Mutational fitness effects can be measured with relatively high accuracy in viruses due to their small genome size, which facilitates full-length sequencing and genetic manipulation. Previous work has shown that animal and plant RNA viruses are very sensitive to mutation. Here, we characterize mutational fitness effects in single-stranded (ss) DNA and ssRNA bacterial viruses. First, we performed a mutation-accumulation experiment in which we subjected three ssDNA (ΦX174, G4, F1) and three ssRNA phages (Qβ, MS2, and SP) to plaque-to-plaque transfers and chemical mutagenesis. Genome sequencing and growth assays indicated that the average fitness effect of the accumulated mutations was similar in the two groups. Second, we used site-directed mutagenesis to obtain 45 clones of ΦX174 and 42 clones of Qβ carrying random single-nucleotide substitutions and assayed them for fitness. In ΦX174, 20% of such mutations were lethal, whereas viable ones reduced fitness by 13% on average. In Qβ, these figures were 29% and 10%, respectively. It seems therefore that high mutational sensitivity is a general property of viruses with small genomes, including those infecting animals, plants, and bacteria. Mutational fitness effects are important for understanding processes of fitness decline, but also of neutral evolution and adaptation. As such, these findings can contribute to explain the evolution of ssDNA and ssRNA viruses.
The fitness effects of mutations are the raw material for natural selection. It has been shown that point mutations typically have strongly deleterious effects in plant and animal RNA viruses, whereas cellular organisms are comparatively more robust. Here, we characterize the fitness effects of random mutations in DNA viruses and compare them with those found in RNA viruses, using six phage species of similar genome sizes. To achieve this goal, we introduced mutations by chemical and site-directed mutagenesis, identified the genetic changes by sequencing, and quantified their fitness effects using growth-rate assays. In all cases, mutations had a strong average impact on fitness. We conclude that mutational sensitivity is a general property of viruses with small genomes and discuss the evolutionary implications of these findings.
Citation: Domingo-Calap P, Cuevas JM, Sanjuán R (2009) The Fitness Effects of Random Mutations in Single-Stranded DNA and RNA Bacteriophages. PLoS Genet 5(11): e1000742. https://doi.org/10.1371/journal.pgen.1000742
Editor: David J. Begun, University of California Davis, United States of America
Received: July 22, 2009; Accepted: October 26, 2009; Published: November 26, 2009
Copyright: © 2009 Domingo-Calap et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was financially supported by grant BFU2008-03978/BMC from the Spanish MICINN (www.micinn.es) to RS, by the Juan de la Cierva and Ramon y Cajal research programs from the Spanish MICIIN to JMC and RS, respectively, and by a fellowship from the Generalitat Valenciana (www.gva.es) to PDC. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Mutational fitness effects are important for understanding the genetic variability of populations, the relative roles of natural selection and drift, the origin of sex and recombination, or the ability to produce evolutionary innovations, among other processes –. Further, they are of practical relevance to several fields, including complex human disease  and conservation genetics . A simple parameter describing mutational fitness effects is the mean selection strength, defined as the average change in fitness caused by random mutations. However, it is also important to determine their variance and the shape of their statistical distribution. A classic approach to achieve these goals is the mutation-accumulation experiment, in which lines derived from a founder clone are propagated at the smallest possible population size to minimize selection, thereby allowing mutations to accumulate ,. A more direct and powerful approach consists of genetically engineering random mutants, although this has been done far less often due to the greater difficulty of the task .
Some progress has been made in characterizing mutational fitness effects. For instance, in Escherichia coli more than 90% of gene knock-outs are viable  and random insertions caused by transposition reduce fitness by 3% on average . In nematodes, the vast majority of nucleotide substitutions have very small fitness effects , whereas in humans few amino acid substitutions have effects greater than 10% and about 30% evolve neutrally . However, the distribution of mutational fitness effects is complex and large differences between species may exist . For example, in some RNA viruses random nucleotide substitutions reduce fitness by nearly 50% on average and up to 40% are lethal ,. This extreme mutational sensitivity contrasts with the greater robustness of cellular organisms. Mutational robustness is thought to emerge from the presence of alternative metabolic pathways, genetic redundancy, or modularity, but these buffering mechanisms are usually not found in RNA viruses due to their compact genomes .
Viruses are a unique experimental system for characterizing mutational fitness effects because it is relatively easy to engineer sets of random single-nucleotide substitutions. This has been done previously for single-stranded (ss) RNA viruses, but not for ssDNA viruses. These two types of viruses have similar genome sizes and therefore might be expected to be equally sensitive to mutation. On the other hand, given the obvious ecological , evolutionary , and genetic  differences between ssDNA and ssRNA viruses, differences in robustness would not be surprising. For instance, ssRNA viruses show higher mutation rates –, possibly selecting for greater robustness –, although this might have also promoted genome compression –. Second, replicase genes represent a large portion of the genome of RNA viruses but are not encoded by ssDNA viruses, which use host DNA polymerases. Since replicase genes are typically highly constrained, this might also lead to differences in robustness between ssRNA and ssDNA viruses. Finally, some ssDNA viruses encode two scaffolding proteins, apparently introducing some redundancy in the process of capsid assembly .
Here, we compare mutational fitness effects in ssDNA and ssRNA viruses using six phages which can infect the same E. coli strain under identical environmental conditions : the microviruses ΦX174 and G4, the inovirus F1 (ssDNA viruses), the alloleviviruses Qβ and SP, and the levivirus MS2 (ssRNA viruses) (Figure 1). First, we carried out a mutation-accumulation experiment in which we subjected phages to plaque-to-plaque passages and chemical mutagenesis, quantified changes in fitness, and sequenced the ancestral and evolved genomes. Second, we used site-directed mutagenesis to determine the mean strength of selection more accurately and to characterize the distribution of fitness effects in ΦX174 and Qβ. We demonstrate that ssDNA and ssRNA phage exhibit very similar mutational fitness effects.
Genome sizes are 5.4 kb, 5.6 kb, 6.4 kb, 4.2 kb, 4.3 kb, and 3.6 kb, respectively. Protein-coding regions represent between 91% (F1) and 96% (ΦX174) of the total genome, respectively. Arrows indicate the location of each cistron and colors represent broad functional categories (blue: replication; green: particle assembly; yellow: coat; red: lysis/extrusion; grey: other/unknown). However, several genes are multifunctional; for instance A2 is involved in phage assembly and lysis. The location of each of the mutations fixed in the mutation-accumulation lines is shown within white bars, each bar corresponding to an independent line (see Table S1 for details). The three DNA phages have circular genomes but are shown in linear form for convenience. The genomes of ΦX174 and G4 are shown collinearly to indicate gene homology. However, by convention, the first position of the ΦX174 genome is assigned to the unique Pst I cleavage site (vertical dashed line).
Preliminarily, we adapted each phage to our laboratory conditions by performing serial passages at high population sizes, until all lineages reached stable fitness values. For each phage, three independent mutation-accumulation lineages were initiated by picking single plaques at random from the adapted populations. Phage supernatants from plaques were subjected to chemical mutagenesis using nitrous acid and plated to isolate new plaques, a protocol that was continued until plaque sizes became drastically reduced. Under these conditions, selection is minimized and therefore, except for highly deleterious or lethal mutations, random genetic drift and mutational pressure are the only factors driving the fixation of mutations . Growth rates were then assayed and the fitness values relative to the non-mutated genotype were obtained as growth rate ratios.
The change in fitness was homogeneous among species (Table 1, nested ANOVA: P = 0.769) and there were no significant differences between the DNA and RNA groups (P = 0.605). After sequencing the six ancestral and the 18 evolved genomes (GenBank accession numbers GQ153912-GQ153935) we found that DNA and RNA phages had accumulated 157±3 and 146±1 nucleotide substitutions in total, respectively (Table 1, Figure 1, nested ANOVA: P = 0.802). We calculated the per-mutation effect by dividing the fitness loss of each lineage by the number of mutations accumulated and again, this did not reveal significant differences between DNA and RNA phages (P = 0.870). We therefore conclude that selection strength is similar in the two groups. F1 appeared to show the highest level of robustness, but differences between species were non-significant (P = 0.088) and a Tukey's post-hoc test indicated that the six phages formed a single coherent group.
However, the number of passages required to reach similar fitness losses or mutation numbers was higher on average for DNA phages (Table 1, nested ANOVA, P = 0.010). This might be due to their lower spontaneous mutation rates or to lower susceptibility to the mutagen. In both groups, the proportion of transitions was high, and this excess was significantly more marked in DNA phages (94.9% versus 84.9%, Fisher's exact test: P = 0.004). Since nitrous acid induces mainly transitions, this suggests that, in DNA phages, most substitutions were caused by the mutagen, whereas in RNA phages there was a greater contribution of spontaneous mutation. This could explain why more passages were required in the former group, although differences in susceptibility to the mutagen or in the proportion of transitions among spontaneous mutations cannot be discarded. Importantly, the higher transition/transversion ratio of DNA phages did not result in a significantly higher fraction of synonymous substitutions (nested ANOVA, P = 0.484) and therefore is unlikely to have biased the above selection strength estimates.
We constructed 45 clones of ΦX174 and 42 clones of Qβ carrying single-nucleotide substitutions, choosing the target site and the substitution at random. First, we amplified the viral genomes by PCR using mutagenic primers, transfected E. coli with the PCR products, picked single plaques, and verified the presence of the mutation by sequencing. Some transfected cultures failed to form plaques, suggesting that the engineered mutation was lethal for the virus. We first sequenced the region of the mutagenesis PCR product flanking the target site to verify that there were no additional changes. Then, control assays were carried out in which the entire mutagenesis protocol was repeated using PCR primers that did not carry the mutation. By comparing the numbers of plaques formed in mutagenesis and control assays, we showed that 9 and 12 mutations were lethal in ΦX174 and Qβ, respectively (Figure S1), i.e. a lethal fraction of pl = 0.20 and pl = 0.29, respectively. These two proportions did not differ significantly (Fisher's exact test, P = 0.454). All lethal mutations found in ΦX174 produced amino-acid changes, whereas in the case of Qβ, one was synonymous (U2379A) and one intergenic (G1329A).
We measured the growth rate of viruses carrying viable substitutions to obtain the distribution of mutational fitness effects. Growth rates were also determined for the above control samples and corrected means and variances were obtained by subtracting the mean and variance of the control group from the grand mean and variance of the mutants. In ΦX174, this yielded a mean selection strength of = −0.301 including all (lethal or viable) mutations, with variance V(s) = 0.162. For viable mutations only, = −0.126 and V(sv) = 0.047. In Qβ, = −0.359, V(s) = 0.181, = −0.103 and V(sv) = 0.018 (Table 2). There were no significant differences in mean selection strength between the two phages (nested ANOVA for viable mutations: P = 0.633; Mann-Whitney test for all mutations: P = 0.336). The above correction using controls implies that our inferences of the mean and variance should be free of experimental error or bias resulting from the presence of additional mutations or changes in the assay conditions, although it is still possible that the fitness values of some individual mutants were altered by the presence of additional mutations.
The distribution of fitness effects of viable mutations was highly skewed, with deleterious substitutions of small effect being more abundant than those of large effect (Figure 2), a property shared by most model systems studied . Focusing on viable mutations with negative s-values, we performed non-linear regression to characterize the shape of the distribution. In ΦX174, an exponential model with an expected mean selection strength against deleterious mutations of = −0.186±0.005 described the data satisfactorily (R2 = 0.989) and was as accurate as a Gamma (R2 = 0.989) or a Beta (R2 = 0.985) model (partial F-test, P>0.25 in both cases). Other statistical models were not considered here. In Qβ, the data were better described by the Gamma distribution (R2 = 0.975) with an expected = −0.136±0.008. The Beta model provided a similarly good fit (R2 = 0.973), whereas the exponential model was slightly worse (R2 = 0.965, P = 0.025). Finally, we must emphasize that mutants with positive values were not used for these inferences, which implies that they were assumed to be neutral or beneficial. Neglecting beneficial mutations, the fraction of deleterious to total mutations should thus be pd = (−pl)/sd and the fraction of neutral to total mutations simply pn = 1−pl−pd (Table 2). However, classifying mutations as neutral or deleterious is somewhat unnatural since the difference between a deleterious mutation of infinitesimally small effect and a neutral mutation is merely formal.
Previous site-directed mutagenesis studies have shown that most random nucleotide substitutions are strongly deleterious in animal and plant ssRNA viruses. In Vesicular stomatitis virus (VSV), pl = 0.40, = −0.48 and = −0.13 , whereas in the unrelated Tobacco etch virus (TEV), pl = 0.41, = −0.49, and = −0.13 . After performing similar experiments with phage Qβ, we confirm that ssRNA viruses are extremely sensitive to mutation in general. Roughly speaking, the probability that a random single nucleotide substitution is lethal for an ssRNA virus is one third or higher, while viable mutations reduce fitness by 10–13% on average. The observed fraction of lethals was lower in Qβ than in VSV or TEV, which might reflect real biological differences or may be a consequence of methodological issues. Concerning the shape of the distribution, viable mutations of small effect are more abundant that those of large effect. The specific statistical model that better describes the data varies, but the Gamma and the Beta distributions are generally accurate. However, larger mutant collections would be needed to increase the statistical power of inferences about the distribution of mutational fitness effects.
We analyzed ΦX174 using the same methodology and under the same environment as Qβ to make the comparison between ssDNA and ssRNA viruses. The two phages did not significantly differ in the fraction of lethal mutations or the average strength of selection, suggesting that ssDNA and ssRNA viruses are similarly sensitive to mutation. The strong parallel shown by viruses as different as ΦX174 and Qβ probably stems from the fact that both have small genomes with few and short non-coding regions, several multifunctional proteins, and little genetic redundancy ,. Differences between ΦX174 and Qβ were only minor. For instance, deleterious (non-lethal) effects seemed to follow a simple exponential law in ΦX174 whereas Qβ deviated from this model, probably because there were more mildly deleterious mutations and fewer strongly deleterious ones (Figure 2, Table 2), although the fraction of lethals might be higher in the latter.
The similar robustness shown by ssDNA and ssRNA viruses was also supported by the mutation-accumulation experiment, in which we examined three phage species of each group and found no significant differences in mean selection strength. Differences between species within the two groups were weak or absent, although F1 appeared to be the most robust phage. Related to this, it is worth noting that F1 has fewer overlapping genes. This might be related to the fact that inoviruses have filamentous capsids, which are structurally less constrained than icosahedral ones and can hence accommodate larger genomes ,. However, further work is needed to elucidate whether F1 really differs from the other phages in terms of robustness.
It is possible to compare directly the mean selection strengths estimated by mutation-accumulation and site-directed mutagenesis, since the same viruses and the same environment were used. Because lethal mutations cannot be sampled during plaque-to-plaque passages we expected the former to be lower. However, even after excluding lethals, the values obtained by site-directed mutagenesis experiments were approximately twofold higher. The most likely explanation for this discrepancy is that, since the population size within plaques necessarily exceeds one, some degree of selection is inevitable in mutation-accumulation experiments. Also, given that several mutations accumulated in each lineage, the observed fitness values were dependent on genetic interactions. In RNA viruses, mutations tend to have weaker effects as they accumulate (antagonistic epistasis) and the same might hold for ssDNA viruses and small genomes in general , probably leading to an underestimate of the mean selection strength in mutation-accumulation experiments.
The fitness effects of random mutations are relevant to evolution in many ways. For instance, they determine the fraction of nucleotide sites that evolves neutrally and thus the rate at which populations diverge through random genetic drift. Further, neutral and deleterious mutations can also determine the rate of adaptive evolution indirectly. The observation that mutations tend to be highly deleterious indicates that there are no efficient buffering mechanisms, which might result in large phenotypic variation and strong selection for beneficial mutations too –. Consistent with this view, the fixation of big-benefit mutations has been reported in several phages – and low mutational robustness has been associated with faster adaptation in VSV . Also, everything else being equal, greater effects of deleterious mutations might result in faster adaptation because strongly deleterious mutations are removed more efficiently from populations, favoring the spread of beneficial ones . The connection between mutational robustness and evolvability is controversial, however. By reducing phenotypic variation, robustness should facilitate the accumulation of genetic variation and may foster evolutionary innovation upon changes in the environment or the genetic background over the long-term , , –, a prediction that has also been supported experimentally ,.
The fitness effects of random mutations are not only relevant to neutral evolution and adaptation, but obviously, also to processes of fitness decline such as Muller's ratchet  or lethal mutagenesis . Viruses experience frequent transmission bottlenecks during which random mutations may fix in the population. The consequences of this process for viral fitness are highly dependent on how deleterious are these mutations on average . A substantial proportion of the variation observed in natural populations of RNA viruses comprises transient deleterious mutations, but most fail to reach fixation , possibly due to their strongly negative impact on fitness. Mutational fitness effects are also important for understanding viral extinction through mutagenesis  and the clinical use of mutagens as ribavirin to treat viral infections makes this subject of practical importance.
A conceptual dichotomy between DNA and RNA viruses has been established based on the observation that the latter evolve faster . However, it has been shown more recently that some ssDNA viruses, including parvoviruses , anelloviruses  and geminiviruses  can match the evolutionary rates of RNA viruses and rapidly adapt to new hosts (although this has not been tested experimentally yet). Despite mutating faster than double-stranded DNA viruses or cellular organisms, ssDNA viruses are less error-prone than ssRNA viruses –. Therefore, the rate at which spontaneous mutations occur may not fully explain why ssRNA and ssDNA viruses evolve at similar rates. Additional factors as, for instance, the fitness effects of mutations, should be considered.
Materials and Methods
Bacteriophages and cells
Bacteriophages ΦX174, G4, F1, Qβ, SP, and MS2 and the E. coli C strain IJ1862  were obtained from Prof. James J. Bull (University of Texas). The Qβ infectious clone pBRT7Qβ was provided by Dr. René C. Olsthoorn (Leiden University). General biology of the six phages can be found elsewhere .
Each phage was serially passaged at high population sizes in IJ1862 cells. For each transfer, ∼105 particle forming units (pfu) were inoculated into 0.5 mL LB medium containing IJ1862 cells at their exponential growth phase. The appropriate cell density varied depending on the virus lytic activity and growth rate (OD600 = 0.7 for ΦX174 and G4, 0.15 for F1, Qβ, and MS2, and 0.05 for SP). Infected cultures were incubated in agitation (650 rpm) at 37°C in a Thermomixer 24-tube shaker (Eppendorf) and harvested during the late exponential growth phase of the virus (∼109 pfu/mL). Cells were removed by centrifugation and supernatants were aliquoted, stored at −70°C, and titrated using LB medium solidified with soft agar (0.7%). Initial and final titers were used to calculate growth rates and to adjust sampling times for the next passage accordingly. We performed 60–80 passages for each phage. In all cases, no significant changes in growth rate were observed during the last 30 passages, indicating that phages had reached a quasi-maximal fitness under these experimental conditions.
Three independent mutation-accumulation lines were seeded for each phage by picking random plaques from the high-fitness populations. Each plaque was resuspended in 50 µL LB and stored at −70°C. Lines were propagated plaque-to-plaque as previously described  and after each passage, phages were mutagenized with nitrous acid. To do so, four volumes of 0.3 M acetate buffer pH 4.3 were mixed with one volume of 5 M sodium nitrite and 50 µL of this solution were immediately added to 4 µL of phage-containing supernatant. The exposure time was adjusted to obtain a maximal titer loss. The mutagenesis reactions were quenched by adding 200 µL of 1 M acetate buffer pH 4.3 containing 0.1 mg/mL BSA and 100 µL of this final mix were plated without dilution. After incubation at 37°C, a single lysis plaque was picked, resuspended in 50 µL, aliquoted, stored at −70°C, titrated, and used for the next round of chemical mutagenesis.
The High Pure Plasmid Isolation kit (Roche) was used to purify ΦX174 DNA and the pBRT7β plasmid from partially lysed E. coli IJ1862 cells and from an overnight culture of E. coli K12 previously transformed with this cDNA, respectively. For PCR-based mutagenesis, full-length PCR amplicons were obtained from 500 pg of template using Phusion high-fidelity DNA polymerase (New England Biolabs, error rate provided by the manufacturer: 4.4×10−7 per base per replication round) and a pair of adjacent, divergent, 5′ phosphorylated primers, one of which carried the desired nucleotide substitution. PCR products were circularized using the Quick T4 ligase (New England Biolabs) and E. coli IJ1862 competent cells were transfected by the heat-shock method (42 C, 30 s) in the presence of CaCl2 100 mM. The transfected cells were immediately plated onto LB plates using soft agar and individual plaques were picked after 5-9 hours of incubation at 37°C, resuspended in LB, and stored at −70°C. To verify that the desired mutation had been introduced and that no additional changes were present in the region flanking the target site, (RT)-PCR was performed directly from the resuspended plaque. Moloney murine leukemia virus reverse transcriptase (Fermentas) was used for cDNA synthesis and Phusion DNA polymerase for PCR. The products were column-purified and sequenced using sequence-specific primers.
In cases where transfection yielded no plaques or phages recovered from plaques had not incorporated the mutation, the entire protocol was repeated and, after three consecutive failures, the mutation was classified as a candidate lethal. In all of these cases, the number of plaques obtained from transfection assays was abnormally low, suggesting that the mutation was lethal for the virus and that the few observed plaques came from the template wild-type DNA. To confirm lethality, we first sequenced the region of the PCR-based mutagenesis product flanking the target site to check that the mutation was present and that no additional changes had appeared in this region. Then, we designed new primers that did not carry the nucleotide substitution but were otherwise identical to those used for the mutagenesis reaction, and performed the PCR, circularization and transfection steps exactly as above. Lethality was assessed based on the comparison between the numbers of plaques recovered from mutagenesis versus control reactions (Figure S1).
Similar control assays were carried out to estimate the contribution of non-desired mutations and other sources of error to the inferred distribution of mutational fitness effects. For each phage, we transfected three of the above control PCR-based mutagenesis products and picked 11–12 random plaques from each, yielding 36 and 33 plaques in total for ΦX174 and Qβ, respectively. The relative fitness of each clone was determined as above and we obtained the mean and variance for each phage. These values were subtracted from the means and variances obtained for the real mutants. This allowed us to account for experimental error and, in particular, to control for any potential mutations present in the DNA templates, arising during PCR amplification, or during plaque growth.
To measure growth rates, ∼104 pfu were inoculated into 0.5 mL of LB medium containing IJ1862 cells at their exponential growth phase. Infected cultures were incubated in agitation (650 rpm) at 37°C in a Thermomixer 24-tube shaker and harvested when the wild-type reached an estimated titer of 108 pfu/mL. Cells were removed by centrifugation and the supernatants were aliquoted, stored at −70°C and titrated. The growth rate (r) was calculated as the increase in log-titer per hour. Relative fitness was obtained as W = ri/r0, where i and 0 refer to the mutant and wild-type, respectively (notice that this fitness measure is in log-scale), and the fitness effect (selection coefficient) was calculated as s = W−1. In all cases, mutant and wild-type samples were assayed in the same experimental block, and experimental blocks were done in triplicate.
Sequences were obtained using the Sanger method (Applied Biosystems). In general, we sequenced column-purified plaque-(RT)-PCRs directly. In cases where this yielded low-quality readings, we cloned the PCR fragments using the Zero Blunt TOPO PCR Cloning kit (Invitrogen) and sequenced the inserts with vector-based and internal primers, discarding mutations that were not present in at least 3/5 clones.
For the mutation-accumulation experiment, we used a univariate linear model with two factors: Genetic material (G, main fixed factor) and Phage species (S, random factor nested within G), whereas the line was the experimental replicate. The variables analyzed were the number of passages, the number of mutations fixed, and the expected fitness effect per mutation. For each variable v, the model can be written as νijk = μ+Gi+Sij+lijk, where μ is the grand mean and l is the line (error term).
To test for differences in average fitness between the ΦX174 and Qβ mutant collections, we used a univariate linear model with three factors: Phage species (S, main fixed factor), Assay type (A, site-directed mutagenesis or control, fixed factor nested within S), and Plaque (P, each of the mutant or control plaques, random factor nested within A). Hence, each experimental determination can be expressed as sijkl = μ+Si+Aij+Pijk+εijkl, where μ is the grand mean, is positive for mutagenesis assays and negative for control assays, and ε is the error term.
Non-linear regressions were performed to estimate and V(sd) using the Levenberg-Marquardt algorithm implemented in SPSS v12.
Confirmation of lethal mutations. Site-directed mutagenesis reactions and control reactions (using non-mutagenic but otherwise identical primers) were loaded in 0.8% agarose gels and stained with ethidium bromide. For each candidate mutation, the left and right lanes contain the mutagenesis and control reactions, respectively. Bands of the expected size (5.4 kb for ΦX174 and ca. 7.0 kb for Qβ insert plus the vector) were obtained in all cases (plus additional, non-specific, lower molecular weight bands which did not interfere with transfections). The number of plaques obtained after transfection of E. coli cells with these PCR producuts are shown below the brackets. (A) For ΦX174, all mutations except A402T were confirmed as lethal. (B) For Qβ, all mutations except G780U, G1110A, and G2587U were confirmed.
(0.43 MB PDF)
Genetic changes found in ΦX174, G4, F1, Qβ, SP and MS2 mutation accumulation lines (three lines per phage). The final passage number, the relative fitness and its standard error (SEM), the nucleotide substitutions found (genomic position and substitution), the genes where they map, and the associated amino acid change (amino acid number within gene) are shown. Notice that some mutations fall at overlapping genes (the respective amino acid substitutions are indicated in these cases). Genomic positions in ΦX174 are numbered following the convention that nucleotide 1 corresponds to the Pst I cleavage site. Sequences covered nearly the whole genome (>98%) in all cases except for line SP.1 in which we failed to sequence genomic regions 1–167, 653–823, 2277–2633, and 2985–3186, and line MS2.3 in which we failed to sequence region 3048–3569.
(0.03 MB PDF)
ΦX174 point mutants obtained by site-directed mutagenesis. For each mutant, the genomic position of the nucleotide substitution (relative to the Pst I cleavage site), the mutated gene, the amino acid change, and the relative fitness effect ± SEM are shown. Assays were done in triplicate and fitness values were corrected to account for the presence of additional mutations and other sources of experimental error (see Materials and Methods). Notice that some mutations fall within regions at overlapping genes (the respective amino acid substitutions are indicated in these cases).
(0.02 MB PDF)
Qβ point mutants obtained by site-directed mutagenesis. For each mutant, the genomic position of the nucleotide substitution, the mutated gene, the amino acid change, and the relative fitness effect ± SEM are shown. Assays were done in triplicate and fitness values were corrected to account for the presence of additional mutations and other sources of experimental error (see Materials and Methods). Notice that some mutations fall within regions with overlapping genes (the respective amino acid substitutions are indicated in these cases).
(0.02 MB PDF)
We thank Jim Bull and René C. Olsthoorn for supplying the viruses, cells, and molecular clones; Jeff Clune, Siobain Duffy, Holly Wichman, and Jim Bull for helpful comments; and Paqui de la Iglesia, Concha Hueso, and Loles Catalán for technical assistance.
Conceived and designed the experiments: RS. Performed the experiments: PDC JMC RS. Analyzed the data: RS. Wrote the paper: RS.
- 1. de Visser JA, Hermisson J, Wagner GP, Ancel ML, Bagheri-Chaichian H, et al. (2003) Perspective: Evolution and detection of genetic robustness. Evolution 57: 1959–1972.JA de VisserJ. HermissonGP WagnerML AncelH. Bagheri-Chaichian2003Perspective: Evolution and detection of genetic robustness.Evolution5719591972
- 2. Ohta T (1992) The nearly neutral theory of molecular evolution. Annu Rev Ecol Syst 23: 263–286.T. Ohta1992The nearly neutral theory of molecular evolution.Annu Rev Ecol Syst23263286
- 3. Peck JR, Barreau G, Heath SC (1997) Imperfect genes, Fisherian mutation and the evolution of sex. Genetics 145: 1171–1199.JR PeckG. BarreauSC Heath1997Imperfect genes, Fisherian mutation and the evolution of sex.Genetics14511711199
- 4. Wagner A (2005) Robustness and evolvability in living systems. New Jersey: Princeton University Press. A. Wagner2005Robustness and evolvability in living systems.New JerseyPrinceton University Press
- 5. Eyre-Walker A, Woolfit M, Phelps T (2006) The distribution of fitness effects of new deleterious amino acid mutations in humans. Genetics 173: 891–900.A. Eyre-WalkerM. WoolfitT. Phelps2006The distribution of fitness effects of new deleterious amino acid mutations in humans.Genetics173891900
- 6. Schultz ST, Lynch M (1997) Mutation and extinction: the role of variable mutational effects, synergistic epistasis, beneficial mutations, and degree of outcrossing. Evolution 51: 1363–1371.ST SchultzM. Lynch1997Mutation and extinction: the role of variable mutational effects, synergistic epistasis, beneficial mutations, and degree of outcrossing.Evolution5113631371
- 7. Keightley PD (2004) Comparing analysis methods for mutation-accumulation data. Genetics 167: 551–553.PD Keightley2004Comparing analysis methods for mutation-accumulation data.Genetics167551553
- 8. Mukai T (1964) The genetic structure of natural populations of Drosophila melanogaster. I. spontaneous mutation rate of polygenes controlling viability. Genetics 50: 1–19.T. Mukai1964The genetic structure of natural populations of Drosophila melanogaster. I. spontaneous mutation rate of polygenes controlling viability.Genetics50119
- 9. Eyre-Walker A, Keightley PD (2007) The distribution of fitness effects of new mutations. Nat Rev Genet 8: 610–618.A. Eyre-WalkerPD Keightley2007The distribution of fitness effects of new mutations.Nat Rev Genet8610618
- 10. Baba T, Ara T, Hasegawa M, Takai Y, Okumura Y, et al. (2006) Construction of Escherichia coli K-12 in-frame, single-gene knockout mutants: the Keio collection. Mol Syst Biol 2: 2006.T. BabaT. AraM. HasegawaY. TakaiY. Okumura2006Construction of Escherichia coli K-12 in-frame, single-gene knockout mutants: the Keio collection.Mol Syst Biol22006
- 11. Elena SF, Ekunwe L, Hajela N, Oden SA, Lenski RE (1998) Distribution of fitness effects caused by random insertion mutations in Escherichia coli. Genetica 102/103: 349–358.SF ElenaL. EkunweN. HajelaSA OdenRE Lenski1998Distribution of fitness effects caused by random insertion mutations in Escherichia coli.Genetica102/103349358
- 12. Davies EK, Peters AD, Keightley PD (1999) High frequency of cryptic deleterious mutations in Caenorhabditis elegans. Science 285: 1748–1751.EK DaviesAD PetersPD Keightley1999High frequency of cryptic deleterious mutations in Caenorhabditis elegans.Science28517481751
- 13. Carrasco P, de la Iglesia F, Elena SF (2007) Distribution of fitness and virulence effects caused by single-nucleotide substitutions in Tobacco ech virus. J Virol 81: 12979–12984.P. CarrascoF. de la IglesiaSF Elena2007Distribution of fitness and virulence effects caused by single-nucleotide substitutions in Tobacco ech virus.J Virol811297912984
- 14. Sanjuán R, Moya A, Elena SF (2004) The distribution of fitness effects caused by single-nucleotide substitutions in an RNA virus. Proc Natl Acad Sci USA 101: 8396–8401.R. SanjuánA. MoyaSF Elena2004The distribution of fitness effects caused by single-nucleotide substitutions in an RNA virus.Proc Natl Acad Sci USA10183968401
- 15. Elena SF, Carrasco P, Daròs JA, Sanjuán R (2006) Mechanisms of genetic robustness in RNA viruses. EMBO Rep 7: 168–173.SF ElenaP. CarrascoJA DaròsR. Sanjuán2006Mechanisms of genetic robustness in RNA viruses.EMBO Rep7168173
- 16. Holmes EC (2008) Evolutionary history and phylogeography of human viruses. Annu Rev Microbiol 62: 307–328.EC Holmes2008Evolutionary history and phylogeography of human viruses.Annu Rev Microbiol62307328
- 17. Duffy S, Shackelton LA, Holmes EC (2008) Rates of evolutionary change in viruses: patterns and determinants. Nat Rev Genet 9: 267–276.S. DuffyLA ShackeltonEC Holmes2008Rates of evolutionary change in viruses: patterns and determinants.Nat Rev Genet9267276
- 18. Ahlquist P (2006) Parallels among positive-strand RNA viruses, reverse-transcribing viruses and double-stranded RNA viruses. Nat Rev Microbiol 4: 371–382.P. Ahlquist2006Parallels among positive-strand RNA viruses, reverse-transcribing viruses and double-stranded RNA viruses.Nat Rev Microbiol4371382
- 19. Cuevas JM, Duffy S, Sanjuán R (2009) Point mutation rate of bacteriophage ΦX174. Genetics, in press. JM CuevasS. DuffyR. Sanjuán2009Point mutation rate of bacteriophage ΦX174.Genetics, in press
- 20. Drake JW, Charlesworth B, Charlesworth D, Crow JF (1998) Rates of spontaneous mutation. Genetics 148: 1667–1686.JW DrakeB. CharlesworthD. CharlesworthJF Crow1998Rates of spontaneous mutation.Genetics14816671686
- 21. Sniegowski PD, Gerrish PJ, Johnson T, Shaver A (2000) The evolution of mutation rates: separating causes from consequences. BioEssays 22: 1057–1066.PD SniegowskiPJ GerrishT. JohnsonA. Shaver2000The evolution of mutation rates: separating causes from consequences.BioEssays2210571066
- 22. Sanjuán R, Cuevas JM, Furió V, Holmes EC, Moya A (2007) Selection for robustness in mutagenized RNA viruses. PLoS Genet 3: e93.R. SanjuánJM CuevasV. FurióEC HolmesA. Moya2007Selection for robustness in mutagenized RNA viruses.PLoS Genet3e93
- 23. Schuster P, Swetina J (1988) Stationary mutant distributions and evolutionary optimization. Bull Math Biol 50: 635–660.P. SchusterJ. Swetina1988Stationary mutant distributions and evolutionary optimization.Bull Math Biol50635660
- 24. Wilke CO, Wang JL, Ofria C, Lenski RE, Adami C (2001) Evolution of digital organisms at high mutation rates leads to survival of the flattest. Nature 412: 331–333.CO WilkeJL WangC. OfriaRE LenskiC. Adami2001Evolution of digital organisms at high mutation rates leads to survival of the flattest.Nature412331333
- 25. Belshaw R, Pybus OG, Rambaut A (2007) The evolution of genome compression and genomic novelty in RNA viruses. Genome Res 17: 1496–1504.R. BelshawOG PybusA. Rambaut2007The evolution of genome compression and genomic novelty in RNA viruses.Genome Res1714961504
- 26. Eigen M, McCaskill J, Schuster P (1988) Molecular quasi-species. J Phys Chem 92: 6881–6891.M. EigenJ. McCaskillP. Schuster1988Molecular quasi-species.J Phys Chem9268816891
- 27. Holmes EC (2003) Error thresholds and the constraints to RNA virus evolution. Trends Microbiol 11: 543–546.EC Holmes2003Error thresholds and the constraints to RNA virus evolution.Trends Microbiol11543546
- 28. Chen M, Uchiyama A, Fane BA (2007) Eliminating the requirement of an essential gene product in an already very small virus: scaffolding protein B-free ΦX174, B-free. J Mol Biol 373: 308–314.M. ChenA. UchiyamaBA Fane2007Eliminating the requirement of an essential gene product in an already very small virus: scaffolding protein B-free ΦX174, B-free.J Mol Biol373308314
- 29. Bull JJ, Badgett MR, Springman R, Molineux IJ (2004) Genome properties and the limits of adaptation in bacteriophages. Evolution 58: 692–701.JJ BullMR BadgettR. SpringmanIJ Molineux2004Genome properties and the limits of adaptation in bacteriophages.Evolution58692701
- 30. Belshaw R, Gardner A, Rambaut A, Pybus OG (2008) Pacing a small cage: mutation and RNA viruses. Trends Ecol Evol 23: 188–193.R. BelshawA. GardnerA. RambautOG Pybus2008Pacing a small cage: mutation and RNA viruses.Trends Ecol Evol23188193
- 31. Calendar R (2006) The bacteriophages. Oxford University Press. R. Calendar2006The bacteriophages.Oxford University Press
- 32. van der Ende A, Teertstra R, Weisbeek PJ (1982) Initiation and termination of the bacteriophage ΦX174 rolling circle DNA replication in vivo: packaging of plasmid single-stranded DNA into bacteriophage ΦX174 coats. Nucl Acids Res 10: 6849–6863.A. van der EndeR. TeertstraPJ Weisbeek1982Initiation and termination of the bacteriophage ΦX174 rolling circle DNA replication in vivo: packaging of plasmid single-stranded DNA into bacteriophage ΦX174 coats.Nucl Acids Res1068496863
- 33. Sanjuán R, Elena SF (2006) Epistasis correlates to genomic complexity. Proc Natl Acad Sci USA 103: 14402–14405.R. SanjuánSF Elena2006Epistasis correlates to genomic complexity.Proc Natl Acad Sci USA1031440214405
- 34. Frank SA (2007) Maladaptation and the paradox of robustness in evolution. PLoS ONE 2: e1021.SA Frank2007Maladaptation and the paradox of robustness in evolution.PLoS ONE2e1021
- 35. Lenski RE, Barrick JE, Ofria C (2006) Balancing robustness and evolvability. PLoS Biol 4: e428.RE LenskiJE BarrickC. Ofria2006Balancing robustness and evolvability.PLoS Biol4e428
- 36. Wagner A (2008) Robustness and evolvability: a paradox resolved. Proc Biol Sci 275: 91–100.A. Wagner2008Robustness and evolvability: a paradox resolved.Proc Biol Sci27591100
- 37. Bull JJ, Badgett MR, Wichman HA (2000) Big-benefit mutations in a bacteriophage inhibited with heat. Mol Biol Evol 17: 942–950.JJ BullMR BadgettHA Wichman2000Big-benefit mutations in a bacteriophage inhibited with heat.Mol Biol Evol17942950
- 38. Rokyta DR, Joyce P, Caudle SB, Wichman HA (2005) An empirical test of the mutational landscape model of adaptation using a single-stranded DNA virus. Nat Genet 37: 441–444.DR RokytaP. JoyceSB CaudleHA Wichman2005An empirical test of the mutational landscape model of adaptation using a single-stranded DNA virus.Nat Genet37441444
- 39. Rokyta DR, Beisel CJ, Joyce P, Ferris MT, Burch CL, Wichman HA (2008) Beneficial fitness effects are not exponential for two viruses. J Mol Evol 67: 368–376.DR RokytaCJ BeiselP. JoyceMT FerrisCL BurchHA Wichman2008Beneficial fitness effects are not exponential for two viruses.J Mol Evol67368376
- 40. Cuevas JM, Moya A, Sanjuán R (2009) A genetic background with low mutational robustness is associated with increased adaptability to a novel host in an RNA virus. J Evol Biol 22: 2041–2048.JM CuevasA. MoyaR. Sanjuán2009A genetic background with low mutational robustness is associated with increased adaptability to a novel host in an RNA virus.J Evol Biol2220412048
- 41. Orr HA (2000) The rate of adaptation in asexuals. Genetics 155: 961–968.HA Orr2000The rate of adaptation in asexuals.Genetics155961968
- 42. Bloom JD, Labthavikul ST, Otey CR, Arnold FH (2006) Protein stability promotes evolvability. Proc Natl Acad Sci USA 103: 5869–5874.JD BloomST LabthavikulCR OteyFH Arnold2006Protein stability promotes evolvability.Proc Natl Acad Sci USA10358695874
- 43. Ciliberti S, Martin OC, Wagner A (2007) Innovation and robustness in complex regulatory gene networks. Proc Natl Acad Sci USA 104: 13591–13596.S. CilibertiOC MartinA. Wagner2007Innovation and robustness in complex regulatory gene networks.Proc Natl Acad Sci USA1041359113596
- 44. Kim Y (2007) Rate of adaptive peak shifts with partial genetic robustness. Evolution 1847–1856.Y. Kim2007Rate of adaptive peak shifts with partial genetic robustness.Evolution18471856
- 45. McBride RC, Ogbunugafor CB, Turner PE (2008) Robustness promotes evolvability of thermotolerance in an RNA virus. BMC Evol Biol 8: 231.RC McBrideCB OgbunugaforPE Turner2008Robustness promotes evolvability of thermotolerance in an RNA virus.BMC Evol Biol8231
- 46. Butcher D (1995) Muller's ratchet, epistasis and mutation effects. Genetics 141: 431–437.D. Butcher1995Muller's ratchet, epistasis and mutation effects.Genetics141431437
- 47. Bull JJ, Sanjuán R, Wilke CO (2007) Theory of lethal mutagenesis for viruses. J Virol 81: 2930–2939.JJ BullR. SanjuánCO Wilke2007Theory of lethal mutagenesis for viruses.J Virol8129302939
- 48. Pybus OG, Rambaut A, Belshaw R, Freckleton RP, Drummond AJ, et al. (2007) Phylogenetic evidence for deleterious mutation load in RNA viruses and its contribution to viral evolution. Mol Biol Evol 24: 845–852.OG PybusA. RambautR. BelshawRP FreckletonAJ Drummond2007Phylogenetic evidence for deleterious mutation load in RNA viruses and its contribution to viral evolution.Mol Biol Evol24845852
- 49. Holland JJ, Spindler K, Horodyski F, Grabau E, Nichol S, et al. (1982) Rapid evolution of RNA genomes. Science 215: 1577–1585.JJ HollandK. SpindlerF. HorodyskiE. GrabauS. Nichol1982Rapid evolution of RNA genomes.Science21515771585
- 50. Shackelton LA, Parrish CR, Truyen U, Holmes EC (2005) High rate of viral evolution associated with the emergence of carnivore parvovirus. Proc Natl Acad Sci USA 102: 379–384.LA ShackeltonCR ParrishU. TruyenEC Holmes2005High rate of viral evolution associated with the emergence of carnivore parvovirus.Proc Natl Acad Sci USA102379384
- 51. Umemura T, Tanaka Y, Kiyosawa K, Alter HJ, Shih JW (2002) Observation of positive selection within hypervariable regions of a newly identified DNA virus (SEN virus - 1). FEBS Lett 510: 171–174.T. UmemuraY. TanakaK. KiyosawaHJ AlterJW Shih2002Observation of positive selection within hypervariable regions of a newly identified DNA virus (SEN virus - 1).FEBS Lett510171174
- 52. Duffy S, Holmes EC (2008) Phylogenetic evidence for rapid rates of molecular evolution in the single-stranded DNA begomovirus Tomato yellow leaf curl virus. J Virol 82: 957–965.S. DuffyEC Holmes2008Phylogenetic evidence for rapid rates of molecular evolution in the single-stranded DNA begomovirus Tomato yellow leaf curl virus.J Virol82957965
- 53. Chao L (1990) Fitness of RNA virus decreased by Muller's ratchet. Nature 348: 454–455.L. Chao1990Fitness of RNA virus decreased by Muller's ratchet.Nature348454455