Metagenomics survey of major metabolic network of sunflower microbiota

Abstract

The microbial communities inhabiting the root, termed the rhizosphere, are in a symbiotic association with their host. However, the plant-microbe interaction study at its current stage is still an evolving field of science. Though still largely unexplored, the soil consists of a metabolically active microbiome where microorganisms are abundant. They provide beneficial services to the host plant, such as protection against pathogens and mineral acquisition enhancement, which in the long run help in plant growth and health. Therefore, it is imperative to explore the microbes inhabiting this niche and what they do at the functional level. This study was designed to use high throughput sequencing and computational approaches to study the soil microbiome taxonomical and functional traits induced by the cultivation of sunflower. Sunflower rhizosphere and bulk soil samples collected from Palmietfontein (R1) Bloemhof rhizosphere (R2) and bulk soil Palmietfontein (B1) Bloemhof bulk (B2) in South Africa were used for the shotgun metagenomic sequencing using Illumina HiSeq platform. MG-RAST was used for the taxonomical and functional analysis of the metagenomic sequences. The metagenomes were then mapped against the SEED and KEEG subsystems database where gene descriptions were assigned. The most dominant domains were bacteria, eukaryote, and archaea. In the R1 sample bacteria accounted for 98.82% of the obtained sequences, followed by eukaryote, which accounted for 0.81% and archaea was 0.29%. 98.47% of R2 sequences belonged to bacteria, 1.23% to eukaryote and 0.2% to archaea. Sequences in bulk soil samples were assigned to bacteria 98.61% and 98.53%, eukaryote 0.82% and 1.05%, and archaea 0.48% and 0.34% in B1 and B2 samples, respectively. The most dominant phyla in the rhizosphere, which also shared same features as the bulk soil were the Actinobacteria, Proteobacteria, Acidobacteria, Bacteroidetes, Planctomycetes, and Verrucomicrobia. There was no significant difference (p value > 0.05) in the structural

composition of the alpha diversity of the sunflower rhizosphere and bulk soil and also the two locations sampled while there was a significant difference in beta diversity, which was visualized using principal coordinate analysis (PCoA), which explained axis 1 and 2 with a combined variation of 85.90%. The metabolic pathways and functional attributes in the sunflower microbial community were also determined. There was no significant difference (p value >0.05) within the functional diversity categories of the sunflower rhizosphere and bulk soils. However, PCoA, which explained the β functional diversity between the sunflower rhizosphere and bulk soil showed a clear separation between them. This was ascertained using the analysis of similarities (ANOSIM), which revealed that there was a significant difference in the functional categories of the sunflower rhizosphere and bulk soils and also there was significant difference in the two locations (p value = 0.01; R= 0.58). The PCoA showed that axis 1 and 2 had combined community variation of 52.39%. Principal component analysis (PCA) showed the distribution of the functional categories for both microhabitats. Motility and chemotaxis, photosynthesis, stress response, iron acquisition and metabolism, and cell wall and capsule placed the sunflower rhizosphere (R1) apart from B2, R2, and B1. Strikingly from our study, functional annotation of the genes revealed genes coding for nitrogen fixation (nifH), siderophore production, 1-aminocyclopropane-1-carboxylate (ACC) deaminase (acdS), mineral phosphate solubilization (ppx/gppA), exopolysaccharide biosynthesis glycosyltransferase (epsF, exoF), and high temperature stress response genes (htrA). From previous reports, these genes have been discovered to help in the growth and health of plants. Location R1 was more enriched in genes potentially needed for plant growth and development, which means sunflower would most likely thrive well there. Our canonical correspondence analysis to determine whether physicochemical variables drive the microbiome revealed calcium was the best predictor for structural diversity, while phosphorus was the best predictor for

functional gene categories. These results indicated that the microbial community structure and function were closely correlated with environmental factors. This study shows that microbial consortia have the potential to be efficiently used as bioinoculants to optimize crop growth and health.