Bacteria and their viruses, phage, are the most abundant and genetically diverse group of organisms on earth. Given their prevalence, it is no wonder that recent studies have found their interactions important for ecosystem function, as well as the health of humans. Unfortunately, because of technical challenges with studying microbes, some of the most basic questions on their interactions, such as who infects whom, and how their relationships evolved in the first place, remain unanswered. Here I report six studies on bacterial-phage interactions, each focused on understanding their pattern and the underlying biophysical, ecological, and evolutionary processes that shape them. To do this, I tested a number of hypotheses using laboratory experiments and analyses of natural microbial diversity. First, I tested whether Esherichia coli cultured without phage would counter-intuitively evolve new interactions with phage. Typically bacterial traits responsible for phage resistance have pleiotropic consequences on growth, therefore as a side-effect of adapting to an abiotic environment, bacteria may also evolve to become more or less vulnerable to their parasites. After 45,000 generations of laboratory culturing without phage, E. coli gained resistance to lambda phage, gained sensitivity to a mutant T6 phage, and remained resistant to wild type T6. Each response was explained by understanding the pleiotropic costs or benefits of mutations that confer resistance. Because of pleiotropy, interactions may even evolve in the absence of one player.For the rest of my studies I examined how interactions evolve when host and parasite co-occur. First, I found that when E. coli and phage lambda are cocultured, E. coli evolves resistance by reducing the number of phage genotypes that can infect it, whereas, lambda evolves to increase the number of bacterial genotypes it can infect. This antagonism produces a interaction matrix with a nested form where less derived host-ranges fall one within another. To determine whether this nested pattern is an artifact of the labratory environment, or if the pattern is general to natural communities, I performed a metaanalysis on already published phage-bacterial interaction matrices. The majority of networks were significantly nested (28 of 38). Lastly, I examined the molecular basis of E. coli resistance to lambda and found that resistance often evolves through mutations in E. coli's lamB, the gene for the phage receptor. Also, the strength of resistance is correlated with how the mutation perturbs the orientation specific features of the protein structure, primarily loop four which extends out of the cell membrane. For the final two chapters, I studied whether lambda could evolve to target a novel receptor and the evolutionary consequences of such an innovation. Under particular laboratory conditions, E. coli evolves resistance by down-regulating LamB, which sets the stage for lambda to evolve the necessary mutations to exploit a new protein receptor. When allowed to coevolve under this condition, lambda evolved to exploit another outer-membrane protein, OmpF. This new function is the result of a particular combination of four mutations in J, the gene for the protein ligand lambda uses to bind to its host. Once lambda evolves this novel interaction, an evolutionary arms-race begins that drives rapid diversification of the bacteria and phage. Overall, my studies show that coevolution between bacteria and phage, whether it be in the lab or in nature, produces nested interactions matrices. Secondly, antagonistic coevolution is a creative process able to generate new genotypes of host and parasite and promote the evolution of novel function. Lastly, costs for resistance have many important effects, from determining whether resistance will evolve or be lost, to the generation of diversity.
Read
