ª
Oncogene (2002) 21, 5940 – 5945 2002 Nature Publishing Group All rights reserved 0950 – 9232/02 $25.00 www.nature.com/onc
The DNA repair protein, O6-Methylguanine-DNA methyltransferase is a proteolytic target for the E6 human papillomavirus oncoprotein Kalkunte S Srivenugopal*,1 and Francis Ali-Osman1 1
Section of Molecular Therapeutics, Department of Neurosurgery, The University of Texas MD Anderson Cancer Center, Houston, Texas, TX 77030-4009, USA
We have previously shown that O6-methylguanine-DNA methyltransferase (MGMT), a DNA repair protein that protects tissues against toxic and carcinogenic effects of alkylating agents, is degraded through ubiquitinationdependent proteolysis. Here, we investigated the role of the human papillomavirus (HPV) E6 protein in MGMT degradation. In three pairs of isogenic human tumor cell lines in which a member of each pair expressed the E6 protein through stable transfection (HCT116/HCT116E6, MCF7/MCF7-E6, and RKO/RKO-E6), we found a consistent 40 – 55% reduction in the MGMT protein level and its activity in all E6-expressing cells compared with the parent cells (P=50.05). E6 expression did not, however, alter the levels of MGMT mRNA. Addition of the recombinant MGMT (rMGMT) protein to extracts of HCT116/E6 cells resulted in the binding of E6 to MGMT. Further, the purified E6 protein promoted the degradation of rMGMT in rabbit reticulocyte lysates. Immunoprecipitation assays showed the presence of a ternary protein complex between MGMT, E6, and the cellular ubiquitin-ligase E6-associated protein (E6-AP). Transient transfection of the p53-null H1299 lung tumor cells with an E6 construct also down-regulated the MGMT. The MGMT protein also showed structural features that are compatible for interaction with the E6, and E6-AP components. Collectively, these data suggest that the oncogenic E6 proteins enhance the ubiquitindependent proteolysis of MGMT. Oncogene (2002) 21, 5940 – 5945. doi:10.1038/sj.onc. 1205762 Keywords: MGMT; DNA repair; ubiquitin; proteasome; alkylating agents
Human papillomavirus (HPV), and in particular, the high-risk types HPV16 and HPV18 are etiologically linked to nearly 90% of cervical cancers (Bosch et al., 1995; zur Hausen, 1999; Rapp and Chen, 1998) and 20 – 30% of head and neck cancers (Key et al., 1999).
*Correspondence: KS Srivenugopal, Department of Neurosurgery, Box 64, UT MD Anderson Cancer Center, Houston, Texas, TX 77030, USA; E-mail:
[email protected] Received 26 February 2002; revised 5 June 2002; accepted 14 June 2002
The transforming properties of high-risk HPVs reside primarily in two genes, E6 and E7, which are consistently expressed in HPV-positive cervical cancers and derived cell lines (Schwarz et al., 1985; Baker et al., 1987). E6 and E7 contribute to the oncogenic process by inactivating tumor suppressor proteins, mainly, by targeting them for degradation through the ubiquitin-proteasome pathway (Scheffner, 1998). The ubiquitin pathway normally regulates physiological levels of the retinoblastoma (pRb) and p53 proteins (Wang et al., 2001; Maki et al., 1996). What the HPV oncoproteins do is to enhance the degradation of these proteins through the same pathway by accelerating the ubiquitin conjunction step through a redirection of the substrate specificity (Scheffner et al., 1994; Talis et al., 1998). While E7 binds to pRb and facilitates its destruction (Boyer et al., 1996), the oncogenic E6 proteins specifically stimulate the degradation of p53 (Scheffner et al., 1990, 1994). In this process, E6 recruits and utilizes the cellular ubiquitin-protein ligase called E6-AP (E6 Associated Protein) for introducing the polyubiquitin tag on the p53 protein, which is subsequently degraded into peptides by the 26S proteasome (Huibregtse et al., 1991; Talis et al., 1998; Scheffner, 1998). It appears that E6, by itself, does not bind to p53 significantly, but rather functions as an adapter or auxiliary protein that facilitates p53 binding to E6-AP (Scheffner et al., 1994). Recent studies have shown that the HPV E6 proteins also promote the E6-AP-dependent degradation of many other proteins, such as c-myc (Gross-Mesilaty et al., 1998); Bak (Thomas and Banks, 1998); hDlg, the human homolog of the Drosophila melanogaster tumor suppressor Discs large (Dlg) (Gardiol et al., 1999); a novel GAP protein called E6TP1 (Gao et al., 1999); and the human scribble (vartul) protein (Nakagawa and Huibregtse, 2000). These findings suggest that besides the depletion of p53 caused by E6, the degradation of other proteins may also contribute to the HPV-induced transformation and carcinogenesis. Pertinent in this context is the DNA repair performed by O6-methylguanine-DNA methyltransferase (MGMT, EC 2.1.1.63, also called O6-alkylguanineDNA alkyltransferase) which protects the cellular genome from the mutagenic actions of endogenous carcinogens and clinically active alkylating agents (Mitra and Kaina, 1993; Pegg, 2000). MGMT functions by a stoichiometric and suicidal reaction
E6-mediated degradation of MGMT KS Srivenugopal and F Ali-Osman
5941
mechanism in which the alkyl groups bound to the O6 position of guanine are transferred to a cysteine in its active site; this reaction results in the direct restoration of the normal guanine structure and self-inactivation of the MGMT (Pegg, 2000). MGMT has been shown to prevent mutations in a number of target genes including the p53 gene (Wolf et al., 2001; Hussain et al., 1994), and the ras oncogene (Zaidi et al., 1995; Esteller et al., 2000). MGMT is also highly expressed in human cancers, and this bestows resistance of tumors to many anticancer alkylating agents, because the methylguanine and O6-chloroethylguanine lesions induced by methylating agents (temozolomide, dacarbazine, and procarbazine) and the 2-chloroethyl nitrosoureas (e.g., 1,3 bis-(2-chloroethyl)-1-nitrosourea; BCNU), are excellent substrates for MGMT (Gerson and Willson, 1995). Consequently, the inhibition of MGMT by pseudosubstrates, such as, O6-benzylguanine (BG), which inactivates and effectively depletes cellular MGMT, has emerged as a promising strategy to enhance the cytotoxicity of alklyating agents (Dolan and Pegg, 1997). In previous studies, we showed that, after inactivation by O6-benzylguanine and BCNU, MGMT becomes a target for polyubiquitination and subsequent degradation by the proteasome (Srivenugopal et al., 1996). Our studies have also suggested that the cellular turnover of the MGMT protein occurs through the ubiquitin proteolytic pathway (Ali-Osman et al., 2001), although the specific ubiquitin conjugases
involved in this process have not been identified. Our recent study on MGMT gene expression in p53proficient and -deficient tumor cells (Srivenugopal et al., 2001), did indicate that HPV E6 may control MGMT protein levels. We showed significantly higher MGMT levels in the E6-expressing H460 lung cancer cells compared to the non-E6 parent cells (Srivenugopal et al., 2001), however, MGMT expression in other E6 expressing cell lines was found to be consistently lower, and this provided the basis for the current investigation. As a first step, we quantitated MGMT expression in four human cancer cell lines that had been stably transfected with the E6 oncogene encoded by HPV16 or HPV18. The cell lines used, HCT116, RKO, PA1, and MCF7, all possess wild-type p53 (Beroud and Soussi, 1998), and E6 expression has been shown to induce a p53-deficient state in them. As a result, these pairs of cell lines have been used for investigating p53 function in an isogenic setting. Figure 1a shows the levels of MGMT activity in these cells. HCT116 had the highest MGMT levels, with the other three cell lines showing lower, but similar, levels. E6 expression resulted in a consistent and significant reduction in MGMT activity in all the cell lines (P50.05), ranging from threefold decrease in the HCT116/E6 cells to about a 2.5-fold decrease in the other E6 expressing cells (Figure 1a). Consistent with this reduced activity, and the stoichiometric mechanism of MGMT in which one protein molecule consumes one O6-alkylguanine
b a
c
Figure 1 The DNA repair activity of MGMT and its protein levels are markedly reduced in HPV E6 transfected human tumor cell lines. Stable cell lines expressing the HPV E6 oncogene and their untransfected counterparts were obtained as follows: HCT116/E6 and RKO/E6 colon cancer cells from Dr Michael Kastan (St Jude Hospital, Memphis, TN, USA); PA1/E6 ovarian cancer cells (Wu et al., 1998) from Dr Wafik el-Deiry (University of Pennsylvania, Philadelphia, PA, USA); and MCF7/E6 breast cancer cells (Gupta et al., 1997) from Dr Albert Fornace (National Institutes of Health, Bethesda, MD, USA). Cells were cultured in DMEM with 10% serum and harvested in their exponential growth phase for experiments. (a) Quantitation of MGMT activity. Transfer of 3H-labeled methyl groups from the O6 position of guanine in DNA to the MGMT protein was measured as described previously (Srivenugopal et al., 2000). Cell extracts (50 – 200 mg of protein) prepared in MGMT assay buffer were supplemented with [3H] DNA enriched for O6-methylguanine (3 mg; 10 000 c.p.m.), incubated for 30 min at 378C. The reactions were quantitated after acid hydrolysis of the DNA substrate and counting of the protein bound radioactivity. The mean of specific activity (pmoles 3H transferred/mg protein) obtained in four independent experiments in each cell line was computed, and rendered relative to each other. (b) Western analysis of MGMT protein in parent and E6-expressing cell lines. The immunoblots were probed sequentially with monoclonal antibodies to MGMT and actin proteins. (c) Northern blot analysis of MGMT. Fifteen mg of total RNA from each cell line was electrophoresed and the blots were hybridized with 32P-labeled full-length cDNA probes for MGMT and actin (Srivenugopal et al., 2001) Oncogene
E6-mediated degradation of MGMT KS Srivenugopal and F Ali-Osman
5942
lesion, MGMT protein levels in E6-transfected cells were markedly lower (Figure 1b). There was no difference in the steady-state mRNA levels for MGMT between parent and E6-expressing cell lines (Figure 1c), indicating that altered transcription of the MGMT gene did not underlie the decrease in MGMT protein. Therefore, the case of H460-E6 cells, which showed higher levels of MGMT than the parent controls (Srivenugopal et al., 2001) appears an exception to the general rule; the reasons for this discrepancy are unclear. However, it is possible that MGMT may be subjected to enhanced transcriptional and post-translational (e.g. a reversible phosphorylation) regulation in H460/E6 cells. Moreover, unexpected findings have been reported in HPV-positive cancer cells previously; these include the induction of Gadd45 (a p53-targeted gene) by ionizing radiation (Butz et al., 1999), and the resistance of p53 sequestered in the cytoplasm to E6dependent proteolysis (Isaacs et al., 1999). Direct evidence for the involvement of E6 in the downregulation of MGMT was sought by examining the degradation of 14C-labeled recombinant MGMT protein in rabbit reticulocyte lysates in the presence or absence of purified E6-GST fusion protein. The reticulocyte lysates are enriched in ubiquitin-conjugated enzymes including E6-AP and proteasome and have been widely used to investigate protein degradation in vitro (Wajnberg and Fagan, 1989; Gross-Mesilaty et al., 1998). Figure 2 shows that the 14C-MGMT disappeared in a timedependent fashion in this proteolytic system (lanes 2 and 3), and that the addition of the E6 protein greatly enhanced the rate of MGMT degradation (lanes 4 – 7). It has been shown previously that E6 protein does not significantly interact physically with the proteins that are targeted for degradation. Instead, E6 binds to E6-AP, the resulting complex then binds the target protein (Huibregtse et al., 1991; Talis et al., 1998; Scheffner, 1998) and catalyzes the conjugation of ubiquitin chains.
Figure 2 In vitro degradation of recombinant MGMT is stimulated by E6 protein. Proteolysis of human MGMT protein was examined in rabbit reticulocyte lysates with or without the purified HPV18 E6 protein. The GST-E6 protein was a kind gift from Dr M Pawelita of Deutsches Krebsforschungszentrum, Heidelberg, Germany (Sehr et al., 2001). The reaction mixtures containing 20 ml of rabbit reticulocyte lysate (Promega, Madison, WI, USA) and 14C-labeled human recombinant MGMT protein expressed and purified from E. coli (1.5 mg; 5000 c.p.m.) (Srivenugopal et al., 2001) were incubated with or without GST-E6 protein (0.2 mg) at 338C for times shown. The reactions were stopped with the addition of SDS – PAGE sample buffer, after which the samples were boiled for 5 min and subjected to electrophoresis on 12% gels. The gels were dried and autoradiographed. Degradation assays were also performed in the presence of GST alone (as a control for GST-E6), and this did not alter the proteolytic pattern of 14C-MGMT (not shown). Ret. Lysate, reticulocyte lysate Oncogene
To examine the interaction of E6 and E6-AP with MGMT, we performed a novel pull-down assay as well as an immunoprecipitation study. In the pull-down assay, a histidine-tagged recombinant MGMT (rMGMT) protein was incubated with the crude extracts from HCT116 and HCT116/E6 cells, followed by selective purification of the rMGMT and interacting proteins on a metal chelator resin (Srivenugopal et al., 2002) and Western analysis. As shown in Figure 3a, a 16kDa band corresponding to the E6 protein was specifically associated with rMGMT in E6-expressing cells, presumably through the E6-AP present in the cell extracts. To further identify the components interacting with MGMT, we immunoprecipitated the MGMT protein from HCT116 and HCT116/E6 cells, resolved the complexes by SDS – PAGE, and performed Western blot analysis for the MGMT, E6, and E6-AP proteins. Both the E6 and E6-AP proteins were detected in MGMT immunocomplexes isolated from HCT116/E6 cells (Figure 3b), a feature consistent with the known mechanisms of the HPV E6-promoter proteolysis. Additional evidence for the involvement of the E6-AP in MGMT degradation was obtained by immunodepletion experiments. Extracts from HCT116 and HCT116/ E6 cells were incubated with monoclonal antibodies to E6-AP to deplete the ubiquitin ligase activity associated with the protein. The resulting extracts were tested for their ability to degrade the 14C-MGMT protein in the presence of Mg2+, ATP and ubiquitin, which are all required for efficient sub-conjugation and proteasomal reactions (Skhedy et al., 1994; Ali-Osman et al., 2001). Immunodepletion of E6-AP in HCT116/E6 cell extracts resulted in a 40% decrease in the proteolysis of MGMT, while the depleted extracts from HCT116 parent cells did not differ significantly in their ability to degrade the MGMT protein (Figure 3c). The p53 tumor suppressor is the most extensively studied target of E6-driven ubiquitination-dependent proteolysis, and the human tumor cell lines used in this study (Figure 1) have served as experimental models in various p53-related investigations. Consequently, the question of whether the decrease in MGMT we observed in E6-expressing cells is mechanistically linked to the up-regulated proteolysis of p53 is very relevant. For example, down-regulation of b-catenin in p53-overexpressed cells (Sadot et al., 2001) and enhanced levels of p53 in b-catenin-overexpressed cells (Damalas et al., 1999) has been attributed to an interference in the ubiquitin-mediated proteolysis of these two proteins. To clarify this, we transiently transfected the HPV E6 expression plasmid into the p53-null H1299 human lung cancer cells (Chen et al., 1993) and quantitated the MGMT protein and activity levels. We previously showed that H1299 cells express high levels of MGMT (Srivenugopal et al., 2001). E6 expression in these cells induced a significant decrease in the specific activity of MGMT; the attenuation of MGMT activity correlated with the amount of E6 expression vector transfected (Figure 4a). The loss of MGMT activity in these experiments was due to a decrease in its protein content (Figure 4b), in
E6-mediated degradation of MGMT KS Srivenugopal and F Ali-Osman
5943
a
b
c
accordance with the mechanism of action known for E6. p53 also has the ability to modulate MGMT gene transcription. In previous studies, we and others have shown that overexpression of wild-type p53 transcriptionally represses the MGMT gene (Srivenugopal et al., 2001; Harris et al., 1996). Therefore, the p53 deficiency induced by E6 is unlikely to have contributed to the observed MGMT down-regulation, and the lack of alteration in MGMT mRNA levels in E6-expressing cells (Figure 1c) supports this notion. Rather, the reduced MGMT content we observed after the transient transfection of H1299 cells (Figure 4a,b) and after stable transfection of E6 into the four cell lines (Figure 1a,b) confirm a p53-independent degradation of MGMT by the E6 oncoprotein. a
b
c Figure 3 Evidence for the binding of MGMT with E6 protein in vitro and the association of MGMT, E6 and E6-AP proteins in vivo. (a) A pull-down assay was performed to show the physical association of MGMT and E6 proteins. HCT116 parent and HCT116/E6 cells were sonicated in the buffer containing 50 mM Tris-HCl (pH 8.0), 0.5 mM dithiothreitol, 0.1 mM EDTA, and 150 mM NaCl, and extracts were prepared by centrifugation. Human recombinant MGMT protein (2 mg) expressed with an N-terminal hexahistidine-tag (Srivenugopal et al., 2000) was then added to the extracts (300 mg of protein in a total volume of 150 ml), and the mixtures incubated at 258C for 1 h. Next, a 15 ml slurry of the Talon resin (Clonetech, Palo Alto, CA, USA) was added to selectively bind the his-tagged MGMT and associated proteins (Sirvenugopal et al., 2002). The resin was retrieved by centrifugation, washed with Tris-buffer saline, and solibulized in 50 mM Tris-HCl containing 2% SDS. The samples were boiled, divided into two portions, electrophoresed separately on 12% SDS-polyacrylamide gels, and Western blotted. The blots were probed with antibodies to MGMT (Srivenugopal et al., 2001) or E6 proteins (Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA), and visualized by enhanced chemiluminiscence (ECL). (b) Co-precipitation of MGMT with E6-AP and E6 proteins in extracts of HCT116/E6 cells. The parent HCT116 and E6-expressing cells were lysed in Tris-buffered saline containing proteinase inhibitors and Nonidet P-40. The extracts were then precleaned with protein A-agarose, and the MGMT protein was immunoprecipitated using a monoclonal antibody as described previously (Srivenugopal et al., 2000). The immunoprecipitates were solublized, and then each sample was split into three portions, electrophoresed on SDS-polyacrylamide gels, and Western blots prepared. The blots were probed with antibodies to MGMT, E6-AP (Oncogene Research Products, Boston, MA, USA), and E6 proteins. (c) Reduced degradation of 14C-MGMT protein in HCT116/E6 cell extracts after immunodepletion of E6-AP. Extracts from HCT116 and HCT116/E6 cells were prepared in 40 mM Tris-HCl (pH 7.8) and 0.5 mM dithiothreitol. Aliquots of 150 mg protein (40 ml vol) were mixed with protein A-Sepharose beads containing 1.5 mg of anti-E6-AP monoclonal antibody or non-immune serum (IgG), and kept at 48C for 1 h. After removal of the beads by centrifugation, the supernatants were supplemented with MgCl2 (4 mM), ATP (1 mM), ubiquitin (0.5 mg) and [14C]MGMT (1.5 mg; 7000 c.p.m.), and incubated for 40 min at 378C (Ali-Osman et al., 2001). Degradation of MGMT was assessed by SDS – PAGE followed by autoradiography. The top panel shows a representative autoradiographic pattern. Lane 1 contained the 14C-MGMT substrate alone, lanes 2 and 3 had the HCT116 cell extract without and with E6-AP depletion re-
Figure 4 p53-induced degradation of MGMT by E6 and the effect of proteasome inhibition. (a) Down-regulation of MGMT protein after transient transfection of E6 in p53-null H1299 human lung cancer cells. H1299 cells (American type Culture Collection, Rockville, MD, USA) were plated in 10 cm dishes for 24 h before the experiments and transfected with the pCMV-E6 vector (kindly provided by Dr Kathleen Cho, University of Michigan, Ann Arbor, MI, USA; Keiss et al., 1993) using Lipofectamine Plus Reagent (GIBCO – BRL, Gaithersburg, MD, USA). The cells were treated with the E6 construct (1 – 3 mg) or vector DNA alone in the presence of 2% serum for 10 h, which was then replaced with regular medium (DMEM-10% serum). Forty-eight hours after transfection, the cells were used for quantitating MGMT activity, as described in the legend for Figure 1. (b) Western blot analysis of MGMT in H1299 cells transfected with the E6 expression plasmid. Extracts from the untransfected H1299 cells (lane 1) and cells transfected with 3 mg pcDNA3.1E6 for 24 h (lane 2) and 48 h (lane 3) were electrophoresed, blotted, and probed with antibodies to MGMT. (c) Increased accumulation of ubiquitinated MGMT in lactacystin-treated HCT116/E6 cells. HCT116 cells expressing the E6 oncoprotein and the HCT116 parent cells were treated with 5 mM lactacystin (Calbiochem, San Diego, CA, USA) for 3 h. Next, the cells were lysed in the presence of 5 mM N-ethylmaleimide (NEM), and MGMT protein was immunoprecipitated as described previously (Srivenugopal et al., 1996). NEM inhibits the proteasome and deubiquitinating enzymes and enhances the detection of ubiquitinated proteins. The immunoprecipitates were electrophoresed on SDS gels, Western blotted and probed with MGMT antibodies. The bands higher than 22 kDa represent the ubiquitinated MGMT (Ub-MGMT) protein
spectively, and lanes 3 and 4 had HCT116/E6 cell extracts without and with E6-AP depletion respectively. Bottom panel – the MGMT bands were quantitated by densitometry and the per cent MGMT protein degraded was calculated. The mean values from two independent experiments are shown Oncogene
E6-mediated degradation of MGMT KS Srivenugopal and F Ali-Osman
5944
Additional evidence for the increased ubiquitination of the MGMT protein in E6-expressing cells was obtained using lactacystin, a specific inhibitor of proteasomal activity (Fenteany and Schreiber, 1998). In this experiment, HCT116/E6 and HCT116 cells were exposed to 5 mM lactacystin for 3 h and the cellular MGMT protein was immunoprecipitated and Western blotted (Figure 4c). During this brief period of proteasome inactivation, no changes were apparent in the ubiquitination pattern of MGMT in the parent HCT116 cells (lanes 1 and 2). However, the intensity of the higher molecular-weight protein bands (422 kDa) which represent the ubiquitin-conjugated MGMT (Srivenugopal et al., 1996) were significantly increased by lactacystin in the HCT116/E6 cells (lane 4). The greater accumulation of ubiquitinated MGMT following proteasome inhibition in E6 expressing cells is consistent with the expected increase of ubiquitin conjugation catalyzed by the E6/E6-AP complexes. The presence of higher molecular weight MGMT in non-E6 HCT116 cells suggests the normal turnover of this protein through the ubiquitin proteolytic route. Although a 3 h exposure of HCT116 cells to lactacystin did not induce alterations in the ubiquitinated MGMT protein species (lanes 1 and 2 in Figure 4c), such changes were evident after longer exposures (48 h) to the proteosome inhibitor. A closer examination of the human MGMT protein revealed some structural features that may serve as potential recognition elements for the E6-AP and E6 proteins. These include (i) an L2G box (consensus sequence LLGXXXS/T) which was identified in the multicopy maintenance protein 7 (Mcm7), and was shown to interact with an identical sequence present in the E6-AP for mediating efficient ubiquitin conjugation (Kuhne and Banks, 1998), and (ii) a PDZ-like domain, which serves as a recognition site for the binding of oncogenic E6 proteins (Thomas et al., 2001). We identified in the human MGMT protein an L2G sequence, 31IKLLGKGTSAAD42, that is identical to that observed in Mcm7. Further, the secondary structure of MGMT deduced by X-ray crystal studies (Wibley et al., 2000) reveals a partial, but typical PDZlike domain at its N-terminus, that is composed of three b-strands connected to a a-helix (Doyle et al., 1996). It is also noteworthy that the L2G box present in MGMT is located just next to the N-terminal b-sheet in the putative PDZ domain, and that among mammalian MGMTs, only human MGMT has the L2G sequence. PDZ domains are globular structures that function as specific protein-recognition modules in a variety of cellular processes including the binding of carboxyterminus of E6 proteins (Harris and Lim, 2001; Thomas et al., 2001; Lee et al., 1997). In fact, a computer search program (http://scansite.mit.edu; Yaffe et al., 2001) identified the sequence, 35GKGTSAADAVEVPAP48 which immediately follows the L2G box in MGMT, as a class 3 PDZ site, originally identified in the neuronal nitric oxide synthase (nNOS) (Schepens et al., 1997). These structural considerations lend support to our observations that MGMT is a target for the E6/E6AP system, however, the actual involvement of these Oncogene
motifs remains to be confirmed through mutation studies. Our study raises the question as to whether MGMT is a physiological substrate for E6-AP in cells not infected with HPV. It is pertinent however, to note that p53, which also contains an L2G box (Kuhne and Banks, 1998), does not appear to be a substrate for E6AP in HPV-negative cells (Talis et al., 1998). The persistence of the MGMT protein, with up to 50% remaining intact in E6-transfected cells (Figure 1), is interesting and warrants discussion. Although stable expression of the E6 gene induces a consistent decrease in p53 protein levels in tumor cells, p53 is typically not completely eliminated, with 10 – 25% of the control protein levels still persisting (Gupta et al., 1997). In addition, in HeLa cells, and other cervical cancer cells that express the E6 gene through a natural infection with HPV, p53 exists at low steady-state levels and is readily detectable (Talis et al., 1998; Scheffner et al., 1991; Butz et al., 1995). Further, transfection of the E6 gene markedly down-regulates the cellular levels of other targets, namely the E6TP1, c-myc, and scribble protein, but not their complete disappearance (Gao et al., 1999; Gross-Mesilaty et al., 1998; Nakagawa and Huibregtse, 2000).The extent of protein elimination may also depend on the cellular levels of E6-AP and the binding affinities of the E6-AP/E6 complexes with the target protein. Further, the longer half-life of cellular MGMT (412 h) (Brent et al., 1991) may also contribute to the ubiquitination reaction catalyzed by the E6/E6-AP system. In agreement with these observations, the HPV-positive HeLa cells do contain moderate levels of MGMT (Zhang et al., 2001). The E6-induced deficiency of MGMT has important implications for alkylation-induced DNA damage and repair in HPV-susceptible human cancers. The E6induced reduction of MGMT is likely to be beneficial for the chemotherapy of HPV-positive cancers, and this needs to be explored. On the flip side, the reduced repair of alkylation damage may contribute to increase genomic instability and accelerate the progression of cervical carcinogenesis. The finding that HPV-infected cells also contain decreased levels of the glutathione Stransferase M1, a key xenobiotic metabolizing enzyme (Chen and Nirunsuksiri, 1999), is consistent with such a notion. Abbreviations
MGMT, O6-methylguanine-DNA methyltransferase; HPV, human papillomavirus; BCNU, 1,3-bis(2-chloroethyl)-1-nitrosourea; rMGMT, recombinant MGMT protein; E6-AP, E6-Associated Protein Acknowledgements We thank Dr Jiang Shou for initial experiments and technical assistance. We are grateful to Drs Michael Kastan, Albert Fornace Jr, and Wafik el-Deiry for supply of E6-transfected cell lines. This work was supported by grants from the National Institutes of Health (CA-74321), Pediatric Brain Tumor Foundation of the United States, and the Association for Reseach of Childhood Cancer to KS Srivenugopal.
E6-mediated degradation of MGMT KS Srivenugopal and F Ali-Osman
5945
References Ali-Osman F, Shou J, Mullapudi SR and Srivenugopal KS. (2001). Proc. Am. Assoc. Cancer Res., 42, 572. Baker CC, Phelps WC, Lindgren V, Braun MJ, Gonda MA and Howley PM. (1987). J. Virol., 61, 962 – 971. Beroud C and Soussi T. (1998). Nucleic Acids Res., 26, 200 – 224. Boyer SN, Wazer DE and Bond V. (1996). Cancer Res., 56, 4620 – 4624. Bosch FX, Manos MM, Munoz N, Sherman M, Jansen AM, Peto J, Schiffman MH, Moreno V, Kurman R and Shah KV. (1995). J. Natl. Cancer Inst., 87, 796 – 802. Brent TP, von Wronski MA, Remack JS and Bigner DD. (1991). Proc. Am. Assoc. Cancer Res., 32, 421 (abstract 2505). Butz K, Shahabeddin L, Geisen C, Spitkovsky D, Ullmann A and Hoppe-Seyler F. (1995). Oncogene, 10, 927 – 936. Butz K, Whitaker N, Denk C, Ullman A, Geisen C and Hoppe-Seyler F. (1999). Oncogene, 18, 2381 – 2386. Chen JY, Funk WE, Wright JW and Minna JD. (1993). Oncogene, 8, 2159 – 2166. Chen C and Nirunsuksiri W. (1999). Carcinogenesis, 20, 699 – 703. Damalas A, Ben-Ze’ev A, Simcha I, Shtutman M, Leal JF, Zhurinsky J, Geiger B and Oren M. (1999). EMBO J., 18, 3054 – 3063. Dolan ME and Pegg AE. (1997). Clin. Cancer Res., 3, 837 – 847. Doyle DA, Lee A, Lewis J, Kim E, Sheng M and MacKinnon R. (1996). Cell,, 85, 1067 – 1076. Esteller M, Toyota M, Sanchez-Cespedes M, Capella G, Peinado MA, Watkins DN, Issa JP, Sidransky D, Baylin SB and Herman JG. (2000). Cancer Res., 60, 2368 – 2371. Fenteany G and Schreiber SL. (1998). J. Biol. Chem., 273, 8545 – 8548. Gao Q, Srinivasan S, Boyer SN, Wazer DE and Band V. (1999). Mol. Cell Biol., 19, 733 – 744. Gardiol D, Kuhne C, Glaunsinger B, Lee SS, Javier R and Banks L. (1999). Oncogene, 18, 5487 – 5496. Gerson SL and Willson JKV. (1995). Hematol. Oncol. Clin. North Am., 9, 431 – 450. Gross-Mesilaty S, Reinstein E, Bercovich B, Tobias KE, Schwartz AL, Kahana C and Ciechanover A. (1998). Proc. Natl. Acad. Sci. USA, 95, 8058 – 8063. Gupta M, Fan S, Zhan Q, Kohn KW, O’Connor PM and Pommier Y. (1997). Clin. Cancer Res., 3, 1653 – 1660. Harris BZ and Lim WA. (2001). J. Cell Sci., 114, 3219 – 3231. Harris LC, Remack JS, Houghton PJ and Brent TP. (1996). Cancer Res., 56, 2029 – 2032. Huibregtse JM, Scheffner M and Howley PM. (1991). EMBO J., 10, 4129 – 4135. Hussain SP, Aguilar F and Cerutti P. (1994). Oncogene, 9, 13 – 18. Isaacs JS, Barrett JC and Weissman BE. (1999). Mol. Carcinogen., 24, 70 – 77. Keiss TD, Slebos RJ, Nelson WG, Kastan WG, Plunkett BS, Han SM, Lorincz AT, Hendrick L and Cho KR. (1993). Proc. Natl. Acad. Sci. USA, 90, 3988 – 3992. Key LD, Adler-Storthz K, Mitchell MF, Clayman GL and Chen Z. (1999). Oral. Oncol., 35, 415 – 420. Kuhne C and Banks L. (1998). J. Biol. Chem., 273, 34302 – 34309.
Lee SS, Weiss R and Javier R. (1997). Proc. Natl. Acad. Sci. USA, 94, 6670 – 6675. Maki CG, Huibregtse JM and Howley PM. (1996). Cancer Res., 56, 2649 – 2654. Mitra S and Kaina B. (1993). Prog. Nucleic Acids Res. Mol. Biol., 44, 109 – 142. Nakagawa S and Huibregtse JM. (2000). Mol. Cell Biol., 20, 8244 – 8253. Pegg AE. (2000). Mutat Res., 462, 83 – 100. Rapp L and Chen JJ. (1998). Biochem. Biophys. Acta, 1378, F1 – F19. Sadot E, Geiger B, Oren M and Ben-Ze’ev A. (2001). Mol. Cell Biol., 21, 6768 – 6781. Scheffner M. (1998). Pharmacol. Ther., 78, 129 – 139. Scheffner M, Huibregtse J and Howley P. (1994). Proc. Natl. Acad. Sci. USA, 91, 8797 – 8801. Scheffner M, Munger K, Byrne JC and Howley PM. (1991). Proc. Natl. Acad. Sci. USA, 88, 5523 – 5527. Scheffner M, Werness B, Huilbregtse J, Levine A and Howley P. (1990). Cell, 63, 1129 – 1136. Schepens J, Cuppen E, Wieringa B and Hendriks W. (1997). FEBS Lett., 409, 53 – 56. Schwarz E, Freese UK, Gissman W, Mayer B, Roggenbuck A, Stremlau H and zur Hausen H. (1985). Nature, 314, 111 – 114. Sehr P, Zumbach K and Pawlita M. (2001). J. Immunol. Methods, 253, 153 – 162. Skhedy D, Gonen H, Bercovich B and Ciechanover A. (1994). FEBS Lett., 348, 126 – 130. Srivenugopal KS, Mullapudi SRS, Shou J, Hazra TK and Ali-Osman F. (2000). Cancer Res., 60, 282 – 287. Srivenugopal KS, Shou J, Mullapudi SRS, Lang Jr FF, Rao JS and Ali-Osman F. (2001). Clin. Cancer Res., 7, 1398 – 1409. Srivenugopal KS, Yuan X-H, Friedman HS and Ali-Osman F. (1996). Biochemistry, 35, 1328 – 1334. Srivenugopal KS, Mullapudi SRS and Ali-Osman F. (2002). Cancer Lett., 181, 87 – 93. Talis AL, Huibregtse JM and Howley PM. (1998). J. Biol. Chem., 273, 6439 – 6445. Thomas M and Banks L. (1998). Oncogene, 17, 2943 – 2954. Thomas M, Glaunsinger B, Pim D, Javier R and Banks L. (2001). Oncogene, 20, 5431 – 5439. Wajnberg EF and Fagan JM. (1989). FEBS Lett., 243, 141 – 144. Wang J, Sampath A, Raychaudhuri P and Bagchi S. (2001). Oncogene, 20, 4740 – 4749. Wibley JEA, Pegg AE and Moody PCE. (2000). Nucleic Acids Res., 28, 393 – 401. Wolf P, Hu YC, Doffek K, Sidransky D and Ahrendt SA. (2001). Cancer Res., 61, 8113 – 8117. Wu GS, Saftig P, Peters C and El-Diery WS. (1998). Oncogene, 16, 2177 – 2183. Yaffe MB, Leparc GG, Lai J, Obata T, Volinia S and Cantley LC. (2001). Nat. Biotechnol., 19, 348 – 353. Zaidi NH, Pretlow TP, O’Riordan MA, Dumenco LL, Allay E and Gerson SL. (1995). Carcinogenesis, 16, 451 – 456. Zhang Q, Ohannesian DW, Kreklau EL and Erickson LC. (2001). J. Pharmacol. Exp. Ther., 298, 141 – 147. zur Hausen H. (1999). Semin. Cancer Biol., 9, 405 – 411.
Oncogene
