| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Priority Reports |
Divisions of 1 Cancer Genomics and Proteomics and 2 Signaling Biology, Ontario Cancer Institute and Departments of 3 Computer Science and 4 Medical Biophysics, University of Toronto, Toronto, Ontario, Canada and 5 Terry Fox Laboratory, BC Cancer Research Centre, Vancouver, British Columbia, Canada
Requests for reprints: Linda Z. Penn, Division of Cancer Genomics and Proteomics, Ontario Cancer Institute, University of Toronto, 610 University Avenue, Room 9-628, Toronto, Ontario, Canada M5G 2M9. Phone: 416-946-2276; Fax: 416-946-2840; E-mail: lpenn{at}uhnres.utoronto.ca.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
Myc can potentiate apoptosis by activating p53 and evidence shows that abrogation of the p53 pathway is a key event in tumorigenesis (3). In many tumors harboring deregulated Myc, however, cells are surprisingly resistant to apoptosis even without p53 mutations (4). These findings suggest that there must be oncogenes that can inhibit p53 and cooperate with Myc to promote transformation. Here, we report CUL7 as a novel oncogene that cooperates with Myc to promote transformation.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
Constructs. The plasmid pSK-KIAA0076 was obtained from the Kazusa DNA Research Institute (Chiba, Japan). The KIAA0076 insert (CUL7) was cloned into pBabeMNiresGFP retroviral vector at EcoRI/XhoI and into pBabepuro at EcoRI/SalI. pBabeMNiresGFP and pBabepuro retroviral constructs were modified by ligation of a FLAG tag at the NH2 terminus of CUL7. pBabeMNiresGFP-Hu-cMyc was described previously (5). To generate MSCV-YFP-p53DD, p53DD insert was first obtained by digesting pSPp53DD (kind gift of Dr. Moshe Oren, Weizman Institute of Science, Rehovot, Israel) at BamH1/Sma1 site and then cloned into BglII/HpeI sites of pMSCV-YFP.
Retroviral production and infection. Infectious replication-deficient ecotropic retroviral particles were produced in Phoenix Eco cells and used to infect Rat1, Rat1a, and SHEP N-Myc cells expressing ecotropic receptor (5).
Functional screen. Rat1MycERtmv-myc cells were either mock infected or infected with a retroviral library of cDNA from the T47D breast carcinoma cell line in the retroviral vector pCTV1 (6). Apoptotic conditions were maintained by changing the medium to 0.06% FBS and adding 0.1 µmol/L hydroxytamoxifen to activate MycERtm. The surviving cells were then replated and subjected to a second round of Myc-induced apoptosis. Subsequently, surviving colonies were picked and the retrovirally transduced cDNA was isolated from these cells by PCR of genomic DNA using primers specific for the provirus as described (6).
Immunoblot analysis. 
-CUL7 antibodies were raised against a synthetic COOH-terminal peptide RGVPYASCTATQSFSTFR (human CUL7 amino acids 1,681–1,698) in rabbits. Immunoblots were probed with the COOH-terminal–specific CUL7 antibody or p53 antibody (DO-1; Santa Cruz Biotechnology). p53DD was immunoblotted with p53 antibody (Pab421) and c-Myc was probed with 9E10 antibody. Bcl2 was probed with polyclonal anti-Bcl2 antibody (kind gift of Dr. David Andrews, Department of Biochemistry, McMaster University, Hamilton, Ontario, Canada).
Cell death assay. Rat1MycERtmv-myc (106) cells were seeded in duplicate. After 24 h, medium was changed to 0.06% FBS and 0.1 µmol/L hydroxytamoxifen. Cells were harvested at 72 h, stained with propidium iodide, and analyzed by flow cytometry. For the CUL7 RNA interference (RNAi) experiments, SHEP Tet21/N-Myc MSCV-YFP and SHEP Tet21/N-Myc MSCV-YFP-p53DD cells were first infected with lentivirus expressing one of pLKO-Puro-luciferase small interfering RNA (siRNA), pLKO-Puro-CUL7 siRNA #2, or pLKO-Puro-CUL7 siRNA #5. One day after infection, the cells were selected with puromycin at 1 µg/mL concentration for 2 days. Cells were then seeded in triplicate on tissue culture plates, and 3 days after the seeding, both floating and adherent cells in triplicate tissue culture plates were harvested and analyzed by propidium iodide staining and flow cytometry. Cell death (% pre-G1) was assessed by measurement of subdiploid DNA content.
Colony growth in soft agar. Anchorage-independent growth assays were conducted as described for Rat1a cells (7). The number of Rat1a colonies measuring >20 cells was assessed 3 weeks later. For SHEP Tet21/N-Myc cells, 104 cells were seeded in 10% FBS-RPMI 1640 + 0.3% noble agar in 35-mm Petri dishes and counted at the end of a 2-week period.
Coimmunoprecipitation. U2OS cells were exposed to 0.5 µmol/L etoposide for 12 h and then harvested using immunoprecipitation lysis buffer (8). Extracts were immunoprecipitated with antibodies against CUL7 or p53 (DO-1). Immune complexes were collected with protein G-agarose beads and washed four times with wash buffer containing 5% sucrose, 5 mmol/L Tris-HCl (pH 7.4), 5 mmol/L EDTA, 0.5 mol/L NaCl, and 0.7% NP40.
CUL7 knockdown with RNAi. Oligos for CUL7 RNAi #2 were 5'-TGAGATCCTAGCTGAACTG-3' (forward) and 5'-CAGTTAAGCTAGGATCTCA-3' (reverse). Oligos for CUL7 RNAi #5 were 5'-AGAACTCCGCTACAGGGAATT-3' (forward) and 5'-AATTCCCTGTAGCGGAGTTCT-3' (reverse). The control RNAi was against luciferase (9). Oligos were annealed and ligated into pLKO.1-puro, and together with pMD.G, pMDLg/pRRE and pRsv-Rev were transfected into 293TV cells to produce lentiviruses (10).
Purification of glutathione S-transferase-p53. Full-length p53 was cloned into GST2TK (Amersham) and purified as described previously (11).
In vitro translation and glutathione S-transferase pull-down assay. 35S-labeled CUL7 proteins were prepared in reticulocyte lysate using the TNT system (Promega). Equal amounts of glutathione S-transferase (GST) and GST-p53 (10 µg) immobilized on glutathione-Sepharose beads (GE) were incubated with labeled in vitro translated CUL7 proteins for 2 h at 4°C in the binding buffer [25 mmol/L Tris (pH 8.0), 150 mmol/L NaCl, 1 mmol/L EDTA, 1 mmol/L DTT, 10% glycerol, 0.2% NP40, 1 mmol/L phenylmethylsulfonyl fluoride]. The beads were washed extensively with the same buffer. The bound proteins were released with SDS sample buffer, separated by SDS-PAGE, and detected by autoradiography.
Generation of GST-p53 fusion proteins and use in CUL7(1-375) pull-down assays. Purified 1-70, 311-393, and 82-360 GST-fused p53 proteins were incubated with His-tagged CUL7(1-375) and then with the GST beads at 4°C for 1 h. After extensive washing with assay buffer, bound proteins were eluted with 20 mmol/L reduced glutathione and detected after SDS-PAGE by Coomassie Brilliant Blue.
Bioinformatics. The unprocessed data from microarray studies of tumor-normal differential expression in multiple sites, including lung (12, 13), prostate (14, 15), glioblastoma (16), and pancreas (14), were acquired. Following robust multichip analysis (RMA) preprocessing (17), we assessed differential CUL7 expression by comparing mean CUL7 mRNA levels in tumor samples with those in normal tissues separately for each study using a two-tailed t test with Welch's correction for unequal variances. Significant elevation (P < 0.001) of CUL7 mRNA was found in all studies of non–small cell lung cancers (12, 13). To determine if patient outcome is also related to CUL7 expression, we retrieved the unprocessed data for a large study of biomarkers in non–small cell lung cancer (18). Following RMA preprocessing (17), patients were median dichotomized according to CUL7 mRNA levels. The linear effect of CUL7 on survival was calculated using Cox proportional hazards modeling and the Wald test. All analysis was done in the R statistical environment (version 2.3.1) using the affy (version 1.10.0) and survival (version 2.29) packages.
|
Results and Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
2.7 million cDNAs) derived from a human breast carcinoma cell line, T47D. Two independent alleles of Myc were expressed in these cells to maximize apoptosis and reduce the outgrowth of false-positive clones. MycERtm was activated with the addition of 4-hydroxytamoxifen to potentiate serum withdrawal-induced apoptosis, and the surviving cells were allowed to recover before being replated and subjected to a second round of selection. Cells that survived the second treatment were expanded and the cDNAs were recovered by PCR and tested for apoptosis-suppressing activity. From this screen, we identified and characterized three positive hits, including the well-established antiapoptotic Bcl2 and CUL7, a relatively new and unusually large member of the cullin family of proteins. Cullins, including CUL7, function as scaffold proteins to form either a Skp1-cullin-F-box (SCF) or SCF-like complex, which is involved in the process of protein ubiquitylation (8).
|
Interestingly, results of our functional screen suggest a novel biological role for CUL7 in the etiology of cancer as an antiapoptotic oncogene that can cooperate with Myc in transformation. To further explore this hypothesis, it was important to first validate the screen and determine whether CUL7 can inhibit Myc-potentiated apoptosis in the cells used in the original screen. Rat1MycERtmv-myc cells were infected with replication-incompetent control retrovirus and retrovirus carrying FLAG-CUL7 or Bcl2. The cells were then subjected to apoptosis induced by serum depletion and Myc activation, and the percentage of pre-G1 cells (cell death) was determined by flow cytometry and propidium iodide staining. Like Bcl2, CUL7 was able to significantly inhibit Myc-potentiated apoptosis (Fig. 1B) when overexpressed (Fig. 1C).
CUL7 blocks both c-Myc–potentiated and N-Myc–potentiated apoptosis. To evaluate the functional role of CUL7 in independent cell systems, CUL7 or Bcl2 was first ectopically introduced into Rat1a cells expressing either control empty vector [green fluorescent protein (GFP) control] or ectopic c-Myc (Myc control; Fig. 2A and B ). Rat1a cells are a transformed clone of Rat1 cells that are able to grow in an anchorage-independent manner when Myc is ectopically expressed in these cells (25). Rat1a cells expressing either control empty vector or ectopic c-Myc were subjected to serum withdrawal-induced apoptosis. As shown in Fig. 2A and B, cells expressing either ectopic CUL7 or Bcl2 were significantly more resistant to c-Myc–potentiated apoptosis compared with the Myc alone control (Myc control).
|
Because Myc-induced apoptosis is broadly categorized as being p53 dependent or independent, we investigated which of these two pathways CUL7 targets. To determine whether p53 was essential for CUL7 to regulate apoptosis, we exogenously expressed control empty vector (MSCV-YFP) or a COOH-terminal fragment of p53 (MSCV-YFP-p53DD) into SHEP N-Myc neuroblastoma cells. Ectopic expression of p53DD, which encodes the COOH-terminal region (amino acids 302–390) of p53, binds to endogenous, wild-type, full-length p53 and significantly reduces p53 activity. To confirm that p53DD effectively functioned as a p53 dominant negative, we treated both SHEP YFP and SHEP p53DD YFP cells with doxorubicin for 24 h. As shown in Supplementary Fig. S1, expression of p53DD completely inhibited p53 from inducing one of its target genes, PUMA. SHEP YFP and SHEP p53DD YFP cells were then subsequently transduced with two independent CUL7 siRNAs to knock down endogenous CUL7 expression at the level of both protein (Fig. 2D, lanes 2, 3, 5, and 6) and mRNA (Supplementary Fig. S2A). Interestingly, p53 expression was consistently up-regulated in response to the knockdown of CUL7 (Fig. 2D, compare lanes 1–3). As mentioned above, SHEP N-Myc cells underwent significant cell death when CUL7 expression was knocked down (Fig. 2C, columns 1–3); however, when these cells expressed p53DD (Fig. 2D, lanes 4–6), the degree of apoptosis was decreased to basal levels (Fig. 2C, columns 4–6). Similar results were obtained when the experiment was repeated with terminal deoxynucleotidyl transferase–mediated dUTP nick end labeling staining as an alternative apoptosis assay (Supplementary Fig. S3). Furthermore, knockdown of CUL7 in ectopic Myc-expressing HCT116 cells (Supplementary Fig. S2B) promoted p53-dependent apoptosis triggered by 5-fluorouracil (Supplementary Fig. S4). Together, these data show that the antiapoptotic effect of CUL7 is p53 dependent.
CUL7 interacts with p53 both in vivo and in vitro. Because our data link CUL7 to p53-mediated apoptosis, we next investigated if CUL7 interacts with endogenous wild-type p53 in vivo. The U2OS osteosarcoma cells are often used to study endogenous wild-type p53 because low basal p53 levels can be induced in response to etoposide. Thus, coimmunoprecipitation assays were conducted in etoposide-treated U2OS cells using either control preimmune serum or CUL7-specific antibody and immunoprecipitated proteins were then resolved by SDS-PAGE and immunoblotted using an anti-p53 (DO-1) antibody. Endogenous p53 was coimmunoprecipitated using the anti-CUL7 antibody (Fig. 3A, left ), suggesting that endogenous CUL7 interacts with endogenous p53. The p53:CUL7 interaction was confirmed by reciprocal coimmunoprecipitation, which showed that CUL7 was coimmunoprecipitated using an anti-p53 antibody but not with an anti-GFP antibody (Fig. 3A, right).
|
To determine whether the physical interaction between CUL7 and p53 is direct, a His-tagged CUL7 fragment (residues 1–375) and three GST-tagged p53 fragments (1-70, 82-360, and 311-393) were expressed in bacteria and purified. As shown in Fig. 3C, when His-CUL7(1-375) was coincubated with each GST-p53 fragment and subjected to GST pull-down assays, His-CUL7(1-375) strongly bound both GST-p53(82-360) and GST-p53(311-393) but not GST-p53(1-70). These data show that CUL7 can bind to p53 directly in the absence of other proteins, such as the Skp1 or F-box proteins, which are required for SCF complex formation. Thus, CUL7 directly binds (Fig. 3C) and regulates p53 expression (Fig. 2D) by a mechanism that remains unclear (21).
CUL7 cooperates with Myc in transformation. Having shown that CUL7 is able to inhibit apoptosis induced by both c-Myc and N-Myc in a p53-dependent manner, we next tested the hypothesis that CUL7 functions as an oncogene and collaborates with Myc in transformation. Anchorage-independent growth was assessed in cells with and without ectopic expression of CUL7. Rat1a cells were first infected with control retrovirus or retrovirus containing c-Myc, and these two cell lines were then further infected with either control retrovirus or retroviruses containing FLAG-CUL7 or Bcl2. These Rat1a cells were then seeded in soft agar for 3 weeks under either growth-promoting (10% serum) or apoptotic conditions (2% serum; ref. 7). Under growth-promoting conditions, the colony number was elevated in response to ectopic Myc expression as expected, and ectopic expression of CUL7 or Bcl2 in addition to ectopic expression of c-Myc did not result in increase in colony numbers (data not shown). By contrast, under apoptotic conditions, Rat1a cells coexpressing ectopic c-Myc with Bcl2 or CUL7 formed a significantly greater number of colonies compared with those cells expressing ectopic c-Myc alone (Fig. 4A ). These data show that CUL7 cooperates with c-Myc to drive transformation specifically under apoptotic conditions, which is consistent with the antiapoptotic properties of CUL7 described here (Figs. 1B and 2A and C).
|
As a negative regulator of apoptosis, CUL7 may be deregulated in human cancers. To test this hypothesis, we used in silico analysis of multiple public microarray studies (12–16). Interestingly, CUL7 mRNA was significantly elevated relative to normal in two studies (12, 13) of non–small cell lung cancer but not in prostate (14, 15), glioblastoma (16), and pancreas (Fig. 4C; ref. 14). This suggests that overexpression of CUL7 mRNA confers a survival advantage to tumors in a tissue-specific manner. Indeed, our data are consistent with the finding that the most profound effect of CUL7 knockout was on lung development and that this resulted in postnatal death (19). Because CUL7 is significantly overexpressed in lung tumors compared with normal lung, this raised the possibility that CUL7 expression is also associated with disease progression. To evaluate this possibility, we used a recent microarray study of prognostic markers in non–small cell lung cancers (18). We split the patient population into two groups based on their CUL7 expression levels and found that patients with elevated CUL7 mRNA levels showed significantly (P = 0.02) reduced survival (Fig. 4D). Specifically, a 2-fold increase in CUL7 expression confers a 2.8-fold increase in hazard for the patient. Similar results were observed with both alternative CUL7 ProbeSets and a second array data set (12). These primary tumor data (Fig. 4C and D) provide strong evidence that CUL7 expression is deregulated in primary lung cancers and that this deregulation is associated with poor patient outcome.
Over the last 5 years since CUL7 was first described (8), it has been shown that CUL7 can function as a regulator of cell proliferation. CUL7 knockout MEFs displayed a slow growth phenotype (19), overexpression of CUL7 accelerated cell proliferation in a p53-dependent manner (21), and the knockdown of CUL7 induced p53-dependent growth arrest (28). Moreover, 3M syndrome, a human disease caused by intrauterine growth retardation, is linked to various CUL7 mutations (20).
Our study clearly indicates that CUL7 is an oncogene that can play a role in the etiology of human cancer. Unlike the role of CUL7 in cell growth, the role of CUL7 in apoptosis has remained largely unaddressed. Although it has been reported that expression of mouse CUL7 triggers apoptosis when ectopically expressed in NIH 3T3 cells (22), we have shown that ectopic expression of human CUL7 in rat cells protected them from Myc-potentiated apoptosis. We have also shown that knockdown of human CUL7 triggered p53-dependent apoptosis in several cell lines derived from a variety of human cancers, including neuroblastoma and colon. Interestingly, the role of CUL7 in the control of cell growth and death seems functionally linked to p53 (24).
The mechanism by which CUL7 regulates p53 remains unclear; however, two working models under consideration propose that CUL7 controls p53 ubiquitylation and/or subcellular localization. CUL7 can function as an E3 ubiquitin ligase (8, 20), suggesting that CUL7 may negatively regulate p53 by ubiquitylating and targeting it for degradation. This model is consistent with a recent report indicating that DNA damage can induce CUL7 expression leading to a decrease in p53 expression (28). However, two independent groups showed that overexpression of CUL7 and p53 in vivo did not result in an increase of ubiquitylated p53 (21, 28). By contrast, in vitro ubiquitylation experiments by Andrews et al. (21) showed that CUL7 can monoubiquitylate or diubiquitylate p53. The second mechanistic model under consideration is based on the amino acid similarity of CUL7 with PARC, a protein reported to inhibit p53 activity by anchoring it in the cytoplasm (29). In particular, these proteins are founding members of the new CPH structural domain shown to be important for direct interaction with p53 (Fig. 3; ref. 27). However, Andrews et al. showed that coexpression of CUL7 together with p53 did not significantly alter p53 localization, suggesting that CUL7 does not inhibit p53 activity in a similar manner as PARC. Taken together, it is not yet clear how CUL7 regulates the p53 tumor suppressor.
In this article, we not only show that CUL7 can bind directly to p53 and protect cells from p53-dependent apoptosis but that CUL7 can cooperate with both c-Myc and N-Myc to promote anchorage-independent growth in multiple cell systems. This is an interesting finding as it suggests that CUL7 may play an important role in the transformation of normal cells to tumor cells. Indeed, we show that CUL7 is also overexpressed in human lung cancers and that increased expression is correlated with poor patient outcome. Other proteins deregulated in lung cancer, such as EGFR, have been specifically targeted pharmacologically (i.e., gefitinib), resulting in improved clinical outcome (30). Because our data indicate that CUL7 is also highly deregulated in lung tumors and associated with poor patient outcome, pharmacologic inhibition of CUL7 may be a valuable strategy for treating lung cancers, as well as other human cancers, with deregulated CUL7 expression.
|
Acknowledgments |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
We thank Dr. Moshe Oren for the p53DD plasmid, M. Schwab and S. Hann for providing us with SHEP Tet21/N-Myc and Rat1a cell lines, Dr. David Andrews for anti-Bcl2 antibody, and our colleagues in the Penn laboratory for helpful discussions.
|
Footnotes |
|---|
Current address for M. Shago: Department of Paediatric Laboratory Medicine, The Hospital for Sick Children, 555 University Avenue, Toronto, Ontario, Canada M5G 1X8.
Received 2/15/07. Revised 7/ 4/07. Accepted 8/31/07.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  X. Zhao, R. J. Madden-Fuentes, B. X. Lou, J. M. Pipas, J. Gerhardt, C. J. Rigell, and E. Fanning Ataxia Telangiectasia-Mutated Damage-Signaling Kinase- and Proteasome-Dependent Destruction of Mre11-Rad50-Nbs1 Subunits in Simian Virus 40-Infected Primate Cells J. Virol., June 1, 2008; 82(11): 5316 - 5328. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
