Consistent divergence and drainage-specific convergence of gut
microbiomes
We investigated variation in the guppy gut microbiome along an
evolutionary ‘chronosequence’ of time-since-introduction, hypothesizing
that gut microbiota tracks the evolution of fish traits such as gut
morphology. Our hypothesis was supported in that, when HP fish were
introduced into LP sites, their gut morphology changed to resemble that
of LP native populations, and their microbiomes tracked this change.
Coevolution of host and microbiome has been observed in
macroevolutionary studies (Ley et al. 2008; Muegge et al.2011), but not always in replicate, independent divergences in the wild
(Sevellec et al. 2018; Härer et al. 2020). We uniquely
show that coevolution not only occurs but can be quite rapid; in our
study, HP guppies translocated to LP environments after 5-6 years
(approximately 15-18 guppy generations) were already more similar to
native LP fish, despite their genetic background being more similar to
their HP source populations (Fitzpatrick et al. 2015). This rapid
microbiome divergence was driven by shifts in phenotypic traits and
environment, suggesting that these factors are equally important for
driving microbiome as diet and host genetics, which are more commonly
studied.
While we observed dramatic and consistent divergence from HP
microbiomes toward LP native microbiomes, even in the first 5 years
after translocation, we did not observe consistent convergence of
introduced populations with LP natives in every drainage. Although we
cannot be sure, lack of convergence with LP natives does not appear to
due to insufficient time. Some of the most recent introduced populations
were indistinguishable from LP native microbiomes (Guanapo drainage),
while one of the oldest introductions (El Cedro, 57 years in LP
environment) remained distinct from LP natives. Abiotic differences
among sites most strongly explained why LP and introduced microbiomes
were not more similar. But while these site-specific effects led Sullamet al. (2015) to rule out parallel microbiome co-evolution in
this system (because LP populations did not converge) our larger study
suggests this should not be ruled out. Rather, the consistent divergence
from HP source, drainage-dependent convergence with LP natives, and
potential mechanisms for microbiomes to impact nutrition (see third
section) suggest the gut microbiome affects host evolution while in the
context of other agents influencing microbiome assembly (e.g. horizontal
transmission). We further point out that parallelism at the functional
level could occur without documenting consistent parallelism at the
taxonomic level, especially given high functional redundancy in
microbial communities (Louca et al. 2017).