Phylogenetic conservatism of thermal traits explains dispersal limitation and genomic differentiation of Streptomyces sister-taxa

Abstract

The latitudinal diversity gradient is a pattern of biogeography observed broadly in plants and animals but largely undocumented in terrestrial microbial systems. Although patterns of microbial biogeography across broad taxonomic scales have been described in a range of contexts, the mechanisms that generate biogeographic patterns between closely related taxa remain incompletely characterized. Adaptive processes are a major driver of microbial biogeography, but there is less understanding of how microbial biogeography and diversification are shaped by dispersal limitation and drift. We recently described a latitudinal diversity gradient of species richness and intraspecific genetic diversity in Streptomyces by using a geographically explicit culture collection. Within this geographically explicit culture collection, we have identified Streptomyces sister-taxa whose geographic distribution is delimited by latitude. These sister-taxa differ in geographic distribution, genomic diversity, and ecological traits despite having nearly identical SSU rRNA gene sequences. Comparative genomic analysis reveals genomic differentiation of these sister-taxa consistent with restricted gene flow across latitude. Furthermore, we show phylogenetic conservatism of thermal traits between the sister-taxa suggesting that thermal trait adaptation limits dispersal and gene flow across climate regimes as defined by latitude. Such phylogenetic conservatism of thermal traits is commonly associated with latitudinal diversity gradients for plants and animals. These data provide further support for the hypothesis that the Streptomyces latitudinal diversity gradient was formed as a result of historical demographic processes defined by dispersal limitation and driven by paleoclimate dynamics.

Fig. 1

Streptomyces lineages NDR and SDR were partitioned across sites with respect to latitude. The phylogenetic tree was constructed from multiple whole-genome alignments using maximum likelihood and a GTRGAMMA model of evolution. Scale bar represents nucleotide substitutions per site. Bootstrap values for all internal nodes are 100%. Colored clades depict the northern-derived (NDR) and southern-derived (SDR) populations. The remaining strains belong to the intermediate groups (INT). Strain names reflect the location they were isolated from (Table S1). Genome NBRC13350 is the type strain Streptomyces griseus subsp. griseus NBRC 13350. The geographic distribution of strains s is illustrated in the right panel. Sample locations are labeled with their site codes, which are found in Table S1 and are colored to reflect population ranges (color figure online)

Fig. 2

NDR and SDR lineages exhibit substantial genome-wide genetic differentiation as revealed by pairwise values of genetic differentiation (F_ST) and average nucleotide identity (ANI) for 2778 single-copy core genes. Kernal density distributions for gene-by-gene ANI and F_ST are displayed on the corresponding axes

Fig. 3

Overall genomic diversity is lower in NDR compared with SDR as assessed by pairwise analysis of 2778 single-copy core genes with respect to average nucleotide identity (ANI) a, percentage of segregating sites b, and per site nucleotide diversity c. Boxplots show the interquartile range of values. The median is indicated by the central horizontal line, whiskers indicate the 1.5 interquartile range beyond the box boundaries, and black points indicate outlying values. Colored circles further illustrate the distributions of diversity metrics (color figure online)

Fig. 4

NDR exhibits a tradeoff in thermal traits with higher relative growth rate than SDR at 4 ˚C and lower relative growth rate at 23 ˚C, 30 ˚C, and 35 ˚C. Relative growth rate is defined as (GR_str–GR_ref)/GR_ref. Circles depict the population-level mean values, and error bars depict the population-level standard deviations. The same trend is observed in absolute growth rate (Figure S1), but the use of relative growth rate standardizes the effect size for differences in growth rate across temperature