Expression of Microcystis Biosynthetic Gene Clusters in Natural Populations Suggests Temporally Dynamic Synthesis of Novel and Known Secondary Metabolites in Western Lake Erie

Abstract

Microcystis spp. produce diverse secondary metabolites within freshwater cyanobacterial harmful algal blooms (cyanoHABs) around the world. In addition to the biosynthetic gene clusters (BGCs) encoding known compounds, Microcystis genomes harbor numerous BGCs of unknown function, indicating a poorly understood chemical repertoire. While recent studies show that Microcystis produces several metabolites in the lab and field, little work has focused on analyzing the abundance and expression of its broader suite of BGCs during cyanoHAB events. Here, we use metagenomic and metatranscriptomic approaches to track the relative abundance of Microcystis BGCs and their transcripts throughout the 2014 western Lake Erie cyanoHAB. The results indicate the presence of several transcriptionally active BGCs that are predicted to synthesize both known and novel secondary metabolites. The abundance and expression of these BGCs shifted throughout the bloom, with transcript abundance levels correlating with temperature, nitrate, and phosphorus concentrations and the abundance of co-occurring predatory and competitive eukaryotic microorganisms, suggesting the importance of both abiotic and biotic controls in regulating expression. This work highlights the need for understanding the chemical ecology and potential risks to human and environmental health posed by secondary metabolites that are produced but often unmonitored. It also indicates the prospects for identifying pharmaceutical-like molecules from cyanoHAB-derived BGCs. IMPORTANCE Microcystis spp. dominate cyanobacterial harmful algal blooms (cyanoHABs) worldwide and pose significant threats to water quality through the production of secondary metabolites, many of which are toxic. While the toxicity and biochemistry of microcystins and several other compounds have been studied, the broader suite of secondary metabolites produced by Microcystis remains poorly understood, leaving gaps in our understanding of their impacts on human and ecosystem health. We used community DNA and RNA sequences to track the diversity of genes encoding synthesis of secondary metabolites in natural Microcystis populations and assess patterns of transcription in western Lake Erie cyanoHABs. Our results reveal the presence of both known gene clusters that encode toxic secondary metabolites as well as novel ones that may encode cryptic compounds. This research highlights the need for targeted studies of the secondary metabolite diversity in western Lake Erie, a vital freshwater source to the United States and Canada.

Keywords: Lake Erie; Microcystis; cyanoHABs; metagenomics; metatranscriptomics; secondary metabolites; strain diversity.

FIG 1

Overview of the 2014 western Lake Erie cyanoHAB. (A) Map of western Lake Erie and the sampling sites used for this study. WE2 and WE12 are close to the Ohio coast and Maumee River and are considered nearshore stations, while WE4 is located more centrally in the basin and considered offshore. (B) Various bloom metrics for the 2014 bloom, including cyanobacterial biomass as measured by both chlorophyll α (blue) and phycocyanin (green) concentrations, as well as measured concentrations of particulate microcystins, soluble reactive phosphorus, and nitrate. Panel A was generated via QGIS using the Open Street Map (OSM) (https://wiki.osmfoundation.org/wiki/Main_Page).

FIG 2

Relative abundance of biosynthetic gene clusters (BGCs), including the complete mcy operon described previously (33). Relative abundance is presented as estimated relative proportion of the Microcystis population containing the BGC (see Materials and Methods). Predicted secondary metabolites are indicated on the right vertical axis) and in the legend. On 4 August 2014 at WE2, the mcy operon and T3PKS gene clusters had a relative proportion greater than 1, indicating more copies of BGC genes than 16S rRNA genes.

FIG 3

Comparison of PKS-M BGCs identified in WLE MAGs to the known microginin-encoding BGCs from Microcystis aeruginosa LEGE 91341. (A) Both PKS-M gene clusters are incomplete/fragmented compared to the complete cluster, where PKS-M contains conserved additional synthesis and hypothetical protein-encoding genes, while PKS-M 2 lacks these genes and contains a putative nuclease gene. Both PKS-M clusters contain a fragmented hybrid NRPS-PKS-encoding gene found in the microginin-encoding pathway. (B) Read mapping from a sample collected on 4 August 2014 at WE2 against the complete microginin biosynthesis pathway in LEGE 91341. Coverage is observed throughout with notable variation in sequence identity and evenness.

FIG 4

Gene schematics for select cryptically annotated PKS-containing gene clusters. (A) The iterative PKS gene cluster contains transport genes, a putative enediyne biosynthesis gene, a thioesterase gene, 2/3 genes of unknown function that are part of the enediyne biosynthesis cassette, and putative nuclease genes. (B) The T3PKS cluster contains several transport genes and a cytochrome b₅ binding gene. (C) The MIC 1 cluster contains several core biosynthesis and transport genes and a tryptophan halogenase gene.

FIG 5

Phylogenetic tree consisting of protein sequences from the putatively identified PKSE from the WLE MAGs, putatively identified PKSEs from previous studies (45, 73), and PKSEs from known biosynthesis pathways that generate enediyne-containing molecules. The PKSE from the WLE MAGs is colored red, the putative PKSEs identified in other cyanobacteria are colored green, and a black circle indicates a known synthesized product associated with the PKSE enzyme.

FIG 6

Relative abundance of transcripts from BGCs throughout the 2014 bloom separated by sampling station. Relative transcript abundance was calculated by determining the number of reads mapped to a BGC via specific cutoff parameters, normalized to the number of reads mapped to an entire reference Microcystis genome to determine “expression effort” (see Materials and Methods).