Marine microbial community structure assessed from combined metagenomic analysis and ribosomal amplicon deep-sequencing

ABSTRACT

The microbial taxonomic composition of the three domains of life in two coastal plankton samples was assessed by random total community metagenomic sequencing and PCR-based rDNA amplicon deep-sequencing in order to compare the resulting diversity and investigate possible limitations and complementarities of each method. The various universal primer sets, used to amplify different hypervariable rDNA regions, revealed the same major high-level taxonomic groups in Bacteria and unicellular Eukaryota, and showed a scarce Archaea apparent richness. However, significant differences were found between the different primer sets (p-value < 0.05, with the Kolmogorov–Smirnov test), regarding both operational taxonomic unit (OTU) richness and relative abundance of the major high-level taxonomic groups detected. Based on the metagenomic approach, the phylum Bacteroidetes dominated the prokaryotic community, followed by Proteobacteria, while the detected eukaryotic unicellular taxa belonged to the groups of Alveolata, Fungi, Chlorophyta, Stramenopiles and Phaeophyceae. These groups were found to carry genes typically found in microbial communities, which are linked to DNA, RNA and protein metabolism and the synthesis of nucleotides, amino acids, carbohydrates and vitamins. Although our findings suggest that the total community metagenomic approach can provide a more comprehensive picture of the planktonic microbial community structure, a number of issues associated with this approach emerged. These issues include the still relatively high cost compared to amplicon sequencing, the possible low coverage of the full marine diversity, the insufficiency of databases for other gene markers than the small subunit gene, and the bias towards bacterial sequences because of their higher abundance relative to eukaryotes in marine environments.

KEYWORDS:

• Gene pool
• marine microbial community
• metagenome
• pyrosequencing
• SSU rRNA

Acknowledgements

This study was supported by the ‘Nord-Pas de Calais’ FRB-DEMO (FRB_2013) and the ANR-ROME (ANR 12 BSV7 0019 02) projects, and the SOMLIT network. We are grateful to Eric Lecuyer for running the logistics and the SOMLIT sampling and Elsa Breton for validating the physical and chemical values. The authors would also like to thank those at www.englisheditor.webs.com for their help in the English proof-reading of this paper. We are thankful to the three referees who helped improve the original manuscript with their constructive comments.

Disclosure statement

No potential conflict of interest was reported by the authors.