OUP user menu

Dynamics of microcystin-producing and non-microcystin-producing Microcystis populations is correlated with nitrate concentration in a Japanese lake

Mitsuhiro Yoshida , Takashi Yoshida , Yukari Takashima , Naohiko Hosoda , Shingo Hiroishi
DOI: http://dx.doi.org/10.1111/j.1574-6968.2006.00496.x 49-53 First published online: 1 January 2007


Temporal changes in hepatotoxin microcystin-producing and non-microcystin-producing Microcystis aeruginosa populations were examined in Lake Mikata, Japan. To monitor the densities of the total M. aeruginosa population and the potential microcystin-producing subpopulation, we used a quantitative real-time PCR assay targeting the phycocyanin intergenic spacer and the microcystin synthetase gene (mcyA), respectively. During the sampling period, the ratio of the mcyA subpopulation to the total M. aeruginosa varied considerably, from 0.5% to 35%. When surface nitrate concentrations increased, there was a rise in the relative abundance of the mcyA subpopulation. This was a positive correlation with the nitrate concentrations (r=0.53, P<0.05, n=14); whereas temperature and ortho-phosphate had no significant correlation with the presence of mcyA. Our data suggest that high nitrate loading may be a significant factor promoting the growth of the microcystin subpopulations within M. aeruginosa communities in Lake Mikata.

  • Microcystin
  • Microcystis
  • peptide synthetase
  • population dynamics
  • real-time PCR
  • toxic cyanobacteria


Cyanobacteria frequently form blooms in eutrophic lakes, ponds and reservoirs. A major bloom-forming component, the genus Microcystis, can produce a potent hepatotoxin called microcystin (Carmichael, 1995). Microcystin specifically inhibits eukaryotic protein phosphatase types 1 and 2A (MacKintosh et al., 1990; Yoshizawa et al., 1990). There have been several reports of deaths in wild and domestic animals as well as humans due to this acute poisoning, which causes massive hepatic hemorrhage (Beasley et al., 1989; Jochimsen et al., 1998; Pouria et al., 1998).

Previously, individual species within the unicellular Microcystis group were determined on the basis of morphological features observed microscopically, such as cell size and cell arrangement in colonies (Komárek & Anagnostidis, 1986; Komárek, 1991); however, these features change during cultivation (Otsuka et al., 2000). Therefore, recently, the Microcystis species have been combined as M. aeruginosa according to bacteriological taxonomic criteria (Kondo et al., 2000; Otsuka et al., 2001). Several attempts have been made to link microcystin production to morphological characteristics or molecular phylogeny (Otsuka et al., 1999; Tillett et al., 2001); however, none of these analyses revealed a correlation with toxicity.

The gene cluster responsible for microcystin biosynthesis has been characterized. Its products are peptide synthetases and polyketide synthases, a putative ABC transporter, and tailoring enzymes (Nishizawa et al., 1999, 2000; Tillett et al., 2000). Both microcystin-producing and non-microcystin-producing strains can be isolated from the same water source (Ohtake et al., 1989; Vézie et al., 1998), and reports show the composition of microcystin-producing and non-producing populations within a M. aeruginosa species change over time (Kurmayer & Kutzenberger, 2003; Janse et al., 2004). In addition, using a quantitative competitive PCR with selective primers targeting the microcystin synthetase gene, mcyA (Yoshida et al., 2003), a temporary disappearance of the mcyA subpopulation was observed during M. aeruginosa blooms (Yoshida et al., 2005). No studies have been published concerning the effect of environmental factors on the internal dynamics of the total communities in fresh water.

Therefore, the aim of this study was to determine whether there are environmental factors controlling temporal changes in microcystin-producing and non-producing M. aeruginosa populations in Lake Mikata. We performed two independent real-time PCR assays to evaluate the internal dynamics of M. aeruginosa blooms. One assay was used to quantify the total M. aeruginosa population using the phycocyanin intergenic spacer (PC-IGS) that was previously used to examine the total M. aeruginosa numbers (Kurmayer & Kutzenberger, 2003). A second real-time PCR assay was used to quantitatively detect the potentially microcystin-producing genotypes using mcyA-specific primers. We monitored the relative size of the microcystin subpopulation compared with the total population in relation to various environmental factors through a field survey of M. aeruginosa blooms in Lake Mikata.

Materials and methods

Lake water sampling

Sampling was performed at a fixed point in Lake Mikata (35°33′ N, 135°53′ E) from July 2004 to October 2005. Water samples from the surface layer (0.5 m) were collected once or twice per month. In situ water temperature was determined using a digital meter (YSI Model 85; Yellow Springs Instrument, Yellow Springs, OH). To determine the major dissolved inorganic nutrients, nitrate (NO3–N) and orthophosphate (PO4–P), 50 mL of water sample was filtered through a glass fiber filter (GF/F; Whatman, Maidstone, Kent, UK) and analyzed using an autoanalyzer (TRAACS 2000; BRAN+LUEBBE, Elmsford, NY) according to the manufacturer's instructions. For quantitative PCR, a 100-mL sample was sonicated (47 Hz, 300 W) to remove gas vesicles and the cells were harvested by centrifugation at 14 400 g for 10 min.

DNA extraction

DNA isolation was performed using a combination of the potassium xanthogenate-sodium dodecyl sulfate and phenol/chloroform/isoamylalcohol procedures as described previously (Tillett & Neilan, 2000; Yoshida et al., 2005). Purified DNAs were suspended in 30 µL of TE buffer.

Quantitative real-time PCR

The real-time PCR assay was used to quantify two genetic elements, the PC-IGS and mcyA regions. For the PC-IGS, the primers used were 188F (5′-GCT ACT TCG ACC GCG CC-3′) and 254R (5′-TCC TAC GGT TTA ATT GAG ACT AGC C-3′), which are specific for M. aeruginosa as described previously (Kurmayer & Kutzenberger, 2003). Microcystis aeruginosa mcyA gene-specific primers were used: M1r-F (5′-AGC GGT AGT CAT TGC ATC GG-3′) and M1r-R (5′-GCC CTT TTT CTG AAG TCG CC-3′) were designed using the mcyA sequence of M. aeruginosa strains NIES298 and NIES102 (DDBJ/EMBL/GenBank accession numbers AB092804 and AB092805, respectively) with the clustalw version 1.7 programs (Thompson et al., 1994). The primer sequence specificities were confirmed using a BLAST search of the DDBJ/EMBL/GenBank databases (Altschul et al., 1997).

External standards used to determine the PC-IGS and mcyA copy numbers were prepared using the genomic DNA from M. aeruginosa strain NIES298. Copy numbers of both genes were estimated on the basis of DNA fragment sizes (66 bp for the PC-IGS and 107 bp for mcyA). A 10-fold dilution series of the DNAs was prepared and amplified with the PC-IGS gene and mcyA real-time PCR assays. For both genes, linear regression equations to obtain cycle threshold (Ct) values were calculated as a function of the known DNA copy numbers.

The real-time PCR assay was performed in a volume of 50 µL containing 25 µL 2 × SYBR Green Realtime PCR Master Mix (Toyobo, Osaka, Japan), 0.5 pmol each primer, 1 µL DNA from a standard strain or lake water sample, and filled up to 50 µL with sterile ultra-pure water. For mcyA, 0.2 pmol of each primer was used. The PCR was conducted with an iCycler iQ Real Time PCR Detection System (Bio-Rad Laboratories, Hercules, CA) using the icycler iq optical system Software (version 3.0a) (Bio-Rad Laboratories). Amplification was performed as follows: The first step was an initial preheating for 1 min at 95°C. For PC-IGS, the initial preheating step was followed by 35 cycles: 95°C for 30 s, 60°C for 30 s, and 84°C for 30 s. For mcyA, the initial preheating step was followed by 35 cycles: 95°C for 30 s, 65°C for 30 s, and 83°C for 30 s. The melting temperature for the real-time PCR products was determined using the manufacturer's software. All tests were performed in triplicate.

Statistical analysis

The relative abundance of the mcy subpopulation within the M. aeruginosa population was log transformed to normalize the data, and linear correlations with temperature, NO3–N concentration (log NO3–N), and PO4–P concentration (log PO4–P) were determined by the correlation coefficient (r) with Microsoft Excel.

Results and discussion

The real-time PCR data show log linear relationships using both the PC-IGS and mcyA gene copies when the genomic DNA from M. aeruginosa NIES298 was used as a template (Fig. 1). The detection range of PC-IGS copy numbers was from 3.4 × 101 to 3.4 × 108 copies in the reaction mixture; and the detection range of mcyA copy numbers was from 7.5 × 101 to 7.5 × 108. The lowest detectable numbers of PC-IGS and mcyA gene copies were 11 and 25 copies mL−1, respectively.


Standard curves obtained by the PC-IGS gene and mcyA real-time PCR assays with the Microcystis aeruginosa strain NIES298 as a function of gene copy numbers. Each data point shows the threshold cycle (Ct) of standard DNA samples performed in triplicate. Amplification efficiency was calculated as follows: e=10−1/S−1, where S is the slope. Error bars represent the standard deviations.

With the real-time PCR assays, the densities of the total M. aeruginosa population and the potential microcystin-producing subpopulation were examined for the sampling period during 2004 and 2005. Melting curves of all of the environmental the PC-IGS and mcyA real-time PCR products showed a peak at approximately 89°C and 86°C, respectively, corresponding to the melting temperature of the M. aeruginosa standard strain (data not shown). The environmental real-time PCR data are shown in Fig. 2. Throughout the field survey, copy numbers of all M. aeruginosa genotypes with the PC-IGS gene varied between 7 × 102 and 1 × 106 copies mL−1. The subpopulation with the mcyA genotypes was found in every sample at 2.6 × 101 to 1 × 104 copies mL−1. In 2004, the ratio of the copy numbers of mcy genotypes to the total M. aeruginosa was 0.5–7.7%. The total variation in the relative abundance of microcystin genotypes in 2005 was larger at 0.8–35%.


Abundance of the total Microcystis aeruginosa and the mcy subpopulation in Lake Mikata during the growth periods of 2004 and 2005. DNA copies per milliliter were determined using real-time PCR. The percentage of mcy is the relative DNA copy numbers of the mcy subpopulation compared to the total copy number of M. aeruginosa determined with primer set 188F/254R. Error bars represent the standard deviations in triplicate.

In previous studies, the mcyA quantitative competitive PCR data showed that the proportion of microcystin genotypes to the total M. aeruginosa varied considerably in Lake Mikata (Yoshida et al., 2003; Yoshida et al., 2005). As the population size of the total M. aeruginosa was enumerated by direct microscopic observation, which depends on morphological characteristics, non-colony-forming M. aeruginosa numbers were not obtained, and therefore these count densities may have been low (Yoshida et al., 2003; Yoshida et al., 2005). The detection of all M. aeruginosa cells using an indirect fluorescent antibody technique with a M. aeruginosa-specific antiserum demonstrated that the numbers of M. aeruginosa cells detected using immunological probes were twice as high as those observed using simple microscopy in Lake Mikata, suggesting that the direct microscopic conventional counting method is not necessarily accurate enough to determine the environmental M. aeruginosa dynamics (Hiroishi et al., 2004). As shown in Fig. 2, our field surveys using two independent real-time PCR assays found that the mcyA genotype proportions fluctuated greatly, between 0.5% and 35%. Similarly, Kurmayer & Kutzenberger (2003) show temporal dynamics of 1–38% for the occurrence of a microcystin synthetase gene, mcyB, in a natural M. aeruginosa population in Lake Wannsee. As an alternative approach, we used the ITS genotyping to identify potentially microcystin-producing and non-producing genotypes in Lake Mikata, and the microcystin genotypes were also detected when the relative abundance of the mcyA subpopulation was relatively high (M. Yoshida et al., submitted for publication). In addition, Janse (2004) report ITS DGGE profiles of M. aeruginosa communities from a Dutch lake, which showed the different relative abundance of strains containing the microcystin and mcyB between bloom stages. These observations suggest there is an internal dynamic at the intra-species level during a M. aeruginosa bloom succession.

The hydrological data collected in this study are shown in Table 1. During the study period, the surface water temperature varied between 12°C and 32°C, with an average of 24°C. The surface PO4–P and NO3–N concentrations fluctuated; 0.003–0.032 and 0.004–0.449 mg L−1, respectively. There was a rapid increase in NO3–N concentration during the second half of September 2004, when concentrations rose from 0.011 to 0.449 mg L−1 and the relative abundance of the subpopulation with mcyA genotypes was 12 times (5 October) greater than at the 15 September sampling (Table 1). In 2005, NO3–N increased to a peak of c. 0.118 mg L−1 on 23 August, which also coincided with a rapid rise (8.5–35%) in the relative abundance of microcystin genotypes. The relative abundance of the microcystin subpopulation was significantly positively correlated with NO3–N concentrations (r=0.53, P<0.05, n=14). The other physicochemical factors, temperature and PO4–P, during the survey were also considered in the correlation analysis (Table 1); however, we found no significant correlation (r=−0.30 and r=0.25, respectively). Our data suggest high NO3–N loading may be a significant factor in increasing the growth of microcystin genotypes within the M. aeruginosa communities of Lake Mikata.

View this table:

Sampling date and physicochemical parameters in Lake Mikata

Sampling date (month/day/year)Temperature (°C)PO4–P (mg L−1)NO3–N (mg L−1)%mcyA subpopulation
  • Percentage of the DNA copy numbers of the mcyA subpopulation to the total Microcystis aeruginosa as determined by the real-time PCR assay (Fig. 2).

Kurmayer & Kutzenberger (2003) reported that the relative abundance of mcyB genotypes was seasonally stable during the M. aeruginosa succession from the lowest densities in winter to the highest densities in summer and did not depend on the thermal influence during the study period. Here, our data also show no significant correlation was observed between water temperature and relative abundance of microcystin genotypes in Lake Mikata (Table 1). Therefore, the relative abundance of this subpopulation probably does not follow temperature trends, whereas increasing NO3–N and relative increased abundance of microcystin genotypes in Lake Mikata does. The relationship between PO4–P and relative abundance of microcystin genotypes was not observed. It has been demonstrated that lower nutrient concentrations favor the growth of two non-microcystin-producing M. aeruginosa strains instead of two microcystin-producing ones, whereas the microcystin-producing strains grew better at high nutrient concentrations than the non-producing ones (Vézie et al., 2002). More isolate studies are required to confirm whether both populations have the different ecological traits. The regulation of the internal dynamics of M. aeruginosa populations is anticipated to be complicated by a broad range of environment conditions including nutrient supply. The biological function(s) of microcystin are currently unknown; however, there are many theories including siderophoric scavenging of and binding to trace metal (such as iron) and quorum sensing (Utkilen & Gjølme, 1995; Dittmann et al., 2001). We isolated a cyanophage from Lake Mikata that specifically infects only a microcystin-producing M. aeruginosa strain (Yoshida et al., 2006). These other factors may also influence seasonal shifts in microcystin-producing and non-producing M. aeruginosa populations and should be considered along with the nutrient content, especially biotic factors like the intraspecific selective lysis caused by phage.


This study was partly supported by Research Fellowships of the Japan Society for the Promotion of Science for Young Scientists (JSPS Research Fellowships for Young Scientists), Japan.


  • Editor: Aharon Oren


View Abstract