Chiang Mai J. Sci. 2014; 41(1) 27

Chiang Mai J. Sci. 2014; 41(1) : 27-38http://epg.science.cmu.ac.th/ejournal/Contributed Paper

Cyanobacterial Composition and MicrocystinAccumulation in Catfish PondSuneerat Ruangsomboon*[a], Wichien Yongmanitchai [b], Paveena Taveekijakarn [a]and Monthon Ganmanee [a][a] Program in Fisheries Science, Faculty of Agricultural Technology, King Mongkut’s Institute of Technology

Ladkrabang, Bangkok 10520, Thailand.[b] Department of Microbiology, Faculty of Science, Kasetsart University, Bangkok 10900, Thailand.*Author for correspondence; e-mail: krsuneer@kmitl.ac.th

Received: 7 December 2012Accepted: 22 April 2013

ABSTRACTCyanobacterial blooms frequently occur in catfish ponds, but the potential effects of

its secondary metabolites on cultured animals in ponds and human beings remain largelyunknown. In the present study, high chlorophyll-a levels (1.28-2.06 mg/L) indicated massivephytoplankton blooms in the studied ponds, with a maximum density of 52 × 106 cells/L ofpotentially toxic cyanobacteria, composed mainly of Microcystis aeruginosa, Oscillatoria spp.,Lyngbya sp., Pseudanabaena spp., Anabaena spp., Anabaenopsis sp., and Cylindrospermopsis sp.However, microcystins were not detected by liquid chromatography-mass spectrometry insamples of water and fish. Instead, microcystin-RR in phytoplankton was estimated to bebetween 0.014 and 0.202, and correlated positively with the density of Anabaena spp. (r=0.33,p<0.05), alkalinity (r=0.37, p<0.05), and ammonia (r=0.48, p<0.01). Microcystin was notdetected in the fish in this study, but microcystin accumulation was found in 43% of analyzedphytoplankton samples. Therefore, long-term monitoring is recommended in order to fullyunderstand microcystin accumulation patterns at all degrees of phytoplankton bloom.

Keywords: catfish, fish pond, microcystin, phytoplankton bloom, toxic cyanobacteria

1. INTRODUCTIONAquaculture is one of the fastest growing

animal food sectors around the world, andcultured fish supplies a portion of the animalprotein intake for the human population [1].In Thailand, carnivorous catfish species(Clarias macrocephalus vs. C. gariepinus) havebecome major commercial breeding varietiesand are mainly cultured in soil ponds for along period of time. Especially in the lastdecade (2000-2010), their farming has grown

intensively and rapidly, with a 237% overallincrease in fish production [2].

As a tropical region of Southeast Asiawith high temperatures, Thailand is favorablefor cyanobacteria growth, since rapidcyanobacteria growth has been shown to beassociated with such temperatures. Forexample, Wang et al. [3] reported that thebloom of toxic cyanobacteria in freshwaterbodies is associated with high levels of

28 Chiang Mai J. Sci. 2014; 41(1)

nutrients, light intensity, temperature, and pH.Also, the conditions of high fish-stockingdensities and feeding rates during catfishculture are suitable for the prolificdevelopment of cyanobacteria. It is knownthat cyanobacteria can produce a number ofsecondary metabolites and compounds.Among these, microcystin is considered tobe one of the most dangerous toxins, as it isknown to be a potent hepatotoxin that maycause liver damage in humans, fish, andother organisms, and may promote tumorgrowth [4].

A study has shown that microcystinproduction by cyanobacteria increasedproportionally to the amount of nutrients(e.g., total phosphorus) [3]. Microcystin isreleased from cyanobacterial cell lysis;hence, a high concentration of dissolvedmicrocystin may occur in water immediatelyafter the collapse of cyanobacterial blooms.Microcystin contents in fish may varyaccording to both their exposure time andbloom duration. Therefore, the degree ofrisk of human exposure to microcystin viaconsumption of fish may vitally depend onthe accumulation of microcystin in fish [5].Smith et al. [6] revealed that about 47%of the studies investigating microcystinaccumulation in edible tissues of aquaticanimals exceeded the Tolerable Daily Intakeof 0.04 mg/kg/bw/d, and suggested thathuman intoxication is possible through bothintensive and extensive aquaculture.

It is a general view that microcystinconcentration in carnivorous fish is lowerthan that of phytoplanktivorous fish [7]because the latter directly ingest cyanobacteria,whereas the former accumulate microcystinvia accidental ingestion, passive assimilationthrough gills [8], and transdermal adsorption[9]. Although there are extensive studies onmicrocystin bioaccumulation in fish, mostof these focused on phytoplanktivorous

fish, notably tilapia and carp [10-12]. Bycomparison, less information was producedfor carnivorous fish, such as catfish [8, 13].To date, there is still no published informationregarding both the occurrence of toxin-producing cyanobacteria and the quantities oftoxins found in catfish-culturing systems [13].Moreover, cyanobacterial toxins in pondwater and farmed animals are occasionallyinvestigated after a die-off event [8], but arerarely regularly monitored. Regular,frequent monitoring could provide accurateinformation about cyanobacterial toxins thatcould prevent fish mortality, as ambienttoxin concentrations can vary over a seasonand this accumulation can occur at sub-lethalconcentrations [14].

Thus, our concern is whether massivecyanobacterial blooms in catfish ponds mayproduce microcystins during the culturingprocess, and if so, how they may affect humanhealth. Thus, the goals of this study are four-fold: 1) to investigate species composition andabundance of cyanobacteria; 2) to measurethe concentration of microcystins (MC-LR,-RR, -YR) in three levels of phytoplankton,water, and fish; 3) to assess the potential riskto humans from microcystin contamination;and 4) to detect the key environmental factorsthat regulate the occurrence of microcystinin catfish ponds.

2. MATERIALS AND METHODS2.1 Site Description

There were 3 catfish ponds (Clariasmacrocephalus vs. C. gariepinus) used in thisstudy. All are 60 × 80 × 1.5 m in size. Theseponds are located in Klong Suan Sub-district,Bang Bo District, Samut Prakan Province,central Thailand. At this location, theaverage temperature, relative humidity, andrainfall were 22-42°C, 73% and 1743 mm/y,respectively. The locations of ponds were:pond 1, 13°41′52.08′′N, 100°50′53.34′′E;

Chiang Mai J. Sci. 2014; 41(1) 29

pond 2, 13°41′49.62′′N, 100°50′52.68′′E andpond 3, 13°41′47.04′′N, 100°50′52.08′′E.They are all soil ponds and are managedby private owners. Their water supplies aremainly from the influent of the KlongPravatebureerom Canal. During the studyperiod, all ponds were isolated (no extrawater entering each of these ponds).The culturing period was from April (summer,the time of cyanobacteria bloom) to October(pond 1-rainy season) and November(pond 2, 3-winter season). The total culturingperiods for Pond 1, 2, and 3 were 24, 28, and32 weeks, respectively, based on the situationof fish size and market demand.

2.2 SamplingPhytoplankton, water, and fish were

sampled every two weeks during the culturingperiod. Water samples were collected fromfive locations per pond at about 30 cm waterdepth and 1 m off the shores [15]. The volumeof water collected for microcystins analysiswas one liter per sample. All samples werepreserved on ice and then stored at -20°C inthe laboratory.

Water quality parameters weredetermined to describe the environmentalconditions of the sampling area.Measurements of all parameters were carriedout every two weeks (30 cm depth; 8 a.m.).Surface water temperature and transparencywere measured using a Celsius thermometerand a Secchi disc. pH and conductivity weremeasured using a digital meter. Dissolvedoxygen (DO), biological oxygen demand(BOD), nitrate nitrogen (NO3

-), and nitritenitrogen (NO2

-) were determined followingthe methodology of APHA [15]. Totalalkalinity, hardness, nitrogen (NH3),orthophosphate (PO4

2-), and total suspendedsolids (TSS) were analyzed following Stirling[16]. Chlorophyll and dry weight weredetermined following Becker [17].

2.3 Phytoplankton Sampling andIdentification

A 15 mm-mesh plankton net was usedto collect phytoplankton. At each location,a volume of 20 L of surface (depth 0-30 cm)water was passed through the net andfinally concentrated into a 100 ml sample.A 50 ml subsample was taken and preservedin 4% formaldehyde. In the laboratory,species identification were performedby using a Nikon (E 200) microscope atmagnifications of 200X and 400X. Afteridentification, the cell counting of filamentouscyanobacteria (Oscillatoria, Cylindrospermopsis,Lyngbya, Arthrospira, Raphidiopsis) wasenumerated filament-by-filament, andestimated by the average number of cellsper filament. For colonial cyanobacteria(Microcystis), cell numbers were estimated bycolony size and average number of cells percolony.

A volume of 100 L of surface water waspassed through plankton net and finally aconcentrated sample was stored at -20°Cfor microcystin analysis.

2.4 Fish SamplingFifteen fish of each pond were captured

with multi-mesh gillnets every two weeks.These fish were measured, weighed andsacrificed immediately, then dissected in thefield into muscle and liver and immediatelyfrozen at -20°C condition.

2.5 Microcystin ExtractionWater: A water sample (1 L) was filtered

through Whatman TM GF/C filter paper(0.22 mm pore size; Millipore).

Phytoplankton: Freeze dried cells (~ 0.1g dw) were homogenized and extracted threetimes with 2 ml of 75% methanol for 15 minusing ultrasound. The extract was centrifugedat 5,000 rpm and the supernatant was

30 Chiang Mai J. Sci. 2014; 41(1)

evaporated and re-dissolved with 20%methanol.

Fish: Dried samples of fish, liver, andmuscle (~ 0.5 g dw for each tissue) werehomogenized and extracted three times with10 mL of 70% methanol with the additionof 1% v/v trifluoroacetic acid for 5 minusing ultrasound. The extract was centrifugedat 5,000 rpm and the supernatant wasevaporated and re-dissolved with 20%methanol.

The water sample and solution ofphytoplankton and fish extracted weredirectly passed through a C18 cartridge(Waters sep-pak Millipore), which had beenpreconditioned by washing with 3 ml of100% methanol and 3 mL of distilled water.The column was washed with 1 mL ofdistilled water and 1 mL of 20% methanol.The final step column was washed with0.4 mL of 80% methanol and 0.6 mL ofdistilled water and used for the final detectionand identification of microcystins byliquid chromatography-mass spectrometry(LC-MS).

2.6 Microcystins Analysis by LC-MSQualitative and quantitative analysis of

microcystins were performed using LC-MS(Shimadzu, Kyoto, Japan) system equippedwith an auto sampler (SIL-HT), a LC pump(LC-10AD), a photo - diode array detector(SPD-M10A) and MS (2010A) using ZorbaxEclipse XDB-C18 column (2.1×150 mm,Agilent, USA) and 55% methanol in 0.1%formic acid as mobile phase at a flow rateof 0.2 mL/min with 10 μl injection. Theidentification and quantification of the toxinare determined by comparison with an

authentic toxin standard (Kanto Chemical, Co.Ltd., Tokyo, Japan). The quality control in eachanalytical batch were reagent blank spikerecovery and cross-check toxin standard.The limit of detection for the microcystin was0.25 μg/L. Standards of pure microcystins(MC-LR, -YR, and -RR) were purchasedfrom Pure Chemical Industries (Cica-Reagent).

2.7 StatisticsAll values are expressed as mean ±

standard deviation. Significant differences offish growth were determined using analysisof variance (ANOVA) with 95% confidence.A Pearson correlation test was used to establishassociation between different variables.Significance was accepted for p < 0.05.

3. RESULTS AND DISCUSSION3.1 Fish Growth and Water Quality

The growth of catfish was notsignificantly different among the three ponds(Figure 1), with average weights of 305.0 ±10.6, 312.6 ± 14.6, and 331.2 ± 17.4 g/bodyfor ponds 1, 2, and 3, respectively; however,the water quality parameters revealedsignificant pond variation (Table 1).

There was no mass mortality during theculturing period, a result indicating thatcyanobacterial blooms did not significantlyaffect the survival of fish. Water quality alsodid not affect survival. Even though thelevels of ammonia, pH, and DO (Table 1) inthe ponds were not optimal at all times, thefish still survived intact. The optimumconcentration of ammonia for aquatic animalsdid not exceed 0.2 mg/L; the optimumpH values for fish production is between 6and 9, with DO not lower than 5.0 mg/L[18].

Chiang Mai J. Sci. 2014; 41(1) 31

Figure 1. Average weight of catfish in various culturing ponds containing dense phytoplanktonbloom.

Table 1. Minimum, maximum and mean of physico-chemical variables of water from catfishpond.

(T= water temperature, Cond=conductivity, Alk=alkalinity, Hard=hardness, Tran=transparency, Ch-a=chlorophyll-a,DW=dry weight)

3.2 Species Diversity and Abundance ofPhytoplankton

Chlorophyll-a levels (1.28-2.06 mg/L)were higher than those of intensiveaquaculture reported (0.075 mg/L,sometimes 0.3-1mg/L) [6], indicating densephytoplankton blooms in the studiedcatfish ponds. Among these ponds, pond 1had the highest chlorophyll-a level,followed by ponds 3 and 2 (Table 1, Figure

2a), corresponding to the density oftotal cyanobacteria (Table 2, Figure 2b).The cyanobacterial blooms were generallydominated by Microcystis aeruginosa, Arthrospiraplatensis, Pseudanabaena spp., and Oscillatoria spp.

Thirteen cyanobacterial genera wereidentified in catfish ponds, of which sevengenera (Oscillatoria, Microcystis, Lyngbya,

Pond 1minimum-maximummeanPond 2minimum-maximummeanPond 3minimum-maximummean

pH

7.0-9.18.1

5.3-8.26.9

3.6-7.46.4

T(°C)

32.8-38.335.0

26.0-40.134.4

25.0-38.733.6

Cond(ms/cm2)

0.97-2.761.56

1.33-2.962.19

1.94-4.662.79

DO(mg/L)

6.1-27.817.2

2.8-24.28.0

1.6-14.45.1

BOD(mg/L)

13.3-173.359.6

8.3-174.739.5

1.5-172.042.7

Alk(mg/L)

32.7-205.0137.5

80.0-198.392.6

53.3-170.3111.6

Hard(mg/L)

96.7-170.3122.0

108.3-193.3163.0

118.7-217.7167.9

TSS(g/L)

0.54-2.621.58

1.12-3.121.96

1.13-3.462.08

Tran(cm)

4.0-19.711.3

4.3-28.313.4

6.7-31.713.5

NH3(mg/L)

1.27-8.484.98

0.31-7.342.60

7.47-21.8211.77

NO2-

(mg/L)

0.01-0.530.17

0.01-0.260.12

0.01-0.350.15

NO3-

(mg/L)

0.12-2.351.06

0.15-2.521.02

0.11-3.951.18

PO42-

(mg/L)

0.08-2.240.99

0.03-0.850.22

0.03-1.780.33

Ch-a(mg/L)

0.38-3.572.06

0.42-2.021.28

0.35-3.611.68

DW(g/L)

0.05-0.360.12

0.03-0.170.09

0.04-0.410.13

32 Chiang Mai J. Sci. 2014; 41(1)

Pseudanabaena, Anabaena, Anabaenopsis, andCylindrospermopsis) can produce a variety oftoxins of strain-dependent microcystins [14].The maximum densities of potentially toxiccyanobacteria in ponds 1, 2, and 3 were53 × 106, 2.8 × 106, and 25 × 106 cells/L,respectively (Figure 2c).

In all fish ponds, the maximum densitiesof M. aeruginosa, Pseudanabaena spp., Oscillatoriaspp., and Anabaena spp. occurred mostly atthe 12th-16th week of the studied period(Figure 2d-g). The maximum density ofM. aeruginosa (49.5 × 106 cells/L) in the fishpond was higher than those reported byKankaanpaa et al. [14] in a prawn pond,where the maximum density of M. aeruginosa

was 3.5 × 104 cells/L [14].There were differences among the

three fish ponds in terms of both thespecies composition of cyanobacteria foundand their relative abundance. It wassurprising to find that not only non-fixingcyanobacteria, but also fixing cyanobacteria,strongly depended on an adequate amountof ammonia and orthophosphate in orderto survive (Tables 2 and 3). This assumptionwas supported by the result that themaximum densities of Anabaena spp. andM. aeruginosa were found in pond 1, whichhad the highest concentration of ammoniaand orthophosphate, but not in pond 2,which had the lowest (Table 1).

Table 2. Average density of cyanobacteria (cells/L) in catfish pond.

- not found * potentially toxic cyanobacteria

Cyanobacteria Pond 1 Pond 2 Pond 3Chroococcus spp. 213 569 346Cyanosarcina sp. 412 678 310Gloeocapsa pleurocapsoides 9,089 4,767 110Merismopedia spp. 779,510 39,574 27,086Microcystis aeruginosa* 9,715,337 - 4,189,560Anabaena spp.* 20,850 - 4,135Anabaenopsis sp. * - 2500 -Cylindrospermopsis sp.* 587 - -Raphidiopsis spp. 833 - -Lyngbya sp. * 6,121 - -Oscillatoria spp.* 1,330,959 827,090 587,678Pseudanabaena spp.* 167,934 14,374 961,135Arthrospira platensis 6,281,386 3,600,002 733,857total potentially toxiccyanobacteria 11,241,788 843,964 5,742,508total cyanobacteria 18,313,231 4,489,554 6,504,217

Chiang Mai J. Sci. 2014; 41(1) 33

Figure 2. Variation of chlorophyll-a (a), total cyanobacteria (b), potentially toxic cyanobacteria(c), Microcystis aeruginosa (d), Pseudanabaena spp. (e), Oscillatoria spp. (f), and Anabaena spp. (g) incatfish pond.

Table 3. Pearson correlation between microcystin content in phytoplankton, total cyanobacteria,cyanobacteria produced microcystin and nutrient.

* Correlation is significant at the 0.05 level** Correlation is significant at the 0.01 level (MC=microcystin, Cyan=totalcyanobacteria, Os=Oscillatoria, Micro=Microcystis, Lyng=Lyngbya, Cylin=Cylindrospermopsis, Ana=Anabaena, Pseuda=Pseudanabaena,tox cyan=potentially toxic cyanobacteria)

MCCyanOs

MicroLyngCylinAna

PseudaNH3NO2

-

NO3-

PO42-

tox cyan

MC1

0.060.020.06-0.02-0.080.33*.056

0.48**0.30-0.17-0.150.06

Cyan0.06

10.69**0.90**0.21-0.13

0.79**.1220.070.110.02

0.44**0.92**

Os0.02

0.69**1

0.49**0.15-0.15

0.50**.093-0.200.12-0.040.27

0.56**

Micro0.06

0.90**0.49**

1-0.06-0.08

0.79**.0370.240.14-0.020.35*

0.99**

Lyng-0.020.210.15-0.06

1-0.030.12.015-0.10-0.190.22

0.44**-0.05

Cylin-0.08-0.13-0.15-0.08-0.03

10.07.0030.010.39*-0.19-0.09-0.09

Ana0.33*

0.79**0.50**0.79**

0.120.07

1-.0580.280.27-0.14

0.39**0.80**

Pseuda0.060.120.090.040.020.01-0.06

10.200.080.02-0.090.04

NH30.48*0.07-0.200.24-0.100.010.28.197

10.30-0.12-0.090.21

NO2-

0.300.110.120.14-0.190.39*0.27.0770.30

1-0.45**

-0.180.15

NO3-

-0.170.02-0.04-0.020.22-0.19-0.14.016-0.12

-0.45**1

0.31*-0.02

PO42-

-0.150.44**

0.270.35*

0.44**-0.09

0.39**-.091-0.09-0.180.31*

10.36*

tox cyan0.06

0.92**0.56**0.99**-0.05-0.09

0.80**.0430.210.15-0.020.36*

1

34 Chiang Mai J. Sci. 2014; 41(1)

3.3 Microcystins in Water, Fish, andPhytoplankton

Microcystins were not detectedthroughout the farming period in thesamples of water and fish (muscle and liver)from all catfish ponds. These resultscontrast with those from ponds with massiveblooms in other regions. For example,Zimba and Grimm [13] reported thatapproximately 50% of channel catfishponds in the southeastern United States hada detectable level of microcystin in thewater that was beyond the level (1 ng/mL)recommended by the World HealthOrganization as safe for human consumption[1].

Although we did not detect microcystinin these waters, the water quality indicatedthat the studied ponds were of eutrophicconditions (Table 1), which should befavorable for toxic cyanobacteria blooms.This is possibly due to the low density oftoxic cyanobacteria in the studied ponds.

For example, catfish ponds from Zaria ofNigeria, where the microcystins in the waterwere 0.6-5.89 mg/L [19], contained 12.5 to25 times more toxic cyanobacteria than foundin our studied ponds.

The concentrations of microcystin weredetected only in phytoplankton, withmicrocystin-RR values of 0.014-0.202 mg/gdw for pond 1, 0.024-0.141 mg/g dw forpond 3, but no detection in pond 2(Figure 3). It is probable that both M. aeruginosaand Anabaena spp. could contribute to theproduction of microcystins (ponds 1 and 3),since they did not occur in pond 2 (Figure 2d,g). Although there were a higher densitiesof M. aeruginosa and Anabaena spp. in thepresent study than those reported byKankaanpaa et al. [14] in a prawn pond, themicrocystin levels in this study were lower(0.5-1.2 mg/kg dw) [14], indicating thatmicrocystin concentration depends not onlyon cyanobacterial density, but also oncyanobacterial strain.

Figure 3. Variation of intracellular microcystin in phytoplankton of catfish pond.

Chiang Mai J. Sci. 2014; 41(1) 35

The results of this study show that onlymicrocystin-RR (MC-RR) was detected inphytoplankton; MC-LR and MC-YR werenot detected. Different microcystins havedifferent toxicity profiles. Among these, MC-RR has the lowest toxicity [20]; however,standard criteria of MC-RR for humanconsumption have not been proposed bythe World Health Organization.

The highest concentrations of MC-RRin ponds 1 and 3 were at the 24th and 32nd

week of the farming period, respectively,and MC-RR was not detected in pond 2throughout the sampling period (Figure 3).These variations could be a consequence ofdifferent environmental conditions, as well aschanges in the dominance of cyanobacterialspecies and strain composition amongsampling periods [7, 21].

The time period of the maximum densityof potentially toxic cyanobacteria (Figure 2c)did not simultaneously correspond to themaximum microcystin content (Figure 3).This is possibly due to the fact that a smallportion of secondary metabolites are lostduring the maximum cyanobacterial(exponential) growth [22].

The bloom of toxic cyanobacteria inother regions of Thailand has been reported.For example, a toxic cyanobacterium,M. Aeruginosa, with 5.86 × 10-4 g/g microcystinconcentration was present in a channel inthe Northeast region [23]. Also present ina reservoir in the same region wereCylindrospermopsis sp., Microcystis sp., Oscillatoriasp., and Pseudanabaena sp., with microcystinconcentrations in water of 1.09-1.23 g/L[24]. In the Northern region, the toxiccyanobacteria Microcystis spp. were dominantin prawn ponds, with a microcystinconcentration of 0.44 g/kg dw [25]. Induck-husbandry ponds and the water of adam in the central region, M. aeruginosa wasdominant, with microcystin concentrations

of 0.7-0.8 mg/g dw [26].

3.4 Correlation AnalysisMicrocystin content in phytoplankton

was highly correlated with the density ofAnabaena spp. (r=0.33, p<0.05) (Table 3).This may indicate that Anabaena spp. was themajor potentially cyanobacteria-producingmicrocystin in the studied ponds. Normallywith mixed cyanobacterial blooms ineutrophic water, it is uncommon to identifyonly one species as a microcystin producer;however, Vareli et al. [27] reported that onlyone genus, Microcystis sp., from mixedcyanobacterial blooms was responsible forall the microcystin production in a lake.

Densities of M. aeruginosa, Lyngbya sp., andAnabaena spp. were significantly correlatedwith orthophosphate (r=0.35, p<0.05; r=0.44,p<0.01; and r=0.39, p<0.01, respectively),whereas the density of Cylindrospermopsis sp.correlated with nitrite (r=0.39, p<0.05) (Table3). The microcystin content in phytoplanktonwas positively correlated with ammonia(r=0.48, p<0.01) and alkalinity (r=0.37,p<0.05), and negatively correlated with TSS(r=-0.36, p<0.05) and transparency (r=-0.35,p<0.05). No correlation with nitrate ororthophosphate was observed.

It has been reported that manyenvironmental factors, such as nitrogen,phosphorus, iron, pH, light intensity,and temperature, may affect microcystinproduction [28, 29]. In this study, we foundthat alkalinity and ammonia correlated withthe amount of microcystin in phytoplankton(Tables 3 and 4). This result supports thehypothesis that nitrogen can promotegrowth and microcystin production of toxiccyanobacteria [30], since ammonia plays amajor role as a nitrogen source for microcystinsynthesis in cyanobacteria of catfish ponds.

Kankaanpaa et al. [14] reported that thepeak of the cyanobacterial bloom was

36 Chiang Mai J. Sci. 2014; 41(1)

obviously boosted by high nutrient content,especially phosphate availability. We founda similar result, as the density of potentiallytoxic cyanobacteria was significantlycorrelated with orthophosphate (Table 3).Although it is common to associate themassive development of cyanobacteria withincreases in nutrients, it has not always beenpossible to establish a clear relationshipbetween nutrient levels and toxicity. Someauthors even claim that the production oftoxins is not necessarily connected witheutrophication [31].

In this study, microcystin was detectedonly in phytoplankton, so catfish mightget microcystin only via ingested toxiccyanobacteria. However, there was nodetection of microcystin in fish muscle andliver at all, even during the intensive catfishproduction in which the cyanobacteria wereblooming for more than 60-90% of theculturing period. It is unlikely that the catfishdid not ingest the toxic cyanobacteria at all,but it might be that they did not ingest enough,and for a long enough time, for detection

The results of this study showed thatthere was no accumulation of microcystin infish cultivated in these ponds, which were fedand managed by private owners. The situationis very different for fish feeding in naturalecosystems. In the ponds, catfish were fed withcommercial feed; thus, they could partiallyavoid ingesting the toxic cyanobacteria.In contrast to these ponds, catfish in anatural habitat that has toxic cyanobacteriablooms are more likely to accumulate moremicrocystin because they need to continuallyfeed on available food. Therefore, thechance of accidentally taking in toxiccyanobacteria is higher than that for thefish in our ponds.

It seems that carnivorous catfish that livein bodies of water with toxic cyanobacterialblooms may accumulate lower microcystin

than phytoplanktivorous fish, and highmicrocystin accumulation in phytoplanktivorous fish has been previously reported.For example, microcystin concentration washighest in the muscle of omnivorous fish,followed by phytoplanktivorous fish, andwas lowest in carnivorous fish [7]. Nyakairuet al. [12] reported that microcystinaccumulation in herbivorous fish (Oreochromisniloticus) in Lake Mburo was higher than inomnivorous fish (Lates niloticus), and suggestedthat omnivorous fish have more alternativesources of food.

In this study, after catfish were harvested,the water with dense, toxic cyanobacterialbloom was drained into a canal that was alsoused for plant (mimosa, banana, mango,coconut, etc.) farms. After cyanobacterial celllysis, microcystin can be immediately releasedand could accumulate in plant tissues [32].This is of special concern, and considerationshould be taken, since some treatment maybe needed before draining the water fromcatfish farms.

4. CONCLUSIONSThis study was an attempt to continuously

monitor the distribution of microcystin incatfish ponds in Thailand. The microcystinsin water and catfish (muscle and liver)were not sufficient to pose a risk tohumans; however, the degree of bloomtoxicity may vary, and accumulation maystill reach a toxic level. Therefore, long-termmonitoring is recommended in order tounderstand microcystin accumulationpatterns at all degrees of cyanobacterialbloom throughout the country. Theconsequences of the production and potentialliberation into the water of intracellularmicrocystin must be taken into consideration,along with other attempts to elucidate theirinfluence in the final concentration ofdissolved microcystin in water and to

Chiang Mai J. Sci. 2014; 41(1) 37

ascertain the potential effects on othercarnivorous fish.

ACKNOWLEDGEMENTSThis work was supported by grants from

the King Mongkut’s Institute of TechnologyLadkrabang.

REFERENCES

[1] WHO,Water sanitation cyanobactoxins.http://www.who. Accessed 20 January2012.

[2] Fisheries Statistics of Thailand,Department of Fishery. http://www.fisheries.go.th. Accessed 13 February2012.

[3] Wang X., Parkpian P., Fujimoto N.,Ruchirawat K.M., DeLaune R.D. andJugsujinda, A., Environmentalconditions associating microcystinsproduction to Microcystis aeruginosain a reservoir of Thailand, J. Environ.Sci. Health A-Tox. Hazard Subst.Environ. Eng., 2002; 37: 1181-1207.

[4] Dawson R.M., The toxicology ofmicrocystins, Toxicon., 1998; 36: 953-962.

[5] Chen J., Xie P., Li L. and Xu J., Firstidentification of the hepatotoxicmicrocystins in the serum of achronically exposed humanpopulation together with indication ofhepatocellular damage, Toxicol. Sci.,2009; 108: 81-89.

[6] Smith J.L., Boyer G.L. and ZimbaP.V., A review of cyanobacterialodoruous and bioactive metabolites:Impacts and management alternativesin aquaculture, Aquaculture, 2008;280: 5-20.

[7] Zhang D., Xie P., Liu Y. and Qiu T.,Transfer, distribution and bioaccumu-lation of microcystins in the aquaticfood web in Lake Taihu, China, withpotential risks to human health, Sci.Total Environ., 2009; 407: 2191-2199.

[8] Zimba P.V., Khoo L., Gaunt P.S.,Brittain S. and Carmichael W.W.,Confirmation of catfish, Ictalurus

punctatus Ralfinesque, mortality fromMicrocystis toxins, J. Fish. Diseases,2001; 24: 41-47.

[9] Ibelings B.W. and Chorus I., Accumulationof cyanobacterial toxins in freshwater“seafood” and its consequences forpublic health: A review, Environ.Pollut., 2007; 150: 177-192.

[10] Chen J., Xie P., Zhang D. and Lei H.,In situ studies on the distributionpatterns and dymanics of microcystinsin a biomanipulation fish-bighead carp(Aristichthys nobilis), Environ. Pollut.,2007; 147: 150-157.

[11] Deblois C.P., Aranda-Rodriguez R.,Giani A. and Bird D.F., Microcystinaccumulation in liver and muscle oftilapia in two large Brazilianhydroelectric reservoirs, Toxicon.,2008; 51: 435-448.

[12] Nyakairu G.W.A., Nagawa C.B. andMbabazi J., Assessment ofcyanobacteria toxins in freshwater fish:A case study of Murchison Bay (LakeVictoria) and Lake Mburo, Uganda,Toxicon., 2010; 55: 939-946.

[13] Zimba P.V. and Grimm C.C., Asynoptic survey of musty/muddy odormetabolites and microcystin toxinoccurrence and concentration insoutheastern USA channel catfish(Ictalurus punctatus Ralfinesque)production ponds, Aquaculture, 2003;218: 81-87.

[14] Kankaanpaa H.T., Holliday J.,Schroder H., Goddard T.J., VonFister R. and Carmichael W.W.,Cyanobacteria and prawn farming innorthern New South Wales, Australia-a case study on cyanobacteria diversityand hepatotoxin bioaccumulation,Toxicol. Appl. Pharmacol., 2005; 203:243-256.

[15] APHA, Standard Methods for theExamination of Water and WasteWater, 21st Edn., American PublicHealth Association, Washington DC,2005.

[16] Stirling H.P., Chemical and BiologicalMethods of Water Analysis forAquaculturists, Institute of Aquaculture,University of Stirling, Stirling, 1985.

38 Chiang Mai J. Sci. 2014; 41(1)

[17] Becker E.W., Microalgae: Biotechnologyand Microbiology, Cambridge UniversityPress, New York, 1994.

[18] Boyd C.E. and Tucker C.S., PondAquaculture Water Quality Management,Kluwer Academic Publishers,Massachusetts, 1998.

[19] Chia A.M., Oniye S.J., Ladan Z.,Lado Z., Pila A.E., Inekwe V.U. andMmerole J.U., A survey for thepresence of microcystins inaquaculture ponds in Zaria, northern-Nigeria: Possible public healthimplication, Afr. J. Biotechnol.,2009; 22: 6282-6289.

[20] Gupta N., Pant S.C., VijayaraghavanR. and Rao P.V.L., Comparativetoxicity evaluation of cyanobacterialcyclic peptide toxin microcystinvariants (LR, RR, YR) in mice,Toxicol., 2003; 188: 285-296.

[21] Mez K., Beattie K.A., Codd G.A.,Hanselmann K., Hauser B., Naegeli H.and Preisig H.R., Identification of amicrocystin in benthic cyanobacterialinked to cattle deaths on alpinepastures in Switzerland, Eur. J.Phycol., 1997; 32: 111-117.

[22] Zimba P.V., Dionigi C.P. and MillieD.F., Evaluating the relationshipbetween photopigments and secondarymetabolite accumulation incyanobacteria, J. Phycol., 1999; 35:1422-1429.

[23] Sengpracha W., Suvannachai N. andPhutdhawong W., Microcystin LRcontent in Microcystis aerugionsa Kutzcollected from Sri Sakhett, Thailand,Chiang Mai J. Sci., 2006; 33: 231-236.

[24] Somdee T., Kaewkhiaw K. andSomdee A., Detection of toxiccyanobacteria and quantification ofmicrocystins in four recreational waterreservoirs in Khon Kaen, Thailand,KKU Res. J., 2013; 18: 1-8.

[25] Prommana R., Peerapornpisal Y.,Whangchai N., Morrison L.F.,Metcalf J.S., Ruangyuttikarn W.,Towprom A. and Codd G.A.,Microcystins in cyanobacterial blooms

from two freshwater prawn(Macrobrachium rosenbergii) ponds innorthern Thailand, Science Asia, 2006;32: 365-370.

[26] Mahakhant A., Sano T., RatanachotP., Tong-a-ram T., Srivastava V.C.,Watanabe M.M. and Kaya K.,Detection of microcystins fromcyanobacterial water blooms inThailand freshwater, Phycol. Res.,1998; 46: 25-29.

[27] Vareli K., Pilidis G., MavrogiorgouM.C., Briasoulis E. and Sainis I.,Molecular characterization ofcyanobacterial diversity and yearlyfluctuations of microcystin loads in asuburban Mediterranean lake (LakePamvotis, Greece), J. Environ. Monit.,2009; 11: 1506-1512.

[28] Dai R., Liu H., Qu J., Zhao X. andHou Y., Effects of amino acids onmicrocystin production of theMicrocystis aeruginosa, J. Hazard.Mater., 2009; 161: 730-736.

[29] Tonk L., Welker M., Huisman J., andVisser P.M., Production ofcyanopeptolins, anabaenopeptins, andmicrocystins by the harmfulcyanobacteria Anabaena 90 andMicrocystis PCC 7806, Harmful Algae,2009; 8: 219-224.

[30] Davis T.W., Berry D.L., Boyer G.L.and Gobler C.J., The effects oftemperature and nutrients on thegrowth and dynamics of toxic andnon-toxic strain of Microcystis duringcyanobacteria blooms, Harmful Algae,2009; 8: 715-725.

[31] Willen T. and Mattsson R., Water-blooming and toxin-producingcyanobacteria in Swedish fresh andbrackish waters, 1981-1995,Hydrobiologia, 1997; 353: 181-192.

[32] McElhiney J., Lawton L.A. and LeifertC., Investigations into the inhibitoryeffects of microcystins on plantgrowth, and the toxicity of planttissues following exposure, Toxicon.,2001; 39: 1411-1420.