Evolutionary dynamics of bacteria in the gut microbiome within and across hosts

Abstract

The structure and function of the gut microbiome are shaped by a combination of ecological and evolutionary forces. While the ecological dynamics of the community have been extensively studied, much less is known about how strains of gut bacteria evolve over time. Here we show that with a model-based analysis of existing shotgun metagenomic data, we can gain new insights into the evolutionary dynamics of gut bacteria within and across hosts. We find that long-term evolution across hosts is consistent with quasi-sexual evolution and purifying selection, with relatively weak geographic structure in many prevalent species. However, our quantitative approach also reveals new between-host genealogical signatures that cannot be explained by standard population genetic models. By comparing samples from the same host over ∼6 month timescales, we find that within-host differences rarely arise from the invasion of strains as distantly related as those in other hosts. Instead, we more commonly observe a small number of evolutionary changes in resident strains, in which nucleotide variants or gene gains or losses rapidly sweep to high frequency within a host. By comparing the signatures of these mutations with the typical between-host differences, we find evidence that many sweeps are driven by introgression from existing species or strains, rather than by de novo mutations. These data suggest that bacteria in the microbiome can evolve on human relevant timescales, and highlight the feedback between these short-term changes and the longer-term evolution across hosts. INTRODUCTIONThe gut microbiome is a complex ecosystem comprised of a diverse array of microbial organisms. The abundances of different species and strains can vary dramatically based on diet (1), host-species (2), and the identities of other co-colonizing taxa (3). These rapid shifts in community composition suggest that individual gut microbes may be adapted to specific environmental conditions, with strong selection pressures between competing species or strains. Yet while these ecological responses have been extensively studied, much less is known about the evolutionary forces that operate within populations of gut bacteria, both inside individual hosts, and across the larger host-associated population. This makes it difficult to predict how rapidly strains of gut microbes will evolve new ecological preferences and traits when faced with environmental challenges, and how the genetic fingerprint of the community will change as a result.The answers to these questions depend on two different types of information. At a mechanistic level, we must understand the functional traits that are under selection in the gut, and the range of genetic mutations that can alter these traits. Although it can be challenging to measure such selection pressures in vivo, comparative genomics (4, 5), experiments in model organisms (6, 7), and high-throughput screens (8, 9) are starting to provide valuable information about the functional traits required to thrive in the gut environm...

Linking influenza virus evolution within and between human hosts

“…We tracked the fraction of nonsynonymous mutations at a range of mutation frequencies within and between hosts to determine how long deleterious, nonsynonymous mutations persist in the global viral population (Figure 5A). This analysis draws on the same logic that underlies the classical McDonald-Kreitman test for positive selection (Bhatt et al, 2010(Bhatt et al, , 2011Garud et al, 2019;McDonald and Kreitman, 1991;Rand and Kann, 1996). We expect purifying selection to purge deleterious nonsynonymous mutations from the viral population before they reach high mutation frequencies.…”

Can the microbiome influence host evolutionary trajectories?

“…For example, strain-level variation in honey bee microbes influences the response to pathogens or different diets for the hosts 56 . Strain-level analyses also provide insights into the stability of the microbiome across the lifetime of the host [57][58][59] as well transmission dynamics 60,61 . Metagenomes, which fully characterize the extended genotype, will provide crucial insights into functional variation of the microbiome 43,56,62,63 .…”

To dereplicate or not to dereplicate?

“…Although variant identification errors do remain, which are tool and likely database and metagenomic dataset dependent, this has been reported to be as low as 0.1% (25). While potential issues with these approaches have not been fully evaluated, analyses focusing on populations where the dominant strain can be more readily resolved have been able to go as far as tracking in situ bacterial evolution in environmental biofilms and the human gut (30,31).…”