Structure and dynamics of genome-wide diversity in Prochlorococcus

Abstract

The capability of microbes to thrive in myriad environments has its foundation in the diversity of microbial genomes. Here we explore adaptation and diversification through the lens of the marine cyanobacterium Prochlorococcus, which comprises a group of closely-related ecotypes that together perform most of the primary production in low-nutrient regions of the world oceans. Prochlorococcus was one of the first microbes in which a genomic basis for ecological differentiation was characterized, in the distinction between high- and low-light adapted ecotypes. It is clear, however, that other axes of differentiation are important, including temperature, nutrient availability, and biotic interactions. This thesis seeks to characterize salient aspects of genomic diversity in Prochlorococcus and to advance understanding of the ecological and evolutionary forces that shape this variation. We show that closely related isolates harbor remarkably dissimilar gene complements, and much of this variation is concentrated in specific genome regions, termed islands, that appear to have arisen through phage-mediated gene transfer. Several island-encoded genes likely play important metabolic roles, as inferred from their strong and specific upregulation under stress conditions. A region of the genome involved in phosphate assimilation has highly variable gene content that appears to reflect oceanic phosphate availability. Accordingly, we find extreme differences between strains in the transcriptional response to phosphate starvation. Using metagenomics approaches, we describe high coexisting diversity in natural Prochlorococcus populations. Nevertheless, this diversity is structured: a core genome of universal single-copy genes is augmented by a flexible genome.

(cont.) The population genome changes with water depth, reflecting genotypic variation among ecotypes and within the dominant ecotype. Finally, we show that the transcriptomes of wild Prochlorococcus correlate strongly with transcriptomes in culture as measured by microarrays. Genes of unknown function are among the most highly expressed in the wild. Several highly expressed genes show signatures of intragenic recombination, a process that likely influences their diversity and function. Overall, this work demonstrates that environmental factors such as light, temperature, nutrient availability, and interspecies interactions each leave different marks in the genome over different scales of time and space. Understanding microbial evolution requires that we dissect diversity over these multiple scales.