Metabarcoding of the phytotelmata of

background

methodsWater from 10 individuals was sampled. Five individuals had abundant surrounding vegetation, and five had little or no vegetation. We extracted DNA and amplified seven hypervariable regions of the 16S gene (V2, V4, V8, V3-6, 7-9). Metabarcoding sequencing was performed on the Ion Torrent PGM platform. Taxonomic identity was assigned by the binning reads and coverage between hit and query from the reference database of at least 90%. Putative metabolic functions of the bacterial families were assigned mainly using the FAPROTAX database. The dominance patterns in each site were visualized with rank/abundance curves using the number of Operational Taxonomic Units (OTUs) per family. A percentage similarity analysis (SIMPER) was used to estimate dissimilarity between the sites. Relationships among bacterial families (identified by the dominance analysis and SIMPER), sites, and their respective putative functions were analyzed with shade plots.

resultsA total of 1.5 million useful bacterial sequences were obtained. Sequences were clustered into OTUs, and taxonomic assignment was conducted using BLAST in the Greengenes databases. Bacterial diversity was 23 phyla, 52 classes, 98 orders, 218 families, and 297 genera. Proteobacteria (37%), Actinobacteria (19%), and Firmicutes (15%) comprised the highest percentage (71%). There was a 68.3% similarity between the two sites at family level, with 149 families shared. Aerobic chemoheterotrophy and fermentation were the main metabolic functions in both sites, followed by ureolysis, nitrate reduction, aromatic compound degradation, and nitrogen fixation. The dominant bacteria shared most of the metabolic functions between sites. Some functions were recorded for one site only and were related to families with the lowest OTUs richness. Bacterial diversity in the