Table Of Contents

This Page

Notes on “Explaining microbial population genomics through phage predation” by Rodrigues-Valera et al. (2009)

“The remarkable differences that have been detected by metagenomics in the genomes of strains of the same bacterial species are difficult to reconcile with the widely accepted paradigm that periodic selection within bacterial populations will regularly purge genomic diversity by clonal replacement. We have found that many of the genes that differ between strains affect regions that are potential phage recognition targets. We therefore propose the constant-diversity dynamics model, in which the diversity of prokaryotic populations is preserved by phage predation. We provide supporting evidence for this model from metagenomics, mathematical analysis and computer simulations. Periodic selection and phage predation dynamics are not mutually exclusive; we compare their predictions to shed light on the ecological circumstances under which each type of dynamics could predominate.”

I do not understand this: “with the advent of genomics, it has become apparent that the genomic diversity in prokaryotes arises more from having different sets of genes than from allelic differences at the same loci”

“A comparison of two strains, which may belong to highly related lineages within a single species, might reveal that approximately 10–35% of the genome content (typically in the range of 500–1,000 genes) is present in only one of the two strains. [...] Is this diversity important for the ecology and environmental adaptation of the different lineages, or is it just the result of junk DNA accumulation that has yet to be pruned by regular sweeps of natural selection? What are the evolutionary forces that preserve this degree of diversity within highly related populations?”

“The accepted models of bacterial population genetics state that a low phenotypic diversity is expected in asexual (or rarely sexual) microbial populations owing to purges involving fitter mutants, called periodic selection events.”

“It has been claimed that the same kinds of dynamics are behind the genetic coherence of natural prokaryotic taxonomic units or ‘ecotypes’. Ecotypes are defined as “populations that are genetically cohesive and ecologically distinct”. Cohesion is mostly ascribed to “periodic selection events that recurrently purge each ecotype of its genetic diversity”. Divergence can become permanent when a mutation (or recombination) event places an organism into a new ecological niche and establishes a new ecotype.”

“Periodic selection will therefore keep the populations in each ecotype homogeneous and divergent from those in other ecotypes, making these population “the fundamental units of ecology and evolution”, in other words, a stable ecotype.”

“We know that prokaryotes can easily acquire foreign DNA, and phages and plasmids can be easily transferred between bacteria and the DNA inserted into the chromosome of the new host. In fact, some (or most) periodic selection events could be due to the acquisition of new genes rather than mutation. However, it seems unlikely that the neutral accumulation of a large number of new genes could occur in a clone before it is purged during a clonal sweep, and such a phenomenon might have alternative explanations.”

“We propose that the main factor that generates diversity, which has been largely overlooked in many previous models, is the role of a crucial ecological factor: the presence of bacteriophages. In nature, prokaryotic cells must deal with a strong predation pressure that is mainly viral in origin, and therefore their fitness is measured not only by their adaptation to the available resources and the physical conditions (the niche) but also by their adaptation to the biotic environment. Protozoan grazing might also contribute but is less pervasive than viral attack.”

“Phages depend heavily on proper selection of the target cell, and for that they rely on a prominent structure for target recognition; preventing this recognition is therefore the first line of defence for bacteria.”

“Let us consider an idealized aquatic habitat where organic nutrients are dissolved and in which a single prokaryotic population is present. A large diversity of phage sensitivity types is required to avoid catastrophic lysis of the population. Phage receptor types are each recognized by a different phage lineage. The phage receptor types differ in the gene clusters that control the synthesis of complex surface components such as the O chain or the pili, which are exposed extracellularly and are targets for phage recognition, analogous to a lock-and-key system. Periodic selection occurs when a new adaptive mutant (or recombinant) arises in the population and natural selection causes the mutant and its nearly clonal descendants to replace all competing variants in the population16. However, we predict that the increase in number of that fitter lineage would alter the predator–prey equilibrium, and the number of phages that target the receptor encoded by this lineage would also increase (FIG. 2a). This would select against the invasive clone, which would eventually be replaced by the original ‘normal fitness’ lineages. In this way, a constant high level of diversity of lineages would be maintained steadily.”

“Under CD dynamics, no dominant lineage is found in a population”

“a corollary of this model is that, throughout the history of these clonal lineages, each lineage will acquire different but complementary capabilities for niche exploitation. As a consequence, a more efficient exploitation of the resources by the community is expected, and a better functioning of the ecosystem will be achieved. For example, one prokaryotic clone cannot contain even a fraction of the transporters required to internalize the chemical diversity of the organic compounds that are contained in a single eukaryotic cell. However, an ensemble of lineages carrying different sets of transporters could exploit every single one of them.”

“We have considered a simplified case in which one phage infects one cell type; however, it must be noted that several experiments indicate that host–virus interactions in natural systems are more complex, with each phage infecting different bacterial strains with different efficiencies (in other words, the binding probability between bacteriophages and cells is graded).” See Forde, S. E., Thompson, J. N., Holt, R. D. & Bohannan, B. J. Coevolution drives temporal changes in fitness and diversity across environments in a bacteria– bacteriophage interaction. Evolution 62, 1830–1839 (2008). In this paper, there are some interesting references (classic model studies, it seems) to papers on bacteria-bacteriophage interactions like Levin et al. (1977)

This model is nice but it is still not a model of the distribution described by Suttle (2005).

“In the presence of phage predation pressure and environmental changes, the lineage that feeds on the most abundant substrate (or that uses more than one substrate efficiently) undergoes substantial fluctuations in density. (FIG. 3d) By contrast, lineages using the least-abundant substrates show constant densities through time, undergoing only slight variations. Because the amount of density fluctuation is directly related to the probability of extinction, selection will favour less fit cells that use single, low-concentration substrates. This paradoxical selection against the fitter cells will give rise to lineages that feed on all accessible substrates, regardless of their availability. As a consequence, the ecosystem is expected to be more efficient.”

“Note that a high bacterial diversity cannot be the consequence of the availability of different resources alone: it is only in the presence of phages that all cell types reach a similar average density regardless of the substrate that they use; in the absence of phage predation, the exploitation of less favoured niches is selected against.”

“Under PS, a dominant cell type would be expected to arise, and this dominant lineage would change periodically. Conversely, under CD dynamics, phages would keep dominant lineages in check, and therefore many coexisting cell types would be found at any one time. In addition, PS predicts that genomic variability among ecotypes is driven by niche exploitation, whereas CD predicts that it is driven by phage avoidance”

“The initial evidence for antagonistic dynamics between phages and bacteria came from mathematical modelling, as well as experimental studies [see ref. 21,22,46,47]. [...] Recent data on the temporal variation in phage and bacterial diversity also show oscillations that are consistent with a constant control of abundant genotypes by their infecting phages. [...] dominant genotypes are not found over time, and there is a dynamic equilibrium of functionally redundant microbial and viral strains that continuously replace each other in a killthe- winner manner, thereby maintaining a stable metabolic potential and taxonomical signal.”

Basically, we need to refine the model and add different values for the parameters to reflect the r-selection and K-selection dynamics of Suttle (2005).

See Shoresh, N., Hegreness, M. & Kishony, R. Evolution exacerbates the paradox of the plankton. Proc. Natl Acad. Sci. USA 105, 12365–12369 (2008).

“The CD dynamics counter-intuitively predicts that the strain of highest fitness would be selected against and that phage predation would select for cell types of lower fitness that use any substrate regardless of its availability”

“the fitness concept under PS dynamics is dependent on the efficiency with which a resource is used, but the evolutionary success of a strain under CD dynamics is dependent on the fitness of the other strains that co-inhabit the environment and is therefore a relative concept. This is analogous to the outcome of game theory models for optimal market strategies: there is no optimal economic strategy per se; instead, the best market tactic depends on what the competitors are doing. This thinking has been successfully applied to evolutionary theory, giving rise to models that conclude that, once an optimal proportion of strategies is achieved in a population, those proportions are in equilibrium over time.”

“The influence of phage predation on bacterial diversity requires that bacterial populations interact with each other; therefore, hostassociated niches can act as physical barriers that prevent direct cell competition and phage dispersal. [...] CD dynamics are also not expected to occur in physically constrained microbial communities such as biofilms, in which populations cannot interact with each other or invade other niches, apart from their role in constraining phage attack and dispersal.”

“if predation pressure is so intense, we would predict that genes that provide phage resistance would be among the fastest evolving in free-living species in order to counteract the fast phage adaptability, giving rise to an evolutionary arms race analogous to that which occurs between the immune system and surface antigens of bacterial pathogens.”

This paper has been referenced by Winter et al. (2010).