Stimulation of Poliovirus Synthesis in a HeLa Cell-Free In Vitro Translation-RNA Replication System by Viral Protein 3CD pro

Transcription

1 JOURNAL OF VIROLOGY, May 2005, p Vol. 79, No X/05/$ doi: /jvi Copyright 2005, American Society for Microbiology. All Rights Reserved. Stimulation of Poliovirus Synthesis in a HeLa Cell-Free In Vitro Translation-RNA Replication System by Viral Protein 3CD pro David Franco, 1 Harsh B. Pathak, 2 Craig E. Cameron, 2 Bart Rombaut, 3 Eckard Wimmer, 1 and Aniko V. Paul 1 * Department of Molecular Genetics and Microbiology, School of Medicine, Stony Brook University, Stony Brook, New York ; Department of Biochemistry and Molecular Biology, Pennsylvania State University, University Park, Pennsylvania ; and Department of Microbiology and Hygiene, Vrije Universiteit Brussel, B-1090 Brussels, Belgium 3 Received 20 August 2004/Accepted 30 December 2004 The plus-strand RNA genome of poliovirus serves three distinct functions in the life cycle of the virus. The RNA is translated and then replicated, and finally the progeny RNAs are encapsidated. These processes can be faithfully reproduced in a HeLa cell-free in vitro translation-rna replication system that produces viable poliovirus. We have previously observed a stimulation of virus synthesis when an mrna, encoding protein 3CD pro, is added to the translation-rna replication reactions of poliovirus RNA. Our aim in these experiments was to further define the factors that affect the stimulatory activity of 3CD pro in virus synthesis. We observed that purified 3CD pro protein also enhances virus synthesis by about 100-fold but has no effect on the translation of the polyprotein. Optimal stimulation is observed only when 3CD pro is present early in the incubation period. The stimulation, however, is abolished by a mutation either in the RNA binding domain of 3CD pro,3c pro R84S/ I86A, or by each of two groups of complementary mutations R455A/R456A and D339A/S341A/D349A at interface I in the 3D pol domain of 3CD pro. Surprisingly, virus synthesis is strongly inhibited by the addition of both 3C pro and 3CD pro at the beginning of incubation. We also examined the effect of other viral or cellular proteins on virus synthesis in the in vitro system. No enhancement of virus synthesis occurred with viral proteins 3BC, 3ABC, 3BCD, 3D pol, and 3C pro or with cellular protein PCBP2. These results suggest that 3CD pro has to be present in the reaction at the time the replication complexes are assembled and that both the 3C pro and 3D pol domains of the protein are required for its activity that stimulates virus production. * Corresponding author. Mailing address: Department of Molecular Genetics and Microbiology, School of Medicine, Stony Brook University, Stony Brook, NY Phone: (631) Fax: (631) apaul@notes.cc.sunysb.edu. The RNA genome of poliovirus (PV), a prototype of Picornaviridae, is used in three important processes in the viral life cycle: (i) it is an mrna, which directs the synthesis of a polyprotein; (ii) it serves as a template for RNA synthesis; and (iii) the progeny RNA associates with the nonstructural proteins during virus assembly. In the past, several different approaches have been used to elucidate the individual steps in poliovirus replication. The most complex and difficult method involves studying biochemical reactions in the infected cell itself. The second in complexity uses crude replication complexes isolated from infected cells to decipher the stages in the replication of the viral RNA (vrna). The simplest system utilizes purified components to conduct detailed biochemical analyses of individual reactions in RNA synthesis. The disadvantage of the latter approach is that it has not yet been possible to reconstitute active replication complexes that synthesize VPg-linked RNAs of plus and minus polarity or to produce infectious virions. The discovery more than a decade ago that authentic PV can be made de novo in HeLa cell-free translation RNA-replication reactions (22) has provided an important new tool to study individual steps in the life cycle of the virus, except those involving virus attachment, entry into the cell, and uncoating. The coupled translation-rna replication system utilizes extracts from uninfected HeLa cells, which support the translation of input viral RNA, yielding all of the necessary viral proteins (22), the formation of membranous replication complexes (3, 4, 11, 23), the uridylylation of VPg (24, 33), the synthesis of plus- and minus-strand RNAs (4, 22), and the encapsidation of the progeny viral RNA (22). Although the in vitro cell-free translation-rna replication reactions mimic, in large part, the processes observed in virus-infected cells, there are also significant differences between the two systems. The in vitro reactions are programmed with large amounts of viral RNA ( 0.5 g of RNA or RNA molecules per reaction) when compared to the RNA of a single or of a few viruses, which are initially replicated in an infected cell. In spite of this, the yield of virus in the in vitro reaction is low compared to PV-infected HeLa cells. The low yield of virus in the in vitro system, compared to in vivo, might be attributed to differences in the membranous structures where RNA replication takes place (11) or to a lack of sufficient quantities of active viral proteins for efficient RNA synthesis or encapsidation to occur. Particle instability might also contribute to the low titers of infectious virus in the in vitro reactions. Viral RNA replication in the infected host cell is carried out primarily by the RNA-dependent RNA polymerase 3D pol in conjunction with other viral and cellular proteins (reviewed in reference 34). Replication takes place on small vesicles, which are derived from membranes of the host cell and which are associated with the nonstructural proteins of the virus. The incoming viral RNA is first transcribed into a complementary minus strand yielding a double-stranded replicative intermedi- 6358

2 VOL. 79, 2005 STIMULATION OF PV SYNTHESIS BY VIRAL PROTEIN 3CD pro 6359 FIG. 1. Genomic structure of poliovirus and processing of the P3 domain of the polyprotein. The single-stranded RNA genome of poliovirus is shown with the terminal protein VPg attached to the 5 end of the RNA and the poly(a) tail at the 3 end. The 5 NTR and the 3 NTR are shown with single lines. The polyprotein (open box) contains structural (P1) and nonstructural (P2, P3) domains that are cleaved into individual polypeptides. Processing of the P3 domain by proteinase 3C pro /3CD pro is shown enlarged. ate. In the next step, the minus strand is used as a template for the synthesis of the progeny plus strands. In addition to the RNA polymerase, the other viral proteins most likely involved in RNA replication are a small membrane-bound protein, 3A, and its precursor, 3AB, the terminal protein and primer for RNA synthesis VPg, and the multifunctional proteinase 3C pro / 3CD pro. Protein 3CD pro, the precursor of both proteinase 3C pro and RNA polymerase 3D pol, is derived from the P3 domain of the poliovirus polyprotein (Fig. 1), primarily by a proteolytic cleavage at a Q/G cleavage site between 3CD pro and 3AB. 3CD pro is a multifunctional polypeptide, with roles both in polyprotein processing and in RNA replication. As a proteinase, 3CD pro is required for the processing of the structural precursor domain P1 of the polyprotein (19, 46) and for the cleavage of a cellular protein (36). Its role in replication is related to its ability to bind RNA and to form ribonucleoprotein complexes with cisacting RNA elements of the viral genome. It forms complexes with the 5 -terminal cloverleaf structure of the PV RNA either in the presence of 3AB (16) or in the presence of cellular protein poly(rc) binding 2 protein (PCBP2) (1, 2, 13, 28). The 3CD pro /PCBP2 complex has been proposed to have a role both in the switch from translation to replication (12) and also in the circularization of the genome (17). The RNA binding activity of 3CD pro is also important for an interaction with the cre(2c) RNA element, during the uridylylation of VPg by 3D pol (33, 45). Finally, 3CD pro, in a complex with 3AB, interacts with the 3 nontranslated region (NTR)-poly(A) segment of the PV genome (16), but the biological significance of this interaction has not yet been determined. In the in vitro cell-free system, translation of the polyprotein and processing of the viral proteins are optimal during the first 3 h postincubation, but RNA replication is relatively slow, reaching a maximum between 8 and 9 h (41). The formation of infectious virus reaches a peak after a 12-h incubation period. Since 3CD pro has been speculated to play a role in the switch from translation to replication (12), we reasoned that at least one of the reasons for the observed imbalance between translation and RNA synthesis or encapsidation in the in vitro system could be the lack of sufficient quantities of 3CD pro. Indeed, when we cotranslated 3CD pro mrna with the viral RNA (vrna) template, we observed a 100-fold increase in virus yield compared to reactions in which only the template RNA was added (41). However, under the same conditions, only a 3-fold enhancement of viral RNA synthesis by 3CD pro mrna was observed (41), which we attributed to an imbalance between translation and replication in vitro. An alternate explanation of the findings is that the large increase in virus yield is due the effect of 3CD pro on an additional step or steps of the viral life cycle. The aim of this study was to further define how 3CD pro enhances virus yield in the in vitro cell-free translation-rna replication reactions programmed with viral RNA. We have now shown that 3CD pro protein can stimulate virus production just as well as the translation products of its mrna. Our results suggest that the RNA binding activity of 3CD pro is required for its stimulatory activity and that both the 3C pro and 3D pol domains of the protein are required for function. In addition, we have tested several mature and precursor proteins of the P3 domain of the viral polyprotein and cellular protein PCBP2 and found that they all lacked the ability to stimulate virus synthesis in vitro. MATERIALS AND METHODS Cells and viruses. HeLa R19 cell monolayers and suspension cultures of HeLa S3 cells were maintained in Dulbecco s modified Eagle s medium supplemented with 5% fetal bovine calf serum. Poliovirus was amplified on HeLa R19 cells as described before. The infectivity of virus stocks was determined by plaque assays on HeLa R19 monolayers as described by Lu et al. (21). Preparation of poliovirus RNA. Virus stocks were grown and they were purified by CsCl gradient centrifugation (21). Viral RNA was isolated from the purified virus stocks with a 1:1 mixture of phenol and chloroform. The purified RNA was precipitated by the addition of 2 volumes of ethanol. Preparation of HeLa cytoplasmic extracts. HeLa S10 extracts were prepared as previously described (7, 22), except for the following modifications: (i) packed

3 6360 FRANCO ET AL. J. VIROL. cells from 2 liters of HeLa S3 cells were resuspended in 0.8 to 1.0 volume (relative to packed cell volume) of hypotonic buffer; (ii) the final extracts were not dialyzed. Translation-RNA replication reactions with HeLa cell extracts and plaque assays. Viral RNA (500 ng) was translated at 34 C in the presence of unlabeled methionine, 200 M each CTP, GTP, UTP, and 1 mm ATP in a total volume of 25 l (22, 23). After incubation for 12 to 15 h, the samples were diluted with phosphate-buffered saline and were added to HeLa cell monolayers (22, 23). Virus titers were determined by plaque assay, as described previously (22, 23). Preinitiation RNA replication complexes. Preinitiation RNA replication complexes were prepared as described previously (3), except for some minor modifications. Translation-RNA replication reactions lacking initiation factors were incubated for 4 h at 34 C in the presence of 2 mm guanidine HCl. The complexes were isolated by centrifugation, resuspended in 50 l HeLa S10 translation-rna replication reaction mixture without viral RNA, and incubated for 11 h at 34 C. Proteins. The following proteins with a C-terminal His tag were expressed in Escherichia coli and purified by nickel column chromatography (QIAGEN): (1) cellular protein PCBP2 (28); poliovirus proteins 3CD pro (3C pro H40A) (33), 3C pro (C147G) (31), 3CD pro (3C pro H40G; 3D pol D339A/S341A/D349A) (31), and 3ABC(3C pro H40A) (A. Paul and E. Wimmer, unpublished results). The expression in E. coli of poliovirus 3BC(3C pro C147G) and 3BCD(3C pro C147G) with a C-terminal His tag, and of untagged 3CD pro (C147G) will be described elsewhere (H. B. Pathak and C. E. Cameron, unpublished results). The expression and purification of 3CD pro (3C pro H40G; 3D pol R455A, R456A) was described previously (31). Poliovirus protein 3D pol was expressed in E. coli from plasmid pt5t3d and was purified as described before (30). Construction of plasmids. Poliovirus sequences were derived from plasmid pt7pvm, which contains the full-length (nucleotides 1 to 7525) plus strand poliovirus cdna sequence (40). All constructs were sequenced to ensure their accuracy. FIG. 2. Stimulation of virus production in the in vitro translation- RNA replication system by 3CD pro. In vitro translation-rna replication reactions and plaque assays were carried out as described in Materials and Methods. Where indicated, 3CD pro mrna or purified 3CD pro (3C pro H40A) protein was added to the translation reactions. (A) Effect of 3CD pro mrna concentration. The amount of 3CD pro mrna added to the translation reactions at t 0 h is indicated in the figure. (B) Effect of purified 3CD pro protein concentration. The amount of purified 3CD pro (3C pro H40A) protein added to the translation reactions at t 0 h was varied as indicated in the figure. From three different experiments, the average value for the stimulation of virus synthesis by 3CD pro was 130-fold. (C) Comparison of the stimulatory activities of 3CD pro mrna and purified 3CD pro (3C pro H40A) protein. Virus production was measured with optimal concentrations of either 3CD pro mrna (1.4 g/ml) or 3CD pro (3C pro H40A) protein (5.5 nm) added to the translation-rna replication reactions. (i) plop315ser. Plasmid plop315ser contains the 3CD pro coding sequence preceded by a translation start codon and the T7 promoter sequence (41). It was digested with EcoRI prior to transcription by T7 RNA polymerase. (ii) plop315ser(3c pro R84S/I86A). An EcoRI-to-BglII fragment of pt7pvm (3C pro R84S/I86A) was ligated into similarly restricted plop315ser. The plasmid was digested with EcoRI and was transcribed by T7 RNA polymerase. Transcription and in vitro translation. All plasmids were linearized with EcoRI prior to transcription by T7 RNA polymerase. The transcript RNAs were purified by phenol-chloroform extraction and ethanol precipitation. Translation reactions (25 l) containing 8.8 Ci of 35 S-TransLabel (ICN Biochemicals) were incubated for 12 h at 34 C (22, 23). The samples were analyzed by electrophoresis on sodium dodecyl sulfate-12% polyacrylamide gels, followed by autoradiography. RESULTS We have previously shown that the addition of 3CD pro mrna to translation-rna replication assays, programmed with viral RNA, results in about 100-fold stimulation of virus production (41). The aim of the experiments presented here was to extend those studies and to obtain further information about the mechanism by which 3CD pro exerts its function. 3CD pro mrna stimulates virus production in the in vitro translation-rna replication system. In this study, we confirmed our previous observation (41) that the addition of 3CD pro mrna along with viral RNA at the time translation commences increases the virus yield about 100-fold from 6

4 VOL. 79, 2005 STIMULATION OF PV SYNTHESIS BY VIRAL PROTEIN 3CD pro 6361 FIG. 3. 3CD pro has to be added early to the translation-rna replication reactions to exert optimal stimulatory activity. In vitro translation- RNA replication reactions and plaque assays were carried out as described in Materials and Methods. Purified 3CD pro (3C pro H40A) protein was added to the reactions at the indicated times. (A) Effect of time of 3CD pro addition. The time at which 3CD pro (H40A) protein (5.5 nm) was added to the translation reactions was varied as indicated. (B) Addition of purified 3CD pro (3C pro H40A) protein to preinintiation replication complexes. Preinitiation replication complexes (PIRC) were made as described in Materials and Methods. The complexes were resuspended in either the absence (column 1) or presence (column 2) of purified 3CD pro (3C pro H40A) (5.5 nm). In column 3, 3CD pro (3C pro H40A) (5.5 nm) was added at t 0 h before the formation of the PIRC PFU/ g RNA ( PFU/ml) (Fig. 2A, column 1) to PFU/ g RNA ( PFU/ml) (Fig. 2A, columns 4 to 7). The extent of stimulation is strongly dependent on the concentration of 3CD pro mrna used, with an optimal range between 1.4 to 11.2 g RNA/ml of reaction. As the RNA concentration is increased further, first the extent of stimulation is reduced and then the virus production is inhibited to a level below the value obtained in the absence of 3CD pro (Fig. 2A, compare column 1 with columns 8 and 9). Purified 3CD pro protein enhances the virus yield in translation-rna replication reactions. The enhancement of virus synthesis in the in vitro translation-rna replication assay can be achieved not only by the cotranslation of 3CD pro from an mrna but also by the addition of purified 3CD pro protein, regardless of whether it carries a C-terminal His tag (3CD pro 3C pro H40A) (Fig. 2B) or is untagged (3CD pro 3C pro C147G) (data not shown). In both cases, the proteolytic activity of 3CD pro is inactivated through a mutation, as indicated. The extent of stimulation is dependent on protein concentration, and optimal stimulation is reached at 300 to 400 ng/ml (4.5 to 5.5 nm) 3CD pro (Fig. 2B). At their optimal concentrations, purified 3CD pro (3C pro H40A), just like 3CD pro mrna, leads to a 100-fold increase in virus yield (Fig. 2C, compare column 1 with columns 2 and 3). These results confirm our previous finding that translation of the 3CD pro mrna to yield the corresponding polypeptide is required for its enhancing activity (41). The 3CD pro (3C pro H40A) protein used in these experiments had no effect on the overall translation or processing of the viral polyprotein (see Fig. 7, lane 2). It should be noted that, with a typical HeLa cell extract, the yield of virus varies between different experiments from to PFU/ g RNA, which corresponds to to PFU/ml. In reactions supplemented with optimal concentrations of either 3CD pro mrna or purified protein, the fluctuation in virus yield is less pronounced. It is consistently in the range of to PFU/ g RNA ( to PFU/ml). Therefore the extent of stimulation by 3CD pro varies from 20- to 200-fold. Because of the relatively large variation in virus yield in reactions lacking extra 3CD pro, we have included such a control in all of our experiments. The data shown on all figures represent an average of two or more experiments. 3CD pro has to be added early to retain its optimal enhancing activity in virus synthesis. In an effort to determine the time at which 3CD pro has to be added to the reactions to exert its stimulatory function during the growth cycle of the virus, we have tested the effect of adding purified 3CD pro (3C pro H40A) protein at various times after the reactions were incubated. As shown in Fig. 3A, 3CD pro (3C pro H40A) can be added either at 0 h or 2 h of incubation to retain its optimal enhancing activity in virus production (compare column 1 with columns 2 and 3). If added at 4 h, the stimulatory effect of 3CD pro is already somewhat diminished (Fig. 3A, compare columns 2 and 3 with column 4), and if added at 6 or 8hofincubation, the enhancement is nearly completely abolished (Fig. 3A, compare columns 2 and 3 with columns 5 and 6). 3CD pro does not stimulate virus synthesis when added to isolated preinitiation replication complexes. Addition of guanidine HCl, a reversible inhibitor of poliovirus replication, to in vitro translation-rna replication reactions completely inhibits RNA synthesis (22). By supplementing such reactions with this drug, it is possible to isolate preinitiation replication complexes (3, 4). This drug blocks the initiation of minus-strand RNA synthesis but has no effect on poliovirus

5 6362 FRANCO ET AL. J. VIROL. FIG. 4. Mutations both in the 3C pro and 3D pol domains of 3CD pro affect the stimulation of virus synthesis. In vitro translation-rna replication reactions and plaque assays were carried out as described in Materials and Methods. (A) Either wild-type or mutant 3C pro (R84S/I86A) 3CD pro mrnas (1.4 g/ml) were added to the reactions at t 0 h as indicated. From three different experiments, the average value for the stimulation of virus synthesis by 3CD pro (3C pro R84S/I86A) was 1.3-fold. (B) CD pro (3C pro H40A), 3CD pro (3C pro H40G; 3D pol R455A/R456A), or 3CD pro (3C pro H40G; 3D pol D339A/S341A/D349A) purified proteins (5.5 nm) were added to the reactions at t 0 h as indicated. From two different experiments, the average value for the stimulation of virus synthesis by 3CD pro (3C pro H40G/3D pol R455A/R456A) protein was 1.0-fold and by 3CD pro (3C pro H40G; 3D pol D339A/S341A/D349A) it was 1.4-fold. polyprotein translation and processing. Upon the removal of guanidine, the synthesis of negative-strand RNA is initiated in a synchronous manner followed by the production of plusstrand RNA (3, 4). To evaluate the effect of 3CD pro on virus synthesis by preinitiation replication complexes, we have carried out the translation of vrna in the presence of 2 mm guanidine HCl for 4 h. The complexes were isolated by centrifugation, the pellets were resuspended in HeLa extracts, and the incubation was continued for 11 h. When purified 3CD pro (3C pro H40A) protein was added at the beginning of the incubation, the yield of virus was 100-fold higher than in reactions which contained no extra 3CD pro (Fig. 3B, compare column 1 with column 3). On the other hand, the addition of 3CD pro (3C pro H40A) to the resuspended preinitiation complexes had no effect on the production of infectious virions (Fig. 3B, compare column 1 with column 2). These results, as well as those presented in Fig. 3A, suggest that 3CD pro has to be present in the reaction at an early time point in the viral life cycle so that it can be incorporated into replication complexes that are subsequently required for RNA replication and/or encapsidation. RNA binding by 3CD pro is required for its stimulatory activity in virus synthesis. The 3CD pro polypeptide is the precursor of both proteinase 3C pro and the RNA polymerase 3D pol. Although both the proteinase and RNA binding sequences of 3CD pro reside in the 3C pro domain, the two proteins differ greatly with respect to these activities. In functioning as a proteinase, only 3CD pro and not 3C pro has the ability to cleave the structural precursor P1 (19, 46). When functioning as an RNA binding protein at the cloverleaf, in concert with PCBP2 (13, 28), or with 3AB (16), 3CD pro has an enhanced ability to form a functional ribonucleoprotein complex over 3C pro. The stimulatory activity of 3CD pro in virus production in the in vitro translation-rna replication reactions is not due to its proteolytic activity. As pointed out before, the purified protein used in our experiments is inactive as a proteinase due to a mutation (3C pro H40A) in its active site. To test the possibility that the RNA binding activity of 3CD pro is required for its enhancing function, we mutated the RNA binding domain of the protein (3C pro R84A/I86A) (5, 14). As shown in Fig. 4A, translation of 3CD pro (3C pro R84A/I86A) mrna along with the viral RNA template had no significant stimulatory effect on virus production (compare column 2 with column 3). This indicates that the RNA binding domain of 3C pro, in the context of 3CD pro, is required for the enhancing function of the protein. The 3D pol domain of 3CD pro is also important for its stimulatory activity in virus synthesis. It has been previously shown that the 3D pol domain of 3CD pro modulates both the proteinase and RNA binding activities of the protein (6, 29). The RNA polymerase itself is a multifunctional protein, which has two types of synthetic activities. It has the ability to elongate RNA primers on suitable templates (10) and to covalently link UMP to the hydroxyl group of a tyrosine in VPg (33). The crystal structure for 3D pol revealed that this enzyme possess a typical nucleic acid polymerase structure containing a palm, thumb, and finger subdomains (15). A considerable amount of evidence has accumulated thus far suggesting that 3D pol oligomerizes and that the oligomeric forms of the protein are important for function (18, 30). Two charged amino acids, R455 and R456 (8), located in the thumb domain along interface I interact with D339/S341/D349 in the palm subdomain of another polymerase molecule, and this interaction is essential for stability along interface I (31). It was recently shown that the R455A/R456A mutations in the thumb domain of 3D pol strongly reduce the ability of the polymerase to catalyze 3CD pro -stimulated uridylylation of VPg on the cre(2c) element but the 3D pol palm mutations have no detrimental effect

6 VOL. 79, 2005 STIMULATION OF PV SYNTHESIS BY VIRAL PROTEIN 3CD pro 6363 FIG. 5. Effect of 3ABC, 3BC, 3BCD, 3C pro, and PCBP2 on virus production in the translation-rna replication system. In vitro translation- RNA replication reactions and plaque assays were carried out as described in Materials and Methods. Purified viral protein was added (t 0h) to the translation-rna replication reactions (40 ng/ml and 400 ng/ml). The effect of each protein on virus synthesis was tested at least twice, and the data shown are an average of the two experiments. (A) The effect of 3ABC, 3BC, 3BCD, or 3C pro on virus production. Column 2, 3CD pro (3C pro H40A), 5.5 nm; columns 3 and 4, 3ABC(3C pro H40A), 2.5 nm and 12.5 nm, respectively; columns 5 and 6, 3C pro (C147G), 2 nm and 20 nm, respectively; columns 7 and 8, 3BC(3C pro C147G), 1.8 nm and 18 nm, respectively; columns 9 and10, 3BCD(3C pro C147G), 0.54 nm and 5.4 nm, respectively. (B) Effect of adding PCBP2 with or without 3CD pro on virus production. Purified PCBP2 and/or purified 3CD pro (3C pro H40A) was added as indicated in the figure. on the reaction (31). Interestingly, in the context of 3CD pro, neither of these groups of mutations in 3D pol altered the ability of the protein to stimulate the uridylylation reaction in vitro (31). However, about 2-fold-higher concentrations of the mutant proteins were required to achieve optimal RNA binding of a cre(2c) RNA probe in vitro than what was observed with the wild-type 3CD pro protein (Pathak and Cameron, unpublished results). In addition, the thumb mutant 3CD pro proteinase exhibited a modest reduction in the processing of the VP0/VP3 precursor in in vitro translation reactions of mutant genomic transcript RNAs (31). To determine whether the 3CD pro protein, carrying the 3D pol R455A/R456A mutations, possesses normal stimulating activity of virus production in the in vitro system, we have compared its enhancing activity to that of 3CD pro (3C pro H40A). As shown in Fig. 4B, there is no stimulation of virus production in the presence of the mutant 3CD pro (3D pol R455A/R456A) protein (compare column 1 with column 3), while there is a 100-fold enhancement of virus yield when 3CD pro (3C pro H40A) is added to the reaction (compare column 1 with column 2). These results suggest that the integrity of interface I in 3D pol, in the context of 3CD pro, is important for the protein s stimulatory activity. If this is true, then one would expect that 3CD pro carrying the palm mutations D339A/S341A/D349A in the 3D pol domain, would exhibit the same phenotype as the thumb mutant protein 3CD pro (3D pol R455A/R456A). This hypothesis was confirmed by the finding that 3CD pro carrying the 3D pol palm mutations D339A/ S341A/D349A also lacks stimulatory activity in virus production (Fig. 4B, compare column 1 with column 4). It should be noted that increasing the concentration of the mutant 3CD pro polypeptides from the standard concentration (5.5 nm) to 11 nm and 22 nm did not enhance virus synthesis (data not shown). Since not only 3CD pro (3C pro H40A) but also both mutant 3D pol -3CD pro polypeptides contain a mutation (H40G) in the proteinase 3C pro active site, the observed differences cannot be due to the altered proteolytic properties of the proteins. Effect of other viral and cellular proteins on the enhancement of virus production in the in vitro translation-rna replication system. Protein 3CD pro is the precursor of both proteinase 3C pro and the RNA polymerase 3D pol. Our finding that the RNA binding activity of 3CD pro is required for its enhancing properties and the fact that its RNA binding site is located in the 3C pro domain suggested to us the possibility that 3C pro or some of its precursors (3ABC, 3BC, 3BCD) might also stimulate virus synthesis in the in vitro translation-rna replication assay. Surprisingly, the addition of comparable amounts of purified 3C pro (C147G), 3BC (3C pro C147G), 3BCD(3C pro C147G), or 3ABC(3C pro H40A) proteins to the in vitro translation-rna replication reactions did not lead to an increase in viral yield (Fig. 5A). None of the proteins had any effect on the translation of the viral RNA template or processing of the polyprotein (data not shown). Cellular protein PCBP2 plays a role in both PV translation initiation and in viral RNA synthesis (42). This protein specifically interacts with two domains of the poliovirus 5 NTR, the cloverleaf structure, and domain IV of the internal ribosome

7 6364 FRANCO ET AL. J. VIROL. entry site (13, 28). Since PCBP2 is known to bind to the cloverleaf structure as a complex with 3CD pro (2), we have tested the possibility that the addition of both proteins together to the in vitro translation-rna replication reactions would result in even further stimulation of virus production. We observed that when PCBP2 was added at t 0htothe translation-rna replication reactions alone, at concentrations of 1 nm to 100 nm, it had no effect on virus yield (Fig. 5B, compare column 1 with column 4, and data not shown). However, when added together with 3CD pro at 100 nm ( 20-fold molar excess over 3CD pro ), PCBP2 totally blocked the stimulatory activity of 3CD pro (Fig. 5B, compare column 1 with columns 2 and 3). Exogenously added PCBP2 at the same concentrations had no effect on the translation of the viral RNA or processing of the polyprotein (data not shown). A possible explanation of these observations is that the added 3CD pro is sequestered with PCBP2 either at the 5 cloverleaf in a ternary complex (13, 28) or in a complex with poly(a) binding protein (PABP) (17) and that one or both of these complexes lack stimulatory activity in virus production. Protein 3C pro inhibits the enhancing activity of 3CD pro in virus production. Since 3CD pro is the precursor of both 3C pro and 3D pol, we have tested the possibility that supplying the translation-rna replication reactions with the two mature cleavage products, instead of the precursor, would also enhance virus production. However, no stimulation of virus synthesis can be achieved by adding to the in vitro reactions purified 3D pol and 3C pro together (Fig. 6, compare column 1 with column 6). Neither of the two proteins when added alone to the reactions has an effect on virus yield (Fig. 6, compare column 1 with columns 7 and 8). Surprisingly, when protein 3C pro is included in the reactions along with 3CD pro, there is a striking inhibition of virus production (Fig. 6, compare column 1 with column 4). Indeed, the yield of infectious virions in the reactions is reduced about 100-fold when compared to reactions to which no 3CD pro was added. The reason for the inhibitory activity of 3C pro is not yet understood, but it is likely related to a competition between the two proteins for some RNA sequence/structure that is required for replication and/or encapsidation. The fact that 3C pro by itself has no effect on virus yield suggests that it is not able to compete for that RNA sequence/structure with 3CD pro that is made in cis from the input viral RNA. Neither 3C pro alone, nor a combination of 3C pro with 3CD pro (3C pro H40A), has any detectable effect on the translation and processing of the viral polyprotein (Fig. 7, compare lane 1 with lanes 5 and 6). The translation reactions shown in Fig. 7 were incubated for 12 h at 34 C, but we also obtained the same results at earlier time points of incubation (data not shown). We have already shown in Fig. 3A that 3CD pro (H40A) protein has to be added to the translation-rna replication reactions early in the incubation to retain its optimal stimulatory activity. To determine whether 3C pro also has to be added to the reactions early in the incubation period to inhibit the stimulatory function of 3CD pro, we used preinitiation replication complexes. Table 1 shows that when 3C pro and 3CD pro are added together at t 0 h in the presence of guanidine HCl, followed by the isolation of preinitiation replication complexes and resuspension in HeLa extracts, virus yield is strongly inhibited. Note that in this case the polypeptides were added FIG. 6. Effect of adding 3CD pro together with 3C pro and/or 3D pol on virus production in translation-rna replication reactions. In vitro translation-rna replication reactions and plaque assays were carried out as described in Materials and Methods. Purified 3CD pro (3C pro H40A), 3D pol,or3c pro (C147G), each at 400 ng/ml, was added to the reactions at t 0 h as indicated in the figure. The molar concentration of 3CD pro (3C pro H40A) was 5.5 nm, that of 3C pro (C147G) was 20 nm, and that of 3D pol was 7.7 nm. The average value obtained from three different experiments for the effect of 3C pro on virus synthesis was 1.0-fold; the inhibitory effect of 3C pro and 3CD pro added together was 60-fold. while the viral RNA was translated in the presence of guanidine HCl. On the other hand, if 3CD pro is added at t 0 h but 3C pro is added at the time the isolated preinitiation complexes are resuspended, it has no effect on the stimulatory activity of 3CD pro on virus production (Table 1). These results confirm our previous finding that, for exogenously added 3CD pro to possess the ability to stimulate virus production, it already has to be present in the reactions at the time the replication complexes are assembled. DISCUSSION In this study, we have confirmed and extended our previous work on the stimulation of virus production by 3CD pro mrna in the in vitro cell-free translation-rna replication system, programmed by viral RNA. Our finding that purified 3CD pro stimulates virus yield just as well as the addition of its mrna proves that the function of 3CD pro occurs at the level of protein rather than RNA. Since the addition of 3CD pro to the in vitro reactions has no effect on the translation of viral proteins from the input RNA and on the processing of the polyprotein, we conclude that the enhancing activity of this protein is involved at a later stage of the viral growth cycle, such as RNA replication, encapsidation, or both. It should be noted, however, that the protein has to be added to the reactions during the first 2 to 4 h postincubation, which is the time of optimal translation, to retain its maximal stimulatory activity. In addition, after the replication complexes have been formed from the newly made viral proteins in the presence of 2 mm guanidine HCl, 3CD pro loses its ability to stimulate virus production. These observations suggest that to be fully active the exog-

8 VOL. 79, 2005 STIMULATION OF PV SYNTHESIS BY VIRAL PROTEIN 3CD pro 6365 FIG. 7. In vitro translation of vrna is not affected by the addition of 3CD pro or 3C pro. In vitro translation reactions of viral RNA were incubated for 12 h at 34 C, and the products were analyzed (see Materials and Methods). Purified proteins (400 ng/ml) or mrna (1.4 g/ml) was added to the reactions at t 0 h. Lane 1, vrna; lane 2, 5.5 nm 3CD pro (3C pro H40A) protein; lane 3, 3CD pro (3C pro R84S/I86A) mrna; lane 4, 5.5 nm 3CD pro (3C pro H40G; 3D pol R455A/R456A) protein; lane 5, 20 nm 3C pro (C147G) protein; lane 6, 5.5 nm 3CD pro (3C pro H40A) and 20 nm 3C pro (C147G) proteins. enously added 3CD pro has to be present in the reactions at the time the replication complexes are assembled. 3CD pro functions in the viral growth cycle both as a proteinase and as an RNA binding protein. In an effort to elucidate the mechanism by which 3CD pro stimulates virus synthesis, we have mutated the RNA binding site of the protein (3C pro R84S/I86A) (5, 14) and showed that this abolished its stimulatory activity. Since the purified 3CD pro we use in these in vitro reactions is proteolytically inactive, its stimulatory activity is likely to be, at least in part, due to its RNA binding properties. Interaction of 3CD pro with RNA is known to be important for RNA replication, minimally at two different locations within the poliovirus genome. These are the cloverleaf (1, 2, 13, 28, 35) and the cre(2c) element (33, 45). 3CD pro also binds to the 3 NTR-poly(A) region of the RNA genome, but the relevance of this interaction for RNA replication has not yet been demonstrated (16). Whether the binding of any of these RNA sequences/structures by 3CD pro is a prerequisite for the stimulation of virus synthesis in the in vitro system is not yet known. The observation that 3C pro is inactive in the stimulation but together with 3CD pro strongly inhibits virus production might be explained by a competition between these two proteins for the same essential RNA sequence/structure either alone or in a complex with other proteins during RNA replication and/or encapsidation. The fact that both 3C pro and 3CD pro form a ribonucleoprotein complex at the cloverleaf (2), but only the latter is biologically relevant, suggests the possibility that interaction with the cloverleaf is at least one of the steps involved in the stimulatory activity of 3CD pro in virus production. In contrast to the 5 cloverleaf, the cre(2c) RNA element binds 3C pro and 3CD pro equally well (45) and both proteins stimulate VPg uridylylation in vitro (31), suggesting that this reaction is not likely a process involved in the enhancement of virus production by 3CD pro. The observation that 3D pol and 3C pro individually cannot replace 3CD pro in its stimulatory properties indicates that the 3D pol domain of 3CD pro is also required for its function. This conclusion was confirmed by the observation that two 3CD pro proteins, each containing mutations in the 3D pol domain (R455A/R456A or D339A/S341A/D349A), possess very little, if any, enhancing activity in virus production in the in vitro system. Interestingly, about 2-fold-higher concentrations of the mutant 3CD pro proteins were required to achieve optimal RNA binding in vitro than with the wild-type protein. However, in our experiments a similar increase in mutant 3CD pro protein concentration did not enhance the stimulation of virus synthesis in vitro, suggesting that the inability of mutant proteins to stimulate virus synthesis is not simply due to their reduced RNA binding activities. It should be noted that these two mutations disrupt oligomerization of 3D pol along interface I; hence, it is possible that an interaction of 3CD pro molecules or of 3CD pro with 3D pol is required for the stimulatory activity of the protein. However, in the yeast two-hybrid system, 3CD pro homopairs or 3CD pro and 3D pol exhibit only weak interactions (44). The observation that neither the palm nor the thumb 3D pol mutant, in the context of 3CD pro, is defective in enhancing VPg uridylylation in vitro (31), but they lack stimulating activity in virus synthesis, further confirms our conclusion that the cre(2c) templated uridylylation reaction is not affected by the exogenously added 3CD pro. The primary processing of the P3 domain of the polyprotein yields 3AB and 3CD pro, while alternate minor processing cascades produce 3ABC and 3D pol or 3A and 3BCD (20). None of the other viral proteins tested, which are potential precursors of 3C pro (3BC, 3BCD, 3ABC), possess the ability to stimulate virus production in the in vitro translation-rna replication reactions. It should be noted that 3ABC and 3BC are normally not detectable in PV-infected HeLa cell lysates (26) or in vitro translation reactions (22), but the possibility cannot be excluded that the reason for this is their short half-life. 3BCD can TABLE 1. Effect of adding 3C pro (C147G) to preinitiation replication complexes a 3CD pro (0 h) Effect on vrna at time shown 0h 3C pro 6h Infectivity (PFU/ g vrna) a Preinitiation complexes were made from reactions incubated in either the absence or presence of 3CD pro (3C pro H40A) (400 ng/ml, 5.5 nm) for 4 h (see Materials and Methods). 3C pro (C147G) (400 ng/ml, 20 nm) was added either at t 0 h or at two hours after resuspension of the complexes (t 6 h).

9 6366 FRANCO ET AL. J. VIROL. be observed both in vivo and in vitro in small amounts, particularly during the early hours of translation of poliovirus RNA (32). In other picornaviruses, such as encephalomyocarditis virus (EMCV) or hepatitis A virus (HAV), 3ABC is one of the major processing products (27, 37). The exact functions, if any, of these large precursors are not yet known. Recent observations, however, suggest that 3BC can substitute for VPg as a substrate for 3D pol in the uridylylation reaction (C. Cameron et al., unpublished results). Cellular protein PCBP2, which is known to form a complex with 3CD pro and bind to the cloverleaf (2, 13, 28), also has no effect on virus yield in the in vitro translation-rna replication system, but at high concentrations, it inhibits the stimulatory activity of the exogenously added 3CD pro. This result might be related to the sequestering of the exogenously added 3CD pro into a complex with PCBP2 at the cloverleaf (13, 28) or with PABP (17) that possesses no stimulatory activity in virus synthesis. It has been previously reported that poliovirus RNA replication in vivo requires protein translation in cis through an internal region of the genome (25). Other studies, in addition, showed that the formation of the poliovirus replication complex in vivo requires coupled viral translation, vesicle production, and viral RNA synthesis (9). These results can be interpreted to mean that during virus infection the proteins translated from the input viral RNA are essentially the only ones that assemble and form the replication complex. This conclusion is in agreement with other mutational studies that showed that mutations in nonstructural proteins couldn t be efficiently complemented in trans, or if they were, they represented only certain activities of a multifunctional protein (38, 39, 43). The reason for this might be the formation of a tightly enclosed replication complex in cis, which sequesters its components and prevents their exchange with proteins located on the outside (9). Our studies suggest that, in the vitro translation-rna replication system, one or more functions of 3CD pro can be provided in trans. The HeLa cell-free translation-rna replication system offers an easy way to investigate those steps in the life cycle of poliovirus, which include the synthesis and processing of the polyprotein, RNA replication, and encapsidation. In these experiments, we have analyzed the factors that affect the stimulatory properties of 3CD pro in virus production. We are now extending these studies to determine the effect extra 3CD pro directly on minus- and plus-strand RNA synthesis and on encapsidation (unpublished data). ACKNOWLEDGMENTS We thank K., Kirkegaard for plasmid pt5t3d and B. L. Semler for the PCBP2 expression clone. We are grateful to D. W. Kim for his help in the preparation of HeLa cell extracts and for helpful discussions. This work was supported by two grants from the National Institute of Allergy and Infectious Diseases (E. Wimmer, R37 AI ; and C. Cameron, AI053531). REFERENCES 1. Andino, R., G. E. Rieckhof, and D. Baltimore A functional ribonucleoprotein complex forms around the 5 end of poliovirus RNA. Cell 63: Andino, R., E. Rieckhof, P. L. Achacoso, and D. Baltimore Poliovirus RNA synthesis utilizes an RNP complex formed around the 5 end of viral RNA. EMBO J. 12: Barton, D. J., E. P. Black, and J. B. Flanegan Complete replication of poliovirus in vitro: preinitiation RNA replication complexes require soluble cellular factors for the synthesis of VPg-linked RNA. J. Virol. 69: Barton, D. J., and J. B. Flanegan Synchronous replication of poliovirus RNA: initiation of negative-strand RNA synthesis requires the guanidine-inhibited activity of protein 2C. J. Virol. 71: Blair, W. S., T. B. Parsley, H. P. Bogerd, J. S. Towner, B. L. Semler, and B. R. Cullen Utilization of a mammalian cell-based RNA binding assay to characterize the RNA binding properties of picornavirus 3C proteinases. RNA 4: Cornell, C. T., and B. L. Semler Subdomain specific functions of the RNA polymerase region of poliovirus 3CD polypeptide. Virology 298: Cuconati, A., A. Molla, and E. Wimmer Brefeldin A inhibits cell-free, de novo synthesis of poliovirus. J. Virol. 72: Diamond, S. E., and K. Kirkegaard Clustered charged-to-alanine mutagenesis of poliovirus RNA-dependent RNA polymerase yields multiple temperature-sensitive mutants defective in RNA synthesis. J. Virol. 68: Egger, D., N. Teterina, E. Ehrenfeld, and K. Bienz Formation of the poliovirus replication complex requires coupled viral translation, vesicle production, and viral RNA synthesis. J. Virol. 74: Flanegan, J. B., and D. Baltimore Poliovirus-specific primer-dependent RNA polymerase able to copy poly(a). Proc. Natl. Acad. Sci. USA 74: Fogg, M. H., N. L. Teterina, and E. Ehrenfeld Membrane requirements for uridylylation of the poliovirus VPg protein and viral RNA synthesis in vitro. J. Virol. 77: Gamarnik, A., and R. Andino Switch from translation to replication in a positive strand RNA virus. Genes Dev. 12: Gamarnik, A. V., and R. Andino Interactions of viral protein 3CD and poly(rc) binding protein with the 5 nontranslated region of the poliovirus genome. J. Virol. 74: Hammerle, T., A. Molla, and E. Wimmer Mutational analysis of the proposed FG loop of poliovirus proteinase 3C identifies amino acids that are necessary for 3CD cleavage and might be determinants of a function distinct from proteolytic activity. J. Virol. 66: Hansen, J. L., A. M. Long, and S. C. Schultz Structure of the RNAdependent RNA polymerase. Structure 5: Harris, K., W. Xiang, L. Alexander, W. S. Lane, A. V. Paul, and E. Wimmer Interaction of the poliovirus polypeptide 3CD pro with the 5 and 3 termini of the poliovirus genome: identification of viral and cellular cofactors needed for efficient binding. J. Biol. Chem. 269: Herold, J., and R. Andino Poliovirus RNA replication requires genome circularization through a protein-protein bridge. Mol. Cell 7: Hobson, S. D., E. S. Rosenblum, O. C. Richards, K. Richmond, K. Kirkegaard, and S. C. Schultz Oligomeric structures of poliovirus polymerase are important for function. EMBO J. 20: Jore, J., B. de Geus, R. J. Jackson, P. Pouwels, and B. Walk Poliovirus protein 3CD is the active protease for processing of the precursor protein P1 in vitro. J. Gen. Virol. 69: Lawson, M. A., and B. L. Semler Alternate poliovirus nonstructural processing cascades generated by primary sites of 3C proteinase cleavage. Virology 191: Lu, H.-H., C.-F. Yang, A. D. Murdin, M. H. Klein, J. J. Harber, O. M. Kew, and E. Wimmer Mouse neurovirulence determinants of poliovirus type 1 strain LS-a map to the coding regions of capsid protein VP1 and proteinase 2A pro. J. Virol. 68: Molla, A., A. V. Paul, and E. Wimmer Cell-free de novo synthesis of poliovirus. Science 254: Molla, A., A. V. Paul, and E. Wimmer Effects of temperature and lipophilic agents on poliovirus formation and RNA synthesis in a cell-free system. J. Virol. 67: Murray, K. E., and D. J. Barton Poliovirus CRE-dependent VPg uridylylation is required for positive-strand RNA synthesis but not for negative-strand RNA synthesis. J. Virol. 77: Novak, J. E., and K. Kirkegaard Coupling between genome translation and replication in an RNA virus. Genes Dev. 8: Pallansch, M. A., O. M. Kew, B. L. Semler, D. R. Omilianowski, C. W. Anderson, E. Wimmer, and R. R. Rueckert Protein processing map of poliovirus. J. Virol. 49: Parks, G. D., J. C. Baker, and A. C. Palmenberg Proteolytic cleavage of encephalomyocarditis virus capsid region substrates by precursors to the 3C enzyme. J. Virol. 63: Parsley, T. B., J. S. Towner, L. B. Blyn, E. Ehrenfeld, and B. L. Semler Poly((rC) binding protein 2 forms a ternary complex with the 5 -terminal sequences of poliovirus RNA and the viral 3CD proteinase. RNA 3: Parsley, T. B., C. T. Cornell, and B. L. Semler Modulation of the RNA binding and protein processing activities of poliovirus polypeptide 3CD by the viral RNA polymerase domain. J. Biol. Chem. 274: Pata, J. D., S. C. Schultz, and K. Kirkegaard Functional oligomerization of poliovirus RNA dependent RNA polymerase. RNA 1: Pathak, H. B., S. K. Ghosh, A. W. Roberts, S. D. Sharma, J. D. Yoder, J. J. Arnold, D. W. Gohara, D. J. Barton, A. V. Paul, and C. E. Cameron