Determining the presence of cyanotoxins in water reservoirs of Anuradhapura, using molecular and bioassay methods

This study was performed to detect the presence of cyanotoxin producing organisms in the reservoirs of the Anuradhapura Distrct. Water samples were collected from Kala wewa, Tissa wewa and Nuwara wewa in Anuradhapura. Environmental and cultured samples were subjected to PCR to detect cyanobacteria, Microcystis, microcystin, Cylindrospermopsis and cylindrospermopsin, targeting the specific 16S rRNA and cpc, mcyA and mcyE, cylindrospermopsin synthetase (rpoC1) and cylindrospermopsin specific peptide synthase (PS) genes, respectively. All tested samples were positive for unicellular and five were positive for filamentous cyanobacteria. In all three reservoirs, positive amplification was observed for mcyA/ mcyE genes indicating the presence of cyanobacterial strains with genetic potential to produce microcystins. A positive amplification resulted in samples cultured from Nuwara wewa for rpoC1 gene confirming the presence of Cylindrospermopsis species in the reservoir. Further more, an environmental sample from Nuwara wewa was positive for PS gene confirming the presence of toxic Cylindrospermopsis species, and thereby cylindrospermopsin in that reservoir. Bioassay was performed using Artemia salina to test the toxicity of water samples. Filtrates and filters were assayed and in all three tanks, higher toxicities were recorded for filters of both environmental and cultured samples. Among filters, the highest toxicity was recorded in two environmental samples from Nuwara wewa with a mean LD 50 value of -694 ppm. Among cultured samples, the highest toxicity of -1128.28 ppm was recorded from Kala wewa. According to molecular and bioassay findings, all three lakes have a wide distribution of cyanobacterial species with toxin generating ability and it might be a potential risk factor for the health issues that prevail in the Anuradhapura District.


INTRODUCTION
Most chemical contaminants in drinking water become health concerns only after exposure for long periods. A number of chemical contaminants in drinking water, have been shown to cause adverse health effects in humans after prolonged exposure.
Health impairments are shown in numerous anecdotal reports of irritations of the skin and/or mucous membranes and in documented cases of illness after exposure through drinking water as well as accidental swallowing or aspiration of scum material (Chorus et al., 2000).
Among the cyanotoxins, microcystins are the predominant toxins in freshwater lakes worldwide, which have been implicated in several cases of animal and human intoxications (Kuiper-Goodman et al., 1999). Progressive active liver injury (Falconer et al., 1983) and the potential for promotion of tumour growth are two aspects of microcystin damage. A number of human deaths have been reported resulting from exposure to the cyanobacterial toxin, microcystin and both cylindrospermopsin and microcystin have been detected in drinking water (Byth, 1980;Falconer et al., 1983;Hawkins et al., 1985;Falconer, 2005). Cylindrospermopsin is a general cytotoxin that blocks protein synthesis (Falconer, 1996), and the first clinical symptoms of cylindrospermopsin are kidney and liver failure. In contrast to the pure toxin, crude extracts of the organism also cause injury to the lungs, adrenals and intestine. This indicates that other toxins may also be present in the organism. Since clinical symptoms may become manifest only several days after exposure, it will often be difficult to determine a cause-effect relationship. Patients intoxicated with cylindrospermopsin via drinking water in an incident in Australia, escaped death only through skilled and intensive hospital care (Falconer, 2005).
This research study was aimed at identifying the cyanotoxins present in certain water bodies of the Anuradhapura District.

METHODS AND MATERIALS
Site selection and sample collection: Kala wewa, Tissa wewa and Nuwara wewa in the District of Anuradhapura were selected for water sample collection in June 2010 (Table 1). All samples were collected in sterile brown glass containers. The samples were collected from the water surface and the water column (~ 0.1 -15m), to represent the whole water body (three replicates were collected from each site). The temperature and pH were recorded using a thermometer and a pH meter, respectively. Morphological observations were also made of each sample, using a compound light microscope (Olympus BH-2) (400-1,000×) and standard morphological study methods.
All the amplifications were carried out in Techne TC 3000 DNA Thermal Cycler. A single cycle of initial denaturation for 5 min at 94 0 C followed by 35 incubation cycles each consisting of 1 min at 94 0 C, 1 min at 60 0 C, 56 0 C and 59 0 C for 16S rRNA gene, cpc gene, mcyA gene and mcyE gene, respectively and 1min at 72 0 C followed by 15 min extension at 72 0 C. A touchdown PCR was carried out with a single cycle of initial denaturation for 5 min at 94 0 C followed by 5 incubation cycles each consisting of 1 min at 94 0 C, 1 min at 62 0 C and 1 min at 72 0 C and another 5 cycles at 60 0 C and 30 cycles at 58 0 C followed by 15 min extension at 72 0 C for the amplification of cylindrospermopsin specific peptide synthase (PS) gene. Further, another touchdown PCR was carried out as above, with 5 cycles at 61 0 C, 5 cycles at 58 0 C and 30 cycles at 56 0 C for the amplification of cylindrospermopsin synthetase gene. Aliquots of the resulted amplified products were electrophoresed in 1.5 % agarose gels containing 10 μg/mL ethidium bromide and documented through a gel documentation system (Syngene, UK).

Bioassay detection:
a) Extraction of toxins: In order to detect the dissolved, intracellular and total cyanotoxins in water, environmental and cultured samples ( Table 2) were filtered using Whatman's membrane filter papers. Each membrane with the residue was processed according to the protocol described by Fastner et al. (1998), vacuum dried at 45 0 C and the toxin was extracted into 100 % methanol. Each filter and filtrate was then subjected to the bioassay. % of final mortality = % of mortality -% of mortality in control

Site selection and sample collection
This study was performed to determine the presence of cyanotoxins in three main reservoirs (Kala wewa, Tissa wewa and Nuwara wewa) in the Anuradhapura District. These reservoirs are often used for drinking, recreational purposes, bathing, fishing and various other water related activities. Furthermore, they are the main water sources for irrigation. According to the data, mean temperatures of 27 0 C, 28 0 C and 27 0 C and mean pH of 8.5, 8.34 and 8.7 were observed in Nuwara wewa, Tissa wewa and Kala wewa, respectively (Table 1). Under microscope, Cylindrospermopsis species were recorded as the dominant cyanobacterial species in the environmental water samples collected from all three reservoirs along with M. aeruginosa, Anabaena spp., Chroococcus spp., Phormidium spp., Microcystis spp., Oscillatoria spp., Limnothrix spp., Lyngbia spp., Calothrix spp., Anabaenopsis spp., and Merismopedia spp., which were comparatively moderate to low in numbers. However, in culture samples in BG11, BG11 0 and BG11 0 C, no Cylindrospermopsis species were observed except Phormidium spp., Microcystis spp., Limnothrix spp., Lyngbia spp., and Merismopedia spp., Moreover, in the cultures in MLA medium, which lacks the inorganic nitrogen, Cylindrospermopsis was observed only in Nuwara wewa sample and also were biased to other filamentous forms with a minority of Merismopedia. Although Cylindrospermopsis is a filamentous heterocyst former, the reason for this contrast can be due to the culture media used for culturing (BG11, BG11 0 and BG11 0 C), which did not facilitate the growth of Cylindrospermopsis species. Further, it is well known that most of the microorganisms are non cultivable (Dadheech et al., 2009).

Molecular detection
The method employed for genomic DNA extraction resulted in high quality DNA in satisfactory amounts for amplification. According to the study done by Boutte et al. (2005), the primers, which are cyanobacterial specific CYA781 Ra and CYA781 Rb were used separately to identify heterocyst forming filamentous and unicellular non heterocyst forming filamentous cyanobacteria, respectively and equimolar mixture of CYA781Ra + CYA781Rb to identify the presence of cyanobacterial community in those reservoirs. According to the gel profiles obtained, all DNA samples  Ten individuals of A. salina nauplii were placed in each well of the polystyrene plate and the extracted toxins were added to each well with artificial seawater medium to get a concentration series in duplicates. A control was also performed with artificial seawater. The plates were incubated at room temperature for 24 h (Meyer et al., 1982). After incubation, dead larvae in each well (or those swimming in a fixed position) were counted using a hand lens and the percentage mortality was calculated using the mean of two replicates. Results were expressed as percentage of mortality using the following formula and LD 50 value was analysed with StatPlus 2009 Probit Analysis software.
% of mortality = Number of dead shrimps in well after 24 h × 100 Number of shrimps initially placed in well submitted to PCR reactions with cyanobacterial specific oligonucleotide primers of CYA359F forward and CYA781Rb reverse primer yielded the unique fragment of ~450 bp (Table 3 and Figure 1a), indicating the presence of unicellular cyanobacteria in those reservoirs. However, the use of the forward primer CYA359F and the reverse primer CYA781Ra yielded ~450 bp fragment only in AN 8 from the environmental sample of the Nuwara wewa (Table 3) and AKI-C, AN7-C, AT5-C, and AT6-C samples, which were cultured in MLA medium (Table 3 and Figure 1b).These observations along with microscopic observations confirmed the presence of filamentous forms, which could be either heterocyst formers or non-heterocyst formers. Furthermore, all the samples yielded the ~450 bp unique fragment with CYA 359F and the equimolar mixture of CYA781Ra + CYA 781Rb with different intensities (Table 3 and Figure 1b) signifying the presence of unicellular and filamentous cyanobacteria in those water bodies. The reason for these different intensities might be due to the competition of reverse primers to the same template, resulting in less complete genetic fingerprints (Boutte et al., 2005).
In addition, positive amplification of 500 bp in all the samples with cpc_arF and cpc_arR primers except AK3-C and AT5-C representing the presence of phycocyanin gene confirmed the presence of cyanobacteria (Table 3 and Figure 2).
AN8 and AT5 environmental samples gave positive amplification for the mcyA gene, yielding the unique    Table 3). This indicates the presence of cyanobacterial strains that have the genetic potential to produce microcystins. Further, all the samples gave positive amplification for the mcyE gene, yielding the unique ~250 bp except AK1 environmental sample and AK3-C, which was cultured in MLA medium (Table 3 and Figure 3). This indicates the presence of Microcystis strains in all three reservoirs that have the genetic potential to produce microcystins.
Only AN7-C DNA sample, which was cultured in MLA medium gave positive amplification with rpoC1 -1 forward and rpoC1 -T reverse primer, confirming the presence of cylindrospermopsin synthetase gene and therefore, the presence of Cylindrospermopsis species. Although the expected size was 609 bp for the Cylindrospermopsis raciborskii, the only amplified product size was ~250 bp (Table 3 and Figure 4). The microscopic identification of the samples indicated the presence of Cylindrospermopsis species, and therefore, the resulting PCR product might be due to the presence of Cylindrospermopsis species other than Cylindrospermopsis raciborskii. A specific ~597 bp amplification was observed in AN8 environmental DNA sample with M13 forward and M14 reverse primers, while a ~300 bp unspecific amplification was observed in AK1-C, AN7-C and AT6-C DNA samples, which were cultured in MLA medium (Table 3 and Figure 5), indicating the presence of cylindrospermopsin specific peptide synthase (PS) gene and therefore, confirmed the presence of toxin producing Cylindrospermopsis species and also cylindrospermopsin. However, unspecific amplifications might be due to the other strains of Cylindrospermopsis, which also contain the peptide synthase (PS) gene.

Bioassay detection
The bioassay detection is based on the lethal effect of cyanotoxins on the brine shrimp, A. salina. This bioassay provides a viable tool since it is a representative species of the zooplankton community. According to the results obtained for bioassay, higher toxicity for A. salina after 24 hours was recorded in the filters compared to the filtrates (Table 4 and 5). Moreover, 50 % of environmental samples showed higher toxicity compared to the cultured samples.

bp 250 bp
Sample No. Sample code Total mortality % of final after 24h mortality  (Figures 6 and 7), except AN6 and AN7 environmental samples and AK1 and AT5, which were cultured in BG11 medium (Table 5). However, for the 2 samples AT5 and AK1 cultured in BG11 medium, the mortality was independent from the concentration (Table 5). Among cultured samples in BG11 and BG11 0 C media, the highest toxicity was recorded in the Nuwara wewa samples.
Beside chemical analyses of cyanobacterial toxins, a wide range of biological laboratory methods have been developed for detection and identification of the cyanotoxins in cyanobacterial biomass and raw waters. Since analytical analysis cannot describe the real toxic effect of complex mixtures of chemicals, relatively simple and low cost methods provide necessary additional information on overall toxicity, and allow rapid evaluation of the potential hazard for humans, animals or ecosystems. A specific biological screening method can be selected, depending on the type of information required. However, Among the filtrates, the highest mortality of 66 % was recorded in AK1 filtrate, which was cultured in MLA medium. A considerable amount of mortality (60 %) was also recorded in AK1 sample cultured in BG11 0 C (Table 4). Further, in filter samples, a sigmoid dose   it is important to note that there cannot be one single method providing adequate information on toxicity of all types of cyanotoxins. Therefore, analysis of chemical composition in the algal extract is important to verify the toxicity. However, the most widely used invertebrate bioassay available for cyanotoxins is the brine shrimp (Artemia salina) assay (Kiviranta et al., 1991;Lawton et al., 1994). This is a useful toxicity screening method used in very basic laboratories as it requires little expertise and does not require culture maintenance or special equipment.
This research study focused on cyanotoxins, because most males (from childhood to adulthood) in the Dry Zone, often use lake water for recreational purposes, bathing and washing after farming, fishing and various water related activities. Recreational use of lakes, rivers and canals is the most common route of exposure to cyanobacterial toxins. As an example, in 1959 in Canada, 12 people became ill with headaches, nausea, muscular pains and acute gastroenteritis symptoms (Dillenberg & Dehnel, 1959). The illness was noted in people who had been swimming in a lake contaminated by toxins from Anabaena, before the intoxication. In England in 1989, 10 of 20 soldiers became ill after swimming and canoetraining in water with a heavy bloom of Microcystis (Turner et al., 1990) and two cases of pneumonia were noted among the soldiers. According to Mehra et al. (2009), an additional aspect influencing the toxicity of cyanobacterial blooms is the age of the victim. Children are more vulnerable for several reasons: they drink more water per unit of body weight, are less likely to have a choice of the source of drinking water and are more susceptible to damage that takes a considerable time to develop, such as environmentally induced carcinomas. It is a common sight in the Dry Zone tanks, where lots of children, mainly boys spend hours in the water and therefore, the direct ingestion of cyanotoxic water by the swimmers for a long period, might be the cause for the mysterious chronic kidney disease, cancers and other human illnesses recorded in the Dry Zone.
Healthy people are less susceptible to cyanotoxicosis than people with hepatitis, alcoholism or kidney damage. Patients who undergo kidney dialysis are especially susceptible to toxic damage from cyanobacteria. In 1996 in Caruaru, Brazil, 131 dialysis patients were exposed to microcystins via water used for dialysis; 56 of them died and the others had typical symptoms of harm from microcystins such as nausea, vomiting and painful, excessive enlargement of the liver (Jochimsen et al., 1998). The episode of deaths of haemodialysis patients at the Caruaru dialysis clinic emphasized the importance of haemodialysis water as an exposure route for microcystins (Codd, 2000).
According to the investigations in all three reservoirs, Cylindrospermopsis species were recorded as the dominant cyanobacterial species along with a vast diversity of cyanobacterial species in moderate numbers. Except Chroococcus, the rest of the cyanobacterial species which have been recorded, has the genetic potential to generate toxins. Microcystins, which is a most common class of toxins worldwide, is mainly produced by M. aeruginosa and other species of Microcystis, and these were moderately present in all three lakes. The mechanisms of toxicity of the diverse cyanobacterial toxins include nerve ion-channel blocking, neuromuscular blockade, anti-acetyl cholinesterase activity, antiphosphatase activity and protein synthesis inhibition. The liver is particularly damaged by the latter two activities, as well as the gastro intestinal tract and, in the case of protein synthesis inhibition, the kidneys. The anti-phosphatase activity of microcystins and nodularin has significant potential for stimulating tumour growth. This capacity is clearly demonstrated experimentally, and its mechanism is under detailed investigation. Cyanobacterial poisons can affect not only hepatocytes but also other types of cells. Microcystins and nodularins can inhibit the activity of protein phosphatases PP1 and PP2A in any eukaryotic cell they can get into, but usually need the bile acid carrier to cross cell membranes. Apoptotic changes have been observed after microinjection of microcystins or nodularins into human embryo kidney HEK 293, Swiss 3T3 mouse embryo fibroblast, breast carcinoma cell line MCF-7 and rat promyelocytic IPC-81 leukemia cells  (Fladmark et al., 1999). Morphological changes such as membrane budding, cell shrinkage and organelle redistribution have been seen in human fibroblasts, human endothelial cells, human epithelial cells, human lymphocytes and rat promyelocytes (Mcdermott et al., 1998;Mankiewicz et al., 2001). Microcystins also have a clastogenic effect in human lymphocytes, connected with a dose-related increase of chromosomal breakage (Repavich et al., 1990;Premazzi & Volterra, 1993). According to a previous study, the cytotoxic alkaloid cylindrospermopsin produced by Cylindrospermopsis species, were the dominant species in three lakes and it generally blocks the protein synthesis (Hawkins et al., 1985). The first clinical symptoms of poisoning are kidney and liver failure; it also causes damage to the spleen, intestine, heart and thymus. Cylindrospermopsin is especially dangerous because clinical symptoms may become manifest only several days after exposure, so the toxic effects can be difficult to correlate. Tumor promotion by exposure to cyanobacterial toxins poses an added risk, and in some areas of the Dry Zone the number of deaths due to cancer has increased sharply within the last few years.
According to the findings of this study, all three lakes show a wide distribution of cyanobacterial species with toxin generating ability and it might be a major risk factor for various human illnesses recorded in the Dry Zone. Therefore, more epidemiological studies are required to explore the relationship existing between human illnesses and cyanotoxins in water.
In conclusion, molecular detection data and bioassays have confirmed the presence of toxic cyanobacteria in all three sampled water reservoirs. Further, molecular data confirmed the presence of cyanobacterial strains that have the genetic potential to produce microcystins and more importantly, the presence of Cylindrospermopsis and also the presence of cylindrospermopsin producing Cylindrospermopsis species. The bioassay results confirmed the presence of lethal cyanotoxins in collected water samples. However, there is a need for extensive studies to identify the types of cyanobacteria, cyanotoxins, their variants, compositions and the level of toxicity present in those water reservoirs to correlate it with the present issue of human illnesses.