Self-floating capsule of algicidal bacteria Bacillus sp. HL and its performance in the dissolution of Microcystis aeruginosa.

Algicidal bacteria is considered as an efficient and environmentally friendly approach to suppress Microcystis aeruginosa (M. aeruginosa). However, algicidal bacteria in natural water is limited during the practical application due to the interference of external factors and the low reuse capability. In this study, a bio-degradation capsule for M. aeruginosa is prepared by bio-compatible sodium alginate (SA) compositing with eco-friendly ethyl cellulose (EC) to improve the property and reuse capability of algicidal bacteria. Bacterial strain HL was well immobilized and the capsule was obtained with 2% of SA, 3% of calcium chloride (CaCl2) and 3% of EC. It has been proved that capsules immobilizing bacteria HL shows considerable advantage over traditional bio-treatment systems (free-living bacteria) and good reusable performance. A better algicidal rate of 77.67% ± 1.14% at 7th day was obtained with the use of capsule embedding 50 mL of algicidal bacteria, enhanced by 11.05% comparing with same amount of free-living bacteria. Moreover, the algicidal rate of M. aeruginosa still reached 68.57% ± 2.88% after three times repetitive use. The effect of capsules on the fluorescence and antioxidant system of M. aeruginosa indicated that the photosystems were irreversibly damaged and the antioxidant response of superoxide dismutase (SOD), peroxidase (POD) and catalase (CAT) were significantly induced. Overall, capsules prepared in this study can provide a desirable environment for algicidal bacteria HL and ensure algicidal bacteria to in-situ work well in inhibiting booms of algae.


Introduction
Water eutrophication aroused by the excessive discharge of nutrients and phosphorus has resulted in the overgrowth of harmful algal blooms (HABs) worldwide ). M. aeruginosa, one kind of the cyanobacteria (Schmidt et al. 2020), not only in uences negatively the water ecosystem (Lu et al. 2021b) but also produces toxic microcystins (MCs) , which can pose toxic effect on liver, nervous and genital system (Zhang et al., 2020a). Up to now, many strategies have been applied to control M. aeruginosa, such as photocatalysis (Fan et al. 2021, hydrogen peroxide , copper sulfate (Anderson 2009), etc. Though these methods are fast and effective in the inhibition of M. aeruginosa, the potential threat to aquatic environment and the secondary pollution limit their large-scale application (Zhu et al. 2020). Currently, an enormous amount of effort has been devoted to developing a biological way to inhibiting the algae, different kinds of the bacteria, which are able to dissociate the algae, are found and applied to control M. aeruginosa. The reported algicidal bacteria species involved Bacillus sp. (Xuan et al. 2017), Acinetobacter (Yi et al. 2015), Streptomyces (Yu et al. 2019), etc. It was reported that M. aeruginosa could be inhibited by algicidal bacteria through direct feeding or secreting algicidal compounds and they can have a substantial effect on M. aeruginosa (Lu et al. 2021a).
The preparation of microbial inhibition reagents for the algae is usually conducted by gathering algicidal bacteria from water, soil and organisms, then enriching and culturing process. Thus, most of current microbial algal inhibition reagents included algicidal bacteria and culture medium ).
Usually, microbial algal inhibition reagents are used by directly pouring in natural water, algicidal bacteria were then released and dispersed, which could inhibit HABs effectively. However, the environment and conditions for algicidal bacteria, including temperature, water velocity and natural enemies, are complex. Especially, it is hard to ensure algicidal bacteria to be the predominant bacterial population in natural micro-ecosystem, and may disappear due to intense competition between them. Giving a desirable community environment for algicidal bacteria is thus essential to keep the activity and effect of microbial inhibition reagents.
Capsule immobilization of algicidal bacteria is a potential solution , Wang &Coyne 2020), which can provide favorable environment for algicidal bacteria and improve the algicidal effect. For example, Shewanella sp. IRI-160 was immobilized into several carriers (Agarose, alginate hydrogel, cellulosic sponge, and polyester foam) and used in inhabiting harmful dino agellates, a higher activity against the target species compared to free-living bacteria (Wang &Coyne 2020) was obtained. It is considered that a suitable immobilized carrier material is critical for keeping the activity of immobilized bacteria (Chen et al. 2013 Medrano et al. 2013), and keeping microbial inhibition reagents enriching in upper water can take fully advantage of algicidal bacteria. It is assumed that SA was employed as the loaded material for greater biocompatibility and environmental safety to combine the gathered algicidal bacteria and form the capsule. Furthermore, ethyl cellulose (EC) was insoluble in water and exhibits good chemical, physical and mechanical properties (Lin et al. 2018). EC was chosen to coat on the surface of the capsule to keep the capsule oating on the water.
It is supposed that algicidal bacteria enriching in the oating capsule structure is a feasible approach to provide a desirable microbial population.
In current study, algicidal bacteria HL, was merged by the protoplast fusion, exhibited a better performance in the inhibition of M. aeruginosa. The algicidal bacteria HL was mixed with SA to prepare the capsule, and EC was coated on the surface of the capsule to keep oating. Scanning electron microscope (SEM) was used to characterize the morphology of algicidal bacteria HL in the capsule. The algicidal e ciency of prepared capsules on algae cells and chlorophyll-a was evaluated and compared with free-living bacteria. The algae inhibition process of capsules was speculated by the changes in algal uorescence and oxidative stress biomarkers in M. aeruginosa. This study can provide an effective approach for improving the e ciency of algicidal bacteria in the microbial treatment of HABs. Materials And Methods 2.1. Experimental algae and bacteria M. aeruginosa (No. FACHB-905) was purchased from the Institute of Hydrobiology, Chinese Academy of Sciences (Wuhan, China). The detailed cultivation steps are described in Text S1. Bacillus sp. HL, formed by the protoplast fusion of algicidal bacteria R1 and denitrifying phosphorous accumulating bacteria B8. The detailed cultivation steps and identity information are described in Figure S1 and Table S1-S2. According to the growth curve of algicidal bacteria HL, HL was cultured in beef-protein medium for 12 h at the late logarithmic growth stage and cultures were diluted with beef-protein solid medium to OD 600 of 1.5 before use. The supernatant was discarded after centrifugation of the diluted culture, and the bacterial suspension were obtained by the addition of deionized water. The prepared bacterial suspension was mixed with SA solution, and stirred by the magnetic mixer at 750 rpm. SA capsules were prepared by the dropping injection of the obtained mixing slurry. The mixed slurry was poured into the constant pressure hopper (The distance between the exit and the curing agent is set in 10 cm) to obtain spherical capsules. The spherical capsules were remained in the curing agent for 25 min to ensure gel reaction occur completely, and then ltered by strainer, washed with deionized water for three times and dried at 35 ℃.

Preparation of algicidal bacteria capsule
The dried capsules were placed in the EC solution and stirred for 25 min, and ltered, were further washed by anhydrous ethanol and deionized water for three times. The prepared capsules were dried in the oven at 35 ℃ and nally stored in the refrigerator at 4 ℃.

Factors affecting performance of algicidal bacteria capsule
SA concentration was set in 0.5, 1, 1.5, 2, 2.5, 3, 3.5% (w/v). Both CaCl 2 and EC concentration were examined by 1, 2, 3, 4, 5% (w/v). 50 mL of the bacterial suspension was mixed with different concentrations solutions of SA, CaCl 2 and EC to prepare capsules with same method. The prepared capsules were introduced to examine the in uences of SA, CaCl 2 and EC concentrations on and 7-day algae dissolution rate of M. aeruginosa. The concentrations of the above embedding bacterial suspensions were 50 mL and capsules prepared above were placed in M. aeruginosa to determine the oating performance and 7-day algae dissolution rate.

Characterization of algicidal bacteria capsule
The prepared capsules were treated with 2.5% glutaraldehyde solution for 3 h at room temperature and washed with deionized water. Then, the capsules were dipped in ethanol of gradient concentration (50%, 60%, 70%, 80%, 90%) for 10 min each. The morphological feature of the prepared capsules was observed by SEM (Nova Nano 450, USA).

Determination of algae growth
The growth of M. aeruginosa was determined by concentrations of cell density and chlorophyll-a. 500 mL of M. aeruginosa at logarithmic growth stage (7.5×10 5 cells·mL − 1 ) were selected for simulating natural water containing algae, and the magnetic stirrer was equipped to simulate the disturbance of the real water body. The concentration of M. aeruginosa was measured at OD 680 every 24 h. The density of algae cells was measured by the optical density method and blood cell counting (Yuan et al. 2020). The cell density was calculated using the following established standard curve: y=168.16x-1.11 R 2 =0.99, where y represents the cell density, 10 5 cells·mL −1 , and x represents the absorbance of M.
aeruginosa measured at 680 nm. The concentrations of cellular pigments, including chlorophyll-a, were determined according to a previous study (Siedlewicz et al. 2020), as shown in Eq. (3). The formula for the inhibition rate of M. aeruginosa was calculated using the formula of IR = (1 where N represents the cell density of the experimental group, cells·mL − 1 , N 0 represents the cell density of the control group, cells·mL − 1 , and IR represents the inhibition ratio, %.

Determination of algal uorescence
The algal uorescence intensity was measured using ow cytometry (BD Accuri TM C6 plus, USA), with an argon excitation laser (15 mW, 488 nm). Capsules were co-cultured in algal solution and BG11 medium was used as the control. On the 7th day, the experimental and control group were added with Propidium iodide (PI, Solarbio-C0080, China) solution and in complete darkness at room temperature for 30 min. PI is a uorescent dye that binds only to DNA in dead or membraneruptured cells (Gumbo et al. 2014, Xiao et al. 2011). The uorescence detection wavelength was set at 560-590 nm. PI uorescence was collected by channel FL2 to detect cell membrane integrity, phycoerythrin (PC) uorescence was collected by channel FL4, the uorescence wavelength was greater than 630 nm, and chlorophyll-a uorescence was collected by channel FL3. The obtained data were analyzed using owjo_v10.

Determination of oxidative stress biomarkers
Algae can enhance the antioxidant effect through the antioxidant enzyme system and improve the resistance of the organism to adversity ). The damage degree of algae cells at the logarithmic growth stage (7.5×10 5 cells·mL − 1 ) at 1th, 4th, 7th day with capsules containing different concentrations of bacterial suspension (50 mL, 100 mL and 150 mL) was analyzed. The control group was set up by capsules without the doping of algicidal bacteria HL. Enzyme activity of the capsules treated algae cells was determined by measuring contents of SOD, POD, CAT and MDA using the corresponding chemical assay kit (Nanjing Jiancheng Institute of Biological Engineering, China).

Reusability
The reusability test (three consecutive cycle) of capsules is conducted by placing capsules into M. aeruginosa, effect of capsules was accessed by determining the concentration of M. aeruginosa. The capsules were taken out every cycle and dried for next cycle experiment.

Statistical analysis
All data in this study were obtained by three replicates and expressed as mean ± standard deviation (n = 3). One-way analysis of variance (ANOVA), and differences were conducted through SPSS 22 (IBM, USA), with statistically signi cant when P < 0.05.  (Fig. 1). Usually, a large amount of free Ca 2+ is still absorbed on the prepared SA capsules surface when the capsules are immersed into the EC solution, and the oxyethyl in EC is coated on the surface of SA capsules due to the electrostatic force. Moreover, part of hydroxyl groups in EC could interact with carboxyl groups in SA and form hydrogen bonds (Zhao et al. 2021). To ensure the oating performance and 7-day algae dissolution rate of capsules, the usage concentrations of SA, CaCl 2 and EC during the preparation process are need to be optimized.

Effect of CaCl 2 concentration
The type and usage concentration of the curing agent affect the stability and activity of immobilized algicidal bacteria (Sarma &Pakshirajan 2011). In this study, CaCl 2 was selected as the cross-link to promote the formation of capsule gel components. 7-day degradation e ciency of M. aeruginosa and oating performance of capsules with 1-5% CaCl 2 were evaluated and in Fig. 2b. It was observed that the maximum 7-day algae dissolution rate of 71.09% ± 0.79% and oating performance of 81.53% ± 0.76% (P < 0.05) with the use of 3 g 100 mL − 1 of CaCl 2 . The amount of Ca 2+ was not enough to cause crosslinking reaction and combine with antiparallel chains at low concentrations of 1-2 g 100 mL − 1 , which led to a poor embedding of algicidal bacteria (42.90% ± 2.64%, 50.68% ± 0.96%) and the fragile structure of capsules. Gelling process usually starts from outer towards inner core of capsules (Bennacef et al. 2021 concentration increased to 4-5 g 100 mL − 1 , which led to a reduction in porosity and the over-sealing internal structure of capsules.

Effect of EC concentration
1-5% of EC was coated on the surface of SA capsules, and the 7-day degradation e ciency of M. aeruginosa and oating performance of capsules were shown in Fig. 2c. It was observed that 68.33% ± 1.06% of M. aeruginosa was dissociated with the use of EC at the concentration of 1 g 100 mL − 1 . As EC concentration increased to 2-3 g 100 mL − 1 , algae dissolution rate was promoted to 70.39% ± 2.22% and 73.45% ± 0.24%, respectively. With further increase in concentration of EC to 5 g 100 mL − 1 , a sharp decline to 28.58% ± 0.78% was observed. The oating rate increased with the increase in EC concentration, and oating performance was signi cantly affected when the usage concentration of EC was at 2-5 g 100 mL − 1 (P < 0.05). 3 g 100 mL − 1 of EC was considered as the suitable concentration for capsule coating. The coating concentration of EC is related with the mechanical strength and biological activity of capsules. Poor surface coating resulted in the sinking of the capsule and excessive coating of EC declined size of pores in capsules, which were not favored to promoting degradation e ciency of capsules for M. aeruginosa.

Characterization of algicidal bacteria capsule
The optimal preparation conditions of algicidal bacteria capsules were set in SA, CaCl 2 , EC usage concentrations of 2%, 3%, 3%, respectively. When bacterial suspension was 150 mL, 87.73% ± 0.21% of capsules is oated and 55.25% ± 2.78% of algicidal bacteria HL is embedded in capsules. Photos and SEM of algicidal bacteria capsules were depicted in Fig. 3. The appearance of prepared algicidal bacteria capsules was white and spherical. The micro surface of algicidal bacteria capsules was rough and porous ( Fig. 3c-d). The porous gaps were favored to cell adhesion and the nutrient diffusion. The morphology of algicidal bacteria HL with rod shaped and a length of about 2 µm was observed. It is clearly showed aggregated rod-shaped bacterial cells were immobilized inside capsules. compared with free-living bacteria. It was found that 7-day algae dissolution rate though with lower embedding concentration of 50 mL still could reach 77.67% ± 1.14%, and algae dissolution rate was promoted by 11.05% comparing with free-living bacteria. These results indicated that capsules could provide a safe and stable environment for algicidal bacteria HL and algae dissolution rate is not impacted by environment factors.

Chlorophyll-a
Chlorophyll-a is an important indicator to re ect growth status of algae. Chlorophyll-a can bind with proteins through noncovalent bonds and deliver light to the central pigment in the reaction by inducing resonance (Zhang et al. 2013). Photosynthetic pigment destruction hinders photosynthesis of algae and inhibits the normal growth of algae. Figure 4 showed in uences of different concentrations of algicidal bacteria (50 mL, 100 mL and 150 mL) on algae dissolution and chlorophyll-a. There was a dramatic increase in the concentration of chlorophyll-a in control group. The concentration of chlorophyll-a declined at 5th day as the introduction of different concentrations of algicidal bacteria. The content of 7-day Chlorophyll-a with the addition of 150 mL embedding algicidal bacteria was 764.01 ± 162.83 mg·m − 3 , and the best algae dissolution rate was 88.54% ± 1.23%, which was higher than free-living bacteria by 14.72%. The change in the concentration of chlorophyll-a was consistent with algae cell. The above results indicated that capsules were more effective in improving algae dissolution.
3.4. Effects of algicidal bacteria capsule on the algal uorescence Figure 5 shows the two-dimensional distribution point diagram of PI/chlorophyll-a uorescence before and after the use of capsules with ow cytometry. In Fig. 5a-b, Q1 region is assigned to PI − / chlorophyll-a + cells, which indicates the cell membrane of M. aeruginosa is intact and in regular physiological state. Q2 region represents PI + / chlorophyll-a + cells, which suggests that though the appearance of algae cells may be same as regular cells, the cell membrane of M. aeruginosa is damaged. Q3 region represents PI + / chlorophyll-a − cells, is usually considered as death cells. Q4 region represents PI − / chlorophyll-a − cells, is usually assigned as double negative cells, which indicates most algal cells are dissociated. The dead algae cells refuse to dye PI because the DNA / RNA that can bind to PI in cells is destroyed. It is believed that no PI uorescence is detected and chlorophyll-a uorescence is strong in Q1 zone, and both PI and chlorophyll-a uorescence are detected. PI uorescence is strongly appeared in Q3 zone while chlorophyll-a uorescence is absent. There is no PI uorescence and chlorophyll-a uorescence detected in Q4 zone. 91.0%, 0.25%, 0.39% and 8.39% of M. aeruginosa in the control group were live, cell debris, dead and selective permeabillity, which were displayed in the Q1, Q2, Q3 and Q4 zones. Most dots shift left remarkably, for example, 28.4% and 64.5% of cells shifted to Q2 and Q3 zones, which suggested that the live cells population of M. aeruginosa decreased with the application of capsules. Figure 5c-d and e-f presents chlorophyll-a and PC uorescence of control and capsules respectively. Similar population shifting after treating with capsules (Hadjoudja et al. 2009). As the working of capsules, the fore peak of chlorophyll-a and PC uorescence appeared and back peak decreased, suggesting that algae cells were destroyed and even scrapped.

Effects of algicidal bacteria capsule on the antioxidant system of M. aeruginosa
Antioxidant reaction is crucial for algae to resist interference from the external environment ). Effects of capsules on the antioxidant system of algae is helpful to reveal the process of algicidal bacteria capsules dissociating M. aeruginosa. Figure 6 shows in uences of different embedding algicidal bacteria HL concentrations (50 mL, 100 mL, 150 mL) on the antioxidant enzyme system of M. aeruginosa by accessing activities of SOD, POD, CAT and MDA on the 1th day, 4th day and 7th day.
Exogenous substance induces increase in various antioxidant active enzymes remove the harmful components and strengthen the defense function of algae. On the 1th and 4th of the experiment, SOD, POD and CAT show the similar tendency, which signi cantly increased compared with control group (P < 0.05, Fig. 6a-c), and the highest SOD, POD and CAT activity was observed at the 4th day with the treatment of 150 mL algicidal bacteria HL embedding capsules group. While the decreases in SOD, POD and CAT in the late period of the experiment, SOD, POD and CAT decreased to varying degrees, which also con rmed that the antioxidant enzyme system of algae had been greatly traumatized or even incapacitated with the usage of capsules. This result is similar to previous experiments (de la Rosa et al. 2020), in which SOD increases signi cantly initially by exogenous substances, while SOD content will decrease when it exceeds the threshold value. SOD is generally considered as the rst barrier of the antioxidant system in algae cells and deal with the excess superoxide and convert it into H 2 O 2 in order to avoiding the damage of reactive oxygen species (ROS) and superoxide (Yu et al. 2021 MDA is often recognized as an indicator of the degree of membrane oxidative damage. During the experiment, MDA content in all treatment groups were signi cantly higher than in the control group (Fig. 6d, P < 0.05), which indicated that oxidative stress reactions occurred in M. aeruginosa after the treatment of capsules. The signi cant increase for MDA in the 4th and 7th days indicated that the degree of lipid peroxidation of algae was deeper. The disruptive membrane system stability of M. aeruginosa is one of the main mechanisms of damage to algae cells, which indicated that free radical formation and removal of dynamic balance is disrupted, the excess free radicals play a destructive role in algae cell macromolecules, decreasing their unsaturation, membrane uidity and increasing membrane permeability ).

Reusability
The reusability of capsules is also important during practical applications. Therefore, the stability of algicidal bacteria capsules (50 mL) were evaluated by three repetitive recycle experiments and results were represented in Fig. 7. The algicidal bacteria capsules after the experiment were recovered and immersed in deionized water to remove residual M. aeruginosa. The dissolution e ciency of M. aeruginosa by algicidal bacteria capsules declined with the increase in the use time, which can be attributed to the pore block and structure collapse of capsules (Daâssi et al. 2014). Even so, three repeated experimental results demonstrated that the dissolution e ciency of M. aeruginosa by algicidal bacteria capsules was still higher than 68%, which is just reduced by 10% than rst time. It is considered that the prepared capsules are stable in the dissolution e ciency of M. aeruginosa, which suggests that the capsules can be recovered and be used repetitively. The prepared capsules in this study can be used in the control algae in summer by throwing into the lake or rivers at June and July. The capsules in the lakes or rivers cannot be salvaged in the next two or three months. Most of algicidal bacteria can remain and survive due to desirable environment and popular community in the capsules, and play signi cant role in the dissolution of algae for ages. Thus, algae blooms in summer can be effectively inhibited by throwing capsules in the target waters, and capsules can be salvaged over the summer and reused next year.

Conclusions
In this study, the capsule was prepared by sodium alginate mixing with algicidal bacteria well and ethyl cellulose to protect bacteria from leaking and from environmental change. Experiments results demonstrated good retention of Bacillus sp. HL in SA-EC capsules. With the optimal concentrations of SA, CaCl 2 , EC (w/v) and bacterial suspension at 2%, 3%, 3% and 150 mL, the oating and embedding rate of capsules were 87.73% ± 0.21% and 55.25% ± 2.78%, respectively. In the simulated treatment of M. aeruginosa, capsules were signi cantly more effective than free-living bacteria in dissolution of both algal cells and chlorophyll-a. Moreover, though the capsules are reused repetitively for 3 times, more than 68% of M. aeruginosa biodegradation rate are still obtained. The results of algal uorescence indicated that the photosynthetic mechanisms of M. aeruginosa were signi cantly inhibited by algicidal bacteria capsules. The synchronous increase in content of SOD, POD and CAT indicated that capsules could cause oxidative stress for M. aeruginosa, and signi cant increases in MDA content, suggesting ROS were not completely removed in a short time and lipid peroxidation damage occurred. Overall, this study revealed that capsules immobilized algicidal bacteria HL may be an environmentally friendly and reusable approach for the control of M. aeruginosa.

Declarations
Data availability The datasets used and/or analysed during the current study may be made available from the corresponding author on reasonable request.

Funding
This work was nancially supported by Major science and technology projects for water pollution control and treatment (2017ZX07202-004) and Changshu city science and technology development project (Social development, CS202005). Ethics approval and consent to participate Not applicable.

Consent for publication
Not applicable.

Figure 1
Preparation procedure and mechanism of algicidal bacteria capsule Effects of (a) SA, (b) CaCl2, (c) EC usage concentrations on the 7th algae dissolution rate and oating performance of capsules    Algae dissolution rates of capsules during three repetitive times

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. Supplementaryinformation.docx