Enhanced expression of vascular endothelial growth factor ... - Nature

70 downloads 52 Views 169KB Size Report
Enhanced expression of vascular endothelial growth factor (VEGF) plays a .... Laboratories), or a polyclonal anti-FGF-2 antibody (5 ng/ml; Upstate Biotechnology, Lake Placid, NY, .... Conference on Malignant Mesothelioma-Therapeutic Op-.
ã

Oncogene (2002) 21, 2896 ± 2900 2002 Nature Publishing Group All rights reserved 0950 ± 9232/02 $25.00 www.nature.com/onc

Enhanced expression of vascular endothelial growth factor (VEGF) plays a critical role in the tumor progression potential induced by simian virus 40 large T antigen Alfonso Catalano1, Mario Romano*,2, Stefano Martinotti3 and Antonio Procopio*,1 1

Institute of Experimental Pathology, University of Ancona, Faculty of Medicine, Via Ranieri, 60131 Ancona, Italy; 2Department of Human Pathology, University of Messina, 98125 Messina, Italy; 3Department of Oncology and Neuroscience, `G. D'Annunzio' University, Chieti, Italy

Vascular endothelial growth factor (VEGF), an important angiogenic factor, regulates cell proliferation, di€erentiation, and apoptosis through activation of its tyrosine-kinase receptors, such as Flt-1 and Flk-1/Kdr. Human malignant mesothelioma cells (HMC), which have wild-type p53, express VEGF and exhibit cell growth increased by VEGF. Here, we demonstrate that early transforming proteins of simian virus (SV) 40, large tumor antigen (Tag) and small tumor antigen (tag), which have been associated with mesotheliomas, enhanced HMC proliferation by inducing VEGF expression. SV40-Tag expression potently increased VEGF protein and mRNA levels in several HMC lines. This e€ect was suppressed by the protein synthesis inhibitor, cycloheximide. Inactivation of the VEGF signal transduction pathway by expression of soluble form of Flt-1 inhibited Flk-1/Kdr activation and HMC proliferation induced by SV40 early genes. Experiments with SV40 mutants revealed that SV40-Tag, but not -tag, is involved in the VEGF promoter activation. However, concomitant expression of SV40-tag enhanced Tag function. In addition, SV40-Tag expression sustained VEGF induction in colon carcinoma cell line (CCL)-233, which have wild-type p53, but not in CCL-238, which lack functional p53. These data indicate that VEGF regulation by SV40 transforming proteins can represent a key event in SV40 signaling relevant for tumor progression. Oncogene (2002) 21, 2896 ± 2900. DOI: 10.1038/sj/ onc/1205382 Keywords: SV40; Tag; VEGF; mesothelioma

Increased expression of angiogenic cytokines, such as ®broblast growth factor (FGF), platelet-derived growth factor (PDGF), and vascular endothelial growth factor (VEGF), has been shown in numerous malignant

*Correspondence: A Procopio; E-mail: [email protected] and M Romano; E-mail: [email protected] Received 26 September 2001; revised 29 January 2002; accepted 7 February 2002

disorders, including mesotheliomas (Kumar-Singh et al., 1999). Among these, VEGF, whose crucial role in tumor angiogenesis is well established (Hanahan and Folkman., 1996), acts also as a potent mitogen and survival factor for human malignant mesothelioma cells (HMC). In fact, VEGF regulates HMC proliferation through its tyrosine-kinase receptors activation (i.e. Flt-1 and KDR/¯k-1) (Strizzi et al., 2001). In addition, VEGF is the main e€ector of 5-lipoxygenase action on HMC survival (Romano et al., 2001). Thus, VEGF plays a key role by regulating multiple mechanisms of tumor progression, such as formation of new blood vessels and proliferation of neoplastic cells. The development of mesothelioma is associated with a history of asbestos exposure (Browne, 2001). However, other factors may intervene in the pathogenesis of this tumor. The presence of nucleic acids and proteins of Simian Virus 40 (SV40) has been observed in most of mesotheliomas (Carbone et al., 1994). Moreover, SV40 infection represents a negative prognostic cofactor for patients a€ected by mesothelioma (Procopio et al., 2000). SV40 encodes two transforming proteins, the large tumor antigen (Tag) and the small tumor antigen (tag). SV40-Tag, a 90 kDa nuclear phosphopolypeptide, is essential for virus growth and sucient to induce mesothelial cell transformation in the absence of cell lysis (Bocchetta et al., 2000). Although SV40-Tag displays pleotropic actions on multiple potential mechanisms of cell transformation (DeCaprio, 1999), it has been proposed that it may facilitate cell transformation by binding and inactivating p53 and retinoblastoma (Rb) tumor suppressor proteins (Carbone et al., 1997; De Luca et al., 1997). However, the molecular consequences of these associations are only beginning to be described. In the present report, we asked whether a relationship might exist between SV40-related proteins and VEGF expression. To this end, SV40-negative HMC (MPP89) (Orengo et al., 1999) were transiently transfected with an SV40-Tag expression vector, pw2dl (Bocchetta et al., 2000). As shown in Figure 1a, a timedependent expression of SV40-Tag was observed. The 88 and 90 kDa polypeptides detected by the Tag antibody correspond to the previously described Tag

SV40 large T antigen regulates VEGF expression A Catalano et al

phosphoisoforms (Paley, 1996). When it was examined the e€ect of SV40-Tag expression on VEGF secretion in MPP89, a selective temporal relationship between SV40-Tag expression and increased VEGF production was observed (Figure 1b). Moreover, 96 h after SV40Tag transfection, MPP89 displayed maximum increase in all VEGF protein isoforms (Figure 1a) and returned to basal levels within 6 days (results not shown). In contrast, SV40-Tag did not alter the levels of other HMC-derived angiogenic factors, such as FGF-type 2 and PDGF (Figure 1a). SV40-Tag also increased VEGF mRNA levels by ®vefold and eightfold after 48 and 72 h, respectively (Figure 1c). Similar results were obtained using three additional distinct HMC

lines (results not shown). Notably, VEGF mRNA expression induced by SV40-Tag was suppressed by the protein synthesis inhibitor cycloheximide (Figure 1d), indicating that the accumulation of VEGF mRNA induced by SV40-Tag is dependent on newly synthesized proteins. In the speci®c case of HMC, VEGF also acts as a potent autocrine growth factor. Thus, to establish the relevance of VEGF regulation by SV40-Tag, we antagonized VEGF activity in Tag-expressing HMC using an adenoviral vector encoding a soluble form of Flt-1 (Ad.sFlt-1) (Takayama et al., 2000). sFlt-1 expression signi®cantly inhibited VEGF accumulation in conditioned media of SV40-Tag-expressing MPP89

2897

Figure 1 SV40-Tag expression increased VEGF production in HMC. MPP89 were transfected with 3.0 mg pw2dl. (a) Cellular extracts were collected for the indicated times after transfection. Seventy-®ve mg of soluble protein extracts were separated by 10% SDS polyacrylamide gel electrophoresis and transferred to Immobilon-P membranes (Millipore, Bedford, MA, USA). VEGF, PDGF, and FGF-2 proteins were detected by autoradiography with a polyclonal anti-VEGF antibody (1 : 1000 dilution; Santa Cruz Biotechnology, Santa Cruz, CA, USA), a polyclonal anti-PDGF antibody (1 : 40 000 dilution; Je€rey Weidner, Merck Research Laboratories), or a polyclonal anti-FGF-2 antibody (5 ng/ml; Upstate Biotechnology, Lake Placid, NY, USA), as previously described (Gaetano et al., 2001). For analysis of SV40-Tag expression, 50 mg of protein extract was separated by 7% SDS polyacrylamide gel electrophoresis and Tag proteins were detected with a rabbit polyclonal anti-Tag antibody (Santa Cruz Biotechnology), as described (Carbone et al., 1997). Protein concentration was determined by standard Bradford protein assay (BioRad Lab., Hercules, CA, USA). Extracts from the Tag-expressing 1166 cells and human umbilical vein endothelial cells were used as positive controls (+) for Tag and VEGF expression, respectively. (b) Conditioned media of MPP89 transfected with pw2dl (+ Tag) or with the control plasmid, pw101 (7 Tag) were collected at the indicated times and VEGF was measured using an ELISA assay (R&D System Europe Ltd., Abingdon, UK) according to manufacturer's instructions. Results are from two independent experiments with duplicates. (c) Total RNA was extracted using the TRIzol reagent (Life Technologies, Milan, Italy). RNA (15 ± 20 mg) was resolved using a 1.2% agarose gel containing 6.3% formaldehyde, transferred to a HybondTM-N nylon membrane (Amersham) and hybridized with a 32P-labeled cDNA probe containing the 522 bp of the human VEGF sequence common to all known VEGF isoforms (Oncogene Research Products, Cambridge, MA, USA). All Northern blots were reprobed with random primed 32P-labeled human b-actin cDNA. Fold increase was calculated by densitometry after normalization for b-actin expression. Basal expression at time `0' was assumed as 1.0. RNA from a melanoma cell line (SK-Mel-147) was utilized as a positive control (+). (d) MPP89 transfected with pw2dl for 48 h, were treated with 25 mg/ml of cycloheximide (CHX) for additional 3 h, prior to total RNA extraction. VEGF mRNA was assessed by Northern blot. Results are from one representative experiment (n=2) Oncogene

SV40 large T antigen regulates VEGF expression A Catalano et al

2898

(P=0.03) (Figure 2a). Moreover, it abrogated both Flk-1/Kdr phosphorylation and DNA synthesis induced by SV40-Tag in HMC (Figure 2b,c). These results strongly indicate that VEGF signaling induced by SV40-Tag contributes to cell cycle modulation promoted by SV40. To investigate the mechanisms of VEGF expression by SV40 early genes, a VEGF promoter-luciferase construct, covering the promoter region 72018 to +50 (Finkenzeller et al., 1997), was transiently co-transfected into MPP89 with SV40 plasmids harboring various combinations of the SV40 transforming proteins. As shown in Figure 3a, a signi®cant increase in luciferase activity was denoted following transfection of the plasmid expressing both SV40 oncoproteins (T/t) and the plasmid expressing Tag alone (T/7) (P50.001

Figure 2 Impact of Ad.sFlt-1 on VEGF signaling triggered by SV40-Tag in HMC. Adenovirus vectors were ampli®ed in 293 cells and titered by plaque forming assay according to standard protocols. Viral stocks were maintained at 7808C for long-term storage before use. HMC were transfected with 4.0 mg of pw2dl (+ Tag) or with pw101 (7 Tag), incubated for 24 h, then infected with Ad.sFlt-1 or Ad.Null (50 p.f.u.6cell) for additional 48 h. (a) VEGF levels in conditioned media were measured by ELISA. Results are from n=3 (*P50.001). (b) Cell lysates were separated by 7.5% polyacrylamide gel electrophoresis and transferred to nitrocellulose membrane. Blots were exposed to PY20 anti-phosphotyrosine monoclonal antibody (1 : 1000) (Santa Cruz Biotechnology) for 1 h at room temperature (upper panels), stripped and re-blotted with an anti-¯k-1/Kdr antibody (lower panels). The bands were visualized using the ECL Western blotting system (Amersham Int., Bucks, UK). Results are from a single experiment representative of three. (c) 46104 cells/ml were seeded into 24-well plates and grown to 70% con¯uency over 24 h. Cells were made quiescent with FCS-free medium for 12 h before transfection experiments. At the end of the incubations, [3H]-Thymidine (0.5 mCi/ml) was added for additional 4 h. Details of the [3H]-Thymidine uptake assay are reported elsewhere (Strizzi et al., 2001). Results are mean+s.d. from n=3 with duplicates (*P50.05) Oncogene

Figure 3 Activation of a VEGF promoter construct by SV40 early genes. A VEGF promoter construct (2.0 mg) and plasmids (from 1.0 to 4.0 mg) harboring SV40 early genes: pw2, expressing both Tag and tag (T/t); pw2dl, expressing Tag alone (T/7); pw2t, expressing tag alone (7/t); and pw101, expressing only a mutated tag (7/7) were co-transfected using the lipofectamine-plus reagent (Life Technologies, Milan, Italy) into MPP89 (a) or A549 (ATCC, Manassas, VA, USA) (b). pCDNA3, containing the CMV immediate-early enhancer/promoter region coupled to b-galattosidase (0.5 mg), was also transfected as an internal control. After 24 h from transfection cells were lysed and tested for luciferase activity using the Double-LuciferaseTM Assay System (Promega, Milan, Italy). Results were normalized for bgalattosidase activity, measured by Galacto-Light PlusTM (TROPIX Inc., Bedford, MA, USA) according to manufacturer's instructions. Results represent mean+s.d. from three independent experiments (*P50.05; **P50.01)

SV40 large T antigen regulates VEGF expression A Catalano et al

2899

Figure 4 Analysis of p53 impact on SV40-Tag-induced VEGF expression. (a) MPP89, transfected with the VEGF promoter construct (2.0 mg) together with 3.0 mg of the SV40-Tag-expressing plasmid (+ Tag) or the control plasmid (7 Tag) were exposed to 10 mM of menadione for 8 h. Luciferase activity was determined as described in the legend to Figure 3. Results represent mean+s.d. from n=3 (*P50.05). (b) VEGF release was assessed by ELISA (n=3; *P=0.03). (c) CCL-238 (black bars) and CCL233 (gray bars) were transfected as above and VEGF promoter activity determined. Results were from n=3

and P=0.03, respectively). Thus, tag may act sinergistically with Tag to activate the VEGF promoter. In contrast, the vectors expressing tag alone (7/t) or a mutated tag, that did not bind protein phosphatase 2A (7/7), did not in¯uence VEGF transcription. These ®ndings are consistent with previous results showing that tag may enhance cell transformation potential of Tag by inhibiting protein phosphatase 2A activity (Bocchetta et al., 2000). Similar e€ects were observed when the lung cancer-derived cell line, A549, was transfected with the SV40 plasmids (Figure 3b), indicating that VEGF regulation by SV40 is not restricted to HMC. Thus, SV40 by its early genes expression can stimulate VEGF transcription, but, considering that VEGF mRNA accumulation depended on newly synthesized proteins (see Figure 1d), additional factors may intervene. Binding and inactivation of the tumor suppressor p53 by SV40-Tag has been observed in HMC (Carbone et al., 1997). Moreover, p53 can suppress VEGF expression (Zhang et al., 2000). Therefore, we asked whether p53 was partially involved in VEGF regulation by SV40. To answer this question we ®rst examined the

impact of menadione, a drug that selectively disrupts Tag/p53 nuclear complex (Gonin et al., 1999), on VEGF promoter activation induced by SV40-Tag. Menadione drastically reduced either VEGF promoter activity and protein levels induced by SV40-Tag (Figure 4a,b). As expected, menadione increased p53mediated p21WAF1/CIP1 expression (results not shown). Then, to con®rm that p53 was involved in the SV40Tag-mediated VEGF expression, we used a wild-type p53 (+/+) (CCL-233) and a p53-null (7/7) (CCL238) colon cancer cell lines, as recipient cells. We observed that SV40-Tag expression sustained VEGF induction in CCL-233, but not in CCL-238 (Figure 4c). Together these results indicate that VEGF expression induced by SV40-Tag is involved in the cell cycle perturbation caused by p53 inactivation in response to SV40 infection. The mechanisms of VEGF regulation remain to be fully investigated. Previous studies have shown that p53 represses VEGF transcription by preventing the binding of Sp-1 to the VEGF promoter (Zhang et al., 2000). More recently the involvement of Src kinase activity in p53 inhibition of VEGF transcription has Oncogene

SV40 large T antigen regulates VEGF expression A Catalano et al

2900

been assessed (Pal et al., 2001). In addition, also p16 and Rb family members can inhibit VEGF expression (Claudio et al., 2001). Therefore, it is possible that cell cycle regulatory proteins generally acting in the G1 phase of the cell cycle could have similar e€ects, or that these proteins control distinct pathways with a common end point. Since SV40 simultaneously can repress both p53 and Rb function to accomplish cellular transformation, VEGF up-regulation could represent a common molecular strategy for SV40 to regulate proliferation and tumor progression. However, the roles of these DNA tumor viruses in tumorigenesis need to be studied more closely. Our results suggest that the impact of SV40-Tag on VEGF expression is abrogated by blocking protein synthesis and is dependent by p53 status. Thus, the SV40-Tag/p53 binding may deregulate the synthesis of inhibitors or activators of VEGF transcription. Recent observation from our group points at nitric oxide synthase 2 as a key regulator of p53-VEGF interactions in HMC (Catalano et al., unpublished). However, the possibility that SV40-Tag can also induce a subset of transcription factors acting on the VEGF promoter in a p53-independent fashion cannot be ruled out.

In conclusion, we present here the ®rst evidence of a molecular link between SV40 expression and VEGF regulation involving p53 inactivation. These ®ndings may be particularly relevant for a better understanding of the mechanisms of tumor progression and open new prognostic and therapeutic perspectives for SV40related neoplasms such as mesotheliomas.

Acknowledgments This work was presented in part at the International Conference on Malignant Mesothelioma-Therapeutic Options and role of SV40: An Update, Chicago, Illinois (USA), April, 2001. It was supported by grants from the Associazione Italiana per la Ricerca sul Cancro (AIRC) and from the Italian Ministero dell' UniversitaÁ e della Ricerca Scienti®ca (ex 60%) to A Procopio. A Catalano was supported by a fellowship from FIRC. We thank Dr G Finkenzeller and Dr J Abraham for making the VEGF promoter construct, Dr MC Capogrossi for providing the adenovirus vectors, Dr G Vianale and Dr Strizzi for technical assistance and Dr M Bocchetta for providing the SV40 plasmids.

References Bocchetta M, Di Resta I, Powers A, Fresco R, Tosolini A and Testa JR. (2000). Proc. Natl. Acad. Sci. USA, 97, 10214 ± 10219. Browne K. (2001). Ann. Occup. Hyg., 45, 327 ± 329. Carbone M, Pass HI, Rizzo P, Marinetti M, Di Muzio M, Mew DJ, Levine AS and Procopio A. (1994). Oncogene, 9, 1781 ± 1790. Carbone M, Rizzo P, Grimley PM, Procopio A, Mew DJ, Shridhar V, de Bartolomeis A, Esposito V, Giuliano MT, Steinberg SM, Levine AS, Giordano A and Pass HI. (1997). Nat. Med., 3, 908 ± 912. Claudio PP, Stiegler P, Howard CM, Bellan C, Minimo C, Tosi GM, Rak J, Kovatich A, De Fazio P, Micheli P, Caputi M, Leoncini L, Kerbel R, Giordano GG and Giordano A. (2001). Cancer Res., 61, 462 ± 468. De Luca A, Baldi A, Esposito V, Howard CM, Bagella L, Rizzo P, Caputi M, Pass HI, Giordano GG, Baldi F, Carbone M and Giordano A. (1997). Nat. Med., 3, 913 ± 916. DeCaprio JA. (1999). Biologicals, 27, 23 ± 28. Finkenzeller GA, Sparacio A, Technau D and Siemeister G. (1997). Oncogene, 15, 669 ± 676. Gaetano C, Catalano A, Illi B, Felici A, Minucci S, Palumbo R, Facchiano F, Mangoni A, Mancarella S, Muhlhauser J and Capogrossi MC. (2001). Circ. Res., 88, E38 ± E47. Gonin S, Diaz-Latoud C, Richard MJ, Ursini MV, Imbo A, Manero F and Arrigo AP. (1999). Oncogene, 18, 8011 ± 8023.

Oncogene

Hanahan D and Folkman J. (1996). Cell, 86, 353 ± 364. Kumar-Singh S, Weyler J, Martin MJ, Vermeulen PB and Van Marck E. (1999). J. Pathol., 189, 72 ± 78. Orengo AM, Spoletini L, Procopio A, Favoni RE, De Cupis A, Ardizzoni A, Castagneto B, Ribotta M, Betta PG, Ferrini S and Mutti L. (1999). Eur. Respir. J., 13, 527 ± 534. Pal S, Datta K and Mukhopadhyay D. (2001). Cancer Res., 61, 6952 ± 6957. Paley EL. (1996). Carcinogenesis, 17, 939 ± 945. Procopio A, Strizzi L, Vianale G, Betta P, Puntoni R, Fontana V, Tassi G, Gareri F and Mutti L. (2000). Genes Chromosomes Cancer, 29, 173 ± 179. Romano M, Catalano A, Nutini M, D'Urbano E, Crescenzi C, Claria J, Davi G and Procopio A. (2001). FASEB J., 15, 2326 ± 2336. Strizzi L, Catalano A, Vianale G, Orecchia S, Casalini A, Tassi G and Procopio A. (2001). J. Pathol., 193, 468 ± 475. Takayama K, Ueno H, Nakanishi Y, Sakamoto T, Inoue K, Shimizu K, Oohashi H and Hara N. (2000). Cancer Res., 60, 2169 ± 2177. Zhang L, Yu D, Hu M, Xiong S, Lang A, Ellis LM and Pollock RE. (2000). Cancer Res., 60, 3655 ± 3661.