Spatial patterns of bacterial taxa in nature reflect ecological traits of deep branches of the 16S rRNA bacterial tree

Abstract

Whether bacteria display spatial patterns of distribution and at which level of taxonomic organization such patterns can be observed are central questions in microbial ecology. Here we investigated how the total and relative abundances of eight bacterial taxa at the phylum or class level were spatially distributed in a pasture by using quantitative PCR and geostatistical modelling. The distributions of the relative abundance of most taxa varied by a factor of 2.5-6.5 and displayed strong spatial patterns at the field scale. These spatial patterns were taxon-specific and correlated to soil properties, which indicates that members of a bacterial clade defined at high taxonomical levels shared specific ecological traits in the pasture. Ecologically meaningful assemblages of bacteria at the phylum or class level in the environment provides evidence that deep branching patterns of the 16S rRNA bacterial tree are actually mirrored in nature.

Evaluating community–environment relationships along fine to broad taxonomic resolutions reveals evolutionary forces underlying community assembly

“…These findings support the hypothesis of phylogenetic niche conservatism and further suggest that broader taxonomic classification may balance the distribution uncertainty associated with finer taxonomic units and strengthen niche-related signals. Overall, these results are consistent with the current notion of ecological coherence in deep prokaryotic branches (Philippot et al, 2009(Philippot et al, , 2010, which suggests that members of the same prokaryotic clades generally maintain similar ecological characteristics over evolutionary time (Martiny et al, 2015).…”

“…Here we used eight prokaryotic data sets to show that systematically investigating community structure at fine to broad taxonomic resolutions is practical and meaningful. Specifically, our detected patterns (Figure 3) reinforce the current notion of ecological coherence in deep prokaryotic branches (Philippot et al, 2009(Philippot et al, , 2010, which suggest that broad-level taxonomic classification might be useful for making generalizations about the biogeographic distribution of prokaryotic taxa, at least in the context of some environmental drivers. Through these cases, we demonstrate that tracking the strength of community-environment relationships along taxonomic ranks might serve as a means to uncover macroevolutionary patterns, adding a new and important perspective to the evidence regarding niche conservatism or niche divergence in community assembly (Figure 1).…”

“…Conversely, breaking down broader taxa into finer units may result in more noise than information, as the distributions and abundances of those ecologically equivalent units exhibit a high degree of randomness caused by neutral-stochastic processes instead of nichedeterministic processes (Lozupone et al, 2012;Prosser, 2012;DeLong, 2014). Actually, previous studies have indicated that despite the considerable variability in species-level composition, prokaryotic communities in a given habitat usually have stable phylum-level composition as well as similar functional attributes (Fierer et al, 2007;Lauber et al, 2009;Burke et al, 2011;Fan et al, 2012;Lozupone et al, 2012); these findings further support the view that there is ecological coherence in deep prokaryotic branches (Philippot et al, 2009(Philippot et al, , 2010) and potentially many fine-level taxa are functionally redundant and fulfill the same ecological role (Lozupone et al, 2012;Prosser, 2012;DeLong, 2014).…”

“…We observed a sharp increase of explained variation when grouping otu97 into otu94 and broader taxa (for example, at least 50% increase in R 2 from otu97 to otu94 for soil cases; Figure 3), suggesting that under conditions of strong habitat filtering, some sequence-derived finely resolved taxa might exhibit similar responses to environmental conditions and behave as the same ecologically cohesive units (Philippot et al, 2009(Philippot et al, , 2010. Previous studies have hypothesized ecological equivalence for finely resolved taxa such that a particular niche may be occupied by any of the suitable taxa drawn from a large pool of candidates (Lozupone et al, 2012;Prosser, 2012;DeLong, 2014).…”

Towards soil geostatistics

“…Lark et al [8] reviewed and illustrated this approach in a paper for the soil science community, who have taken up the approach widely. See, for example, work in soil hydrology [9], soil pollution [10] and in soil biology [11]. The solution to the technical problem of estimating the fixed effects and random effects coefficients that bedevilled classical geostatistics is important because it immediately opens up the possibility of using the fixed effects component of the model as a vehicle to incorporate soil knowledge.…”