Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 5 (1), 127

Models of Microbiome Evolution Incorporating Host and Microbial Selection

Affiliations

Models of Microbiome Evolution Incorporating Host and Microbial Selection

Qinglong Zeng et al. Microbiome.

Abstract

Background: Numerous empirical studies suggest that hosts and microbes exert reciprocal selective effects on their ecological partners. Nonetheless, we still lack an explicit framework to model the dynamics of both hosts and microbes under selection. In a previous study, we developed an agent-based forward-time computational framework to simulate the neutral evolution of host-associated microbial communities in a constant-sized, unstructured population of hosts. These neutral models allowed offspring to sample microbes randomly from parents and/or from the environment. Additionally, the environmental pool of available microbes was constituted by fixed and persistent microbial OTUs and by contributions from host individuals in the preceding generation.

Methods: In this paper, we extend our neutral models to allow selection to operate on both hosts and microbes. We do this by constructing a phenome for each microbial OTU consisting of a sample of traits that influence host and microbial fitnesses independently. Microbial traits can influence the fitness of hosts ("host selection") and the fitness of microbes ("trait-mediated microbial selection"). Additionally, the fitness effects of traits on microbes can be modified by their hosts ("host-mediated microbial selection"). We simulate the effects of these three types of selection, individually or in combination, on microbiome diversities and the fitnesses of hosts and microbes over several thousand generations of hosts.

Results: We show that microbiome diversity is strongly influenced by selection acting on microbes. Selection acting on hosts only influences microbiome diversity when there is near-complete direct or indirect parental contribution to the microbiomes of offspring. Unsurprisingly, microbial fitness increases under microbial selection. Interestingly, when host selection operates, host fitness only increases under two conditions: (1) when there is a strong parental contribution to microbial communities or (2) in the absence of a strong parental contribution, when host-mediated selection acts on microbes concomitantly.

Conclusions: We present a computational framework that integrates different selective processes acting on the evolution of microbiomes. Our framework demonstrates that selection acting on microbes can have a strong effect on microbial diversities and fitnesses, whereas selection on hosts can have weaker outcomes.

Keywords: Commensal and pathogen; Diversity; Microbiome; Selection.

Conflict of interest statement

Ethics approval and consent to participate

Not applicable

Consent for publication

Not applicable

Competing interests

The authors declare that they have no competing interests.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Figures

Fig. 1
Fig. 1
Host reproduction and microbial community assembly under models of selection. a A schematic showing the action of host selection on numbers of offspring. Each small circle represents a host with the first row representing the parental host population and the second row representing the offspring host population. The color indicates the fitness status of hosts which is completely determined by the microbes hosts carry (blue, fit; red, unfit), and the dotted lines indicate the parent-offspring relationship between generations. Fitter hosts produce relatively more offspring, weighted by parental fitness. b The relative abundance of microbial OTUs in host offspring, taking account of microbial fitness values. The letters “P,” “E,” “S,” and “O” stand for parental, environmental, source, and offspring microbial communities, respectively. The values of Y% and X% represent the percentage of host contribution under mixed environment (ME), and the percentage of parental inheritance under mixed acquisition (MA), respectively. Two yellow ellipses represent the source community (either parental, environmental, or mixed microbiomes depending on the value of X), and the offspring microbiome with a blue arrow indicating how microbes are sampled. Each small circle stands for a microbe with its color indicating the fitness status (green, fit; red, unfit but abundant in source community; blue, unfit and rare in source community). Fitter microbes and abundant microbes are more likely to be acquired by offspring
Fig. 2
Fig. 2
From phenomes to host/microbe fitnesses. The diagram shows how our models connect collections of traits (i.e., phenomes) of microbes to host fitness and microbe fitness. a When HS is acting, each trait (represented by a triangle) has a fitness score affecting host fitness (red − 1, green + 1, blue 0). We determined the host absolute fitness and relative fitness with the average fitness score of all microbial traits associated with its microbiome and a host selection coefficient (see “Methods” for details). b When TMS/HMS is acting, each trait has another fitness score that affects microbe fitness (red − 1, green + 1, blue 0). The difference between HMS and TMS is that the fitness score of a certain trait is universal for all hosts under TMS but varies from host to host under HMS. Similarly, microbial absolute fitness and relative fitness are also determined with the average fitness score of all microbial traits possessed by the microbe and a microbial selection coefficient (see “Methods” for details)
Fig. 3
Fig. 3
Alpha diversity patterns under models of microbiome evolution with HMS. Each heatmap corresponds to a combination of host and microbial selection parameters both of which have four different levels: 1 (no corresponding selection), 10, 100, and 1000. Microbial selection is implemented as HMS. For each heatmap, horizontal and vertical axes represent percentages of parental contribution and pooled environmental contribution, respectively. The scales of axes are linear with ranges from 0 to 1. The color bar on the right of the heatmaps indicates the corresponding values for diversity (warm color, high diversity; cold color, low diversity)
Fig. 4
Fig. 4
Alpha diversity patterns under models of microbiome evolution with TMS. With a similar layout, all heatmaps are also plotted in the same way as those in Fig. 3 except that microbial selection here is implemented as TMS, not HMS
Fig. 5
Fig. 5
Composition of beneficial, commensal, and pathogenic microbes in host population under different selective models. Each barplot corresponds to a combination of host and microbial selection parameters both of which have four different levels: 1 (no corresponding selection), 10, 100, and 1000. Microbial selection is implemented as HMS. Each bar represents results averaged from five replicate simulations, and color indicates the types of microbes (blue, beneficial; green, commensal; red, pathogenic). Categories on the horizontal axes refer to different combinations of microbiome acquisition and environmental community assembly processes: MA(0)*ME(0) indicates no contributions from parents either directly or to the environment, MA(50)*ME(50) indicates 50% contribution of the parent to the offspring microbiome and 50% of parent to the environment, and MA(90)*ME(90) indicates 90% parental contribution to offspring microbiome and 90% parental contribution to environmental microbial community. Analysis of variance (ANOVA) tests suggest that the compositional changes of three types of microbes are statistically significant as parental contribution and HMS increases (p value < 0.0001) but not as HS increases
Fig. 6
Fig. 6
The effects of different models on host fitness. The vertical axis in each figure represents the log final fold change (after 200,000 generations) of average host fitnesses (from five replicates) with respect to the initial levels. The categories on the horizontal axes represent different selective models; colors label different host parental contributions to offspring or environmental microbiomes (see Fig. 5 for description). Each barplot corresponds to a combination of host and microbial selection parameters, both of which have three different levels: 10, 100, and 1000. Host fitness is most strongly affected when HS and HMS apply (even when parental contributions to offspring or environmental microbial communities are absent). Compared with HS and HMS, the improvement of HS alone on host fitness under high parental contribution becomes unobservable
Fig. 7
Fig. 7
The effects of different models on microbe fitness. With similar layout, barplots are also plotted in the same way as those in Fig. 6 except that the vertical axis represents the log final fold change (after 200,000 generations and from 5 simulations) of average microbe fitnesses (from 5 replicates) with respect to the initial levels. Microbial fitness is strongly influenced by any selective model in which microbial selection operates, i.e., HMS, TMS, HS and HMS, and HS and TMS

Similar articles

See all similar articles

Cited by 7 PubMed Central articles

See all "Cited by" articles

References

    1. Zeng Q, Sukumaran J, Wu S, Rodrigo A. Neutral models of microbiome evolution. PLoS Comput Biol. 2015;11:e1004365. doi: 10.1371/journal.pcbi.1004365. - DOI - PMC - PubMed
    1. Burns AR, Zac Stephens W, Stagaman K, Wong S, Rawls JF, Guillemin K, Bohannan BJ. Contribution of neutral processes to the assembly of gut microbial communities in the zebrafish over host development. ISME J. 2015;10(3):655–64. - PMC - PubMed
    1. Dumbrell AJ, Nelson M, Helgason T, Dytham C, Fitter AH. Relative roles of niche and neutral processes in structuring a soil microbial community. ISME J. 2010;4:337–345. doi: 10.1038/ismej.2009.122. - DOI - PubMed
    1. McCafferty J, Muhlbauer M, Gharaibeh RZ, Arthur JC, Perez-Chanona E, Sha W, Jobin C, Fodor AA. Stochastic changes over time and not founder effects drive cage effects in microbial community assembly in a mouse model. ISME J. 2013;7:2116–2125. doi: 10.1038/ismej.2013.106. - DOI - PMC - PubMed
    1. Li L, Ma ZS. Testing the neutral theory of biodiversity with human microbiome datasets. Sci Rep. 2016;6:31448. doi: 10.1038/srep31448. - DOI - PMC - PubMed

Publication types

LinkOut - more resources

Feedback