Identification of novel VHL target genes and relationship to ... - Nature

Oncogene (2005) 24, 4549–4558

& 2005 Nature Publishing Group All rights reserved 0950-9232/05 $30.00 www.nature.com/onc

Identification of novel VHL target genes and relationship to hypoxic response pathways Esther N Maina1, Mark R Morris1,2, Malgorzata Zatyka1, Raju R Raval3, Rosamonde E Banks4, Frances M Richards1,2, Claire M Johnson5 and Eamonn R Maher*,1,2 1 Section of Medical & Molecular Genetics, Department of Paediatrics and Child Health, University of Birmingham, The Medical School, Birmingham B15 2TT, UK; 2Cancer Research UK Research Group, University of Birmingham, The Medical School, Birmingham B15 2TT, UK; 3Wellcome Trust Centre for Human Genetics, University of Oxford, Roosevelt Drive, Oxford OX3 7BN, UK; 4Cancer Research UK Clinical Centre, St James’s University Hospital, Leeds LS9 7TF, UK; 5Discovery Biology, Pfizer Global R&D, Ramsgate Road, Sandwich, Kent CT13 9NJ, UK

Upregulation of hypoxia-inducible factors HIF-1 and HIF-2 is frequent in human cancers and may result from tissue hypoxia or genetic mechanisms, in particular the inactivation of the von Hippel–Lindau (VHL) tumour suppressor gene (TSG). Tumours with VHL inactivation are highly vascular, but it is unclear to what extent HIFdependent and HIF-independent mechanisms account for pVHL tumour suppressor activity. As the identification of novel pVHL targets might provide insights into pVHL tumour suppressor activity, we performed gene expression microarray analysis in VHL-wild-type and VHL-null renal cell carcinoma (RCC) cell lines. We identified 30 differentially regulated pVHL targets (26 of which were ‘novel’) and the results of microarray analysis were confirmed in all 11 novel targets further analysed by real-time RT–PCR or Western blotting. Furthermore, nine of 11 targets were dysregulated in the majority of a series of primary clear cell RCC with VHL inactivation. Three of the nine targets had been identified previously as candidate TSGs (DOC-2/DAB2, CDKN1C and SPARC) and all were upregulated by wild-type pVHL. The significance for pVHL function of two further genes upregulated by wild-type pVHL was initially unclear, but re-expression of GNG4 (G protein gamma-4 subunit/ guanine nucleotide-binding protein-4) and MLC2 (myosin light chain) in a RCC cell line suppressed tumour cell growth. pVHL regulation of CDKN1C, SPARC and GNG4 was not mimicked by hypoxia, whereas for six of 11 novel targets analysed (including DOC-2/DAB2 and MLC2) the effects of pVHL inactivation and hypoxia were similar. For GPR56 there was evidence of a tissuespecific hypoxia response. Such a phenomenon might, in part, explain organ-specific tumorigenesis in VHL disease. These provide insights into mechanisms of pVHL tumour suppressor function and identify novel hypoxiaresponsive targets that might be implicated in tumorigen-

*Correspondence: ER Maher, Section of Medical and Molecular Genetics, Department of Paediatrics and Child Health, University of Birmingham, The Medical School, Edgbaston, Birmingham, B15 2TT, UK. E-mail: [email protected] Received 23 September 2004; revised 17 January 2005; accepted 8 February 2005; published online 11 April 2005

esis in both VHL disease and in other cancers with HIF upregulation. Oncogene (2005) 24, 4549–4558. doi:10.1038/sj.onc.1208649 Published online 11 April 2005 Keywords: VHL; HIF gene expression; microarray hypoxia; renal cell carcinoma

Introduction von Hippel–Lindau disease (VHL) (OMIM 193300) is a dominantly inherited familial syndrome characterized by the development of vascular tumours (haemangioblastomas) in the retina and central nervous system, renal cell carcinoma (RCC), phaeochromocytoma and pancreatic endocrine tumours (Maher et al., 1990; Kaelin and Maher, 1998). Somatic inactivation of the VHL tumour suppressor gene (TSG) by allele loss, mutation or promoter hypermethylation occurs in most sporadic clear cell RCC (Foster et al., 1994; Gnarra et al., 1994; Clifford et al., 1998). The VHL TSG gene product complexes with elongin C, elongin B and Cul2 proteins to form a ubiquitin ligase complex (VCBC) with structural similarities to the yeast SCF protein complex (Skpl-Cdc53/Col1-F-box) (Pause et al., 1997; Lonergan et al., 1998; Iwai et al., 1999; Lisztwan et al., 1999). The VCBC complex targets the a regulatory subunits of hypoxia inducible factors HIF-1 and HIF-2 for oxygen-dependent proteolysis (Maxwell et al., 1999; Cockman et al., 2000; Ohh et al., 2000; Tanimoto et al., 2000). HIF-1 is a basic helix–loop–helix heterodimeric transcription factor that plays a critical role in cellular responses to hypoxia. In VHL defective cells, HIF-a subunits are constitutively stabilized, leading to overexpression of HIF-1 and HIF-2 and an extensive range of hypoxia-inducible mRNAs including those involved in energy metabolism, angiogenesis and apoptosis (e.g. glucose transporter (GLUT-1) and vascular endothelial growth factor (VEGF)). Both HIF-dependent and HIF-independent mechanisms have been implicated in pVHL tumour suppressor

Identification of novel VHL target genes EN Maina et al

4550

activity (Kondo et al., 2002, 2003; Maranchie et al., 2002; Mack et al., 2003; Zimmer et al., 2004). However, the complex genotype–phenotype correlations observed in VHL disease suggest that pVHL has multiple functions. Accordingly, a range of putative pVHL functions have been described including regulation of cell cycle exit control (Pause et al., 1998), fibronectin binding and extracellular matrix assembly (Ohh et al., 1998), mRNA stability (Gnarra et al., 1996; Levy et al., 1996; Knebelmann et al., 1998), microtubule stability (Hergovich et al., 2003) and interaction with atypical PKCs (Pal et al., 1998; Okuda et al., 2001). However, some of these functions may be HIF-dependent. Analysis of global mRNA expression patterns by oligonucleotide microarray technology or SAGE analyses has provided an unbiased approach to identifying pVHL target genes (Wykoff et al., 2000; Zatyka et al., 2002; Jiang et al., 2003; Wykoff et al., 2004). As the identification of novel pVHL targets might provide further insights into mechanisms of pVHL tumour suppressor activity, we used high-density oligonucleotide arrays to identify differentially expressed genes in pVHL wild-type and null RCC cell lines. We then investigated in detail a subset of novel candidate pVHL target genes to determine whether they were regulated by HIF-dependent or -independent mechanisms.

Results array analysis of VHL þ and VHL null RCC4 cell line

Expression

To identify novel pVHL targets, we used the Affymetrix U95v2 GeneChip microarrays to assay 12 600 mRNA transcripts in stable RCC4 transfectants expressing empty vector or full-length wild-type human pVHL (RCC4/VHL). To investigate the consistency between experiments, we performed two independent experiments and three clones in total of each cell line (RCC4/ VHL- and RCC4/VHL þ ) were analysed. As illustrated in Figure 1, there was a good correlation between the

results of the two experiments and each of the two clones. We identified 30 genes (0.2 of total genes analysed) that demonstrated a X2-fold difference between VHL þ and VHL- cells in both experiments and in the three different clones. In all, 26 of the candidate pVHL target genes had not been reported previously (Table 1), whereas four had (VEGF, transgelin, Differentiated embryo chondrocyte 1 (DEC1) and Cyclin D1(CCND1), transgelin and DEC1) (Gnarra et al., 1996; Iliopoulos et al., 1996; Wykoff et al., 2000; Zatyka et al., 2002). In all, 11 of the candidate novel pVHL targets GNG4 (G protein gamma-4 subunit/guanine nucleotide-binding protein-4), GPR56 (G-proteincoupled receptor), KIAA0838 GLS (KIAA0838), DOC-2/DAB2 (differentially expressed in ovarian carcinoma-2/disabled-2), P311, CDKN1C (cyclin-dependent kinase 1C inhibitor), MLC2 (myosin light chain), SPARC (secreted, protein, rich in cysteine), Claudin 4, E3 (E3) and MCT3 (monocarboxylate transporter) were selected for further investigation. 8 of 11 were upregulated by wild-type pVHL and 3 were downregulated. Confirmation of gene expression microarray analysis in a subset of genes Real-time PCR analysis of 11 novel candidate pVHL target genes To confirm the expression array results we used real-time PCR analysis. Thus, total RNA was prepared from stably transfected RCC4 cell lines

Table 1 Gene

Primer sequence

MLC2

F: 50 CATCCAATGTCTTCGC 30 R: 50 CCGGTACATCTCGTCC 30 F: 50 TTTGCAGTGCTGATGGTCTC 30 R: 50 TGGTGTAGTCCCGAGGTTTC 30 F: 50 GACGCAGAAGAGTCCACCAC 30 R: 50 CCTGCTGGAAGTCGTAATCC 30 F: 50 CATGCACTGCACTTCTTCGT 30 R: 50 AAGCCTGCATTTTCAGCAGT 30 F: 50 CAGAAGACGAACCAGAGACAAA 30 R: 50 CCCCAAGGAGGACAAATC 30 F: 50 ACAAGTGCAACCAATGGTCA 30

GPR56 CDKN1C P311 GLS DOC-2/ DAB2 GNG4 Claudin 4 SPARC MCT3 E3

Figure 1 Technical variability of the Affymetrix GeneChip data set. The data are represented as log/log scatter plots of the expression values provided by GENECHIP software. The upper and lower boundaries represent a twofold change in expression. Data from VHL-null clone 1 are plotted against expression data from VHL-null clone 2 for the same genes. The data show correlation with few outliers Oncogene

Primer sequence for novel target genes

VEGF PAI-1 b-Actin

R: 50 CCGGTTGTCTGTCACATCAC 30 F: 50 AGCAGGGGCAGTAGAATGAA 30 R: 50 GGATGGGTGTTGGTCTCACT 30 F: 50 CTCCATGGGGCTACAGGTAA 30 R: 50 AGCAGCGAGTCGTACACCTT 30 F: 50 GTGCAGAGGAAACCGAAGAG 30 R: 50 TCATTGCTGCACACCTTCTC 30 F: 50 GCACCCACAAGTTCTCCAGT 30 R: 50 CAAAATCAGGGAGGAGGTGA 30 F: 50 GGGCTGACTTCAATCCATGT 30 R: 50 GCTGCCCAGAATACCAATGT 30 F: 50 CAGCGCAGCTACTGCCATCCAATCGAGA30 R: 50 GCTTGTCACATCTGCAAGTACGTTCGCTTA 30 F: 50 AGATCGAGGTGAACGAGAGTGGCACG 30 R: 50 TTTGTCCCAGATGAAGGCGTCTTTCC 30 F: 50 GGCGACGAGGCCCAGA 30 R: 50 CGATTTCCCGCTCGGC 30


4551

expressing vector alone (RCC4/VHL) or wild-type pVHL (RCC4/VHL þ ). Expression of target genes was normalized to b-actin and results for the 11 genes analysed are shown in Table 1. Real-time PCR analysis confirmed the results of the microarray analysis for all 11 genes. Western blot analysis of VHL target genes To determine whether pVHL regulation of target gene transcripts was reflected in protein levels, Western blots of RCC4/VHL þ and RCC4/VHL protein extracts were probed with anti-CDKNIC and anti-MCT-3 antibodies. In each case the effect of pVHL on the transcripts was mirrored by the effect on protein levels. Thus pVHL upregulated CDKNIC and downregulated MCT3 protein expression (Figure 2). Transient transfection of VHL into RCC4 cells We transiently transfected empty vector (pCDNA3.1, Invitrogen) or the wild-type pVHL (pCDNA 3.1-VHL (1– 213) HA) into RCC4 cells to determine whether the results of transient transfection were consistent with those for stable transfection. The expression of four genes was analysed (CDKN1C, P311, Claudin 4 and E3) by real-time PCR and these demonstrated similar results in transient and stable transfections (Table 2). VHL target gene expression in primary tumours The relationship between pVHL regulation of target genes in an isogenic cell line and in primary RCC with VHL inactivation (somatic VHL mutation73p25 allele loss) was analysed in up to nine paired normal–tumour pairs. For CDKN1C (n ¼ 8), GNG4 (n ¼ 8), GPR56

(n ¼ 8), MCT3 (n ¼ 4), MLC2 (n ¼ 6), P311 (n ¼ 8) and SPARC (n ¼ 7), all tumours tested demonstrated altered expression consistent with the cell line data (results showed threshold of X2-fold differential expression). For DOC-2/DAB2 6/7 clear cell RCC with VHL inactivation demonstrated downregulation compared to matched normal tissue. However, E3 and CDN4 demonstrated altered expression in only a minority of the primary tumours tested (4/9 and 1/3, respectively) (data not shown). Analysis of the tumour-suppressing activity of MLC2 and GNG4 in RCC In order to determine the relevance of pVHL upregulation of MLC2 and GNG4 for pVHL tumour suppressor function, we determined the effect of MLC2 and GNG4 expression on RCC cell line growth and migration. Inhibition of colony formation by MLC2 and GNG4 in tumour cell lines Colony formation assays were used to test for the growth-inhibitory effect of MLC2 and GNG4 in a RCC cell line with VHL inactivation and loss of MLC2 and GNG4 expression. Empty plasmid pCDNA3.1 or pcDNA3.l containing the full open reading frame of MLC2 and GNG4 was transfected into the RCC SKRC45 tumour cell lines. Cells were selected with geneticin (G418), 48 h after transfection, and resistant colonies developing 14–21 days later were stained with 0.4% crystal violet. The number of G418resistant colonies after transfection with GNG4 was reduced by 72% (s.d.72.6%, P ¼ 0.009) compared with transfection with empty vector control, in three independent experiments (Figure 3a). Colony formation efficiency after transfection with MLC2 was reduced by 61.3% (s.d.77.7%, P ¼ 0.002) compared with transfection with empty vector control in three independent experiments (Figure 3b). Re-expression of MLC2 and GNG4 and cell motility Following growth in serum-free media, confluent dishes of stable SKRC45-empty vector and SKRC45-wtMLC2 and -GNG4 were scratched with a 200 ml pipette tip. After 24 h, both the SKRC45 cells re-expressing MLC2 and GNG4 and SKRC45-pCDNA 3.1 cells had fully invaded the resulting ‘wound’, suggesting that GNG4 and MLC2 TSGs are not involved in the regulation of cell mortility.

Figure 2 Western blot analyses of CDKN1C and MCT3 protein levels. CDKN1C was shown to be upregulated by wild-type pVHL, while MCT3 showed a downregulation in wild-type pVHL. Two clones representing VHL þ and VHLnull were used. These analyses were standardized against b-actin

Regulation of VHL target genes by oxygen To determine whether the novel pVHL target genes were likely to be regulated by HIF-dependent or -independent mechanisms, we investigated whether they demonstrated evidence of hypoxic induction. Thus, we analysed the effect of hypoxia on expression of the 11 target genes in a pVHL þ RCC4 cell line such that target gene expression was compared in hypoxia (1%) and normoxia (21%) by real-time PCR to expression of the known hypoxia-inducible pVHL target gene, PAI-1. As expected, expression of PAI-1 was upregulated by hypoxia, matching the effect that VHL inactivation Oncogene


4552 Table 2 Results for 26 novel VHL target genes and four known pVHL target genes Accession no.

Gene

Description

Microarray Fold change

Real time-PCR Fold change

Signalling/Transporter

MCT3

U31382 U53446 U81800

Guanine nucleotide-binding protein Mitogen-responsive phosphoprotein Monocarboxylate transporter

—a 3.9 10.5

21 10 5.5

Metabolism/catalytic AB020645 KIAA0836 U78302 DECR D88667 CST Z52244 HO-1 X51956 ENO2 L07956 GBE1

b-Actin Glutaminase 2,4 dienoyl-coA reductase Cerebroside sulfotransferase Heme oxygenase 1 Enolase 2 (gamma neuronal) 1,4-alpha-glucan branching enzyme

5.8 2 3.4 2.6 8.3 2.1

6

Proliferation/adhesion/differentiation X78947 CTGF J03040 SPARC/Osteonectin

Connective tissue growth factor Secreted protein, acidic, cysteine rich

3.2 —a

Contraction/calcium M95787 Transgelinb J02854 MLC-2 D17408 Calponin

22 kDa smooth muscle protein 20 kDa myosin light chain Calponin 1, basic smooth muscle

Transcription AF055376 L07648

C-MAF MXI 1A

Short form transcription factor Max-interacting protein 1

—a 2.8

Apoptosis AF002697 U37518

BNIP3 TRAIL

E1B 19k/Bcl-2-binding protein Nip3 TNF-related apoptosis inducing ligand

4.9 3.1

Cell cycle U22398 M64349

CDKN1C CCND1b

Cyclin-dependent kinase inhibitor 1C Cyclin D1

4.6 3.6

Angiogenesis AF024710

VEGFb

Vascular endothelial growth factor

3.6

Others J02931 M34715 AB000712 U30521 AJ011001 K03000 U53445 U03036 AB004066

E3 PSG1 Claudin 4 P311 TM7XN1 ALDH1 Doc1 LUCA1/HYAL1 STRA13 (DEC1)b

Coagulation factor III,Tissue factor Pregnancy-specific beta-1-glycoprotein hCPE-R for CPE-receptor

6 2.1 4.3 2.5 2 4.6 5.1 —a 2.9

G-Protein gamma-4 subunit DOC-2/DAB2 MCT3

G-protein-coupled receptor (GPR56) Aldehyde dehydrogenase 1 member 1A Ovarian cancer downregulated myosin heavy chain Putative tumour suppressor Differentiated embryo-chondrocyte expressed gene 1

3.8 2.5 8.45

3

9

8

30 3 3 2

() Downregulated by wild-type pVHL. aOnly detected in wild-type pVHL. bKnown pVHL target genes. In all, 11 of the novel candidate VHL target genes were confirmed by quantitative real-time PCR

has on this gene. Similarly, three target genes downregulated by wild-type pVHL (Claudin 4, E3 and MCT3) were all induced by hypoxia. Of the remaining eight novel target genes that were upregulated by wild-type pVHL, hypoxia downregulated DOC-2/DAB2, MLC2 and GPR56. Three novel pVHL target genes demonstrated no significant hypoxic response (GNG4, GLS and SPARC), while two genes that were upregulated by wild-type VHL were also upregulated by hypoxia (P311 and CDKN1C). As CCND1 has previously been demonstrated to be Oncogene

hypoxia inducible in a RCC cell line expressing wildtype pVHL but not in bladder or breast cell lines, we analysed the effect of hypoxia on gene expression in EJ29 (bladder cancer cell line) and HBL-100 (breast cancer cell line) for PAI-1, MCT3, MLC2, P311, CDKN1C, DOC-2/DAB2, GPR56 and GNG4. Except for GPR56, the results in RCC4 matched those in EJ-28 or HBL100. However, GPR56 demonstrated no response to hypoxia in the two non-RCC cell lines, but was downregulated by hypoxia in a RCC4 cell line expressing wild-type VHL (Table 3).


4553

a

120

% Colony growth

100 80 60 40 20 0 pCDNA3.1

pCDNA3.1 GNG4

% Colony growth

b 120 100 80 60 40 20 0 pCDNA3.1 MLC2

pCDNA3.1

Figure 3 (a, b) Representation of colony suppression effect of wild-type GNG4 (pcDNA3.1 GNG4) compared to empty vector control (pcDNA3.1) (see top panel) and wild-type MLC2 (pcDNA3.1 MLC2) compared to empty vector control (pcDNA3.1) (see lower panel). Studies were performed in the SKRC45 RCC cell line that does not express endogenous MLC2 or GNG4

Table 3

Effect of transient transfection of pVHL into RCC4-null on expression of four novel candidate VHL target genes

Gene

Accession no.

Claudin 4 E3 CDKN1C P311

AB000712 J02931 U22398 U30521

Fold change k2 k6 m2 m2

k (fold-change) downregulation by wild-type VHL. m (fold-change) upregulation by wild-type VHL

Discussion We used high-density oligonucleotide expression arrays to examine differences in mRNA expression resulting from stable transfection of wild-type pVHL into a VHL defective RCC cell line. Using a significance threshold of two fold differential expression, we identified 30 genes

that were differentially regulated. Although other groups have used similar strategies to identify pVHL targets (Wykoff et al., 2000, 2004; Zatyka et al., 2002; Jiang et al., 2003; Staller et al., 2003) only four (VEGF, CCND1, transgelin and DEC1) of the 30 genes had been previously identified as a pVHL target. We selected 11 of the 26 novel target genes for further analysis and verified the microarray expression results by real-time quantitative RT–PCR or Western analysis in all cases. Thus, it would appear that our experimental approach had a low false-positive rate. However, we did not detect candidate pVHL target genes that have been identified in previous microarray studies (e.g. CD59, integrin alpha-3, JM4, LRP1) (Wykoff et al., 2000, 2004; Zatyka et al., 2002; Jiang et al., 2003). Possible reasons for these differences include variables such as the specific cell lines used, the microarray chips employed and the different thresholds selected for analysis. However, we note that although Wyckoff et al. (2004) and Jiang et al. (2003) both analysed the 768-0 RCC cell line different targets were identified and differences between studies may also relate to methodological differences such as sample preparation and hybridization. As a pVHL target gene is not necessarily relevant for pVHL tumour suppressor activity, we reviewed whether the known or putative functions of the 11 novel pVHL targets were consistent with that expected for a role in pVHL tumour suppressor activity. Thus, a priori we might expect that the eight genes (GNG4, GPR56, GLS, DOC-2/DAB2, P311, CDKN1C, MLC-2 and SPARC) upregulated by wild-type pVHL would have antioncogenic properties and the three genes (E3, Claudin 4 and MCT3) downregulated by wild-type pVHL would have oncogenic functions. DOC-2/DAB2 is a mitogen-responsive phosphoprotein that is thought to be an essential component of the wnt/TGF-beta signaling pathway (Hocevar et al., 2001; Wang et al., 2002). DOC-2/DAB2 is a potent tumour suppressor in many cancer types and is downregulated in ovarian, breast, prostate and colorectal cancers (Mok et al., 1994; Schwahn and Medina, 1998; Tseng et al., 1999; Kleeff et al., 2002; Wang et al., 2002). DOC-2/DAB2 forms a protein complex with hDAB2IP (human DAB2 interactive protein) that modulates the Ras-mediated signaling pathway (Wang et al., 2002). This complex has been detected in the basal cell population of the prostate gland (Tseng et al., 1999; Wang et al., 2002) and may orchestrate the differentiation and proliferation potential of these cells during gland development. pVHL-null RCC cells are reported to have a defect in cell cycle exit control (Pause et al., 1998) and CCND1 has been identified as a pVHL target (Horiuchi et al., 2002; Zatyka et al., 2002; Baba et al., 2003). CDKN1C is a maternally expressed imprinted gene that is mutated in a subset of patients with the Beckwith–Wiedemann syndrome (BWS) (Hatada et al., 1996; Lam et al., 1999). BWS is characterized by prenatal and/or postnatal overgrowth and susceptibility Oncogene


4554

to embryonal cancers such as Wilms’ tumours (Maher and Reik, 2000). Somatic CDKN1C mutations have not been reported in human cancers, but epigenetic silencing of CDKN1C has been reported in lymphoid and solid cancers (Kikuchi et al., 2002; Li et al., 2002), although RCC has not yet been analysed. CDKN1C induces cell cycle arrest by inhibiting the activity of cyclin-dependent kinases, so both genetic and functional evidence implicate CDKN1C as a candidate TSG. SPARC, also called osteonectin or BM-40, is a calcium-binding matricellular glycoprotein, which is expressed in tissue undergoing remodeling (Bornstein and Sage, 2002). Although its precise role is unclear, SPARC plays a modulatory role in cell–matrix interaction (Lane and Sage, 1994), induces cell rounding, blocks cell spreading and adhesion, and inhibits endothelial cell migration (Sage et al., 1989; Hasselaar and Sage, 1992; Kupprion et al., 1998). SPARC has been reported to have antiangiogenic activity in cancer (Chlenski et al., 2002) and to induce apoptosis and reduce in vitro growth and tumorigenesis of cancer cells (Yiu et al., 2001). P311 encodes a highly conserved protein that is upregulated by pVHL. Although the precise function of P311 is unknown, P311 mRNA expression is downregulated by Met and its ligand, hepatocyte growth factor/scatter factor (HGF). While P311 has not been implicated in renal tumorigenesis previously, germlineinactivating mutations in MET cause hereditary papillary RCC (Kuroda et al., 2003) and pVHL inhibits HGF-induced invasion and branching morphogenesis in renal carcinoma cells (Koochekpour et al., 1999). These observations suggest a synergy between the loss of VHL function and Met signaling. The potential tumorigenic roles of the remaining three genes upregulated by pVHL introduction were not clear. Thus, GLS (KIAA0838) catalyses the first reaction in the primary pathway for the renal catabolism of glutamine. GLS has three isoforms (kga (analysed here), gam and gac) that are produced by alternative splicing. Kga is expressed predominantly in brain and kidney. It has been suggested that increased rates of proliferation and levels of malignancy in colorectal tumour cells may be attributable to downregulation of one or more isoforms of GLS, accompanied by increased glucose metabolism (Turner and McGivan, 2003). GNG4 is a member of the G protein gamma subunits multigene family (Kleuss et al., 1993), which is required for coupling of the muscarinic receptor to voltage-sensitive calcium channels (Kalyanaraman et al., 1998). To investigate whether GNG4 has tumour suppressor activity, we transformed wild-type GNG4 into a RCC cell line with loss of GNG4 expression. Restoration of GNG4 expression was associated with reduced ability to form colonies, but there was no apparent effect on cell migration. Similarly, MLC2 is known to play an important role in regulation of both smooth muscle and non-smooth muscle cells and has been shown to be downregulated in many transformed cell lines (Kumar and Chang, 1992). Again, we found Oncogene

that restoration of MLC2 expression in a VHL-defective cell line without MLC2 expression resulted in suppression of in vitro colony formation with no apparent effect on cell migration. These results suggest that both MLC2 and GNG4 can be considered to be candidate TSGs implicated in pVHL tumour suppressor activity. Three of 11 confirmed pVHL targets were downregulated by wild-type pVHL. If these were germane to pVHL tumour suppressor activity, we would predict that they would function to promote tumour growth. E3 (E3, coagulation factor III or thromboplastin), a transmembrane glycoprotein, is a receptor for factor VII/VIIa, and is the main initiator of blood coagulation (Edgington et al., 1991). E3 is also involved in a variety of cellular processes, including intracellular signaling, cellular proliferation and the development of blood vessels. E3 expression appears to be a marker for malignant angiogenesis in human cancers (Contrino et al., 1996; Guan et al., 2002; Ohta et al., 2002). Thus, from the described putative functions of E3, downregulation of E3 by pVHL would be expected to suppress tumorigenesis. Claudin 4 was also downregulated by wild-type pVHL. The tight junction (TJ) and its adhesion molecules, claudins, are responsible for the barrier function of epithelia (Furuse et al., 2002). Claudin 4 (CDN4) encodes an integral membrane protein and is a component of tight junction strands. It is reported to be overexpressed in most pancreatic cancer tissues and cell lines and also in several other gastrointestinal tumours (Michl et al., 2001), so again the proposed functional effects of Claudin 4 would predict that wild-type pVHL should downregulate (rather than upregulate) Claudin 4 expression, if regulation of Claudin 4 was relevant to VHL tumour suppression. Interestingly, both E3 and CDN4 did not consistently demonstrate altered expression in primary RCC with VHL inactivation. Finally, the proton-linked MCT3 catalyses the rapid transport across the plasma membrane of many monocarboxylates such as lactate, pyruvate, branched-chain oxo-acids derived from leucine, valine and isoleucine, and the ketone bodies acetoacetate, beta-hydroxybutyrate and acetate. MCTs are ubiquitously distributed among many tissues. Transmembrane lactate transport is mediated mainly by MCTs, so MCT function is important for pH regulation. An acidic extracellular pH is a fundamental property of the malignant phenotype and overexpression of carbonic anhydrase IX by VHL-inactivated RCC has been suggested to contribute to the tumour microenvironment by maintaining extracellular acidic pH and helping cancer cells grow and metastasize (Ivanov et al., 2001). Further investigation is required to determine if MCT3 downregulation by pVHL is relevant to pVHL function. The known or putative functions of the majority of novel pVHL-target genes we identified suggest that changes in their expression following pVHL inactivation would contribute to tumour development. The extent to which HIF-1 and HIF-2 dysregulation accounts for tumour development after VHL inactivation is unclear.


4555

Thus Kondo et al. (2003) reported that HIF-2 overexpression per se promoted tumorigenesis, but Maranchie et al. (2002) found that normoxic stabilization of HIF-1 was not sufficient to promote tumorigenesis. Further, genotype–phenotype correlation suggests that pVHL has multiple functions. In the current study, six of 11 confirmed pVHL targets investigated were demonstrated to be hypoxia-regulable such that the results of HIF overexpression would mimic the effects of VHL inactivation. Previously it has been demonstrated that CCND1 is hypoxia-inducible in RCC cell lines, but not in non-renal cell lines (e.g. breast and bladder) (Horiuchi et al., 2002; Zatyka et al., 2002; Baba et al., 2003). We found that GPR56 was suppressed by VHL inactivation and hypoxia in a RCC cell line, but hypoxia did not affect expression in breast or bladder cancer cell lines. This finding demonstrates that CCND1 is not unique and that a distinct hypoxiarelated signalling pathway exists in renal cells. Such a pathway may play a critical role in determining the tissue-specific effects of VHL inactivation. We identified five novel VHL targets that were not hypoxia-responsive (GNG4, KIAA0838 and SPARC) or for which hypoxiainduced changes in gene expression were opposite to those seen with VHL inactivation (P311 and CDKN1C). Further work is required to investigate the precise significance of these genes for pVHL tumour suppression, and their precise mechanism of regulation. However, SPARC and CDKN1C have previously been implicated as TSGs and we have demonstrated that reexpression of GNG4 in a RCC cell line suppressed tumour cell growth. Thus, these findings provide evidence that both HIF-dependent and -independent mechanisms contribute to pVHL tumour suppressor activity. While it is generally accepted that HIF dysregulation is closely linked to the characteristic vascularity of VHLrelated tumours, the importance of tumour cell growth is less well accepted. Although pVHL has a critical role in regulating proteosomal degradation of target proteins such as the HIF-a subunits, this mechanism cannot directly be implicated in regulation of target genes which were identified by changes in RNA expression. We note that VHL inactivation and RASSF1A, another TSG implicated in the pathogenesis of RCC, have similar effects on SPARC expression (Ivanov et al., 2001; Morrissey et al., 2001; Agathanggelou et al., 2003), consistent with a role for this candidate TSG in RCC development. Six of the 11 confirmed novel pVHL target genes appeared to be regulated by HIF-dependent mechanisms. HIF-dependent downregulation of DOC2/DAB2, MLC2, etc. has wide-ranging implications because of the involvement of the HIF transcription factors in both VHL-related tumours and in intratumoral hypoxic responses in many non-VHL-related human cancers. Thus, CXCR4 (a chemokine receptor implicated in organ-specific metastastic processes) was demonstrated to be regulated by pVHL in a HIFdependent manner (Staller et al., 2003). These findings underscore the importance of understanding the mechanisms of pVHL tumour suppression.

Materials and methods Cell culture and transfection Stable transfection VHL gene expression constructs containing empty vector (pCDNA3.1, Invitrogen) or the wild-type pVHL (pCDNA 3.1-VHL (1–213). HA) were created and established as stable transfectants into a VHL-defective RCC4 cell line lacking endogenous VHL (Clifford et al., 2001). Cells were grown in DMEM supplemented with 10% foetal calf serum (FCS), L-glutamine (2 mM), penicillin (50 IU/ml), streptomycin sulphate (50 mg/ml) and G418 (1 mg/ml). Cells approaching confluence were harvested and the resulting RNA used for gene expression analysis by real-time PCR. Transient transfection Transient transfection was similarly set up using pCDNA 3.1VHL (1–213) or pCDNA 3.1. Cells were grown in 60 mm dish for 24hrs in DMEM supplemented with 10% FCS, Lglutamine (2 mM), penicillin (50 IU/ml) and streptomycin sulphate (50 mg/ml). Media was replenished 4 h before transfection. Plasmid (10 mg) was added to 200 ml of 250mM CaCl2 and mixed with an equal volume of 2 HEPES phosphate buffer. After 10 min incubation, the mixture was added to each dish containing 4 ml of media. The growth media was replenished 24 h post-transfection and left for another 24 h before the cells were harvested for RNA preparation. Oligonucleotide array expression analysis Poly (A) mRNA purification, biotin labelling of cRNA and hybridization were performed according to the manufacturer’s protocols. Double-stranded cDNA was synthesized from 5 mg poly (A) mRNA using reverse transcriptase Avian Myeloblastosis Virus Superscript Choice system (Gibco BRL) with a T7-poly dT primer. cDNA (1 mg) was in vitro transcribed (Ambion, Austin, TX, USA) in the presence of biotinylated UTP and CTP (Enzo Diagnostics). Target for hybridization was prepared by combining 15 mg of fragmented cRNA with hybridization buffer (100 mM MES, 1 M NaCl, 20 mM EDTA and 0.01% Tween-20) containing 50 pM control oligonucleotide B2, control cRNA cocktail, 0.1 mg/ml herring sperm DNA and 0.5 mg/ml acetylated BSA. Target samples were hybridized for 16 h at 451C to a set of oligonucleotide arrays (HG-U95Av2) Affymetrix, Santa Clara, CA, USA) containing probes for 12 600 human genes. Arrays were washed at 301C with nonstringent buffer 6 SSPET (3 M NaCl, 0.2 M NaH2PO4, 0.02 M EDTA, 0.005% Triton X-100, 0.01% Tween-20), then at 501C with stringent buffer (100 mM, 0.1 M NaCl, 0.01% Tween-20). Arrays were then stained with streptavidin-phycoerythrin (Molecular Probes). Fluorescence intensities were captured with a laser confocal scanner (Hewlett-Packard) and were analysed with the microarray software, MAS 4.0 software (Affymetrix). Each hybridization was first analysed using the Absolute Analysis software and then by comparative analysis between the cell lines. The absolute analysis yielded the average fluorescence difference and an absolute call of present or absent for each transcript. In the comparative analysis, transcripts were considered as significantly altered over control when the ratio of average fluorescence difference of experimental transcript to baseline was twofold. Analysis of RNA by real-time quantitative RT–PCR Total RNA from each sample used on the GeneChip was treated with DNase 1(Qiagen) according to the manufacturer’s Oncogene


4556 protocol. Additional clones were also used to assess consistency of results across different RCC4 cell line clones. DNase-1 (1 mg)-treated total RNA was used for cDNA synthesis. Reverse transcription was performed using 1.25 U/ml Multiscribe Transcriptase, 2.5 mM random hexamers, 500 mM per dNTP, 5.5 mM MgCl2 and 0.4 U/ml RNase inhibitor (Applied Biosystems). Gene-specific primers were designed using Primer3-Output software. As SYBR Green 1 is not specific and binds to primer dimers formed nonspecifically during all PCR reactions, preliminary experiments were carried out with each primer pair to determine the annealing temperature that yielded the greatest amount of specific product (Table 4). Relative standard curves were generated from dilutions of cDNA from the VHL wild-type RCC4 cell line. Dilutions (1 : 5, 1 : 10, 1 : 20, 1 : 50 and 1 : 100) of cDNA were used to construct a relative standard curve. Each 25 ml PCR mixture contained 12.5 ml SYBR Green PCR Master Mix (1 SYBR Green Buffer, 3 mM MgCl2, 200 nM dNTP mix, 0.6275 U Amplitaq Gold, 0.25 U AmpErase UNG), 2.5 ml cDNA, 25– 500 nM forward and reverse HPLC purified primers. Cycle conditions for the ABI 7900 HT were 501C for 2 min, 951C for 10 min and 40 cycles of 951C for 15 s and the specified annealing temperature for 1 min. Amplification was followed by melting curve analysis generated by heating the sample to 951C programmed for 10 s followed by cooling down to 601C for 15 s and slowly heating the samples at 11C/s to 951C while the fluorescence was measured continuously. Each reaction was run in triplicate, and a negative control without cDNA template was run with every assay to assess the overall specificity. Data were analysed with Microsoft Excel to generate raw expression values. Standard curves were generated from cDNAs of different known dilution factors. Threshold cycle (CT) was determined as the cycle number over background noise. The CT values were used to calculate and plot linear regression lines by plotting the logarithm of template concentration (X-axis) against the corresponding threshold cycle (Y-axis). The quality of the slope was judged from the correlation coefficient (r). The slope of the line was used to determine the efficiency of target amplification. Western analysis Protein was prepared from cell pellets. Cells were grown to B70% confluence, washed with ice-cold PBS and harvested by

Table 4 Gene

Plasminogen activatior inhibitor (PAI-1)a Claudin 4 E3 MCT3 DOC-2/DAB2 MLC2 GPR56 GLS GNG4 SPARC P311 CDKN1C

scraping. Cell pellets were homogenized in lysis buffer (150 mM NaCl, 50 mM Tris, 5 mM EDTA, 1% NP40 and protease inhibitors) and incubated on ice for 5 min. Lysates were centrifuged for 10 min at 13 000 r.p.m./41C and stored at 701C. Protein samples (20–40 mg) were separated on a 12% sodium dodecyl sulphate–polyacrylamide gel and electroblotted to transblot polyvinylidene difluoride membrane (Hybond-P; Amersham). Anti-CDKN1C (BD Pharmingen) and anti-MCT3 (Santa Cruz) were applied followed by secondary horseradish peroxidase conjugate, respectively, and visualization by the enhanced chemiluminescence detection system (Amersham Bioscience). b-Actin was used to show equal loading. VHL target gene expression in primary tumours Nine clear cell RCC tumour/normal pair mRNA samples were analysed for target gene expression by quantitative real-time PCR as described above. All tumours were clear cell RCC with VHL inactivation (VHL mutations or promoter methylation and allele loss). RNA was extracted using TriReagent (Sigma), according to the manufacturer’s protocol. The RNA was analysed by real-time quantitative PCR as described earlier. Approval for sample collection was obtained from the relevant Ethics committees. Plasmid constructs and growth suppression analysis The MLC2 and GNG4 expression constructs were made by cloning their full-length human coding regions into the EcoR1– BamHII sites of pCDNA3.1 vector (Invitrogen). In all, 1 mg of empty vector or expression vector was transfected using FuGENE 6 (Roche) according to the manufacturer’s instructions, into 5 104 SKRC45 cells growing in six-well plates. At 48 h after transfection, cells were seeded in a serial dilution and maintained in DMEM and 10% FCS supplemented with 50 mg/ ml G418 (Gibco). Surviving colonies were counted 14–21 days later, after staining with 0.4% crystal violet. ‘Wound healing’ was assayed by growing 5 104 SKRC45 cells stably expressing MLC2 and GNG4 or the empty vector in DMEM supplemented with10% FCS. After 48 h, when the cells were approaching confluence, the media was replaced with serum-free DMEM. Cells were maintained in serum-free media in each well of a six-well plate for 24 h. Wells containing cells expressing either MLC2 or GNG4 or the empty vector

Regulation of novel VHL target genes by hypoxia

Accession no.

X04429 AB000712 J02931 U81800 U53446 J02854 AJ011001 AB020645 U31382 J03040 U30521 U22398

Effect of VHL inactivation

m m m m k k k k k k k k

Effect of hypoxia on gene expression RCC4

EJ-28

HBL-100

m1.9 m2.4 m2.1 m2.1 k3.0 k2.0 k4.3 21.1 21.0 21.0 m4.4 m1.8

m1.7 — — — k4.5 — 21.0 — 21.4 — m1.74 —

m2.2 — — m23 — k2.2 21.2 — 2 0.9 — — m1.7

m ¼ expression upregulated by hypoxia (fold induction). k ¼ expression downregulated by hypoxia (fold induction). 2 ¼ expression unaffected by hypoxia (fold induction). — ¼ not expressed or not analysable because of very low expression. aGene already known to be hypoxia inducible for information – not for Publication Oncogene


4557 were then scratched with a 200 ml pipette tip and grown in serum-free medium for a further 24 h. The extent of ‘wound healing’ was observed microscopically. VHL target gene regulation by hypoxia Details of methods and cells used for hypoxic response studies have been reported previously (Zatyka et al., 2002). Real-time

RT–PCR was performed to determine whether the target genes were regulated by hypoxia. The data were normalized to bactin. Acknowledgements We thank BBSRC, Pfizer and Cancer Research UK for financial support.

References Agathanggelou A, Bieche I, Ahmed-Choudhury J, Nicke B, Dammann R, Baksh S, Gao B, Minna JD, Downward J, Maher ER and Latif F. (2003). Cancer Res., 63, 5344–5351. Baba M, Hirai S, Yamada-Okabe H, Hamada K, Tabuchi H, Kobayashi K, Kondo K, Yoshida M, Yamashita A, Kishida T, Nakaigawa N, Nagashima Y, Kubota Y, Yao M and Ohno S. (2003). Oncogene, 22, 2728–2738. Bornstein P and Sage EH. (2002). Curr. Opin. Cell Biol., 14, 608–616. Chlenski A, Liu S, Crawford SE, Volpert OV, DeVries GH, Evangelista A, Yang Q, Salwen HR, Farrer R, Bray J and Cohn SL. (2002). Cancer Res., 62, 7357–7363. Clifford SC, Cockman ME, Smallwood AC, Mole DR, Woodward ER, Maxwell PH, Ratcliffe PJ and Maher ER. (2001). Hum. Mol. Genet., 10, 1029–1038. Clifford SC, Prowse AH, Affara NA, Buys CH and Maher ER. (1998). Genes Chromosomes Cancer, 22, 200–209. Cockman ME, Masson N, Mole DR, Jaakkola P, Chang GW, Clifford SC, Maher ER, Pugh CW, Ratcliffe PJ and Maxwell PH. (2000). J. Biol. Chem., 275, 25733–25741. Contrino J, Hair G, Kreutzer DL and Rickles FR. (1996). Nat. Med., 2, 209–215. Edgington TS, Mackman N, Brand K and Ruf W. (1991). Thromb. Haemost., 66, 67–79. Foster K, Crossey PA, Cairns P, Hetherington JW, Richards FM, Jones MH, Bentley E, Affara NA, Ferguson-Smith MA and Maher ER. (1994). Br. J. Cancer, 69, 230–234. Furuse M, Hata M, Furuse K, Yoshida Y, Haratake A, Sugitani Y, Noda T, Kubo A and Tsukita S. (2002). J. Cell Biol., 156, 1099–1111. Gnarra JR, Tory K, Weng Y, Schmidt L, Wei MH, Li H, Latif F, Liu S, Chen F and Duh FM. (1994). Nat. Genet., 7, 85–90. Gnarra JR, Zhou S, Merrill MJ, Wagner JR, Krumm A, Papavassiliou E, Oldfield EH, Klausner RD and Linehan WM. (1996). Proc. Natl. Acad. Sci. USA, 93, 10589–10594. Guan M, Su B and Lu Y. (2002). Mol. Biotechnol., 20, 123–129. Hasselaar P and Sage EH. (1992). J. Cell Biochem., 49, 272–283. Hatada I, Ohashi H, Fukushima Y, Kaneko Y, Inoue M, Komoto Y, Okada A, Ohishi S, Nabetani A, Morisaki H, Nakayama M, Niikawa N and Mukai T. (1996). Nat. Genet., 14, 171–173. Hergovich A, Lisztwan J, Barry R, Ballschmieter P and Krek W. (2003). Nat. Cell Biol., 5, 64–70. Hocevar BA, Smine A, Xu XX and Howe PH. (2001). EMBO J., 20, 2789–2801. Horiuchi A, Imai T, Shimizu M, Oka K, Wang C, Nikaido T and Konishi I. (2002). Anticancer Res., 22, 2697–2702. Iliopoulos O, Levy AP, Jiang C, Kaelin Jr WG and Goldberg MA.. (1996). Proc. Natl. Acad. Sci. USA, 93, 10595–10599. Ivanov S, Liao SY, Ivanova A, Danilkovitch-Miagkova A, Tarasova N, Weirich G, Merrill MJ, Proescholdt MA, Oldfield EH, Lee J, Zavada J, Waheed A, Sly W, Lerman MI and Stanbridge EJ. (2001). Am. J. Pathol., 158, 905–919.

Iwai K, Yamanaka K, Kamura T, Minato N, Conaway RC, Conaway JW, Klausner RD and Pause A. (1999). Proc. Natl. Acad. Sci. USA, 96, 12436–12441. Jiang Y, Zhang W, Kondo K, Klco JM, St Martin TB, Dufault MR, Madden SL, Kaelin Jr WG and Nacht M. (2003). Mol Cancer Res., 1, 453–462. Kaelin Jr WG and Maher ER.. (1998). Trends Genet., 14, 423–426. Kalyanaraman S, Copeland NG, Gilbert DG, Jenkins NA and Gautam N. (1998). Genomics, 49, 147–151. Kikuchi T, Toyota M, Itoh F, Suzuki H, Obata T, Yamamoto H, Kakiuchi H, Kusano M, Issa JP, Tokino T and Imai K. (2002). Oncogene, 21, 2741–2749. Kleeff J, Huang Y, Mok SC, Zimmermann A, Friess H and Buchler MW. (2002). Dis. Colon Rectum, 45, 1242–1248. Kleuss C, Scherubl H, Hescheler J, Schultz G and Wittig B. (1993). Science, 259, 832–834. Knebelmann B, Ananth S, Cohen HT and Sukhatme VP. (1998). Cancer Res., 58, 226–231. Kondo K, Kim WY, Lechpammer M and Kaelin Jr WG. (2003). PLoS Biol., 1, 439–444. Kondo K, Klco J, Nakamura E, Lechpammer M and Kaelin Jr WG. (2002). Cancer Cell., 1, 237–246. Koochekpour S, Jeffers M, Wang PH, Gong C, Taylor GA, Roessler LM, Stearman R, Vasselli JR, Stetler-Stevenson WG, Kaelin Jr WG, Linehan WM, Klausner RD, Gnarra JR and Vande Woude GF. (1999). Mol. Cell. Biol., 19, 5902–5912. Kumar CC and Chang C. (1992). Cell Growth Differ., 3, 1–10. Kupprion C, Motamed K and Sage EH. (1998). J. Biol. Chem., 273, 29635–29640. Kuroda N, Toi M, Hiroi M and Enzan H. (2003). Histol. Histopathol., 18, 487–494. Lam WW, Hatada I, Ohishi S, Mukai T, Joyce JA, Cole TR, Donnai D, Reik W, Schofield PN and Maher ER. (1999). J. Med. Genet., 36, 518–523. Lane E and Sage EH. (1994). FASEB J., 8, 163–173. Levy AP, Levy NS and Goldberg MA. (1996). J. Biol. Chem., 271, 25492–25497. Li Y, Nagai H, Ohno T, Yuge M, Hatano S, Ito E, Mori N, Saito H and Kinoshita T. (2002). Blood, 100, 2572–2577. Lisztwan J, Imbert G, Wirbelauer C, Gstaiger M and Krek W. (1999). Genes Dev., 13, 1822–1833. Lonergan KM, Iliopoulos O, Ohh M, Kamura T, Conaway RC, Conaway JW and Kaelin Jr WG. (1998). Mol. Cell Biol., 18, 732–741. Mack FA, Rathmell WK, Arsham AM, Gnarra J, Keith B and Simon MC. (2003). Cancer Cell, 3, 75–88. Maher ER and Reik W. (2000). J. Clin. Invest, 105, 247–252. Maher ER, Yates JR, Harries R, Benjamin C, Harris R, Moore AT and Ferguson-Smith MA. (1990). Q. J. Med., 77, 1151–1163. Maranchie JK, Vasselli JR, Riss J, Bonifacino JS, Linehan WM and Klausner RD. (2002). Cancer Cell, 1, 247–255. Oncogene


4558 Maxwell PH, Wiesener MS, Chang GW, Clifford SC, Vaux EC, Cockman ME, Wykoff CC, Pugh CW, Maher ER and Ratcliffe PJ. (1999). Nature, 399, 271–275. Michl P, Buchholz M, Rolke M, Kunsch S, Lohr M, McClane B, Tsukita S, Leder G, Adler G and Gress TM. (2001). Gastroenterology, 121, 678–684. Mok SC, Wong KK, Chan RK, Lau CC, Tsao SW, Knapp RC and Berkowitz RS. (1994). Gynecol. Oncol., 52, 247–252. Morrissey C, Martinez A, Zatyka M, Agathanggelou A, Honorio S, Astuti D, Morgan NV, Moch H, Richards FM, Kishida T, Yao M, Schraml P, Latif F and Maher ER. (2001). Cancer Res., 61, 7277–7281. Ohh M, Park CW, Ivan M, Hoffman MA, Kim TY, Huang LE, Chau V and Kaelin Jr WG. (2000). Nat. Cell Biol., 2, 423–427. Ohh M, Yauch RL, Lonergan KM, Whaley JM, StemmerRachamimov AO, Louis DN, Gavin BJ, Kley N, Kaelin Jr WG, Illopoulos and O. (1998). Mol. Cell, 1, 959–968. Ohta S, Wada H, Nakazaki T, Maeda Y, Nobori T, Shiku H, Nakamura S, Nagakawa O, Furuya Y and Fuse H. (2002). Anticancer Res., 22, 2991–2996. Okuda H, Saitoh K, Hirai S, Iwai K, Takaki Y, Baba M, Minato N, Ohno S and Shuin T. (2001). J. Biol. Chem., 276, 43611–43617. Pal S, Claffey KP, Cohen HT and Mukhopadhyay D. (1998). J. Biol. Chem., 273, 26277–26280. Pause A, Lee S, Worrell RA, Chen DY, Burgess WH, Linehan WM and Klausner RD. (1997). Proc. Natl. Acad. Sci. USA, 94, 2156–2161.

Oncogene

Pause A, Lee S, Lonergan KM and Klausner RD. (1998). Proc. Natl. Acad. Sci. USA, 95, 993–998. Sage H, Vernon RB, Funk SE, Everitt EA and Angello J. (1989). J. Cell Biol., 109, 341–356. Schwahn DJ and Medina D. (1998). Oncogene, 17, 1173–1178. Staller P, Sulitkova J, Lisztwan J, Moch H, Oakeley EJ and Krek W. (2003). Nature, 425, 307–311. Tanimoto K, Makino Y, Pereira T and Poellinger L. (2000). EMBO J., 19, 4298–4309. Tseng CP, Ely BD, Pong RC, Wang Z, Zhou J and Hsieh JT. (1999). J. Biol. Chem., 274, 31981–31986. Turner A and McGivan JD. (2003). Biochem. J., 370 (Part 2), 403–408. Wang Z, Tseng CP, Pong RC, Chen H, McConnell JD, Navone N and Hsieh JT. (2002). J. Biol. Chem., 277, 12622–12631. Wykoff CC, Pugh CW, Maxwell PH, Harris AL and Ratcliffe PJ. (2000). Oncogene, 19, 6297–6305. Wykoff CC, Sotiriou C, Cockman ME, Ratcliffe PJ, Maxwell P, Liu E and Harris AL. (2004). Br. J. Cancer., 90 (6), 1235–1243. Yiu GK, Chan WY, Ng SW, Chan PS, Cheung KK, Berkowitz RS and Mok SC. (2001). Am. J. Pathol., 159, 609–622. Zatyka M, da Silva NF, Clifford SC, Morris MR, Wiesener MS, Eckardt KU, Houlston RS, Richards FM, Latif F and Maher ER. (2002). Cancer Res., 62, 3803–3811. Zimmer M, Doucette D, Siddiqui N and Iliopoulos O. (2004). Mol. Cancer Res., 2, 89–95.