The CXXC Zn binding motifs of the human papillomavirus ... - Nature

Oncogene (1998) 16, 1085 ± 1089  1998 Stockton Press All rights reserved 0950 ± 9232/98 $12.00

SHORT REPORT

The CXXC Zn binding motifs of the human papillomavirus type 16 E7 oncoprotein are not required for its in vitro transforming activity in rodent cells Joris Braspenning1, Antonio Marchini1, Valentina Albarani1, Laura Levy2, Francesca Ciccolini1, Caterina Cremonesi1, Robert Ralston3, Lutz Gissmann1 and Massimo Tommasino1 1

Angewandte Tumorvirologie, Deutsches Krebsforschungszentrum, Heidelberg, Germany; 2Chiron Corporation, Virology Department, Emeryville, California, USA; 3GeneMedicine Inc. The Woodlands, Texas, USA

The conserved region 3 (CR3) of the E7 protein of human papillomaviruses contains two CXXC motifs involved in zinc binding and in the homodimerization of the molecule. Studies have suggested that the intact CXXC motifs in the CR3 of HPV16 and HPV18 E7 are required for the in vitro transforming activity of these proteins. CR3 also contains a low anity pRb binding site and is involved in the disruption of the E2F/Rb1 complex. E7 is structurally and functionally related to Adenovirus E1A protein, which also has two CXXC motifs in CR3. However, the Ad E1A transforming activity appears to be independent of the presence of such domains. In fact, this viral protein exists in vivo as two dierent forms of 289 and 243 amino acids. The shorter Ad E1A form (Ad E1A243), where both CXXC motifs are deleted by internal splicing, retains its in vitro transforming activity. We have investigated if the HPV16 E7 CR3 can be functionally replaced by the Ad E1A243 CR3, which lacks both CXXC motifs. A chimeric protein (E7/E1A243) containing the CR1 and CR2 of HPV16 E7 fused to the CR3 of Ad E1A 243 was constructed. The E7/E1A243 while not able to homodimerize in the S. cerevisiae two-hybrid system retains several of the properties of the parental proteins, HPV16 E7 and Ad E1A. It associates with the `pocket' proteins, induces growth in soft agar of NIH3T3 cells and immortalizes rat embryo ®broblasts. These data suggest that the CXXC motifs in CR3 of E7 do not play a direct role in the transforming properties of this viral protein but probably are important for maintaining the correct protein con®guration. Keywords: Human papillomavirus type 16 E7 protein; E7-E1A243 chimeric protein; CXXC motifs; cellular transformation; E7 homodimerization

In the last two decades the association of certain types of human papillomavirus (HPV) with cervical cancer, e.g. HPV type 16 and 18 has been clearly demonstrated (zur Hausen, 1991). Several lines of evidence suggest that the products of two early genes, E6 and E7, play a key role in the induction of malignant transformation

Correspondence: M Tommasino The ®rst two authors have equally contributed to this work Received 30 July 1997; revised 1 October 1997; accepted 3 October 1997

of the host cells. For instance, HPV DNA has been found to be integrated into the genome of the malignant cells in most of the cases (Choo et al., 1987; Cullen et al., 1991). This viral DNA integration results in the disruption of various viral genes with consistent preservation of only the early E6 and E7 genes, which are actively transcribed (Schwarz et al., 1985). The E7 proteins of the `high risk' HPV16 and 18 genotypes are able to induce focus formation and growth in soft agar in a variety of established rodent ®broblast lines, e.g. NIH3T3 and Y31 cells (for review see Mansur and Androphy, 1993). They also cooperate with the activated ras gene product to transform primary rodent cells and with E6 to immortalize primary human keratinocytes, the natural host of the virus (Mansur and Androphy, 1993). E7 is structurally and functionally related to other viral oncoproteins, such as the Adenovirus E1A (Ad E1A) and the SV40 large T antigen (Tag) Phelps et al., 1988; Vousden and Jat, 1989). These viral oncoproteins exert their transforming activity by interacting with and altering the function of a number of cellular proteins. The Ad E1A and HPV E7 associate with the so called `pocket' proteins, pRb, p107 and p130 (Whyte et al., 1988, 1989; Dyson et al., 1989; Davies et al., 1993; Hu et al., 1995), which are involved in controlling the activity of several transcription factors, including proteins of the E2F family (Kouzarides, 1995). The pRb/E7 interaction results in the inactivation of the regulatory function of the `pocket' protein with consequent constitutive activation of the transcription factors (reviewed in Tommasino and Crawford, 1995). On the basis of its homology with Ad E1A, HPV16 E7 protein can be divided into three domains: conserved region 1 (CR1), conserved region 2 (CR2) and conserved region 3 (CR3). The disruption of the E2F/pRb by Ad E1A protein is mediated by two independent pRb binding domains, the low and high anity binding sites, which are located in the CR1 and CR2 of Ad E1A, respectively (Fattaey et al., 1993; Ikeda and Nevins, 1993). In HPV16 E7, while the high anity `pocket' protein binding site is located in CR2 as in Ad E1A, the low anity site is present in the CR3 (Patrick et al., 1994). In agreement with this, independent studies have shown that the HPV16 E7 CR3, together with CR1 and CR2, is required for the disruption of E2F/pRb complex (Huang et al., 1993; Wu et al., 1993). Another feature of the HPV16 E7 CR3 is the presence of two cys-x-x-cys motifs (CXXC)

E7 CXXC Zn binding motifs and transformation J Braspenning et al

1086

Figure 1 Schematic representation of the chimeric protein E7/E1A243 construction. The E7/E1A243 gene was constructed by overlapping PCR. The N-terminal domain of HPV16 E7 and the C-terminal domain of E1A243 were separately ampli®ed by PCR using the following oligos as indicated in the ®gure: (a) GAATTCATGCATGGAGATACACCT, (b) TTCAGACACAGGACCCTCTTCATCCTCCTCCTCTGA, (c) TCAGAGGAGGAGGATGAAGAGGGTCCTGTGTCTGAA (d) AAGCTTTTATGGCCTGGGGCGTTT. The two ampli®ed fragments were subsequently mixed and a second PCR reaction was performed using oligos (a) and (d). EcoRI and HindIII sites were added upstream and downstream the ORF, respectively. The ampli®ed chimeric gene was cloned into the retroviral vector pMXSV neo-18, and the nucleotide sequence was veri®ed by dideoxy chaintermination methods. Boxed regions indicate the Ad E1A and HPV16 E7 conserved regions 1, 2, 3 (CR1, 2 and 3). The length of the proteins in amino acid, the position of the fusion between Ad E1A and HPV16 E7 (underlined numbers) are also indicated. The black bars indicate the position of the epitopes recognized by the speci®c monoclonal antibodies for HPV16 E7 (6D) and Ad E1A (M73)

which are involved in the binding of a zinc molecule (Barbosa et al., 1989). Studies on E7 proteins of HPV type 16 and 18 have indicated that the binding of zinc through the CXXC motifs mediates the dimerization of the viral protein (McIntyre et al., 1993; Clemens et al., 1995; Zwerschke et al., 1996). Point mutations in one of the two CXXC motifs of E7 severely impair its ability to homodimerize and to co-operate with ras in the transformation of rat embryo ®broblasts (McIntyre et al., 1993). Thus, homodimerization appears to be required for the biological activity of the E7 protein. Although Ad E1A CR3 also contains two CXXC motifs in the CR3, its transforming activity is independent of the presence of such domains. In fact, Ad E1A protein can exist in vivo as two dierent forms of 289 or 243 amino acids. The shorter form of Ad E1A protein (Ad E1A243), where both CXXC motifs have been deleted by internal splicing, retains its in vitro transforming activity. To investigate if the E7 in vitro transforming activity can also be independent of the presence of the two CXXC motifs, we have constructed a chimeric protein containing the CR1 and CR2 of HPV16 E7 fused to the CR3 of Ad E1A 243. As shown in Figure 1, the junction between the two viral proteins, HPV16 E7 and Ad E1A243, is immediately after the HPV16 E7 CR2. The chimeric gene (E7/E1A243) was generated by overlapping PCR

(see legend Figure 1), cloned into a retroviral vector (pMXSV neo-18) and sequenced. Retroviral construct expressing E7/E1A243 gene was calcium transfected into an ecotropic producer cell line (GPE) and selected in G418 for 3 weeks. The progeny virus was then used to infect NIH3T3 cells in the presence of polybrene (8 mg/ml). Cells were grown in G418 and after 4 weeks single colonies were isolated. In order to check if the expression of the E7/E1A243 gene leads to the synthesis of a stable protein, we have determined the level of the chimeric protein in the total cell extracts by direct immunoblotting. The M73 monoclonal antibody that recognizes the C-terminal regions of Ad E1A was used (see Figure 1) (Harlow et al., 1985). A protein band of the expected molecular weight was detected in the extract of E7/E1A243 synthesizing cells, but not in the extract of uninfected NIH3T3 cells (Figure 2a). Although the chimeric protein was easily detected in the cellular extract, it was still essential to demonstrate whether the E7/E1A243 protein retains the ability to localize in the same sub-cellular compartments of the parental proteins. Therefore, we analysed the cellular localization of E7/E1A protein by immuno¯uorescence. As is shown in Figure 2b, a nuclear staining was observed only with the M73 antibody, while a monoclonal antibody reacting with an epitope in HPV16 E7 CR3 (see also Figure 1) did not give any speci®c staining. Next we checked if the E7/E1A243


a kDa

1

2

43 —

29 — E7/E1A243

protein is able to induce growth in soft agar of NIH3T3 cells. Three independent NIH3T3 established cell lines expressing the E7/E1A243 were plated on soft agar plates. After 4 weeks colonies were readily observed. These NIH3T3 cells still expressed the E7/ E1A243 gene and resembled the colonies obtained after infection of NIH3T3 cells with a retroviral construct expressing the HPV16 E7 gene (Figure 3) (data not shown). Another property of HPV16 E7 and Ad E1A proteins is the ability to immortalize primary rodent cells (for review see Mansur and Androphy, 1993). We also investigated if the E7/E1A243 protein retains the HPV16 E7 activity to immortalize rat embryo

pMXSVNeo-18

18 —

b pMXSVNeo-18 E7/E1A243

pMXSVNeo-18/HPV16 E7

M73

pMXSVNeo-18/E7/E1A243

6D Figure 2 E7/E1A243 is a stable protein and is localized in the nucleus. (a) immunoblot analysis of NIH3T3 cell expressing the E7/E1A243 gene. Protein gel was performed as previously described (Laemmli, 1970). For immunoblots proteins were electrotransferred onto PVDF membrane (Du Pont) and incubated with an anti E1A monoclonal antibody (M73). The protein bands were detected using peroxidase conjugated goat anti mouse IgG serum (DAKO) and ECL detection kit (Amersham). Line 1: 50 mg of NIH3T3 cell extract; line 2: 50 mg of NIH3T3 E7/E1A243 cell extract. (b) immuno¯uorescent staining of NIH3T3 cell expressing the E7/E1A243 gene. Cells were ®xed with cold methanol (20 min at 7208C) and stained with M73 hybridoma supernatant (dil 1/10) or with 6D from ascites (dil 1/200). Fluorescein-conjugated anti-mouse antiserum was used as secondary antibody. The stained coverslips were examined on a Zeiss Axiphot microscope

Figure 3 The chimeric protein E7/E1A243 induce growth in soft agar of NIH3T3 cells. NIH3T3 cells, after retroviral infection, were plated in 0.35% soft agar (FMC BioProducts) and DMEM supplemented with 10% calf serum (Sigma). After four weeks the plates were analysed for the presence of colonies

1087


1088

a

pMXSVNeo-18/HPV16 E7

b pMXSVNeo-18/E7/E1A243

Figure 4 Immortalization of Rat Embryo Fibroblasts by the E7/ E1A243 protein. REF were infected with retroviruses expressing HPV16 E7 or E7/E1A423 genes as indicated in the ®gure. After 4 ± 5 weeks under selection in G418, single colonies were isolated and grown for more than twenty passages

Table 1 Immortalization of rat embryo ®broblasts Plasmid pMXSVneo-18 pMXSVneo-18 16E7 pMXSVneo-18 E7/E1A243

No. of cultures immortalized/No. experiments 0/3 3/3 3/3

The table indicates the times that rat embryo ®broblasts were immortalized by the empty vector, by the vector expressing HPV16 E7 gene or the chimeric gene, E7/E1A243

®broblasts (REF). Immortalized clones were successfully established from cells infected with retroviruses expressing HPV16 E7 or the chimeric gene, but not from cells infected by empty retrovirus (Table 1 and Figure 4). The two CXXC motifs of E7 CR3 mediate the homodimerization of the viral protein, most likely through the binding of a molecule of zinc. HPV16 E7 CR3 was able to dimerize in the yeast `two hybrid system', but only when both of the CXXC motifs were left intact (Clemens et al., 1995; Zwerschke et al., 1996). In an independent study, McIntyre et al. (1993) showed that substitution of one of the cysteines in the CXXC motifs abolishes the ability of HPV18 E7 to homodimerize and severely impairs the E7 in vitro

Figure 5 E7/E1A243 protein does not homodimerize in the yeast `two hybrid' system. (a) schematic diagrams of the fusion proteins used in the yeast `two hybrid' assays. The dierent genes were fused as indicated in the Figure with the Gal4 DNA binding domain (Gal4bd) in the pGBT9 vector (Clontech) or with Gal4 transcription activator domain (Gal4ad) in the pGAD424 vector (Clontech). (b) b-Galactosidase activity in yeast. The strain S. cerevisiae SFY526 was transformed by the lithium acetate procedure with constructs showed in the Figure 5a in dierent combinations as indicated in the Figure 5b. The b-Galactosidase activity was measured as described in Clontech manual book of the yeast `two hybrid' system. The data are the mean of two independent experiments

transforming activity. Since E7/E1A243 retains the transforming properties of E7 in rodent cells, it was important to determine if the chimeric protein is also able to homodimerize. For this scope, we utilized the `two hybrid' method in Saccharomyces cerevisiae which has been previously used to demonstrate E7/E7 dimerization (Clemens et al., 1995; Zwerschke et al., 1996). The HPV16 E7 and E7/E1A243 genes were


fused to the transcription activation domain and to the DNA binding domain of the Gal4 protein (see Figure 5a). In this assay, activation of the b-Galactosidase reporter gene is strictly dependent on the level of interaction between the two fusion proteins. As shown in Figure 5b the E7/E1A243 protein did not homodimerize although it was able to interact with the `pocket' protein p107. The low level of bGalactosidase activity, which was detected in cell expressing the HPV16 E7 protein or the E7/E1A243 protein fused to the DNA binding domain of the Gal4 (see Figure 5b), is due to the intrinsic transcriptional activation properties of the HPV16 E7 N-terminus (CR1 and CR2) (Zwerschke et al., 1996). The present study demonstrates that the CR3 of Ad E1A is able to functionally replace the CR3 of HPV16 E7 protein. In fact, the E7/E1A243 retains the several biological properties of Ad E1A and HPV16 E7, e.g. cellular localization and the transforming activity in rodent cells. Most interestingly, the chimeric protein has lost the ability to homodimerize. Thus, the CXXC motifs in E7 protein, through the binding of zinc, are most likely required to maintain the correct folding of the viral protein, rather than playing a direct role via homodimerization in the E7-induced transforming processes. It has been shown that the dissociation of E2F/ `pocket' protein complexes by Ad E1A and HPV16 E7

requires two independent pRb binding sites with low and high anity (Fattaey et al., 1993; Ikeda and Nevin, 1993; Huang et al., 1993; Wu et al., 1993; Patrick et al., 1994). The low anity sites are located in dierent regions in Ad E1A and HPV16 E7, being present in CR1 and CR3, respectively (see also Figure 1). Consequently, the chimeric protein E7/E1A should not contain the low anity pRb binding site. We have demonstrated that E7/E1A243 protein retains the ability to interact with p107 and is fully competent in disrupting a speci®c G1 phase p107/E2F complex in NIH3T3 cells (Zerfass et al., 1995). Other, new, unidenti®ed domains of the two viral proteins may contribute in the dissociation of the E2F/`pocket' protein complexes. Future studies on the E7/E1A243 protein would provide new information which will allow the identi®cation of such domains.

Acknowledgements We express our gratitude to Prof H zur Hausen for his constant support during the course of this work. We would also like to thank Dr R Tindle for the 6D anti-E7 antibody, Dr E Harlow for M73 anti-E1A antibody and Drs P Jansen-DuÈrr, S Reshkin and I Zehbe for their constructive comments on the manuscript. AM and VA were supported by a fellowship from Associazione Italiana per la Ricerca sul Cancro.

References Barbosa MS, Lowy DR and Schiller JT. (1989). J. Virol., 63, 1404 ± 1407. Choo KB, Pan CC and Han SH. (1987). Virology, 161, 259 ± 261. Clemens KE, Brent R, Gyuris J and Munger K. (1995). Virology, 214, 289 ± 293. Cullen AP, Reid R, Csmpion M and Lorincz AT. (1991). J. Virol., 65, 606 ± 612. Davies R, Hicks R, Crook T, Morris J and Vousden K. (1993). J. Virol., 67, 2521 ± 2528. Dyson N, Howley PM, Munger K and Harlow E. (1989). Science, 243, 934 ± 937. Fattaey AR, Harlow E and Helin K. (1993). Mol. Cell. Biol., 13, 7267 ± 7277. Hu TH, Ferril SC, Snider AM and Barbosa MS. (1995). Int. J. Oncol., 6, 167 ± 174. Harlow E, Franza BR and Schley C. (1985). J. Virol., 55, 533 ± 546. Huang PS, Patrick DR, Edwards G, Goodhart PJ, Huber HE, Miles L, Garsky VM, Oli A and Heimbrook DC. (1993). Mol. Cell. Biol., 13, 953 ± 960. Ikeda MA and Nevins JR. (1993). Molecular and Cellular Biology, 13, 7029 ± 7035. Kouzarides T. (1995). Semin. Cancer Biol., 6, 91 ± 98. Laemmli UK. (1970). Nature, 227, 680 ± 685. Mansur CP and Androphy EJ. (1993). Biochim. Biophys. Acta, 1155, 323 ± 345.

McIntyre MC, Frattini MG, Grossman SR and Laimins LA. (1993). J. Virol., 67, 3142 ± 3150. Patrick DR, Oli A and Heimbrook DC. (1994). J. Biol. Chem., 269, 6842 ± 6850. Phelps WC, Yee CL, MuÈ nger K and Howley PM. (1988). Cell, 53, 539 ± 547. Schwarz E, Freese UK, Gissmann L, Mayer W, Roggenbuck B, Stremlau A and zur Hausen H. (1985). Nature, 314, 111 ± 114. Tommasino M and Crawford L. (1995). Bioessays, 17, 509 ± 518. Vousden KH and Jat PS. (1989). Oncogene, 4, 153 ± 158. Whyte P, Ruley HE and Harlow E. (1988). J. Virol., 62, 257 ± 265. Whyte P, Williamson NM and Harlow E. (1989). Cell, 56, 67 ± 75. Wu EW, Clemens KE, Heck DV and Munger K. (1993). J. Virol., 67, 2402 ± 2407. Zerfass K, Levy L, Cremonesi C, Ciccolini F, Jansen-DuÈrr P, Crawford L, Ralston R and Tommasino M. (1995). J. Gen. Virol., 76, 1815 ± 1820. Zur Hausen H. (1991). Virology, 184, 9 ± 13. Zwerschke W, Joswig S and Jansen-DuÈrr P. (1996). Oncogene, 12, 213 ± 220.

1089