney cancer tumor suppres- 5′-CAGTTCCTGAGGAAGAACTAATCCAAAGATACTAGCTTT sor genes on 14q, in addition to HIF1α, with perhaps some GCAG-3′; acting through haploinsufficiency. Alternatively, these small 5′-GGAACATGATGGTTAACTTTTTCAAGCAGTAGG-3′; deletions might have been passenger, rather than driver, 5′-CATGATTTACATTTCTGATAATGAGAACAAATACATGGGA mutations. TTAAC-3′; and 5′-CTGCACTCAATC AAGAAGTTTCA TTAAAATT ′ It will be of interest to determine whether pVHL-defective AGAACCAAATCC-3 , respectively. α clear cell renal carcinomas that retain wild-type HIF1α ex- Lentiviral shRNA vector pLKO.1, lentiviral HIF1 shRNA vector (TRCN0000003810, target sequence: 5′-GTGATGAAAGAATTACCG pression use alternative mechanisms to circumvent the tu- ′ ′ α AAT-3 ; TRCN0000003811, target sequence: 5 -CGGCGAAGTAAAG mor suppressor activity of HIF1 . We note, for example, that ′ ′ PLAGL1 AATCTGAA-3 ; TRCN0000003809, target sequence: 5 -CCAGTTAT maps to a region of 6q frequently deleted in VHL- GATTGTGAAGTTA-3′), and lentiviral HIF2α (EPAS1) shRNA vec- associated neoplasms and sporadic clear cell renal carcino- tor (TRCN0000003806, target sequence: 5′-CAGTACCCAGACGGA mas (27, 46, 47). Moreover, it will be important to determine TTTCAA-3′) were obtained from the Broad Institute TRC shRNA whether retention of HIF1α expression alters the response library. of pVHL-defective tumors to targeted agents that directly or indirectly target HIF. In this regard, it is possible that the Immunoblot Analysis salutary effects of rapamycin-like mTOR inhibitors (rapa- Cells were lysed with 1 × EBC buffer [50 mM Tris (pH 8.0), logs) in kidney cancer are partially mitigated by their ability 120 mM NaCl, 0.5% Nonidet P-40] supplemented with a protease to downregulate HIF1α (48), especially in light of a recent inhibitor cocktail (Complete; Roche Applied Science, Indianapolis, report predicting that HIF2α would be relatively resistant to Indiana), resolved by SDS-PAGE (30 μg/lane), and transferred to nitrocellulose membranes (Bio-Rad). Primary antibodies were rab- such agents (49). bit polyclonal anti-HIF1α (NB100-479; Novus, Littleton, Colorado), mouse monoclonal anti-HIF1α (BD Transduction Laboratories), mouse monoclonal anti-HA (HA-11; Covance Research Product), and METHODS mouse monoclonal antitubulin (B-512; Sigma-Aldrich). Bound an- Cell Lines tibody was detected with horseradish peroxidase (HRP)–conjugated HK-2, 786-O, A704, 769-P, Caki-2, RCC4, and A498 cells were pur- secondary antibodies (Pierce, Rockford, IL) and Immobilon Western chased from the American Type Culture Collection. UMRC-2 (50), chemiluminescent HRP substrate (Millipore). UMRC-6 (50), and UOK101 (51) cells were provided by Drs. Bert Northern Blot Analysis Zbar and Martson Linehan (National Cancer Institute, Bethesda, Maryland). SLR20, SLR21, SLR23, SLR24, SLR25, and SLR26 cells Total RNA was isolated using TRIzol reagent (Invitrogen), re- were supplied by Drs. Mark A. Rubin and Kirsten Mertz (Weill solved by agarose-formaldehyde gel electrophoresis (10 μg RNA per Cornell Medical College, New York, New York; ref. 52). SKRC-20 cells lane), transferred to nitrocellulose membranes, and probed with a HIF1α α 32 (53) were provided by Drs. Gerd Ritter and Beatrice Yin (Memorial AgeI-PstI fragment labeled with [ - P]dCTP using a Prime-It Sloan-Kettering Cancer Center, New York, New York). DNA from II Random Primer Labeling Kit (Stratagene). Hybridizations were these cell lines was subjected to SNP analysis by us previously (27), performed in QuikHyb (Stratagene) according to the manufacturer’s and the cell lines were monitored regularly for mycoplasma con- instructions. Signals were detected by X-ray film. tamination. No other validation was performed. Whenever possible, freshly thawed vials of cells were used at early passage. HK2 im- Multiplex Ligation-Dependent Probe Analysis mortalized human renal epithelial cells were maintained in keratino- Multiplex ligation-dependent probe amplification was performed cyte serum-free medium supplemented with 0.05 mg/mL of bovine as described in ref. 31, using the probe sets listed in Supplementary pituitary extract and 5 ng/mL of human recombinant epidermal Table S1 and reagents provided by MRC-Holland (54). growth factor. Renal carcinoma cells 786-O, A498, RCC4, UMRC- The capillary electrophoresis and peak height determination 2, UMRC-6, and UOK101 were maintained in Dulbecco’s modified of amplification products were performed by Mei Lin at the DNA Eagle’s medium containing 10% FBS; 769-P, A704, SKRC-20, SLR20, Sequencing Facility, Brigham and Women’s Hospital, Boston, SLR21, SLR23, SLR24, SLR25, and SLR26, in RPMI-1640 containing Massachusetts. Briefly, amplification products were 10× diluted 10% FBS; and Caki-2, in McCoy’s 5A containing 10% FBS. Following in Hi-Di Formamide (ABI, Foster City, California) containing retroviral or lentiviral infection, cells were maintained in the presence 1/16 volume of ROX500 size standard (ABI) and then were separated of puromycin (2 μg/mL) or G418 (500 μg/mL), depending on the by size on an ABI 3100 Genetic Analyzer (ABI). Electropherograms vector. All cells were kept at 37°C in 10% CO2. were analyzed by GeneMapper v3.5 (ABI), and peak height data were exported to an Excel table. Plasmids The wild-type and variant HIF1α cDNAs (HIF1α∆3–4, Cell Proliferation Assays HIF1α∆2–6, and HIF1α∆5–10) were PCR amplified from HK-2, Renal carcinoma cells that had been infected to inducibly express cells per well; 6 wells 500ف) SKRC-20, A498, and SLR26 cells, respectively, with a 5′ primer that HIF1α were plated in 96-well plates introduced a BamH1 site and an HA epitope and a 3′ primer that per condition and time point) in RPMI-1640 media supplemented

232 | CANCER DISCOVERY AUGUST 2011 www.aacrjournals.org

HIF1α Is a 14q Kidney Cancer Suppressor Gene RESEARCH ARTICLE with 5% FBS in the presence or absence of the indicated amount of average signal ratio ≥ 1.3 between 2 subgroups were selected as posi- doxycycline, with a media change every 2 days. The number of viable tively regulated by HIF1α. This gene set was evaluated by GSEA (58) cells per well at each time point was measured using an XTT assay across a panel of 52 ccRCC tumors with biallelic inactivation of VHL (Cell Proliferation Kit II; Roche, cat. no. 11465015001) according to (with 32 non-HIF1α deletion tumors and 20 HIF1α deletion tumors) the manufacturer’s instructions. Spectrophotometric absorbance at for which expression data were previously obtained (GSE14994; ref. 450 nm was measured 6 hours after the addition of the XTT labeling 27). We used the GSEA parameters: weighted scoring, signal-to-noise reagent/electron coupling reagent, using a microtiter plate reader metric, and gene set permutations. (Perkin Elmer Life and Analytical Science). Renal carcinoma cells stably infected with lentiviral shRNAs were Disclosure of Potential Conflicts of Interest plated, in triplicate, in 6 well plates (104 cells/well) in RPMI (RCC4, W.G. Kaelin owns equity in, and consults for, Fibrogen, Inc., which SLR25 and UMRC-2) or McCoy’s 5A (Caki-2) supplemented with is developing drugs that modulate HIF activity. No other potential 5% FBS. Cells were trypsinized and collected at the indicated time conflicts of interest were disclosed. points. The number of viable cells, as determined by Trypan blue staining, was determined with a hemocytometer. Acknowledgments Renal carcinoma 769-P cells that had been infected to inducible We thank Drs. Mark A. Rubin and Kirsten Mertz at Weill Cornell express tumor-derived HIF1α mutants were plated, in triplicate, in 6 Medical College for the SLR20, SLR21, SLR23, SLR24, SLR25, and well plates (5,000 cells/well) in RPMI with 5% FBS. Cells were trypsin- SLR26 cell lines; Drs. Gerd Ritter and Beatrice Yin at Memorial ized and collected at the indicated time points. The number of viable Sloan-Kettering Cancer Center for the SKRC-20 cell line; Dr. cells was determined by using an automated cell counter (Countess, Marston Linehan at NCI for ccRCC tumor samples; and Dr. David Invitrogen). Kwiatkowski for advice on the MLPA assays. Xenograft Assays and Bioluminescence Grant Support Orthotopic and s.c. growth of tumor cells was as described in ref. 55. For RCC4 cells, 1 × 106 viable RCC4 HIF1α shRNA cells (right This work was supported by NIH (W.G. Kaelin), Howard Hughes kidney) and 1 × 106 viable RCC4 scrambled shRNA cells (left kidney) Medical Institute (W.G. Kaelin), Doris Duke Charitable Foundation were injected orthotopically into Swiss nude mice (Taconic, Hudson, (W.G. Kaelin), Dana-Farber/Harvard Cancer Center Kidney Cancer New York). Bioluminescent detection and quantification of tumor Career Development Award (C. Shen; funded by Genentech). W.G. burden were performed as described in ref. 55. For each mouse, total Kaelin is a Doris Duke Distinguished Clinical Investigator and photons from the right kidney were divided by total photons from Howard Hughes Medical Institute Investigator. the left kidney and normalized to the ratio for that mouse the first week after tumor cell injection. For UMRC-2 cells, 2 × 107 viable Received May 2, 2011; revised May 25, 2011; accepted June 2, 2011; UMRC-2 HIF1α shRNA cells (right side) and 2 × 107 viable UMRC-2 published OnlineFirst June 7, 2011. scrambled shRNA cells (left side) were injected s.c. into Swiss nude mice. The mice were sacrificed 4 weeks after injection, and tumors REFERENCES were excised and weighed. For Caki-2 cells, 1 × 107 viable Caki-2 HIF1α shRNA cells (right side) and 1 × 107 viable Caki-2 scrambled 1. Jemal A, Siegel R, Xu J, Ward E. Cancer statistics, 2010. CA Cancer shRNA cells (left side) were injected s.c. into Swiss nude mice. The J Clin 2010;60:277–300. mice were sacrificed 16 to 20 weeks after injection, and tumors were 2. Kim WY, Kaelin WG. Role of VHL gene mutation in human cancer. J Clin Oncol 2004;22:4991–5004. excised and weighed. 3. Kaelin WG. Von Hippel-Lindau disease. Annu Rev Pathol 2007; The 5-week-old female NOD scid gamma mice (The Jackson 2:145–73. Laboratory, Bar Harbor, Maine) were orthotopically injected with 4. Iliopoulos O, Levy AP, Jiang C, Kaelin WG, Jr., Goldberg MA. × 6 1 10 viable 769-P cells (1:1:1:1 mixture of 769-P cells stably infected Negative regulation of hypoxia-inducible genes by the von Hippel– α α with retroviruses encoding wild-type HIF1 or 3 different HIF1 Lindau protein. Proc Natl Acad Sci U S A 1996;93:10595–9. variants). The mice were randomized to chow that did or did not 5. Li L, Zhang L, Zhang X, Yan Q, Minamishima YA, Olumi AF, contain 6 g doxycycline/kg chow (Bioserv) 1 week after tumor cell et al. Hypoxia-inducible factor linked to differential kidney cancer implantation and were sacrificed 6 to 8 weeks later. Then 300 ng of risk seen with type 2A and type 2B VHL mutations. Mol Cell Biol genomic DNA, extracted from the injected kidney with the DNeasy 2007;27:5381–92. Blood & Tissue Kit (Qiagen), was PCR amplified using a forward 6. Maxwell PH, Wiesener MS, Chang GW, Clifford SC, Vaux EC, primer (5′-ATGTACCCATACGATGTTCCAGATTACGC-3′) based Cockman ME, et al. The tumour suppressor protein VHL targets on the sequence encoding the shared HA epitope tag and a reverse hypoxia-inducible factors for oxygen-dependent proteolysis. Nature primer (5′-TGTGCTTTGAGGACTTGCGCTTTC-3′) based on the 1999;399:271–5. HIF1α C-terminus. 7. Rini BI. New strategies in kidney cancer: therapeutic advances through understanding the molecular basis of response and resis- Microarray and GSEA tance. Clin Cancer Res 2010;16:1348–54. Gene expression profiling was performed at the Microarray Core at 8. Gordan JD, Lal P, Dondeti VR, Letrero R, Parekh KN, Oquendo CE, Dana-Farber Cancer Institute, using Gene 1.0ST Arrays (Affymetrix), et al. HIF-alpha effects on c-Myc distinguish two subtypes of spo- and the data were deposited in the Gene Expression Omnibus data- radic VHL-deficient clear cell renal carcinoma. Cancer Cell 2008; base (GSE27415). In brief, total RNA was extracted using the RNeasy 14:435–46. 9. Mandriota SJ, Turner KJ, Davies DR, Murray PG, Morgan NV, Sowter Mini Kit with on-column DNase digestion (Qiagen). Biotin-labeled HM, et al. HIF activation identifies early lesions in VHL kidneys: cRNA was prepared from 1 μg of total RNA, fragmented, and hybrid- evidence for site-specific tumor suppressor function in the nephron. ized to a Human Gene 1.0 ST Array. The arrays were scanned and Cancer Cell 2002;1:459–68. raw expression data normalized using DNA-Chip Analyzer (dChip) 10. Raval RR, Lau KW, Tran MG, Sowter HM, Mandriota SJ, Li JL, et al. custom software (56). Genes were assessed for significantly altered Contrasting properties of hypoxia-inducible factor 1 (HIF-1) and t expression between H1H2 and H2 subgroups by -test analysis using HIF-2 in von Hippel–Lindau–associated renal cell carcinoma. Mol P ≤ the software GENE-E (57). The 71 genes with values 0.05 and an Cell Biol 2005;25:5675–86.

AUGUST 2011 CANCER DISCOVERY | 233

RESEARCH ARTICLE Shen et al.

11. Kondo K, Klco J, Nakamura E, Lechpammer M, Kaelin WG, Jr. TSC2 using MLPA, and genotype-phenotype correlations. Hum Inhibition of HIF is necessary for tumor suppression by the von Genet 2007;121:389–400. Hippel–Lindau protein. Cancer Cell 2002;1:237–46. 32. Harvey K, Tapon N. The Salvador-Warts-Hippo pathway–an emerg- 12. Maranchie JK, Vasselli JR, Riss J, Bonifacino JS, Linehan WM, ing tumour-suppressor network. Nat Rev Cancer 2007;7:182–91. Klausner RD. The contribution of VHL substrate binding and HIF1- 33. Tapon N, Harvey KF, Bell DW, Wahrer DC, Schiripo TA, Haber alpha to the phenotype of VHL loss in renal cell carcinoma. Cancer DA, et al. salvador Promotes both cell cycle exit and apopto- Cell 2002;1:247–55. sis in Drosophila and is mutated in human cancer cell lines. Cell 13. Kondo K, Kim WY, Lechpammer M, Kaelin WG, Jr.Inhibition of 2002;110:467–78. HIF2alpha is sufficient to suppress pVHL-defective tumor growth. 34. Hu CJ, Wang LY, Chodosh LA, Keith B, Simon MC. Differential roles PLoS Biol 2003;1:E83. of hypoxia-inducible factor 1alpha (HIF-1alpha) and HIF-2alpha in 14. Zimmer M, Doucette D, Siddiqui N, Iliopoulos O. Inhibition of hy- hypoxic gene regulation. Mol Cell Biol 2003;23:9361–74. –/– poxia-inducible factor is sufficient for growth suppression of VHL 35. Dalgliesh GL, Furge K, Greenman C, Chen L, Bignell G, Butler A, tumors. Mol Cancer Res 2004;2:89–95. et al. Systematic sequencing of renal carcinoma reveals inactivation 15. Purdue MP, Johansson M, Zelenika D, Toro JR, Scelo G, Moore of histone modifying genes. Nature 2010;463:360–3. LE, et al. Genome-wide association study of renal cell carcinoma 36. Biswas S, Troy H, Leek R, Chung YL, Li JL, Raval RR, et al. Effects of identifies two susceptibility loci on 2p21 and 11q13.3. Nat Genet HIF-1alpha and HIF2alpha on growth and metabolism of clear-cell 2011;43:60–5. renal cell carcinoma 786-0 xenografts. J Oncol 2010;2010:757908. 16. Rankin EB, Rha J, Unger TL, Wu CH, Shutt HP, Johnson RS, et al. 37. Carmeliet P, Dor Y, Herbert JM, Fukumura D, Brusselmans K, Hypoxia-inducible factor-2 regulates vascular tumorigenesis in mice. Dewerchin M, et al. Role of HIF-1alpha in hypoxia-mediated apoptosis, Oncogene 2008;27:5354–8. cell proliferation and tumour angiogenesis. Nature 1998;394:485–90. 17. Kim WY, Safran M, Buckley MR, Ebert BL, Glickman J, Bosenberg 38. Blouw B, Song H, Tihan T, Bosze J, Ferrara N, Gerber HP, et al. The M, et al. Failure to prolyl hydroxylate hypoxia-inducible factor alpha hypoxic response of tumors is dependent on their microenviron- phenocopies VHL inactivation in vivo. EMBO J 2006;25:4650–62. ment. Cancer Cell 2003;4:133–46. 18. Gordan JD, Bertout JA, Hu CJ, Diehl JA, Simon MC. HIF-2alpha pro- 39. Covello KL, Simon MC, Keith B. Targeted replacement of hypoxia- motes hypoxic cell proliferation by enhancing c-myc transcriptional inducible factor-1alpha by a hypoxia-inducible factor-2alpha knock- activity. Cancer Cell 2007;11:335–47. in allele promotes tumor growth. Cancer Res 2005;65:2277–86. 19. Zhang H, Gao P, Fukuda R, Kumar G, Krishnamachary B, Zeller KI, 40. Bracken CP, Fedele AO, Linke S, Balrak W, Lisy K, Whitelaw ML, et al. et al. HIF-1 inhibits mitochondrial biogenesis and cellular respira- Cell-specific regulation of hypoxia-inducible factor (HIF)-1alpha and tion in VHL-deficient renal cell carcinoma by repression of C-MYC HIF-2alpha stabilization and transactivation in a graded oxygen en- activity. Cancer Cell 2007;11:407–20. vironment. J Biol Chem 2006;281:22575–85. 20. Semenza GL. Targeting HIF-1 for cancer therapy. Nat Rev Cancer 41. Yan Q, Bartz S, Mao M, Li L, Kaelin WG, Jr.The hypoxia-inducible 2003;3:721–32. factor 2alpha N-terminal and C-terminal transactivation domains 21. Mitsumori K, Kittleson JM, Itoh N, Delahunt B, Heathcott RW, cooperate to promote renal tumorigenesis in vivo. Mol Cell Biol Stewart JH, et al. Chromosome 14q LOH in localized clear cell renal 2007;27:2092–102. cell carcinoma. J Pathol 2002;198:110–4. 42. Sheth SS, Bodnar JS, Ghazalpour A, Thipphavong CK, Tsutsumi S, 22. Kaku H, Ito S, Ebara S, Ouchida M, Nasu Y, Tsushima T, et al. Tward AD, et al. Hepatocellular carcinoma in Txnip-deficient mice. Positive correlation between allelic loss at chromosome 14q24-31 Oncogene 2006;25:3528–36. and poor prognosis of patients with renal cell carcinoma. Urology 43. Ohta S, Lai EW, Pang AL, Brouwers FM, Chan WY, Eisenhofer G, 2004;64:176–81. et al. Downregulation of metastasis suppressor genes in malignant 23. Alimov A, Sundelin B, Wang N, Larsson C, Bergerheim U. Loss of pheochromocytoma. Int J Cancer 2005;114:139–43. 14q31-q32.2 in renal cell carcinoma is associated with high malig- 44. Abdollahi A. LOT1 (ZAC1/PLAGL1) and its family members: mecha- nancy grade and poor survival. Int J Oncol 2004;25:179–85. nisms and functions. J Cell Physiol 2007;210:16–25. 24. Klatte T, Rao PN, de Martino M, LaRochelle J, Shuch B, Zomorodian 45. Mancarelli MM, Zazzeroni F, Ciccocioppo L, Capece D, Po A, N, et al. Cytogenetic profile predicts prognosis of patients with clear Murgo S, et al. The tumor suppressor gene KCTD11REN is regulated cell renal cell carcinoma. J Clin Oncol 2009;27:746–53. by Sp1 and methylation and its expression is reduced in tumors. Mol 25. Beroukhim R, Mermel CH, Porter D, Wei G, Raychaudhuri S, Cancer 2010;9:172. Donovan J, et al. The landscape of somatic copy-number alteration 46. Lemeta S, Salmenkivi K, Pylkkanen L, Sainio M, Saarikoski ST, across human cancers. Nature 2010;463:899–905. Arola J, et al. Frequent loss of heterozygosity at 6q in pheochromocy- 26. Varela I, Tarpey P, Raine K, Huang D, Ong CK, Stephens P, et al. toma. Hum Pathol 2006;37:749–54. Exome sequencing identifies frequent mutation of the SWI/SNF 47. Lemeta S, Pylkkanen L, Sainio M, Niemela M, Saarikoski S, complex gene PBRM1 in renal carcinoma. Nature 2011;469:539–42. Husgafvel-Pursiainen K, et al. Loss of heterozygosity at 6q is frequent 27. Beroukhim R, Brunet JP, Di Napoli A, Mertz KD, Seeley A, Pires MM, and concurrent with 3p loss in sporadic and familial capillary he- et al. Patterns of gene expression and copy-number alterations in von mangioblastomas. J Neuropathol Exp Neurol 2004;63:1072–9. Hippel–Lindau disease–associated and sporadic clear cell carcinoma 48. Brugarolas J, Kaelin WG, Jr.Dysregulation of HIF and VEGF is a of the kidney. Cancer Res 2009;69:4674–81. unifying feature of the familial hamartoma syndromes. Cancer Cell 28. Yoshimoto T, Matsuura K, Karnan S, Tagawa H, Nakada C, Tanigawa 2004;6:7–10. M, et al. High-resolution analysis of DNA copy number altera- 49. Toschi A, Lee E, Gadir N, Ohh M, Foster DA. Differential depen- tions and gene expression in renal clear cell carcinoma. J Pathol dence of hypoxia-inducible factors 1 alpha and 2 alpha on mTORC1 2007;213:392–401. and mTORC2. J Biol Chem 2008;283:34495–9. 29. Morris MR, Hughes DJ, Tian YM, Ricketts CJ, Lau KW, Gentle D, 50. Grossman HB, Wedemeyer G, Ren LQ. Human renal carcinoma: et al. Mutation analysis of hypoxia-inducible factors HIF1A and characterization of five new cell lines. J Surg Oncol 1985;28:237–44. HIF2A in renal cell carcinoma. Anticancer Res 2009;29:4337–43. 51. Anglard P, Trahan E, Liu S, Latif F, Merino MJ, Lerman MI, et al. 30. Shinojima T, Oya M, Takayanagi A, Mizuno R, Shimizu N, Murai Molecular and cellular characterization of human renal cell carci- M. Renal cancer cells lacking hypoxia inducible factor (HIF)-1alpha noma cell lines. Cancer Res 1992;52:348–56. expression maintain vascular endothelial growth factor expression 52. Tatsumi T, Herrem CJ, Olson WC, Finke JH, Bukowski RM, Kinch through HIF-2alpha. Carcinogenesis 2007;28:529–36. MS, et al. Disease stage variation in CD4+ and CD8+ T-cell reactivity 31. Kozlowski P, Roberts P, Dabora S, Franz D, Bissler J, Northrup H, to the receptor tyrosine kinase EphA2 in patients with renal cell car- et al. Identification of 54 large deletions/duplications in TSC1 and cinoma. Cancer Res 2003;63:4481–9.

234 | CANCER DISCOVERY AUGUST 2011 www.aacrjournals.org

HIF1α Is a 14q Kidney Cancer Suppressor Gene RESEARCH ARTICLE

53. Ebert T, Bander NH, Finstad CL, Ramsawak RD, Old LJ. 57. GENE-E [homepage on the Internet]. Cambridge (MA): The Broad Establishment and characterization of human renal cancer and nor- Institute of MIT and Harvard. [cited 2010 Aug 6]. Available from: mal kidney cell lines. Cancer Res 1990;50:5531–6. http://www.broadinstitute.org/cancer/software/GENE-E/ 54. MRC-Holland MLPA.com [homepage on the Internet]. Amsterdam, 58. Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, The Netherlands: MRC-Holland; ©2011. [cited 2011 May 2]. Gillette MA, et al. Gene set enrichment analysis: a knowledge-based Available from: http://www.mlpa.com approach for interpreting genome-wide expression profiles. Proc 55. Zhang Q, Gu J, Li L, Liu J, Luo B, Cheung HW, et al. Control of cy- Natl Acad Sci U S A 2005;102:15545–50. clin D1 and breast tumorigenesis by the EglN2 prolyl hydroxylase. Cancer Cell 2009;16:413–24. 56. Wong WH, Li C. Cheng Li Lab [database on the Internet]. Boston (MA): Harvard School of Public Health and Dana-Farber Cancer Institute. [cited 2011 Apr 30]. Available from: http://www.dChip.org/

AUGUST 2011 CANCER DISCOVERY | 235

Genetic and Functional Studies Implicate HIF1α as a 14q Kidney Cancer Suppressor Gene

Chuan Shen, Rameen Beroukhim, Steven E. Schumacher, et al.

Cancer Discovery 2011;1:222-235. Published OnlineFirst June 7, 2011.

Updated version Access the most recent version of this article at: doi:10.1158/2159-8290.CD-11-0098

Supplementary Access the most recent supplemental material at: Material http://cancerdiscovery.aacrjournals.org/content/suppl/2011/06/16/2159-8290.CD-11-0098.DC 1

Cited articles This article cites 55 articles, 18 of which you can access for free at: http://cancerdiscovery.aacrjournals.org/content/1/3/222.full#ref-list-1

Citing articles This article has been cited by 38 HighWire-hosted articles. Access the articles at: http://cancerdiscovery.aacrjournals.org/content/1/3/222.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerdiscovery.aacrjournals.org/content/1/3/222. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.