Species-dependent variation in sensitivity of Microcystis species to copper sulfate: implication in algal toxicity of copper and controls of blooms

Copper sulfate is a frequently used reagent for Microcystis blooms control but almost all the previous works have used Microcystis aeruginosa as the target organism to determine dosages. The aim of this study was to evaluate interspecific differences in the responses of various Microcystis species to varying Cu2+ concentrations (0, 0.05, 0.10, 0.25, and 0.50 mg L−1). The half maximal effective concentration values for M. aeruginosa, M. wesenbergii, M. flos-aquae, and M. viridis were 0.16, 0.09, 0.49, and 0.45 mg L−1 Cu2+, respectively. This showed a species-dependent variation in the sensitivity of Microcystis species to copper sulfate. Malonaldehyde content did not decrease with increasing superoxide dismutase content induced by increasing Cu2+, suggesting that superoxide dismutase failed to reduce Cu2+ damage in Microcystis. Considering the risk of microcystin release when Microcystis membranes are destroyed as a result of Cu2+ treatment and the stimulation effects of a low level of Cu2+ on growth in various species, our results suggest that copper sulfate treatment for Microcystis control could be applied before midsummer when M. aeruginosa and M. viridis are not the dominant species and actual amount of Cu2+ used to control M. wesenbergii should be much greater than 0.10 mg L−1.

(2 mg L −1 ) 20 . Whereas, a recent study reported that a safe Cu 2+ concentration (0.16-0.64 mg L −1 ) would lyse Microcystis cells and release microcystins from the cells 11,21,22 . Therefore, more care should be taken when recommending a safe Cu 2+ dose and more information on the effects of Cu 2+ on growth, physiology, and cell integrity of both toxic and non-toxic Microcystis species is required.
It is noteworthy that almost all of the previous studies used M. aeruginosa as the target organism to determine the Cu 2+ dose for Microcystis control [10][11][12] . However, several species have been recorded in the Microcystis genus 23 and the growth, physiology, and toxicity of Microcystis species varies greatly 24 . The effects of temperature, nutrients, and iron on growth of various Microcystis species differs significantly 25,26 . Moreover, M. wesenbergii, M. flos-aquae, and M. viridis have been reported as the dominant species in lakes besides M. aeruginosa 27,28 . Succession is always observed in these species in lakes 29 . A serious and important question is whether or not we can use the Cu 2+ dose determined from M. aeruginosa to control all Microcystis species in lakes.
Even though species-dependent variation in algal sensitivity to chemical compounds has been widely reported 30,31 , significant differences could not be inferred because the species used in the previous study came from a different genus; however, we are talking about species in the same genus. The aim of this study was to evaluate interspecific differences in the responses of various Microcystis species to varying Cu 2+ concentrations. The efficiency of primary conversion of light energy of PS II (Fv:Fm), cell viability (analyzed by 2,3,5-triphenyltetrazolium chloride (TTC) reduction), superoxide dismutase (SOD), and malonaldehyde (MDA) were analyzed to assess the physiological status of various Microcystis species to varying Cu 2+ concentrations since the mechanisms by which Cu 2+ inhibits growth of Microcystis had been well studied 18,19,21 .

Materials and Methods
Organisms. Four Microcystis species were provided by the Freshwater Algae Culture Collection of the Institute of Hydrobiology, Chinese Academy of Sciences. The identification numbers and origins of the four species are listed in Table 1. All strains were unicellular and purified through the dilution method and axenically cultivated in BG-11 medium for more than 3 months. Experimental design. Each strain was batch-cultured axenically in triplicate in 150 mL sterilized liquid BG-11 medium in a 250-mL conical flask under a 12:12-h light:dark cycle. All cultures were prepared in triplicate. The medium was treated with varying amounts of copper sulfate and the Cu 2+ concentrations were 0.05, 0.10, 0.25, and 0.50 mg L −1 . The culture without copper sulfate treatment was used as the control. The initial cell density of Microcystis was 100 × 10 4 cells mL −1 . The light intensity was 50 μ mol photons m −2 s −1 and the culture time was 4 days. The cell density and efficiency of primary conversion of light energy of PS II (Fv:Fm) was analyzed daily. At the end of the experiment, cell viability (analyzed by 2,3,5-triphenyltetrazolium chloride (TTC) reduction), superoxide dismutase (SOD), and malonaldehyde (MDA) were analyzed.
Cell counting. The cells were counted at least three times in a hemocytometer at 400× magnification with an optical microscope (Olympus CX31; Olympus Corporation, Japan). Counting was stopped when three counts that differed by less than 10% had been obtained. The final cell density was calculated from the average of the three counts.
Biochemical analysis. A 10-mL sample was injected into a 10-mL centrifugal tube. All of the tubes were left undisturbed in the dark at room temperature for 15 min. The sample was then analyzed by AquaPen-P 100 (Photon Systems Instruments, Czech Republic) to determine the Fv:Fm value. TTC, SOD, and MDA were analyzed according to the methods described by Hong et al. 32 , Choo et al. 33 , and Kong et al. 34 , respectively. Data analysis. All of the data are presented as mean ± SD. The specific growth rate was calculated by where D t is the cell density at time t, D 0 is the cell density in the initial logarithmic growth phase, and t is the duration of the logarithmic growth phase. In the current study, the value of t was 4. The half maximal effective concentration (EC 50 ) was determined on day 4 at the 50% inhibition rate according to the relationship between inhibition rate and concentration of Cu 2+ . The inhibition rate was calculated by equation (2): where D 4,0 is the cell density in the control on day 4, D 4,c is the cell density on day 4 treated with concentration c of Cu 2+ .   Microcystis SOD and MDA content. In the control, SOD content in M. wesenbergii was greater than that in the other species (Fig. 4). With increasing Cu 2+ concentration, SOD content significantly increased in all four species (P < 0.05). In the highest Cu 2+ treatment, SOD content was 9.    (Fig. 2), which was consistent with the order of EC 50 . It can be seen that Fv:Fm ratio of M. aeruginosa decreased with increase of Cu 2+ concentration but Fv:Fm ratio of M. flos-aquae and M. Wesenbergii decreased in first 2 days of application of copper sulfate, and then increased as the days progress. For M. viridis, the ratio was at par with the control in all the treatments. These differences would be because of variations in sensitivity to Cu 2+ among various species. Although, it was reported that high level Cu 2+ reduced the electron transfer rate of the PS II system in M. aeruginosa 22,36 , inhibition of PS II system may not be the only way by which copper sulfate controls other Microcystis species. The decrease in the relative TTC reduction reflected cell damage exposed to Cu 2+ . Our results showed that cells of all the Microcystis species were damaged when exposed to Cu 2+ except for M. viridis (Fig. 3). This result was consistent with the results of growth and Fv:Fm ratio.

Microcystis
SOD may be crucial to the growth inhibition of Microcystis 37,38 . M. wesenbergii had the highest SOD content in the control compared with the other Microcystis species (Fig. 4). However, the M. wesenbergii EC 50 value was the lowest. Moreover, M. aeruginosa, M. flos-aquae, and M. viridis EC 50 values varied greatly but their SOD contents were similar. The MDA content was considered an indicator of cell injury and increasing MDA indicated damage of cytomembrane 39 . In the current study, the values did not decease with increasing SOD content induced by increasing Cu 2+ . All of the above results suggest that SOD failed to reduce Cu 2+ damage in Microcystis in the current study. Both enzymatic and non-enzymatic antioxidants of Microcystis played important roles in tolerating oxidative damage 37,38 . Therefore, non-enzymatic antioxidants such as reduced glutathione (GSH) and ascorbic acid (AsA) would be important for Microcystis spp. to counteract the oxidative stress induced by Cu 2+ 38 .
The initial M. wesenbergii SOD and MDA content was significantly higher than that of the other species. Temperature may have affected this, given that the optimal temperature for M. wesenbergii growth is approximately 30 °C 25 and the temperature in our experiment was much lower (25 °C). Both SOD and MDA content increased with increasing concentrations of Cu 2+ in the current study. Similar result was also reported by Chen et al. 39 and Shao et al. 40 . However, it was considered that SOD and MDA had an rough inverse relationship 38 . It might be because that MDA was a continuously accumulating material but SOD varied against time 41 . As shown  in Fig. 2, damage from Cu 2+ was highest on first day and then the physiological activity was improving later. Therefore, the relationship between SOD and MDA on day 4 was irregular.
Extracellular polysaccharide (EPS) release is another protective response against chemical compounds in algae including microcystins 42 , salt 43 , and heavy metals [44][45][46] . Li et al. 47 suggested that EPS is an important strategy to reduce Cu 2+ damage because -COO − and some amino groups in EPS can absorb heavy metals effectively 48 . Xu et al. 49 demonstrated that EPS content was significantly lower in M. wesenbergii than the other three Microcystis species under standard culture conditions similar to ours. It could be deduced from their results that M. wesenbergii was the most sensitive species. This conclusion is also supported by our results (Table 2). Therefore, the species-dependent variation in the sensitivity of Microcystis species to copper sulfate in the current study may have been the result of variations in EPS content in different Microcystis species. Forni et al. 50 reported that the content of polysaccharide of M. viridis was much higher than other Microcystis species. This difference would cause variation in growth and physiology of Microcystis when treated with copper sulfate. It was noticed that the standard deviation of cell density obtained with M. viridis was much higher than other species and this result supported above deduction. In addition, it was also found that the standard deviation of TTC, SOD and MDA obtained with M. viridis was very high. Polysaccharide may be the main interfering substance for analysis of above enzyme activity. However, the error range was still within the equivalent range reported by some other researchers 37,39,40 .
The growth curves of M. flos-aquae and M. viridis exposed to 0.05 and 0.10 mg L −1 Cu 2+ were higher even than control (Fig. 1). This was due to that Cu 2+ are essential micronutrient for Microcystis 51 . Additionally, it has also been well documented that low-level contaminants promote Microcystis growth [52][53][54] . However, the sensitivity or tolerance to heavy metals varies amongst different algae and this variation caused that the beneficial concentration of Cu 2+  In addition, the species of copper from different copper algaecides should also be considered 56 .
The risk of microcystin release when Microcystis membranes are destroyed by Cu 2+ treatment is an important concern in the application of copper sulfate to control Microcystis. Trace Cu 2+ (0.16-0.50 mg L −1 ) 10,11,21 can result in cell lysis and microcystin release. The most sensitive species, M. wesenbergii, potentially produces microcystins and other toxins 57,58 . This species always dominates in summer in lakes with high biomass 27 . Nevertheless, it has been reported as a non-microcystin production species in China and other countries 59,60 . Therefore, 0.10-0.16 mg L −1 Cu 2+ would reduce M. wesenbergii growth without cell lysis and microcystin release. Additionally, this Cu 2 + dose would not promote growth in other Microcystis species.
M. aeruginosa and M. viridis colonies may produce large amounts of microcystins 58,61 . The EC 50 values of these two species to Cu 2+ in this study were 0.16 and 0.45 mg L −1 , respectively. These concentrations may induce cell lysis and microcystin release. Reports of toxin production are rare in M. flos-aquae (sometimes identified as M. ichthyoblabe) 25,57 . This species always dominates in lakes before early summer and the biomass is lower compared with M. wesenbergii in midsummer 62 . M. flos-aquae growth would be inhibited by 1 mg L −1 Cu 2+ , which is safe for drinking water and there is no risk of microcystin release. Therefore, copper sulfate treatment for Microcystis control could be applied before midsummer when M. aeruginosa and M. viridis are not the dominant species. The dose of copper sulfate should be evaluated according to the dominant Microcystis species.
The EC 50 value is also affected by initial cell density 63,64 and Microcystis phenotype 36,47 . In the current study, all of the strains were unicellular and the initial cell density was the same. The effects of initial cell density and Microcystis phenotype on EC 50 were excluded. However, the EC 50 values obtained in the current study should be re-evaluated because Microcystis always exists as colonies in lakes with varying cell densities 65 . The effects of temperature on heavy metal tolerance in Microcystis should also be considered 66 .

Conclusions
Our results reveal species-dependent variation in the sensitivity of Microcystis species to copper sulfate. The 96-h-EC 50 value of Microcystis species to Cu 2+ was in the order of M. flos-aquae > M. viridis > M. aeruginosa > M. wesenbergii and ranged from 0.09 to 0.49 mg L −1 . MDA content did not decrease with increasing SOD content induced by increasing Cu 2+ , suggesting that SOD failed to reduce Cu 2+ damage to Microcystis in the current study. However, the species-dependent variation in the sensitivity of Microcystis species to copper sulfate may have resulted from variations in EPS content in different Microcystis species. M. flos-aquae and M. viridis growth were promoted when exposed to 0.05 and 0.10 mg L −1 Cu 2+ . Our results suggest that copper sulfate treatment for Microcystis control could be applied before midsummer when M. aeruginosa and M. viridis are not the dominant species and the actual amount of Cu 2+ used to control M. wesenbergii should be much greater than 0.10 mg L −1 , while M. flos-aquae and M. viridis control in lakes requires further investigation.