Removal of Microcystis aeruginosa cells using the dead cells of a marine filamentous bacterium, Aureispira sp. CCB-QB1

Introduction

Harmful blooms of cyanobacteria (cyanoHABs) by toxic cyanobacterial species have been recognized in many countries (Briand et al., 2003) and generate direct economic damages because of the pollution of freshwater environments, as sources of drinking water, agriculture, fishing, and industry water (Hamilton, Salmaso & Paerl, 2016). CyanoHABs also seriously affect health problems in humans and animals because of the production of toxins such as microcystin and anatoxin (Azevedo et al., 2002; Katırcıoğlu, Akin & Atici, 2004). For instance, microcystin and its derivatives disrupt the structure and function of the liver and cause deaths by liver hemorrhage (Azevedo et al., 2002; Katırcıoğlu, Akin & Atici, 2004). CyanoHABs occur in many countries in tropical and temperate regions. In addition, it is known that climate change, such as rising CO₂ and global warming, is likely to stimulate the development of harmful cyanobacterial blooms in eutrophic waters (Visser et al., 2016).

To remove microalgae, including cyanobacteria cells, from the water column, several types of flocculants are used. The flocculants are mainly classified into three groups, such as inorganic flocculants (e.g., polyaluminum chloride (PAC) and ferric chloride (FeCl₃)), organic synthetic flocculants (e.g., polyacrylamide derivatives), and some kinds of clay or bioflocculants mainly consisting of polysaccharides produced by microorganisms (Salehizadeh & Shojaosadati, 2001). Inorganic flocculants, such as PAC and FeCl₃, were low-cost materials and exhibited a high removal efficiency on M. aeruginosa cells (Li et al., 2015; Sun et al., 2012; Zhang et al., 2011). However, a high dosage of the flocculants is required for its high removal efficiency. In addition, the sludge containing inorganic flocculants consisted of heavy metals is toxic to the environment (Fast, Kokabian & Gude, 2014). Especially, aluminum may be one of the risk factors for Alzheimer’s disease (Flaten, 2001; Fast, Kokabian & Gude, 2014).

Polyacrylamide derivatives are extensively used to treat natural water and sewage because of their high removal efficiency on pollutants (Kurenkov, Hartan & Lobanov, 2002). Polyacrylamide derivatives are also used in removing microalgae, including M. aeruginosa (Pugazhendhi et al., 2019; Cai et al., 2021). However, acrylamide is considered a probable human carcinogen and mutagen in vitro and in vivo experiments using animals (Dearfield et al., 1988; Dearfield et al., 1995). Thus, there are concerns about their toxicity and less biodegradability.

To date, bioflocculants, which are non-toxic and biodegradable compounds, have attracted attention as an alternative to chemical flocculants. Bioflocculants are biological polymers, mainly extracellular polysaccharides (EPS) or proteins, produced by living organisms, particularly plants and microbes. As the plant-based natural coagulants, some seeds and fruit/ peels, such as Cicer arietinum, Moringa oleifera, and cactus, are reported (Kumar, Othman & Asharuddin, 2017). Bioflocculants composed of EPS produced by bacteria belonging to genera Bacillus (Deng et al., 2003; Zheng et al., 2008), Corynebacterium (He, Li & Chen, 2004), Klebsiella (Wang et al., 2007), Paenibacillus (Li et al., 2013), Proteus (Xia et al., 2008) and Pseudomonas (Azzam & Tawfik, 2015; Liu et al., 2016) were also found during past decades. These bioflocculants were exhibited high removal efficiency (80–99.6%) on negatively charged clay particles, kaolin. In addition, several investigations demonstrated that the bioflocculants were helpful for removing M. aeruginosa. For instance, Sun and colleagues reported that bioflocculant, EPS-1, produced by Bacillus amyloliquefaciens could remove approximately 88% of M. aeruginosa cells (Sun et al., 2015a). Bioflocculant produced by Pseudomonas aeruginosa ZJU1 combined with kaolin, CaCl₂, and KAI(SO₄)₂ also exhibited high removal efficiency on M. aeruginosa cells (Sun et al., 2015b). The investigation conducted by Zhang et al. (2019) demonstrated that the bioflocculant produced by Halobacillus sp. strain H9 also successfully removed M. aeruginosa cells. Besides plants and microbes, another type of polysaccharide, chitosan, produced from shells of crustaceans has also attracted much attention as a bioflocculant to remove M. aeruginosa. The study conducted by Park et al. (2020) demonstrated that chitosan fiber removed almost 89% of M. aeruginosa cells. Ma et al. (2016) also reported that chitosan-aluminum chloride combined coagulants exhibited high removal efficiency (>90%) on M. aeruginosa cells. Thus, bioflocculants consisted of non-toxic and biodegradable biopolymers might be useful for removing M. aeruginosa cells. However, the high cost of the natural macromolecules, especially EPS from microorganisms, limits their application.

In a previous investigation, we reported that a marine filamentous bacterium, Aureispira sp. CCB-QB1, formed cell aggregates in the presence of CaCl₂ (Furusawa, Hartzell & Navaratnam, 2015). In addition, the dead cells of Aureispira sp. CCB-QB1, which has a negatively charged cell surface, could absorb negatively charged clay particles, kaolin, and form aggregates with kaolin particles in the presence of CaCl₂ (Hasyimah, Furusawa & Amirul, 2021). Our investigation demonstrated that the precipitation might be occurred by neutralizing the negative charge of both kaolin and the dead cells by adding a divalent cation, CaCl₂. Based on the result, we predicted that M. aeruginosa cells might also be precipitated by the dead cells in the presence of CaCl₂ (Hasyimah, Furusawa & Amirul, 2021). Furthermore, the preparation of the dead cells was more straightforward than that of known bioflocculants consisted of polysaccharides because the precipitation and purification steps required in known bioflocculants were not necessary to prepare the dead cells. This information indicates that the dead cells have the potential for an inexpensive flocculant compared with the other bioflocculants based on extracellular polymers. In this study, to confirm our prediction, we performed to remove M. aeruginosa cells by using the dead cells from aqueous samples under various conditions. To the best of our knowledge, this is the first report to remove the M.aeruginosa using the dead cells of a filamentous bacterium.

Material and Methods

Statistical analysis

All results were represented by the mean ± standard deviation (SD). All data were subjected to one-way analysis of variance (ANOVA) with Tukey tests to examine significant differences between individual mean values. All statistical tests were carried out using SPSS Version 27.0 (IBM). Differences were considered statistically significant at P values <0.05. Independent t-test was used to compare the differences between samples with and without CaCl₂ and samples with and without FeCl₃. The analysis was also conducted by using SPSS Version 27.0.

Results

Aggregate formation of the dead cells with NISE 102 cells

To test the absorption ability of the dead cells of CCB-QB1 against NISE 102 cells, the dead cells of CCB-QB1 and NISE 102 cells were mixed in 10 mM HEPES buffer (pH 7.6) with and without 7 mM CaCl₂. Figure 1A showed that both cells formed small aggregates within 3 min incubation period only in the presence of CaCl₂. In contrast, the cell aggregates were not observed in the absence of CaCl₂ (Fig. 1A). In light microscopic observation, isolated cells of NISE 102 and QB1 were observed in the absence of CaCl₂ (Fig. 1B, top). On the other hand, NISE 102 cells were surrounded and trapped by many CCB-QB1 cells in the presence of CaCl₂ (Fig. 1B, bottom), indicating that the dead cells of CCB-QB1 are capable of removing NISE 102 cells.

Effect of CaCl₂ on the removal efficiency

To confirm the effect of CaCl₂ against the removal of NISE 102 cells, 50, 20, 5, and 1 mM of CaCl₂ were added to the suspension of NISE 102 cells. Although we expected that high CaCl₂ concentration conditions, 50 mM and 20 mM, stimulated robust removal of the cells compared to 5 mM samples, the removal rate of the three samples was quite similar (75. 11, 76.44, and 75.44%, respectively) (P > 0.05) (Fig. 3). On the other hand, the sample added of 1 mM of CaCl₂ exhibited significantly lower removal efficiency (30.05%) (P < 0.05) than those of samples containing a higher concentration of CaCl₂ (Fig. 3). The results showed that the minimum effective concentration of CaCl₂ for removing NISE 102 cells was 5 mM, and the concentration was used in the following experiments.

SEM observation of cell aggregates

The aggregates were observed under SEM to investigate how dead cells formed the aggregates with NISW 102 cells. In the sample without FeCl₃, big aggregates were observed (Fig. 5A), and many NISE 102 cells were trapped in the aggregates (Fig. 5B). The dead cells that possessed filamentous structures got entangled in each filamentous cell. The dead cells tightly attached NISE 102 cells, and the entangled cells captured NISE 102 cells (Figs. 5C and 5F). Big aggregates were also observed in the sample with FeCl₃ (Fig. 5D). Many NISE 102 cells were also trapped in the aggregates as well as the sample without FeCl₃ (Fig. 5E). Compared to the sample without FeCl₃, the cells were packed more tightly in the aggregates (Fig. 5F).

Discussion

In this study, we tried to apply the dead cells of CCB-QB1 for removing harmful algae, M. aeruginosa. Our previous investigations demonstrated that CCB-QB1 cells aggregated in the presence of 7 mM CaCl₂ (Furusawa, Hartzell & Navaratnam, 2015), and the dead cells of CCB-QB1 formed aggregates with kaolin clay particles in the same condition (Hasyimah, Furusawa & Amirul, 2021). In keeping with these data, we expected that the dead cells of CCB-QB1 could also form aggregates and precipitate on M. aeruginosa in the presence of CaCl₂. To confirm the hypothesis, the dead cells of CCB-QB1 and NISE 102 cells were mixed in the buffer with or without CaCl₂. As shown in Fig. 1A, although no cell aggregates were observed in the absence of CaCl₂, the dead cells could make aggregates with NISE 102 cells in the presence of CaCl₂. In addition, the removal efficiency of NISE 102 cells decreased under low CaCl₂ concentration (1 mM) (Fig. 3), indicating that CaCl₂ is crucial for removing NISE 102 cells. Several investigators described that the cell surface of M. aeruginosa is negatively charged in water (Hadjoudja, Deluchat & Baudu, 2010; Shi et al., 2016). Generally, the net charge of the bacterial cell surface is also negative (Corpe, 1970; Li et al., 2019; Pajerski et al., 2019). Our previous investigations also indicated that the cell surface of CCB-QB was negatively charged (Furusawa, Hartzell & Navaratnam, 2015). However, the dead cells of CCB-QB1 could adsorb negatively charged materials, such as kaolin clay particles, in the presence of CaCl₂ because Ca²⁺ might help neutralize the electrical forces (Hasyimah, Furusawa & Amirul, 2021). Sato and colleagues described that the addition of EPS produced by cyanobacteria and 1,000 mg/L of calcium (approximately 9 mM) was stimulated the colonization of M. aeruginosa cells (Sato et al., 2017). The addition of EPS and calcium decreased the zeta potential of M. aeruginosa cells and promoted its colonization. However, when only EPS was added, the zeta potential hardly decreased (Sato et al., 2017), resulting in inhibiting the colonization. These investigations revealed that Ca²⁺ plays an important role in neutralizing the charge of M. aeruginosa cells. Thus, it is expected that the negative charge of NISE 102 and the dead cells might also be neutralized by Ca²⁺, and the neutralization may eliminate repulsive electron forces between the dead cells and NISE 102 cells. As a result, Ca²⁺ might facilitate the biosorption of the dead cells against NISE 102 cells.

FeCl₃ was used as a flocculant for removing M. aeruginosa cells from aqueous samples (Li et al., 2015; Hao et al., 2016; Gao et al., 2012). In this study, we also performed to use FeCl₃ as a coagulant. Figure 4A showed that the removal efficiency of NISE 102 cells by a low concentration of FeCl₃ was enhanced by adding CaCl₂. Figure 5F also showed that the dead cells were more tightly aggregated in the presence of CaCl₂ and FeCl₃. The absorption of Fe(OH)₃ generated from FeCl₃ by organic matters could explain this result (Amuda & Amoo, 2007). Besides that, Liu et al. reported that Ca²⁺ increased the zeta potential of Fe(OH)₃ and facilitated Fe(OH)₃ floc aggregation (Liu et al., 2007). These investigations indicated that the increase of zeta potential and the promotion of floc aggregation by Fe(OH)₃ via the addition of Ca²⁺ enhance the removal of NISE 102 cells. Figure 4A demonstrated that a high removal efficiency (>90%) of NISE 102 cells was observed when the dosage of FeCl₃ was more than 50 mg/L. (Li et al., 2015) also described that the addition of 50 mg/L of FeCl₃ exhibited almost the maximum removal efficiency on M. aeruginosa cells. Hao and colleagues also showed that M. aeruginosa cells could aggregate when the dosage of FeCl₃ was 175 mg/L (Hao et al., 2016). Thus, the addition of a high dosage of FeCl₃ was required to remove M. aeruginosa cells. Figure 4B demonstrated that the addition of only 10 mg/l of FeCl₃ significantly improved the removal efficiency. These results also revealed that the combination of the dead cells and CaCl₂ could reduce the dosage of FeCl₃ for removing M. aeruginosa cells.

Sun and colleagues reported that maximum removal efficiency by bioflocculants from B. amyloliquefaciens strain DT and P. aeruginosa strain ZJU1 was reached within 10 min (Sun et al., 2015a; Sun et al., 2015b). The bioflocculant from Klebsiella pneumonia took 1.6 h for floc formation with M. aeruginosa cells (Xu et al., 2016). In contrast, in this study, the maximum removal efficiency by the dead cells was reached within 3 min (Fig. 2). This result revealed that the dead cells were able to quickly remove M. aeruginosa cells compared to other bioflocculants based on EPS from microorganisms. Another advantage of the usage of the dead cells is that the preparation process of the dead cells is more straightforward than that of other bioflocculants based on EPS from microorganisms. Generally, bioflocculant-producing bacteria were cultured for 2–3 days in a nutrient-rich broth. For instance, several types of saccharides (5.0 g/L sucrose, 2.0 g/L glucose, and 2.0 g/L maltose) were added to the culture broth to produce a bioflocculant from P. aeruginosa ZJU1 (Sun et al., 2015b). In addition, two volumes of cold ethanol were required to recover the bioflocculant released into the culture supernatant. Finally, the crude bioflocculant was centrifuged to harvest and dried by heat treatment or freeze-drying. In contrast, the CCB-QB1 cells were cultured in a seawater medium without adding any carbon sources. The incubation period to obtain the cells was 24 h (Hasyimah, Furusawa & Amirul, 2021). After 10 min UV treatment, the dead cells were able to use as a bioflocculant (Hasyimah, Furusawa & Amirul, 2021). Thus, the preparation procedure of the dead cells was quite simple in comparison with the other bioflocculants, suggesting that the production cost of the dead cells might be cheaper than that of known bioflocculants consisted of EPS from microorganisms.

It was reported that some plant-based natural coagulants also successfully removed M. aeruginosa cells. For example, protein-based coagulant extracted using a salt solution (1 M NaCl or 1 M CaCl₂) from Moringa oleifera seeds exhibited a high removal rate (78.9%) on M. aeruginosa cells (Carvalho et al., 2016). Moreti and colleagues also reported that a saline extract from M. oleifera seeds could remove M. aeruginosa cells effectively (de Oliveira Ruiz Moreti et al., 2019). Interestingly, the extract could adsorb and remove microcystin-LR (de Oliveira Ruiz Moreti et al., 2019). The preparation procedure of the extract using saline solution was more straightforward than that of bioflocculants consisted of EPS. However, many pretreatment processes, such as dehusking, powdering, and drying, were required. Of course, the large agricultural land and long cultivating period to obtain seeds are necessary for culturing M. oleifera. In contrast, the biomass production of CCB-QB1 is faster and easier than the plant.

Conclusion

In this study, the dead cells of a marine filamentous bacteria, Aureispira sp. CCB-QB1, was used as a bioflocculant for removing a harmful freshwater microalga, M. aeruginosa. The dead cells were removed about 75% of M. aeruginosa cells with 3 min incubation in the presence of CaCl₂. The optimization of CaCl₂ concentration for removing the M. aeruginosa cells showed that the optimal concentration of CaCl₂ was 5 mM. The removal efficiency by inorganic flocculant, FeCl₃, was improved by the addition of CaCl₂. Besides that, the combination of the dead cells and CaCl₂ could reduce the dosage of FeCl₃ for removing M. aeruginosa cells. These results indicated that the dead cells have the potential for a novel and alternative bioflocculant.

Supplemental Information