Growth and photosynthesis responses of microcystin (MC)- and non-MC-producing Microcystis strains during co-culture with the submerged macrophyte Myriophyllum spicatum

The frequency and intensity of cyanobacterial blooms, especially Microcystis blooms, are increasing worldwide because of nutrient enrichment and global warming (Paerl et al. 2016). Microcystis are the most common bloom-dominated species in eutrophic waters, and they are distributed from temperate to tropical regions worldwide (Xiao et al. 2018). Microcystis-dominated blooms are composed of microcystin (MC)- and non-MC-producing populations (Liu et al. 2021). Their composition proportion varies depending on the waterbody, season, and bloom formation stage, driven by environmental factors such as light, CO₂ level, and phosphorus concentrations (Lei et al. 2015). The intraspecific composition also affects the accurate assessment and effective elimination of cyanobacterial blooms. However, the responses of MC- and non-MC-producing Microcystis strains to biotic and abiotic stresses in aquatic ecosystems remain poorly understood to date.

Submerged plants are crucial primary producers that maintain the integrity and stability of shallow aquatic ecosystems. They can significantly reduce turbidity by preventing sediment re-suspension (Bai et al. 2020). In addition, these plants maintain the good biological structure of aquatic ecosystems and provide a good living environment for other aquatic organisms (Wood et al. 2017). They absorb excess nutrients from water (Ferreira et al. 2018) and exert allelopathic inhibition on algal growth (Hilt et al. 2006). Submerged plants inhibit cyanobacteria by releasing allelochemicals, which can destroy the structure of cyanobacteria cells and effectively inhibit the occurrence of cyanobacterial blooms (Yu et al. 2019), without obvious effects on eukaryotic phytoplankton (Zhu et al. 2010).

The reestablishment of submerged vegetation is an important strategy to restore the structure and function of eutrophic aquatic ecosystems (Bai et al. 2020). The underlying physiological mechanisms include affecting photosynthesis, breaking cell membranes, reducing enzyme activities, and disturbing ultrastructure and gene expression (Gao et al. 2017). However, few studies investigated the intraspecific responses of cyanobacteria such as Microcystis to the stress from submerged macrophytes in the same ecological niche. The responses of MC- and non-MC-producing Microcystis strains to plant allelochemicals including tannic acid and pyrogallol were reported but with inconsistent results for the two allelochemicals (Švanys et al. 2016; Gao et al. 2020a). It is well-known that submerged macrophytes could produce and release various allelochemicals simultaneously to affect target organisms (Nakai et al. 2012). Whether MC-producing or non-MC-producing strains are more sensitive to the allelopathy from submerged macrophytes is not clear. The co-existence status of plants and cyanobacteria in eutrophic waterbodies must be simulated to verify the abovementioned phenomenon in plant-cyanobacterial co-culture systems. Plant-cyanobacteria co-culture systems were previously used to explore the inhibition effects and physiological mechanisms of submerged plants on different cyanobacterial species (Körner & Nicklisch 2002). However, such systems have been less used to investigate the strain responses to the stress from submerged macrophytes simulating conditions in actual shallow waterbodies.

Photosynthesis is the most basic and important physiological and ecological characteristic of phytoplankton. Chlorophyll fluorescence parameters can accurately reflect the photosynthesis of Microcystis aeruginosa (Campbell et al. 1998). Chlorophyll fluorescence transients and subsequent JIP tests have been widely used to reflect the structure and function of photosystem II (PSII) in phototrophic phytoplankton (Ni et al. 2012). The current study aimed to investigate the responses of MC- and non-MC-producing Microcystis strains to the stress from the submerged macrophyte M. spicatum, with strong allelopathic inhibition activity (Nakai et al. 1999), under co-culture conditions. The cell concentrations of Microcystis were set to be higher than those in previous studies to mimic the dense Microcystis blooms (Liu et al. 2016). The PSII performance dynamics of both Microcystis strains during 12-day experiments were measured.

Fresh M. spicatum plants were obtained from the Honghu Lake (N29.827°, E113.476°) in Hubei Province, and cultivated in an aquarium with 10 cm sediments in our greenhouse until new roots grew. Well-grown and uniform plant tissues were selected and cleaned for subsequent preculture in 1/10 diluted BG11 medium (Table 1) under climate-controlled conditions at 22 ± 3 °C, light at 2,000 lux with 12 h light/12 h dark cycles in the lab before the experiments. Our preliminary experiment proved that 1/10 diluted BG11 medium was suitable for the growth of M. spicatum.

One MC-producing strain (FACHB 915) and one non-MC-producing Microcystis strain (FACHB 1005) were obtained from the Freshwater Algae Culture Collection of the Institute of Hydrobiology (FACHB), the Chinese Academy of Sciences. Their MC-producing ability was also identified by FACHB. Microcystis was precultured in 1/10 diluted BG11 medium under the same conditions as M. spicatum, and the cells in the exponential phase after 5-day culture were prepared for the experiments.

Two co-culture experimental groups (PMC−, PMC+), two Microcystis mono-culture groups (SMC-, SMC+) and one M. spicatum mono-culture group (P) were established to compare the responses of two Microcystis strains to M. spicatum. Sterilized 1/10 diluted BG11 medium was prepared at 1 L per beaker. Plant apical tissues of M. spicatum with 12 cm length were cultivated at 2.0 ± 0.05 g fresh weight per litre in the beakers of the P, PMC− and PMC+ groups in sextuplicate. The prepared MC- and non-MC-producing Microcystis cells were then added to the beakers of the PMC+ and PMC− groups, respectively, at a final cell concentration of 4.50 ± 0.06 × 10⁶ cells mL⁻¹ in the range of dense bloom populations in eutrophic waters (Liu et al. 2016). The same amount of Microcystis cells was added into the beakers of the SMC+ and SMC- groups in triplicate as the Microcystis mono-culture control. All the beakers were covered with breathable sealing membranes, and cultured under the conditions described above for 12 days. The subsamples were collected every 3 days to measure the growth and photosynthesis parameters of both Microcystis strains.

Fresh Microcystis fluid samples were dark-adapted for 20 min and then subjected to an AquaPen-C 100 fluorometer (Photon Systems Instruments, Drasov, Czech Republic) to record O-J-I-P chlorophyll a fluorescence transient in accordance with the manufacturer's instructions. The JIP-test parameters were automatically extracted to assess the PS II performance of MC- and non-MC-producing Microcystis strains in different experimental groups (Table 2). The fluorescence yields at 50 μs, 2 ms, and 50 ms were denoted as F_o, F_j, and F_i, respectively (Appenroth et al. 2001). V_j represents the relative variable fluorescence intensity at the J step, φ_Po represents the maximum quantum yield for primary photochemistry, Ψo represents the probability that a trapped exciton moves an electron into the electron transport chain beyond Q_A (at t = 0) and φEo represents the quantum yield for electron transport (at t = 0). The specific energy flux ratios in the electron transport chain included parameters ABS/RC, TRo/RC, ETo/RC, and DIo/RC, representing the energy fluxes ratio, the trapped energy flux ratio, the electron transport flux ratio, and the dissipated energy flux ratio per reaction centre, respectively (Wang et al. 2021). PI_ABS represents the performance index on an absorption basis (Ni et al. 2012).

Microsoft Excel 2019 and Origin 2021b were used for data analysis in the study. Data are presented as the means ± standard deviations of samples. After determining the normality using Shapiro-Wilks test and homogeneity using Bartlett's test, one-way ANOVA was performed among the experimental groups, and multiple comparisons were performed using the least significance difference test. Significant difference was considered at p < 0.05. Difference significances between groups on the same day are represented by uppercase letters (i.e., A, B, C, D) and difference significances within groups at different timepoints are represented by lowercase letters (i.e., a, b, c, d, e).

The cell density of both Microcystis strains co-cultured with the plant M. spicatum was significantly lower than that in the corresponding mono-culture controls from day 9 (p < 0.05), when the cell density in the PMC− treatment dropped to a lower level than the initial value (Figure 1(a)). The corresponding inhibition ratio of the non-MC-producing Microcystis strains was significantly higher than that of the MC-producing strains from day 9. The inhibition rate reached 69.70% for the non-MC-producing Microcystis strains and 37.50% for the MC-producing strain on day 12 (Figure 1(b)).

The OJIP-curves of both strains in the co-culture groups were beneath those in the corresponding mono-culture groups. Fluorescence no longer increased when it reached the J step, and the O-J-I-P rise changed to the O-J (J = I = P) rise in the PMC− group on day 12 (Figure 2). The F_j values increased in all treatments, but were lower in the co-culture groups than in the mono-culture groups on day 12. In the PMC− group, the F_j value decreased by 25.13% compared with the SMC- group. In the PMC+ group, the value decreased by 22.81% based on that in the SMC+ group. By contrast, the relative variable fluorescence intensity (V_j) increased over time, from 0.21 to 0.39 in the PMC− group and from 0.19 to 0.28 in the PMC+ group (Table 3, p< 0.05). The V_j values in the PMC− and PMC+ groups were 2.3-fold and 1.3-fold those in the SMC- and SMC+ groups, respectively. The F_i values increased over time in the four groups. However, the F_i values in the co-culture groups were always lower than those in the corresponding mono-culture groups. The decreasing ratio was 26.14% for the PMC− group and 36.14% for the PMC+ group in comparison with the SMC- and SMC+ groups, respectively.

The φ_Po values, the maximum quantum yield for primary photochemistry, were not different among all experimental groups in the beginning (p > 0.05). But they differed significantly in the end of the experiment through 12-day changes at varying degrees (p < 0.05). For non-MC-producing Microcystis, the φ_Po in the PMC− group began to drop significantly from day 3, and it decreased by 31.03% compared with that in the SMC- group on day 12. For MC-producing Microcystis, the φ_Po in the PMC+ group began to drop significantly from the 6th day, and it decreased by 22.86% compared with that in the SMC+ group on day 12. A stable state without significant changes in Ψ_o during the experiment was observed in the SMC- and SMC+ groups. However, the Ψ_o in the co-culture groups decreased. In the PMC− group, the value dropped significantly from 0.79 to 0.61, with a decreasing ratio of 25.61% compared with that in the SMC- group. The Ψ_o decreased from 0.81 to 0.72 in the PMC+ group, with a decreasing ratio of 7.69% compared with that in the SMC+ group. The quantum yield for electron transport (φ_Eo) increased from 0.24 on day 0 to 0.41 for the SMC- group and 0.36 for SMC+ group on day 12, which were significantly higher than in the PMC− and PMC+ groups (Table 4, p< 0.05). The decreasing ratio was 43.90% for the PMC− group and 19.44% for the PMC+ group.

The ABS/RC values maintained stable status with a small fluctuation in the SMC- and SMC+ groups, but the values decreased significantly from 2.87 on day 0 to 2.66 on day 12 in the PMC− group. The ABS/RC value in the PMC+ group did not decrease significantly during the experiment, but was significantly higher than that in the SMC+ group on days 6 and 9 (p < 0.05). The ABS/RC value for MC-producing Microcystis was significantly higher than that for the non-MC-producing strain (Figure 3(a), p < 0.05). The TRo/RC values in the SMC- and SMC+ groups slightly changed during the experiment, whereas they decreased from 1.62 on day 0 to 1.56 in the PMC− group and 1.58 in the PMC+ group on day 12 (Figure 3(b)). No significant differences were observed between the PMC− and PMC+ groups (p > 0.05). The ETo/RC values gradually decreased in all treatments. The largest decline occurred in the PMC− group, followed by the PMC+ group, and the smallest drop appeared in the SMC+ group. The ETo/RC values in the PMC− and PMC+ groups were 85.71% and 93.85% of those in the SMC- and SMC+ groups on day 12, respectively (Figure 3(c)). The DIo/RC values in the SMC- and SMC+ groups showed no significant changes during the experiment (p > 0.05). However, the DIo/RC values in the PMC− group decreased significantly from 1.18 on day 0 to 1.10 on day 12, whereas those in the PMC+ group increased from 1.71 on day 0 to 1.75 on day 12 (Figure 3(d)).

The performance index (PI_ABS) values increased from 0.48 on day 0 to 0.63 on day 12 in the SMC- group, and from 0.56 to 0.80 in the SMC+ group. However, the PI_ABS values in the PMC− group decreased significantly from 0.49 to 0.46. Although the PI values in the PMC+ group increased from 0.56 to 0.67, the increasing degree was obviously smaller than that in the SMC+ group (Table 5). The PI_ABS values in the PMC− and PMC+ groups decreased by 26.98% and 16.25%, respectively, in comparison with those in the corresponding mono-culture controls.

The different responses of MC- and non-MC-producing Microcystis to co-existing M. spicatum were proven from the growth and photosynthesis aspects of Microcystis in the present study. Previous studies found that cyanobacteria such as Microcystis were more sensitive to submerged plants than green algae (Jasser 1995; Hilt et al. 2006). In addition, MC- and non-MC-producing Microcystis strains showed different responses to nutrients and light (Davis et al. 2009; Lei et al. 2015). This study demonstrated that non-MC-producing Microcystis strains were more easily inhibited by the co-existence of submerged plants. This finding was consistent with the studies using the plant allelochemicals tannic acids and 12 Microcystis strains worldwide (Švanys et al. 2016), but contrary to the previous investigation using pyrogallol and three Microcystis strains isolated in Australia, where the toxigenic M. aeruginosa strain was more sensitive to pyrogallol than non-toxigenic strains (Gao et al. 2020a). The plant-Microcystis co-culture systems used in the current study reflected the interaction environments of cyanobacteria and submerged plants in actual waterbodies. The interaction and battle between the plants and Microcystis under co-culture conditions were close to the actual situation in eutrophic water bodies, which was not considered in the exposure experiments using pure plant allelochemicals. Submerged macrophytes could release various allelochemicals to inhibit Microcystis under co-culture conditions, while Microcystis strains release allelochemicals to affect plants simultaneously. The MC-producing Microcystis strain could release MCs. MCs have been widely reported to consistently affect photosynthesis of aquatic macrophytes (Zhang et al. 2022). We indeed observed more damage of submerged macrophyte M. spicatum co-existing with the MC-producing Microcystis strain than with the non-MC-producing strain in the present study. This might be the reason why MC-producing Microcystis strain demonstrated more resistance to the stress from M. spicatum than the non-MC-producing strain. Different responses of MC- and non-MC-producing Microcystis strains to co-existing M. spicatum indicated that attention should be paid to the intraspecies composition and toxigenic levels of cyanobacteria when using restoration measures with submerged vegetation to prevent and control cyanobacteria blooms in shallow eutrophic waterbodies.

The photosynthetic pigments (e.g., chl a), photosynthetic oxygen evolution and maximum quantum yield of PSII (φ_Po) are frequently used as parameters to investigate the photosynthesis activity of cyanobacteria under the stress from plants under co-existing conditions (Körner & Nicklisch 2002; Zhu et al. 2010). Chlorophyll fluorescence transients (OJIP curves) and subsequent JIP tests were used to reveal the status and behavior of PSII in MC- and non-MC-producing Microcystis strains when co-existing with M. spicatum in the present study. The OJIP transients reflected the successive reduction of the electron acceptor pools of PSII (Govindjee 1995). The J step and I step represent an accumulation of Q_A⁻Q_B and Q_A⁻Q_B⁻ forms, respectively. Lower F_j and F_i values of the MC- and non-MC-producing Microcystis strains when co-existing with M. spicatum than in mono-cultures indicated that the electron transfer from the primary bound plastoquinone Q_A⁻ to the second bound plastoquinone Q_B was blocked by the stress from M. spicatum.

All JIP-test parameters indicated more serious damage to the PSII behavior and performance of non-MC-producing Microcystis cells than on MC-producing strain from M. spicatum. The parameters φ_Po and φ_Eo represent the maximum quantum yield for primary photochemistry and quantum yield for electron transport, respectively. The parameter Ψ_o reflects the quantum efficiency of electron transfer from Q_A⁻ to Q_B (Appenroth et al. 2001). The effect differences on φ_Eo and Ψ_o between the MC- and non-MC-producing Microcystis strains were larger than those on φ_Po. It further indicated that electron transport efficiency of PSII in the non MC-producing Microcystis cells was much more affected by co-existing M. spicatum than that in MC-producing cells, and the acceptor side of the electron transport chain might be the specific target. This result is consistent with the effects of salinity stress on PSII in Ulva lactuca (Xia et al. 2004).

The structure and function of PSII differ among various phototrophs. The fluorescence transients of cyanobacteria Anabaena sp., Planktothrix rubescens, and Limnothrix redekei are different (Strassert & Srivastava 1995). The chlorophyll fluorescence parameters of M. aeruginosa and Chlorella pyrenoidosa are different (Liu et al. 2019). Among the four parameters to reflect the specific energy fluxes ratios in the electron transport chain, the initial TRo/RC and ETo/RC values for both strains in each experimental group were similar, but the initial ABS/RC and DIo/RC values differed obviously between two strains. This result indicates differences in PSII behavior between Microcystis strains. The parameters ABS/RC, TRo/RC, ETo/RC and DIo/RC represent the energy fluxes ratio, the trapped energy flux ratio, the electron transport flux ratio, and the dissipated energy flux ratio per reaction centre, respectively (Wang et al. 2021). During the experiment, the inhibition activity of M. spicatum on TRo/RC was weak for both strains, but the inhibition on ETo/RC was much higher for the non-MC-producing Microcystis strain than for the MC-producing strain. This parameter, together with φ_Eo and Ψ_o, provided multiple evidence that the electron transfer of non-MC-producing Microcystis was hampered more seriously than that of the MC-producing strain during the co-culture with M. spicatum.

The performance index on an absorption basis, PI_ABS, is the combination of three independent parameters, namely, RC/ABS, φ_Po, and ψ_o. This index is more sensitive than φ_Po for Microcystis to water soluble substances of Dendranthema indicum flowers (Gao et al. 2020b). It is also sensitive for other aquatic phototrophs to biotic and abiotic environmental stresses (Wang et al. 2021). The inhibition ratio of co-existing M. spicatum on PI_ABS for non-MC-producing Microcystis was 1.7-fold that for MC-producing Microcystis, calculated according to the decreasing ratio in comparison with the corresponding mono-culture groups on day 12. However, the φ_Po values did not show such big differences between the two strains. The parameter RC/ABS in plant-Microcystis co-culture treatments showed an increasing trend in comparison with that in mono-culture controls. Hence, the value of the parameter PI_ABS was largely determined by ψ_o in the current study. The higher inhibition ratio of M. spicatum on ψ_o and PI_ABS suggested the electron transfer chain of the non-MC-producing Microcystis was much more sensitive than that of MC-producing Microcystis to co-existing M. spicatum. Here, one submerged plant was tested; whether the electron transfer chain of the non-MC-producing Microcystis is more sensitive to other submerged macrophytes than MC-producing Microcystis needs further investigation.

The growth and photosynthesis of both Microcystis strains were inhibited to varying degrees when co-cultured with the submerged plant M. spicatum. Chlorophyll a fluorescence transients and subsequent JIP-test parameters demonstrated more serious damage to the efficiency and energy flux of the electron transport of PSII in the non-MC-producing Microcystis than in the MC-producing strain. The electron acceptor side of the electron transport chain might be the specific target. These results provided detailed photosynthetic evidence that the strain composition and MC-producing ability of Microcystis affected their interactions with submerged macrophytes and the elimination of Microcystis-dominated blooms.

The work was funded by the National Program for the introduction of high-end foreign experts (G2021026024 L), National Natural Science Foundation of China (No. U1904124), major public welfare projects in Henan Province (201300311300), and Breeding Project of Henan Normal University (HNU2021PL05). We also appreciate the two anonymous reviewers for their helpful comments and suggestions to improve our manuscript.

All relevant data are included in the paper or its Supplementary Information.