Characterization of self-aggregation in Bacillus licheniformis strain RK14

Abstract

Purpose

Some Bacillus licheniformis strain isolated from the environment has been reported to form aggregates during the suspension culture under carbon limitation. The aim of this study is to characterize the aggregation process of B. licheniformis RK14 strain isolated from the soil.

Methods

B. licheniformis RK14 was cultured in a glucose-free lysogeny broth (LB) to confirm the aggregate formation under glucose limitation conditions and compared to model B. licheniformis strains. To characterize the aggregation process of RK14, microscopic analyses and time-lapse observations were conducted. Thioflavin T (ThT) was used to assess the involvement of transient cell elongation in aggregate formation.

Results

Although RK14 did not form aggregate when cultured in a glucose-rich γ-PGA production medium, it was found to self-aggregate when grown in a glucose-free LB. The optimal temperature for aggregation was approximately 40℃. Microscopic analysis showed that the aggregates were composed of viable cells with cell–cell cluster-like structures. Time-lapse observations clarified elongation of individual cells after 1 h of culture, followed by a return to rod-shape while maintaining the aggregation state. Addition of ThT at micromolar concentration inhibits aggregate formation, probably due to the suppression of cell elongation.

Conclusions

The aggregation process of RK14 was similar to the pellicle and submerged biofilm formation process of B. subtilis in terms of morphological changes. It was concluded that transient cell elongation is critical for aggregate formation of RK14 strain.

Background

Flocculation is an aggregation phenomenon of bacterial cells in which they form flocs or clumps. Numerous microorganisms have been found to have floc-forming capabilities (Salehizadeh and Shojaosadati 2001) that can be applied to wastewater treatment. In an activated sludge, the components of flocs typically include polysaccharides, polynucleotides, and proteins (Tago and Aida 1977). However, cells may also aggregate to form cell–cell cluster-like structures without producing excessive extracellular matrix, as in the case of flocculating yeast (Soares 2011). In flocculating yeast, a specific lectin-like protein (Flo protein), only present in flocculating Saccharomyces cerevisiae cells, recognizes and interacts with the carbohydrate residues of α-mannans of neighboring cells. Calcium ions enable the lectins to adopt their active conformations (Soares 2011). There are several advantages of using flocculent yeast strains in alcoholic beverage production (wine, cachaça, and sparkling wine), renewable fuel production (bio-ethanol), biotechnology (production of heterologous proteins), and environmental applications (bioremediation of heavy metals).

Some Bacillus licheniformis strain isolated from the environment have been reported to form aggregates during suspension culture under carbon limitation condition (da Silva et al. 2005). This is a phenomenon not seen in laboratory strains of B. licheniformis or B. subtilis. In that study, although culture engineering based on the substrate feeding condition have been well studied, the aggregate itself have not been characterized from a biological point of view, and the biomolecules that play an essential role in aggregate formation still need to be clarified. In general, Bacillus species have been reported to form biofilms on solid surfaces (Randrianjatovo-Gbalou et al. 2017; Arnaouteli and Bamford 2021). Biofilms are composed of closely associated cells encapsulated in an extracellular matrix mainly composed of polysaccharides, proteins, and DNA. The proteinaceous matrix of Bacillus species biofilms contains amyloid-like fibrils, which have been shown to play a determinant role in the structural integrity of the biofilm (Arnaouteli and Bamford 2021). Furthermore, most of Bacillus species form a pellicle biofilm that develops at the air–liquid interface under static culture conditions. This pellicle formation is a feature of obligate aerobes and a means of efficient oxygen acquisition. Although the mechanism of biofilm and pellicle formation in Bacillus species have been well reported, there is, to our knowledge, no report of characterization and mechanism of aggregate formation of B. licheniformis in suspension culture.

Meanwhile, we previously isolated an RK14 strain of Bacillus licheniformis with high γ-polyglutamic acid (γ-PGA) productivity from soil (Liu et al. 2017). γ-PGA from RK14 cells has a high molecular mass, and an enantiomeric composition that differs from those obtained from other B. licheniformis strains. RK14 strains produced 6.9 g/L γ-PGA without aggregate formation when cultured in semi-synthetic media, mainly containing fructose, glycerol, and L-glutamate at 37 °C.

In this study, B. licheniformis RK14 was cultured in glucose-free lysogeny broth (LB) to confirm the aggregate formation under glucose limitation conditions and compared to model B. licheniformis strains. To characterize the aggregation process of RK14, microscopic analyses and time-lapse observations were conducted. We finally examined the involvement of transient cell elongation in aggregate formation.

Results and discussion

Self-aggregation of B. licheniformis RK14

In our previous study, the RK14 strain was cultured in glucose-rich 3YD or YFEb medium for γ-PGA production (Liu et al. 2017). In this study, the RK14 strain was first cultured in glucose-free LB at 37 °C under shaking culture conditions. During suspension culture, we found that the RK14 strain forms aggregates. After 18 h, the precipitated aggregates were collected into microtubes and photographed as shown in Fig. 1a. The LB culture had significantly much more precipitated aggregates than the 3YD or YFEb cultures. Microscopic observation showed aggregation in LB culture of about several hundred µm² (Fig. 1b), and greater magnification clearly showed the cell–cell cluster-like structure as reported previously (da Silva et al. 2005). Evaluation of cell growth showed similar growth in LB (OD₆₀₀ = 4.9) and in 3YD medium (OD₆₀₀ = 4.4), while OD₆₀₀ = 9.0 and much higher in YFEb medium.

Fig. 1

Aggregate formation and cell growth of Bacillus licheniformis strains RK14, ATCC9945, and ATCC14580 at 37 °C. a OD₆₀₀ and photos of aggregates in microtubes at 18 h with different culture media. OD₆₀₀ was measured after removing aggregates through precipitation. b Microscopic images of aggregates formed by strain RK14 after 18 h of culture in LB

To evaluate whether this phenomenon is unique to strain RK14, we also cultured B. licheniformis ATCC14580 (Rey et al. 2004), which does not produce γ-PGA, and ATCC9945 (Mitsunaga et al. 2016), the typical γ-PGA producing strain, under the same conditions. The ATCC14580 strain grew well in LB and reached an OD₆₀₀ = 7.5, however, no precipitant was observed (Fig. 1a). The ATCC9945 strain also grew well (OD₆₀₀ = 7.7), and a very small amount of precipitate was observed. These results suggest that aggregate formation is unique to the RK14 strain cultured in LB and is not directly related to γ-PGA production.

To examine the effect of culture temperature on aggregate formation, the RK14 strain was cultured in LB at temperatures ranging from 30 to 50 °C. Culturing at 30 °C greatly decreased the amount of aggregate compared with that at 37 °C even though cell growth was similar. Aggregate formation at 42 °C was similar to that at 37 °C. A further temperature increase to 50 °C resulted in decreased aggregate formation despite promoting growth. The amount of aggregate was quantitatively determined by dry weight. As shown in Fig. 2b, the dry weight of aggregates at 37 °C and 42 °C were approximately 0.6 and 0.7 g/L, respectively, and significantly higher than those at both lower (30 °C) and higher (50 °C) temperatures. Thus, the optimal temperature for aggregate formation is approximately 40 °C.

Fig. 2

Aggregate formation and cell growth of the RK14 strain cultured under different temperature conditions in LB. a OD₆₀₀ and photos of aggregates in microtubes at 18 h. b Dry weight of aggregates formed by strain RK14 after 18 h. Data were obtained from 3 independent experiments. Vertical bars indicate standard deviations

Observation of cell aggregates of strain RK14

Aggregate formation is specific for the RK14 strain cultured in LB, and optimal aggregate formation occurs at approximately 40 °C. Furthermore, microscopic observation showed that aggregates have a cluster-structure. Considering that bacterial aggregate formation is usually induced when microbial cells are damaged or partial killed by physiological stress (Ojima et al. 2015, 2021), we evaluated cell viability of the RK14 strain aggregates with live/dead straining. In this assay, live cells were fluorescent green (Calcein-AM) and dead cells were fluorescent red (PI), and three-dimensional images were observed with the CLSM. As shown in Fig. 3, most cells were fluorescent green and only a small number of cells were red, suggesting that the aggregates are mainly composed of viable cells. The merged image showed that dead cells were dispersed within the aggregate structure, indicating that aggregates do not originate from dead cell clumps.

Fig. 3

Confocal scanning laser microscopy images of strain RK14 aggregates with live/dead staining. Green fluorescence shows the living cells stained with calcein-AM; red fluorescence shows the dead cells stained with PI

Next, we performed time-lapse microscopic observations with photographs taken every hour. Figure 4 shows photographs of morphological changes at 1, 2, 4, 6, 8, and 10 h. After 1 h of culture, aggregates had already started to form, and microscopic images showed that most cells are several times longer than planktonic cells (Fig. 4a). The higher magnification image showed the presence of very long cells, especially at the edges of aggregates. At 2 h, the aggregates still included long cells. To confirm whether these long cells were elongated single cells or a chain of several cells, the aggregate at 2 h was stained with FM4-64, a plasma-membrane dye. Visualization of membranes with FM4-64 revealed that these long cells had a clear division plane for each cell unit, suggesting that the long cells were cell chains (Fig. 4b). It seems that these chains were tightly connected to form clusters (Fig. 4a) and many chains are still observed in the aggregate structure. At 6 h, longer chains were observed at the center of aggregates, whereas shorter chains increased at the edges of aggregates, indicating that cell separation had occurred. At 8 h, most cells had returned to rod-shape and at 10 h most chains had disappeared, although the cells maintained the cluster structure. This temporal change in cell morphology was previously reported for B. subtilis pellicle formation (Kobayashi 2007), during which cell chains are formed, followed by clustering and degradation of cells chains. While pellicles are floating aggregates that form at the liquid–air interface, the aggregate formation in this study occurs submerged during shaking test-tube culture. From this point of view, the mechanism of submerged biofilm formation by B. subtilis was recently reported (Sanchez-Vizuete et al. 2022). In that report, cells adhering to the surface firstly form elongated chains before being suddenly fragmented and released as free motile cells in the medium. That switching coincided with an oxygen depletion in the well which preceded the formation of the pellicle at the liquid–air interface. Residual bacteria still associated with the solid surface at the bottom of the well started to express matrix genes under anaerobic metabolism to build the typical biofilm protruding structures. It was concluded that a floating pellicle and a submerged biofilm can form consecutively in the same system. In this study, oxygen supply seems to be higher than when cultured in static wells because the aggregate formation occurs during shaking test-tube culture. It is possible that the higher oxygen supply inhibits the release of free motile cells and leads to the maintenance of aggregates. Thus, the mechanism of aggregate formation in this study may be related to that of floating pellicle or submerged biofilm formation reported previously.

Fig. 4

Morphological changes in RK14 cells during aggregate formation. The RK14 strain was grown under shaking culture conditions in LB at 37 °C. a Time-lapse observation of aggregates at 1, 2, 4, 6, 8, and 10 h. b Aggregates at 2 h are stained with FM4-64_, a membrane-specific dye

Effect of inhibition of cell elongation on aggregate formation

It was confirmed that transient cell elongation is important for aggregate formation of RK14 strain. From the aspect of cell elongation, recent single cell measurements revealed that Thioflavin T (ThT) decreased the elongation rate of B. subtilis at micromolar concentration (Han and Payne 2022). Therefore, we hypothesized that ThT might act as an inhibitor against aggregate formation due to the suppression of cell elongation. Aggregate formation of RK14 strain was examined in the presence of ThT. According to the previous report (Han and Payne 2022), ThT was added in LB at the range from 10 to 100 µM, and cell growth and aggregate formation were determined at 18 h. As results, the OD₆₀₀ slightly decreased with 10, 20, and 50 µM ThT compared to without addition (Fig. 5). These results are corresponding to those of B. subtilis in the previous paper (Han and Payne 2022). At 100 µM, OD₆₀₀ value was 0.1, indicating that higher concentration of ThT has an anti-proliferative effect on RK14 strain. The aggregate formation slightly decreased at 10 µM and dramatically decreased at 20 µM. At 50 µM, no aggregates were observed. Considering that the OD₆₀₀ values at 10 and 20 µM are comparable, ThT seems to act as an inhibitor of aggregation rather than suppressing proliferation. In the previous paper, single cell analysis revealed that the elongation rate decreased by half even at 10µM ThT, suggesting that ThT inhibits aggregate formation by suppressing cell elongation. These results support the conclusion that transient cell elongation is key factor for aggregate formation of RK14 strain.

Fig. 5

Aggregate formation and cell growth of the RK14 strain cultured at 37 °C in LB in the presence of thioflavin T. OD₆₀₀ values and photos of aggregates were obtained at 18 h

Conclusions

In this study, self-aggregation of B. licheniformis RK14 was observed when cultured with LB at 37 to 42 °C with shaking. This phenomenon is not seen in laboratory strains of B. licheniformis or B. subtilis. The aggregates were composed of viable cells with cell–cell cluster-like structures. Time-lapse observation revealed that the individual cells grew longer after 1 h of culture and then returned to rod-shape over time, which is similar to previously reported morphological changes during B. subtilis pellicle formation. Aggregate formation was inhibited by ThT due to the suppression of cell elongation. It was concluded that transient cell elongation is critical during aggregation process of RK14 strain.

Materials and methods

Bacterial strains and media

Bacterial strains used in this study are listed in Table 1. The Bacillus licheniformis RK14 strain, which produces γ-PGA, was the same as that used in previous studies (Liu et al. 2017). B. licheniformis RK14 was originally isolated from the soil. A laboratory B. licheniformis strain ATCC14580, which does not produce γ-PGA, and the typical γ-PGA producing strain ATCC9945 were also used. Preculture of the B. licheniformis strains was performed in lysogeny broth (LB; 10 g/L Bacto^TM Tryptone, 5 g/L yeast extract, and 10 g/L NaCl). Aggregation test cultures of the B. licheniformis strains were performed in LB, typical culture medium 3YD (30 g/L yeast extract and 20 g/L D-glucose), and γ-PGA production medium YFEb (30 g/L yeast extract, 20 g/L D-fructose, 60 g/L sodium L-glutamate⋅H₂O, 20 g/L glycerol, and 30 g/L NaCl). All media were initially adjusted to pH 7.0 using NaOH solution.

Table 1 Bacterial strains used in this study

Aggregation test culture of each B. licheniformis strain

For aggregation test culture, B. licheniformis was precultured in LB for 18 h at 37 °C and subsequently inoculated into test tubes containing 4 mL fresh LB, 3YD, or YFEb to an optical density (OD) at 600 nm = 0.01 and cultured at 37 °C with shaking at 140 strokes/min for 18 h. Thioflavin T (0 ~ 100 µM) was added in LB as an inhibitor for cell elongation (Han and Payne 2022). The effect of temperature was evaluated by culturing at 30, 42, and 50 °C, in addition to 37 °C. Cell growth was measured by OD₆₀₀. After 18 h of culture, the test tubes containing the aggregates and cell suspension were left to stand for 15 min. After 15 min, aggregates precipitated, whereas suspension cells remained in the supernatant. The OD₆₀₀ of the supernatant was measured as an indicator of cell growth. Precipitates were resuspended in a 0.9% (w/v) NaCl solution and left to stand for 15 min. This step was repeated twice to remove cells loosely adsorbed onto the aggregates. The aggregates were transferred to microtubes and photographed using a digital camera. For quantitative assays, the aggregates were collected from the culture broth in ten test tubes and dried at 80 °C and the constant dry weight of aggregates determined.

Live/dead staining and FM4-64 staining

For live/dead staining, the aggregates were resuspended in phosphate-buffered saline (PBS) and then stained with 0.7µM Calcein-AM and 1.3µM propidium iodide (PI) in PBS for 15 min. The excess dye was washed with PBS. The aggregates were observed using a confocal laser scanning microscope (CLSM) (model DM6000B) with TCS SP8 software (Leica, Germany), with excitation at 488 nm and emission at 515 nm for Calcein-AM, and excitation at 538 nm and emission at 617 nm for PI.

For FM4-64 staining, the aggregates were fixed with 4% paraformaldehyde in phosphate buffer (PB) for 10 min. After washing with PBS, the aggregates were stained with 2 µg/mL FM4-64 (Kobayashi 2007) and then observed with the CLSM with excitation at 505 nm and emission at 725 nm.

Time-lapse observation of aggregation

For time-lapse observation of aggregate formation, strain RK14 was inoculated into test tubes containing 4 mL fresh LB to an OD₆₀₀ = 0.01 and cultured at 37 °C with shaking at 140 strokes/min. The aggregate samples were removed from the test tubes and observed under a microscope (BX50, Olympus) every hour.