The phenotype-fitness map in experimental evolution of phages

Controlling Recombination to Evolve Bacteriophages

Recombination among different phages sometimes facilitates their ability to grow on new hosts. Protocols to direct the evolution of phage host range, as might be used in the application of phage therapy, would then benefit from including steps to enable recombination. Applying mathematical and computational models, in addition to experiments using phages T3 and T7, we consider ways that a protocol may influence recombination levels. We first address coinfection, which is the first step to enabling recombination. The multiplicity of infection (MOI, the ratio of phage to cell concentration) is insufficient for predicting (co)infection levels. The force of infection (the rate at which cells are infected) is also critical but is more challenging to measure. Using both a high force of infection and high MOI (>1) for the different phages ensures high levels of coinfection. We also apply a four-genetic-locus model to study protocol effects on recombinant levels. Recombinants accumulate over multiple generations of phage growth, less so if one phage outgrows the other. Supplementing the phage pool with the low-fitness phage recovers some of this 'lost' recombination. Overall, fine tuning of phage recombination rates will not be practical with wild phages, but qualitative enhancement can be attained with some basic procedures.

Genome-scale top-down strategy to generate viable genome-reduced phages

Efforts have been made to reduce the genomes of living cells, but phage genome reduction remains challenging. It is of great interest to investigate whether genome reduction can make phages obtain new infectious properties. We developed a CRISPR/Cas9-based iterative phage genome reduction (CiPGr) approach and applied this to four distinct phages, thereby obtaining heterogeneous genome-reduced mutants. We isolated and sequenced 200 mutants with loss of up to 8-23% (3.3-35 kbp) of the original sequences. This allowed the identification of non-essential genes for phage propagation, although loss of these genes is mostly detrimental to phage fitness to various degrees. Notwithstanding this, mutants with higher infectious efficiency than their parental strains were characterized, indicating a trade-off between genome reduction and infectious fitness for phages. In conclusion, this study provides a foundation for future work to leverage the information generated by CiPGr in phage synthetic biology research.

Modeling the Directed Evolution of Broad Host Range Phages

Background: The host ranges of individual phages tend to be narrow, yet many applications of phages would benefit from expanded host ranges. Empirical methods have been developed to direct the evolution of phages to attack new strains, but the methods have not been evaluated or compared for their consequences. In particular, how do different methods favor generalist (broad host range) phages over specialist phages? All methods involve exposing phages to two or more novel bacterial strains, but the methods differ in the order in which those hosts are presented through time: Parallel presentation, Sequential presentation, and Mixed presentation. Methods: We use a combination of simple analytical methods and numerical analyses to study the effect of these different protocols on the selection of generalist versus specialist phages. Results: The three presentation protocols have profoundly different consequences for the evolution of generalist versus specialist phages. Sequential presentation favors generalists almost to the exclusion of specialists, whereas Parallel presentation does the least so. However, other protocol attributes (the nature of dilution between transfers of phages to new cultures) also have effects on selection and phage maintenance. It is also noted that protocols can be designed to enhance recombination to augment evolution and to reduce stochastic loss of newly arisen mutants.

An Optimal Lysis Time Maximizes Bacteriophage Fitness in Quasi-Continuous Culture

Optimality models have a checkered history in evolutionary biology. While optimality models have been successful in providing valuable insight into the evolution of a wide variety of biological traits, a common objection is that optimality models are overly simplistic and ignore organismal genetics. We revisit evolutionary optimization in the context of a major bacteriophage life history trait, lysis time. Lysis time refers to the period spanning phage infection of a host cell and its lysis, whereupon phage progenies are released. Lysis time, therefore, directly determines phage fecundity assuming progeny assembly does not exhaust host resources prior to lysis. Noting that previous tests of lysis time optimality rely on batch culture, we implemented a quasi-continuous culture system to observe productivity of a panel of isogenic phage λ genotypes differing in lysis time. We report that under our experimental conditions, λ phage productivity is maximized around optimal lysis times ranging from 60 to 100 min, and λ wildtype strain falls within this range. It would appear that natural selection on phage λ lysis time uncovered a set of genetic solutions that optimized progeny production in its ecological milieu relative to alternative genotypes. We discuss this finding in light of recent results that lysis time variation is also minimized in the strains with lysis times closer to the λ wild-type strain. IMPORTANCE Optimality theory presents the idea that natural selection acts on organismal traits to produce genotypes that maximize organismal survival and reproduction. As such, optimality theory is a valuable tool in guiding our understanding of the genetic constraints and tradeoffs organisms experience as their genotypes are selected to produce optimal solutions to biological problems. However, optimality theory is often critiqued as being overly simplistic and ignoring the roles of chance and history in the evolution of organismal traits. We show here that the wild-type genotype of a popular laboratory model organism, the bacteriophage λ, produces a phenotype for a major life history trait, lysis time, that maximizes the reproductive success of bearers of that genotype relative to other possible genotypes. This result demonstrates, as is rarely shown experimentally, that natural selection can achieve optimal solutions to ecological challenges.

Superinfection exclusion: A viral strategy with short-term benefits and long-term drawbacks

Viral superinfection occurs when multiple viral particles subsequently infect the same host. In nature, several viral species are found to have evolved diverse mechanisms to prevent superinfection (superinfection exclusion) but how this strategic choice impacts the fate of mutations in the viral population remains unclear. Using stochastic simulations, we find that genetic drift is suppressed when superinfection occurs, thus facilitating the fixation of beneficial mutations and the removal of deleterious ones. Interestingly, we also find that the competitive (dis)advantage associated with variations in life history parameters is not necessarily captured by the viral growth rate for either infection strategy. Putting these together, we then show that a mutant with superinfection exclusion will easily overtake a superinfecting population even if the latter has a much higher growth rate. Our findings suggest that while superinfection exclusion can negatively impact the long-term adaptation of a viral population, in the short-term it is ultimately a winning strategy.

Viral social behaviour has recently been receiving increasing attention in the context of ecological and evolutionary dynamics of viral populations. One fascinating and still relatively poorly understood example is superinfection or co-infection, which occur when multiple viruses infect the same host. Among bacteriophages, a wide range of mechanisms have been discovered that enable phage to prevent superinfection (superinfection exclusion) even at the cost of using precious resources for this purpose. What is the evolutionary impact of this strategic choice and why do so many phages exhibit this behaviour? Here, we conduct an extensive simulation study of a phage population to address this question. In particular, we investigate the fate of viral mutations arising in an environment with a constant supply of bacterial hosts designed to mimic a “turbidostat,” as these are increasingly being used in laboratory evolution experiments. Our results show that allowing superinfection in the long-term yields a population which is more capable of adapting to changes in the environment. However, when in direct competition, mutants capable of preventing superinfection experience a very large advantage over their superinfecting counterparts, even if this ability comes at a significant cost to their growth rate. This indicates that while preventing superinfection can negatively impact the long-term prospects of a viral population, in the short-term it is ultimately a winning strategy.

Layman N, Redwood AJ, Chan B, Bull JJ. Methods for measuring the evolutionary stability of engineered genomes to improve their longevity

Diverse applications rely on engineering microbes to carry and express foreign transgenes. This engineered baggage rarely benefits the microbe and is thus prone to rapid evolutionary loss when the microbe is propagated. For applications where a transgene must be maintained for extended periods of growth, slowing the rate of transgene evolution is critical and can be achieved by reducing either the rate of mutation or the strength of selection. Because the benefits realized by changing these quantities will not usually be equal, it is important to know which will yield the greatest improvement to the evolutionary half-life of the engineering. Here, we provide a method for jointly estimating the mutation rate of transgene loss and the strength of selection favoring these transgene-free, revertant individuals. The method requires data from serial transfer experiments in which the frequency of engineered genomes is monitored periodically. Simple mathematical models are developed that use these estimates to predict the half-life of the engineered transgene and provide quantitative predictions for how alterations to mutation and selection will influence longevity. The estimation method and predictive tools have been implemented as an interactive web application, MuSe.

Intra-Population Competition during Adaptation to Increased Temperature in an RNA Bacteriophage

Evolution of RNA bacteriophages of the family Leviviridae is governed by the high error rates of their RNA-dependent RNA polymerases. This fact, together with their large population sizes, leads to the generation of highly heterogeneous populations that adapt rapidly to most changes in the environment. Throughout adaptation, the different mutants that make up a viral population compete with each other in a non-trivial process in which their selective values change over time due to the generation of new mutations. In this work we have characterised the intra-population dynamics of a well-studied levivirus, Qβ, when it is propagated at a higher-than-optimal temperature. Our results show that adapting populations experienced rapid changes that involved the ascent of particular genotypes and the loss of some beneficial mutations of early generation. Artificially reconstructed populations, containing a fraction of the diversity present in actual populations, fixed mutations more rapidly, illustrating how population bottlenecks may guide the adaptive pathways. The conclusion is that, when the availability of beneficial mutations under a particular selective condition is elevated, the final outcome of adaptation depends more on the occasional occurrence of population bottlenecks and how mutations combine in genomes than on the selective value of particular mutations.

Recurrent evolution of high virulence in isolated populations of a DNA virus

Hosts and viruses are constantly evolving in response to each other: as a host attempts to suppress a virus, the virus attempts to evade and suppress the host's immune system. Here, we describe the recurrent evolution of a virulent strain of a DNA virus, which infects multiple Drosophila species. Specifically, we identified two distinct viral types that differ 100-fold in viral titer in infected individuals, with similar differences observed in multiple species. Our analysis suggests that one of the viral types recurrently evolved at least four times in the past ~30,000 years, three times in Arizona and once in another geographically distinct species. This recurrent evolution may be facilitated by an effective mutation rate which increases as each prior mutation increases viral titer and effective population size. The higher titer viral type suppresses the host-immune system and an increased virulence compared to the low viral titer type.

Optimum Threshold Minimizes Noise in Timing of Intracellular Events

How the noisy expression of regulatory proteins affects timing of intracellular events is an intriguing fundamental problem that influences diverse cellular processes. Here we use the bacteriophage λ to study event timing in individual cells where cell lysis is the result of expression and accumulation of a single protein (holin) in the Escherichia coli cell membrane up to a critical threshold level. Site-directed mutagenesis of the holin gene generated phage variants that vary in their lysis times from 30 to 190 min. Observation of the lysis times of single cells reveals an intriguing finding-the noise in lysis timing first decreases with increasing lysis time to reach a minimum and then sharply increases at longer lysis times. A mathematical model with stochastic expression of holin together with dilution from cell growth was sufficient to explain the non-monotonic noise profile and identify holin accumulation thresholds that generate precision in lysis timing.

Transcript degradation and codon usage regulate gene expression in a lytic phage

Many viral genomes are small, containing only single- or double-digit numbers of genes and relatively few regulatory elements. Yet viruses successfully execute complex regulatory programs as they take over their host cells. Here, we propose that some viruses regulate gene expression via a carefully balanced interplay between transcription, translation, and transcript degradation. As our model system, we employ bacteriophage T7, whose genome of approximately sixty genes is well annotated and for which there is a long history of computational models of gene regulation. We expand upon prior modeling work by implementing a stochastic gene expression simulator that tracks individual transcripts, polymerases, ribosomes, and ribonucleases participating in the transcription, translation, and transcript-degradation processes occurring during a T7 infection. By combining this detailed mechanistic modeling of a phage infection with high-throughput gene expression measurements of several strains of bacteriophage T7, evolved and engineered, we can show that both the dynamic interplay between transcription and transcript degradation, and between these two processes and translation, appear to be critical components of T7 gene regulation. Our results point to targeted degradation as a generic gene regulation strategy that may have evolved in many other viruses. Further, our results suggest that detailed mechanistic modeling may uncover the biological mechanisms at work in both evolved and engineered virus variants.

Predicting Evolution of the Transcription Regulatory Network in a Bacteriophage

Prediction of evolutionary trajectories has been an elusive goal, requiring a deep knowledge of underlying mechanisms that relate genotype to phenotype plus understanding how phenotype impacts organismal fitness. We tested our ability to predict molecular regulatory evolution in a bacteriophage (T7) whose RNA polymerase (RNAP) was altered to recognize a heterologous promoter differing by three nucleotides from the wild-type promoter. A mutant of wild-type T7 lacking its RNAP gene was passaged on a bacterial strain providing the novel RNAP in trans. Higher fitness rapidly evolved. Predicting the evolutionary trajectory of this adaptation used measured in vitro transcription rates of the novel RNAP on the six promoter sequences capturing all possible one-step pathways between the wild-type and the heterologous promoter sequences. The predictions captured some of the regulatory evolution but failed both in explaining 1) a set of T7 promoters that consistently failed to evolve and 2) some promoter evolution that fell outside the expected one-step pathways. Had a more comprehensive set of transcription assays been undertaken initially, all promoter evolution would have fallen within predicted bounds, but the lack of evolution in some promoters is unresolved. Overall, this study points toward the increasing feasibility of predicting evolution in well-characterized, simple systems.

Combinatorial Approaches to Viral Attenuation

Live viral vaccines rely on attenuated viruses that can successfully infect their host but have reduced fitness or virulence. Such attenuated viruses were originally developed through trial and error, typically by adaptation of the wild-type virus to novel conditions. That method was haphazard, with no way of controlling the degree of attenuation or the number of attenuating mutations or preventing evolutionary reversion. Synthetic biology now enables rational design and engineering of viral attenuation, but rational design must be informed by biological principles to achieve stable, quantitative attenuation. This work shows that in a model system for viral attenuation, bacteriophage T7, attenuation can be obtained from rational design principles, and multiple different attenuation approaches can be combined for enhanced overall effect.

Attenuated viruses have numerous applications, in particular in the context of live viral vaccines. However, purposefully designing attenuated viruses remains challenging, in particular if the attenuation is meant to be resistant to rapid evolutionary recovery. Here we develop and analyze a new attenuation method, promoter ablation, using an established viral model, bacteriophage T7. Ablation of promoters of the two most highly expressed T7 proteins (scaffold and capsid) led to major reductions in transcript abundance of the affected genes, with the effect of the double knockout approximately additive of the effects of single knockouts. Fitness reduction was moderate and also approximately additive; fitness recovery on extended adaptation was partial and did not restore the promoters. The fitness effect of promoter knockouts combined with a previously tested codon deoptimization of the capsid gene was less than additive, as anticipated from their competing mechanisms of action. In one design, the engineering created an unintended consequence that led to further attenuation, the effect of which was studied and understood in hindsight. Overall, the mechanisms and effects of genome engineering on attenuation behaved in a predictable manner. Therefore, this work suggests that the rational design of viral attenuation methods is becoming feasible.

IMPORTANCE Live viral vaccines rely on attenuated viruses that can successfully infect their host but have reduced fitness or virulence. Such attenuated viruses were originally developed through trial and error, typically by adaptation of the wild-type virus to novel conditions. That method was haphazard, with no way of controlling the degree of attenuation or the number of attenuating mutations or preventing evolutionary reversion. Synthetic biology now enables rational design and engineering of viral attenuation, but rational design must be informed by biological principles to achieve stable, quantitative attenuation. This work shows that in a model system for viral attenuation, bacteriophage T7, attenuation can be obtained from rational design principles, and multiple different attenuation approaches can be combined for enhanced overall effect.

Modeling the probability distribution of the bacterial burst size via a game-theoretic approach

Based on previous studies, empirical distribution of the bacterial burst size varies even in a population of isogenic bacteria. Since bacteriophage progenies increase linearly with time, it is the lysis time variation that results in the bacterial burst size variations. Here, the burst size variation is computationally modeled by considering the lysis time decisions as a game. Each player in the game is a bacteriophage that has initially infected and lysed its host bacterium. Also, the payoff of each burst size strategy is the average number of bacteria that are solely infected by the bacteriophage progenies after lysis. For calculating the payoffs, a new version of ball and bin model with time dependent occupation probabilities (TDOP) is proposed. We show that Nash equilibrium occurs for a range of mixed burst size strategies that are chosen and played by bacteriophages, stochastically. Moreover, it is concluded that the burst size variations arise from choosing mixed lysis strategies by each player. By choosing the lysis time and also the burst size stochastically, the released bacteriophage progenies infect a portion of host bacteria in environment and avoid extinction. The probability distribution of the mixed burst size strategies is also identified.

Viral attenuation by engineered protein fragmentation

A possible but untested method of viral attenuation is protein fragmentation, engineering wild-type proteins as two or more peptides that self-assemble after translation. Here, the bacteriophage T7 was engineered to encode its essential RNA polymerase as two peptides. Initial fitness was profoundly suppressed. Subjecting the engineered virus to over 100 generations of adaptation by serial transfer resulted in a large fitness increase, still remaining below that of evolved wild-type. The fitness increase was accompanied by three substitutions in the fragmented peptides as well as six mutations in other parts of the genome, but the fragmentation was retained. This study thereby demonstrates the feasibility of using gene fragmentation as a possibly permanent method of attenuation, but the initial fitness of the engineered genome may be a poor measure of its fitness on extended adaptation.

Reduced Protein Expression in a Virus Attenuated by Codon Deoptimization

A general means of viral attenuation involves the extensive recoding of synonymous codons in the viral genome. The mechanistic underpinnings of this approach remain unclear, however. Using quantitative proteomics and RNA sequencing, we explore the molecular basis of attenuation in a strain of bacteriophage T7 whose major capsid gene was engineered to carry 182 suboptimal codons. We do not detect transcriptional effects from recoding. Proteomic observations reveal that translation is halved for the recoded major capsid gene, and a more modest reduction applies to several coexpressed downstream genes. We observe no changes in protein abundances of other coexpressed genes that are encoded upstream. Viral burst size, like capsid protein abundance, is also decreased by half. Together, these observations suggest that, in this virus, reduced translation of an essential polycistronic transcript and diminished virion assembly form the molecular basis of attenuation.

How Good Are Statistical Models at Approximating Complex Fitness Landscapes?

Fitness landscapes determine the course of adaptation by constraining and shaping evolutionary trajectories. Knowledge of the structure of a fitness landscape can thus predict evolutionary outcomes. Empirical fitness landscapes, however, have so far only offered limited insight into real-world questions, as the high dimensionality of sequence spaces makes it impossible to exhaustively measure the fitness of all variants of biologically meaningful sequences. We must therefore revert to statistical descriptions of fitness landscapes that are based on a sparse sample of fitness measurements. It remains unclear, however, how much data are required for such statistical descriptions to be useful. Here, we assess the ability of regression models accounting for single and pairwise mutations to correctly approximate a complex quasi-empirical fitness landscape. We compare approximations based on various sampling regimes of an RNA landscape and find that the sampling regime strongly influences the quality of the regression. On the one hand it is generally impossible to generate sufficient samples to achieve a good approximation of the complete fitness landscape, and on the other hand systematic sampling schemes can only provide a good description of the immediate neighborhood of a sequence of interest. Nevertheless, we obtain a remarkably good and unbiased fit to the local landscape when using sequences from a population that has evolved under strong selection. Thus, current statistical methods can provide a good approximation to the landscape of naturally evolving populations.

Genetically Determined Variation in Lysis Time Variance in the Bacteriophage φX174

Researchers in evolutionary genetics recently have recognized an exciting opportunity in decomposing beneficial mutations into their proximal, mechanistic determinants. The application of methods and concepts from molecular biology and life history theory to studies of lytic bacteriophages (phages) has allowed them to understand how natural selection sees mutations influencing life history. This work motivated the research presented here, in which we explored whether, under consistent experimental conditions, small differences in the genome of bacteriophage φX174 could lead to altered life history phenotypes among a panel of eight genetically distinct clones. We assessed the clones' phenotypes by applying a novel statistical framework to the results of a serially sampled parallel infection assay, in which we simultaneously inoculated each of a large number of replicate host volumes with ∼1 phage particle. We sequentially plated the volumes over the course of infection and counted the plaques that formed after incubation. These counts served as a proxy for the number of phage particles in a single volume as a function of time. From repeated assays, we inferred significant, genetically determined heterogeneity in lysis time and burst size, including lysis time variance. These findings are interesting in light of the genetic and phenotypic constraints on the single-protein lysis mechanism of φX174. We speculate briefly on the mechanisms underlying our results, and we discuss the potential importance of lysis time variance in viral evolution.

Challenges in predicting the evolutionary maintenance of a phage transgene

BACKGROUND

In prior work, a phage engineered with a biofilm-degrading enzyme (dispersin B) cleared artificial, short-term biofilms more fully than the phage lacking the enzyme. An unresolved question is whether the transgene will be lost or maintained during phage growth - its loss would limit the utility of the engineering. Broadly supported evolutionary theory suggests that transgenes will be lost through a 'tragedy of the commons' mechanism unless the ecology of growth in biofilms meets specific requirements. We test that theory here.

RESULTS

Functional properties of the transgenic phage were identified. Consistent with the previous study, the dispersin phage was superior to unmodified phage at clearing short term biofilms grown in broth, shown here to be an effect attributable to free enzyme. However, the dispersin phage was only marginally better than control phages on short term biofilms in minimal media and was no better than control phages in clearing long term biofilms. There was little empirical support for the tragedy of the commons framework despite a strong theoretical foundation for its supposed relevance. The framework requires that the transgene imposes an intrinsic cost, yet the transgene was intrinsically neutral or beneficial when expressed from one part of the phage genome. Expressed from a different part of the genome, the transgene did behave as if intrinsically costly, but its maintenance did not benefit from spatially structured growth per se - violating the tragedy framework.

CONCLUSIONS

Overall, the transgene was beneficial under many conditions, but no insight to its maintenance was attributable to the established evolutionary framework. The failure likely resides in system details that would be used to parameterize the models. Our study cautions against naive applications of evolutionary theory to synthetic biology, even qualitatively.

Realistic three dimensional fitness landscapes generated by self organizing maps for the analysis of experimental HIV-1 evolution

Human Immunodeficiency Virus type 1 (HIV-1) because of high mutation rates, large population sizes, and rapid replication, exhibits complex evolutionary strategies. For the analysis of evolutionary processes, the graphical representation of fitness landscapes provides a significant advantage. The experimental determination of viral fitness remains, in general, difficult and consequently most published fitness landscapes have been artificial, theoretical or estimated. Self-Organizing Maps (SOM) are a class of Artificial Neural Network (ANN) for the generation of topological ordered maps. Here, three-dimensional (3D) data driven fitness landscapes, derived from a collection of sequences from HIV-1 viruses after “in vitro” passages and labelled with the corresponding experimental fitness values, were created by SOM. These maps were used for the visualization and study of the evolutionary process of HIV-1 “in vitro” fitness recovery, by directly relating fitness values with viral sequences. In addition to the representation of the sequence space search carried out by the viruses, these landscapes could also be applied for the analysis of related variants like members of viral quasiespecies. SOM maps permit the visualization of the complex evolutionary pathways in HIV-1 fitness recovery. SOM fitness landscapes have an enormous potential for the study of evolution in related viruses of “in vitro” works or from “in vivo” clinical studies with human, animal or plant viral infections.

Fitness benefits of low infectivity in a spatially structured population of bacteriophages

For a parasite evolving in a spatially structured environment, an evolutionarily advantageous strategy may be to reduce its transmission rate or infectivity. We demonstrate this empirically using bacteriophage (phage) from an evolution experiment where spatial structure was maintained over 550 phage generations on agar plates. We found that a single substitution in the major capsid protein led to slower adsorption of phage to host cells with no change in lysis time or burst size. Plaques formed by phage isolates containing this mutation were not only larger but also contained more phage per unit area. Using a spatially explicit, individual-based model, we showed that when there is a trade-off between adsorption and diffusion (i.e. less 'sticky' phage diffuse further), slow adsorption can maximize plaque size, plaque density and overall productivity. These findings suggest that less infective pathogens may have an advantage in spatially structured populations, even when well-mixed models predict that they will not.

Lethal mutagenesis failure may augment viral adaptation

Lethal mutagenesis, the attempt to extinguish a population by elevating its mutation rate, has been endorsed in the virology literature as a promising approach for treating viral infections. In support of the concept, in vitro studies have forced viral extinction with high doses of mutagenic drugs. However, the one known mutagenic drug used on patients commonly fails to cure infections, and in vitro studies typically find a wide range of mutagenic conditions permissive for viral growth. A key question becomes how subsequent evolution is affected if the viral population is mutated but avoids extinction--Is viral adaptation augmented rather than suppressed? Here we consider the evolution of highly mutated populations surviving mutagenesis, using the DNA phage T7. In assays using inhibitory hosts, whenever resistance mutants were observed, the mutagenized populations exhibited higher frequencies, but some inhibitors blocked plaque formation by even the mutagenized stock. Second, outgrowth of previously mutagenized populations led to rapid and potentially complete fitness recovery but polymorphism was slow to decay, and mutations exhibited inconsistent patterns of change. Third, the combination of population bottlenecks with mutagenesis did cause fitness declines, revealing a vulnerability that was not apparent from mutagenesis of large populations. The results show that a population surviving high mutagenesis may exhibit enhanced adaptation in some environments and experience little negative fitness consequences in many others.

Evolutionarily stable attenuation by genome rearrangement in a virus

Live, attenuated viruses provide many of the most effective vaccines. For the better part of a century, the standard method of attenuation has been viral growth in novel environments, whereby the virus adapts to the new environment but incurs a reduced ability to grow in the original host. The downsides of this approach were that it produced haphazard results, and even when it achieved sufficient attenuation for vaccine production, the attenuated virus was prone to evolve back to high virulence. Using bacteriophage T7, we apply a synthetic biology approach for creating attenuated genomes and specifically study their evolutionary stability. Three different genome rearrangements are used, and although some initial fitness recovery occurs, all exhibit greatly impaired abilities to recover wild-type fitness over a hundred or more generations. Different degrees of stable attenuation appear to be attainable by different rearrangements. Efforts to predict fitness recovery using the extensive background of T7 genetics and biochemistry were only sometimes successful. The use of genome rearrangement thus offers a practical mechanism of evolutionary stable viral attenuation, with some progress toward prediction.

Adaptive regulatory substitutions affect multiple stages in the life cycle of the bacteriophage φX174

BACKGROUND

Previously, we showed that adaptive substitutions in one of the three promoters of the bacteriophage φX174 improved fitness at high-temperature by decreasing transcript levels three- to four-fold. To understand how such an extreme change in gene expression might lead to an almost two-fold increase in fitness at the adaptive temperature, we focused on stages in the life cycle of the phage that occur before and after the initiation of transcription. For both the ancestral strain and two single-substitution strains with down-regulated transcription, we measured seven phenotypic components of fitness (attachment, ejection, eclipse, virion assembly, latent period, lysis rate and burst size) during a single cycle of infection at each of two temperatures. The lower temperature, 37°C, is the optimal temperature at which phages are cultivated in the lab; the higher temperature, 42°C, exerts strong selection and is the condition under which these substitutions arose in evolution experiments. We augmented this study by developing an individual-based stochastic model of this same life cycle to explore potential explanations for our empirical results.

RESULTS

Of the seven fitness parameters, three showed significant differences between strains that carried an adaptive substitution and the ancestor, indicating the presence of pleiotropy in regulatory evolution. 1) Eclipse was longer in the adaptive strains at both the optimal and high-temperature environments. 2) Lysis rate was greater in the adaptive strains at the high temperature. 3) Burst size for the mutants was double that of the ancestor at the high temperature, but half that at the lower temperature. Simulation results suggest that eclipse length and latent period variance can explain differences in burst sizes and fitness between the mutant and ancestral strains.

CONCLUSIONS

Down-regulating transcription affects several steps in the phage life cycle, and all of these occur after the initiation of transcription. We attribute the apparent tradeoff between delayed progeny production and faster progeny release to improved host resource utilization at high temperature.

Evolutionary stability of a refactored phage genome

Engineered genetic systems are commonly unstable; if propagated, they evolve to reverse or modify engineered elements because the elements impair fitness. A goal of synthetic biology is thus to anticipate and avoid detrimental engineering, but little is yet known about which types of elements cause problems in different contexts. In prior work, 30% of the genome of bacteriophage T7 was “refactored” by the insertion or modification of 65 short sequences that included a useful restriction enzyme site in order to, among other goals, separate genes and their translational initiation regions from each other and from other genetic elements. Although gene sequences and known important regions of regulatory elements were kept intact, the translational efficiency of some genes or element regulatory function might have been compromised. We adapted the refactored phage for rapid growth in two conditions, observing fitness and sequence evolution. As anticipated from the original work, refactoring had major fitness effects in both environments, but most of the fitness costs were recovered upon adaptation. The evolved phages retained 60–70% of the design elements, suggesting they had only minor fitness effects. Approximately half the elements that were lost lie within large deletions commonly observed during adaptation of the wild-type genome. Some elements were lost or modified in parallel between the adaptations without affecting T7 gene sequences, but no obvious correlates can be made. Nevertheless, experimental adaptations are useful for identifying specific synthetic design problems, and we suggest that experimental evolution in conjunction with alternative engineering may also be useful in overcoming those problems.

Evolutionary recovery of a recombinant viral genome

It is well appreciated that the evolutionary divergence of genes and genomes from a common ancestor ultimately leads to incompatibilities if those genomes are hybridized. Far less is known about the ability and nature of compensatory evolution to yield the recovery of function in hybrid genomes. Here the major capsid gene of the bacteriophage T7 (40-kb dsDNA) was replaced with the homologous gene of either T3 or K11, each 22% different at the protein level from the T7 homolog. Initial fitness was moderately impaired for the T3 exchange, but the K11 exchange was not viable without a compensatory change in the T7 scaffolding protein. Subsequent adaptation of the transgenic phages led to nearly complete fitness recoveries. Compensatory changes were few, mostly in the transgene and its main interacting partner, the scaffolding protein gene. The large magnitude of fitness recovery with relatively few mutations suggests that the fitness costs of hybridizations and horizontal gene exchanges between moderately diverged genomes can potentially be short-lived through compensatory evolution.

Slow fitness recovery in a codon-modified viral genome

Extensive synonymous codon modification of viral genomes appears to be an effective way of attenuating strains for use as live vaccines. An assumption of this method is that codon changes have individually small effects, such that codon-attenuated viruses will be slow to evolve back to high fitness (and thus to high virulence). The major capsid gene of the bacterial virus T7 was modified to have varying levels of suboptimal synonymous codons in different constructs, and fitnesses declined linearly with the number of changes. Adaptation of the most extreme design, with 182 codon changes, resulted in a slow fitness recovery by standards of previous experimental evolution with this virus, although fitness effects of substitutions were higher than expected from the average effect of an engineered codon modification. Molecular evolution during recovery was modest, and changes evolved both within the modified gene and outside it. Some changes within the modified gene evolved in parallel across replicates, but with no obvious explanation. Overall, the study supports the premise that codon-modified viruses recover fitness slowly, although the evolution is substantially more rapid than expected from the design principle.

A common, non-optimal phenotypic endpoint in experimental adaptations of bacteriophage lysis time

Background

Optimality models of evolution, which ignore genetic details and focus on natural selection, are widely used but sometimes criticized as oversimplifications. Their utility for quantitatively predicting phenotypic evolution can be tested experimentally. One such model predicts optimal bacteriophage lysis interval, how long a virus should produce progeny before lysing its host bacterium to release them. The genetic basis of this life history trait is well studied in many easily propagated phages, making it possible to test the model across a variety of environments and taxa.

Results

We adapted two related small single-stranded DNA phages, ΦX174 and ST-1, to various conditions. The model predicted the evolution of the lysis interval in response to host density and other environmental factors. In all cases the initial phages lysed later than predicted. The ΦX174 lysis interval did not evolve detectably when the phage was adapted to normal hosts, indicating complete failure of optimality predictions. ΦX174 grown on slyD-defective hosts which initially entirely prevented lysis readily recovered to a lysis interval similar to that attained on normal hosts. Finally, the lysis interval still evolved to the same endpoint when the environment was altered to delay optimal lysis interval. ST-1 lysis interval evolved to be ~2 min shorter, qualitatively in accord with predictions. However, there were no changes in the single known lysis gene. Part of ST-1's total lysis time evolution consisted of an earlier start to progeny production, an unpredicted phenotypic response outside the boundaries of the optimality model.

Conclusions

The consistent failure of the optimality model suggests that constraint and genetic details affect quantitative and even qualitative success of optimality predictions. Several features of ST-1 adaptation show that lysis time is best understood as an output of multiple traits, rather than in isolation.