Effects of mixing intensity on colony size and growth of Microcystis aeruginosa

Mixing is an integral environmental factor that affects lake ecosystems. For the cyanobacterium Microcystis, colony size is important with respects to migration velocity, how cells respond to grazing pressure, light attenuation, nutrient uptake and growth. To understand how mixing shapes colony size and the growth of Microcystis, we measured the effects of different current velocities (0, 0.16, 0.32, 0.64, and 1.28 m s−1) on M. aeruginosa in Lake Taihu. After 24 h of continuous mixing, the mean colony sizes of M. aeruginosa in the controls, 0.16, 0.32, 0.64, and 1.28 m s−1 groups were 23.6, 50.1, 92.9, 67.8, and 37.3 μm, respectively. Colony sizes of M. aeruginosa in all treatment groups were significantly larger than those in controls. As well, the concentration of soluble extracellular polysaccharide and bound extracellular polysaccharides of M. aeruginosa in all treatment groups were significantly higher than those in controls. Except for the highest level of mixing (1.28 m s−1), the growth rate of M. aeruginosa was significantly higher than that in controls. This study suggested that mixing intensity over short time periods can significantly influence colony size and the growth of M. aeruginosa.


Introduction
The formation of cyanobacterial blooms is one of the most troubling symptoms of eutrophication (Sommaruga et al., 2009). Cyanobacterial blooms alter ecological processes and decrease the economic value of impacted waters. Controlling blooms induced by eutrophication remains a challenge facing water quality and water supply managers globally (Harke et al., 2016).
Many cyanobacterial blooms in eutrophic waters are dominated by the non N 2 -fixing colonial genus Microcystis. Field investigations have demonstrated that large colonies of Microcystis are a main component of many blooms. In eutrophic Lake Taihu (China), investigations have found that large colonies (colony size > 38-50 mm) were dominant during Microcystis bloom events (Cao and Yang, 2010;Zhu et al., 2014;Qin et al., 2018), while small colonies (colony size < 50 mm) were dominant in the water column during non-blooming periods (Wu and Kong, 2009;Cao and Yang, 2010). Large Microcystis colonies also dominate during bloom periods in others eutrophic lakes (Sabart et al., 2013). Studies have shown that larger colonies of Microcystis have advantages in upwards floating speed , the ability to resisting grazing stress (Oliver and Ganf, 2000), the ability to capture light (Kirk, 1975;Robarts and Zohary, 1984), and nutrient uptake (Shen and Song, 2007). Larger colonies formation by Microcystis better enables cells to access optimal light and nutrient environments by floating on the water surface as a thick "scum" (Reynolds, 2006;Yamamoto et al., 2011;Qin et al., 2018). The above investigations have suggested that Microcystis colony size was of an important facts affecting blooms formation in eutrophic fresh waters.
In fresh waters, Microcystis populations mostly exist as colonies during blooms (Wu and Kong, 2009;Cao and Yang, 2010;Qin et al., 2018). However, in lab, Microcystis colonies usually transform to unicells or paired-cells in growth medium. How Microcystis unicells might transform to colonies remains unclear. Studies have demonstrated many factors that can induce unicell to colony transformation of Microcystis, including biotic and abiotic factors, e.g., zooplankton grazing (Burkert et al., 2001), bacteria (Wang et al., 2016b), microcystins (Sedmak and Eler sek, 2006), high light intensity (Xiao, 2011), temperature and phosphorus (Zhu et al., 2016;Duan et al., 2018), and the presence of heavy metals (Bi et al., 2013). Although a lot of research has been conducted, the mechanism of colony formation of Microcystis remains unknown.
Mixing is an integral environmental factor that affects lake ecosystems. Several studies have reported that mixing was an important factor affecting colony size of colonial Microcystis (Regel et al., 2004;Wang et al., 2016c;Xiao et al., 2016;Yang et al., 2017;Li et al., 2018). However, the quantitative effect of mixing on colony size of Microcystis remains unclear. To understand this, we established simulation experiments to measure the effects of different mixing intensity (0, 0.16, 0.32, 0.64, and 1.28 m s À1 ) on M. aeruginosa isolated from Lake Taihu. The result of this research shed light on quantitative effect of mixing on colony size and growth of Microcystis and potentially blooms in shallow lakes like Lake Taihu.

Materials and methods
Single colony of Microcystis aeruginosa, one of the dominant species of Microcystis in Lake Taihu (China), was isolated from lake water in Meiliang Bay (dominated by Microcystis bloom) in Lake Taihu in September 2016 and maintain in BG-11 medium (Rippka et al., 1979). After November 2016, unialgal cultures of M. aeruginosa were transferred to modified BG-11 medium (where TN = 50 mg L À1 and TP = 2.5 mg L À1 ). Until the beginning of our experiment, the M. aeruginosa cultures persisted as a mixture of single-cells, paired-cells and small colonies for a period of five months (∼4.83 Â 10 6 cells mL À1 ). At the beginning of the experiment, 150 mL inoculums of the exponentially growing M. aeruginosa (∼4.83 Â 10 6 cells mL À1 ) were transferred to 500 mL Erlenmeyer flasks containing 200 mL of modified BG-11 medium. Considering the current velocities of Lake Poyang (0.075-1.34 m s À1 ) (Lai et al., 2015), Lake Chaohu (0.002-0.109 m s À1 ) (Wang et al., 2016a), and Lake Taihu (0.005-0.077 m s À1 ) , different mixing intensities were designed as following: 0, 50, 100, 200, and 400 rpm, which approximate current velocities of 0, 0.16, 0.32, 0.64, and 1.28 m s À1 (Camacho et al., 2007;Rodríguez et al., 2009), respectively. For treatments, continuous mixing was maintained for 24 h while the 0 rpm groups were considered as the controls. All treatments were maintained in triplicate. Next, treatment groups were put into 500 mL flasks then put on four shaker incubators (50, 100, 200 and 400 rpm) for 24 h at 25°C under dark to simulate the effect of the mixing induce by windwave on M. aeruginosa. Mixing was generated on four horizontally oscillating shaking incubators. After shaking treatments, cultures were maintained in a "quiescent state" (i.e., no shaking). Controls were kept quiescent during the entire experimental period. Finally, after continuing mixing for 24 h, all groups were put in incubator at 25°C under cool white fluorescent lights at an intensity of 40.5 mol m À2 s À1 with a light-dark period of 12:12 h. The total concentration of nitrogen and phosphorus in all groups nutrient were TN = 50 mg L À1 and TP = 2.5 mg L À1 at the start of the shaking.
Samples were collected at 0, 1, 3, 5, 7, 9, and 11th days into this experiment to measure EPS (extracellular polysaccharide), colony size and abundance of M. aeruginosa. The content of soluble extracellular polysaccharide (sEPS) and bound extracellular polysaccharide (bEPS) were quantified spectrophotometrically by the anthrone method (Herbert et al., 1971) using glucose as standard. Samples of M. aeruginosa (5 mL) were preserved with Lugol's iodine solution: these samples were concentrated to 1 mL after 5 mL of M. aeruginosa settled for 48 h.M. aeruginosa colonies in the concentrated samples were measured (400x magnification) with a Nikon E200 microscope and QCapture Pro software (QImaging, Surrey BC, Canada). To determine mean colony size, at least 100 colonies of M. aeruginosa were measured. The abundance of M. aeruginosa was for at least 100 units for unicell and paired-cells, at least 100 colonies for 3-10 cell colony À1 and 10-100 cells colony À1 and 100 colonies for > 100 cells colony À1 . The concentrations of Chla were determined by spectrophotometry.

Statistical analyses
One-way analysis of variance (ANOVA) was used to test the differences in EPS, abundance, and colony size of M. aeruginosa between treatments and controls. All analyses were made using the SPSS19.0 computer programs.

The EPS of M. aeruginosa
EPS (extracellular polysaccharides) have been shown to be important in colony formation of Microcystis (Yang et al., 2008). In this study, there was no significant difference between the mean value of sEPS (soluble extracellular polysaccharides), and bEPS (bound extracellular polysaccharides) of M. aeruginosa in controls and that in 0.16, 0.32, 0.64, and 1.28 m s À1 groups during this experiment (P > 0.05). However, on the first day after continuing mixing for 24 h, the concentration of sEPS for M. aeruginosa in 0.16 m s À1 (1.70 pg cell À1 ), 0.32 m s À1 (1.78 pg cell À1 ), 0.64 m s À1 (1.70 pg cell À1 ), and 1.28 m s À1 (1.70 pg cell À1 ) were significantly higher than that in controls (1.44 pg cell À1 ) (P < 0.01) (Fig. 3). Also, the concentration value of bEPS of M. aeruginosa in 0.16 m s À1 (0.38 pg cell À1 ), 0.32 m À1 (0.45 pg cell À1 ), 0.64 m s À1 (0.39 pg cell À1 ), and 1.28 m s À1 (0.36 pg cell À1 ) groups were significantly higher than that in controls (0.26 pg cell À1 ) (P < 0.05). The concentration of sEPS and bEPS for M. aeruginosa in all groups gradually decreased with the time after the first day in this experiment (Fig. 3).

The abundance and growth of M. aeruginosa
The mean abundance of M. aeruginosa in controls, 0.16, 0.32, 0.64, and 1.28 m s À1 groups was 4. 05, 4.79, 5.17, 4.48, and 4.30 Â 10 6 cells mL À1 (Fig. 4a), respectively. The mean abundance of M. aeruginosa in 0.16, 0.32, and 0.64 m s À1 groups was significantly higher than those in controls and in the 1.28 m s À1 groups (P < 0.01) (Fig. 4a). The growth rates of M. aeruginosa in controls, 0.16, 0.32, 0.64, and 1.28 m s À1 groups was 0.227 (±0.006), 0.271 (±0.007), 0.298 (±0.006), 0.240 (±0.007), and 0.220/d (±0.007) (Fig. 5b), respectively. The growth rates of M. aeruginosa in 0.16, 0.32, 0.64 m s À1 groups were significantly higher than that in controls and in 1.28 m s À1 groups (P < 0.05), while no significantly different between the growth rates of M. aeruginosa in 1.28 m s À1 groups and that in controls (P > 0.05). The variations of abundance and Chla of M. aeruginosa in this study were showed in Figure 5. The abundance of M. aeruginosa in 0.32 m s À1 groups (8.87 Â 10 6 cells mL À1 ) was the highest among all groups at the end of experiment. Similar were found in Chla of M. aeruginosa in 0.32 m s À1 groups (2212 mg L À1 ) (Fig. 5).
Before mixing, no colonies with > 100 cells of M. aeruginosa were found in any treatment groups or the controls (Tab. 1). After mixing for 24 h, the proportion of cells within > 100 cell colonies relative to total abundance of M. aeruginosa increased from 0 to 9.68 in 0.16 m s À1 groups, from 0 to 32.13 in 0.32 m s À1 groups, from 0 to 28.29 in 0.64 m s À1 groups, and from 0 to 15.64% in 1.28 m s À1 groups on the first day, respectively. However, no colonies with > 100 cells of M. aeruginosa were found in control groups on the first day. In contrast, on the first day, the proportion of cells in 10-100 cell colonies of M. aeruginosa significantly decreased in all treatment groups (Tab. 1) after mixing, e.g., in the 0.32 m s À1 groups, the cell abundance proportion of 10-100 cells colony to total abundance of M. aeruginosa decreased from 57.00% to 30.13% (Tab. 1). This suggested that > 100 cell colonies of M. aeruginosa may have come from the aggregation of 3-100 cells colony by mixing, especially the 10-100 cells colony.

Discussion
In this study, we found that mixing intensities (0.16-0.64 m s À1 ) favor increased colony sizes for M. aeruginosa. Wang et al. (2016c) reported that colony size of Microcystis significantly increased after mixing driven by Typhoon Soulik in Lake Taihu. Yang et al. (2017) found that simulating mixing (24 h) significantly enlarged the colony sized of Microcystis in Lake Taihu. However, O'Brien et al. (2004) reported that Microcystis colonies collected from field broke up into smaller colonies (< 200 mm) after mixing.  study showed that no colonies conformation was found after mixing. Li et al. (2018) reported that three species of Microcystis colonies collected from Lake Taihu broke up into smaller colonies after mixing. Overall these previous observations have resulted in conflicting results concerning the effects of mixing on colony formation by Microcystis.
One important variable to be considered is the starting conditions of Microcystis for each of these studies. In the current experiment, the mean colony size of M. aeruginosa was 23.6 mm at the onset of experiment, while it was above 200 mm in O' Brien's (2004) study. In  study, the Microcysits was only single-cells, while those were a mixture of single-cells, paired-cells and small colonies of M. aeruginosa in this study. Similarly in our study the mixing used was continuous over 24 h, while it was 30 min in Li's (2018) study. In total the above studies showed that whether the colonies of Microcystis aggregate or disaggregate after mixing may depend on the mixing intensity, the mixing time and colony size of Microcystis.
EPS are mainly found in mucilage or the cell's sheath, and it can affect the "stickiness" of the cell surface, contributing to colony formation in Microcystis (Yang et al., 2008;Li et al., 2013;Zhu et al., 2014). Research has shown that the concentration of EPS in Microcystis colonies was significantly higher than in single cells (Li et al., 2013). Small colonies of Microcystis may come from the division of mother cell and adhesion via EPS (Kessel and Eloff, 1975). It is thought that sEPS (soluble extracellular polysaccharides) may increase cell adhesiveness, while the bEPS (bound extracellular polysaccharides) may prevent daughter cells from separating after cell division (Li et al., 2013). In this study, the concentrations of bEPS and sEPS of M. aeruginosa in 0.16, 0.32, 0.64, and 1.28 m s À1 were significantly higher than that in controls after   continuing mixing for 24 h (P < 0.05). The increased bEPS and sEPS after mixing may explain why colony size of M. aeruginosa enlarged in all treatment in this study, especially in 0.32 m s À1 groups.
Many studies have shown that stress conditions can lead to EPS production and releasing by Microcystis spp., including grazing of plankton (Yang et al., 2008), high concentration of calcium (Wang et al., 2011) and microcystin-RR . Gan et al. (2012) found that microcystin-RR induced EPS in the culture medium and up regulated genes related to polysaccharide biosynthesis, but had no effect on the cell growth rate. Li et al. (2007) founded that allelopathy material eathyl-2-methyl acetoacetate (EMA) produced by reed could raise the respiration rate of M. aeruginosa, causing the CO 2 concentration raising in its culture flask, lowering the photosynthesis action rate of M. aeruginosa. In this study, mixing is one of stress conditions, so EPS content in the mixing treated groups was much higher than that of the control after mixing for 24 h in the dark and no significant difference was observed in both cell density and Chla. Also, we infer that respiration rate in all treated groups would raise to satisfy the energy requirement of EPS production and releasing of Microcystis spp. in the mixing treated groups.
The abundance and growth rates of M. aeruginosa in 0.16, 0.32, and 0.64 m s À1 groups were significantly higher than that in controls groups and in 1.28 m s À1 groups (P < 0.05), while the abundance and growth rates were not significantly different between the 1.28 m s À1 groups and controls (P > 0.05). Regel et al. (2004) demonstrated that low mixing intensities had no effect on the growth rate of M. aeruginosa, but high mixing intensity decreased growth rates. Yan et al. (2008) reported that certain mixing intensities stimulated the growth of M. aeruginosa, but high mixing intensities inhibited the growth of M. aeruginosa under low nutrient availability and all mixing intensities had no significant effect on the growth of M. aeruginosa under eutrophic nutrient. Jiang et al. (2012) found that low mixing intensity stimulated the growth of M. aeruginosa while high mixing intensity inhibited growth. Turbulence is reported to decrease the diffusive boundary layer around cells, theoretically increasing nutrient diffusion to cells (Lazier and Mann, 1989) and potentially increasing metabolic activity. Regel et al. (2004) found that low mixing intensities increased the esterase activity and cell viability of M. aeruginosa, but these were decreased at high mixing intensities. In this study, the values of ETR max (the potential maximum photosynthetic rate) and I k (half saturation light intensity) of M. aeruginosa in 0.16, 0.32, and 0.64 m s À1 groups were significantly higher than those in controls and 1.28 m s À1 groups (P < 0.01) in the 3rd day after mixing (Fig. S1). This may explain why growth rates of M. aeruginosa in 0.16, 0.32, 0.64 m s À1 groups were significantly higher than that in controls groups and in 1.28 m s À1 groups. Moreover, the values of MDA and SOD activities of M. aeruginosa in 1.28 m s À1 groups were significantly higher than those in controls and the 0.16, 0.32, and 0.64 m s À1 groups after mixing (Fig. S2). This information suggested that the hurt by high intensity mixing (1.28 m s À1 ) on M. aeruginosa maybe lead to the lower growth of M. aeruginosa.
Taihu Lake, the third largest freshwater lake in China is shallow and hypereutrophic. Microcystis blooms have occurred from May to October every year since the 1950s and have been larger and more severe since the 1980s (Qin, 2008). M. aeruginosa has been one of the dominant species of Microcystis in Lake Taihu (Yang et al., 2009). The mixing induced by wind-wave action in Lake Taihu is frequent due to the lakes shallow depth, large surface area and large fetch. This mixing leads to reduced transparency and nutrient resuspension into the water column from the sediments (Qin, 2008). Moreover, the mixing can increase the colony size of Microcystis. During the Typhoon Soulik (July 12-13, 2013), the mean current velocities was 0.0361 m s À1 (range 0.0063-0.1031 m s À1 ) in Lake Taihu (Wu et al., 2015). And the mean colony size of Microcystis in Lake Taihu significantly increased from 32.8 mm in pre-typhoon period to 69.4 mm in post-typhoon period within 48 h (Qin et al., 2018). The increasing of colony size of M. aeruginosa by mixing favors upward movement of cells, enhancing exposure to light and subsequently growth and biomass accumulation (Cao and Yang, 2010;Yamamoto et al., 2011;Qin et al., 2018). This may explain why M. aeruginosa consistently becomes the dominant species of phytoplankton in Lake Taihu.