Identification and Characterization of a Poliovirus Capsid Mutant with Enhanced Thermal Stability

Viruses spread by the fecal-oral route need to maintain viability in the environment to ensure transmission. Previous work indicated that bacteria and bacterial surface polysaccharides can stabilize viral particles and enhance transmission. To explore factors that influence viral particle stability, we isolated a mutant poliovirus that is heat resistant. This mutant virus does not require feces for stability at most temperatures but can be stabilized by feces at very high temperatures. Even though the mutant virus is heat resistant, it is susceptible to inactivation by treatment with bleach. This work provides insight into how viral particles maintain infectivity in the environment.

ABSTRACT

Enteric viruses, including poliovirus, are spread by the fecal-oral route. In order to persist and transmit to a new host, enteric virus particles must remain stable once they are in the environment. Environmental stressors such as heat and disinfectants can inactivate virus particles and prevent viral transmission. It has been previously demonstrated that bacteria or bacterial surface glycans can enhance poliovirus virion stability and limit inactivation from heat or bleach. While investigating the mechanisms underlying bacterially enhanced virion thermal stability, we identified and characterized a poliovirus (PV) mutant with increased resistance to heat inactivation. The M132V mutant harbors a single amino acid change in the VP1 capsid coding that is sufficient to confer heat resistance but not bleach resistance. Although the M132V virus was stable in the absence of bacteria or feces at most temperatures, M132V virus was stabilized by feces at very high temperatures. M132V PV had reduced specific infectivity and RNA uncoating compared with those of wild-type (WT) PV, but viral yields in HeLa cells were similar. In orally inoculated mice, M132V had a slight fitness cost since fecal titers were lower and 12.5% of fecal viruses reverted to the WT. Overall, this work sheds light on factors that influence virion stability and fitness.

IMPORTANCE Viruses spread by the fecal-oral route need to maintain viability in the environment to ensure transmission. Previous work indicated that bacteria and bacterial surface polysaccharides can stabilize viral particles and enhance transmission. To explore factors that influence viral particle stability, we isolated a mutant poliovirus that is heat resistant. This mutant virus does not require feces for stability at most temperatures but can be stabilized by feces at very high temperatures. Even though the mutant virus is heat resistant, it is susceptible to inactivation by treatment with bleach. This work provides insight into how viral particles maintain infectivity in the environment.

INTRODUCTION

Enteric viruses such as poliovirus (PV) cause mild to severe diseases in humans. PV is a nonenveloped, single-stranded 7.5-kb positive-sense RNA virus in the Enterovirus genus of the Picornaviridae family. Spread by the fecal-oral route, PV replicates in the gastrointestinal tract and can disseminate and cause neuronal damage and subsequent paralysis in the host. Although highly successful vaccines prevent poliomyelitis in most countries, PV serves as a useful model system to understand fundamental aspects of virology.

PV’s 30-nm icosahedral capsid is comprised of 60 copies of each of the viral capsid proteins VP1, VP2, VP3, and VP4 (1). VP1, VP2, and VP3 form a network that encompasses the surface of the capsid, while VP4 is on the interior of the virion (1, 2). VP1 contains a highly conserved lipid moiety pocket that is associated with virion uncoating (3). Like those of many nonenveloped viruses, PV’s structure is dynamic at physiological temperatures, and internal capsid amino acids are reversibly exposed in a process called “breathing” (4–6). These transient events may be an important precursor to the normal entry process. However, these conformational changes can also lead to irreversible premature RNA release prior to cell entry.

Previous work elucidating capsid stability determinants has led to the identification of thermal stability mutants and stabilizing compounds (7–11). For example, Adeyemi et al. identified two capsid mutations in VP1 (V87A and I194V) that enhance virion stability by limiting premature uncoating and release of viral RNA at high temperatures (9), and work from Shiomi et al. also suggested that VP1 mutation V87A contributes to virion heat resistance (8).

Because PV is transmitted by the fecal-oral route, PV particle stability in the environment is crucial for viral persistence and transmission to new hosts. Furthermore, recent studies revealed that the intestinal microbiota promotes PV infection by increasing virion stability and cell attachment (12–14). A PV mutant with reduced lipopolysaccharide binding (VP1-T99K) had a transmission defect, exhibiting reduced stability in feces compared to that of wild-type (WT) PV, without displaying any defects in cell attachment and replication in vitro or pathogenesis in mice. Moreover, it was determined that PV interacts with bacteria by binding to bacterial surface glycans, including lipopolysaccharide and peptidoglycan (12, 13), enhancing virion stability by preventing premature RNA release (12). While it is clear that binding to bacteria or bacterial surface glycans enhances thermal stability of PV virions, the specific interaction determinants between bacterial components and virion capsids—and therefore the mechanism of thermal stabilization by bacteria—remain unknown.

In this study, we exploited a PV mutant that is thermostable in the absence of bacteria to explore mechanisms of bacterially mediated PV thermostability. By selecting for variants that maintained infectivity after heat exposure, we identified a single amino acid substitution in the VP1 viral capsid protein, VP1-M132V, that is sufficient to increase virion thermal stability. We found that although the M132V mutant is generally stable independently of bacteria or feces, at higher temperatures fecal contents rescued M132V viruses from thermal inactivation. The M132V mutation did not protect the virion from other environmental stressors, such as bleach inactivation, suggesting that the M132V phenotype is specific to heat resistance. Finally, we examined whether enhanced virion stability confers a fitness cost. Although M132V PV had reduced specific infectivity and RNA uncoating, viral yields in HeLa cells were similar. However, there was evidence for a fitness defect of M132V PV in orally inoculated mice.

(This article was submitted to an online preprint archive [15].)

RESULTS

Identification of a thermostable PV mutant.

Previously we found that binding to bacteria or bacterial surface polysaccharides can reduce thermal inactivation of PV (12, 13). Additionally, we showed that a single amino acid substitution in the VP1 viral capsid protein, T99K, diminished lipopolysaccharide binding and virion stabilization by lipopolysaccharide (12). To further define factors that influence virion stability, we sought to select for PV mutants with increased thermal stability in the absence of bacteria by using repeated exposure to elevated temperatures followed by amplification of viable viruses. PV was incubated at 43°C in phosphate-buffered saline (PBS) for 6 h, followed by quantification of viable virus by plaque assay. Under these conditions, the infectivity of WT PV was reduced by 99.99% compared to that of no-heat-treatment controls (Fig. 1A). Remaining viable viruses were amplified in HeLa cells, followed by incubation at 43°C in PBS for 6 h. This cycle was repeated a total of 10 times. By passage 5 (P5), the heat-exposed viruses had 2-fold-increased viability after heat treatment compared to the WT virus, and by passage 8 (P8), the heat-exposed viruses had 500-fold-increased viability after heat treatment compared to the WT virus (Fig. 1A). After passage 10, we isolated RNA from the heat-passaged viruses, performed reverse transcription-PCR (RT-PCR), and sequenced the entire capsid-coding region (VP1, -2, -3, and -4). Sequence alignment revealed a single amino acid change, M132V in the VP1 capsid protein (Fig. 1B), along with two silent mutations in VP1. The VP1-M132V mutation is buried inside the capsid within the 8-stranded β-barrels of VP1 (Fig. 1C).

FIG 1.

Selection and identification of a thermostable PV mutant. (A) Thermal stability profiles for WT PV, passage 5 and 8 viruses, and the VP1-M132V mutant. Viruses were incubated in PBS at 43°C for 6 h, followed by plaque assays. Titers from time zero were compared to postincubation titers to determine the percent input PFU. For the WT and M132V mutant, n = 6; for the P5 and P8 viruses, n = 1 because the remainder of the samples were used for the next passage. (B) Genome schematic of PV, with the M132V mutation in VP1 indicated in red. (C) Poliovirus structure showing one 5-fold symmetry axis, with the location of VP1-M132V highlighted in red. The expanded view shows a ribbon model of VP1 with the position of M132 amino acid shown in red. Data are means ± SEM. **, P < 0.01.

A single amino acid change, VP1-M132V, is sufficient for thermal resistance.

To confirm that the VP1-M132V mutation confers thermal resistance, we cloned the M132V mutation into the PV infectious clone and examined viability by plaque assay after incubation at 43°C for 6 h. As shown in Fig. 1A, the M132V amino acid change was sufficient to increase infectivity more than 1,000-fold compared to WT PV. As a second method to quantify virion stability using a physical/cell-free approach, we examined whether the M132V mutation alters viral RNA release using a particle stability thermal release assay (PaSTRY) (12, 16). Gradient-purified WT or M132V PV was mixed with SYBR green and heated in a real-time PCR machine with fluorescence monitoring. SYBR green binds viral RNA upon release from capsids and can be used to determine the temperature at which RNA release occurs. WT PV generated a peak of fluorescence intensity, indicating RNA release at 51°C (Fig. 2). However, for M132V PV the fluorescence intensity peak shifted to 78°C (Fig. 2). Overall, these data indicate that the M132V mutation in VP1 enhances viral stability by limiting RNA release.

FIG 2.

Particle stability thermo release assay (PaSTRy). Gradient-purified viruses were mixed with SYBR green and placed in a real-time machine, where samples were heated from 25°C to 99°C on a 1% stepwise gradient with fluorescent monitoring. Peaks of SYBR green fluorescence indicate virion RNA release. (A) Plots of SYBR green fluorescence over increasing temperatures. AU, arbitrary units. (B) Quantification of RNA release temperatures for three independent PaSTRy experiments. Data are means ± SEM.

Feces can stabilize M132V PV at higher temperatures.

Since PV spreads by the fecal-oral route, virion stability in feces is important for transmission to a new host. Previously we demonstrated that feces stabilize WT PV (12). Because M132V PV has enhanced thermal stability, we suspected that M132V virions do not require feces for stability in the environment. To examine environmental stability of M132V and WT PV, we collected feces from uninfected C57BL/6 PVR-IFNAR^−/− mice, resuspended them in PBS, and incubated this mixture with 10⁵ PFU of M132V or WT PV at 37°C prior to quantification of viable virus by plaque assay over several time points. As shown in Fig. 3A, WT PV was stabilized by feces over time, consistent with our previous findings that bacteria or bacterial components in feces enhance virion stability (12, 13). In contrast, M132V PV was stable even in the absence of feces, although fecal components slightly increased the virion stability of the mutant at a late time point (day 8) (Fig. 3A).

FIG 3.

Thermal stability of WT and M132V PVs in the presence or absence of feces. (A) Thermal stability profiles of WT and M132V PVs at 37°C. Feces were collected from uninfected C57BL/6 PVR-IFNAR^−/− mice and resuspended in PBS prior to mixing with either WT or M132V PV. Samples were incubated at 37°C and plaque assays were performed at 0, 2, 4, and 8 days postincubation (n = 6). (B) Thermal stability profiles of WT and M132V PVs at various temperatures in the presence or absence of feces during 6 h of incubation. Plaque assays were performed pre- and post-heat treatment (n = 4). Data are means ± SEM. Statistical analysis was performed using 2-way ANOVA. An asterisk indicates that one or more of samples was statistically significantly different (P < 0.05) from the WT PBS group. Data are means ± SEM.

Since we selected for the M132V mutant at 43°C, it was possible that stability by fecal components was masked by the mutant’s inherent stability at 37°C. To investigate this, we examined the stability of WT and M132V PVs during 6-h incubations at higher temperatures in the presence or absence of feces. As shown in Fig. 3B, after 6 h of incubation at 43°C in PBS, WT PV viability was reduced >1,000-fold, whereas M132V PV viability was reduced <2-fold. Feces stabilized WT PV up to an incubation temperature of 47°C. While M132V PV was stable in PBS up to 47°C, exposure to feces allowed recovery of 20% viable virus at 49°C and 1% viable virus at 51°C. Thus, for both WT and M132V viruses, feces stabilized viruses at temperatures that are damaging to the virions.

WT and M132V PVs have equivalent inactivation following bleach treatment.

Hyperstable virions could be dangerous to human health, as they can persist longer in the environment and potentially resist inactivation by disinfectants such as bleach (12). Given that M132V PV is more thermostable than WT PV, we wanted to determine if the mutant is resistant to bleach inactivation. We treated 10⁵ PFU of either WT or M132V PV with 0.0001% bleach for 1 min, the bleach was neutralized with sodium thiosulfate, and we performed plaque assays to determine the amount of viable virus before and after treatment. WT and M132V PVs had equivalent degrees of inactivation following bleach treatment (Fig. 4). This result suggests that the M132V mutation confers resistance to heat but not resistance to bleach.

FIG 4.

Inactivation of WT and M132V PVs in the presence of bleach. WT and M132V PVs were treated with 0.0001% fresh bleach for 1 min, followed by neutralization. Plaque assays were performed to determine the amount of viable virus before and after treatment. Data are means ± SEM.

M132V PV has reduced RNA uncoating in cultured cells and reduced specific infectivity.

Because the M132V mutation can increase viral stability by delaying RNA release during heat exposure and RNA release is critical for infection, we wanted to determine whether this mutant has a defect in uncoating its RNA genome during infection. To quantify uncoating efficiency, we used neutral red-containing viruses. Viruses propagated in the presence of neutral red dye are light sensitive due to cross-linking of virion RNA with dye in the viral particle. However, upon RNA release into the cytosol during infection, the dye diffuses and infection is no longer inhibited by light exposure. Thus, by comparing viral yields of light- versus dark-exposed cells, the percent uncoating can be determined. We infected HeLa cells in the dark with approximately 100 PFU of neutral red M132V or WT PV, and at 30, 60, or 120 min postinfection cells were either kept in the dark or exposed to light, followed by addition of an agar overlay. The percent uncoating was calculated by dividing the number of PFU on light-exposed cells by the number of PFU on dark-exposed cells and multiplying by 100. As shown in Fig. 5, M132V PV had approximately 2-fold less uncoating than WT PV at all time points tested, indicating that M132V PV has a slight uncoating defect.

FIG 5.

RNA uncoating efficiency during infection of HeLa cells. HeLa cells were infected with ∼100 PFU of neutral red/light-sensitive WT or M132V PV in the dark. Cells were exposed to light or dark for 15 min at the indicated time points, followed by addition of an agar overlay for plaque assay. The graph represents the percentage of uncoated virus (ratio of light-exposed PFU to dark-exposed PFU × 100). Data are means ± SEM. *, P < 0.05.

Due to its uncoating defect, we hypothesized that M132V PV has reduced specific infectivity compared with that of WT PV. To test this, we quantified viral RNA by quantitative RT-PCR (qRT-PCR) as a proxy for particle number. In two independent experiments, our WT PV stock averaged 177 RNA copies per PFU, while the M132V PV stock averaged 992 RNA copies per PFU. Therefore, M132V PV particles are 5.6-fold less infectious than WT PV particles.

Rates of replication of M132V and WT PVs are comparable in cultured cells.

Given that M132V PV had reduced RNA uncoating and specific infectivity, we wanted to determine if this mutant has a growth defect compared to WT PV. We first compared the replication of M132V and WT PV in cell culture using single-cycle growth curve assays. HeLa cells were infected at a multiplicity of infection (MOI) of 10, and viral titers were determined over time by plaque assay. As shown in Fig. 6A, we were surprised to find that the M132V mutation did not significantly affect viral yields in HeLa cells. To determine whether a replication defect emerges over multiple cycles of replication, we infected HeLa cells at an MOI of 0.01 and quantified viral yields over a 48-h time course. Again, the M132V mutation did not reduce viral yields compared to those of the WT (Fig. 6B). Moreover, plaque sizes were similar for the WT and M132V viruses, suggesting that the M132V virus does not have a major replication defect (Fig. 6C). These data suggest that, in spite of M132V virus having a defect in RNA uncoating and reduced specific infectivity, replication of M132V PV is comparable to that of WT PV.

FIG 6.

Comparison of WT and M132V PV replication in HeLa cells. (A) Single-cycle growth curve assay. HeLa cells were infected with WT or M132V PV at an MOI of 10 at 37°C. Samples were harvested at various time points and yields were quantified by plaque assays (n = 6). (B) Multicycle growth curve assay. HeLa cells were infected with WT or M132V PV at an MOI of 0.01 at 37°C. Samples were harvested at various time points and yields were quantified by plaque assays (n = 6). (C) Representative plaques of WT and M132V PVs, stained at approximately 48 hpi.

Fitness of M132V PV in mice.

Next we wanted to determine whether the M132V mutant has a fitness defect in mice by examining viral shedding and pathogenesis in orally inoculated C57BL/6 PVR-IFNAR^−/− mice. These mice express the human poliovirus receptor (PVR) and are deficient for the interferon alpha/beta (IFN-α/β) receptor (17), which confers oral susceptibility to PV. Mice were orally inoculated with 10⁸ PFU of either WT or M132V PV, fecal samples were collected at 24, 48, and 72 h postinoculation (hpi), and titers were determined by plaque assay. As shown in Fig. 7A, we found that fecal titers were similar for the WT and M132V viruses, although M132V PV titers were lower than WT PV titers at 48 hpi. However, the levels of pathogenesis were equivalent for the WT and M132V PVs (Fig. 7B). It is possible that the M132V mutation reverted to the WT during replication in mice, which could mask potential replication and pathogenesis defects. To determine the reversion frequency of M132V PV in mice, we performed sequence analysis on cloned RT-PCR products from 72-hpi fecal samples from two M132V PV-inoculated mice (Table 1). A majority of cloned sequences, 21 of 24, contained the M132V mutation. However, 3 of 24 cloned sequences had reverted to M132. Given that 12.5% of fecal viruses had undergone reversion of the M132V mutation at 72 hpi and M132V PV titers were reduced at 48 hpi, it is likely that the mutant has reduced fitness in vivo.

FIG 7.

Replication and pathogenesis of WT and M132V PV in orally inoculated mice. (A) Fecal shedding kinetics of WT and M132V viruses. C57BL/6 PVR-IFNAR^−/− mice were orally inoculated with 10⁸ PFU of WT or M132V virus, and feces were collected at 24, 48, and 72 hpi prior to quantification of viral titer by plaque assay. (B) Survival of C57BL/6 PVR-IFNAR^−/− mice following oral inoculation of WT or M132V poliovirus. Survival curves were not significantly different (P > 0.5) by log rank test. Data in each panel are from two independent experiments with 18 to 20 animals for each virus. Data are means ± SEM. **, P < 0.01.

TABLE 1.

Reversion frequency of M132V PV in mice^a

Mouse	Frequency of M132 in mice inoculated with M132V (%)
1	1/10 (10)
2	2/14 (14.3)
Total	3/24 (12.5)

^aPV RT-PCR products from 72-hpi fecal samples from M132V-inoculated mice were cloned into plasmids, followed by sequence analysis to determine whether the codon was mutant (valine/GTG) or had reverted to the WT (methionine/ATG).

DISCUSSION

Virion stability in the environment is crucial for transmission of PV and other enteric viruses. Previous work has shown that the microbiota can enhance PV replication, shedding, and pathogenesis in mice and that the microbiota increases virion stability and cell attachment (12, 13). Specifically, our lab has shown that binding to the bacterial surface components lipopolysaccharide and peptidoglycan enhanced PV thermostability. How bacteria promote virion stability is still unclear. To explore mechanisms and consequences of virion stabilization, we selected for and characterized a hyperstable PV mutant.

Following repeated exposure to heat, we identified a single amino acid change, M132V, in capsid protein VP1 that limited thermal activation of PV. VP1-M132 is buried in the capsid interior. Although M132 is located in a relatively conserved region of VP1, leucine is present in this position in other picornaviruses, such as coxsackievirus B3, Aichi virus, and Mengo virus.

Other thermostable PV mutants have been identified by other groups. For example, Adeyemi et al. and Shiomi et al. both found that VP1-V87A confers heat resistance and Adeyemi et al. also found that VP1-I194V confers heat resistance (8, 9). Ours is the first study to identify VP1-M132V as a stability determinant, and neither VP1-V87A nor VP1-I194V was selected during our heat passage experiments. These results suggest that multiple amino acids in VP1 can contribute to heat resistance and that there are multiple paths to generate heat-resistant PV variants.

Our work has shown that bacteria and bacterial glycans limit inactivation from both heat and bleach treatment, which led us to hypothesize that bacterial glycans bind virions and limit virion conformational changes and “breathing” that can lead to premature RNA release (12, 13). Although the specific binding site(s) of bacterial components on viral capsids have not been identified, it is highly likely that surface-exposed residues are involved. Importantly, M132V PV is not defective in bacterial binding (E. R. Aguilera et al., submitted for publication), and exposure to feces limits thermal inactivation of M132V PV at very high temperatures (49°C/51°C [Fig. 3B]). Thus, M132V PV does not require feces or bacteria for stabilization at a wide range of temperatures but can be stabilized by feces at very high temperatures.

Although the rates of replication of M132V and WT PVs were equivalent in cultured cells, M132V had an RNA uncoating defect in HeLa cells (Fig. 5) and likely has reduced fitness in mice (Fig. 7; Table 1). Interestingly, the uncoating defect of M132V does not affect viral yields in cultured cells (Fig. 6). This could be due to enhanced stability of M132V progeny virions that offsets potentially lower total yields than with the WT. In mice, M132V PV fecal titers were 7-fold lower than WT PV titers at 48 hpi, but titers were comparable at 72 hpi (Fig. 7). However, since 12.5% of 72-hpi viruses in M132V-inoculated mice had undergone reversion to the WT (Table 1), the titer differences at 48 hpi may be more representative when considering fitness costs. Additionally, the slightly reduced M132V PV pathogenesis in Fig. 7 was not statistically significant, but given the reversion frequency, M132V virulence may be reduced. This could be tested in future studies with a virus engineered to be less revertable (e.g., ATG to GTA as the M132V mutation rather than ATG to GTG).

MATERIALS ANDMETHODS

Viruses and cells.

Serotype 1 Mahoney PV cell culture infections and plaque assays were performed using HeLa cells as previously described (18). HeLa cells were propagated in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% calf serum. Virus stocks from cellular lysates were derived from low-passage stocks. For PaSTRy experiments, PV stocks were CsCl gradient purified as previously described (13).

Selection, sequencing, and cloning of M132V PV.

To select for thermostable PV mutants, 10⁶ PFU of PV was incubated in PBS supplemented with 100 μg/ml of CaCl₂ and 100 μg/ml of MgCl₂ (PBS+) for 6 h at 43°C. Remaining viable viruses were amplified using HeLa cells for 12 to 16 h (until cytopathic effects were visible). Sequential heating and amplification of viable viruses were performed a total of 10 times. Viral RNA was isolated using TRIzol as recommended by the manufacturer’s protocol (Sigma), and cDNA covering the capsid-coding region was synthesized using SuperScript II reverse transcriptase (Invitrogen). PCRs were performed to amplify the coding regions of VP1 to -4, and products were sequenced by the University of Texas Southwestern Sequencing Core. The sequenced region revealed one amino acid change, M132V (ATG to GTG), in VP1 and two silent mutations in VP1 in some isolates (T135 [ACC to ACT and N203 [AAC to AAT]). To generate a virus containing only the VP1-M132V mutation, a 486-bp fragment containing the M132V mutation (and no other amino acid changes) was generated by SnaBI/NheI digestion of an RT-PCR product from the passage 10 virus. This fragment was cloned into a new PV plasmid at nucleotides (nt) 2470 (NheI) and 2956 (SnaBI). The new plasmid was sequenced and confirmed (nt 2470 to 2956). To generate M132V virus, the plasmid was transfected into HeLa cells along with a plasmid encoding the T7 DNA-dependent RNA polymerase to produce viral stocks as previously described (13).

PV thermal stability assays.

Feces were collected from 4- to 6-week-old female C57BL/6 PVR-IFNAR^−/− mice (17). Fecal pellets were resuspended in PBS+ to a final concentration of 0.0641 g/ml. For titer-based stability assays in Fig. 4A, 10⁵ PFU of PV was mixed with PBS+ or fecal slurry and incubated at 37°C for 2, 4, 6, or 8 days. For Fig. 4B, 10⁸ PFU of PV was mixed with PBS+ or fecal slurry and incubated at the indicated temperatures for 6 h. Titers were quantified by plaque assays using HeLa cells. PV titers from pre- and post-heat treatment were calculated to determine the percent input PFU. For PaSTRy experiments, one microgram of CsCl gradient-purified virus was mixed with SYBR Green II (10× final concentration) and buffer (10 mM HEPES [pH 8.0], 200 mM NaCl) to a final volume of 30 μl. Samples were heated from 25°C to 99°C on a 1% stepwise gradient with fluorescent monitoring using an ABI 7500 real-time instrument (12).

Bleach inactivation assay.

Bleach inactivation assays were performed as previously described (12). Briefly, 10⁵ PFU of each virus was mixed with PBS+ and incubated at 37°C for 1 h. After incubation, samples were processed in 0.0001% fresh bleach (diluted immediately before the experiment) for 1 min. Bleach was neutralized by adding equal volume of 0.01% sodium thiosulfate (Sigma). Plaque assays were performed to determine the amount of viable virus before and after treatment.

Viral uncoating assays.

Uncoating was quantified using neutral red PVs as previously described (18). Briefly, HeLa cells were infected with a total of ∼100 PFU of neutral red/light-sensitive WT or M132V PV. At the desired time postinfection, cells were either kept in the dark or exposed to light for 15 min. Light-exposed and unexposed viruses were quantified by plaque assays to determine the amount of uncoated virus by dividing light-exposed PFU by unexposed PFU and multiplying by 100. Data are means ± standard errors of the means (SEM).

Quantifying virus-specific infectivity.

To quantify the relative specific infectivity of WT and M132V PVs, RNA was extracted from ∼1 × 10⁹ PFU of each stock using Tri Reagent LS (Sigma) and a Zymo Direct-zol RNA kit, and quantification of PV RNA was performed using quantitative reverse transcription-PCR (qRT-PCR). Reverse transcription was performed with Superscript II (Invitrogen) using an antisense primer (5′-CGCAAGGTGCAATTGCAACGCAG-3′). Quantitative PCR (qPCR) was performed using 5 μl of the cDNA reaction mixture mixed with SYBR green PCR master mix reagent (Applied Biosystems) and a 10 μM concentration of each primer (5′-GTCGTCCCTCTTTCGACACCCAGAG-3′ and 5′-GGTTAGGTCAGATGCTTGAAAGCAT-3′) to amplify the VP1 capsid region using an Applied Biosystems 7500 instrument. Cycling conditions were 1 cycle for 2 min at 50°C and 10 min at 95°C, followed by 40 cycles, with 1 cycle consisting of 15 s at 95°C and 1 min at 60°C. The qPCRs were performed in triplicate from two independent RNA preparations and quantified using a standard curve generated with poliovirus plasmid DNA samples. Specific infectivity was determined by dividing 1 × 10⁹ PFU by the relative amount of RNA.

Growth curve assays.

Six-well plates were seeded with 2 × 10⁶ HeLa cells/well. Cells were inoculated with either WT or M132V PV at an MOI of 10 (for single-cycle infections) or 0.01 (for multicycle infections). After a 30-min incubation at 37°C, the inoculum was aspirated from the cells. Cell monolayers were washed with PBS and 3 ml of fresh medium was added. After 2, 4, 6, or 8 hpi at 37°C (for single-cycle infections) or 4, 8, 24, or 48 hpi at 37°C (for multicycle infections), cells were washed with PBS, trypsinized, and pelleted. Cells were freeze-thawed three times and the released intracellular viruses were quantified by plaque assay using HeLa cells.

Mouse experiments.

All animals were handled according to the Guide for the Care and Use of Laboratory Animals (19). All mouse studies were performed at UT Southwestern (Animal Welfare Assurance no. A3472-01) using protocols approved by the local Institutional Animal Care and Use Committee in a manner designed to minimize pain, and any animals that exhibited severe disease were euthanized immediately. Four- to 5-week-old female and C57BL/6 PVR-IFNAR^−/− mice were orally inoculated with 10⁸ PFU WT or M132V PV. Feces were collected at 24, 48, and 72 hpi and viral shedding in feces was quantified by plaque assays. Survival of infected animals was also monitored for 12 days postinoculation (dpi). Because PV disease is progressive and irreversible in this system, mice were euthanized at the first sign of disease, and these time points are shown in Fig. 7B.

Frequency of M132V reversion to the WT.

To determine the frequency of reversion of M132V mutation to the WT, RNA was extracted from feces collected from animals infected with M132V PV at 72 hpi using Tri Reagent LS (Sigma) and Zymo Direct-zol RNA kit. cDNA synthesis was performed using Superscript II (Invitrogen). A PCR product containing the VP1-132 region was cloned into a plasmid using the Zero blunt TOPO cloning kit (Invitrogen) and transformed into TOP10 cells. Plasmids isolated from 24 different colonies were sequenced to determine the percentage of the M132V mutant virus that had reverted to the WT.

Statistical analysis.

The differences between groups were examined by unpaired two-tailed Student t tests. A P value of <0.05 was considered statistically significant. WT and M132V PV mouse survival curves were not significantly different (P > 0.5 by log rank test [Fig. 7B]). The differences among the groups in the fecal stability assays (Fig. 3) were assessed by two-way analysis of variance (ANOVA).