Cylindrospermopsin, anatoxin and microcystin cyanobacterial toxins in Singapore reservoirs and historical Irish freshwater; identifying progenitor species, their responses to environmental factors and spatial and temporal variations

Date of Award


Document Type

Doctoral Thesis

Degree Name

Doctor of Philosophy


Physical Sciences

First Advisor

Dr Ambrose Furey

Second Advisor

Dr Mary Lehane


Toxins from cyanobacteria that contaminate raw drinking water catchments are a serious risk to human health. In addition to risk associated with treated drinking water from contaminated reservoirs, there is also a more direct threat from raw water to humans who actively use reservoirs and waterways. The main objective of this Ph.D. is to investigate cyanobacterial toxins and their algal progenitors (producers) in Singapore’s reservoirs and to develop a universal cyanobacterial toxin extraction and detection methods that are fast, robust and efficient. Chapter one in this thesis reviewed cyanobacteria origin, habitat, algal blooms, toxins and producer cyanobacteria genera, toxic effect, target organs, and toxins mechanism of action. The second part of chapter one described in details the latest analytical techniques technologies in liquid chromatography and mass spectrometry applications which includes: sample preparation, sample separation, spectroscopic analysis, and hyphenated techniques. Anatoxin‐a (ANA‐a) is a toxin well known to induce neurotoxicity. There have been several synthetic pathways to obtain ANA‐a, and these are briefly described in chapter two. It is possible to separate into categories the different total synthesis approaches to ANA‐a, according to the methodology used: (a) Synthesis of ANA‐a via ring expansion of tropanes (b) Synthesis of ANA‐a via transannular cyclisation of cyclooctenes (c) Synthesis of ANA‐a via iminium salts (d) Synthesis of ANA‐a via an enol (e) Synthesis of ANA‐a via nitrones (f) Synthesis of ANA‐a via allene (g) Synthesis of ANA‐a via tandem reactions and (h) Synthesis of ANA‐a via enine metathesis. Liquid chromatography (LC)‐UV methods were developed to determine ANA‐a, but this approach was inhibited by the presence of interferences in samples. Anatoxin and its analogues have been the subject of scientific research in different areas. Cyanobacteria toxin groups have caused very serious illnesses in both humans and animals. In 1996, a major toxicological incident happened when hospitalised patients accidently received water that was insufficiently purified, following a cyanobacterial bloom in a local reservoir. One hundred and sixteen patients out of 131 patients receiving hemodialysis treatment at a dialysis centre in Caruaru, Brazil, experienced visual disturbances, nausea, and vomiting. One hundred of those patients suffered from acute liver failure, which resulted in the death of 76 patients. The cause of death was microcystin and cylindrospermopsin toxins that were insufficiently adsorbed by activated carbon filters post water purification and the dissipation of a cyanobacterial bloom. Species known to produce these toxins have been identified in Singapore’s reservoir waters, for risk assessment purposes, it is critical to determine toxins profiles and levels in those reservoirs and to establish the causative algae which will allow for appropriate treatment of water and ensure protection of human and environmental health. Chapter three and four in this thesis assessed the toxin concentrations of various reservoirs and waterbodies in Singapore. A multi- toxin LC-MS/MS method was developed and validated for the analysis of cyanobacteria toxins in water samples, it was carried out on an Agilent HP 1100 series LC system hyphenated with an API3000 triple stage quadrupole mass spectrometer, water samples from 17 reservoirs were monitored monthly over a 12-month period (November 2012–October 2013). This study has undertaken a comprehensive examination of cyanobacteria and cyanotoxin levels and production in Singapore’s reservoirs, the dominant toxins found were microcystins, anatoxin-a and cylindrospermopsin. These toxins were present at many reservoirs and sometimes at concentrations approaching the drinking water guidelines, the microcystins dominant variants were microcystin LR and RR, microcystins were highest with levels above 0.5 μg L-1. Anatoxin-a was also highest at a few different reservoirs with a level above 0.05 μg L-1. Cylindrospermopsin was highest with levels above 0.5 μg L-1. These results indicate that some reservoirs have levels of toxins that approach the drinking water guidelines at various times, some reservoirs would exceeded the WHO drinking water guidelines if total microcystins concentration is added up. On some occasions the total of different toxins (ie. Microcystins + anatoxin-a + cylindrospermopsin) exceeded 1 μg L-1. The effect of multiple toxins acting on a human or organism is relatively unknown but they may have a synergistic effect creating greater risk than the individual concentrations would suggest. Cyanobacterial species were isolated and cultured to determine which species produced particular toxins, some of these cultures were used in experiments to determine their growth responses to environmental conditions (i.e. temperature, light, nutrients) and how this effected toxin production. Three experimental studies, chapter five, examined the growth characteristics and toxin production under different conditions of temperature, light and nutrients for the main Microcystis spp. Isolates. Increasing temperatures to 33°C (postulated mean temperature with climate warming) did not significantly affect the maximum growth rates of most Microcystis strains. Significantly higher growth rates were only observed in one M. ichthyoblabe strain at 33°C and M. flos-aquae at 30°C compared to that at lower temperatures and both were isolated from the same reservoir. MC-RR and MC-LR were produced in varying amounts by all four species of Microcystis. Raised temperatures of 33°C were found to boost total MC cell quotas for three Microcystis strains although a further increase to 36°C (postulated maximum temperature with climate warming) led to a sharp decrease in total MC cell quotas for all five Microcystis strains. Increasing temperature also led to higher MC-LR: MC-RR cell quota ratios in M. ichthyoblabe. Increasing nitrogen and phosphorus concentrations had a significant positive effect on maximum cell yields of all three Microcystis strains, phosphorus played an important role in MC production with increases in phosphorus from medium to high concentrations leading to decreases in MC-RR cell quotas for all three strains at the two highest nitrogen concentrations tested. Low light intensity resulted in higher growth rates compared to that at high light intensity for M. ichthyoblabe while the growth rates of other Microcystis species (M. aeruginosa, M. viridis and M. flos-aquae) were not significantly affected by light intensity. High light intensity led to higher total MC cell quotas of M. ichthyoblabe (TG) and M. flos-aquae compared to that at low light intensity, although the total MC cell quotas of the other three Microcystis strains (M. ichthyoblabe (LP), M. aeruginosa, M. viridis) were similar at all light intensities. The combined effects of light intensity and nutrients on the growth and toxin production of M. ichthyoblabe were tested using in-situ microcosms to better validate the lab experiments. High concentrations of nitrogen and phosphorus led to significantly higher growth compared to that at low concentrations of nitrogen and phosphorus while low phosphorus concentration led to high MC cell quotas, which was similar to the findings of the laboratory study. Microcystis in both laboratory and in-situ experiments had similar growth rates, but toxin production was higher in the in-situ study. Thus, growth results in the laboratory experiments may be comparable to those under natural light and temperature conditions while toxin production results under laboratory conditions may be less similar to toxin production under natural light and temperature conditions. A second LC-MS/MS method was developed and validated for the analysis of cyanobacteria hepatotoxins in the experimental samples, it was carried out on an Agilent HP 1100 series LC system hyphenated with an API3000 triple stage quadrupole mass spectrometer. Some cyanobacteria were isolated and cultured to determine the progenitor species (cyanophytes responsible for production of the toxins). Microcystis species and strains were isolated and a high proportion of these strains were found to produce MC-RR and MC-LR. Other possible toxin-producing cyanobacteria were also isolated including Cylindrospermopsis raciborskii and Planktothricoides raciborskii from two different reservoirs which both produced cylindrospermopsin. The developed LC-MS/MS method was applied for the analysis of cyanobacteria toxins in culture samples from Singapore and also for historic Irish water, chapter six. Those historical samples were stored in -20 °C freezer, to test the robustness and accuracy of the developed triple quadrupole method in this thesis, a screen analysis for target toxins were done for a number of chosen lakes samples to distinguish the variants and concentration of toxins. LC-MS/MS methods for individual cyanobacterial toxin groups are developed when a laboratory has a monitoring programme focused on target analytes from a single toxin class. These toxin specific LC-MS/MS methods can be expanded to include the identification of new analogues that share familiar structural characteristics. Analytical methods validation were carried out in accordance with the requirements of EC and ICH guidelines covering topics; specificity, selectivity, linearity, limits of detection and limits of quantitation, trueness and precision Based on the results of toxin production rates of Microcytis spp. from this study a set of draft guidelines in the unit algal cells ml-1 have been developed for risks associated with drinking water and primary recreational use of water. In summary the most toxic Microcytis species would exceed drinking water guidelines at 7,337 cells ml-1 based on mean cell toxin quota. This study was a proactive approach to the protection of Singapore’s population and its freshwater environment. The study has found a range of toxin producing cyanobacteria to occur in Singapore’s waters however in general the toxin levels are generally at relatively low concentrations in reservoirs. The last chapter in this thesis is chapter 7, bridging the gap between PhD research and R & D in the work place. This chapter demonstrated the knowledge gained and skills which I developed and optimised in relation to: LC-MS/MS, LC-ITMS, analytical chemistry samples extraction, chromatographic separation, method development, method validation and scientific writing. This chapter will outline and highlight the eight methods I have developed and validated since completing my PhD laboratory studies.


Digital copy of PhD cannot be shared due to copyright.

Access Level


This document is currently not available here.