Soil microorganisms (e.g., bacteria, fungi) play key roles in regulating terrestrial ecosystem processes such as nutrient cycling and carbon sequestration. However, despite the growing catalogue of studies that characterize the taxonomic diversity and composition of soil microbial communities, the basic life history traits (e.g., growth rates) of those communities remain largely unknown. The goal of our study was to explore global patterns in the life history traits of soil bacterial communities and identify the dominant environmental drivers of those patterns. To accomplish our goal, we leveraged an existing dataset comprised of 189 soil metagenomes (Illumina HiSeq) sampled across 11 biomes on six continents. We then applied a suite of bioinformatic analyses to estimate four life history metrics for each bacterial community: metagenome GC content, average genome size, average number of 16S rRNA operons per organism, and predicted maximum growth rates. We assessed average genome size by quantifying the mean coverage of 30 prokaryotic single-copy genes and we quantified average number of 16S operons per organism by aligning the metagenomic sequence reads to a custom 16S gene database and then calculating 16S coverage per genome. Finally, we predicted maximum community-averaged growth rates by identifying gene coding regions in the sequence reads and then analyzing codon usage patterns in the highly expressed ribosomal protein genes. We found that all measured life history metrics exhibited clear latitudinal patterns – for example, bacterial communities from tropical latitudes (e.g., rainforests, tropical montane forests) had lower GC content, larger genomes, more 16S copies per organism, and higher maximum growth rates when compared to more arid subtropical/temperate ecosystems (i.e., savanna and mediterranean environments). Interestingly, higher latitude environments (i.e., temperate and boreal forests) were similar to tropical soil communities with respect to life history traits, i.e., low GC content, larger genome size, and high growth rates. This resulted in distinct quadratic patterns of GC content, genome size, and growth rates with increasing latitude. These results indicate that soil bacterial communities in both tropical and high latitude forests can be generally characterized as more copiotrophic (higher growth rates) while more arid subtropical communities are adapted to environmental stress (higher GC content). Finally, random forest regression revealed the dominant drivers of the life history traits to be soil pH, soil organic carbon content, and ecosystem net primary productivity. Overall, our study provides novel information regarding the ecological characteristics and functions of soil bacterial communities at a global scale.
