Search

When cyanobacterial phytoplankton form harmful cyanobacterial blooms (HCBs), the toxins they produce threaten freshwater ecosystems. Hydrogen peroxide is often used to control HCBs, but it is broadly toxic and dangerous to handle. Previously, we demonstrated that glucose addition to lake water could suppress the abundance of cyanobacteria. In this study, glucose was compared to hydrogen peroxide for the treatment of cyanobacterial phytoplankton communities. The six-week study was conducted in the large mesocosms facility at Grand Valley State University’s Annis Water Resources Institute in Michigan. To 1000 L of Muskegon Lake water, glucose was added at either 150 mg or 30 mg glucose/L. Hydrogen peroxide was added at 3 mg/L to two 1000 L mesocosms. And two mesocosms were left untreated as controls. Triplicate 100 mL samples were collected weekly from each mesocosm, which were then filtered and frozen at −80 °C for 16S rRNA amplicon sequencing. The 16S rRNA amplicon sequencing results revealed that hydrogen peroxide treatment quickly reduced the relative abundance of the cyanobacteria compared to the control mesocosms, but the cyanobacteria population returned over the course of the 6-week study. On the other hand, both glucose concentrations caused a rapid proliferation of multiple low abundance proteobacterial and bacteroidotal taxa resulting in notable increases in taxonomic richness over the duration of the study and reducing the relative abundance of cyanobacteria. Although hydrogen peroxide quickly suppressed the cyanobacteria, the population later returned to near starting levels. The glucose suppressed the cyanobacterial phytoplankton apparently by promoting competitive heterotrophic bacteria.

Abstract Cyanobacteria and cyanotoxin producing cyanobacterial blooms are a trending focus of current research. Many studies focus on bloom events in lentic environments such as lakes or ponds. Comparatively few studies have explored lotic environments and fewer still have examined the cyanobacterial communities and potential cyanotoxin producers during ambient, non-bloom conditions. Here we used a metagenomics-based approach to profile non-bloom microbial communities and cyanobacteria in 12 major U.S. rivers at multiple time points during the summer months of 2019. Our data show that U.S. rivers possess microbial communities that are taxonomically rich, yet largely consistent across geographic location and time. Within these communities, cyanobacteria often comprise significant portions and frequently include multiple species with known cyanotoxin producing strains. We further characterized these potential cyanotoxin producing taxa by deep sequencing amplicons of the microcystin E (mcyE) gene. We found that rivers containing the highest levels of potential cyanotoxin producing cyanobacteria consistently possess taxa with the genetic potential for cyanotoxin production and that, among these taxa, the predominant genus of origin for the mcyE gene is Microcystis. Combined, these data provide a unique perspective on cyanobacteria and potential cyanotoxin producing taxa that exist in large rivers across the U.S. and can be used to better understand the ambient conditions that may precede bloom events in lotic freshwater ecosystems.

Mitigating cyanotoxin production is essential to protecting aquatic ecosystems and public health. However, current harmful cyanobacterial bloom (HCB) control strategies have significant shortcomings. Because predicting HCBs is difficult, current HCB control strategies are employed when heavy HCBs have already occurred. Our pilot study developed an effective HCB prediction approach that is employed before exponential cyanobacterial growth and massive cyanotoxin production can occur. We used a quantitative polymerase chain reaction (qPCR) assay targeting the toxin-encoding gene mcyA to signal the timing of treatment. When control measures were applied at an early growth stage or one week before the exponential growth of Microcystis aeruginosa (predicted by qPCR signals), both hydrogen peroxide (H2O2) and the adsorbent hydroxyapatite (HAP) effectively stopped M. aeruginosa growth and microcystin (MC) production. Treatment with either H2O2 (10 mg·L−1) or HAP (40 µm particles at 2.5 g·L−1) significantly reduced both mcyA gene copies and MC levels compared with the control in a dose-dependent manner. While both treatments reduced MC levels similarly, HAP showed a greater ability to reduce mcyA gene abundance. Under laboratory culture conditions, H2O2 and HAP also prevented MC production when applied at the early stages of the bloom when mcyA gene abundance was below 105 copies·mL−1.

To mitigate harmful cyanobacterial blooms (HCBs), toxic algicides have been used, but alternative methods of HCB prevention are needed. Our goal was to test the prophylactic addition of glucose to inhibit HCB development, using Microcystis and the toxin microcystin as the HCB model. Water samples were collected weekly, from 4 June to 2 July, from Harsha Lake in southwestern Ohio during the 2021 algal bloom season. From each weekly sample, a 25 mL aliquot was frozen for a 16S rRNA gene sequencing analysis. Then, 200 mL of Harsha Lake water was added to each of the three culture flasks, and glucose was added to create concentrations of 0 mM (control), 1.39 mM, or 13.9 mM glucose, respectively. The microcystin concentration in each flask was measured after 1 and 2 weeks of incubation. The results showed an 80 to 90% reduction in microcystin concentrations in glucose-treated water compared to the control. At the end of the second week of incubation, a 25 mL sample was also obtained from each of the culture flasks for molecular analysis, including a 16S rRNA gene sequencing and qPCR-based quantification of Microcystis target genes. Based on these analyses, the glucose-treated water contained significantly lower Microcystis and microcystin producing gene (mcy) copy numbers than the control. The 16S rRNA sequencing analysis also revealed that Cyanobacteria and Proteobacteria were initially the most abundant bacterial phyla in the Harsha Lake water, but as the summer progressed, Cyanobacteria became the dominant phyla. However, in the glucose-treated water, the Cyanobacteria decreased and the Proteobacteria increased in weekly abundance compared to the control. This glucose-induced proteobacterial increase in abundance was driven primarily by increases in two distinct families of Proteobacteria: Devosiaceae and Rhizobiaceae. In conclusion, the prophylactic addition of glucose to Harsha Lake water samples reduced Cyanobacteria’s relative abundance, Microcystis numbers and microcystin concentrations and increased the relative abundance of Proteobacteria compared to the control.

Streambank legacy sediments can contribute substantial amounts of sediments to Mid-Atlantic waterways. However, there is uncertainty about the sediment-bound P inputs and the fate of legacy sediment P in surface waters. We compared legacy sediment P concentrations against other streambank sediments and upland soils and evaluated a variety of P indices to determine if legacy sediments are a source or sink of P to surface waters. Legacy sediments were collected from 15 streambanks in the mid-Atlantic USA. Total P and M3P concentrations and % degree of phosphorus saturation (DPS) values for legacy sediments were lower than those for upland soils. % DPS values for legacy sediments were below the water quality threshold for P leaching. Phosphorus sorption index (PSI) values for legacy sediments indicated a large capacity for P sorption. On the other hand, equilibrium phosphorus concentration (EPC0) for legacy sediments suggested that they could be a source or a sink depending on stream water P concentrations. Anoxic conditions resulted in a greater release of P from legacy sediments compared to oxic conditions. These results suggest that legacy sediment P behavior could be highly variable and watershed models will need to account for this variability to reliably quantify the source-sink behavior of legacy sediments in surface waters.