Development of an efficient electroporation method for rhizobacterial Bacillus mycoides strains

In order to develop a method for electroporation of environmental Bacillus mycoides strains, we optimized several conditions that affect the electroporation efficiency of this bacterium. By combining the optimized conditions, the electroporation efficiency of strain EC18 was improved to (1.3±0.6)×105cfu/μg DNA, which is about 103-fold increase in comparison with a previously reported value. The method was further validated on various B. mycoides strains, yielding reasonable transformation efficiencies. Furthermore, we confirmed that restriction/modification is the main barrier for electroporation of this bacterium. To the best of our knowledge, this is the first systematic investigation of various parameters of electroporation of B. mycoides. The electroporation method reported will allow for efficient genetic manipulation of this bacterium.


Introduction
Bacillus mycoides is a spore-forming Gram-positive bacterium that is commonly found in soil and the rhizosphere. It belongs to the B. cereus sensu lato group, which includes B. cereus, B. thuringiensis, and B. anthracis. B. mycoides has received the least attention among this group because it is not a human pathogen as B. anthracis and B. cereus. Moreover, there are several reports on the insecticidal effect of B. thuringiensis, which stimulated further studies using genetic modification (Turchi et al., 2012). The B. mycoides species features a unique filamentous growth pattern, either being rotated clockwise or counterclockwise (Di Franco et al., 2002). Nowadays, more and more studies on B. mycoides are focusing on their plant-growth promoting activities (Ambrosini et al., 2016;Bargabus et al., 2002;Neher et al., 2009).
In order to study the plant interaction mechanism of B. mycoides, it is necessary to be able to perform molecular genetics studies on this bacterial species. A high transformation efficiency is required to establish the genetic manipulation systems, e.g., enabling gene deletion and mutation within the genome of this organism. Several techniques, including phage transduction, protoplast transformation, natural competence, and electroporation have been applied to incorporate exogenous DNA into Bacillus cells (Barlass et al., 2002;Lu et al., 2012;Romero et al., 2006). Among these methods, electroporation is usually the quickest and most reproducible method. Di Franco et al. (2002) modified the electrotransformation protocol of Macaluso and Mettus (1991) for B. thuringiensis, but the implementation on B. mycoides resulted in a very low efficiency (less than 200 cfu/μg plasmid DNA).
Our preliminary experiments show that by applying the method reported by Ehling-Schulz et al. (2005), which was originally developed for B. cereus, the B. mycoides strain EC18 could be transformed, albeit at a low efficiency. Apart from this, no other protocols have been developed on electroporation methods for B. mycoides. This work aimed at developing an efficient electroporation method for several environmental B. mycoides strains. B. mycoides EC18 which was isolated from the endosphere of potato, and displaying potential plant growth-promoting activity (data not shown), was used as a first testing strain. Factors including growth media, growth phase, electroporation buffer, pulse strength, and incubation time that can affect electroporation efficiency were optimized. As a result, a high electroporation efficiency of (1.3 ± 0.6) × 10 5 cfu/μg DNA) was obtained. Furthermore, we evaluated our optimized protocol on other B. mycoides strains as well, which resulted in electroporation efficiencies ranging from (7.3 ± 2.1) × 10 2 to (1.3 ± 0.6) × 10 5 cfu/μg DNA.

Bacterial strains, plasmids, and media
Escherichia coli strains were grown in Luria broth at 37°C, 220 rpm. B. mycoides strains were isolated from a potato field in Wijster (the Netherlands) (Table 3) and were grown at 30°C, 200 rpm in LB. When necessary, 100 μg/ml of ampicillin or 4 μg/ml of chloramphenicol was added to the culture medium. Plasmid DNA from E. coli was purified using the NucleoSpin plasmid isolation kit (Macherey-Nagel GmbH & Co. Düren, Germany) according to the manufacturer's instructions. A Geobacillus-E. coli shuttle vector pNW33N (Bacillus Genetic Stock Centre) isolated from E. coli MC1061 was used as the plasmid for electroporation protocol optimization. pNW33N isolated from E. coli JM110 was used to test the plasmid methylation effects on transformation efficiency.

Preparation of electro-competent cells
B. mycoides cells were grown overnight in an appropriate medium at 30°C, 200 rpm. The culture was diluted with the same medium to obtain an initial optical density at 600 nm (OD 600 ) of about 0.05 then continually grown. The OD 600 was measured by a Genesys 20 spectrophotometer (ThermoSpectonic, USA) during bacterial growth. When the OD 600 reached the appropriate value, the cell culture was transferred into a 50 mL centrifuge tube and cooled on ice for 10 min. Cells were then collected by centrifugation at 4°C, 3000 × g for 10 min. After being washed four times in corresponding electroporation buffer (pre-chilled), the cell pellets were re-suspended in 1 mL of the electroporation buffer. The resulting electro-competent cells were flash-frozen in liquid nitrogen and stored at − 80°C prior to electroporation.

Cell wall-weakening treatment
Serial concentrations of Glycine or DL-threonine were added when bacterial cultures reached an OD 600 = 0.85 for the cell-wall weakening treatment. These treated bacterial cultures were continued to shake for 1 h. After that, electro-competent cells were prepared with the methods mentioned above.

Electroporation
The electroporation was performed as previously reported method (Peng et al., 2009;Turgeon et al., 2006) with slight modification. 100 μL of frozen electro-competent cells was thawed on ice and mixed with 2-3 μL of plasmid DNA (1 μg). Cells mixed with the same amount of deionized water served as negative control. The mixture was loaded into an electroporation cuvette (2 mm electrode gap, pre-chilled) and exposed to a single pulse in a Gene Pulser System (Bio-Rad, USA) with the settings of 25 μF capacitance, 100, 200, or 400 Ω resistance, and voltage ranges between 5 and 12.5 kV cm −1 . After pulse-shock, cells were immediately added with 1 ml of pre-warmed corresponding growth medium and transferred to a 2 mL Eppendorf tube. After 2 h incubation at 30°C, 180 rpm, the dilutions of the recovered cell culture were plated on LB agar with 4 μg/ml of chloramphenicol. The plates were incubated at 30°C overnight and transformation efficiencies (cfu/μg DNA) were calculated by counting the colonies on plates.

Optimization of growth conditions
For determining the optimal growth medium, we transformed B. mycoides according to the protocol published by Ehling-Schulz et al. (2005). Strain EC18 was grown in different media before preparing the electro-competent cells. Media including LB, LBS, LBSP, 2 ×YT and BHIS were chosen according to reported electroporation methods for Bacillus species (Zhang et al., 2011a(Zhang et al., , 2015. When the OD 600 reached a value of 0.6, the competent cells were prepared by washing the cell pellet with increasing concentrations of ice-cold glycerol (2.5, 5, and 10%). Electroporation was performed with 1 μg pNW33N plasmid DNA and recovered for 2 h with the corresponding growth media. The efficiencies were calculated after 1 day (Table 1) on 4 μg/mL chloramphenicol selection media. Among the five tested media, the super rich medium BHIS showed the highest efficiency of (6.2 ± 1.4) × 10 3 cfu/μg plasmid DNA. Zhang et al. (2011a) reported that the transformation efficiency of B. amyloliquefaciens was positively correlated to concentrations of salts, whereas negatively related to the nutritional ingredient concentration. Surprisingly, we observed an opposite effect of the nutritional and salt concentrations on transformation efficiencies in our study. The transformation efficiency by culturing in the LBSP medium is lower than culturing in LBS medium. Moreover, B. mycoides EC18 could not grow in the hypertonic media NCM and M9YE (data not shown). We hypothesized that B. mycoides EC18 is sensitive to salt concentrations.
According to previous reports, cells of B. cereus and B. thuringiensis collected at early growth-stage show better electroporation efficiency than late growth-stage cultures (Peng et al., 2009;Turgeon et al., 2006), while B. subtilis WB800 has high electroporation efficiency of 1.88 × 10 5 cfu/μg DNA at late growth-stage (OD 600 = 2.2-2.3 ) (Lu et al., 2012). To investigate the effects of the growth phase of B. mycoides on the electroporation efficiency, B. mycoides cells were cultured in BHIS medium to OD 600 from 0.3 to 1.8 for competent cell preparations. Our results indicated that when OD 600 is between 0.9 and 1, the highest electroporation efficiency was obtained (Fig. 1A). According to the growth curve (Fig. 1B), cells are in the early stage of exponential growth. A similar phenomenon was also reported for B. subtilis ZK by Zhang et al. (2015). This indicates that cells from the early exponential phase result in higher electroporation efficiency of B. mycoides.

Optimization of cell wall-weakening agents
The cell wall-weakening agents, glycine and DL-threonine, are widely used to improve the electroporation efficiency of environmental strains. These amino acids can reduce the peptidoglycan bonds and loosen up the cell wall by replacing the L-and D-alanine bridges (Hammes et al., 1973). In this study, EC18 cells were first grown in BHIS medium, and when the OD 600 reached about 0.85, glycine and threonine were added at different concentrations (0%, 1%, 2%, 3%, 4% and 5%). After 1 h of additional incubation, the cells were collected and concentrated to obtain competent cells. Electroporation treatments were performed and the greatest transformation efficiency was obtained in the 2% glycine treatment group. The same concentration of threonine also improved the transformation efficiency, which was however lower than that of glycine (Fig. 2). Cell growth rate was slightly reduced in the presence of both glycine and DL-threonine, and high concentrations of glycine treatments resulted in cellular lysis in the samples Table 1 Effect of growth media on the electroporation efficiency of B. mycoides EC18.
(data not shown). These results demonstrate that the transformation efficiency of B. mycoides has been significantly enhanced by a cell wallweakening treatment. The optimal results were obtained at incubations with 2% glycine.

Effects of buffer on the electroporation efficiency
According to previous studies, the composition of the electroporation buffer has a tremendous impact on transformation efficiency. Xue et al. (1998) improved the transformation efficiency of B. subtilis and B. licheniformis by applying electroporation buffers with high osmolarity. It has also been reported that with the addition of trehalose, the transformation efficiency of B. subtilis can be enhanced dramatically to about 100-fold (Lu et al., 2012). In this study, EC18 cells growing in BHIS medium was added with 2% glycine when the cell density reached an OD 600 = 0.85. After 1 h of additional incubation, cells were collected (OD 600~0 .95) and washed four times in different electroporation buffers (Table 2) to obtain competent cells. The transformation efficiency is shown in Table 2. The buffers consisting of glycerol exhibited better transformation efficiencies than those with sucrose. The highest transformation efficiency was obtained with buffer G containing 10% glycerol, 0.25 M sorbitol, and 0.25 M trehalose. However, fewer transformants were obtained when the sorbitol and trehalose concentration increased to 0.5 M. The addition of salts (buffers K, L, and M) showed no improvement on transformation efficiency. Based on these results, buffer G was used throughout.

Effects of the electric field on the electroporation efficiency
A relatively high electrical field is necessary to create pores in the cells and thus provide a temporary pathway for exogenous DNA. After the electric pulse, they gradually reseal and most cells recover. However, if the electrical field is above a certain level, most pores either do not reseal or reseal too slowly to preserve cell viability (Kotnik et al., 2015). To optimize the electric field for B. mycoides EC18, a gradient of field strength (7.5-12.5 kV cm − 1 ) and resistance values (100, 200, and 400 Ω) at 25 μF was tested. The results are shown in Fig. 3, an electric field of 10 kV cm −1 and resistance of 200 Ω led to the optimal transformation efficiency. This electric field is lower than that of B. subtilis (Lu et al., 2012;Zhang et al., 2015) and closely related species, e.g., B. cereus and B. thuringiensis (Peng et al., 2009;Turgeon et al., 2006). These results indicate that B. mycoides is more vulnerable to high electric fields compared to other Bacillus species.

Effects of recovery time
After an electric pulse, the injured bacterial cells need several hours to retain viability and to allow phenotypic expression. To optimize the recovery time, the pulse-shocked cells were immediately transferred to BHIS medium and incubated at 30°C, 180 rpm for 2-7 h. As shown in Fig. 4, the transformation efficiency was increased as the incubation time extended. Remarkably, the transformation efficiency reached 1.2 × 10 5 cfu/μg DNA after 5 h of incubation. Further elongation of the incubation time only led to a slight increase in transformation efficiency. Hence, we used 5 h of recovery time for B. mycoides. Xue et al. (1998) pointed out that the increased transformation efficiency at the longer recovery times might be due to the division of transformed cells. Zhang et al. (2011b) also used a long recovery time of 5 hrs for electroporation of Arthrobacter.

Application of the optimized transformation protocol on other B. mycoides strains and the effects of plasmid methylation
In order to validate the optimized protocol in other environmental B. mycoides isolates, we combined all the improved factors and applied the method on 5 other strains in addition to strain EC18. Since the restriction of methylated DNA in several Bacillus species is known as a major barrier to transformation and genetic manipulation (Sitaraman and Leppla, 2012), we also compared the transformation efficiency of plasmids that were isolated from E. coli methylation strain MC1061 (dam + /dcm + ) and methylation-deficient strain JM110 (dam − /dcm − ). As shown in Table 3, for the plasmid isolated from E. coli MC1061, only strain EC18 showed high transformation efficiency of (1.1 ± 0.6) × 10 5 cfu/μg DNA. Notably, strain SB4 was transformed at a much lower level of efficiency and the other strains were not transformable at all. However, when non-methylated plasmid DNA was used, this protocol can efficiently be applied for all the tested B. mycoides strains. The transformation efficiencies are ranging from (7.3 ± 2.1) × 10 2 to (1.3 ± 0.6) × 10 5 cfu/μg DNA. These results indicate that our optimized protocol can be widely used in B. mycoides strains. Methylation of transmitted DNA can clearly affect the efficiency of transformation, with the exception of strain EC18, which only showed a slight difference between methylated and non-methylated DNA.

Conclusion
This paper describes an efficient electroporation method for several environmental B. mycoides strains, which is reproducible and convenient. By using this method, we obtained transformation efficiency up to (1.3 ± 0.6) × 10 5 cfu/μg DNA. We confirmed that our protocol can be applied to several B. mycoides strains. The method described here facilitates advanced molecular genetics studies in this important biocontrol bacterium.  Turgeon et al. (2006) Cells were grown in BHIS medium to OD 600~0 .85, then incubated 1 h with 2% glycine. Electro-competent cells were prepared using different electroporation buffers. 1 μg of pNW33N plasmid was used for electroporation with the settings 25 μF, 10 kV cm −1 , 200 Ω. Data are shown as mean ± standard deviation of 3 replications. Fig. 3. Effect of electric field on the electroporation efficiency of B. mycoides EC18. Cells were grown in BHIS medium to OD600~0.85, then incubated 1 h with 2% glycine. Electro-competent cells were prepared using buffer G. 1 μg of pNW33N plasmid was used for electroporation with a gradient of field strength (7.5-12.5 kV cm −1 ) and resistance values (100, 200, and 400 Ω) at 25 μF (N = 3, bar = standard deviation). Fig. 4. Effect of incubation time on the electroporation efficiency of B. mycoides EC18. Cells were grown in BHIS medium to OD600~0.85, then incubated 1 h with 2% glycine. Electrocompetent cells were prepared using buffer G. 1 μg of pNW33N plasmid was used for electroporation with the settings 25 μF, 10 kV cm −1 , 200 Ω. Cells were then incubated at 30°C, 180 rpm for 2-7 h (N = 3, bar = standard deviation).