Shifts in soil bacterial communities as a function of carbon source used during anaerobic soil disinfestation

Abstract

Anaerobic soil disinfestation (ASD) is an organic amendment-based management practice for controlling soil-borne plant pathogens. Pathogen suppression appears to be carbon source-dependent and mediated by bacteria that proliferate and produce volatile organic compounds, as well as physico-chemical changes (i.e., elevated temperature, lowered redox potential and pH, release of metal ions) in soil. ASD is under study for adoption in tree crops as a replacement for chemical-fumigation, but its widespread use is limited by incomplete understanding of its suppression mechanisms and high economic costs. The carbon substrate is one component of the ASD process that can be optimized to enhance effectiveness and affordability. While rice bran is currently the standard carbon source used for ASD, we identified three alternative substrates (molasses, mustard seed meal, and tomato pomace) that are similar in efficacy to rice bran at generating and sustaining soil anoxia and reducing populations of introduced plant pathogens. Here, we used replicated ASD field trials to determine if rice bran and the alternative carbon substrates would elicit similar soil bacterial communities (characterized via amplicon sequencing of the 16S rRNA gene v4 region) and to assess if any observed community shifts were consistent across repeated trials. We found significant, but minimal differences in community composition between ASD carbon treatments (F4, 30 = 2.80, P < 0.001, R2 = 0.22) and trials (F1, 30 = 5.24, P < 0.001, R2 = 0.10). In both trials, the abundances of Bacteroidales, Clostridiales, Selenomonadales, and Enterobacteriales increased significantly (> 5 log2 fold change) in all ASD treatments compared to untreated controls. A group of shared core genera belonging to the Clostridiales and Selenomonadales were identified in both trials and constituted 22.6 and 21.5% of the communities. Bacterial taxa that were most responsive to ASD treatments had the genomic potential for denitrification, nitrogen fixation, and fermentation reactions that produce organic acids (such as acetate and butyrate) known to inhibit in vitro growth of plant pathogens based on predicted metagenomes. Together, these results indicate that reproducible and effective implementation of ASD is achievable with alternative carbon substrates to rice bran.