Using Sediment DNA Archives for Interpreting Long-term Cyanobacterial Dynamics in the Anthropocene

Abstract

Climate change and eutrophication, accelerated by anthropogenic activities, have impacted aquatic ecosystems worldwide. These impacts have stimulated the expansion of cyanobacterial blooms which pose severe threats to ecosystem functioning, environmental health, and the economy. However, the long-term effects of environmental change on bloom-forming cyanobacteria are not well understood as traditional paleolimnological approaches are of limited use in the reconstruction of cyanobacterial dynamics through time. Here, sediment DNA (sedDNA) was used to investigate long-term cyanobacterial trends using sediments from two experimental (fertilized L227 and acidified L223) and two reference (L224 and L442) lakes in the Experimental Lakes Area, Canada. First, to determine whether taxonomic bias might arise from the cyanobacterial sediment record, I performed a 1-year incubation experiment comparing the degradation rates of selected cyanobacterial genes under contrasting environmental conditions. Based on first-order linear decay models, Synechococcus sp. (Synechococcales) decayed the slowest under cold, anoxic conditions, followed by Trichormus (Nostocales), then Microcystis (Chroococcales), suggesting differential preservation of DNA. I then compared the quantitative performance of droplet digital polymerase chain reaction (ddPCR) and high-throughput sequencing (HTS) for the analysis of sedDNA and found that the ddPCR results were more consistent with the known history of the lakes. Furthermore, ddPCR showed that cyanobacterial abundance increased over the past century in all study lakes, but the greatest increase was observed in experimentally fertilized L227. HTS revealed shifts in the cyanobacterial community towards Nostocales dominance and a decrease in alpha diversity in response to phosphorus-only additions. An increase in abundance of the mcyE gene (indicative of microcystin producing taxa) was uniquely observed in L227 when nitrogen additions ceased. Heating degree days were important in explaining variation in the cyanobacterial community composition in all lakes, but nutrients had a greater influence on the L227 community. When sediment data were compared to historical surface water phytoplankton records, moderate to strong correlations between the two archives were found, validating the use of sedDNA. This research demonstrated that sedDNA can elucidate cyanobacterial trends at the community, population, and species level over multidecadal timescales in response to environmental change.