Monitoring Algal Toxins in Lake Water by Liquid Chromatography

Analytes were quantified by LC coupled to tandem mass spectrometry in the .... Australia, was as sociated with severe hepatoenteropathy to 148 indigen...
0 downloads 0 Views 149KB Size
Download PDF
Environ. Sci. Technol. 2006, 40, 2917-2923

Monitoring Algal Toxins in Lake Water by Liquid Chromatography Tandem Mass Spectrometry SARA BOGIALLI,† MILENA BRUNO,‡ ROBERTA CURINI,† A N T O N I O D I C O R C I A , * ,† C H I A R A F A N A L I , † A N D A L D O L A G A N AÅ Dipartimento di Chimica, Universita` La Sapienza, Piazza Aldo Moro 5, 00185 Rome, Italy, and Dipartimento d’Igiene Ambientale, Istituto Nazionale di Sanita`, Viale Regina Elena 299, 00161 Rome, Italy

Microcystins (MCs) and cylindrospermopsin (CYL) are potent natural toxins produced by cyanobacteria (bluegreen algae) that grow worldwide in eutrophic freshwaters and cause animal and human water-based toxicoses. The main purpose of this work has been assessing the contamination levels of some MCs and CYL in eutrophic Italian lake (Albano) water. To do this, we have developed an original analytical method involving MC extraction with a sorbent (Carbograph 4) cartridge. CYL is a highly polar compound that is scarcely retained by any sorbent material. To analyze this toxin, we directly injected 0.5 mL of filtered lake water into the liquid chromatography (LC) column. Analytes were quantified by LC coupled to tandem mass spectrometry in the multireaction monitoring mode. The recovery of five selected MCs added to an analytefree lake water sample at three different concentrations (50, 150, and 500 ng/L) ranged between 93 and 103% with RSD values no larger than 8%. Limits of quantification (LOQ) of the five MCs were within the 2-9 ng/L range, whereas the LOQ of CYL was 300 ng/L. The occurrence and abundance of cyanotoxins in Lake Albano was monitored over four months (Sept-Dec 2004) by analyzing water samples collected monthly at the center of the lake and at different depths (from 0 to -30 m). During survey and with the MS/MS system operating in the parent ion scan mode, we individuated two demethylated forms of MCRR and one demethylated variety of MC-LR. Demethylated MC-RRs are known to be even more toxic than MC-RR toward zooplanktic grazers. CYL was the most-abundant toxin during the first three monitoring months. To the best of our knowledge, this is the first work reporting concentration levels of CYL in lake water.

Introduction Cyanobacteria (blue-green algae) are a group of prokaryotic organisms that occur worldwide in environments that include hot springs and under Antarctic ice. A small group of genera produce toxins that have caused both animal (1) and human poisonings (2). In the early 1980s, the chemical structure of * Corresponding author phone: 39-06-49913752; fax: 39-0649913680; e-mail: [email protected]. † Universita ` La Sapienza. ‡ Istituto Nazionale di Sanita `. 10.1021/es052546x CCC: $33.50 Published on Web 04/01/2006

 2006 American Chemical Society

one of these cyanobacterial hepatotoxins, called microcystins (MCs), was determined (3). The structure was elucidated as a cyclic peptide containing seven amino acids. Two amino acids (in positions 2 and 4 of the ring) are variable L-amino acids (R1 and R2, Figure 1). The structural variations are indicated by suffix letters. For example, MC-LR contains leucine (L) in position 2 and arginine (R) in position 4. Later, further MCs were identified with structural variations originating from side-chain modifications other than at position 2 or 4. In cases where the N-methyl-dehydroalanine (Medha) in position 7 is demethylated, (Dha)MCs are formed, whereas demethylation of D-erythro-β-methyl aspartic acid (D-MeAsp) in position 3 leads to (D-Asp)MCs. In addition, variations of the Adda moiety in position 5 have been described (Figure 1). To date, more than 60 MCs have been isolated and characterized (4). The toxicity of MCs is due to severe inhibition of protein phosphatase 1 (PP-1) and 2A (PP-2A), causing mainly functional and structural disturbances of the liver (5). In addition, MCs can also act as tumor promoters (6-8). The danger of MCs is enhanced by the fact that some species form surface scum that can be consumed by animals, thus producing high-level intakes of toxins. More recently, another toxin, the hepatotoxic alkaloid cylindrospermopsin (CYL) (see Figure 1), has also been detected in cyanobacterial blooms. This toxin was retrospectively indicated (2) as the causative agent in a human poisoning incident in 1979, in which a large cyanobacterial bloom on Solomon Dam, Palm Island, Australia, was associated with severe hepatoenteropathy to 148 indigenous people (9). CYL is receiving increasing attention by toxicologists and health authorities because the main producer, Cylindrospermopsis raciborskii, is expanding its geographical extension with a considerable pace. As a matter of fact, this alga was not commonly found in the United States until about 10 years ago when it became a regular component of waterblooms in Florida (10). Because toxic cyanobacterial blooms occur in eutrophic lakes, ponds, and rivers around the world, a large part of the human population becomes exposed to low levels of MCs in their drinking water. To protect consumers from the adverse effects of cyanobacterial peptide toxins, the World Health Organization (WHO) recently proposed a provisional upper limit in drinking water of 1 µg/L for MC-LR (11). However, effective consumer protection requires efficient detection of the whole spectrum of cyanobacterial toxin congeners, many of which are as toxic as MC-LR. Monitoring of water bodies for MCs and CYL poses problems because cyanobacterial blooms may contain complex mixtures of MCs and sometimes several classes of toxins. To elucidate the toxin profile and thus assess the total toxicity of a complex aqueous sample, we require chromatographic techniques. These methods should be able to separate, identify, and quantify individual MCs because MC variants have different toxicities. In the past, methods based on reversed-phase high-performance liquid chromatography (LC) with diode array detection (12) or electrochemical detection (13) have been proposed for analyzing cyanotoxins in real water samples. However, methods created on the basis of the detectors cited above do not offer unequivocal and definitive analyte identification, which is a necessary feature in confirmatory assays. The advent of the electrospray ion (ESI) source coupled to LC has provided a facile method of ionizing nonvolatile thermolabile substances, such as algal toxins, for mass spectrometry detection. Poon and co-workers (14) were the first to use LC-MS with an ESI source for identifying MCs VOL. 40, NO. 9, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

2917

FIGURE 1. General structure of (a) microcystins (MCs) and (b) cylindrospermopsin.

TABLE 1. Multi-Reaction-Monitoring Conditions for Detecting Some Selected Cyanotoxins by Tandem MS compd microcystin-RR desmethylmicrocystin-RR microcystin-YR desmethylmicrocystin-LR microcystin-LR microcystin-LA trimethacarb (IS) microcystin-LW cylindrospermopsin 1,9-diaminononane (IS)

transitions (m/z)

cone voltage (V)

collision energy (eV)

520a > 135 520 > 887 512.5 > 135 523 > 135 523 > 910 1045 > 135 491 > 135 981 > 135 498 > 135 498 > 860 995 > 135 910 > 402.5 909 > 776 194 > 137 1025 > 1007 1025 > 873 416 > 194 416 > 336 159 > 142

35

35 25 35 5 5 70 15 60 15 40 60 25 25 10 30

0-5.8

40 20 12

b

a Double-charged molecular ions are reported in boldface. using 1,9-diaminononane as an internal standard.

b

9

25 30 45 28

5.8-9.0

9.0-13

b

Cylindrospermopsin was analyzed separately by direct injection of the water sample

and other algal toxins. Hummer et al. (15) demonstrated the usefulness of LC-MS equipped with a triple quadrupole in identifying those MCs for which standards are unavailable. To do this, the authors operated the first quadrupole (Q1) in the scan mode after selecting two mass ranges, m/z 9001150 for single-charged MCs and m/z 505-525 for double2918

35 18 18 70 20 70 20 20 70 35

retention window (min)

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 9, 2006

charged MCs, whereas we set the third quadrupole (Q3) at m/z 135 for monitoring the [PhCH2CH(OMe)]+ fragment ion that is generated by the R-cleavage at the methoxy group of the Adda β-amino acid moiety and is a “key-fragment ion” for MCs. Recently, Luckas et al. used tandem MS detection for detecting MCs in the North and Baltic seas (16).

FIGURE 2. MRM LC-MS/MS chromatogram resulting from analysis of a lake water sample spiked with some selected microcystins at a 50 ng/L level. MC stands for microcystin. IS ) internal standard (trimethacarb).

TABLE 2. Recovery of Five Microcystins by Analyzing Increasing Volumes of Lake Water (spike level ) 100 ng/L) recoverya

(RSD) (%)

compd

0.5 L

1L

2L

microcystin-RR microcystin-YR microcystin-LR microcystin-LA microcystin-LW

99 (6) 98 (6) 100 (5) 102 (5) 96 (6)

96 (5) 98 (4) 95 (5) 98 (6) 97 (4)

95 (3) 85 (5) 86 (4) 98 (3) 83 (4)

a

Mean values from quadruplicate experiments.

The purpose of this study has been 2-fold. One has been to develop a very sensitive and reliable method based on LC-tandem MS for analyzing some selected MCs and CYL in freshwater samples at sub-µg/L levels. This method involves solid-phase extraction (SPE) with a Carbograph 4 (a type of graphitized carbon black) cartridge to isolate MCs from water. CYL is a very polar compound with a scarce tendency to be adsorbed on conventional sorbents. This method proposes the direct injection of 0.5 mL of filtered freshwater into the LC-MS/MS apparatus to analyze this toxin. The second objective has been that of assessing concentration levels of the analytes in Lake Albano, which is located in the area of Rome.

Experimental Section Reagents and Chemicals. MC-RR, MC-YR, MC-LR, MCLA, and MC-LW were purchased from Calbiochem (La Jolla, CA). CYL was purchased from Sigma-Aldrich (Milwaukee, WI). Trimethacarb (Riedel-de Hae¨n, Seelze, Germany) is an obsolete insecticide and was used as the internal standard (IS) for quantifying MCs, whereas quantification of CYL was carried out using 1,9-diaminononane (Sigma-Aldrich) as the IS. Individual standard solutions of the five MCs and CYL were prepared by dissolving each compound in water to obtain 25 µg/mL concentrations. After preparation, these solutions were stored at -18 °C in the dark to minimize analyte degradation. They were freshly prepared every two months. A composite working standard solution of MCs was prepared by mixing the above solutions and diluting them with water to obtain analyte concentrations of 1 µg/mL. Solutions of the two ISs (1 mg/mL), trimethacarb and 1,9diaminononane, were separately prepared by dissolving them in acetonitrile. We obtained distinct working solutions of the two ISs at a 1.5 µg/mL concentration by diluting them with 10 mmol/L formic acid-acidified acetonitrile. When unused, all working standard solutions were stored at 4 °C in the dark and were renewed after one working week. Acetonitrile RS of gradient grade was obtained from Carlo Erba (Milan, Italy). Trifluoroacetic acid (TFA) was from Fluka VOL. 40, NO. 9, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

2919

FIGURE 3. MRM LC-MS/MS chromatogram resulting from direct injection into the LC column of 0.5 mL of a lake water sample spiked with cylindrospermopsin at 1000 ng/L. (a) Ion current profile relative to the internal standard (1,9-diaminononane); (b and c) ion current profiles relative to the two reaction transitions selected for cylindrospermopsin. (Bucks, Switzerland). All other solvents and chemicals were of analytical grade (Carlo Erba) and were used as supplied. Site Description. Lake Albano is a volcanic basin inside a crater depression with ellipsoidal shape whose longer axis, NW-SE oriented, is 3.5 km long and the shorter one is 2.3 km long. The lake’s perimeter is 12 km long and its depth reaches 175 m. The mean renewal time is more than 67 years. Lacking emissaries and tributaries, the lake is characterized by an ancient Roman channel carved in the lava rocks through the crater, which has the purpose of regulating the lake level. The lake is liable to a heavy pollution due to human activities, which have already caused the development of algal blooms in the past. Since 1960, the lake level has lowered almost four meters under the artificial emissary. This is one of the reasons for the worsening of trophic conditions, caused especially by the reckless sucking from the water layer of this area. Sample Collection. Over four months of monitoring (Sept-Dec 2004), water samples were taken monthly at one single station located at the center of the lake and at different depths, i.e., 0, 5, 10, 15, 20, 25, and 30 m, using 2.5 L Ruttner bottles. Samples were stored in ice chests and transported to the laboratory, where they were stored at -18 °C for some days and then thawed for analysis. In this way, we determined the total cyanotoxin content in the water/algae samples, as the freeze-thawing operation has the effect of bursting cells with the consequent release of intracellular cyanotoxins (17). Before processing samples, we removed suspended particulate matter by filtration with a 125 mm diameter Black Ribbon 589 paper filter (Schleicher & Schuell, Legnano, Italy) to avoid SPE cartridge plugging. 2920

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 9, 2006

Extraction Apparatus. Extraction cartridges filled with 0.5 g of Carbograph 4 and drilled cylindrical Teflon pistons with indented bases and Luer tips for analyte elution in the back-flushing mode (18) were supplied by LARA, Rome, Italy. The SPE cartridge was fitted into a sidearm filtration flask, and liquids were forced to pass through the cartridge by vacuum (water pump). Before processing aqueous samples, we washed the cartridge with 20 mL of water acidified with HCl (pH 2), followed by 5 mL of distilled water. Sample Preparation. Apart from CYL (see below), MCs were extracted from 0.5 L of lake water. The flow rate at which the aqueous samples passed through the cartridge was maintained at 40-50 mL/min by vacuum. After passage of the aqueous samples through the SPE cartridge, it was washed with 7 mL of distilled water. Water remaining in the cartridge was partially removed by drawing room air through the cartridge for 1 min. Residual water was eliminated by slowly passing 0.5 mL of methanol through the trap. After the passage of methanol, the cartridge was again air-dried. A Teflon piston was then forced into the cartridge until it touched the upper frit; the cartridge was reversed, and analytes were eluted by back flushing the cartridge with 1 mL of methanol followed by 4 mL of 10 mmol/L TFA-acidified CH2Cl2-CH3OH (80:20, v/v) at a flow rate of about 4 mL/ min. When re-extraction of the analytes was performed by forward elution, poor recovery of MC-LW was experienced. The eluate was collected in a 1.4 cm id glass vial with a conical bottom. The last drops of this solvent mixture were collected by further decreasing the pressure inside the vacuum flask.

TABLE 3. Total (intra- and extracellular) Levels (ng/L) of Some Microcystins and Cylindrospermopsin Found in Lake Albano during Four Months of Monitoring (Sept-Dec 2004) depth (m)

desmethyl MC-RR isomer 1a

desmethyl MC-RR isomer 2

desmethyl MC-LRb

MC-RR

MC-YR

ndd 4 nd nd nd nd

189 9240 32 159 9 73

18430 9500 1230 2910 1430 2210

nd nd nd nd nd nd nd

25 nd 109 757 109 nd nd

14900 16000 15500 1520 1400 1280 1580

2 nd nd 20 20 6 6

16 14 12 15 nd nd nd

3700 3600 2500 2870 641 770 410

21 12 20 15 16 34 nd

246 584 1064 700 670 2420 16

CYL

Sept surfacec -5 -10 -15 -20 -25 -30

60 100 27 120 3 5.5

60 80 20 100 3 nd

surface -5 -10 -15 -20 -25 -30

13 nd 28 67 10 2 nd

13 nd 11 28 5 2 nd

surface -5 -10 -15 -20 -25 -30

75 29 37 40 20 10 7

32 8 12 10 11 5 5

surface -5 -10 -15 -20 -25 -30

920 897 1400 890 925 2220 9

260 270 407 230 275 580 6

Oct

Nov

Dec 222 122 637 483 126 1748 nd

nd nd nd nd nd nd nd

a The two demethylated forms of MC-RR were designated isomers 1 and 2, as their structures could not be elucidated. b The lack of concentration data for demethyl MC-LR in the first 3 monitoring months was due to the fact that only in December did we incidentally suspect and evidence the presence of this toxin after the MS/MS system was operated in the parent ion scan mode. c The surface water sample was not analyzed because it was accidentally damaged. d nd ) not determined.

Before solvent removal, 10 µL of the trimethacarb (IS) working solution was added to the extract. Partial solvent removal was carried out in a water bath at 50 °C under a gentle stream of nitrogen. Solvent evaporation was stopped when the extract reached the volume of about 50 µL. Two hundred microliters of a water-acetonitrile solution (70:30, v/v) acidified with 10 mmol/L formic acid was added to the residue, and 50 µL of the final extract was injected into the LC column. CYL is a highly polar compound that is scarcely retained by any sorbent material. To analyze this toxin, we spiked 5 mL of filtered lake water with 15 µL of the IS (1,9diaminononane), and 0.5 mL of the sample was then directly injected in the LC-MS/MS apparatus. Instrumental Analysis. The liquid chromatograph consisted of a Waters pump (model 600E, Milford, MA), a 500 µL injection loop, and HP 5 µm C-18 guard (7.5 × 4.6 mm i.d.) and analytical (250 mm × 4.6 mm i.d.) columns (Alltech, Sedriano, Italy) thermostated at 35 °C and was interfaced by an electrospray ion (ESI) source to a benchtop triplequadrupole mass spectrometer (model micromass 4 MICRO API, Waters). Mobile phase component A was acetonitrile and component B was water; both were acidified with 10 mmol/L formic acid. In any case, the flow rate of the mobile phase was 1.0 mL/min. For chromatographing MCs, the mobile phase gradient profile was as follows (t in min): t0, A ) 35%; t5, A ) 45%; t6, A ) 57%; t11, A ) 67%; t12, A ) 100%; t15, A ) 100%; t16, A ) 35%, t25, A ) 35%. For chromatographing CYL, the mobile phase gradient profile was as follows: t0, A

) 0%; t10, A ) 20%; t11, A ) 100%; t13, A ) 100%; t14, A ) 0%; t23, A ) 0%. A diverter valve led the effluent into the ion source, operating in the positive ion mode, with a flow of 400 µL/min. High-purity nitrogen was used as the drying and curtain gas, whereas high-purity argon was used as the collision gas. Nebulizer gas was set at 650 L/h, whereas the cone gas was at 50 L/h; the ion source and desolvation temperature were maintained, respectively, at 110 and 350 °C. The settings for the gas pressure in the collision cell were set at 3 mbar. The capillary voltage was 3000 V. Cone voltage, collision energy, and other transmission parameters were optimized for each analyte (data are reported in Table 1). The multi-reaction-monitoring (MRM) mode was used for quantification by selecting, when possible, at least two molecular-ion decomposition reactions for each analyte (Table 1). Quantification. Analytes for which standards were available were quantified by the external standard quantification procedure. Standard solutions were prepared at eight levels by using appropriate volumes of the working standard solution. For each analyte, the peak area versus injected amount chart was obtained by measuring at any injected amount the resulting peak area and relating this area to that of the internal standard. The response of the ESI-MS/MS system was linearly related to injected amounts of the analytes up to 300 ng. When amounts of MCs injected from extracts of lake water samples exceeded the upper limit of the linear dynamic range of the detector response, extracts were suitably diluted and re-injected. By following the method described VOL. 40, NO. 9, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

2921

FIGURE 4. MRM LC-MS/MS chromatogram resulting from analysis of a Lake Albano water sample (Sept 2004, -15 m depth) contaminated by two isomers of desmethyl microcystin-RR. Noteworthy is that separation of the two isomers was possible only by replacing methanol (lower) with acetonitrile (upper). As measured by us, the concentration levels of isomers 1 and 2 were, respectively, 27 and 20 ng/L. by Hummert et al. (15), we identified two demethylated varieties of MC-RR and one demethylated form of MC-LR on monitoring cyanotoxins in Lake Albano. Standards of these three MCs were not available to us. Thus, we quantified the above-cited MCs by assigning to them the same molar response factors of the respective fully methylated MCs.

Results and Discussion Recovery Studies. The presence of organic substances in environmental aqueous samples, mainly fulvic acids, can make SPE cartridges less efficient in extracting target compounds because of sorbent saturation effects. The extraction efficiency of the Carbograph 4 cartridge was assessed by extracting increasing volumes of analyte-free lake (Bracciano) water spiked with the analytes at a 100 ng/L level. At each volume, four determinations were performed; the results are reported in Table 2. As can be seen, losses of some MC-YR, MC-LR, and MC-LW were observed when extracting 2 L of the lake water. For the sake of rapidity and in order to avoid loss of the above analytes when analyzing environmental water samples particularly rich in organic material, we suggest processing no more than 0.5 L of lake water. Under this condition, the method sensitivity allows the determination of MCs to be performed at levels lower than 15 ng/L (see Limits of Detection and Quantification of the Method). The accuracy and precision of this method for the five MCs considered was assessed by analyzing an analyte-free 2922

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 9, 2006

lake (Bracciano) water sample spiked with increasing amounts of the above analytes, i.e, 50, 150, and 500 ng/L. At each analyte concentration, we performed four determinations. In any case, recoveries of the analytes were not less than 93% with a RSD not more than 8%. On surveying species and levels of algal toxins in Lake Albano, we identified three demethylated forms of MCs (see below) by the parent ion scan mode (15), for which standards were not available. For this reason, no recovery study of these species could be performed in the conventional way. A way for estimating an eventual failure of our sample preparation protocol in quantitatively recovering these varieties of MCs was that of both halving the lake water sample volume and doubling the volume of the desorbing solvent mixture (see Experimental Section). In both cases, we measured concentration levels of the three MCs mentioned above that were not significantly different from those measured by the procedure reported in the Experimental Section. Limits of Detection and Quantification of the Method. LOQs of the method were estimated from the MRM LCMS/MS chromatograms shown in Figures 2 and 3 that result from analysis of an analyte-free lake (Bracciano) water sample spiked with selected MCs at a 50 ng/L individual concentration and with 1000 ng/L of CYL. After extracting the sum of the ion currents of the transitions selected for each analyte, we smoothed the resulting trace twice by applying the mean smoothing method. Thereafter, the peak height to averaged background noise ratio was measured. The background noise

estimate was based on the peak-to-peak baseline near the analyte peak. LOQs were then calculated on the basis of a minimal accepted value of the signal-to-noise ratio (S:N) of 10. As calculated by us, LOQs of MCs were between 2 ng/L (MC-RR) and 9 ng/L (MC-LW). Although no enrichment step of CYL could be performed by the SPE cartridge, the LOQ of this toxin was estimated to be 300 ng/L. Monitoring Cyanotoxins in Lake Albano. Concentration levels of toxins found in Lake Albano are shown in Table 3. Among the five MCs detected in the water column of the central part of Lake Albano, two isomers of desmethyl-MCRR (Figure 4) were often the most abundant ones. Demethylated MC-RR variants are known as the characterizing toxic markers of the algal species Planktothrix rubescens (1921). At low concentrations, demethylated MC-RRs have been reported to be more toxic than fully methylated MC-RR toward zooplanktic grazers (21). The presence of the dehydrobutyrine residue has been proposed to be the reason for the higher specific toxicity of (D-Asp 3)(Dhb 7)MC-RR, as compared to N-methyldehydroalanine-containing MCs (21). As measured by us, the second most abundant MC was MC-YR, and in December, the (D-Asp)MC-LR was detected as the third most produced MC. Unfortunately, concentration levels of (D-Asp)MC-LR during the first three monitoring months were not measured, as only in December the presence of this toxin was incidentally ascertained with the MS/MS operated in the parent ion scan mode (15). During the Oct-Dec monitoring period and except for that at -30 m, we measured increasing MC concentrations at any depth of Lake Albano. This finding could be traced to the beginning of the winter blooming period of the Planktothrix rubescens species. The maximum concentration of MC-YR in Lake Albano was found in September at a -10 m depth. This situation was probably generated by a subsurface maximum of the cell population bearing a change in the metabolic toxin production. This change was presumably due to specific environmental conditions or to a clone developed in a genetically heterogeneous population. In the first three months of monitoring, CYL was constantly and largely the most-abundant toxin measured in Lake Albano. To the best of our knowledge, this is the first work reporting concentration levels of CYL in lake water. The sudden disappearance of CYL in December or its concentration decrease at levels lower than the LOD of the method (150 ng/L) was not fully clear to us. Hawkins et al. (22), however, have reported that cyanobacteria producing CYL are able to form resistant cysts, called akinetes, that are capable of surviving during winter and spring by reaching freshwater beds until summer comes back. On the basis of toxicological studies, Humpage and Falconer (23) have proposed a guideline value of 1000 ng/L for CYL in drinking water. On monitoring Lake Albano, we measured CYL concentrations that were often much higher than the above-mentioned guideline value. Results of this study indicate that routine monitoring for CYL should be performed in waterbodies used as drinking water supplies.

Literature Cited (1) Carmichael, W. Cyanobacteria secondary metabolitessThe cyanotoxins. J. Appl. Bacteriol. 1992, 72, 445-459. (2) Hawkins, P.; Runnegar, M.; Jackson, A.; Falconer, I. Severe hepatotoxicity caused by the tropical cyanobacterium (bluegreen alga) Cylindrospermopsis raciborskii (Woloszynska) Seenaya and Subba Raju isolated from a domestic water supply reservoir. Appl. Environ. Microbiol. 1985, 50, 1292-1295. (3) Botes, D.; Kruger, H.; Viljoen, C. Isolation and characterization of four toxins from the blue-green alga Microcystis aeruginosa. Toxicon 1982, 20, 945-954. (4) Chorus, I., Bartram, J., Eds.; Toxic Cyanobacteria in Water: A Guide to Public Health Consequences and Their Management in Water Resources and Supplies; World Health Organization, E & FN Spon: London, 1999.

(5) Yoshizawa, S.; Matsushima, R.; Watanabe, M. F.; Harada, K.-I.; Ichihara, A.; Carmichael, W. W.; Fujiiki, H. Inhibition of protein phosphatases by microcystis and nodularin associated with hepatotoxicity. J. Cancer Res. Clin. Oncol. 1990, 116, 609-614. (6) Falconer, I. R. Potential impact on human health of toxic cyanobacteria. Phycologia 1996, 35, 6-11. (7) Nishiwaki-Matsushima, R.; Ohta, T.; Nishiwaki, S.; Suganuma, M.; Kohyama, K.; Ishikawa, T.; Carmichael, W. W.; Fujiki, H. Liver tumor promotion by the cyanobacterial cyclic peptide toxin microcystin-LR. J. Cancer Res. Clin. Oncol. 1992, 118, 420424. (8) Ueno, Y.; Nagata, S.; Tsutsumi, T.; Hasegawa, A.; Watanabe, M.; Park, H. D.; Chen, G.-C.; Yu, S. Z. Detection of microcystins, a blue-green algal hepatotoxin, in drinking water sampled in Haimen and Fusui, endemic areas of primary liver cancer in China, by highly sensitive immunoassay. Carcinogenesis 1996, 17, 1317-1321. (9) Byth, S. Palm Island mystery disease. Med. J. Aust. 1980, 2, 4042. (10) Chapman, A. D.; Schelske, C. L. Recent appearance of Cylindrospermopsis (cyanobacteria) in five hypereutrophic Florida lakes. J. Phycol. 1997, 33, 191-195. (11) WHO. Guidelines for Drinking-Water Quality, 2nd ed.; addendum to Vol. 2, Health Criteria and Other Supporting Information; World Health Organization: Geneva, Switzerland, 1998. (12) Harada, K. I.; Matsuura, K.; Suzuki, M.; Oka, H.; Watanabe, M. F.; Oishi, S.; Dahlem, A. M.; Beasley, V. R.; Carmichael, W. W. Analysis and purification of toxic peptides from cyanobacteria by reversed-phase high-performance liquid chromatography. J. Chromatogr. 1988, 448, 275-283. (13) Meriluoto, J.; Kincaid, B.; Smyth, M. R.; Wasberg, M. Electrochemical detection of microcystins, cyanobacterial peptide hepatotoxins, following high-performance liquid chromatography. J. Chromatogr., A 1998, 810, 226-230. (14) Poon, G. K.; Griggs, L. J.; Edwards, C.; Beattie, K. A.; Codd, G. A.; Pleasance, S.; Dear, G. J. Liquid chromatography-electrospray ionization-mass spectrometry of cyanobacterial toxins. J. Chromatogr. 1993, 628, 215-233. (15) Hummert, C.; Dahlmann, J.; Reinhardt, K.; Dang, H. Ph. H.; Dang, D. K.; Luckas, B. Liquid chromatographysMass spectrometry identification of microcystins in Microcystis aeruginosa strain from lake Thanh Cong, Hhanoi, Vietnam. Chromatographia 2001, 54, 569-575. (16) Luckas, B.; Dahlmann, J.; Erler, K.; Gerdts, G.; Hummert, C.; Hansen, P. D. Overview of key phytoplankton toxins and their recent occurrence in the North and Baltic seas. Environ. Toxicol. 2005, 20, 1-17. (17) Ortea, P. M.; Allis, O.; Healy, B. M.; Lehane, M.; Shuilleabha´in, A. N.; Furey, A.; James, K. J. Determination of toxic cyclic heptapeptides by liquid chromatography with detection using ultraviolet, protein phosphatase assay and tandem mass spectrometry. Chemosphere 2004, 55, 1395-1402. (18) Di Corcia, A.; Marchese, S.; Samperi, R.; Cecchini, G.; Cirilli, L. Determination of phenol pollutants in water at trace levels: extraction by a reversible graphitized carbon black cartridge. J. AOAC Int. 1994, 77, 446-453. (19) Fastner, J.; Erhard, M.; Carmichael, W. W.; Sun, F.; Rinehart, K. L.; Ronicke, H.; Chorus, I. Characterization and diversity of microcystins in natural blooms and strains of the genera Microcystis and Planktothrix from German freshwaters. Arch. Hydrobiol. 1999, 145, 147-163. (20) Barco, M.; Flores, C.; Rivera, J.; Caixach, J. Determination of microcystin variants and related peptides present in a water bloom of Planktothrix (oscillatoria) rubescens in a Spanish drinking water reservoir by LC/ESI-MS. Toxicon 2004, 44, 881886. (21) Blom, J. F.; Robinson, J. A.; Juttner, F. High grazer toxicity of [D-Asp3,(E)-Dhb7]microcystin-RR of Planktothrix rubescens as compared to different microcystins. Toxicon 2001, 39, 19231932. (22) Hawkins, P. R.; Chandrasena, N. R.; Jones, G. J.; Humpage, A. R.; Falconer, J. R. Isolation and toxicity of Cylindrospermopsis raciborskii from an ornamental lake. Toxicon 1997, 35, 341346. (23) Humpage, A. R.; Falconer, I. R. Oral toxicity of the cyanobacterial toxin cylindrospermopsin in male Swiss albino mice: determination of no observed adverse effect level for deriving a drinking water guideline value. Environ. Toxicol. 2003, 18, 94103.

Received for review December 20, 2005. Revised manuscript received February 28, 2006. Accepted March 13, 2006. ES052546X VOL. 40, NO. 9, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

2923