Genetic Robustness and Adaptability of Viruses Phage can develop differing sensitivities to mutations, strongly affecting their adaptive potential in varied environments Robert C. McBride and Paul E. Turner iodiversity reflects billions of years of brittleness are terms used for measuring and adaptation through natural selection, describing the relative accuracy of this geno- whereby environments help to deter- type-to-phenotype translation. In particular, ge- B mine which genetic variants in a pop- netic robustness is the constancy of a phenotype ulation persist. This process leads in the face of changes in the underlying geno- some individuals in a population to possess type. If a mutation changes a phenotype, we traits that prove beneficial in a particular niche consider the gene or, more broadly, the genome, and can help to distinguish them from other to be relatively nonrobust, or brittle, against organisms residing there or elsewhere. Although mutational input. However, if a genome evolution is defined as the change in genetic changes but phenotype remains unaffected, the makeup of a population through time, natural genome is considered genetically robust. Be- selection acts on phenotypes. That is, natural cause other types of robustness are of interest, selection leads indirectly to changes in gene fre- especially environmental robustness that de- quencies by acting on the phenotypes that genes scribes whether a phenotype persists when an produce. environment changes, it is crucial when examin- Thus, the translation of phenotype from geno- ing robustness to define the specific phenotype type is crucial to evolution. Robustness and and perturbation being measured and discussed. Many phenotypic traits could be used to study robustness. From an evolutionary Summary perspective, however, traits that constitute • fitness, the relative ability to survive and Robustness and brittleness describe the relative reproduce, are of greatest interest. accuracy of genotype-to-phenotype translation, which is crucial to evolution. Comprehensive measurement of fitness is prohibitively difficult for many organisms. • Examining the balance between robustness and evolvability, the capacity to adapt, can help in Fortunately, because so many microbes determining whether natural selection shapes grow asexually, microbiologists can readily evolution itself. measure population-level traits for a partic- Robert C. McBride • Despite some perceived obstacles, microbial ular microbial genotype. Thus, two proxies is a Postdoctoral studies, particularly those involving phage, are for genotype fitness are the reproductive Associate in Ecol- helping to fill gaps in our knowledge of robust- growth rate and the numerical size of a ogy and Evolution- ness. population grown from that genotype. ary Biology and • Because pathogenic viruses may adapt to treat- However, typically it is more desirable to Paul E. Turner is ments by developing greater resistance and measure fitness of a microbial genotype us- Associate Professor greater potential to withstand future therapies, ing an experimental assay where two strains of Ecology and Evo- caution is warranted when considering muta- genic therapies. are placed in the same environment and lutionary Biology at their fitness is gauged relative to one an- Yale University, other or to some baseline, such as a single New Haven, Conn. Volume 3, Number 9, 2008 / Microbe Y 409 common-competitor genotype. In this way, re- genetic robustness, i.e., phenotypic constancy searchers can efficiently and accurately measure despite mutational input. Were a population to relative fitness of microbial genotypes, and become optimally adapted to its habitat, any quantify genotype robustness in the face of mutation would be either neutral or deleterious. change. This scenario should then lead to strong selec- tion for genetic robustness to evolve, protecting the phenotype against mutations. In a related Robustness Provides Insights into the sense, because spontaneous mutations are be- Relationship between Phenotype, lieved typically to be deleterious, selection fa- Genotype voring evolution of robustness should be espe- Improved understanding of the fundamental re- cially strong if mutation rates are elevated, even lationship between phenotype and genotype when populations are away from equilibrium. provides a clear reason for studying the evolu- Some convincing data on evolution of robust- tion of robustness. On the one hand, constancy ness stem from studies looking at virtual organ- in the face of environmental and mutational isms—namely, self-replicating computer pro- changes provides obvious benefits to an organ- grams that change randomly and thus “evolve.” ism during replication. Robustness buffers or- According to one such study, elevated mutation ganisms against such perturbations, affording rates can cause robust genotypes to be selec- constancy in terms of cellular function, develop- tively favored over their brittle counterparts, ment, and offspring production. That is, robust- even though robustness against mutations went ness provides reliability in the very currencies by hand-in-hand with lower reproductive fitness. which natural selection judges phenotypes. Thus, the fittest gave way to the “flattest,” with However, rigidity in the face of change may selection favoring those variants having the pose problems. For example, if organisms are greatest phenotypic constancy and residing on steadfast under environmental change, how can flat regions of the fitness landscape (Fig. 1). they possibly adapt to new conditions? Because Other studies successfully examine robustness natural selection acts on phenotypic variation, by following changes in proteins in vitro. robustness that buffers this variation could im- Microbial populations provide another trac- pede evolution. table choice for examining robustness because These conflicting necessities force organisms of their rapid generation times and large popu- to strike a balance between withstanding some lation sizes. However, microorganisms still re- changes and maintaining an ability to adapt to quire extensive time before selection begins to new circumstances. This compromise is the bal- favor variants that evolve increased robustness. ance between robustness and evolvability, the The limiting factor is time needed to achieve an capacity to adapt. By examining this balancing adaptive optimum. For example, fitness in pop- act, we may learn whether evolvability can itself ulations of Escherichia coli continues to change, evolve. Thus, we can explore the intriguing— even after 40,000 generations (20 years), ac- and contentious—idea that natural selection cording to Richard Lenski of Michigan State shapes evolution itself. University in East Lansing and his colleagues. This result suggests a difficulty for researchers in relying on laboratory populations to examine How Is Robustness Studied? selective pressures favoring evolution of robust- Despite a longstanding interest in robustness ness. and extensive mathematical modeling, biolo- Although some theory suggests that evolu- gists have generated few data on this subject tionary changes in robustness might be difficult because studying robustness presents many to observe in laboratory populations of mi- challenges. One hurdle is to identify organisms crobes, these systems remain the most attractive that vary in robustness. Traits favoring robust- means for conducting experiments to examine ness are not expected to be strongly selected evolution. Despite some perceived obstacles, mi- until a biological population reaches evolution- crobial studies are helping to fill gaps in our ary equilibrium—mutation-selection balance— knowledge of robustness. After all, microbes are in a constant environment. immensely successful from an evolutionary This prerequisite should be especially true for standpoint; they thrive in all habitats that sup- 410 Y Microbe / Volume 3, Number 9, 2008 port life, and they are the most plentiful FIGURE 1 denizens of Earth. Over billions of years, microbes have experienced ex- tensive selection to shape robustness. RNA viruses seem particularly ap- propriate for examining genetic robust- ness, a view that mathematical theory supports. According to digital organism experiments, elevated mutation rates are a key prerequisite for populations to adapt by altering their genetic (mu- tational) robustness. Therefore, success may come from studying genetic ro- bustness in biological systems with ele- vated mutation rates, making RNA vi- ruses even better candidates because their mutation rates generally exceed those of other organisms, including DNA viruses, by at least one order of magni- tude. By focusing on RNA systems, it is possible to study the evolution of ro- bustness even when populations are not at an evolutionary equilibrium or are not subject to artificially elevated muta- tion rates through mutagenesis. Can Robustness Evolve? Several independent studies show that robustness can increase or decrease, de- Schematic where brittle and robust organisms are defined by their fitness response to pending on the particulars of the selec- mutational change, using the metaphor of fitness landscapes. Fitness is vertical height tive environment. For example, we used on the landscape. Mutation causes genotypes to move away from their original position a single genotype of the lytic RNA bac- on the horizontal axis. After mutation, brittle individuals experience large changes in fitness as they are “pushed off” the narrow fitness peak. In contrast,