• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of jvirolPermissionsJournals.ASM.orgJournalJV ArticleJournal InfoAuthorsReviewers
J Virol. Feb 1998; 72(2): 1297–1307.
PMCID: PMC124608

A Protein Linkage Map of the P2 Nonstructural Proteins of Poliovirus


The yeast two-hybrid system was used to catalog all detectable interactions among the P2 nonstructural cleavage products of poliovirus type 1 (Mahoney). Evidence has been obtained for specific associations among 2Apro, 2BC, 2C, and 2B. Specifically, 2Apro can interact with itself and 2BC and its cleavage products (2B and 2C) interact in all possible combinations, with the exception of 2C/2C. Detected interactions were confirmed in vitro by a glutathione S-transferase pulldown assay, which allowed us to detect 2C/2C association. trans-dominant-negative mutants of 2B (K. Johnson and P. J. Sarnow, J. Virol. 65:4341–4349, 1991) were examined and were found to retain interaction with wild-type 2B, perhaps reflecting a need for 2B multimerization in viral RNA replication. The multimerization of 2B was examined further by screening a mutagenized library for 2B variants that have lost the ability to bind wild-type 2B. The screen identified two nonconservative missense mutations within a central hydrophobic region, as well as truncations and frameshifts that implicate the C terminus in homointeraction. Introduction of the missense mutations into the genome of the virus conferred a quasi-infectious phenotype, an observation strongly suggesting that the 2B/2B interaction is required for replication of the viral genome.

The reproduction of a positive-sense RNA genome presents special problems for viruses in that the cell does not contain an RNA replication mechanism that can be subverted during the viral life cycle. Unlike DNA viruses, the RNA viruses must establish an RNA replication pathway under the limitations of an error-prone RNA replication process, giving rise to the necessity for small genomes (59). One answer to these constraints is the evolution of minimized open reading frames specifying polyproteins yielding multifunctional protein products. The well-studied positive-sense RNA animal viruses, the Picornaviridae, encode a single polyprotein that is proteolytically processed to yield the final cleavage products plus several cleavage intermediates. Interestingly, some of these intermediates have functions distinct from those of their products (27, 43, 59). In this manner, minimal coding capacity is maximized. However, the intertwining of different functions for the same polypeptide has rendered study of the contribution of a polypeptide chain to replication difficult.

Polyprotein cleavage and the replication cycle have been thoroughly studied for poliovirus, a prototype of Picornaviridae (59). Polioviral RNA replication occurs in association with membranous vesicles which are a prominent feature of the productively infected cell (10, 13). Biochemical and microscopic studies have demonstrated the association of the vesicles with the cleavage products of the P2 precursor, all of which have been shown to be required for the replication process (9, 11, 52, 59). Indeed, both structural and nonstructural viral proteins can be coisolated in the membranous environment of the replication complex, indicating the possible formation of multimolecular complexes (18). P2 itself maps to the central region of the polyprotein (Fig. (Fig.1A)1A) and is processed to yield three different end products (2Apro, 2B, and 2C) and one long-lived precursor (2BC) (27, 59). 2Apro is a cysteine proteinase with a catalytic triad reminiscent of those of serine proteases (28). The enzyme catalyzes tyrosine-glycine cleavage at the junction of the P1 and P2 precursors (54). 2Apro is also involved in the shutoff of host cell translation (51), has an undefined role in the initiation of cap-independent translation of the viral genome (26), and may also function directly in the replication of the genome (38, 41, 62). The functions and characteristics of 2BC and its cleavage products, 2B and 2C, are less well defined. 2C contains a functional nucleoside triphosphate-binding and hydrolysis domain (39, 46) which is reminiscent of a helicase motif (24) and which is required for RNA replication (40, 52a). 2C also binds nucleic acids (45) and contains a zinc finger motif whose integrity is essential for RNA synthesis (44). Exogenous expression of 2B in mammalian cells has been associated with a strong block in secretory transport, permeabilization of the plasma membrane (1, 17, 56, 58), and dissassembly of the Golgi apparatus (47). Similar studies of 2BC and 2C reveal the accumulation of vesicles reminiscent of those seen in infected cells in the case of 2BC (4, 15) and less-recognizable membrane rearrangements (invaginations of the endoplasmic reticulum) by 2C (15). Whether vesicle induction is absolutely required for efficient replication remains an unanswered question, as is the precise function of 2BC and its cleavage products.

FIG. 1
Gene organization and polyprotein processing of poliovirus. (A) Structure of the genomic RNA of poliovirus and its coding region. The 5′ terminus is covalently linked to the VPg peptide (3B). The highly structured 5′ nontranslated region ...

In this study, our aim was to determine if the P2 products can interact with each other and whether some of these interactions are required in genomic replication. We have constructed a Saccharomyces cerevisiae two-hybrid “protein linkage map” of the P2 region as part of a larger project that would ultimately include all the cleavage products of the polioviral polyprotein. On the basis of what was achieved with the protein linkage map of bacteriophage T7 (7), we expected that cataloging the protein-protein interactions among the P2 products would shed new light on previous observations and suggest new roles in replication. Our results indicate that 2BC, 2B, and 2C multimerize in a network of interactions which is consistent with previous observations of trans-dominance by viral mutants of 2B (29). A genetic analysis of the 2B/2B interaction strongly suggests that 2B multimerization is required for the occurrence of viral replication.


Construction of two-hybrid expression vectors and screening for interaction.

The coding sequences of poliovirus type 1 (Mahoney) (PVM) 2Apro, 2B, 2C, and 2BC were amplified by PCR with Pfu polymerase (Stratagene) by using a set of primers meant to introduce an XmaI site at the upstream (5′) end and two stop codons plus a SalI site at the downstream (3′) end of each coding sequence. The basic design of all upstream primers was 5′-GCG-XmaI-G-15 nucleotides (nt) of positive-sense 5′ coding sequence-3′. The basic design of all downstream primers was 5′-GCG-SalI-TTATCA-15 nt of negative-sense 3′ coding sequence-3′. These were used to amplify the appropriate coding sequences from viral cDNA vector pT7PVM (14, 40), which contains unique XhoI and MluI sites in the 2C coding region and a StuI site in the 2B N terminus coding region. The PCR products were precipitated, digested with XmaI and SalI, gel purified by the GeneClean II method (Bio101, Inc.), and ligated into pGAD.GH (Gal4 activation domain, or GAD) (55), pBTM116 (LexA DNA binding protein), and pGBT9 (Gal4 DNA binding domain, or GBD) (6) vectors that had been digested with XmaI and SalI and treated with calf intestinal phosphatase. The 2B PCR product was also ligated into pBTM38, a derivative of pBTM116 in which the ADE2 gene was inserted into the PvuII site. Plasmid DNA was prepared from several subsequent bacterial transformants and screened for the presence of the inserts, and positive clones were sequenced by standard methods at the fusion junction to ensure correct reading frames.

Clones of pGAD.GH and pBTM116 constructs of 2Apro, 2B, 2C, and 2BC were transformed into S. cerevisiae L40-ura3 (48) in every possible pairwise combination. Control plasmids were pGAD.GH-retinoblastoma and GAD empty vector, pBTM116-lamin, and pBTM116-topoisomerase I. This resulted in 28 pairwise transformations which were selected on minimal synthetic plus sucrose (SS) medium lacking Leu and Trp. All two-hybrid media, buffers, and protocols were used as previously described (6, 20). Each plate was subjected to a β-galactosidase filter lift assay to ensure that all or nearly all colonies would show the same activity (6). Paper filters (VWR Scientific) were laid upon the surface of the plate containing transformed colonies, lifted, and frozen for 10 s in liquid N2; upon thawing they were placed on another filter wetted with Z buffer (6) containing 5-bromo-4-chloro-3-indolyl-β-d-galactoside (X-Gal) and incubated at 37°C until the color change was fully developed. From each plate, approximately 12 colonies were then picked and combined into a single streak which was then tested for activity by filter lift, and restreaked onto SS media lacking Leu, Trp, and His to test for HIS3 expression. β-Galactosidase activities in positive pairwise combinations were quantified by colorimetric measurement of Miller units (6) with chlorophenol red-β-d-galactopyranoside and by measurement of absorbance at 574 nm.

The Y153 strain (Table (Table1)1) was transformed in an analogous manner with pGAD.GH and pGBT9 constructs, along with a GAD fusion of an active-site mutant of 2Apro [2Apro(C109S)] (28). Transformants were tested for activity as described above and grown on SS medium lacking Leu, Trp, and His in the presence of 20 mM 3-amino-1,2,4-triazole (an inhibitor of basal levels of HIS3 caused by leaky Gal1-dependent reporter expression) to test for HIS3 reporter expression (6, 20).

Two-hybrid reporter strains and plasmids used

LexA and GAD fusions of 2B mutants 2B(201), 2B(204), and 2B(205) were constructed essentially as described above for wild-type 2B, except PCR products were generated from the full-length cDNAs pT7201 (8), pT7205, and pT72B204, respectively (29).

The mutation carried by the W36 isolate (see Results) was introduced into pBTM116 and pGAD.GH by PCR with coding sequences for 2B and 2BC from pT7PVM(W36) and the appropriate primers as described above, generating LexA and GAD fusions 2B(W36) and 2BC(W36). The integrity of positive clones was confirmed by sequencing the junctions and the mutated codon.

Expression of GST fusion proteins and pulldown assay for interaction.

Glutathione S-transferase (GST) fusion proteins of the P2 cleavage products were cloned and expressed as follows. The pT7PVM vector (14, 40) was used as a template for PCR amplification (with Pfu polymerase) of coding sequences for 2A, 2B, 2C, and 2BC with primers designed to introduce flanking EcoRI and HindIII sites (principally the same design as that described above). The resulting products were digested with EcoRI and HindIII and ligated into the appropriately digested pGEX-KG vector (25) to yield pGEX-KG2A, pGEX-KG2B, pGEX-KG2C, and pGEX-KG2BC.

Positive clones were transformed into the BL-21(DE3) strain of Escherichia coli, and isolated colonies were grown overnight in 2× yeast extract-tryptone (YT) plus ampicillin medium. These were diluted 50-fold and grown in 50 ml of the same medium to an optical density at 600 nm of 1.0, at which point expression was induced with 0.1 mM isopropyl-β-d-galactopyranoside (IPTG) for 4 h at 25°C. Cultures were centrifuged and resuspended in 2.5 ml of ice-cold HBS (20 mM HEPES [pH 7.5], 140 mM NaCl) supplemented with 5 mM 2-mercaptoethanol and 1 μg each of pepstatin A and leupeptin (Sigma) per ml. Cells were lysed by sonication, Triton X-100 was added to a final concentration of 1.5%, and samples were rotated at 4°C for 30 min and centrifuged at 8,000 × g for 10 min. A 50-μl bed volume (bv) of glutathione-Sepharose (GSH) 4B beads (equilibrated in an equal volume of HBS supplemented with 5 mM dithiothreitol and 0.1% Triton X-100; referred to as binding buffer) was added to the supernatant fraction (all bead pipetting was done with cut tips), and the mixture was rotated at 4°C for 45 min. Beads were washed four times with ice-cold binding buffer, resuspended in 50 μl of the same buffer, and stored at 4°C. A 2-μl bv of beads was analyzed by sodium dodecyl sulfate–15% polyacrylamide gel electrophoresis (SDS-PAGE) and Coomassie blue staining to check the expression and recovery of fusion proteins.

To assay for the binding of the GST fusion proteins with 35S-labeled P2 cleavage products, the P2 proteins were translated in vitro in rabbit reticulocyte lysate (Promega) in the presence of 35S-Translabel (ICN Pharmaceuticals) from synthetic transcripts under the control of the encephalomyocarditis virus internal ribosomal entry site (IRES). A total of 1.0 × 106 cpm of trichloroacetic acid-precipitable material (approximately equal to 20 μl of translation reaction mixture) was precleared with a 5-μl bv of empty GSH beads by incubation at 30°C for 20 min, with gentle mixing every 5 min. Following centrifugation for 30 s, the supernatant was diluted with 20 μl of binding buffer and a 10-μl bv of beads plus fusion protein was added. The mixture was incubated for 30 min at 30°C, the supernatant was removed, and an additional 20-μl bv of empty GSH beads was added as a carrier. Beads were washed six times with 1 ml of binding buffer (rotating 10 min at 4°C each time), resuspended in 1× Laemmli buffer, boiled for 5 min, and analyzed by SDS-PAGE and autoradiography.

Construction of a mutagenized GAD-2B library and reverse screening for interaction.

We used buffer conditions derived by Leung et al. (35) to mutagenize the 2B coding sequence by PCR using Taq polymerase (Boehringer Mannheim) at expected misincorporation rates of 0.4, 1.0, and 2.0%. Three different sets of reaction mixtures were cycled 25 times at 95°C for 1 min, 42°C for 1 min, and 72°C for 2 min. Primers used were of the same design as that described above to amplify the 2B sequence. The PCR products were processed as described above and ligated into pGAD.GH. The ligated products were transformed into electrocompetent stocks of E. coli DH10B (GIBCO BRL) by electroporation, and bacterial transformants were grown on selective media. Colonies were counted, scraped into liquid media and grown for 6 h at 37°C. Plasmid DNA was then prepared to obtain the library.

For each library produced at the three different rates, the procedure was repeated and the preparations were combined until each contained >4,700 members. The libraries were then cotransformed into L40-ura3 with pBTM38-2B, and transformants were tested for activity by filter lift assay. For the library constructed at the 0.4% misincorporation rate, the procedure was repeated until >5,000 transformants had been screened. Fifty negative colonies were picked, replated, and retested; 25 of these were then grown in liquid SS medium lacking Leu to facilitate loss of the pBTM38-2B plasmid and plated on low-concentration adenine (10% of normal concentration) SS medium lacking Leu. Colonies of each isolate that lost the plasmid (indicated by a red color) were picked and grown in liquid media, and extracts were prepared and tested for 2B expression by SDS-PAGE and immunoblotting with anti-2B monoclonal antibody. Plasmid DNA was prepared from positive isolates, and the 2B coding region of each plasmid was sequenced by standard methods. Those isolates showing alterations of the amino acid sequence were reintroduced into pGAD.GH and retested for loss of interaction by filter lift and growth on media lacking HIS.

Construction of mutant viral cDNA vectors.

The megaprimer method (16) was used to generate the point mutations seen in isolates W10 and W36 in the corresponding 2B codons of full-length cDNA vector pT7PVM, generating plasmids pT7PVM(W10s) and pT7PVM(W36). Primers that allowed the mutagenized PCR product to be ligated into the XhoI site within the 2C coding region and into the blunt end at the start of the 2B coding region left by StuI digestion were generated. The resulting constructs lost the StuI site and the ensuing amino acid change of isoleucine to leucine at position 2 (I2L) (40); this restored the wild-type amino acid sequence at position 2 of 2B. The LB21 point mutation was introduced by one-step PCR using an upstream primer that encodes that base change into codon 3 of the 2B coding sequence, generating pT7PVM(LB21). To construct an appropriate wild-type control (called pT7PVMwt), we used one-step PCR with the upstream and downstream primers from the above constructions to generate the wild-type sequence at position 3 of the 2B coding sequence.

Mutant viral replicons containing the luciferase gene in place of the P1 region were constructed by ligating the KpnI fragment of the pPV1(M)/Luc construct (36) into the large KpnI fragments (vector background) of the pT7PVMwt, pT7PVM(LB21), pT7PVM(W10s), and pT7PVM(W36) constructs. This resulted in the substitution of the IRES and P1 regions in the full-length viral cDNAs with the IRES and luciferase genes of pPV1(M)/Luc, generating pPV1(M)/Luc(wt), pPV1(M)/Luc(LB21), pPV1(M)/Luc(W10s), and pPV1(M)/Luc(W36).

In vitro translation, transfection, and analysis of viral phenotypes.

Full-length genomic cDNA constructs pT7PVMwt, pT7PVM(W10s), pT7PVM(W36), and pT7PVM(LB21) were linearized with EcoRI and transcribed with T7 polymerase. The resulting RNA was purified, quantified, and translated in cytoplasmic HeLa cell extracts in the presence of 35S-Translabel (ICN Pharmaceuticals) as previously described (42). After overnight incubation, the reaction mixtures were mixed with equal volumes of Laemmli buffer and analyzed by SDS–15% PAGE and autoradiography.

The RNAs were also transfected (in quadruplicate) into subconfluent (80%) monolayers of HeLa R19 by the Lipofectin method (GIBCO BRL). A mock control was exposed to transfection mixture minus RNA. Transfected cultures were incubated in Dulbecco modified Eagle medium (GIBCO BRL) supplemented with 10% fetal bovine serum at 37°C in a 5% CO2 atmosphere. Cells were observed by phase-contrast microscopy for signs of cytopathic effect (CPE). When CPE appeared in 80 to 100% of the cellular monolayer, the culture was frozen at −80°C and subjected to three cycles of freezing and thawing. The resulting supernatants were titrated on confluent (90 to 95%) HeLa R19 monolayers, and the cultures were incubated in nutrient agar overlay for 48 h to determine plaque size and titer by crystal violet staining. One set of plaques was stained with INT-MTT [0.001% p-iodonitrotetrazolium violet–0.005% 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-tetrazolium bromide] (49, 50) at 37°C for 30 min to visualize plaques in viable cultures, and seven plaques each were isolated to purify progeny of PVM(W10s) and PVM(W36). These were used to reinfect HeLa monolayers, and total RNA was extracted from 200-μl aliquots of the resulting supernatants. The 2B sequence from viral RNA was amplified by reverse transcription with avian myeloblastosis virus reverse transcriptase (Promega) and by PCR with Pfu polymerase and was sequenced by standard methods.

The effects of the 2B mutations on viral RNA replication were assayed by transfecting transcript RNAs of all four luciferase-expressing replicons into 35-mm-diameter HeLa R19 monolayers either in the presence or absence of guanidine hydrochloride. The cultures were lysed at different time points with 300 μl of lysis buffer (10 mM Tris-HCl [pH 7.5], 140 mM NaCl, 1.5 mM MgCl2, 0.5% Nonidet P-40), and reporter activity was determined with the luciferase assay system (Promega) on a Gem Optocomp 1 luminometer (MGM Instruments).


Protein linkage within the P2 products.

In an attempt to determine whether protein-protein interactions occur among the polypeptide cleavage products of the P2 region, we have engineered fusions of all P2 products with the GAD and the LexA DNA binding protein specifically for use in LexA-dependent two-hybrid interaction assays (strain L40-ura3; Table Table1).1). Initial studies revealed that the LexA-2Apro fusion by itself was a powerful activator of reporter gene transcription in this system, rendering the construct useless. To circumvent this obstacle, we constructed a fusion of the GBD with 2Apro, for use in Gal4-dependent (Gal1 promoter-containing) two-hybrid assays (strain Y153, Table Table1).1). Subsequently, it became apparent that expression of the GAD-2Apro fusion was most likely toxic in either system, as evidenced by our total failure to recover any viable colonies when this construct was included in transformation experiments. Such toxicity is consistent with the results of previous observations (5, 32). Replacement of the wild-type 2Apro coding sequence with that of a mutant encoding a substitution of the putative active-site cysteine [2Apro(C109S)] (28) resulted in a fusion construct that yielded viable transformants. However, the resulting colonies were still very slow growing compared to those resulting from other transformant pairings.

All possible pairwise combinations of 2BC and its cleavage products were transformed into the L40-ura3 strain, and 2Apro fusion constructs were likewise cotransformed into Y153 with other P2 products. The results of the assays for reporter expression in the L40-ura3 transformation experiments are shown in Fig. Fig.2.2. As is apparent, 2BC and its cleavage products produce a positive signal in every possible combination with the exception of 2C/2C. These interactions resulted in expression of both the lacZ (Fig. (Fig.2A2A and B) and HIS3 (Fig. (Fig.2C)2C) reporter genes. Interestingly, we noted that the GAD-2C fusion failed to give a positive signal in any combination, whereas LexA-2C was positive for interaction with GAD-2B and 2BC (Fig. (Fig.2B).2B). Such “polarity” of two-hybrid interactions is frequently observed (61) and was also noted when the relative amounts of β-galactosidase expression were quantified by measurement of Miller units (Fig. (Fig.2B).2B). For example, the pairing of LexA-2B and GAD-2BC resulted in very weak reporter activity compared to the reciprocal combination.

FIG. 2FIG. 2FIG. 2
Two-hybrid assays for interactions among 2BC and its cleavage products in strain L40-ura3 transformed with LexA and GAD fusions of 2B, 2C, and 2BC. GAD-retinoblastoma (Rb), GAD empty vector (Vector), and LexA-lamin were used as negative controls. (A) ...

In strain Y153, 2Apro is positive for multimerization but negative for interaction with the other P2 products. However, the relative level of activity was very low in the case of 2Apro multimerization (Fig. (Fig.3A),3A), ranging from 2 to 8 Miller units for different sample cultures (Fig. (Fig.3B).3B). We were not able to detect activation of the second reporter gene, HIS3, possibly because the level of expression was too low to allow the transformants to survive on media lacking His under stringent conditions (20 mM 3-amino-1,2,4-triazole; see Materials and Methods). The reason for the weakness of this interaction is not known. It may be related to the use of a mutant sequence of 2Apro in the GAD fusion protein, to an intrinsically weak interaction between 2Apro polypeptide chains, or to an inherent consequence of the two-hybrid assay.

FIG. 3
Two-hybrid assays for interaction of 2Apro with cleavage products of P2 in strain Y153 transformed with GBD and GAD fusions of 2Apro, 2B, 2C, and 2BC. (A) Filter lift assay for lacZ expression. (B) Colorimetric quantitation of the 2Apro multimerization ...

Interaction of putative interacting pairs in vitro.

To obtain corroborating evidence for the interactions detected by two-hybrid analysis, we employed a GST pulldown assay. Fusion proteins of GST with 2Apro, 2B, 2C, and 2BC were generated, expressed in bacteria, and bound to GSH beads. Over the course of several experiments, we noted that the level of recovery of the fusion proteins was highest for 2B and 2C and lowest for 2A. Significant amounts of the proteins were found in the insoluble fractions of the lysed bacterial cells, and in the case of GST-2Apro almost all the expressed protein was insoluble (as assayed by SDS-PAGE and Coomassie blue staining) and only very small amounts (<5%) could be recovered on the beads (data not shown).

Each of the P2 cleavage products was synthesized in vitro in rabbit reticulocyte lysate with 35S-Translabel present. Such reactions were programmed with synthetic transcripts encoding the P2 proteins, under the translational control of the encephalomyocarditis virus IRES. Interactions were assayed by mixing the translation reaction mixtures with the GST fusion protein-bead complexes. 35S-labeled material retained on the beads was visualized by SDS-PAGE and autoradiography. The results are shown in Fig. Fig.4.4. Interactions that were detected by the two-hybrid system could be reproduced with this biochemical assay, although the strengths of positive signals did not necessarily correlate with the corresponding signals from the two-hybrid system. Surprisingly, the 2B/2C interaction, which was relatively strong in yeast, was barely detectable in vitro and the GST-2C/2C pairing was found to be positive in the GST pulldown assay (Fig. (Fig.4A)4A) but negative in the two-hybrid system. The latter observation suggests that the failure to detect a positive 2C/2C signal in yeast was due to the generation of a nonfunctional GAD-2C, a conclusion supported by the lack of interaction in any pairings which involved GAD-2C (in contrast to LexA-2C).

FIG. 4
GST pulldown assay of putative interacting pairs, performed as described in Materials and Methods. Lanes marked “translation products” were loaded with 0.5 μl of in vitro translation reaction mixture with the indicated contents. ...

As was the case for the two-hybrid analysis, the GST pulldown assay detected only a weak interaction signal for the GST-2Apro/2Apro pairing (Fig. (Fig.44B).

Does wild-type 2B multimerize with trans-dominant-negative 2B mutants?

Johnson and Sarnow (29) have reported on the slow-replication phenotype of PVM mutants that contained linker insertions after amino acid 28 of 2B. These mutants, termed 2B(201) (8), 2B(204), and 2B(205) (29) (Fig. (Fig.1B),1B), could not be rescued by complementation. On the contrary, they displayed dosage-dependent negative dominance over the wild-type virus in mixed infections. There are two possible interpretations for the trans-dominant phenotypes of these mutants. (i) An interaction of 2B with a limiting (host or viral) factor is normally required for replication. The mutant 2B then interacts with this factor in competition with the wild-type protein. (ii) The multimerization of wild-type 2B or 2BC is required for replication, and mutant–wild-type heteromeric complexes may be impaired in their normal function. Using the two-hybrid assay described above, we tested the three insertional mutants (29) for interaction with wild-type 2B and found that all three have levels of activity similar to that for the wild-type interaction (Fig. (Fig.5).5). This observation suggests (but does not prove) that the mutant protein exerts trans-dominance by rendering a 2B multimer inactive.

FIG. 5FIG. 5FIG. 5
Two-hybrid tests for the interaction of 2B with trans-dominant-negative mutants in strain L40-ura3 transformed with LexA and GAD fusions of 2B, 2B(201), 2B(204), and 2B(205) (Fig. (Fig.1B).1B). (A) Filter lift assays for lacZ expression. (B) Colorimetric ...

Mutational analysis of 2B multimerization.

In order to test our hypothesis that 2B/2B interactions are essential in poliovirus replication, we constructed a randomly mutagenized library of GAD-2B fusion polypeptides and screened against the wild-type protein for loss of interaction. The ultimate purpose of this “reverse screen” was to identify noninteracting mutants of 2B that could be tested in the context of the 2BC precursor. Moreover, these mutations could be introduced into the genomic cDNA of the virus to ascertain their effects on viral viability.

The coding sequence of PVM 2B contained in infectious poliovirus cDNA was amplified by PCR under conditions previously estimated to allow a misincorporation rate of 0.4% (roughly 1 base change per 291 nucleotides or per 2B coding sequence) (35). Interestingly, the use of higher mutagenesis rates (up to 2%) resulted in libraries that had completely lost interaction with the wild-type protein, probably indicating a high degree of sensitivity of the 2B protein structure to changes in its primary sequence. A library (termed GAD-2BM) in which 17% of the transformed yeast colonies were negative for the 2B/2B interaction was finally selected. Then, 25 isolates were tested for GAD-2BM expression by Western blotting. Of these, 9 were found to be expressing either full-length 2BM or apparently truncated products. Plasmid DNA was prepared from all nine and, following amplification of the DNA in bacteria, the 2BM coding region was sequenced to yield the results outlined in Fig. Fig.11B.

The reverse screening identified one frameshift (LB5), two missense (point) mutations (LB21 and W36), and two nonsense (amber and opal) mutations (W6 and LB3). One isolate (W10) carried both a frameshift and a missense change. In addition, three isolates carried either wild-type sequences or a silent mutation; we assumed that these must have lost the ability to activate transcription through fortuitous changes in the GAD sequence that were unrelated to our library construction, and they were not studied further. To determine whether this was also the case for the other GAD-2BM isolates, the 2BM coding sequences were reintroduced into the GAD vector and reassayed for interaction. Concurrently, we also separated the missense and frameshift mutations carried by W10, thereby producing separate constructs, W10s and W10f, respectively. The results are presented in Fig. Fig.6.6. After reconstruction, the point mutation carried by LB21 had no effect on 2B multimerization, and the shortest truncation of 2B (LB3) retained some activity, while the rest of the reconstructed 2BM isolates were negative for interaction.

FIG. 6
Two-hybrid assays for the interaction of wild-type 2B with mutagenized library isolates that have been reconstructed into pGAD.GH by using strain L40-ura3. The expression of lacZ and HIS3 reporter genes was assayed by filter lift and selection on media ...

The results of the reverse screen indicate that the C-terminal half of the protein is integral to the multimerization, either because it harbors the contact sites or because it is required for the overall structural integrity of 2B. The isolation of two nonconservative point mutations (carried by W10s and W36) in a central region that is predicted to form a hydrophobic β-sheet suggests that the integrity of this specific region is also required for multimerization (Fig. (Fig.11B).

Linkage analysis of a noninteracting 2B mutant.

The mutation carried by W36 (I53N) as a GAD fusion was introduced into LexA-2B, to determine whether the mutant 2B could interact with itself or with the other 2BC products in either polarity. The results of these interaction tests are shown in Fig. Fig.7A.7A. The 2B(W36) mutant does not multimerize and, in the context of this assay, fails to interact with any 2BC product. These results suggest that determinants for multimerization within 2B are also required for interaction with 2C and 2BC.

FIG. 7
Two-hybrid assays for the interaction with mutant 2B and 2BC. (A) Filter lift assays for the interaction of 2B constructs carrying the W36 mutation (I53N) with 2B, 2C, and 2BC, and for the multimerization of 2B(W36). The multimerization of wild-type 2B ...

We also introduced the W36 mutation into GAD and LexA fusion constructs of 2BC to determine the contribution of the 2B moiety to interactions involving the 2BC precursor. The resulting set of positive pairings (Fig. (Fig.7B)7B) indicates, surprisingly, that the mutant 2BC molecule retains an interaction profile similar to that of the wild-type protein. An exception is the negative signal produced by the LexA-2B–GAD-2BC(W36) pairing, which is in contrast to the signal produced by the corresponding wild-type interaction (Fig. (Fig.2A).2A). The most likely reason for this result may be the strength of the reporter signal. The wild-type pairing is one of the weakest detected in the linkage map (as opposed to the reciprocal pairing) (Fig. (Fig.2B).2B). Therefore, it is possible that the presence of the mutation in the 2BC moiety reduces the overall strength of the interaction between LexA-2B and GAD-2BC(W36) below the limits of detectability, resulting in the polarity addressed above. The quantitative measures of β-galactosidase expression in the interactions with 2BC(W36) support this explanation, since they were significantly lower than the corresponding values obtained with wild-type 2BC (data not shown).

The overall scoring of pairings involving 2B(W36) and 2BC(W36) (summarized in Fig. Fig.7C)7C) suggests that the positive interactions that include only 2BC(W36) are only partially dependent on the presence of the wild-type 2B moiety. This conclusion expands the possible role of 2C in these interactions.

Genetic analysis of noninteracting mutants of 2B.

To determine whether the substitution mutations carried by the LB21, W10s, and W36 isolates affect genomic replication, we constructed mutant and wild-type polioviral replicons in which the P1 region was replaced with the firefly luciferase gene. The accumulation of maximal levels of the luciferase reporter protein in cell cultures transfected with transcript RNAs is indicative of genome replication (2, 3, 36), since most of the viral protein in the infected cell is synthesized from newly made RNA. Transfection of the wild-type and LB21 replicons resulted in roughly equivalent levels of luciferase activity, whereas the replicons harboring the W10s and W36 mutations replicated at extremely low or nonexistent levels (Fig. (Fig.8).8). The inclusion of 2 mM guanidine hydrochloride in the culture media resulted in essentially equally low levels of reporter activity among all replicons, an observation demonstrating that the luciferase activities of W10s and W36 replicons in the absence of the drug are indicative of a block in replication. Therefore, we conclude that the mutations in 2B which block multimerization also severely reduce viral RNA replication.

FIG. 8
Luciferase activity in cell cultures transfected with luciferase-expressing viral replicons encoding either the wild-type protein, LB21, W10s, or 2B(W36). Following transfection, cell cultures were maintained either in the absence or presence of 2 mM ...

We then engineered the 2B mutations into the viral genome to determine their effects on viability. Genomic cDNA clones encoding these three mutations and the wild-type protein sequence were transcribed with T7 polymerase, and the resulting RNAs were translated in vitro in a HeLa cytoplasmic extract in the presence of 35S-Translabel (42). SDS-PAGE analysis revealed efficient translation and proteolytic processing patterns (Fig. (Fig.9).9). A slight anomaly in processing was observed in the case of W10 and W36, namely, a slight accumulation of precursor 3BCD (14), and some reduction in the quantity of P3 processing products (3C, 3D, and 3AB) compared to those for the wild type and LB21 was also observed. However, all structural and nonstructural proteins can be easily identified among the products of all four reactions.

FIG. 9
SDS–15% PAGE analysis of in vitro-translated full-length transcript RNAs harboring the LB21, W10s, and W36 mutations, as compared to the wild type. The “Marker” lane consists of the (35S-Translabel) pulse-labeled products ...

Transfection of all RNAs into HeLa R19 cells resulted in drastic differences in the onset of CPEs (Table (Table2).2). For cultures transfected with the wild-type and LB21 RNAs, CPE was observed at between 23 and 25 h and between 21 and 24 h, respectively. Interestingly, LB21 genomic RNA reproducibly produced CPE slightly earlier than the wild-type standard, a phenomenon that we cannot explain at present. The W10s and W36 mutants displayed CPE at 90 to 97 h and at 60 to 66 h, respectively. Of quadruplicate cultures, three transfected with W10s produced CPE, while all four transfected with W36 produced CPE.

Genetic analysis of viral constructs encoding missense mutations from isolates LB21, W10s, and W36.

The cell culture supernatants were examined by a plaque assay, and we observed that the plaque morphologies of LB21, W10, and W36 were indistinguishable from those of the wild-type virus (Table (Table2).2). We also observed equivalent titers in one-step growth experiments (data not shown). The single W10 culture which did not give rise to CPE did not yield any plaques. Seven isolates each of W10s and W36 were plaque purified and amplified once in cell culture. Viral RNAs were then isolated from each culture, and their 2B coding sequences were analyzed. In all cases, the mutated codons of W10 and W36 mutant RNAs had reverted to the wild-type genotype, an observation indicating that the W10s and W36 mutants exhibited a quasi-infectious (qi) phenotype (22).

We have examined the possibility that the observed viruses were wild-type contaminants by repeating the transfections under nutrient agar overlays, reasoning that contamination would give rise to isolated plaques at the same time as (but at a much lower titer than) the wild-type control; we never observed plaques generated by W10s and W36 in such assays, which involved incubation until the monolayers began to lose viability (110 h in both mock and RNA-containing samples). This observation indicates that infectious particles were generated too late in the incubation to yield visible plaques (results not shown). On the other hand, reversion yielding a very small number of wild-type genotypes would yield infectious particles that can readily spread in liquid media, allowing a detectable onset of CPE.


Protein-protein interactions among the cleavage products of the P2 precursor.

The two-hybrid system (21) has recently been used to construct a protein linkage map of bacteriophage T7 in an attempt to catalog all detectable protein-protein interactions among the gene products of the virus (7). Using an analogous strategy, we have carried out experiments to detect protein-protein interactions among the cleavage products of the poliovirus polyprotein specifying the nonstructural proteins. These include some cleavage precursors that may have functions distinct from those of their cleavage products. Previously, a protein linkage map of the products of the P3 precursor was established by using a semi-inducible version of the two-hybrid system (60). In this work, protein linkage mapping among the P2 products was carried out by using two widely adopted versions of the two-hybrid system in which the fusion proteins are expressed constitutively. Our results strongly suggest that functional interactions occur among 2BC and its cleavage products. Moreover, 2Apro may bind to itself.

Interactions detected by two-hybrid pairings were reconstituted in vitro by a GST pulldown assay (Fig. (Fig.4).4). The strength of the interaction signal did not always correspond to that from the two-hybrid observations, as in the case of the 2C/2B interaction (see Results). We also found one interaction (2C/2C) which we were not able to detect in yeast. At present, we cannot explain the variances between the results obtained with the two systems, but these observations underline the need for alternative experimentational approaches in any study designed to detect protein-protein interactions. It is important to note that two-hybrid measurements of interaction strength are not always in correspondence with biochemically determined values, presumably because many different variables affect the level of a two-hybrid signal (i.e., polarity of interaction, stability of fusion proteins, and promoter strength, etc.) (19). Moreover, experiments carried out with different cell preparations (but a constant set of plasmids) may not necessarily produce the same levels of interaction signal. Interestingly, the weakness of the 2C/2B interaction in vitro and the weak signal obtained for the 2B/2BC interaction suggest that the 2BC multimerization is principally due to the 2C/2C interaction, a conclusion supported by two-hybrid studies of 2BC mutants (see below).

The observed interactions from pairings of two-hybrid fusion constructs and from GST pulldown assays can be divided into two classes (Fig. (Fig.10),10), which may be relevant to the functions of the P2 products during viral RNA replication. Class I describes the formation of intermolecular complexes that can be further divided into homomultimeric interactions (2B/2B, 2C/2C, 2BC/2BC, and 2Apro/2Apro) and heteromultimeric interactions (2B/2BC and 2C/2BC). Class II consists of a mode of interaction (2C/2B) that may, in fact, reflect the intrinsic affinity between two separate domains within the uncleaved precursor. Although all two-hybrid interactions are intermolecular, the separate 2C and 2B hybrids may be interacting in a manner which normally exists within the 2BC polypeptide, reflecting intramolecular binding between distinct domains. Such intramolecular associations have been detected in the protein linkage map of bacteriophage T7, on which basis it was suggested that the two-hybrid system may aid in the elucidation of protein folding (7). Of course, the 2C/2B interaction may also be intermolecular in the infected cell, since our analysis does not permit us to differentiate between these possibilities.

FIG. 10
Protein linkage map of the cleavage products of the P2 protein precursor, suggested by data obtained by two-hybrid and GST pulldown assays. See the text for an explanation of the two classes of interaction. Double arrows indicate positive reciprocal interactions, ...

As mentioned above, we were unable to detect an interaction between 2C and 2C in the two-hybrid system. Indeed, we noted that GAD-2C did not interact with any LexA fusion. In contrast, the LexA-2C fusion was positive for interaction with GAD-2B and GAD-2BC. Explanations for this polar phenomenon include the possibility that the noninteracting fusion protein (GAD-2C) is misfolded or inherently unstable, that the expression is poor, and that transport to the nucleus does not occur. The expression of GAD-2C, as assayed by Western blotting of yeast extracts, was efficient (data not shown), eliminating the possibility that the protein was not present. Although we have failed to directly observe 2C/2C interaction, the 2C moiety of 2BC nevertheless appears to contribute to an interaction between 2BC and its cleavage products. This we conclude from the positive signals, albeit weak, that we obtained between 2BC(W36) and 2B, 2C, and 2BC, whereas 2B(W36) had lost interaction with any of the possible partners (Fig. (Fig.7C).7C). These observations suggest that the 2BC/2BC interaction is aided by, or is even dependent on, the 2C moieties and does not occur exclusively through 2B/2B or 2B/2C interaction. Indeed, the 2BC/2BC interaction in vitro appears to be mediated mainly by the 2C moieties (Fig. (Fig.4A).4A). In addition, the overall reduction in the signal may explain the polarity of the 2B/2BC(W36) interaction, which was already weak in the pairing with wild-type 2BC and is absent in the pairing with the mutant.

Polypeptide 2C has been proposed to be a helicase required in RNA replication (23, 24, 59), although experimental evidence for such a function is as yet missing. It is possible that such a function could be performed by 2BC and not by its cleavage product 2C. Oligomerization has been described as being critical to the function of most, but not all, DNA and RNA helicases (30). However, the question of whether 2C and 2BC are helicases must await further biochemical study of these proteins.

Observations by Tolskaya et al. (53) that guanidine-resistant viral mutants of 2C could complement guanidine-sensitive and -dependent mutants under restrictive conditions have been interpreted to suggest a functional oligomerization of 2C, perhaps at the level of 2BC. However, the rescue of a function in trans does not necessarily require oligomerization, merely the presence of the functional version of the protein, to carry out the activity.

The signal for the 2Apro/2Apro interaction, even for mutants debilitated in proteinase activity, was so weak that HIS3 expression was undetectable, a phenomenon that we cannot explain at present. In general, we noted that the version of the two-hybrid system utilizing the Y153 strain (with GAL1 promoter-dependent reporter genes) was significantly less sensitive than the L40-ura3 system. This suggests that the limiting factor in the level of the signal may be the promoter strength of the reporter system. Liebig et al. (37) reported that purified 2Apro of human rhinovirus 2 may exist as a dimer. Work with purified recombinant 2Apro of PVM indicates that under some conditions, the protein can exist as a monomer, dimer, and tetramer, but these forms have not been detected in infected cells (31). Our observation that the GST 2Apro/2Apro interaction is weak but detectable corroborates the two-hybrid results and suggests that the interaction may be intrinsically weak.

Genetic analysis of 2B multimerization.

Two observations have led us to conclude that the 2B/2B interaction is required for poliovirus replication. First, dominant-negative mutants of 2B studied by Johnson and Sarnow (29) interacted with the wild-type protein in the two-hybrid system at levels comparable to those for wild-type 2B multimerization. If the normal function of 2B requires multimerization, the hybrid wild-type–mutant complex would presumably be inactive, resulting in a lower level of replication of wild-type as well as mutant viral genomes. One example of such an effect is that of dominant-negative mutants of the p53 tumor suppressor protein, in which function appears to be lost through the assembly of heterotetramers consisting of wild-type and mutant monomers (12, 33).

Second, two mutations in 2B, identified by randomly mutagenizing 2B and selecting for variants that are negative for 2B multimerization, nearly abolished replication of luciferase-expressing viral replicons. When introduced into the full-length genome of PVM, these mutations conferred a qi phenotype. After transfection of HeLa cells with mutant transcript RNAs, CPE was observed only after prolonged incubation. From these cultures, only virus in which the mutation had reverted to the wild-type sequence was recovered. We interpret this to mean that the 2B mutants, as expressed by translation of the corresponding transcript RNAs in transfected HeLa cells, may still interact weakly, allowing a very low level of genomic replication and reversion. This property is the signature of the qi phenotype (22). Thus, the absence of a 2B/2B interaction in the two-hybrid system covaries with severe impairment in replication. In addition, the linkage analysis of 2BC(W36) suggests that this phenotype may result from loss of function at the level of 2B, since 2BC(W36) can still multimerize and interact with 2B and 2C. The mild processing anomaly seen by in vitro translation of the mutant genomes is unlikely to cause the qi phenotype, since all the nonstructural proteins are produced. For comparison, mutants of coxsackie B3 virus mapping to the 2B/2C cleavage site that exhibited a severe defect in 2B and 2C production were still viable, if slow-growing (57).

The two amino acid changes in the polypeptide chain of 2B that abolish multimerization are located inside a region which is predicted to form a hydrophobic β-sheet (Fig. (Fig.1B;1B; residues 46 to 56). We cannot conclude that the specific residues in question (I-53 and I-54) constitute contact points for interaction. Since the hydrophobicity of the region has changed, the overall structure of the region may be distorted. However, computer predictions indicate that the mutant sequences retain the propensity for a β-sheet, suggesting that the effect of the mutations on the tertiary structure is minimal. Therefore, the general hydrophobicity of this region may be required for multimerization. The fact that the C-terminal truncation mutants were also negative for 2B/2B binding in the two-hybrid system may implicate large segments of the C terminus in 2B multimerization.

Previous work from other groups has shown that exogenously expressed enteroviral 2B can permeabilize bacterial and mammalian cells (1, 17, 34, 56) and can block secretory transport at the endoplasmic reticulum/cis-Golgi step in the latter (17). Permeabilization suggests that 2B may form a pore in cellular membranes; such a structure would almost certainly require oligomerization of the protein, possibly dependent upon the C-terminal hydrophobic domain. The observation that 2B multimerizes in the two-hybrid assay and the broad structural requirements for this multimerization support this model of 2B function. Van Kuppeveld et al. (58) have implicated this property of 2B in the release of progeny virions from the cell during the final stages of infection. What purpose permeabilization would serve in the life cycle of the virus and whether the 2BC precursor would be the true functional entity in this activity (1) are questions that remain to be answered.


We thank Rohit Duggal for critical reading of the manuscript, Rolf Sternglanz for the gift of the L40-ura3 strain, Stanley Fields and his colleagues for providing valuable advice on two-hybrid vectors, protocols, control plasmids, and the Y153 strain, and Kurt Bienz for providing anti-2B monoclonal antibody. We are particularly indebted to Xuemei Cao and Meijia Yang for the pBTM38 plasmid and suggestions.

A.C. was a member of the graduate training program of the Department of Molecular Genetics and Microbiology. T.P. was supported, in part, by the Swiss National Foundation, the Freie Akademische Gesellschaft Basel, and the Theodor Engelmann-Stiftung, Basel, Switzerland. F.L. was supported by a grant from Schering-Plough. This work was supported in part by NIH-5R37AI15122 and AI32100.


1. Aldabe R, Barco A, Carrasco L. Membrane permeabilization by poliovirus proteins 2B and 2BC. J Biol Chem. 1996;271:23134–23137. [PubMed]
2. Alexander L, Lu H-H, Wimmer E. Polioviruses containing picornavirus type 1 and/or type 2 internal ribosomal entry site elements: genetic hybrids and the expression of a foreign gene. Proc Natl Acad Sci USA. 1994;91:1406–1410. [PMC free article] [PubMed]
3. Andino R, Rieckhof G E, Achacoso P L, Baltimore D. Poliovirus RNA synthesis utilizes an RNP complex formed around the 5′-end of viral RNA. EMBO J. 1993;12:3587–3598. [PMC free article] [PubMed]
4. Barco A, Carrasco L. A human virus protein, poliovirus protein 2BC, induces membrane proliferation and blocks the exocytic pathway in the yeast Saccharomyces cerevisiae. EMBO J. 1995;14:3349–3364. [PMC free article] [PubMed]
5. Barco A, Carrasco L. Poliovirus 2Apro expression inhibits growth of yeast cells. FEBS Lett. 1995;371:4–8. [PubMed]
6. Bartel P L, Chien C, Sternglanz R, Fields S. Using the two-hybrid system to detect protein-protein interactions. In: Hartley D A, editor. Cellular interactions in development: a practical approach. Oxford, United Kingdom: Oxford University Press; 1993. pp. 153–179.
7. Bartel P L, Roecklein J A, SenGupta D, Fields S. A protein linkage map of Escherichia coli bacteriophage T7. Nat Genet. 1996;12:72–77. [PubMed]
8. Bernstein H D, Sarnow P, Baltimore D. Genetic complementation among poliovirus mutants derived from an infectious cDNA clone. J Virol. 1986;60:1040–1049. [PMC free article] [PubMed]
9. Bienz K, Egger D, Pasamontes L. Association of polioviral proteins of the P2 genomic region with the viral replication complex and virus-induced membrane synthesis as visualized by electron microscopic immunocytochemistry and autoradiography. Virology. 1987;160:220–226. [PubMed]
10. Bienz K, Egger D, Pfister T. Characteristics of the poliovirus replication complex. Arch Virol Suppl. 1994;9:147–157. [PubMed]
11. Bienz K, Egger D, Troxler M, Pasamontes L. Structural organization of poliovirus RNA replication is mediated by viral proteins of the P2 genomic region. J Virol. 1990;64:1156–1163. [PMC free article] [PubMed]
12. Brachmann R K, Vidal M, Boeke J D. Dominant-negative p53 mutations selected in yeast hit cancer hot spots. Proc Natl Acad Sci USA. 1996;93:4091–4095. [PMC free article] [PubMed]
13. Caliguiri L A, Tamm I. Membranous structures associated with translation and transcription of poliovirus RNA. Science. 1969;166:885–886. [PubMed]
14. Cao X, Kuhn R J, Wimmer E. Replication of poliovirus RNA containing two VPg coding sequences leads to a specific deletion event. J Virol. 1993;67:5572–5578. [PMC free article] [PubMed]
15. Cho M W, Teterina N, Egger D, Bienz K, Ehrenfeld E. Membrane rearrangement and vesicle induction by recombinant poliovirus 2C and 2BC in human cells. Virology. 1994;202:129–145. [PubMed]
16. Datta A K. Efficient amplification using ‘megaprimer’ by asymmetric polymerase chain reaction. Nucleic Acids Res. 1995;23:4530–4531. [PMC free article] [PubMed]
17. Doedens J R, Kirkegaard K. Inhibition of cellular protein secretion by poliovirus proteins 2B and 3A. EMBO J. 1995;14:894–907. [PMC free article] [PubMed]
18. Egger D, Pasamontes L, Bolten R, Boyko V, Bienz K. Reversible dissociation of the poliovirus replication complex: functions and interactions of its components in viral RNA synthesis. J Virol. 1996;70:8675–8683. [PMC free article] [PubMed]
19. Estojak J, Brent R, Golemis E A. Correlation of two-hybrid affinity data with in vitro measurements. Mol Cell Biol. 1995;15:5820–5829. [PMC free article] [PubMed]
20. Fields S. Methods: a companion to Methods in Enzymology. Vol. 5. New York, N.Y: Academic Press; 1993. The two-hybrid system to detect protein-protein interactions; pp. 116–124.
21. Fields S, Song O. A novel genetic system to detect protein-protein interactions. Nature. 1989;340:245–246. [PubMed]
22. Gmyl A P, Pilipenko E V, Maslova S V, Belov G A, Agol V I. Functional and genetic plasticities of the poliovirus genome: quasi-infectious RNAs modified in the 5′-untranslated region yield a variety of pseudorevertants. J Virol. 1993;67:6309–6316. [PMC free article] [PubMed]
23. Gorbalenya A E, Koonin E V, Donchenko A P, Blinov V M. Two related superfamilies of putative helicases involved in replication, recombination, repair and expression of DNA and RNA genomes. Nucleic Acids Res. 1989;17:4713–4730. [PMC free article] [PubMed]
24. Gorbalenya A E, Koonin E V, Wolf Y I. A new superfamily of putative NTP-binding domains encoded by genomes of small DNA and RNA viruses. FEBS Lett. 1990;262:145–148. [PubMed]
25. Guan K L, Dixon J E. Eukaryotic proteins expressed in Escherichia coli: an improved thrombin cleavage and purification procedure of fusion proteins with glutathione S-transferase. Anal Biochem. 1991;192:262–267. [PubMed]
26. Hambidge S J, Sarnow P. Translational enhancement of the poliovirus 5′ noncoding region mediated by virus-encoded polypeptide 2A. Proc Natl Acad Sci USA. 1992;89:10272–10276. [PMC free article] [PubMed]
27. Harris K S, Hellen C U T, Wimmer E. Proteolytic processing in the replication of picornaviruses. Semin Virol. 1990;1:323–333.
28. Hellen C U, Facke M, Krausslich H G, Lee C K, Wimmer E. Characterization of poliovirus 2A proteinase by mutational analysis: residues required for autocatalytic activity are essential for induction of cleavage of eukaryotic initiation factor 4F polypeptide p220. J Virol. 1991;65:4226–4231. [PMC free article] [PubMed]
29. Johnson K L, Sarnow P. Three poliovirus 2B mutants exhibit noncomplementable defects in viral RNA amplification and display dosage-dependent dominance over wild-type poliovirus. J Virol. 1991;65:4341–4349. [PMC free article] [PubMed]
30. Kadare G, Haenni A. Virus-encoded RNA helicases. J Virol. 1997;71:2583–2590. [PMC free article] [PubMed]
31. Kakegawa, H., and E. Wimmer. 1997. Unpublished results.
32. Klump H, Auer H, Liebig H-D, Kuechler E, Skern T. Proteolytically active 2A proteinase of human rhinovirus 2 is toxic for Saccharomyces cerevisiae but does not cleave the homologues of eIF-4g in vivo or in vitro. Virology. 1996;220:109–118. [PubMed]
33. Ko L J, Prives C. p53: function and paradigm. Genes Dev. 1996;10:1054–1072. [PubMed]
34. Lama J, Carrasco L. Expression of poliovirus nonstructural proteins in Escherichia coli cells. Modification of membrane permeability induced by 2B and 3A. J Biol Chem. 1992;267:15932–15937. [PubMed]
35. Leung D W, Chen E, Goeddel D V. A method for random mutagenesis of a defined DNA segment using a modified polymerase chain reaction. Technique. 1989;1:11–15.
36. Li X, Lu H-H, Mueller S, Wimmer E. Functional analysis of poliovirus 2Apro C-terminal domain. 1997. Unpublished data.
37. Liebig H D, Ziegler E, Yan R, Hartmuth K, Klump H, Kowalski H, Blaas D, Sommergruber W, Frasel L, Lamphear B, Rhoads R, Kuechler E, Skern T. Purification of two picornaviral 2A proteinases: interaction with eIF-4g and influence on in vitro translation. Biochemistry. 1993;32:7581–7587. [PubMed]
38. Lu H H, Li X, Cuconati A, Wimmer E. Analysis of picornavirus 2Apro proteins: separation of proteinase from translation and replication functions. J Virol. 1995;69:7445–7452. [PMC free article] [PubMed]
39. Mirzayan C, Wimmer E. Biochemical studies on poliovirus polypeptide 2C: evidence for ATPase activity. Virology. 1994;199:176–187. [PubMed]
40. Mirzayan C, Wimmer E. Genetic analysis of an NTP-binding motif in poliovirus polypeptide 2C. Virology. 1992;189:547–555. [PubMed]
41. Molla A, Paul A V, Schmid M, Jang S K, Wimmer E. Studies on dicistronic polioviruses implicate viral proteinase 2Apro in RNA replication. Virology. 1993;196:739–747. [PubMed]
42. Molla A, Paul A V, Wimmer E. Cell-free, de novo synthesis of poliovirus. Science. 1991;254:1647–1651. [PubMed]
43. Pallansch M A, Kew O M, Semler B L, Omilianowski D R, Anderson C W, Wimmer E, Rueckert R R. Protein processing map of poliovirus. J Virol. 1984;49:873–880. [PMC free article] [PubMed]
44. Pfister T, Wimmer E. Poliovirus 2C contains a zinc finger motif involved in RNA replication. 1997. Unpublished data.
45. Rodriguez P L, Carrasco L. Poliovirus protein 2C contains two regions involved in RNA binding activity. J Biol Chem. 1995;270:10105–10112. [PubMed]
46. Rodriguez P L, Carrasco L. Poliovirus protein 2C has ATPase and GTPase activities. J Biol Chem. 1993;268:8105–8110. [PubMed]
47. Sandoval I V, Carrasco L. Poliovirus infection and expression of the poliovirus protein 2B provoke the dissassembly of the Golgi complex, the organelle target for the antipoliovirus drug Ro-090179. J Virol. 1997;71:4679–4693. [PMC free article] [PubMed]
48. SenGupta D J, Zhang B, Kraemer B, Pochart P, Fields S, Wickens M. A three-hybrid system to detect RNA-protein interactions in vivo. Proc Natl Acad Sci USA. 1996;93:8496–8501. [PMC free article] [PubMed]
49. Shepley, M. 1997. Personal communication.
50. Shepley M P, Sherry B, Weiner H L. Monoclonal antibody identification of a 100-kDa membrane protein in HeLa cells and human spinal cord involved in poliovirus attachment. Proc Natl Acad Sci USA. 1988;85:7743–7747. [PMC free article] [PubMed]
51. Sonenberg N. Regulation of translation by poliovirus. Adv Virus Res. 1987;33:175–204. [PubMed]
52. Takegami T, Semler B L, Anderson C W, Wimmer E. Membrane fractions active in poliovirus RNA replication contain VPg precursor polypeptides. Virology. 1983;128:33–47. [PubMed]
52a. Teterina N L, Kean K M, Gorbalenya A E, Agol V I, Girard M. Analysis of the functional significance of amino acid residues in the putative NTP-binding pattern of the poliovirus 2C protein. J Gen Virol. 1992;73:1977–1986. [PubMed]
53. Tolskaya E A, Romanova L I, Kolesnikova M S, Gmyl A P, Gorbalenya A E, Agol V I. Genetic studies on the poliovirus 2C protein, an NTPase. A plausible mechanism of guanidine effect on the 2C function and evidence for the importance of 2C oligomerization. J Mol Biol. 1994;236:1310–1323. [PubMed]
54. Toyoda H, Niclin N J, Murray M G, Anderson C W, Dunn J J, Studier F W, Wimmer E. A second virus-encoded proteinase involved in proteolytic processing of poliovirus polyprotein. Cell. 1986;45:761–770. [PubMed]
55. Van Aelst L, Barr M, Marcus S, Polverino A, Wigler M. Complex formation between RAS and RAF and other protein kinases. Proc Natl Acad Sci USA. 1993;90:6213–6217. [PMC free article] [PubMed]
56. van Kuppeveld F J, Melchers W J, Kirkegaard K, Doedens J R. Structure-function analysis of coxsackie B3 virus protein 2B. Virology. 1997;227:111–118. [PubMed]
57. van Kuppeveld F J, van den Hurk P J, Zoll J, Galama J M, Melchers W J. Mutagenesis of the coxsackie B3 virus 2B/2C cleavage site: determinants of processing efficiency and effects on viral replication. J Virol. 1996;70:7632–7640. [PMC free article] [PubMed]
58. van Kuppeveld F J M, Hoenderop J G J, Smeets R L L, Willems P H G M, Dijkman H B P M, Galama J M D, Melchers W J G. Coxsackie protein 2B modifies endoplasmic reticulum membrane and plasma membrane permeability and facilitates virus release. EMBO J. 1997;16:3519–3532. [PMC free article] [PubMed]
59. Wimmer E, Hellen C U, Cao X. Genetics of poliovirus. Annu Rev Genet. 1993;27:353–436. [PubMed]
60. Xiang, W., A. Cuconati, D. Hope, K. Kirkegaard, and E. Wimmer. A protein linkage map of poliovirus P3 proteins. Submitted for publication.
61. Xiang W, Cuconati A, Paul A V, Cao X, Wimmer E. Molecular dissection of the multifunctional poliovirus RNA-binding protein 3AB. RNA. 1995;1:892–904. [PMC free article] [PubMed]
62. Yu S F, Benton P, Bovee M, Sessions J, Lloyd R E. Defective RNA replication by poliovirus mutants deficient in 2A protease cleavage activity. J Virol. 1995;69:247–252. [PMC free article] [PubMed]

Articles from Journal of Virology are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...