Environ. Sci. Technol. 2009, 43, 896–901
Kinetics of Microcystis aeruginosa Growth and Intracellular Microcystins Release after UV Irradiation HIROSHI SAKAI,* HIROYUKI KATAYAMA, KUMIKO OGUMA, AND SHINICHIRO OHGAKI Department of Urban Engineering, University of Tokyo, Bunkyo-ku, Tokyo, Japan
Received August 12, 2008. Revised manuscript received November 16, 2008. Accepted November 18, 2008.
The release of intracellular microcystins following ultraviolet (UV) irradiation was studied and modeled. Experimental results indicated that 90 mJ cm-2 of UV fluence were required to inhibit Microcystis aeruginosa growth. The release of intracellular microcystins was also suppressed at higher UV fluence; microcystins concentrations in water did not increase as much in UV-irradiated samples as in controls. A model, based on the following assumptions, was developed to describe the profiles of M. aeruginosa cell number and microcystins concentration in water. Microcystins were contained in M. aeruginosa cells and released only upon cell death. Two types of M. aeruginosa cells existed after UV irradiation; nongrowing cells damaged by UV and growing cells undamaged by UV. To calculate model parameters, these two cell types were counted separately following the addition 0.3 mg L-1 of cephalosporin, a cell wall synthesis inhibitor. Only growing cells are affected by cephalosporin. The model explained the observed data well, suggesting that the model structure was reasonable. The microcystins release model included release from nongrowing and growing cells. The latter declined as M. aeruginosa growth was inhibited by UV. Release from nongrowing cells was delayed, preventing rapid release of microcystins, which could be explained by a larger reaction order of the decay of nongrowing cells. At 600 and 1800 mJ cm-2 UV fluence, intracellular microcystins were decomposed by UV, which led to reduced intracellular microcystins release after UV irradiation.
Introduction The presence of microcystins in water systems is a major threat to human health. Microcystins are contained in Microcystis cells and is released when these cells die. The toxicity of microcystin-LR is severe; the LD50 value is about 50 µg kg-1 for mice, which is lower than that of potassium cyanide (1). Microcystis blooms occur in lakes and reservoirs, creating problems for both drinking water supplies and recreational water use. The World Health Organization microcystins guideline values are 1 µg L-1 in drinking water (2) and 20 µg L-1 in recreational water (3). The largest microcystins outbreak occurred in 1996 in Brazil, where * Corresponding author phone: (81)3-5841-6242; fax: (81)3-58416244; e-mail:
[email protected]. 896
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 3, 2009
microcystins-contaminated tap water was supplied to a hemodialysis center, resulting in the deaths of 88 patients (4, 5). Many other outbreaks of microcystins and other cyanobacterial toxins have occurred; these outbreaks did not result in human deaths, but cases of diarrhea, nausea, and vomiting were reported (6-14). One study indicated that microcystins may promote liver tumors (15). One method to guard against microcystins is to inhibit the growth of Microcystis. Some conventional technologies exist that inhibit Microcystis growth, such as the use of copper sulfate (16, 17) and chlorination (18). However, growth inhibition treatments are always followed by the release of toxins contained in Microcystis cells. Jones and Orr (16) reported that microcystins concentrations rapidly increased from almost 0 to 1 mg L-1 within 3 h of treatment with copper sulfate. Ultraviolet (UV) radiation treatment can inhibit both Microcystis growth and microcystins release. Our previous research (19) demonstrated that UV can inhibit Microcystis growth without causing a significant release of intracellular microcystins. Here we focused on using a model to explain previous research findings and also to examine the effects of UV fluence less than 180 mJ cm-2, which were not applied in previous studies. Release of intracellular microcystins after UV irradiation was investigated experimentally and modeled mathematically. First, Microcystis cell number and microcystins concentration following UV treatment were experimentally investigated. A model was then developed to describe the results. There was some difficulty in modeling two types of Microcystis cells that exhibited different behaviors. Growing intact cells had repaired the UV damage or did not suffer from UV damage. These cells were expected to grow normally and not release microcystins. The other cell type was nongrowing cells, which were damaged by UV. These cells would decay and release intracellular microcystins over time after UV irradiation. These two types of cells were separated and counted to predict the release of microcystins. A new technique using cephalosporin was applied to separate the cells (21, 22). In this manner, model parameters were obtained, which were used in the new model to describe Microcystis cell number and microcystins concentration in water over time after UV irradiation.
Materials and Methods Microorganisms. Axenic cultures of the planktonic bluegreen algae Microcystis aeruginosa (PCC 7806) were obtained from the Pasteur Culture Collection (Institute Pasteur, Paris, France) and grown in BG-11 medium (Sigma, St. Louis, MO). PCC 7806 strain produces Microcystin-LR and the demethylated variants (23). We used a monochromatic lowpressure (LP) UV lamp (15 W × 2, GE/Hitachi, Tokyo, Japan) and a polychromatic medium-pressure (MP) UV lamp (330 W × 1, B410MW, Ebara, Tokyo, Japan). UV irradiation and incubation conditions were same as our previous study (19). Counting Cell Numbers. The number of cells in samples was determined using a fluorescence microscope (BH2, Olympus, Tokyo, Japan) with a plankton-counting chamber (MPC-200, Matsunami Glass, Osaka, Japan), according to Standard Method by Japan Water Works Association (24). Experiments were conducted three times from UV irradiation to subsequent incubation and analysis, and the results are the means of those three experiments. Error bars in the figures show the maximum and minimum of each data set. Intracellular and Extracellular Microcystins Concentrations. Microcystins concentrations were measured im10.1021/es802246x CCC: $40.75
2009 American Chemical Society
Published on Web 01/06/2009
FIGURE 1. Cell number profile of Microcystis aeruginosa PCC 7806 exposed to low-pressure (LP) UV. Values of UV fluence were 0 (O), 30 (b), 60 (4), 90 (2), 120 (0), 180 (9), 600 (]), and 1800 mJ cm-2 ([). Data are the means of three independent exposures to each type of UV lamp; error bars indicate the maximum and minimum values of each data set.
FIGURE 4. Microcystins concentration in water after MP UV. Values of UV fluence were 0 (O), 30 (b), 60 (4), 90 (2), 120 (0), 180 (9), 600 (]), and 1800 mJ cm-2 ([). Data are the means of three independent exposures to each type of UV lamp; error bars indicate the maximum and minimum values of each data set.
TABLE 1. Fitted Parameters for LP and MP UV Samples UV fluence (mJ cm-2) kd (d-1) n(-) k (d-1) MP UV d n(-)
LP UV
0
30 0.30 4 0.30 2
60
90
120
180
600
0.30 0.60 0.60 0.80 0.90 4 70 75 120 25 0.30 0.37 0.50 0.50 0.68 2 4 15 15 1
1800 2.20 1 2.20 1
(Wako, Osaka, Japan). For statistical analysis, t test was performed with three replicates.
Modeling FIGURE 2. Cell number profile of Microcystis aeruginosa PCC 7806 exposed to medium-pressure (MP) UV. Values of UV fluence were 0 (O), 30 (b), 60 (4), 90 (2), 120 (0), 180 (9), 600 (]), and 1800 mJ cm-2 ([). Data are the means of three independent exposures to each type of UV lamp; error bars indicate the maximum and minimum values of each data set.
Model Structure. A mathematical model was developed in two steps; a cell number model was first developed to fit the observed data and then used to develop a microcystins model. Changes in microcystins concentration were mostly from the release from nongrowing cells, although a small portion may have originated from growing cells. X t ) Dt + Gt
(1)
Dt ) D0 · (1 - (1 - e-kdt)n)
(2)
Gt ) G0 · eµt
(3)
(
dDt dM ) m · 10-6 · µ2 · X1 + dt dt
)
(4)
µ ) µ1 - µ2
FIGURE 3. Microcystins concentration in water after LP UV. Values of UV fluence were 0 (O), 30 (b), 60 (4), 90 (2), 120 (0), 180 (9), 600 (]), and 1800 mJ cm-2 ([). Data are the means of three independent exposures to each type of UV lamp; error bars indicate the maximum and minimum values of each data set. mediately before and after UV irradiation, as well as 1, 3, 6, 10, and 14 days (24 ( 2 h) of incubation after UV irradiation. Microcystins concentrations were determined using a microcystin enzyme-linked immunosorbent assay (ELISA) kit
(5)
Xt, total number of cells in the sample (cells mL-1); Dt, number of nongrowing cells in the sample (cells mL-1); Gt, number of growing cells in the sample (cells mL-1); µ, gross specific growth rate (day-1); µ1, net specific growth rate (day-1); µ2, net specific cell decay rate (day-1); kd, specific cell decay rate (day-1); n, reaction order of nongrowing cell decay (-); M: Concentration of microcystins in water in the sample (µgL-1); m, average microcystins content in the cells (fg cell-1). The structure of the cell number model is described in eqs 1, 2, and 3. The cell number of Microcystis is given by the sum of growing cells (Gt) and nongrowing cells (Dt), as in eq 1. The initial value of each parameter at t ) 0 (just after UV irradiation) is expressed as X0, D0, and G0. Dt is determined based on a typical multihit model formula (eq 2), which produces a straight line with a shoulder at n > 1 and without a shoulder at n ) 1. kd is determined by the microscopic VOL. 43, NO. 3, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
897
FIGURE 5. Modeled concentrations of microcystins in water after LP UV. observation of cell number, i.e., the number of cells that disappear. Gt is determined by a simple exponential growth model (eq 3). The structure of the microcystins model is described in eq 4. To calculate the increase of microcystins concentration in the bulk suspension, the average microcystins content m is multiplied by the number of dead cells in the cell number model. The µ2 is a net cell decay rate, defined by eq 5, and the units are converted from fg mL-1 to µgL-1 by multiplying by 10-6. Estimation of Each Parameter Using the Cephalosporin Technique. To develop the model, potentially growing cells (Gt) and nongrowing cells (Dt) needed to be distinguished. This was achieved using the cell wall inhibitor cephalosporin (21, 22). Growing cells fail to divide in the presence of cephalosporin, as it inhibits the synthesis of new cell walls, whereas nongrowing cells are not affected by cephalosporin. As a consequence, growing cells are killed by cephalosporin, and nongrowing cells remain longer than growing cells. Using this technique, the cell number was determined for both types of cells. Cephalosporin was added to a final concentration of 0.3 mgL-1. Two flasks were prepared for this experiment. Two Microcystis suspensions were exposed to the same UV condition and poured into the flasks. Cephalosporin was then added to one of the two flasks, and both flasks were incubated for 14 days. Dt + Gt can be considered the number 898
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 3, 2009
of cells in the flask without cephalosporin. In contrast, Dt is the cell number of the flask incubated with cephalosporin because growing cells are killed by cephalosporin. Subtracting the cell concentration in the latter flask from that in the former flask produces the cell number profile of growing cells Gt. G0 was assumed based on a regression line with a slope factor of µ for all samples. Subtraction of the observed values on day 0 was not included because it took several days for growing cells to begin cell division and subsequently disappear. Among the other parameters, kd and n were estimated from the cell number profile of Dt, and µ2 was estimated from the control results and applied to all samples. The value of m was measured experimentally just prior to UV radiation, and 22.8 fg cell-1 were applied to all but the 600 and 1800 mJ cm-2 UV samples.
Results Number of Cells. Figure 1 presents a profile of the number of M. aeruginosa cells during incubation after LP UV exposure. Initial cell number was set around 1 × 106 cells mL-1 based on the cell number in an actual lake algal bloom (25). The number of M. aeruginosa cells behaved differently depending on the UV fluence. With an LP UV fluence of 30 or 60 mJ cm-2, the cell number profile was not markedly different from that of the control. For samples with an LP UV fluence of 90-180 mJ cm-2, cell number was almost constant up to day 10. From day 10 to day 14, cell number increased due
FIGURE 6. Modeled concentrations of microcystins in water after MP UV. to the growth of remaining cells. In samples exposed to 600 mJ cm-2 LP UV, cell number clearly declined from day 3 to day 6. The number increased again after day 6. Samples exposed to 1800 mJ cm-2 LP UV irradiation had continually declining cell numbers to day 6, after which cell number began to increase. The cell number profiles of the MP samples were similar to those of the LP samples, as shown in Figure 2. Microcystins Concentration in Water. Figures 3 and 4 present profiles of microcystins concentrations in bulk water during incubation after UV exposure. In the control sample, microcystins concentration continued to increase over the 14 days of incubation. The samples exposed to 30 or 60 mJ cm-2 of LP UV exhibited almost the same microcystins profile as that of the control. Concentrations in other LP UVirradiated samples gradually increased after LP UV exposure, but not as much as in the control sample. From day 1 to day 6, some samples exhibited higher microcystins concentrations than the control, but this difference was not significant according to t-test analysis. Microcystins concentrations on day 14 were lower in the samples exposed to 600 and 1800 mJ cm-2 than in the samples exposed to 90, 120, and 180 mJ cm-2 LP UV. The microcystins profiles did not differ much between the MP UV and LP UV samples.
Modeling Determination of Parameters. Parameters were determined for growing and nongrowing cells in the cell number model. For growing cells, µ was set at a constant value, the observed value of the control, for all samples. Of the growing-cell parameters, only G0 was calculated by imposing a regression line on the Gt profile results with a slope of µ. D0 was calculated using eq 1 after determining G0. kd and n were fitted for nongrowing-cell parameters. These two parameters were obtained from the observed results of the Dt profile. If we assume a multihit model, the results should show a gentle curve, followed by a straight line with a slope factor of kd. Therefore, kd was first determined from the Dt profile; n was then determined to fit the Dt profile. Table 1 lists the fitted parameters both for LP and MP UV samples, and SI Tables SI1 and SI2 list the calculated parameter values both for LP and MP UV samples. For the microcystins model, the value of m also needed to be determined, and the average microcystins content in the cell was applied. In most samples, this was experimentally determined to be 22.8 fg cell-1. The values of µ and µ2 were determined to be 0.32 day-1 and 0.02 day-1, respectively, based on the control sample results. Calculated Results. The modeled numbers of cells after UV irradiation was compared to the observed data in SI VOL. 43, NO. 3, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
899
Figures SI1 and SI2. In SI2, the calculated line for the control sample was very close to that of the sample treated with 30 mJ cm-2 UV irradiation. The model results generally fit the observed data very well, suggesting that the model structure applied in this research was appropriate. Figures 5 and 6 show the modeled concentrations of microcystins in water. The modeled concentrations were similar to the measured values, indicating that the model structure employed in this research was also reasonable as a microcystins model. One modification was made to parameter m for samples treated with 600 or 1800 mJ cm-2 UV. A value of 22.8 fg cell-1 was initially employed for m; however, the model clearly overestimated the measured value for these samples. Our previous research (19) indicated that intracellular microcystins were degraded by 600 or 1800 mJ cm-2 of UV irradiation. Therefore, the value of m was only changed for those samples, and the modified parameter well described the measured values.
Discussion Observed Growth Inhibition of M. aeruginosa by UV Irradiation. UV exposure was used in this study to inhibit M. aeruginosa growth. Growth of M. aeruginosa was clearly inhibited by UV, based on the cell numbers on day 10 shown in Figures 1 and 2. From day 10 to day 14, cell number increased due to the growth of the remaining cells. This increase can be attributed to the limited residual effect of UV treatment and the closed experimental system with nutrient-rich media. Based on the cell numbers on day 10 (Figures 1 and 2), UV fluence of 30 and 60 mJ cm-2 did not inhibit M. aeruginosa, whereas fluence of 90 to 180 mJ cm-2 did inhibit growth. Interestingly, cell number on day 10 did not differ much between the samples treated with 90 and 180 mJ cm-2. Also, 90 mJ cm-2 of UV inhibited M. aeruginosa growth, while 60 mJ cm-2 of UV did not. Thus, a threshold value might exist between the two values and should be investigated further. Observed Microcystins Concentrations in Water after UV Irradiation. Microcystins concentrations in water did not increase as much in UV irradiated samples as in the control sample (Figures 3 and 4). In the control, microcystins concentrations in water increased. Because microcystins are cyclic heptapeptide, it cannot pass naturally through the cell membrane. A transportation protein would be required for it to be transported across the membrane, but no such protein has been identified (20). Therefore, microcystins concentrations in water increase only when intracellular microcystins are released from dead cells. The increased microcystins concentration in the control indicates the presence of dead cells in the group of growing cells. Microcystins concentrations on day 14 were lower in samples treated with 600 or 1800 mJ cm-2 of UV fluence than in other samples. This difference may be attributed to intracellular microcystins decomposition by UV, as mentioned previously (19). Behavior of Fitted Model Parameters. Of the seven parameters in this model (µ, kd, n, X0, G0, D0, and m), the fitted parameters were kd and n (Table 1). Both kd and n in this study depended on the UV fluence. The value of kd increased as the UV fluence increased. This is natural considering that kd describes the process of cell destruction. The value of n in the multihit model expresses the degree of winding; a smaller n will enhance the cell decay. The value of n reached a maximum around 120-180 mJ cm-2, whereas it was smaller at higher and lower UV fluence. At fluence around 120-180 mJ cm-2, UV may have had some negative effect on cell metabolism, and this UV damage may have subsequently inhibited cell division; thus cells 900
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 3, 2009
remained for a longer time period. At higher fluence of 600 or 1800 mJ cm-2, UV had an immediate destructive effect. The value of n was smaller for MP UV than for LP UV, indicating that the former had a more destructive effect on cells than the latter. Kinetics of Microcystins Release from UV-Irradiated M. aeruginosa Cells. The kinetics of microcystins release from UV-irradiated M. aeruginosa cells was clarified by both experimental results and mathematical modeling. The microcystins concentration in water, M, was formulated in eq 4. Three characteristics of the UV treatments maintained microcystins concentrations in UV-irradiated water that were not significantly higher than that of the control. First, the growth inhibitory effect suppressed microcystins release from growing cells. This is evident in Figures 3 and 4; the growth of samples treated with 0, 30, and 60 mJ cm-2 of UV fluence was not inhibited, and that of samples treated with at least 90 mJ cm-2 of UV fluence was inhibited. This phenomenon was expressed in the model by the smaller X1. Second, intracellular microcystins were gradually released from nongrowing-cells, as shown in Figure 5f, for example, where the microcystins concentration remained low until around day 3 and suddenly increased after day 6. This was caused by the gradual reduction in the number of Microcystis cells and expressed by a larger n in eq 2. Third, the degradation of intracellular microcystins from nongrowing cells occurred only at 600 and 1800 mJ cm-2 of UV irradiation, as shown in Figures 5g and h and 6g and h, and expressed by a smaller m in the model. The model developed here enabled us to clearly understand these phenomena, and the close fit of the model results with the observed data confirmed the applicability of the model. UV treatment was effective at inhibiting both M. aeruginosa growth and microcystins release.
Acknowledgments We thank Japan Society for Promotion Science for their financial aid to this research.
Supporting Information Available Detailed explanation of UV lamp, incubator, incubation media, measurement of UV intensity, and figures showing modeling result of cell number, details of determined parameter values in tables. This material is available free of charge via the Internet at http://pubs.acs.org.
Literature Cited (1) Dawson, R. M. The toxicology of microcystins. Toxicon 1998, 36, 953–962. (2) World Health Organization. Cyanobacterial Toxins: MicrocystinLR. 95-110. In Guidelines for Drinking Water Quality; WHO: Geneva, Switzerland, 1998. (3) World Health Organization. Algae and cyanobacteria in fresh water. 136-158. In Guidelines for Safe Recreational Water Environments; WHO: Geneva, Switzerland, 2003. (4) Jochimsen, E.; Carmichael, W. W.; An, J.; Cardo, D.; Cookson, S.; Holmes, C.; Antunes, M.; Filho, D.; Lyra, T.; Barreto, V.; Azevedo, S.; Jarvis, W. Liver failure and death after exposure to microcystins at a hemodialysis center in Brazil. New Engl. J. Med. 1998, 338, 873–878. (5) Pouria, S.; Andrade, A.; Barbosa, J.; Cavalcanti, R.; Barreto, V.; Ward, C.; Preiser, W.; Poon, G.; Neild, G.; Codd, G. Fatal microcystin intoxication in haemodialysis unit in Caruaru, Brazil. Lancet 1998, 352, 21–26. (6) Berlin, R. Haff disease in Sweden. Acta Med. Scand. 1948, 129, 560–572. (7) Dillenberg, H. O.; Dehnel, M. K. Toxic waterbloom in Saskatchewan. Can. Med. Assoc. J. 1960, 83, 1151–1154. (8) Byth, S. Palm Island Mystery Disease. Med. J. Aust. 1980, 2, 40–42.
(9) Keleti, G.; Sykora, J. L.; Maiolie,L. A.; Doerfler, D. L.; Campbell, I. M. Isolation and characterization of endotoxin from cyanobacteria (blue-green algae). 447-464. In The Water Environment Algal Toxins and Health; Plenum Press: New York, 1981. (10) Bourke, A. T. C.; Hawes, R. B.; Neilson, A.; Stallman, N. D. An outbreak of hepato-enteritis (the Palm Island Mystery Disease) possibly caused by algal intoxication. Toxicon Suppl. 1983, 3, 45–48. (11) Falconer, I. R. Effects on human health of some toxic cyanobacteria (blue-green algae) in reservoirs, lakes, and rivers. Toxic. Assess. 1989, 4, 175–184. (12) Long, E. G.; Ebrahimzadeh, A.; White, E. H.; Swisher, B.; Callaway, C. S. Alga associated with diarrhea in patients with acquired immunodeficiency syndrome and in travelers. J. Clin. Microbiol. 1990, 28, 1101–1104. (13) Turner, P. C.; Gammie, A. J.; Hollinrake, K.; Codd, G. A. Pneumonia associated with contact with cyanobacteria. Br. Med. J. 1990, 300, 1440–1441. (14) Teixeira, M.; Costa, M.; Carvalho, V.; Pereira, M.; Hage, E. Gastroenteritis epidemic in the area of the Itaparica Dam, Bahia, Brazil. Bull. Pan Am. Health Org. 1993, 27, 244–253. (15) 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, 420– 424. (16) Jones, G. J.; Orr, P. T. Release and degradation of microcystin following algicide treatment of a Microcystis aeruginosa bloom in a recreational lake, as determined by HPLC and protein phosphatase inhibition assay. Water Res. 1994, 28, 871–876. (17) Lam, Angeline K.-Y.; Prepas, E. E.; Spink, D.; Hrudey, S. E. Chemical control of hepatotoxic phytoplankton blooms: Implications for human health. Water Res. 1995, 29, 1845–1854.
(18) Daly, R. I.; Ho, L.; Brookes, J. D. Effect of chlorination on Microcystis aeruginosa cell integrity and subsequent microcystin release and degradation. Environ. Sci. Technol. 2007, 41, 4447– 4453. (19) Sakai, H.; Oguma, K.; Katayama, H.; Ohgaki, S. Effects of lowor medium-pressure UV irradiation on the release of intracellular microcystin. Water Res. 2007, 41, 3458–3464. (20) Pearson, L. A.; Hisbergues, M.; Borner, T.; Dittman, E.; Neilan, B. A. Inactivation of an ABC transporter gene, mcyH, results in loss of microcystin production in the cyanobacterium Microcystis aeruginosa PCC 7806. Appl. Environ. Microbiol. 2004, 70, 6370– 6378. (21) Weil, J.; Miramonti, J.; Ladisch, M. R. Cephalosporin C: Mode of action and biosynthetic pathway. Enzyme Microb. Technol. 1995, 17, 85–87. (22) Sakai, H.; Katayama, H.; Oguma, K.; Ohgaki, S. Growth inhibition of Microcystis aeruginosa and associated release of intracellular microcystin by UV irradiation. Environ. Eng. Res. 2007, 44, 531– 538, in Japanese. (23) Tillett, D.; Dittman, E.; Erhard, M.; Dohren, H.; Borner, T.; Neilan, B. A. Structural organization of microcystin biosynthesis in Microcystis aeruginosa PCC7806: An integrated peptidepolyketide synthetase system. Chem. Biol. 2000, 7, 753–764. (24) Japan Water Works Association. Standard Methods for the Examination of Water; JWWA: Japan, 2001; in Japanese. (25) Park, H.-D.; Iwami, C.; Watanabe, M. F.; Harada, K.; Okino, T.; Hayashi, H. Temporal variabilities of the concentrations of intraand extracellular microcystin and toxic Microcystis species in a hypertrophic lake, Lake Suwa, Japan (1991-1994). Environ. Toxicol. Water Qual. 1998, 13, 61–72.
ES802246X
VOL. 43, NO. 3, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
901
