Rapid assessment of phytoplankton assemblages using Next Generation Sequencing and Barcode of Life Data System: a widely applicable HAB-ID toolkit for detecting and monitoring biodiversity loss and harmful algal blooms

Abstract

Harmful algal blooms have important implications for the health, functioning, and services of aquatic ecosystems. Our ability to detect and monitor these events is often challenged by the lack of rapid and cost-effective methods to identify bloom-forming organisms and their potential for toxin production. Here, we developed and applied a combination of DNA barcoding and Next Generation Sequencing (NGS) for the rapid assessment of phytoplankton community composition with a focus on two important indicators of ecosystem health: toxigenic bloom-forming cyanobacteria and impaired planktonic biodiversity. To develop this molecular toolset for identification of cyanobacterial and algal species present in HABs (harmful algal blooms), hereafter called HAB-ID, we achieved three goals: creating a validated reference database, optimizing molecular protocols, and developing original bioinformatics pipeline tailored to uncertainty of algal taxonomy. The BOLD (Barcode of Life Data System) 16S reference database from cultures of 211 cyanobacterial and algal strains representing 102 species with particular focus on bloom and toxin producing taxa was constructed with Sanger sequencing and further refined using Single Molecule Real Time Sequencing (SMRT-sequencing). Using the new reference database of 16S rDNA sequences and constructed mock communities of mixed strains for protocol validation, we developed new NGS primer sets which can recover 16S from both cyanobacteria and eukaryotic algal chloroplasts. We also developed DNA extraction protocols for cultured algal strains and environmental samples, which match commercial kit performance and offer a cost-efficient solution for large scale ecological assessments of harmful blooms while giving benefits of reproducibility and increased accessibility. Our innovative bioinformatics pipeline was designed to handle low taxonomic resolution for problematic genera of cyanobacteria such as the Anabaena-Aphanizomenon-Dolichospermum complex, two clusters of Anabaena (I and II), Planktothrix and Microcystis. This newly developed HAB-ID toolset was further validated by applying it to assess cyanobacterial and algal composition in field samples from waterbodies with recurrent HABs events.

Keywords: 16S rDNA; Harmful algal blooms; Next Generation Sequencing (NGS); Phytoplankton community; Reference database; The Barcode of Life Data Systems (BOLD).

Figure 1. Project phases—from initial optimization to high-throughput analysis of environmental samples.

Figure 2. Class level taxonomy of clustered CCS reads (percent of total for each primer pair) inferred from 16S diversity tool in Geneious software.

Figure 3. Neighbor-joining tree visualized in iTOL, colored ranges correspond to taxa complexes and genera with low resolution used in BLAST search assignments and OTUBLASTparser.py script.

Figure 4. Genera count for four filters from mock community (expected versus microscopy and NGS) and Venn diagram representing overall genera count of submitted strains from four filters and their detection using microscopy and NGS.

Figure 5. Order level recovery with two primer pairs on four mock communities and nine environmental samples processed using two primer sets CYA359F-CYA781R and CYA359Fd-CYA781Rd-euk.

Figure 6. Read coverage (bottom) and taxonomic recovery for genus and species level identifications using NGS (top) from samples collected at four stations in Lake Erie tested using two DNA extraction methods.

GF–CCDB in house protocol, PW–MO-BIO PowerWater; 3-repl and 24-repl correspond to three and 24 replicates in PCR.

Figure 7. Distribution of identification top hits for environmental samples from Lake St.

Clair and Lake Winnipeg collected and extracted in three replicates (A, B, C) and amplified using M13-tailed primer cocktail.

Figure 8. Comparison of genera counts detected by microscopy and NGS for samples collected in August 2015 at four stations in Lake Erie (left) and cell counts/NGS read coverage (right) colored by phylum for both graphs.