Nanopore sequencing for aquaculture bacterial microbiota profiling

Caitlin E. Older1, Fernando Y. Yamamoto2, Matt J. Griffin3, Cynthia Ware3, Brian G. Bosworth1,

Geoffrey C. Waldbieser1

 1Warmwater Aquaculture Research Unit, Agriculture Research Service, U.S. Department of Agriculture, Stoneville, Mississippi, USA; 2Thad Cochran National Warmwater Aquaculture Center, Mississippi Agriculture and Forestry Experiment Station, Department of Wildlife, Fisheries, and Aquaculture, College of Forest Resources, Mississippi State University, Stoneville, MS, USA; 3Thad Cochran National Warmwater Aquaculture Center, Aquatic Research and Diagnostic Laboratory, Department of Pathobiology and Population Medicine, College of Veterinary Medicine, Mississippi State University, Stoneville, MS, USA

ABSTRACT

Bacterial community profiling in aquaculture can be useful for monitoring environmental conditions which may lead to disease outbreaks, understanding the impact of dietary treatments on fish health, and identifying microbes which may be useful as antibiotic alternatives. Microbiota characterization is most commonly performed using short-read platforms (i.e. Illumina) targeting informative fragments of the bacterial 16S rRNA gene. Comparably, long-read platforms (i.e. Oxford Nanopore Technology) enable sequencing of full-length 16S rRNA genes. While short-read platforms are highly accurate, long-read platforms sacrifice sequence quality in exchange for longer read lengths. Herein, a typical short-read approach (Illumina MiSeq; V4 region of 16S rRNA gene) was compared with two nanopore near full-length 16S rRNA protocols (ONT Custom and ONT 16S). Sequencing was performed on a mock community composed of fish-relevant bacteria, in addition to samples obtained from ten channel x blue hybrid catfish (digesta, gill, skin) and their environment (pond and tank). To assess the role of euthanasia on microbiota assessments, five fish were euthanized by overdose of MS-222, while the remaining fish were euthanized by cranial concussion and subsequent pithing. This enabled additional comparisons of these three sequencing protocols in the context of a realistic experimental design. Results from the ONT Custom protocol best recapitulated the theoretical composition of the mock community, followed by the Illumina MiSeq and ONT 16S protocols. At the phylum level, taxonomic composition was consistent within sample type across sequencing methods. However, at the lower levels there were biases for several taxa, particularly in pond samples. Euthanasia method also appeared to introduce bias, primarily on the cutaneous communities. The three sequencing methods consistently identified significant differences in beta diversity and taxonomic composition between euthanasia methods, with Illumina MiSeq revealing the greatest differences. Sample type also influenced results; while all three strategies performed comparably for digesta, skin and water (pond and tank), the ONT strategies underperformed for gill samples. Depending on application, results indicate near full-length 16S rRNA gene nanopore sequencing is a viable option for aquaculture microbiota studies.