Can Alkyl Quaternary Ammonium Cations Substitute H2O2 in Controlling Cyanobacterial Blooms—Laboratory and Mesocosm Studies

Mitigation of harmful cyanobacterial blooms that constitute a serious threat to water quality, particularly in eutrophic water, such as in aquaculture, is essential. Thus, in this study, we tested the efficacy of selected cyanocides towards bloom control in laboratory and outdoor mesocosm experiments. Specifically, we focused on the applicability of a group of cationic disinfectants, alkyltrimethyl ammonium (ATMA) compounds and H2O2. The biocidal effect of four ATMA cations with different alkyl chain lengths was evaluated ex situ using Microcystis colonies collected from a fish pond. The most effective compound, octadecyl trimethyl ammonium (ODTMA), was further evaluated for its selectivity towards 24 cyanobacteria and eukaryotic algae species, including Cyanobacteria, Chlorophyta, Bacillariophyta, Euglenozoa and Cryptophyta. The results indicated selective inhibition of cyanobacteria by ODTMA-Br (C18) on both Chroccocales and Nostocales, but a minor effect on Chlorophytes and Bacillariophytes. The efficacy of ODTMA-Br (C18) (6.4 μM) in mitigating the Microcystis population was compared with that of a single low dose of H2O2 treatments (117.6 μM). ODTMA-Br (C18) suppressed the regrowth of Microcystis for a longer duration than did H2O2. The results suggested that ODTMA-Br (C18) may be used as an effective cyanocide and that it is worth further evaluating this group of cationic compounds as a treatment to mitigate cyanobacterial blooms in aquaculture.


Introduction
Excess input of nitrogen, phosphorus and other nutrients make cyanobacteria proliferate exceptionally. The massive accumulation of cyanobacterial biomass can damage water quality [1]. Some cyanobacteria are known to produce toxic secondary metabolites, such as microcystins, which can be harmful to zooplankton, fish and even humans [2,3]. In fish aquaculture systems, extensive feeding protocols lead to water hyper-eutrophication that facilitates frequent and persistent occurrence of cyanobacterial harmful algal blooms (cyanoHABs). Continuous pond aeration is usually required to eliminate hypoxia or anoxic conditions resulting from large amounts of photosynthetic and respiratory activities (respectively) caused by large water blooms, as well as the associated additional costs in Microorganisms 2021, 9,2258 3 of 14 by fluorescent lamps (Philips, The Netherland). ATMA-Br compounds were purchased from Aladdin (Shanghai, China) and H 2 O 2 (30%) from Sinopharm (China). The chemical formulas of the ATMA compounds used in this study and their molecular weights are shown in Table 1. Each ATMA-Br compound was freshly prepared using a stock solution in distilled water. Table 1. Chemical properties of the four ATMA-Br used in this study. Microcystis colonies were collected from a fish pond on 23 September 2020. Microscopic observation showed that Microcystis viridis and Microcystis wesenbergii were dominating. The geographical location of the fish pond and mesocosm set up are presented in Figure S1.

Alkyl Chain Length and Inhibition of Photosynthesis in Microcystis Colonies
In this experiment, we assessed the effect of ATMAs with various alkyl chain lengths (from C12-C18) on Microcystis colonies collected from a fish pond. The colonies were diluted with 1/10 BG11 to a final chlorophyll a (Chl a) concentration of 300 µg/L medium in 25 cm 3  The cyanocidal effect of ODTMA-Br (C18) was compared to that of H 2 O 2 , a commonly used cyanocide, applying the same experimental setup. The selected ODTMA-Br (C18) concentrations were 1, 2, 6.4 and 10 µM, whereas the H 2 O 2 concentrations used were 36.8, 73.5, 147.1 and 294.1 µM. The photosynthetic activity (Fv/Fm) was measured 6, 24 and 96 h after each treatment.

Exploring the Algaecidal Selectivity of ODTMA-Br (C18)
A total of 24 strains: 12 Cyanobacteria, 7 Chlorophyta, 2 Bacillariophyta, 2 Euglenozoa and 1 Cryptophyta were selected and cultured in six-well plates. Information on these strains is shown in Table 2. The concentrations of the ODTMA-Br (C18) applied were 0 (control), 2 and 4 µM. The photosynthetic parameters: Fo (initial fluorescence yield), and Fm (maximum fluorescence yield), were measured (after 15 min dark adaptation) at 6, 24 and 96 h post the addition of ODTMA-Br (C18) to the wells. The concentrations of Chl a were measured at 96 h. The Fv/Fm (maximum photochemical efficiency) values of the treatment group were compared with those of the control group, and the inhibition ratio of Fv/Fm (calculated by the following formula) was used to assess the ability of ODTMA-Br (C18) to inhibit algal photosynthesis. The higher the ratio, the larger the suppression effect.

Mesocosm Experiment
Microcystis colonies collected from a fish pond were first pooled in a 1500 L barrel for a day of acclimation. On day-0 of the mesocosm experiment, aliquots of the Microcystis colonies were placed into 200 L barrels with a final Chl a concentration of 300 µg/L. In order to simulate the original growth conditions of the collected colonies, the barrels were placed approximately 10 m from the fish pond ( Figure S1). ODTMA-Br (C18) (6.4 µM) or H 2 O 2 (117.6 µM) was added to the relevant barrels. Three untreated barrels were included as controls. The photosynthetic activity was measured on days 1, 3, 5, 7, 9, 13, 15 and 28 and the phytoplankton density and composition were analyzed on days 0, 1, 3, 5, 7, 9, 13, 15 and 28 after the addition of ODTMA-Br (C18) or H 2 O 2 . Withdrawn cell suspensions (5 mL) were filtered through a Whatman GF/F membrane. Pigments were extracted in a 10 mL tube with 80% chilled acetone and kept in darkness for 12 h. Then, the tubes were centrifuged at 4 • C for 15 min at 6000× g. The optical densities of the extracts at 646, 663 and 750 nm were determined using a UV-VIS spectrophotometer (UV-1700, Shimadzu). The concentration of Chl a was calculated according to the method from Lichtenthaler and Wellburn [24].

Measurement of Chl a Fluorescence Transient
To test the adverse effect on photosynthetic processes, the polyphasic Chl a fluorescence transient (the fluorescence induction-decay curve, also known as the Kautsky effect) was determined with a Handy PEA with an actinic light of 3000 µmol photos m −2 s −2 (excitation light, λ = 650 nm). The measuring light intensity is 10% of the actinic light (300 µmol photos m −2 s −2 ). All samples were dark acclimated for 15 min before measurement. The fluorescence signals were recorded within a time period from 10 µs to 2 s, and fluorescence kinetics showed a polyphasic rise over time, known as the O-J-I-P curve [25]. The initial fluorescence level O corresponds to the minimal Chl a fluorescence value (Fo). The J-I transient is caused by the gradual reduction of the primary electron acceptors, QA and QB. The P value is the maximal fluorescence yield (F m ). The calculated parameters, based on the O-J-I-P fluorescence curve, are presented in Table S1 and as a radar plot in Figure S2.

Determination of Phytoplankton Community
Phytoplankton samples (100 mL) from the mesocosm experiments were collected after mixing and immediately fixed with a 1% Lugol solution. The samples were quantified using a counting chamber (0.1 mL) under an optical microscope (Olympus CX41) at 400× magnification. Phytoplankton species were identified according to Hu [26]. The wet weight biomass of each species was calculated according to their morphometric characteristics [27].

Data Analyses
Basic statistical analyses were carried out using the SPSS 25.0 software. One-way analysis of variance (ANOVA) with a multiple post hoc (LSD) test (significance level is 0.05) was used to check the significant difference between the various treatments. All the data are presented as the mean of three replicates. Results visualization was performed using GraphPad Prism 8.0. The Bray-Curtis distance and PerMANOVA analyses were performed through package vegan using R (version 3.6.1).

The Effect of the Alkyl Chain Length on Photosynthetic Activity
It was reported that ATMA cations with longer alkyl chain exhibit stronger cyanocidal effect on unicellular Microcystis cultures [21]. To examine whether this is also the case with Microcystis colonies, the effectiveness of ATMA with alkyl chain length (C12 to C18) was examined using colonies collected in fish ponds. The photosynthetic parameters are shown in Figure 1 and Figure S2. After 3 h, ODTMA inhibited PSII activity stronger than the other ATMA compounds used. The PSII parameters Sm, TRo/CSo and ETo/CSo showed concentration-dependent decreases. The changes of these parameters indicated that the photochemical activity of Microcystis was suppressed, and the effective utilization rate of light energy decreased, resulting in the inhibition of photosynthesis. The alkyl chain length of ATMA refers to a biocidal effect towards different microorganisms. Gilbert and Moore [28] reported that molecular weight and N-alkyl chain length affect the efficacy of many QAC-based antimicrobial systems. The optimum alkyl chain length for killing yeast and filamentous fungi is C12, whereas it was C14 and C16 for gram-positive and gram-negative bacteria, respectively [28]. For cyanobacteria, Wu et al. [21] reported that the effect of ODTMA was strongest among ATMAs with an alkyl chain length from C10-C18. Similarly, to inactivate Microcystis colonies, ODTMA is more effective in our study.  Figure S2. Error bars indicate the standard deviation (n = 3). The asterisk indicates statistically significant differences in comparison with the control: * p < 0.05; ** p < 0.01; *** p < 0.005; **** p < 0.001.

The Algaecidal Effect of ODTMA-Br (C18) on Cyanobacteria and Eukaryotic Algae
To investigate the algaecidal impact and selectivity of ODTMA-Br (C18) towards various cyanobacteria and eukaryotic algae species, we applied two ODTMA-Br (C18) concentrations to 24 species of algae that were commonly observed species in freshwater ecosystems, including fish ponds. Two μM of ODTMA-Br (C18) suppressed the activity of five different cyanobacteria within 6 and 24 h but doubling the concentration to 4 μM inhibited the activity of 12 different cyanobacteria ( Figure 2). In contrast, the ODTMA-Br (C18) concentration used hardly affected the photosynthetic activities (Fv/Fm) of Chlorophyta, Bacillariophyta, Euglenozoa and Cryptophyta ( Figure 2). One-way analysis of variance (ANOVA) with a multiple post hoc (LSD) test (significance level is 0.05) was used to check the significant difference between the various treatments. The results show that  Figure S2. Error bars indicate the standard deviation (n = 3). The asterisk indicates statistically significant differences in comparison with the control: * p < 0.05; ** p < 0.01; *** p < 0.005; **** p < 0.001.

The Algaecidal Effect of ODTMA-Br (C18) on Cyanobacteria and Eukaryotic Algae
To investigate the algaecidal impact and selectivity of ODTMA-Br (C18) towards various cyanobacteria and eukaryotic algae species, we applied two ODTMA-Br (C18) concentrations to 24 species of algae that were commonly observed species in freshwater ecosystems, including fish ponds. Two µM of ODTMA-Br (C18) suppressed the activity of five different cyanobacteria within 6 and 24 h but doubling the concentration to 4 µM inhibited the activity of 12 different cyanobacteria ( Figure 2). In contrast, the ODTMA-Br (C18) concentration used hardly affected the photosynthetic activities (Fv/Fm) of Chlorophyta, Bacillariophyta, Euglenozoa and Cryptophyta ( Figure 2). One-way analysis of variance (ANOVA) with a multiple post hoc (LSD) test (significance level is 0.05) was used to check the significant difference between the various treatments. The results show that 6 and 24 h after a 4 µM ODTMA-Br (C18) treatment, there were significant differences between Cyanobacteria and Chlorophyta (p < 0.01), Cyanobacteria with Bacillariophyta, Euglenozoa and Cryptophyta (p < 0.01). These results suggest that when applied at a low concentration, ODTMA -Br (C18) can selectively inhibit cyanobacteria with minimal effect on algae. The reasons for the distinct sensitivity of cyanobacteria to ODTMA-Br (C18) treatment is not known but is likely related to differences in the composition of the cell wall and cell envelope, affecting the interaction with the ODTMA-Br (C18) cation and, consequently, the function and viability of the cells [29]. It is further postulated that the extent of damage to macromolecule within photosynthetic apparatus differs between cyanobacteria and algae, and hence the fluorescence parameters were much stronger affected in the former, confirming an earlier study [21,22]. Compared to eukaryotic algae, the cyanobacterial photosynthetic apparatus does not segregate into organelles and has a direct connection with the plasma membrane. Therefore, these differences in cell structure might result in a higher susceptibility of cyanobacteria to ODTMA-Br (C18).
Microorganisms 2021, 9, x FOR PEER REVIEW 7 of 14 6 and 24 h after a 4 μM ODTMA-Br (C18) treatment, there were significant differences between Cyanobacteria and Chlorophyta (p < 0.01), Cyanobacteria with Bacillariophyta, Euglenozoa and Cryptophyta (p < 0.01). These results suggest that when applied at a low concentration, ODTMA -Br (C18) can selectively inhibit cyanobacteria with minimal effect on algae. The reasons for the distinct sensitivity of cyanobacteria to ODTMA-Br (C18) treatment is not known but is likely related to differences in the composition of the cell wall and cell envelope, affecting the interaction with the ODTMA-Br (C18) cation and, consequently, the function and viability of the cells [29]. It is further postulated that the extent of damage to macromolecule within photosynthetic apparatus differs between cyanobacteria and algae, and hence the fluorescence parameters were much stronger affected in the former, confirming an earlier study [21,22]. Compared to eukaryotic algae, the cyanobacterial photosynthetic apparatus does not segregate into organelles and has a direct connection with the plasma membrane. Therefore, these differences in cell structure might result in a higher susceptibility of cyanobacteria to ODTMA-Br (C18).

Mesocosm Experiment
To evaluate the cyanocidal performance of ODTMA-Br (C18) under field conditions, a mesocosm experiment was carried out where the effects of ODTMA-Br (C18) and H2O2 were compared.

Morphological and Physiological Changes in Microcystis Colonies
Microcystis colonies were recorded prior to the addition of ODTMA-Br (C18) or H2O2 (day 0), depicting neatly arranged cells embedded in a gelatinous matrix ( Figure S3). Three days post-exposure to either ODTMA-Br (C18) or H2O2, the Microcystis colonies disintegrated, and their pigments degraded ( Figure S3). The morphological changes 9 days post-treatment indicated that the majority of Microcystis colonies exposed to either ODTMA-Br (C18) or H2O2 had been effectively disintegrated and disappeared. The green algae Scenedesmus sp. thrived in both ODTMA-Br (C18) and H2O2-treated groups throughout the mesocosm experiment ( Figure S3 day 28).
Fv/Fm declined rapidly during day 1 and increased from day 3 in both ODTMA-Br (C18) and H2O2 treatments (Figure 4). From day 13, the Fv/Fm values increased markedly in ODTMA-Br (C18) treated mesocosms to ca. 0.7, which is commonly obtained in green algae but not Microcystis cells. The rise in Fv/Fm agrees with the microscopic observation ( Figure S3) that indicated a significant increase of the cell count of Scenedesmus sp. at the end of the mesocosm experiment ( Figure 5).

Mesocosm Experiment
To evaluate the cyanocidal performance of ODTMA-Br (C18) under field conditions, a mesocosm experiment was carried out where the effects of ODTMA-Br (C18) and H 2 O 2 were compared.

Morphological and Physiological Changes in Microcystis Colonies
Microcystis colonies were recorded prior to the addition of ODTMA-Br (C18) or H 2 O 2 (day 0), depicting neatly arranged cells embedded in a gelatinous matrix ( Figure S3). Three days post-exposure to either ODTMA-Br (C18) or H 2 O 2 , the Microcystis colonies disintegrated, and their pigments degraded ( Figure S3). The morphological changes 9 days post-treatment indicated that the majority of Microcystis colonies exposed to either ODTMA-Br (C18) or H 2 O 2 had been effectively disintegrated and disappeared. The green algae Scenedesmus sp. thrived in both ODTMA-Br (C18) and H 2 O 2 -treated groups throughout the mesocosm experiment ( Figure S3 day 28).
Fv/Fm declined rapidly during day 1 and increased from day 3 in both ODTMA-Br (C18) and H 2 O 2 treatments (Figure 4). From day 13, the Fv/Fm values increased markedly in ODTMA-Br (C18) treated mesocosms to ca. 0.7, which is commonly obtained in green algae but not Microcystis cells. The rise in Fv/Fm agrees with the microscopic observation ( Figure S3) that indicated a significant increase of the cell count of Scenedesmus sp. at the end of the mesocosm experiment ( Figure 5).     Similar results have been reported in several other studies where H 2 O 2 was used as cyanocide. Weenink and colleagues [13] proposed that competition by green algae restrained the growth of cyanobacteria in H 2 O 2 -treated waters. In a field experiment, Yang and co-authors [30] found that H 2 O 2 treatment could promote the growth of chlorophytes, and the dominant genera were Coelastrum, Pediastrum, Scenedesmus and Staurastrum. Wang et al. [31] reported that Chlamydomonas sp. thrived after suppression of the Microcystis bloom. Figure 5A shows the proportion of the major phyla found in the mesocosms. Cyanobacteria accounted for more than 90% of the phytoplankton population in the untreated controls during the entire experiment. The cell densities of Cyanobacteria were 1.83 × 10 9 and 2.14 × 10 9 cells/L on days 0 and 28, respectively, in the untreated controls ( Figure 5B). The addition of ODTMA-Br (C18) or H 2 O 2 lowered the cyanobacteria cell count but raised the abundance of eukaryotic algae. These responses of the mesocosm population to H 2 O 2 treatment are compatible with several studies showing that H 2 O 2 can reduce the abundance of cyanobacteria relative to eukaryotic algae [32,33]. Total algal density in the ODTMA-Br (C18) group dropped to the lowest level on day 5, and the proportion of Chlorophyta was significantly higher on day 13. On day 15, Chlorophyta almost overtook that of Cyanobacteria as the dominant species ( Figure 5A). In the H 2 O 2 treatment, the cell density of Cyanobacteria dropped to the lowest level on day 3, the biomass of Bacillariophyta increased on day 1 and that of Chlorophyta increased on day 3. However, there was a significant rise in cyanobacterial biomass from day 5 in the H 2 O 2 group. In terms of the duration of the cyanocidal effect, at the end of the experiment (day 28), despite the presence of a significant count of green algae in the H 2 O 2 group, the cyanobacteria recovered to the original level on day 0. In contrast, cyanobacterial biomass in the ODTMA-Br (C18) group was reduced by nearly 50% compared to day 0. The removal rate of Microcystis cells ( Figure 6) shows that the difference between ODTMA-Br (C18) and H 2 O 2 treatments became distinct from day 7. As an example, on day 28, the Microcystis removal rate in ODTMA-Br (C18) treated mesocosms was 99. 5%, whereas it was only 57.0% after H 2 O 2 .

Impacts on the Long-Term Biodiversity
Similar results have been reported in several other studies where H2O2 was used as cyanocide. Weenink and colleagues [13] proposed that competition by green algae restrained the growth of cyanobacteria in H2O2-treated waters. In a field experiment, Yang and co-authors [30] found that H2O2 treatment could promote the growth of chlorophytes, and the dominant genera were Coelastrum, Pediastrum, Scenedesmus and Staurastrum. Wang et al. [31] reported that Chlamydomonas sp. thrived after suppression of the Microcystis bloom. Figure 5A shows the proportion of the major phyla found in the mesocosms. Cyanobacteria accounted for more than 90% of the phytoplankton population in the untreated controls during the entire experiment. The cell densities of Cyanobacteria were 1.83 × 10 9 and 2.14 × 10 9 cells/L on days 0 and 28, respectively, in the untreated controls ( Figure 5B). The addition of ODTMA-Br (C18) or H2O2 lowered the cyanobacteria cell count but raised the abundance of eukaryotic algae. These responses of the mesocosm population to H2O2 treatment are compatible with several studies showing that H2O2 can reduce the abundance of cyanobacteria relative to eukaryotic algae [32,33]. Total algal density in the ODTMA-Br (C18) group dropped to the lowest level on day 5, and the proportion of Chlorophyta was significantly higher on day 13. On day 15, Chlorophyta almost overtook that of Cyanobacteria as the dominant species ( Figure 5A). In the H2O2 treatment, the cell density of Cyanobacteria dropped to the lowest level on day 3, the biomass of Bacillariophyta increased on day 1 and that of Chlorophyta increased on day 3. However, there was a significant rise in cyanobacterial biomass from day 5 in the H2O2 group. In terms of the duration of the cyanocidal effect, at the end of the experiment (day 28), despite the presence of a significant count of green algae in the H2O2 group, the cyanobacteria recovered to the original level on day 0. In contrast, cyanobacterial biomass in the ODTMA-Br (C18) group was reduced by nearly 50% compared to day 0. The removal rate of Microcystis cells ( Figure 6) shows that the difference between ODTMA-Br (C18) and H2O2 treatments became distinct from day 7. As an example, on day 28, the Microcystis removal rate in ODTMA-Br (C18) treated mesocosms was 99. 5%, whereas it was only 57.0% after H2O2. Currently, one of the major drawbacks of using algaecide for cyanobacterial bloom control is their limit in longevity, an important criterion in the selection of the proper cyanocide to be applied [5]. Most likely, the difference between the two cyanocides used here relates to the fact that H2O2 is rapidly degraded by biological and physicochemical Currently, one of the major drawbacks of using algaecide for cyanobacterial bloom control is their limit in longevity, an important criterion in the selection of the proper cyanocide to be applied [5]. Most likely, the difference between the two cyanocides used here relates to the fact that H 2 O 2 is rapidly degraded by biological and physicochemical processes, whereas the ODTMA-Br (C18), which stays in the aqueous system as micelles or by binding to solid particles, is far more stable [14].

Impacts on the Long-Term Biodiversity
At present, it appears that the variability in H 2 O 2 efficacy depends on the type and density of cyanobacteria present, the water chemistry and the isolation of the water body [15]. In many cases, repeated application of H 2 O 2 is employed to control cyanobacterial blooms during an entire season. A recent study reported that two consecutive applications of a low H 2 O 2 dose were much more efficient against M. aeruginosa strain MGK than a single treatment with a higher H 2 O 2 concentration [34].
The biodiversity reflects the compositional characteristics and function of a community. Increased biodiversity indicates more complicated relationships among species and food web structures, which can result in the community having a greater resilience and stability in response to external environmental fluctuations [35]. The outbreak of cyanobacterial blooms reduces biodiversity and causes damage to ecological functions [36]. The species diversity index of the phytoplankton community in the mesocosms gradually increased following the different algaecide treatments (Figure 7). A marked rise in the Shannon-Wiener index was already observed 7 days after the H 2 O 2 treatment, and the maximum value reached was more than three times higher than the control, but then gradually decreased to 1.0 by the end of the experiment. At that time, after ODTMA-Br (C18) application, the biodiversity was 3.2 times higher than in the control and 1.5 times that of the H 2 O 2 treatment. Given the longer exposure to ODTMA-Br (C18) compared with H 2 O 2 (mentioned above), our results show that ODTMA-Br (C18) has an advantage over H 2 O 2 in the long-term impact on the biodiversity. A similar conclusion is drawn from a Bray-Curtis distance of dissimilarity ( Figure S4) and PerMANOVA analyses (Table S2). The dissimilarity coefficients of the Bray-Curtis distance between ODTMA-Br (C18) and control groups was above 0.4 during the 28-day experiment, suggesting that the composition of the phytoplankton community is distinct between the ODTMA-Br (C18) and control groups. Similarly, the composition of the phytoplankton community of the ODTMA -Br (C18) group differed from the H 2 O 2 group (dissimilarity coefficient > 0.4). However, the dissimilarity coefficient between the H 2 O 2 and control groups were above 0.4 by day 15 and below 0.4 on day 28, indicating the recovery of Microcystis. PerMANOVA, based on the phytoplankton cell density and biomass throughout the mesocosm experiment, showed that there were significant differences in the total phytoplankton composition and Diatom composition of both ODTMA-Br (C18) and H 2 O 2 treatments compared to those in the control (p < 0.05, Table S2). The difference in Chlorophyte composition was significant between the control and H 2 O 2 treated groups, while it was not significant between the control and ODTMA-Br (C18) groups.
Microorganisms 2021, 9, x FOR PEER REVIEW 11 of 14 processes, whereas the ODTMA-Br (C18), which stays in the aqueous system as micelles or by binding to solid particles, is far more stable [14]. At present, it appears that the variability in H2O2 efficacy depends on the type and density of cyanobacteria present, the water chemistry and the isolation of the water body [15]. In many cases, repeated application of H2O2 is employed to control cyanobacterial blooms during an entire season. A recent study reported that two consecutive applications of a low H2O2 dose were much more efficient against M. aeruginosa strain MGK than a single treatment with a higher H2O2 concentration [34].
The biodiversity reflects the compositional characteristics and function of a community. Increased biodiversity indicates more complicated relationships among species and food web structures, which can result in the community having a greater resilience and stability in response to external environmental fluctuations [35]. The outbreak of cyanobacterial blooms reduces biodiversity and causes damage to ecological functions [36]. The species diversity index of the phytoplankton community in the mesocosms gradually increased following the different algaecide treatments (Figure 7). A marked rise in the Shannon-Wiener index was already observed 7 days after the H2O2 treatment, and the maximum value reached was more than three times higher than the control, but then gradually decreased to 1.0 by the end of the experiment. At that time, after ODTMA-Br (C18) application, the biodiversity was 3.2 times higher than in the control and 1.5 times that of the H2O2 treatment. Given the longer exposure to ODTMA-Br (C18) compared with H2O2 (mentioned above), our results show that ODTMA-Br (C18) has an advantage over H2O2 in the long-term impact on the biodiversity. A similar conclusion is drawn from a Bray-Curtis distance of dissimilarity ( Figure S4) and PerMANOVA analyses (Table S2). The dissimilarity coefficients of the Bray-Curtis distance between ODTMA-Br (C18) and control groups was above 0.4 during the 28-day experiment, suggesting that the composition of the phytoplankton community is distinct between the ODTMA-Br (C18) and control groups. Similarly, the composition of the phytoplankton community of the ODTMA -Br (C18) group differed from the H2O2 group (dissimilarity coefficient > 0.4). However, the dissimilarity coefficient between the H2O2 and control groups were above 0.4 by day 15 and below 0.4 on day 28, indicating the recovery of Microcystis. PerMANOVA, based on the phytoplankton cell density and biomass throughout the mesocosm experiment, showed that there were significant differences in the total phytoplankton composition and Diatom composition of both ODTMA-Br (C18) and H2O2 treatments compared to those in the control (p < 0.05, Table S2). The difference in Chlorophyte composition was significant between the control and H2O2 treated groups, while it was not significant between the control and ODTMA-Br (C18) groups.  The dynamics of the phytoplankton population in the mesocosm experiment was reflected in the rise in the eukaryotic algae proportion, especially green algae and diatoms, in response to the application of the cyanocides, likely because they are less susceptible to ODTMA-Br (C18) or H 2 O 2 than cyanobacteria [37]. In the whole lake experiment, the selective damage of H 2 O 2 to cyanobacteria was further aggravated [11,38]. The specific advantage of H 2 O 2 is more significant when two applications of a low concentration of H 2 O 2 are applied [5,34]. Meanwhile, ODTMA-Br (C18) exerts a selective effect on cyanobacteria, green algae and diatoms [21]. The data obtained in this study suggested that ODTMA-Br (C18) might be a cyanocide candidate with a relatively longer duration in suppressing cyanobacterial blooms. Although several studies have reported the (eco)toxicity of QACs [39,40], these compounds are adsorbed into particulate matter rapidly after application [41,42], and the possibility to become bioavailable through leaching from the particulate fraction is very low at environmentally relevant concentrations [43]. Moreover, QACs are susceptible to microbial degradation [44]. In practice, further clarification/assessment of the ecological "fingerprint", impact, fate and risk of ODTMA-Br (C18) treatments are requested before application in the field.

Conclusions
The current study showed that ODTMA-Br (C18) is an efficient and selective cyanocide with good application prospects to control and manage harmful Microcystis blooms in fish ponds. Compared with a single low dose of H 2 O 2 application, a single dose of ODTMA-Br (C18) had a more persistent inactivation of Microcystis colonies under the tested conditions. Moreover, the price nowadays of ODTMA-Br (Br) is fourfold higher than H 2 O 2 . Though QACs have been widely used as cationic disinfectants and surfactants, their direct application in field-controlling cyanobacterial blooms has yet to be evaluated and documented.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/microorganisms9112258/s1, Table S1. JIP test parameters with explanations and equation calculated using data extracted from the O-J-I-P fast Figure 1995. Table S2. PerMANOVA assessment of the differences in total phytoplankton, Chlorophyte, and Diatom communities between every two treatments based on biomass during the 28 days experiment. Figure S1. Photos of the fish pond where Microcystis spp. colonies were collected and site of the mesocosm experiment. The geographic location is 30.53N 114.40E. Figure S2. The photosynthetic parameters measured in Microcystis colonies exposed to different ATMA-Br for 3 h. Figure S3. ODTMA and H 2 O 2 impose morphological and physiological changes in Microcystis colonies. Figure S4. Bray-Curtis distance and Metric multidimensional scaling analyses between the control, ODTMA, and H 2 O 2 groups. Dissimilarity coefficients (a value greater than 0.4 indicates that there was a difference in species composition between the two treatments.
Author Contributions: X.Z., Y.X., Y.J. and L.S. conceived and designed the experiments; X.Z., Y.X. and Y.J. performed the experiments, and analyzed and interpreted the data; X.Z., Y.X. and Y.J. wrote the draft; L.S., A.S., A.K., C.S., F.B., G.D. and L.L. modified the draft. All authors have read and agreed to the published version of the manuscript.