Rapid detection methods and modelling simulations provide new insights into cyanobacteria detection and bloom management in a tropical reservoir
Han Li,Shu Harn Te,Yasaman Tavakoli,Jingjie Zhang,Karina Yew-Hoong Gin,Yiliang He
DOI: https://doi.org/10.1016/j.jenvman.2022.116730
2023-01-15
Abstract:The increasing occurrence of cyanobacteria blooms is of global concern, and is often associated with environmental and socio-economic problems, such as degenerated ecosystems and aquaculture impairment. The diazotrophic cyanobacterium Raphidiopsis raciborskii (R. raciborskii) grows rapidly in the tropics, and produces the toxin, cylindrospermopsin (CYN), which has harmful effects on aquatic organisms. Thus, to protect water quality and ecosystem, it is essential to have rapid and reliable methods for cyanobacteria and R. raciborskii detection and prediction so that early warning can be provided for management. Molecular assays, such as PCR, real-time quantitative PCR (qPCR), two-step PCR assays are accurate and widely used, but still require several hours from sample preparation to data analysis. In this study, insulated isothermal PCR (iiPCR) assays in conjunction with fast DNA extraction method, were developed and verified as a rapid detection assay in detecting cyanobacteria and R. raciborskii within 50 min, and also with high detection accuracy (98.8%) and the overall high agreement level (98.8%, k = 97.5%)) comparing to conventional qPCR assay. However, the limitation of the iiPCR assay is that it only generates qualitative results. Therefore, the quantified iiPCR assay, named as A-iiPCR, by coupling iiPCR device with fluorescence signal catching and interpretation instrument (Andor spectrometer with Solis spectroscopy software) was developed and verified with in situ environmental samples. The fluorescence intensity decreased accordingly with the drop of DNA concentration until reaching 1.32 ng/μL. Also, Delft 3D modelling was established to simulate R. raciborskii change in predicting spatial and temporal variabilities for reservoir management, as the simulated R. raciborskii concentration was the highest at sampling site 1, as well as temporally highest in April and October, posing as the most high-risk location and time periods for R. raciborskii bloom-forming requiring corresponding governance measures.