Community structure and abundance of ACC deaminase containing bacteria in soils with 16S-PICRUSt2 inference or direct acdS gene sequencing

Abstract

Bacteria containing the enzyme 1-aminocyclopropane-1-carboxylate deaminase (ACCD+) can reduce plant ethylene levels and increase root development and elongation resulting in increased resiliency to drought and other plant stressors. Although these bacteria are ubiquitous in the soil, non-culture-based methods for their enumeration and identification are not well developed. In this study we compare two culture-independent approaches for identifying ACCD+ bacteria. First, quantitative PCR (qPCR) and direct acdS sequencing with newly designed gene-specific primers; and second, phylogenetic construction of 16S rRNA amplicon libraries with the PICRUSt2 tool. Using soils from eastern Colorado, we showed complementary yet differing results in ACCD+ abundance and community structure responding to water availability. Across all sites, gene abundances estimated from qPCR with the acdS gene-specific primers and phylogenetic reconstruction using PICRUSt2 were significantly correlated. However, PICRUSt2 identified members of the Acidobacteria, Proteobacteria, and Bacteroidetes phyla (now known as Acidobacteriota, Pseudomonadota, and Bacteroidota according to the International Code of Nomenclature of Prokaryotes) as ACCD+ bacteria, whereas the acdS primers amplified only members of the Proteobacteria phyla. Despite these differences, both measures showed that bacterial abundance of ACCD+ decreased as soil water content decreased along a potential evapotranspiration (PET) gradient at three sites in eastern Colorado. One major advantage of using 16S sequencing and PICRUSt2 in metagenomic studies is the ability to get a potential functional profile of all known KEGG (Kyoto Encyclopedia of Genes and Genomes) enzymes within the bacterial community of a single soil sample. The 16S-PICRUSt2 method paints a broader picture of the biological and biochemical function of the soil microbiome compared to direct acdS sequencing; however, phylogenetic analysis based on 16S gene relatedness may not reflect that of the functional gene of interest.