Metagenômica do microcosmo de solo sob a aplicação de fipronil

Abstract

Fipronil is an insecticide widely used in agriculture for pest control. However, its effect on the bacterial community in soil is still not completely understood. In this study, the goal was to infer the bacterial diversity from the 16S ribosomal RNA gene in microcosm of semideciduous forest soil under application of fipronil for 28 days. Microcosm experiments were performed with the semideciduous forest soil collected from the experimental farm of Embrapa Agropecuária Oeste, with weekly applications of fipronil for four weeks. Assays were performed in three replicates. From the DNA extracted from microcosm soil, the metagenomic library of V4 and V5 hypervariable regions of the 16S ribosomal RNA was prepared and sequencing was carried on Ion Torrent PGM platform. Sequences were processed for analysis on BMP (Brazilian Microbiome Project) and MICCA (Microbial Community Analysis) pipelines with adaptations. Files from the pipelines were run in STAMP software for statistical analysis. Functional prediction was performed using PICRUSt (Phylogenetic Investigation of Communities by Reconstruction of unobserved States) software. Statistical differences occurred on the taxonomic level of class. Both BMP and MICCA pipelines resulted in similar classes, while genus were different. The main difference between both pipelines is the step of Operational Taxonomic Units (OTUs) clustering, which singletons (OTUs represented by a single DNA sequence) are discarded in BMP while MICCA does not discard them. Streptomyces was the genus with the highest difference in proportion for control. This fact may indicate the group is susceptible to fipronil or the pesticide inhibits its growth. However, the effect of fipronil on this genus has not been reported in the literature. The functional prediction presented different enzymes from both pipelines.