Use of PacBio sequencing for metagenomics to revealed taxonomic and metabolic diversity of beneficial plant-growth promoting bacteria associated with the wheat rhizosphere.

Dlamini, S.*¹, Ngwentle, N.², Coutinho, T.², Mehlomakulu, N. N.³, Reva, O.¹

¹ Centre for Bioinformatics and Computational Biology (CBCB), BGM, University of Pretoria
² Centre for Microbial Ecology and Genomics (CMEG), BGM, University of Pretoria
³ Dep. Consumer and Food Sciences, University of Pretoria

Background: The project aimed to study soil and rhizobacterial microbiomes in a South African wheat field to identify new effective biofertilizers and biocontrol agents using 3rd generation NGS approaches and computational tools for metagenomics. The application of PacBio technology for soil metagenomics facilitated the precise identification of microbial taxa and contig assembly. Methods: The wheat rhizosphere soil samples used were collected from an irrigated field (34.08652oS, 20.27532oE) and a field dependent on rainfall (34.08551oS, 20.26628oE) in Swellendam, Western Cape. The samples for sequencing were collected at different growth stages of wheat plant (the terminal spikelet stage, before harvest and before new crop cycle stage). The extracted DNA was used for metabarcoding, followed by amplification with standard bacterial 16S rRNA and fungal ITS primers, as well as for direct shotgun metagenomics. SMRT PacBio Sequel 2 sequencing was used for both metabarcoding and metagenomics. Results: In total, 120 soil samples for 16S and ITS rRNA metabarcoding and 15 samples for shotgun WGS metagenomics were processed. 16S/ITS reads binned by BLASTN against the NCBI nt database revealed 15 244 bacterial and 1 484 fungal taxa. Comparative analysis showed that the microbiome of wheat fields differed significantly from the microbiomes of bulk soil (control). Moreover, the wheat field microbiome evolved with development to mature plants towards a higher abundance of rhizosphere-associated microorganisms, such as Pseudomonas, Flavobacterium and Bacillus. Shotgun reads were assembled using Flye and annotated by Prokka. In total 6 425 contigs were assembled with lengths up to 2 mbp. Representative genes coding for known phylogenetic markers were identified in each contig and binned to taxa using the DIAMOND program. Contigs were grouped by their taxonomic provinces. In total, the contigs were binned into 1 198 bacterial, 16 archaeal and 19 fungal metagenome-assembled genomes (MAGs). Different approaches and software tools were used to identify genes associated with specific fertilization and biocontrol activities of the soil microbiome. The highest number of “green” genes was found in MAGs identified as Flavobacterium spp., Cellvibrio sp., Pantoea sp., Stenotrophomonas sp., Chloroflexota, and Pseudomonas sp. Conclusion: This study allowed the identification of bacterial species, especially in the soil–plant-microbe interaction aiding in the selection of the most appropriate candidates for biofertilizer and biocontrol design. The study also addressed benefits and challenges of application of 3rd generation sequencing technologies for metagenomics.

Keywords: microbiome, functional metagenomicsm, wheat rhizosphere, plant-growth promoting bacteria, metabarcoding