Search tips
Search criteria 


Logo of plosonePLoS OneView this ArticleSubmit to PLoSGet E-mail AlertsContact UsPublic Library of Science (PLoS)
PLoS One. 2010; 5(5): e10735.
Published online 2010 May 20. doi:  10.1371/journal.pone.0010735
PMCID: PMC2873979

Deletion Mutants of VPg Reveal New Cytopathology Determinants in a Picornavirus

Maciej Lesniak, Editor



Success of a viral infection requires that each infected cell delivers a sufficient number of infectious particles to allow new rounds of infection. In picornaviruses, viral replication is initiated by the viral polymerase and a viral-coded protein, termed VPg, that primes RNA synthesis. Foot-and-mouth disease virus (FMDV) is exceptional among picornaviruses in that its genome encodes 3 copies of VPg. Why FMDV encodes three VPgs is unknown.

Methodology and Principal Findings

We have constructed four mutant FMDVs that encode only one VPg: either VPg1, VPg3, or two chimeric versions containing part of VPg1 and VPg3. All mutants, except that encoding only VPg1, were replication-competent. Unexpectedly, despite being replication-competent, the mutants did not form plaques on BHK-21 cell monolayers. The one-VPg mutant FMDVs released lower amounts of encapsidated viral RNA to the extracellular environment than wild type FMDV, suggesting that deficient plaque formation was associated with insufficient release of infectious progeny. Mutant FMDVs subjected to serial passages in BHK-21 cells regained plaque-forming capacity without modification of the number of copies of VPg. Substitutions in non-structural proteins 2C, 3A and VPg were associated with restoration of plaque formation. Specifically, replacement R55W in 2C was repeatedly found in several mutant viruses that had regained competence in plaque development. The effect of R55W in 2C was to mediate an increase in the extracellular viral RNA release without a detectable increase of total viral RNA that correlated with an enhanced capacity to alter and detach BHK-21 cells from the monolayer, the first stage of cell killing.


The results link the VPg copies in the FMDV genome with the cytopathology capacity of the virus, and have unveiled yet another function of 2C: modulation of picornavirus cell-to-cell transmission. Implications for picornaviruses pathogenesis are discussed.


Contrary to initiation of cellular DNA replication which is primed by RNA molecules synthesised by cellular primases [1], viruses use a wide variety of molecular mechanisms to initiate genome replication, that include de novo initiation, priming by proteins or by self generated 3′–ends of templates, and ‘cap–snatching’, among other mechanisms [2]. Protein–primed initiation of genome replication is used by several DNA and RNA viruses and some linear plasmids [3][5]. Picornaviridae is a family of positive strand RNA viruses that use as protein–primer a small peptide of about 20 residues in length, termed VPg or 3B [3], [6], [7]. After replication, the protein–primer VPg remains bound to the genomic RNA encapsidated into viral particles. Picornaviruses encode only one copy of VPg, except foot–and–mouth disease virus (FMDV) that expresses three similar but non–identical copies of VPg (VPg1–3 or 3B1–3) [8] (Figure 1). Each of the three VPgs are found covalently bound to genomic viral RNA [9] and they can be uridylylated in vitro by the viral polymerase, with VPg3>VPg2>VPg1 as the order of substrate efficiency [10]. The biological meaning of this unique in–tandem repetition in an RNA virus is not well understood [11], [12]. Molecular poliovirus clones constructed to express two VPgs delete one of the two copies, and the polyprotein harboring two VPgs underwent aberrant processing [13], [14]. FMDV encoding only VPg3 is infectious in cell culture, showing that one copy of VPg may be sufficient to complete the virus replication cycle [12]. The virus expressing only VPg3 was infectious for hamster and bovine fibroblasts (BHK and FBK cells), but not swine fibroblasts (FPK cells), and was attenuated for swine [12]. FMDVs encoding VPg1 and VPg2, but lacking VPg3 were not viable, suggesting that the presence of VPg3 was essential for FMDV viability [11]. The authors proposed that this loss of viability could be due to a defect in the proteolytic processing of the viral polyprotein precursor lacking VPg3 [11].

Figure 1
Schematic representation of the FMDV genome and of the constructions with one copy of VPg.

Picornaviral proteins are generated by proteolytic processing of a single viral polyprotein which is translated from a single ORF. During and after translation, different cleavages of the viral polyprotein take place, most of them catalysed by the viral protease 3C, resulting in the release of different processing intermediates and mature proteins (reviewed in [15]). Specifically, the capsid precursor (P1) is processed into VP0 (VP4–VP2), VP3 and VP1 which are assembled to form the mature virions. P2 and P3 precursors render non-structural proteins 2A, 2B, 2C, 3A, 3B (VPg), 3C, 3D and several processing intermediates which are required for viral replication. 3D is the viral RNA-dependant RNA polymerase (RdRp) that catalyses genomic RNA synthesis and the critical VPg–uridylylation step at the initiation of replication. 3C and its precursor 3CD stimulate the initial VPg–uridylylation step, an activity additional to their role in polyprotein processing [16][18]. It has been recently proposed that a precursor form of VPg (either 3AB or 3BC) could act as the authentic protein–primer molecule, while processing and release of 3B (VPg) would be a step subsequent to initiation of replication, although these hypotheses are still under discussion [16], [17], [19]. 2C and 3A play also central roles in picornavirus replication. 2C includes NTPase and RNA–binding activities [20][23], acts as an RNA chaperone during picornaviral replication [24], and is involved in viral RNA encapsidation [25], uncoating [26], and in host cell membrane rearrangements required for replication [27][31]. 3A is a membrane protein [32] that can establish interactions with 2C [33], suggesting that 3A and 2C may constitute part of the same protein complexes for some biological processes. In addition, 3AB precursor (3A bound to VPg) may be involved in the recruitment of 3D polymerase to membranes to form replication complexes in which the viral genomes are synthesised [34], [35].

In the present report, we provide evidence of a functional link between VPg and 2C in the release of FMDV from cells. We have constructed FMDVs that encode a single VPg. Some mutants were replication–competent but did not produce plaques on BHK–21 cell monolayers. Passage of these viruses in BHK–21 cells selected for mutants in 2C, 3A and 3B (VPg) that regained the ability to develop plaques. These mutants displayed increased cytopathology and virus shedding into the extracellular medium. The results provide evidence for a function of the triplicated VPg in the detachment of cells from the monolayers, which is the event that precedes cell killing by FMDV [36], [37]. These observations establish functional connections between VPg and non–structural FMDV proteins.


FMDVs with only one VPg copy are replication–competent viruses

Previous studies showed that FMDV encoding VPg3 as the only copy of VPg (but not FMDV encoding only VPg1 and VPg2) is infectious in cell culture [11], [12]. It was not clear whether lethality was due to absence of VPg3 or to absence of the proteolytic cleavage site between VPg3 and 3C [11]. To address whether VPg3 is essential for FMDV replication, and to investigate the function of repeated VPg genes in Aphthoviruses, we have designed FMDVs with only VPg1 (termed V1), only VPg3 (termed V3), and chimeric viruses containing most of VPg1 but a short C-terminal portion of VPg3 that restores the cleavage site between VPg1 and 3C. The chimeric–VPg viruses have been termed V19–4 (first 19 residues from VPg1 and last 4 residues from VPg3) and V15–9 (first 15 residues of VPg1 and last 9 residues from VPg3) (Figure 1). All mutant viruses, except V1, were competent in replication when the corresponding RNAs were transfected into BHK–21 cells, as evidenced by quantification of virus–specific RNA and proteins (Figure 2). The results show that a complete sequence of VPg3 is not essential for viral replication, but VPg1–containing mutants require additional residues from VPg3 in their C–terminus for replication.

Figure 2
Replication of mutant FMDVs in BHK–21 cells.

Viral proteins, measured by metabolic labelling and Western blot analysis, were detected with all mutants except V1 (Figure 2A–C). As expected, all mutant constructs that expressed replication-competent RNAs gave rise to a P3 polyprotein precursor that displayed higher mobility than wild type (WT) P3, due to the deletion of two VPg copies (Figure 2B and 2C). The pattern of processed P3 proteins in the expression products of V19–4 and V15–9 supports an efficient processing of the precursor 3BC into 3B and 3C, in agreement with the introduction of a functional 3C cleavage site in the mutants (Figure 2C). Quantification of total (intracellular and extracellular) viral RNA collected from cell cultures transfected with mutant FMDVs yielded 104–105 viral RNA molecules/cell (vRNA/cell) for mutants V3, V15–9 and V19–4, a value which is 10– to 100–fold lower than that obtained for FMDV WT (Figure 2D). Mutant V1 showed a viral RNA level that was 10,000-fold lower than WT (<102 viral RNA molecules/cell), suggesting that replication of V1 is either drastically reduced or abrogated (Figure 2). It cannot be excluded that the low viral RNA level detected after transfection with V1 RNA could be a remnant of the RNA used in the transfection (104 viral RNA molecules/cell), and it confirms the positive replication of the other mutants. In summary, there is no requirement of a specific VPg sequence for FMDV viability since the three different VPg versions confer replication–competence to a similar extent. However, the presence of a 3C protease cleavage site between VPg and 3C seems to be needed for viral RNA synthesis.

FMDVs expressing a single VPg lack plaque–forming capacity

Samples from the cell culture supernatants of BHK–21 cells transfected with V1, V3 or V19–4 RNAs did not include detectable viral infectivity in a standard FMDV plaque assay (<5 PFU/ml), with plaques visualised at 48 hours post–plating (FMDV WT plaques are readily detectable at 24 hours post–plating) (Figure 2E). Samples from transfection with V15–9 presented detectable infectivity [8(±3)×102 PFU/ml], albeit 1,000- to 10,000-fold lower than WT [3(±2)×106 PFU/ml]. The absence of plaque development with V3 or V19–4 contrasted with the large amounts of total viral genomic RNA detected in the same cell cultures (Fig 2D), and with the observed intracellular levels of viral proteins (Figure 2A–C). The specific infectivity (SI, ratio between viral titre and the number of viral RNA molecules) of viruses expressing only one VPg is lower than 1[ratio]20,000,000, which is at least 500–fold lower than that of FMDV WT (1[ratio]40,000). For standard FMDV C–S8c1, the ratio of infectious particles to total number of physical particles was previously estimated in 1[ratio]7,000 to 1[ratio]10,000 [38], and the ratio decreases when FMDV is subjected to mutagenesis [39]. Thus, viruses expressing a single VPg are competent for viral protein synthesis and RNA synthesis, but defective in plaque development (Figure 2).

The specific infectivity of mutant FMDVs is gradually regained upon passage in BHK–21 cells

To investigate whether the capacity of FMDV mutants V1, V3, V19–4 and V15–9 to form plaques could be restored upon further virus replication, the virus present in supernatants collected from cell cultures initially transfected with 100 ng (series A) or 500 ng (series B) of V1, V3, V19–4 and V15–9 RNA was serially passaged in BHK–21 cells. At the first passage (p1), cytopathology was complete in cells infected with V3 and V15–9, but affected only 10% of the cells infected with V19–4, at 70–90 hours post–infection (hpi), and no cytopathology was observed with the supernatants that should contain the progeny of V1 RNA. For FMDV WT, complete cytopathology was observed at 30 hpi, in agreement with previous observations [40]. Infectivity was detected for all viruses (except for V1) at p1, although mutant viruses yielded lower titres and SI values than WT. Successive passages led gradually to a more extensive cytopathology at shorter times pi, and to increases in viral titre and SI for all mutants tested, except for V1 (Figure 3). Plaque development by FMDV on BHK–21 cell monolayers is a reflection of cell detachment as a manifestation of virus infection. Therefore, in the assays described here, cytopathology refers as cell rounding and detachment as a result of the cellular modifications previously described for BHK–21 cells lytically infected by FMDV [36], [37].

Figure 3
Recovery of infectivity of FMDV mutants expressing one VPg, upon serial passages in BHK–21 cells.

Multiple genomic sites are involved in recovery of the plaque–forming phenotype

To investigate whether plaque–forming capacity was regained via adaptive mutations in the viral genome, the entire genomic consensus sequence [except the genomic 5′ and 3′ termini, and the nucleotides upstream and downstream from of the polyC tract] of six mutant populations at passage 2 (series A and B for mutants V3, V19–4 and V15–9) was determined. Although no mutation was repeated in all populations, amino acid substitution R55W in non–structural protein 2C was present in populations V3B, V19–4B and V15–9B (Table 1). This result suggested that substitution R55W in 2C could be involved in restoring the plaque–forming phenotype in these three populations. The only substitution that affected VPg was E8K found in population V19–4A (Table 1).

Table 1
Substitutions found in FMDV mutants that encode one VPg, upon passage in BHK–21 cells a.

Recovery of the plaque–forming phenotype was not associated with any dominant mutation in two out of six populations at passage 2 (V3A and V15–9A). To determine whether the mutant spectrum of V3A included genomes with the same mutations found in the other lineages, the P2–P3–coding region of virus isolated from 5 individual plaques derived from population V3A at passage 1 was sequenced. Each clone included non–synonymous mutations in 2C and/or 3A, and two clones presented replacement R55W in 2C (Table 1). The results suggest that plaque–development in the VPg mutants can be attained by any of a number of substitutions in non–structural proteins 2C, 3A and/or VPg. Since R55W in 2C was found in several clones and populations, its effect on FMDV progeny production and cytopathology was further investigated.

2C substitution R55W increases the specific infectivity of FMDV expressing a single VPg

To determine whether substitution R55W in 2C could restore the plaque–forming capacity of FMDV mutants with alterations in 3B (FMDV mutants V3, V19–4 and V15–9, Figure 1), the corresponding mutation was introduced in each of the mutant constructs that express one VPg. All mutants regained the plaque–forming phenotype (Figure 4A) that resulted in 17– to 190–fold increase of specific infectivity of each viral RNA transcript (Figure 4B). To minimize the emergence of adaptive mutations during plaque development, serially diluted viral RNA transcripts were transfected into cells, and plaque development was permitted under semisolid agar medium. Plaques were formed by all mutants expressing only one VPg and 2C with R55W, and they were visible at 3 days post–transfection. No plaques were formed by their RNA counterparts expressing wild type 2C, under the same conditions. By 2 days post–transfection, plaques were already detected for V15–9 and V3 with R55W in 2C, but not for V19–4 with R55W in 2C (Figure 4). Thus, 2C substitution R55W is sufficient to restore the plaque–forming capacity of the mutants tested that express one VPg.

Figure 4
Restoration of the plaque–forming phenotype in FMDV mutants by replacement R55W in 2C.

R55W in 2C increases cytopathic effect and results in enhanced viral release from cells

To investigate the effect of R55W 2C in FMDV replication, viral RNA and viral protein levels of FMDVs expressing 2C with either R55 (wild type) or W55 (adaptive substitution) were quantified at different times post–electroporation. Positive RNA replication was observed for all mutants (V3, V19–4 and V15–9 with or without R55W in 2C) with RNA levels reaching an amount at least 10-fold higher than levels quantified at the time of electroporation (Figure 5A). To ascertain the presence of newly synthesised viral RNA molecules, the background RNA value scored at the time of the electroporation was subtracted from each of the values obtained at different times after electroporation. The results (Figure 5A) indicate that the presence of R55W in 2C of the FMDV mutants increased the amount of extracellular viral RNA without any significant increase in the total viral RNA levels (intracellular plus extracellular). The ratio of extracellular to total viral RNA molecules was estimated to be in the range of 0.1 to 0.4 for mutants encoding R55W in 2C, and 0.03 to 0.04 for mutants encoding wild type 2C (Figure 5B).

Figure 5
Effect of replacement R55W in 2C in the release of FMDV RNA from cells.

The selective increase of extracellular RNA suggests increased exit of viral particles. This implies that higher extracellular concentrations of capsid proteins should be found for mutants expressing 2C with R55W than for the same mutants expressing wild type 2C. This was confirmed by determining extracellular levels of capsid proteins VP1 and VP3. First, the synthesis of capsid proteins VP1 and VP3 was ascertained in the electroporated cells by pulse-labeling with [35S] Met-Cys at 3 h to 4 h post-electroporation. The result (Figure 5C) revealed the presence of VP1 and VP3 in the cells infected with mutants V15-9 and V3, with or without R55W in 2C (excluding the cell culture supernatant for the analysis). In contrast, when the extracellular levels of VP1 and VP3 were examined at early times post-electroporation, they were barely detectable for the mutants that expressed 2C with R55 (Figure 5D). The difference tended to diminish when the pulse-labeling was extended for several hours post-electroporation (Figure 5D).

A standard cell killing assay previously developed for FMDV in BHK-21 cells [40], [41] could not be directly applied to measure the effect of substitution R55W in 2C to mutants V19-4, V15-9 and V3 because infectious (cell-detaching) particles could not be produced in sufficient amounts. However, the cell killing assay was applied to wild type FMDV expressing either wild type 2C or 2C with R55W and the results showed 8- to 20-fold increases of cytopathology associated with substitution R55W in 2C (Figure 6A).

Figure 6
Effect of replacement R55W in 2C on FMDV cytopathology.

In other assays, the number of detached cells versus the total number of cells was measured following electroporation with V3, V19–4 and V15–9 RNAs, expressing either 2C wild type or 2C with R55W, as a means of circumventing the requirement of infectious virus particles. The results (Figure 6B) show that the proportion of cells detached was 62±3%, 79±6% and 72±10% in transfections with mutants V19–4, V15–9 and V3 encoding R55W in 2C, respectively, while these values were 26±11%, 45±10% and 58±3% in transfections with the same mutants encoding the wild type 2C, and this difference was statistically significant (p = 0.0002; Student's t–Test). To prevent multiple–step growth conditions that may have affected the interpretation of results, the experiment was carried out by using 5 µg of viral RNA for 106 BHK-21 cells, a sufficient amount to ensure that more than 90% of the cells were efficiently transfected Therefore, substitution R55W in 2C favors both release of viral particles and cytopathology.

Further analyses with FMDV mutants expressing one VPg

The major defect of FMDV mutants V19-4, V15-9 and V3 that express a single VPg is the absence of cytopathology associated with limited release from cells with no impairment of intracellular viral RNA levels. To test whether other viral functions might also be affected in these mutants, cell entry and intracellular viral protein synthesis were also examined.

To determine whether mutants expressing a single VPg could also manifest a defect in cell entry, the capacity of mutants V19-4 and V3 expressing either wild type 2C or 2C with R55W to enter cells was determined. For this purpose, viral samples were collected from cell cultures at 4 hours post–transfection by freeze-thawing the cells to liberate virus, and then applied to a BHK–21 cell monolayer. The input FMDV RNA applied to the monolayer and the FMDV RNA that entered BHK-21 cells were quantitated by real time RT-PCR. The proportion of viral RNA released from transfected cells that entered new cells was 3.0±0.3% for V19-4 and 3.5±2.0% for V3 that expressed wild type 2C; the corresponding values were 2.6±1.3% and 3.0±1.2% for the same mutants that expressed 2C with R55W. These data suggest that defects in plaque-forming capacity observed are not due to a defect in viral entry once the mutant FMDvs have been released from the cells.

To compare intracellular protein synthesis directed by the mutant FMDVs, proteins were metabolically labeled between 3 and 4 hours post–electroporation. BHK–21 cells infected with mutants V3, V19–4 and V15–9 encoding R55W in 2C presented modestly higher levels of viral proteins than the same mutants expressing a wild type 2C (Figure 7). Therefore, a limitation in protein synthesis cannot be related to a defect in viral release from cells.

Figure 7
Viral proteins expressed by mutant FMDVs encoding either wild type 2C or 2C with R55W.

Substitution E8K in VPg contributes also to increased viral release in mutant viruses expressing one VPg

Mutant V19–4 (R55W) showed a delayed plaque formation relative to other mutants encoding R55W in 2C (Figure 4). To investigate whether additional replacements could accelerate plaque formation in this virus, populations V19–4A and V19–4B at passage 2 (depicted in Table 1 and Figure 3) were subjected to 25 additional passages in BHK–21 cells. E8K in VPg1 was the only replacement repeatedly found in the two V19–4 lineages at passage 27 (Table 2). Interestingly, R55W in 2C became dominant during the 25 additional passages in lineage V19–4A in which R55W was not present at passage 2 (compare Tables 1 and and2).2). However, in lineage V19–4B W55 was dominant at passage 2 but it had reverted to R55 by passage 27. This reversion was accompanied by dominance of substitution I85V in 2C (Table 2). Other amino acid replacements were identified in other viral proteins of passage series A and B (Table 2).

Table 2
Mutations in populations V19–4A and V19–4AB after 27 passages in BHK–21 cells a.

Since R55W in 2C and E8K in VPg1 were found together in V19–4A at passage 27, the effect of both substitutions together in the same genome in the sequence context of V19–4 was investigated. The double mutation restored plaque–forming capacity, with minute plaques detected at 48 h post–transfection (Figure 8A). The specific infectivity of V19-4 increased 10- to 100-fold as a result of expressing 2C with R55W alone or together with VPg with E8K (Figure 8B). The extracellular viral RNA was 5– to 11–fold higher for V19–4 (R55W,E8K) than for either V19–4 or V19–4 (R55W) (Figure 8C), and this difference was statistically significant (P<0.0001; unpaired t–Test). Metabolic labelling of cells transfected with V19–4 (R55W,E8K) RNA revealed that viral protein synthesis was increased 2.2– to 4.4–fold relative to V19–4 or V19–4 (R55W). Protein levels were 1.6–fold higher for V19–4 (R55W,E8K) than for wild type FMDV (Figure 8C). These measurements suggest that viral release into the extracellular medium was enhanced by substitution E8K in VPg.

Figure 8
Effect of replacement E8K in VPg on the specific infectivity of V19–4 (R55W).

In summary, FMDV expressing only one VPg is defective in cell–to–cell propagation and this defect is reverted by an increase in viral release, concomitantly with restoration of cytopathology, mediated by substitutions in 2C, VPg or other non-structural proteins.


Two features of RNA virus evolution are relevant to the interpretation of the results with VPg mutants of FMDV reported here. One is the multifunctional nature of many (probably most) proteins encoded in the compact RNA genomes, particularly in the case of picornaviruses in which several intermediates obtained in the processing of the polyprotein play essential roles in the virus life cycle. In the present study, multifunctionality was manifested by the participation of 2C in the compensation of a defect in the release from cells of FMDVs that do not express the 3 VPg versions encoded by wild type FMDV [42]. Such compensation suggests a role of 2C in functions additional to those described for this non–structural protein [20], [21], [24]-[27], [43]. The second relevant feature of RNA virus evolution is the functional connection among different genomic regions, in this case between VPg and other non–structural proteins, regarding expression of phenotypic traits [44].

The results reported here indicate that the replication-competent FMDV mutants that expressed a single VPg were deficient in plaque formation, a defect related to the inability to cause cytopathology. Cell rounding and detachment from the monolayer precede cell death. The mechanism (apoptosis, necrosis, autophagy) by which FMDV kills BHK-21 cells in cell culture has not been elucidated [45], and our attempts to confirm or exclude apoptosis of BHK-21 cells following infection by FMDV have not been conclusive (C. Perales, unpublished observations). We have observed that the first step towards cell death –as measured either by trypan blue staining or by fluorescence-activated cell sorting (FACS) using propidium iodide that gave equivalent quantifications [46]– is cell rounding and detachment of the monolayer. The capacity to produce cytopathology was rapidly restored upon passage of the mutant viruses in BHK-21 cells, that acquired amino acid substitutions in non-structural proteins, including R55W in 2C. The results suggest that substitution R55W in 2C permitted FMDVs that expressed a single VPg to attain a critical number of viral particles released from cells, that are associated with cytopathology. An approximate value of 5 infectious particles released per BHK–21 cell was estimated as the minimum number needed to develop a plaque under our experimental conditions (Table 3). This is the number of particles released per cell estimated for mutant V19–4 (R55W) which is near the threshold of not being competent in plaque development; the plaques formed by this virus were visible only at late times post–transfection. The functional basis for this critical number is not known.

Table 3
Relative amounts of mutant FMDV RNA released from cells a.

Residue 55 of 2C [and also residue 85, substituted in the R55W revertant of V19–4B (see Results)] is located in the region comprised between the expected membrane-binding domain and the predicted NTPase/helicase domain A, and it is a variable residue when 2Cs of different picornaviruses are aligned. Positions 55 and 85 have not been identified as related to resistance to guanidine in FMDV [47], [48] and therefore, from the data presently available no connection between 2C positions 55 and 85 and FMDV replication is apparent. It cannot be excluded that the established role of 2C in cellular membrane rearrangements during picornaviral infection [27][31] might be connected with its effect in virus release and cytopathology.

With the assumption that cytopathology by FMDV precedes cell death, several determinants of BHK–21 cell killing were previously mapped in the IRES, the viral capsid, and non–structural proteins of the virus [40], [49], [50]. A lack of correlation between replicative fitness and BHK–21 cell killing capacity was unveiled by analysis of the behavior of chimeric FMDVs. While fitness determinants were scattered along the genome, cell killing determinants were concentrated in several specific genomic regions, including 2C [40]. Three substitutions in 2C (S80N, T256A and Q263H) enhanced BHK–21 cell killing. Since these substitutions are distant from substitution R55W that multiple sites in 2C are related to FMDV cytopathology.

The molecular basis of the increased cytopathology observed for mutants that encode 2C with substitution R55W remains unknown. Since a modest increase in viral protein synthesis in these mutants is observed despite no increase in viral RNA levels, an appealing hypothesis is that 2C has some involved in translation, and that R55W might enhance protein synthesis modestly but to a level sufficient to contribute to enhanced cytopathology. The effect of substitution R55W could be exerted during polyprotein processing, by rendering 2C more sensitive to polyprotein cleavage mediated by 3C. Distance effects of amino acid substitutions on FMDV poplyprotein processing have been previously documented [51]. Other interpretations are possible. For example, since 2C is a multifunctional protein [20][31], substitution R55W may alter virus-host interactions that lead to enhanced cytopathology, being the slight increase of protein synthesis an indirect consequence of alterations in other viral functions.

Substitution E8K was also found associated with the one-VPg mutants that regained capacity to cause cytopathology. The crystal structure of the complex between 3D and VPg predicts that E8 participates directly in the interaction with the viral polymerase. Although E8 is far from the UMP residue bound to VPg (covalent bond Tyr3-UMP), substitution E8A has been shown to affect drastically the uridylylation of VPg by 3D [52]. The molecular basis of the phenotypic effect of VPg replacement E8K remains unknown.

Despite the fact that the behavior of the one-VPg FMDV mutants studied here suggests a functional connection between 2C and VPg in FMDV, to our knowledge a direct interaction between the two proteins has not been reported in picornaviruses. Nevertheless, PV 2C interacts with the VPg precursor 3AB (3B) [33]. In an FMDV–infected cell, a possible 3AB–2C interaction could be mediated by more than one VPg copy, and the alteration in the number of VPg copies or their amino acid sequence could perturb these 2C–3AB interactions. In addition to the replacements in 2C and 3B, substitutions in 3A were also found in populations and biological clones that regained the plaque–forming phenotype upon passage in BHK–21 cells, supporting that recovery of this phenotype can be mediated by any of a number of readjustments in the interaction among several non–structural proteins (Tables 1 and and22).

The present investigation has established non–structural proteins 2C, 3A and VPg as key determinants for modulating cytopathology in cell culture. For several picornaviruses it has been shown that replacements either in 2C or 3A can be associated with modifications of virulence, cell tropism or host range [40], [53][57]. Altogether, these data highlight the role of non–structural proteins in the adaptability to changing environments during picornavirus infections, with clear implications for viral pathogenesis.

Materials and Methods

Plasmids, cells and viruses

pMT28 is a pGEM–1 plasmid (Promega) that contains the complete cDNA genomic sequence of FMDV serotype C (FMDV WT: C-S8c1 as described in [41], [58], [59]). To produce infectious transcripts, pMT28 was linearised with NdeI (NEB), and transcribed with SP6 polymerase (Promega), following described procedures [60], [61]. Mutant plasmids encoding only one VPg copy were constructed by mutagenic PCR with primers harboring the desired deletion (overlapping upstream and downstream sequences of the region to be deleted). Two amplifications were made and then the DNA products were shuffled to introduce the deletion. The first amplification was performed with a forward primer spanning residues 3988 to 4009 and a mutagenic reverse primer spanning either residues 5980–5971 and 5826–5804 (to obtain the deletion of 5827 to 5970, mutant V1), or residues 5910–5899 and 5757–5735 (deletion of 5758–5898, mutant V3), or residues 5977–5959 and 5814–5792 (deletion of 5815–5958, mutant V19–4), or residues 5955–5944 and 5802–5781 (deletion of 5803–5943, mutant V15–9). The second amplification was performed with an antisense primer spanning either residues 7160 to 7141 and a mutagenic forward primer spanning residues 5816–5826 and 5971–5992 (to obtain the deletion of 5827 to 5970, mutant V1), or residues 5746–5757 and 5899–5912 (deletion of 5758–5898, mutant V3), or residues 5804–5814 and 5959–5992 (deletion of 5815–5958, mutant V19–4), or residues 5782–5802 and 5944–5958 (deletion of 5803–5943, mutant V15–9). Each pair of amplification products was shuffled by PCR amplification with internal primers spanning residues 4189 to 4211 (sense) and 7019 to 6997 (antisense). The mutagenised PCR products were inserted into pMT28, previously digested with BglII and ClaI (located at positions 4199 and 7003, respectively), by recombination using the In–Fusion Dry and Down Mix kit (Clontech).

For the construction of mutants encoding R55W in 2C (that correspond to mutation C4507U) two additional cloning sites, NotI (position 5412) and NsiI (position 6393), were introduced into pMT28, leading to plasmid pMT30. We designed mutagenic primers that introduced substitutions C5412G and C5418G (NotI site) and G6393A (NsiI site), without affecting the coding sequence. To obtain plasmid FMDV WT (R55W), the resulting pMT30 plasmid was digested with BglII (position 4199) and NotI (position 5412). RT–PCR amplification of RNA from population V19–4A at passage 27, that encodes 2C with R55W, was performed with primers spanning residues 4189 to 4211 (sense) and 5432 to 5392 (antisense). The amplification product was introduced into plasmid pMT30 by recombination using the InFusion Dry and Down Mix kit (Clontech).

Mutant FMDVs encoding one VPg and 2C with R55W were constructed by amplification of the mutant plasmids described above (V3, V19–4, V15–9) with primers spanning residues 5403 to 5442 (sense) and 6412 to 6374 (antisense). PCR amplification products were inserted into plasmid FMDV WT (R55W) previously digested with NotI and NsiI (positions 5412 and 6393, respectively) by recombination with InFusion Dry and Down Mix kit (Clontech).

Transfection of viral RNA transcripts into BHK-21 cells

Viral replication of each mutant FMDV was analysed by transfection of the corresponding transcript into BHK-21 cells. For this purpose we used both lipofection and electroporation methods. Lipofection was used to obtain mutant viral samples that were then passaged on BHK–21 cell monolayers, and also in plaque assays of viral transcripts. Electroporation was used to detect intracellular FMDV protein synthesis, and in the quantification of intracellular FMDV RNA synthesis relative to viral RNA release. Lipofection and electroporation were carried out as previously described [41], [62].

Extraction of RNA, cDNA synthesis, PCR amplification, and nucleotide sequencing

RNA was extracted from the supernatants of infected cells by treatment with Trizol (Invitrogen) as previously described [63]. Reverse transcription (RT) was carried out using avian myeloblastosis virus reverse transcriptase (Promega) or Transcriptor reverse transcriptase (Roche), and PCR amplification was performed using EHF DNA polymerase (Roche) as specified by the manufacturer. PCR amplifications to obtain mutant infectious clones were carried out using Pfu Turbo DNA polymerase (Stratagene) because of its high copying fidelity [64]. Nucleotide sequencing was carried out as described [61].

Plaque assays of viral samples

Plaque assays were performed as previously described [65]. Serial dilutions of viral samples were applied to confluent BHK–21 cell monolayers and incubated for 1 h at 37°C. Then, supernatants were removed and semisolid agar medium was added. Plaque development was permitted for 48 h post-infection.

Plaque assay of directly transfected infectious RNA transcripts

Serial dilutions of in vitro transcribed infectious RNA (10, 1 and 0.1 ng) were mixed with lipofectin (Invitrogen) as indicated by the manufacturer. The mixture was added to a monolayer of 2×106 BHK–21 cells and incubated for 2 h at 37°C. Then, the medium was removed, the cell monolayer was washed with DMEM and semisolid agar medium was added. Plaque development was permitted for either 48 or 72 h.

Quantification of viral RNA in cell culture samples

Quantitative real time RT–PCR was carried out using the Light Cycler RNA Master SYBR Green I kit (Roche), according to the instructions of the manufacturer. Oligonucleotides spanning residues 3175 to 3194 (sense orientation) and 3518 to 3496 (antisense orientation), or 4924 to 4944 (sense) and 5026 to 5047 (antisense) were used in the amplification. Quantification was relative to a standard curve obtained with known amounts of FMDV RNA, synthesised by in vitro transcription of the infectious plasmid pMT28 [41]. For the quantification of extracellular FMDV RNA, the supernatants were collected, freed from detached cells by centrifugation, and extracted with Trizol (Invitrogen).

Metabolic labelling of viral proteins

To detect viral protein synthesis in transfected cells, 2 to 8 µg of viral transcript were electroporated into 2×106 BHK–21 cells. Electroporated cells were incubated for 3–4 hours in culture medium (DMEM, 1% FCS). Then, the medium was removed and cells were incubated for 1 h at 37°C in DMEM without Met and Cys, but supplemented with [35S] Met–Cys (Perkin–Elmer) (400 mCi/mmol). After metabolic labeling, cells were collected in 0.1 ml of sample buffer [160 mM Tris–HCl pH 6.8; 2% SDS; 11% glycerol; 0.1 M DTT, 0.033% bromophenol–blue]. The samples were heated at 90°C for 5 min and aliquots were subjected to SDS–PAGE (15% acrylamide).

Western blot assays

Proteins were transferred to a 0.45 µm pore size nitrocellulose membrane (BioRad). Western blots were developed with the following antibodies at a dilution of 1[ratio]2,000: mouse monoclonal anti–2C (1C8) and anti–3C (2D2) (a gift from E. Brocchi, Istituto Zooprofilattico Sperimentale della Lombardia e dell'Emilia Romagna, Brescia, Italy) and rabbit polyclonal anti–3D. Goat anti–rabbit IgG antibody coupled to peroxidase and goat anti–mouse IgG antibody coupled to peroxidase (Pierce) were used at 1[ratio]10,000 dilutions. Each sample was analysed by Western blot to identify virus–specific proteins, following previously described procedures [62].


We are indebted to E. Brocchi for the generous supply of MAbs, and to A. I. de Ávila and E. García–Cueto for expert technical assistance.


Competing Interests: The authors have declared that no competing interests exist.

Funding: Work supported by grants BFU2008-02816/BMC from Ministerio de Ciencia e Innovacion (MICINN), and Fundacion R. Areces. CIBERehd (Centro de Investigacion Biomedica en Red de Enfermedades Hepaticas y Digestivas) is funded by Instituto de Salud Carlos III. AA was supported by an I3P contract from Centro Superior de Investigaciones Cientificas (CSIC) and CP is the recipient of a contract from CIBERehd. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.


1. Lewin B. Sudbury, Massachusetts: Jones & Bartlett Publishers Inc; 2007. Genes.
2. Ball LA. Virus replication strategies. In: Knipe DM, Howley PM, editors. Fields Virology. Philadelphia: LWW; 2007. pp. 119–140. 5th ed.
3. Ferrer-Orta C, Arias A, Agudo R, Perez-Luque R, Escarmis C, et al. The structure of a protein primer-polymerase complex in the initiation of genome replication. EMBO J. 2006;25:880–888. [PubMed]
4. Kamtekar S, Berman AJ, Wang J, Lazaro JM, de Vega M, et al. The phi29 DNA polymerase:protein-primer structure suggests a model for the initiation to elongation transition. EMBO J. 2006;25:1335–1343. [PubMed]
5. Salas M. Protein-priming of DNA replication. Annu Rev Biochem. 1991;60:39–71. [PubMed]
6. Paul AV, Wimmer E. Possible unifying mechanism of picornavirus genome replication; In: Semler BL, editor. Washington DC: ASM Press; 2002. pp. 227–246.
7. Paul AV, van Boom JH, Filippov D, Wimmer E. Protein-primed RNA synthesis by purified poliovirus RNA polymerase. Nature. 1998;393:280–284. [PubMed]
8. Forss S, Schaller H. A tandem repeat gene in a picornavirus. Nucleic Acids Res. 1982;10:6441–6450. [PMC free article] [PubMed]
9. King AM, Sangar DV, Harris TJ, Brown F. Heterogeneity of the genome-linked protein of foot-and-mouth disease virus. J Virol. 1980;34:627–634. [PMC free article] [PubMed]
10. Nayak A, Goodfellow IG, Belsham GJ. Factors required for the Uridylylation of the foot-and-mouth disease virus 3B1, 3B2, and 3B3 peptides by the RNA-dependent RNA polymerase (3Dpol) in vitro. J Virol. 2005;79:7698–7706. [PMC free article] [PubMed]
11. Falk MM, Sobrino F, Beck E. VPg gene amplification correlates with infective particle formation in foot-and-mouth disease virus. J Virol. 1992;66:2251–2260. [PMC free article] [PubMed]
12. Pacheco JM, Henry TM, O'Donnell VK, Gregory JB, Mason PW. Role of nonstructural proteins 3A and 3B in host range and pathogenicity of foot-and-mouth disease virus. J Virol. 2003;77:13017–13027. [PMC free article] [PubMed]
13. Cao X, Kuhn RJ, 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]
14. Cao X, Wimmer E. Genetic variation of the poliovirus genome with two VPg coding units. EMBO J. 1996;15:23–33. [PubMed]
15. Ryan MD, Donnelly MLL, Flint M, Cowton VM, Luke G, et al. Foot-and-mouth disease virus proteinases. In: Sobrino F, Domingo E, editors. Foot-and-mouth disease: current perspectives. Wymondham, UK: Horizon Bioscience; 2004. pp. 53–76.
16. Nayak A, Goodfellow IG, Woolaway KE, Birtley J, Curry S, et al. Role of RNA structure and RNA binding activity of foot-and-mouth disease virus 3C protein in VPg uridylylation and virus replication. J Virol. 2006;80:9865–9875. [PMC free article] [PubMed]
17. Marcotte LL, Wass AB, Gohara DW, Pathak HB, Arnold JJ, et al. Crystal structure of poliovirus 3CD protein: virally encoded protease and precursor to the RNA-dependent RNA polymerase. J Virol. 2007;81:3583–3596. [PMC free article] [PubMed]
18. Paul AV, Peters J, Mugavero J, Yin J, van Boom JH, et al. Biochemical and genetic studies of the VPg uridylylation reaction catalyzed by the RNA polymerase of poliovirus. J Virol. 2003;77:891–904. [PMC free article] [PubMed]
19. Tellez AB, Crowder S, Spagnolo JF, Thompson AA, Peersen OB, et al. Nucleotide channel of RNA-dependent RNA polymerase used for intermolecular uridylylation of protein primer. J Mol Biol. 2006;357:665–675. [PubMed]
20. Rodriguez PL, Carrasco L. Poliovirus protein 2C has ATPase and GTPase activities. J Biol Chem. 1993;268:8105–8110. [PubMed]
21. Rodriguez PL, Carrasco L. Poliovirus protein 2C contains two regions involved in RNA binding activity. J Biol Chem. 1995;270:10105–10112. [PubMed]
22. Banerjee R, Echeverri A, Dasgupta A. Poliovirus-encoded 2C polypeptide specifically binds to the 3′-terminal sequences of viral negative-strand RNA. J Virol. 1997;71:9570–9578. [PMC free article] [PubMed]
23. Mirzayan C, Wimmer E. Biochemical studies on poliovirus polypeptide 2C: evidence for ATPase activity. Virology. 1994;199:176–187. [PubMed]
24. Steil BP, Barton DJ. Cis-active RNA elements (CREs) and picornavirus RNA replication. Virus Res. 2009;139:240–252. [PMC free article] [PubMed]
25. Vance LM, Moscufo N, Chow M, Heinz BA. Poliovirus 2C region functions during encapsidation of viral RNA. J Virol. 1997;71:8759–8765. [PMC free article] [PubMed]
26. Li JP, Baltimore D. An intragenic revertant of a poliovirus 2C mutant has an uncoating defect. J Virol. 1990;64:1102–1107. [PMC free article] [PubMed]
27. Cho MW, 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]
28. Teterina NL, Gorbalenya AE, Egger D, Bienz K, Ehrenfeld E. Poliovirus 2C protein determinants of membrane binding and rearrangements in mammalian cells. J Virol. 1997;71:8962–8972. [PMC free article] [PubMed]
29. 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]
30. Schlegel A, Giddings TH, Jr, Ladinsky MS, Kirkegaard K. Cellular origin and ultrastructure of membranes induced during poliovirus infection. J Virol. 1996;70:6576–6588. [PMC free article] [PubMed]
31. Suhy DA, Giddings TH, Jr, Kirkegaard K. Remodeling the endoplasmic reticulum by poliovirus infection and by individual viral proteins: an autophagy-like origin for virus-induced vesicles. J Virol. 2000;74:8953–8965. [PMC free article] [PubMed]
32. Teterina NL, Rinaudo MS, Ehrenfeld E. Strand-specific RNA synthesis defects in a poliovirus with a mutation in protein 3A. J Virol. 2003;77:12679–12691. [PMC free article] [PubMed]
33. Yin J, Liu Y, Wimmer E, Paul AV. Complete protein linkage map between the P2 and P3 non-structural proteins of poliovirus. J Gen Virol. 2007;88:2259–2267. [PubMed]
34. Xiang W, Cuconati A, Hope D, Kirkegaard K, Wimmer E. Complete protein linkage map of poliovirus P3 proteins: interaction of polymerase 3Dpol with VPg and with genetic variants of 3AB. J Virol. 1998;72:6732–6741. [PMC free article] [PubMed]
35. Lyle JM, Bullitt E, Bienz K, Kirkegaard K. Visualization and functional analysis of RNA-dependent RNA polymerase lattices. Science. 2002;296:2218–2222. [PubMed]
36. Garcia-Briones M, Rosas MF, Gonzalez-Magaldi M, Martin-Acebes MA, Sobrino F, et al. Differential distribution of non-structural proteins of foot-and-mouth disease virus in BHK-21 cells. Virology. 2006;349:409–421. [PubMed]
37. Monaghan P, Cook H, Jackson T, Ryan M, Wileman T. The ultrastructure of the developing replication site in foot-and-mouth disease virus-infected BHK-38 cells. J Gen Virol. 2004;85:933–946. [PubMed]
38. Verdaguer N, Fita I, Domingo E, Mateu MG. Efficient neutralization of foot-and-mouth disease virus by monovalent antibody binding. J Virol. 1997;71:9813–9816. [PMC free article] [PubMed]
39. Perales C, Agudo R, Domingo E. Counteracting quasispecies adaptability: extinction of a ribavirin-resistant virus mutant by an alternative mutagenic treatment. PLoS One. 2009;4:e5554. [PMC free article] [PubMed]
40. Herrera M, Garcia-Arriaza J, Pariente N, Escarmis C, Domingo E. Molecular basis for a lack of correlation between viral fitness and cell killing capacity. PLoS Pathog. 2007;3:e53. [PMC free article] [PubMed]
41. Garcia-Arriaza J, Manrubia SC, Toja M, Domingo E, Escarmis C. Evolutionary transition toward defective RNAs that are infectious by complementation. J Virol. 2004;78:11678–11685. [PMC free article] [PubMed]
42. Carrillo C, Tulman ER, Delhon G, Lu Z, Carreno A, et al. Comparative genomics of foot-and-mouth disease virus. J Virol. 2005;79:6487–6504. [PMC free article] [PubMed]
43. Yin J, Paul AV, Wimmer E, Rieder E. Functional dissection of a poliovirus cis-acting replication element [PV-cre(2C)]: analysis of single- and dual-cre viral genomes and proteins that bind specifically to PV-cre RNA. J Virol. 2003;77:5152–5166. [PMC free article] [PubMed]
44. Domingo E, Escarmís C, Menéndez-Arias L, Perales C, Herrera M, et al. Viral quasispecies: dynamics, interactions and pathogenesis. In: Domingo E, Parrish C, Holland JJ, editors. Origin and Evolution of Viruses. Oxford, UK: Elsevier; 2008. pp. 87–118. 2nd ed.
45. Grubman MJ, Moraes MP, Diaz-San Segundo F, Pena L, de los Santos T. Evading the host immune response: how foot-and-mouth disease virus has become an effective pathogen. FEMS Immunol Med Microbiol. 2008;53:8–17. [PubMed]
46. Sierra S, Davila M, Lowenstein PR, Domingo E. Response of foot-and-mouth disease virus to increased mutagenesis: influence of viral load and fitness in loss of infectivity. J Virol. 2000;74:8316–8323. [PMC free article] [PubMed]
47. Belsham GJ, Normann P. Dynamics of picornavirus RNA replication within infected cells. J Gen Virol. 2008;89:485–493. [PubMed]
48. Pariente N, Airaksinen A, Domingo E. Mutagenesis versus inhibition in the efficiency of extinction of foot-and-mouth disease virus. J Virol. 2003;77:7131–7138. [PMC free article] [PubMed]
49. Baranowski E, Sevilla N, Verdaguer N, Ruiz-Jarabo CM, Beck E, et al. Multiple virulence determinants of foot-and-mouth disease virus in cell culture. J Virol. 1998;72:6362–6372. [PMC free article] [PubMed]
50. Martinez-Salas E, Saiz JC, Davila M, Belsham GJ, Domingo E. A single nucleotide substitution in the internal ribosome entry site of foot-and-mouth disease virus leads to enhanced cap-independent translation in vivo. J Virol. 1993;67:3748–3755. [PMC free article] [PubMed]
51. Escarmis C, Perales C, Domingo E. Biological effect of Muller's Ratchet: distant capsid site can affect picornavirus protein processing. J Virol. 2009;83:6748–6756. [PMC free article] [PubMed]
52. Ferrer-Orta C, Arias A, Escarmis C, Verdaguer N. A comparison of viral RNA-dependent RNA polymerases. Curr Opin Struct Biol. 2006;16:27–34. [PubMed]
53. Wang YF, Chou CT, Lei HY, Liu CC, Wang SM, et al. A mouse-adapted enterovirus 71 strain causes neurological disease in mice after oral infection. J Virol. 2004;78:7916–7924. [PMC free article] [PubMed]
54. Harris JR, Racaniello VR. Changes in rhinovirus protein 2C allow efficient replication in mouse cells. J Virol. 2003;77:4773–4780. [PMC free article] [PubMed]
55. Harris JR, Racaniello VR. Amino acid changes in proteins 2B and 3A mediate rhinovirus type 39 growth in mouse cells. J Virol. 2005;79:5363–5373. [PMC free article] [PubMed]
56. Sanz-Ramos M, Diaz-San Segundo F, Escarmis C, Domingo E, Sevilla N. Hidden virulence determinants in a viral quasispecies in vivo. J Virol. 2008;82:10465–10476. [PMC free article] [PubMed]
57. Nuñez JI, Baranowski E, Molina N, Ruiz-Jarabo CM, Sanchez C, et al. A single amino acid substitution in nonstructural protein 3A can mediate adaptation of foot-and-mouth disease virus to the guinea pig. J Virol. 2001;75:3977–3983. [PMC free article] [PubMed]
58. Toja M, Escarmis C, Domingo E. Genomic nucleotide sequence of a foot-and-mouth disease virus clone and its persistent derivatives. Implications for the evolution of viral quasispecies during a persistent infection. Virus Res. 1999;64:161–171. [PubMed]
59. Escarmis C, Davila M, Charpentier N, Bracho A, Moya A, et al. Genetic lesions associated with Muller's ratchet in an RNA virus. J Mol Biol. 1996;264:255–267. [PubMed]
60. Arias A, Agudo R, Ferrer-Orta C, Perez-Luque R, Airaksinen A, et al. Mutant viral polymerase in the transition of virus to error catastrophe identifies a critical site for RNA binding. J Mol Biol. 2005;353:1021–1032. [PubMed]
61. Sierra M, Airaksinen A, Gonzalez-Lopez C, Agudo R, Arias A, et al. Foot-and-mouth disease virus mutant with decreased sensitivity to ribavirin: implications for error catastrophe. J Virol. 2007;81:2012–2024. [PMC free article] [PubMed]
62. Perales C, Mateo R, Mateu MG, Domingo E. Insights into RNA virus mutant spectrum and lethal mutagenesis events: replicative interference and complementation by multiple point mutants. J Mol Biol. 2007;369:985–1000. [PubMed]
63. Arias A, Ruiz-Jarabo CM, Escarmis C, Domingo E. Fitness increase of memory genomes in a viral quasispecies. J Mol Biol. 2004;339:405–412. [PubMed]
64. Cline J, Braman JC, Hogrefe HH. PCR fidelity of pfu DNA polymerase and other thermostable DNA polymerases. Nucleic Acids Res. 1996;24:3546–3551. [PMC free article] [PubMed]
65. Arias A, Lazaro E, Escarmis C, Domingo E. Molecular intermediates of fitness gain of an RNA virus: characterization of a mutant spectrum by biological and molecular cloning. J Gen Virol. 2001;82:1049–1060. [PubMed]
66. Perales C, Agudo R, Tejero H, Manrubia SC, Domingo E. Potential benefits of sequential inhibitor-mutagen treatments of RNA virus infections. PLoS Pathog. 2009;5:e1000658. [PMC free article] [PubMed]

Articles from PLoS ONE are provided here courtesy of Public Library of Science