PS 35-171 - Evolution of microbial genomic traits associated with adaptation to long-term soil warming

Tuesday, August 16, 2022

5:00 PM – 6:30 PM EDT

Located in: ESA Exhibit Hall

Presenting Author(s)

MC

Mallory J. Choudoir

UMass Amherst
Amherst, Massachusetts, United States

Coauthor(s)

XA

Xiao-Jun Allen Liu

University of Oklahoma, United States
KD

Kristen M. DeAngelis

University of Massachusetts Amherst, United States

Background/Question/Methods

Soil warming due to climate change threatens to disrupt microbial communities and their ecosystem functions. Microbial eco-evolutionary responses to climate change will ultimately determine the flux of carbon between terrestrial and atmospheric systems. Forest soils at the Harvard Forest Long-term Ecological Research (LTER) site (Petersham, MA USA) sustained in situ warming 5 ºC above ambient temperatures for nearly 30 years. Decades of chronic warming decreased the quantity and quality of soil carbon and also altered microbial community structure, composition, and function. Previous studies suggest microbes in heated plots evolved adaptive traits related to utilization of complex carbon substrates and oligotrophic growth. What genomic traits underlie adaptation to soil warming? How do these traits evolve, and at what rate? We built a culture collection of bacteria isolated from heated and control plots and sequenced their genomes. We also generated metagenomic datasets from Harvard Forest soils sampled almost a decade apart. To leverage these data, we mapped metagenomes onto isolate genomes to gain insight into the genetic diversity of the environmental populations our isolates belong to. This project applies a population genetics framework to identify genomic traits related to soil warming, and to infer patterns and rates of molecular evolution.

Results/Conclusions

Comparative genomics of Harvard Forest isolates reveals differences in codon usage bias. We observe a significant interaction between genus and warming treatment (P < 0.001), suggesting potential adaptations related to growth efficiency. By recruiting short reads onto isolate genomes (i.e. reference sequence), we identified Bradyrhizobium spp. with sufficient representation across all 12 metagenomes (6 each heated and control). On average, detection across metagenomes ranges from 82–96% of total gene content per isolate. Approximately 10–40% of the total gene content per isolate recruited reads at ≥10X coverage, and these genomic regions have 18.5X mean coverage. Preliminary analysis shows approximately 4300 genes per genome with single nucleotide variants (SNVs) detected across the reference sequence and both heated and control environmental populations. From SNV data, we quantify allele abundances and the extent of genetic differentiation between heated and control populations. We will also determine the extent of nonsynonymous and synonymous variants that contribute to observed differences in codon usage bias. Together, these data inform a rate estimation of molecular evolution across the warming experiment. Microbial evolutionary parameters are largely absent from global carbon models, and rates of evolution are crucial for understanding the ecosystem-level impacts of irreversible microbial adaptations to environmental stress.