Journal of General Virology

Cell-free synthesis of poliovirus: 14S subunits are the key intermediates in the encapsidation of poliovirus RNA

Yvan Verlinden,¹ Andrea Cuconati,² Eckard Wimmer² and Bart Rombaut¹

¹ Department of Microbiology and Hygiene, Vrije Universiteit Brussel, Laarbeeklaan 103, B-1090 Brussels, Belgium
² Department of Molecular Genetics and Microbiology, School of Medicine, State University of New York at Stony Brook, Stony Brook, NY 11794, USA

In a cell-free system of uninfected HeLa cells, programmed with poliovirus RNA, extraneous radiolabelled 14S subunits assembled with endogenous 14S subunits and interacted with newly synthesized RNA to form virions (160S). This result suggests that 14S subunits are the key intermediates in the encapsidation of poliovirus RNA.

Despite many years of research, the assembly mechanism of poliovirus and other picornaviruses is still not understood (reviewed by Boeyé & Rombaut, 1992; Hellen & Wimmer, 1992; Putnak & Phillips, 1981; Rueckert, 1996). Even after the complete primary structure of the poliovirus genome was elucidated (Kitamura et al., 1981) and the capsid structure of poliovirus was determined at the atomic level (Hogle et al., 1985), important steps in the assembly pathway leading to the infectious virus remain obscure.

With the development of a cell-free system allowing the de novo synthesis of poliovirus (Molla et al., 1991, 1993), a new attractive tool emerged to study poliovirus morphogenesis. A major problem in the study of poliovirus assembly is the physical barrier of the cell membrane (and other membranes), which makes it difficult to manipulate intracellular processes. For the isolation of poliovirus intermediates from an infected cell, an extraction procedure is required. This extraction procedure might alter the conformation of some intermediates, which can lead to artefacts (Rombaut et al., 1982, 1989). In the cell-free system, there is direct access to the virus replication machinery, and no such extraction procedure is required. Moreover, the relationship between poliovirus intermediates in an infected cell is not always easy to elucidate. Pulse–chase experiments carried out in infected cells may be difficult to interpret. In the cell-free system, in contrast, such a relationship between particles might be easier to study. Radiolabelled poliovirus intermediates (isolated from infected cells) can be added to a cell-free system and their behaviour can be followed.

In this paper, evidence is presented that 14S subunits are indeed key assembly intermediates and that they interact, when added to the cell-free system, with de novo-synthesized viral RNA (vRNA) and assemble into virions.

A number of potential intermediates in poliovirus assembly have been identified (5S protomer, 14S subunits or pentamers, 45S particles, procapsids and provirions). Of these particles, the 14S subunits and procapsids (empty capsids isolated from infected cells) have been well characterized.

The 14S subunits or pentamers, which have the composition (VP0–VP1–VP3)₅, are considered to be key assembly intermediates. They have been found in poliovirus-infected cells and in cells infected with other picornaviruses. Poliovirus 14S subunits are capable of self-association to form empty capsids in vitro (outside the cell). Finally, by using a temperature block, they can be made to accumulate in infected cells. After removal of the block, they can be chased into mature virions in the infected cell (reviewed by Boeyé & Rombaut, 1992; Hellen & Wimmer, 1992; Putnak & Phillips, 1981; Rueckert, 1996). Procapsids, which are empty shells of composition [(VP0–VP1–VP3)₅]₁₂, were the first poliovirus-induced particles to be isolated from infected cells. Procapsids, which are sensitive to temperature and pH, have been regarded as poliovirus intermediates because they can be made to accumulate in infected cells by using a guanidine block and, after removal of the drug, can be chased into mature virions (Jacobson & Baltimore, 1968). However, procapsids are not universally present in picornavirus-infected cells. Moreover, procapsids can easily dissociate into 14S subunits, a property that opens the possibility that they can act as a reservoir for 14S subunits (reviewed by Boeyé & Rombaut, 1992; Hellen & Wimmer, 1992; Putnak & Phillips, 1981; Rueckert, 1996).

For our experiments utilizing the cell-free system, radiolabelled 14S subunits and procapsids were isolated from infected cells as described previously (Rombaut et al., 1983). Briefly, HeLa cells were infected with poliovirus type 1, Mahoney, at an m.o.i. of 50. After shut-off of host cell protein synthesis, [³⁵S]methionine was added and metabolic labelling of viral proteins was allowed for 1 h. Cells were then collected and disrupted by freeze–thawing and a cytoplasmic cell extract was prepared following the addition of NP-40. The 14S subunits and procapsids were isolated by sucrose gradient ultracentrifugation. To avoid any possible denaturation of the 14S subunits or the sensitive procapsids, the isolation of particles was carried out at low pH (6.5) and temperature (4 °C) (Rombaut et al., 1983). Determining their polypeptide composition (VP0–VP1–VP3) and their antigenicity (see below) characterized the 14S subunits (11000 c.p.m./µl) and the 74S procapsids (14000 c.p.m./µl), and they were stored at –80 °C until use.

Neutralizing MAbs have been raised against native poliovirus (N1 and N2 MAb) and heat-denatured poliovirus (virions heated for 20 min at 56 °C; H MAb). These neutralizing MAbs have been mapped to interact with four neutralization antigenic sites (Boeyé & Rombaut, 1992; Diamond et al., 1985; Rombaut et al., 1983) at the surface of the virion (Page et al., 1988). Of these neutralizing MAbs, only a subset, designated MAbs N1, recognized 14S subunits, whereas other MAbs, designated N2, do not (Rombaut et al., 1983, 1990 a). N1 MAbs interact with neutralization antigenic sites 1, 2 and 3A; N2 MAb with site 3B. Interestingly, H antibodies also bind 14S subunits (Rombaut et al., 1983). The antigenicity of batches of both 14S subunits and 74S procapsids prepared here was determined with a protein A-aided immunoprecipitation assay (Vrijsen et al., 1983). As expected, 14S subunits had [N1,H] antigenicity and 74S procapsids were [N1,N2] antigenic, being indistinguishable from virions (results not shown). Finally, we determined whether the 14S subunits prepared here were able to self-assemble. It was shown previously that 14S subunits can self-assemble into empty capsids at 37 °C, provided that their concentration is above a certain threshold (Rombaut et al., 1991). As expected, 14S subunits at 1.9 nM protein self-assembled into empty capsids, whereas no self-assembly occurred when 14S subunits were diluted 1:5 (results not shown). It was also shown that procapsids were able to dissociate into 14S subunits at alkaline pH (Rombaut et al., 1982).

The cell-free system for de novo synthesis of poliovirus described by Molla et al. (1991, 1993) is the first system in which infectious virions can be synthesized when programmed only with vRNA. We were interested to study the behaviour of our radiolabelled 14S subunits and 74S procapsids in this cell-free extract of uninfected HeLa cells. The system is assembled from three components: (i) an S-10 extract of uninfected HeLa cells, (ii) salts, tRNAs and NTPs and (iii) an energy-generating system. The mixture is then programmed with poliovirus RNA. The detailed composition of the cell-free system will be described in a subsequent publication (Y. Verlinden, A. Cuconati, E. Wimmer and B. Rombaut, unpublished results). Two new components were added to the cell-free system: reticulocyte lysate and pirodavir (a capsid-binding compound; Janssen Research Foundation). Both components have been shown to increase the efficiency of virus production in a cell-free system (Y. Verlinden, A. Cuconati, E. Wimmer and B. Rombaut, unpublished results). Briefly, the final volume of the reaction mixture was 25 µl. It included 17.5 µl 'master-mix' (55 % cytoplasmic HeLa extract, 1.5 mM ATP, 296.5 µM GTP, 284 µM CTP and UTP, 14.8 mM creatine phosphate, 37 µg/ml creatine phosphokinase, 28 mM HEPES–KOH, pH 7.4, 37 µg/ml calf liver tRNA, 3 % amino acid mixture, 370 µM spermidine, 545 µM magnesium acetate, 1.31 mM magnesium chloride and 159 µM potassium acetate), 1 µl 250 µg/ml pirodavir, 2.5 µl rabbit reticulocyte lysate, 1 µl vRNA (approx. 600 ng/µl vRNA extracted from poliovirus type 1, strain Mahoney) and radiolabelled 14S subunits or 74S procapsids. The mixture was incubated at 34 °C for 15 h and the template RNAs were subsequently destroyed by treatment with RNase.

We first studied the behaviour of the 14S subunits. In all experiments, 3 µl of 14S subunits (33000 c.p.m.) was added to the cell-free system. Three samples were prepared: (i) a control, i.e. addition of vRNA to the cell-free system was omitted (replaced by 1 µl RNase-free water) and radiolabelled 14S subunits were added 1 h after this mock programming; (ii) the cell-free system was programmed with vRNA and radiolabelled 14S subunits were added 1 h post-programming (p.p.); and (iii) the cell-free system was programmed with vRNA and 14S subunits were added 8 h p.p. All three samples were incubated at 34 °C for 15 h and cooled on ice and samples were layered onto a 15–30 % sucrose gradient and centrifuged for 135 min at 180000 g_av at 4 °C in a Centrikon TST 41.14 rotor, in order to separate possible assembled 74S procapsids from virions. Results are presented in Fig. 1. In the absence of programming with vRNA, neither procapsids nor virions were assembled (Fig. 1 A). The initial 14S subunits remained 14S subunits. This is not unexpected, as the final 14S subunit concentration in the cell-free system was below the threshold concentration required for self-assembly (Rombaut et al., 1991). In contrast, when the cell-free system was programmed with vRNA, 14S subunits were assembled into either 74S procapsids (when 14S subunits were added 1 h p.p.; Fig. 1 B) or 74S procapsids and virions (when 14S subunits were added 8 h p.p.; Fig. 1 C). This suggests that the extraneous radiolabelled 14S subunits interact with unlabelled newly synthesized endogenous 14S subunits to form either 74S procapsids (addition of 14 S subunits 1 h p.p.) or virions, after interaction with vRNA (addition of 14S subunits 8 h p.p.) (see below for infectivity data for the newly formed virions).

Fig. 1. Separation of assembled procapsids from virions by sucrose-gradient centrifugation. Radiolabelled 14S subunits (3 µl; 33000 c.p.m.) were added to the cell-free system (25 µl). Incubation was for 15 h at 34 °C. After incubation, samples were cooled on ice and possible assembled procapsids were separated from virions as described in the text. The cell-free system was mock-programmed (no addition of vRNA) and extraneous 14S subunits were added 1 h p.p. (A) or the cell-free system was programmed with vRNA and extraneous 14S subunits were added 1 h p.p. (B) or 8 h p.p. (C).

The reason why 14S subunits are assembled into procapsids when they are added to the cell-free system 1 h p.p. and yet they can assemble into virions when they are added to the cell-free system 8 h p.p. relates to the properties of the cell-free system. In the cell-free system described here, viral protein synthesis, formation of capsid proteins and endogenous 14S subunits is fast, reaching a maximum at 2–3 h p.p. Viral RNA synthesis, however, starts much later and only reaches a maximum at 8–9 h p.p. (Y. Verlinden, A. Cuconati, E. Wimmer and B. Rombaut, unpublished results). This means that, if extraneous 14S subunits are added to the cell-free system 1 h p.p., they interact with the available endogenous 14S subunits and assemble into 74S procapsids, but they can not associate with newly made genomic RNA. This is also exactly what happens if extraneous 14S subunits are added to an infected cell extract, in which no vRNA synthesis is found (Rombaut et al., 1991). However, if extraneous 14S subunits are added to the cell-free system 8 h p.p., a time-point at which newly synthesized viral genomes are available, extraneous and endogenous 14S subunits can interact with the vRNA and assemble into virions. This recalls the hypothesis that only newly synthesized vRNA can be packed into capsids (Molla et al., 1991). Further support for this hypothesis was found in the following experiment. A cell-free system was programmed with vRNA and 2 mM guanidine–HCl was added. The radiolabelled 14S subunits were added 8 h p.p. (as in Fig. 1 C). After further incubation, the particles were separated as described in Fig. 1. Results were identical to those described in Fig. 1(B); i.e. the 14S subunits assembled into 74S procapsids and no virions were formed. This again demonstrates that active vRNA synthesis is required for virion assembly.

The antigenicity of the assembled empty capsids and virions was also determined by protein A-aided immunoprecipitation. MAbs N1 (against sites 2 and 3A) and N2 (against site 3B) were used, as well MAb H (Minor et al., 1986; Rombaut et al., 1990 a). As shown previously, the radiolabelled 14S subunits had [N1,H] antigenicity (see Table 1). The assembled empty capsids (Fig. 1 B) and virions (Fig. 1 C) had [N1,N2] antigenicity, which means that the H epitopes are lost during assembly and that a new neutralizing antigenic site is acquired (N2 or 3B). Consequently, the empty capsids and virions assembled in the cell-free system possess the same antigenic sites as procapsids or virions found in infected cells (Rombaut et al., 1990 a). This again proves that assembly in the cell-free system is identical to the pathway found in vivo. Finally, the infectivity of the virion peak (Fig. 1 C) was determined by a classical plaque assay. The titre of the virions was 7.2x10⁶ p.f.u./ml. The particle to p.f.u. ratio was also determined and found to be approx. 200. The same values were found for endogenous assembled virions.

In a second set of experiments, radiolabelled 74S procapsids isolated from infected cells were used. They were again added to the cell-free system and incubations were performed exactly as described for 14S subunits. However, in none of the experiments was there any shift of radioactivity (results not shown). This confirms our previous hypothesis that empty capsids are dead-end particles. They are formed only when there is a temporary or continuing lack of vRNA (see also Fig. 1 B).

Table 1. Antigenicity of subviral particles

Protein A-aided immunoprecipitation in microtitration plates (Vrijsen et al., 1983) was performed with three MAbs recognizing the main neutralizing sites of poliovirus type 1 and one non-neutralizing antibody specific for H antigen (for details and description of the antibodies, see Rombaut et al., 1990 a). Sedimentation profiles of the different particles are shown in Fig. 1 (A) (14S), (B) (74S) and (C) (160S).

Unfortunately, we were unable to repeat these experiments with purified preparations of the capsid precursor P1 or purified protomers VP0–VP3–VP1, since these polypeptides or their aggregate cannot be isolated in sufficient quantities (Boeyé & Rombaut, 1992).

Previous studies had already yielded substantial evidence that 14S subunits are 'key' assembly intermediates in poliovirus morphogenesis: (i) they are always present in cells infected with poliovirus and other picornaviruses (Putnak & Phillips, 1981); (ii) 14S subunits can assemble into empty capsids (Boeyé & Rombaut, 1992; Putnak & Phillips, 1981); (iii) by using a temperature block, they can be made to accumulate in infected cells and, after removal of the block, they can also be chased into mature virions in the infected cell (Rombaut et al., 1990 b); (iv) 14S subunits display the propensity to associate vRNA (Nugent & Kirkegaard, 1995); and (v) they are associated with the poliovirus replication complex and can be cross-linked to vRNA by UV irradiation (Pfister et al., 1995). In this paper, we have shown unequivocally that 14S subunits interact with newly synthesized vRNA and are assembled into virions.

The role of procapsids in poliovirus morphogenesis, on the other hand, has always been controversial. Although they are found in poliovirus-infected cells and respond to the drug guanidine as if they are precursors (Jacobson & Baltimore, 1968), they are not universally present in cells infected with picornaviruses and are even not present in some tissue culture cells infected with poliovirus (Ghendon et al., 1972). Moreover, they display no propensity to associate vRNA (Nugent & Kirkegaard, 1995) and, finally, they are not associated with the poliovirus replication complex and cannot be cross-linked to vRNA by UV irradiation (Pfister et al., 1995). Our data indicate that procapsids are only assembled when vRNA is depleted. Procapsids are consequently dead-end products.

The authors are grateful to Alfons De Rees, Monique De Pelsmacker, Solange Peeters, Frank Van Der Kelen, Bart Verheyden, Sandra Lauwers, Stephane Steurbaut and Raf Vrijsen for their excellent technical assistance and critical advice. This work was supported in part by grants of the National Institutes of Health (USA) to E. Wimmer.

Boeyé, A. & Rombaut, B. (1992). The proteins of poliovirus. Progress in Medical Virology 39, 139–166.

Diamond, D. C., Jameson, B. A., Bonin, J., Kohara, M., Abe, S., Itoh, H., Komatsu, T., Arita, M., Kuge, S., Nomoto, A., Osterhaus, A. D. M. E., Crainic, R. & Wimmer, E. (1985). Antigenic variation and resistance to neutralization in poliovirus type 1. Science 229, 1090–1093.

Ghendon, Y., Yakobson, E. & Mikhejeva, A. (1972). Study of some stages of poliovirus morphogenesis in MiO cells. Journal of Virology 10, 261–266.

Hellen, C. U. T. & Wimmer, E. (1992). Maturation of poliovirus capsid proteins. Virology 187, 391–397.

Hogle, J. M., Chow, M. & Filman, D. J. (1985). Three-dimensional structure of poliovirus at 2.9 Å resolution. Science 229, 1358–1365.

Jacobson, M. F. & Baltimore, D. (1968). Morphogenesis of poliovirus. I. Association of the viral RNA with coat protein. Journal of Molecular Biology 33, 369–378.

Kitamura, N., Semler, B. L., Rothberg, P. G., Larsen, G. R., Adler, C. J., Dorner, A. J., Emini, E. A., Hanecak, R., Lee, J. J., van der Werf, S., Anderson, C. W. & Wimmer, E. (1981). Primary structure, gene organization and polypeptide expression of poliovirus RNA. Nature 291, 547–553.

Minor, P. D., Ferguson, M., Evans, D. M. A., Almond, J. W. & Icenogle, J. P. (1986). Antigenic structure of polioviruses of serotypes 1, 2 and 3. Journal of General Virology 67, 1283–1291.

Molla, A., Paul, A. V. & Wimmer, E. (1991). Cell-free, de novo synthesis of poliovirus. Science 254, 1647–1651.

Molla, A., Paul, A. V. & Wimmer, E. (1993). Effects of temperature and lipophilic agents on poliovirus formation and RNA synthesis in a cell-free system. Journal of Virology 67, 5932–5938.

Nugent, C. I. & Kirkegaard, K. (1995). RNA binding properties of poliovirus subviral particles. Journal of Virology 69, 13–22.

Page, G. S., Mosser, A. G., Hogle, J. M., Filman, D. J., Rueckert, R. R. & Chow, M. (1988). Three-dimensional structure of poliovirus serotype 1 neutralizing determinants. Journal of Virology 62, 1781–1794.

Pfister, T., Egger, D. & Bienz, K. (1995). Poliovirus subviral particles associated with progeny RNA in the replication complex. Journal of General Virology 76, 63–71.

Putnak, J. R. & Phillips, B. A. (1981). Picornaviral structure and assembly. Microbiological Reviews 45, 287–315.

Rombaut, B., Vrijsen, R., Brioen, P. & Boeyé, A. (1982). A pH-dependent antigenic conversion of empty capsids of poliovirus studied with the aid of monoclonal antibodies to N and H antigen. Virology 122, 215–218.

Rombaut, B., Vrijsen, R. & Boeyé, A. (1983). Epitope evolution in poliovirus maturation. Archives of Virology 76, 289–298.

Rombaut, B., Vrijsen, R. & Boeyé, A. (1989). Denaturation of poliovirus procapsids. Archives of Virology 106, 213–220.

Rombaut, B., Boeyé, A., Ferguson, M., Minor, P. D., Mosser, A. & Reuckert, R. (1990 a). Creation of an antigenic site in poliovirus type 1 by assembly of 14S subunits. Virology 174, 305–307.

Rombaut, B., Vrijsen, R. & Boeyé, A. (1990 b). New evidence for the precursor role of 14S subunits in poliovirus morphogenesis. Virology 177, 411–414.

Rombaut, B., Foriers, A. & Boeyé, A. (1991). In vitro assembly of poliovirus 14S subunits: identification of the assembly promoting activity of infected cell extracts. Virology 180, 781–787.

Rueckert, R. R. (1996). Picornaviridae: the viruses and their replication. In Fields Virology, 3rd edn, pp. 609–654. Edited by B. N. Fields, D. M. Knipe & P. M. Howley. Philadelphia: Lippincott–Raven.

Vrijsen, R., Rombaut, B. & Boeyé, A. (1983). A simple quantitative protein A micro-immunoprecipitation method; assay of antibodies to the N and H antigens of poliovirus. Journal of Immunological Methods 59, 217–220.

JGV Direct table of contents

© 2000 SGM

This article is now available in the November 2000 print issue of JGV (vol. 81, 2751–2754). The complete issue of the journal may be seen in electronic form on JGV Online.