The bacterial strains used in this study are listed in Table 1. B. subtilis BR151MA (Henkin et al., 1991), S. aureus JCM20624, E. coli W3110, P. aeruginosa IFO13275, and S. cerevisiae JCM1499 were used to assess improvements in Cs⁺ resistance. Luria-Bertani (LB), Miller medium (2 mL; BD Difco^™; BD Biosciences, Franklin Lakes, NJ, United States) (pH 7.5 for B. subtilis, pH 7.0 for others) was dispensed into a 14-mL culture tube, inoculated with each single colony, and cultured with shaking at 37°C and 200 rpm for 8 h. This culture solution was used as the pre-culture. A bioshaker BR-43FM (TAITEC Co., Ltd., Koshigaya, Japan) was used for shaking the culture, and the same equipment was used unless otherwise specified.

Various concentrations of MgCl₂ (25–200 mM final concentration) were added to 2 mL of LB medium supplemented with different concentrations of CsCl (100–600 mM final concentration). Ten microliters of the pre-culture solution were inoculated and cultured with shaking at 37°C and 200 rpm, and the OD₆₀₀ was measured after 16 h. Turbidity was measured using a UV-1800 ultraviolet–visible spectrophotometer (Shimadzu Co., Ltd., Kyoto, Japan). For S. cerevisiae, 2 mL of yeast malt (YM) broth (BD Difco^™; BD Biosciences) was dispensed into a 14-mL capacity culture tube, inoculated using a single colony, and cultured with shaking at 25°C and 200 rpm for 24 h. This was used as the pre-culture. Similar to eubacteria, CsCl and MgCl₂ were added to the YM broth; approximately 10 μL of the pre-culture was inoculated and cultured with shaking at 25°C and 200 rpm for 24 h, and turbidity was measured at OD₆₀₀.

Alkaliphilic Microbacterium sp. TS-1 was grown at 30°C in neutral complex (NC medium) and Tris media (Imazawa et al., 2016). Tris medium consisted of 3.63 g L⁻¹ Tris base, 1.47 g L⁻¹ citric acid monohydrate, 0.5 g L⁻¹ yeast extract, 9 g L⁻¹ glucose, and 1% (w/v) trace elements (Cohen-Bazire et al., 1957). Deionized (DI) water was used as the solvent. The pH was adjusted to 8 and 9 using 1 M N-methyl-D-glucamine. The pH was adjusted to 7 using 5 N sulfuric acid (H₂SO₄). Tris medium was used for the monovalent cation resistance test to avoid underestimation of cation influx. The NC medium consisted of 15.5 g L⁻¹ K₂HPO₄, 4.5 g L⁻¹ KH₂PO₄, 0.05 g L⁻¹ MgSO₄•7H₂O, 0.34 g L⁻¹ citric acid, 5 g L⁻¹ peptone, 2 g L⁻¹ yeast extract, 5 g L⁻¹ glucose, and 11.7 g L⁻¹ NaCl. Deionized water was used as a solvent. The final pH was adjusted to the desired value using KOH or H₂SO₄ (Fujinami et al., 2011).

Intracellular Cs⁺, K⁺, and Mg²⁺ concentrations were measured in TS-1 wild-type, Cs⁺-sensitive mutants (Mut4 and Mut5), and B. subtilis. As TS-1 Cs⁺-sensitive mutants and B. subtilis did not grow under high Cs⁺ concentration conditions, CsCl was added after culturing. Intracellular ion concentrations were measured 1 h after the addition of CsCl.

Strain TS-1 was inoculated from a single colony by dispensing 2 mL of neutral complex medium (pH 8.0) into a 14-mL culture tube and cultured at 30°C and 200 rpm for 18 h. The culture solution (100 μL) was inoculated into a medieval composite agar medium and pre-incubated at 37°C for 18 h. The cultured neutral complex agar medium was inoculated into 200 mL of 30 mM Tris medium (pH 8.0) such that the turbidity at OD₆₀₀ was 0.1 and cultured at 37°C and 150 rpm. When the turbidity at OD₆₀₀ reached 0.4, 20-mL portions were dispensed, various concentrations of CsCl (0, 100, 200, 300, and 400 mM) were added, and the cells were further cultured for 1 h. The culture solution was centrifuged at 9,100 × g at 20°C for 3 min to collect the cells. Unless otherwise specified, the TOMY Seiko MX-305 centrifuge (TOMY SEIKO Co., Ltd., Tokyo, Japan) was used. The supernatant was removed, and the cell pellet was suspended in 5 mL of 300 mM sucrose solution. The suspension was centrifuged at 9,000 × g for 3 min at 20°C, and the cell pellet was resuspended in 5 mL of 300 mM sucrose solution. The protein concentration was determined via the Lowry method using a 100-μL aliquot of the suspension. One milligram of protein was used in a volume of 3 μL of bacterial cells. The suspension was then centrifuged at 9,000 × g for 3 min at 20°C. The cell pellet was suspended in 5 mL of 5% trichloroacetic acid (TCA) and incubated at 100°C for 10 min. After that, the mixture was centrifuged at 4°C and 9,000 × g for 5 min, and the supernatant was used for intracellular ion concentration measurement. The Cs⁺ concentration was measured by taking 1 mL of the intracellular ion concentration measurement sample using an atomic absorption photometer (iCE3400; Thermo Fisher Scientific, Waltham, MA, United States). A 1-mL aliquot was taken from the measurement sample, and the K⁺ concentration was measured using an ANA-135 flame photometer (Tokyo Koden Co., Ltd., Tokyo, Japan). Mg²⁺ concentration was measured using a metalloassay magnesium assay LS kit (Metallogenics Co., Ltd., Chiba, Japan) according to the manufacturer’s instructions. First, 4.5 μL of 5 M NaOH was added to 100 μL of the measurement sample to adjust the pH from 2 to 7.

Subsequently, 3 μL of the pH-adjusted measurement sample was added to 250 mL of the coloring solution, vortexed, and allowed to stand at 20°C for 5 min. Absorbance at OD₆₆₀ was then measured. The Mg²⁺ concentration of the samples was calculated from the obtained value using the following formula:

Statistical analysis was performed with BellCurve for Excel (Social Survey Research Information Co., Ltd., Tokyo, Japan). Tukey test data for post hoc analysis of the results in Figures 1, 2 are presented in Supplementary Tables S1, S2, respectively. In Figures 1A–C, multiple comparison analysis was performed between strains after confirming a significant difference by two-way analysis of variance.

Ribosome preparation and fractionation were based on a previously reported method (Takada et al., 2021). Wild-type strain TS-1 and its Cs⁺-sensitive mutants Mut4 and Mut5 were streaked on a pH 8.5 neutral composite agar medium and used as pre-culture (37°C, 18 h). Using an inoculation loop, cells grown on the plate were collected and inoculated (37°C, 200 rpm) in 1,000 mL of Tris medium (pH 8.0). This was used as the culture medium. Cultivation was started at an initial turbidity (OD₆₀₀ = 0.1), and after 4 h of culture initiation, 25 and 50 mL of 8 M CsCl were added. To collect the bacteria, centrifugation (5,000 × g, 4°C, 10 min) was performed using an NA-600C rotor (TOMY SEIKO Co. Ltd.) The cells were resuspended in 10 mL Buffer I and subjected to a French press operation twice at 8,000 psi to disrupt the cells. Buffer I (pH 7.6) consisted of 2.42 g L⁻¹ tris base, 2.15 g L⁻¹ magnesium acetate tetrahydrate, 7.7 g L⁻¹ ammonium acetate, 0.015 g L⁻¹ ammonium acetate, and 8 mL of 0.1 M phenylmethylsulfonyl fluoride (PMSF). The pH was adjusted to 7.6 with 6 N HCl. Undisrupted cells were removed via centrifugation (12,000 × g, 4°C, 15 min), and crude cell debris containing ribosomes in the supernatant was collected. The absorbance of the cell debris at 260 nm was measured using a spectrophotometer and stored at 4°C until ultracentrifugation. Cell lysates were subjected to ultracentrifugation within 3 days of preparation. The same experiment was performed twice to confirm reproducibility.

For B. subtilis, 2 mL of LB medium (pH 7.5) was dispensed into a 14-mL culture tube, inoculated using a single colony, and cultured at 37°C and 200 rpm for 8 h. The culture solution (100 μL) was inoculated onto LB agar medium (pH 7.5) and pre-cultured at 37°C for 16 h. The cultured LB agar medium was inoculated into 1 L of LB medium (pH 7.5) such that the turbidity at OD₆₀₀ was 0.03 and cultured at 37°C and 150 rpm. After 2.5 h of culture initiation, CsCl (0, 200, and 400 mM) and 50 mM MgCl₂ were added, and the mixture was further cultured for 1 h. A crude cell extract was obtained from the culture solution in the same manner as described for strain TS-1.

For strain TS-1 and its Cs⁺-sensitive mutants Mut4 and Mut5, an Ultra-Clear^™ centrifuge tube (14 mL; Beckman Coulter, Brea, CA, United States) was used for sucrose density gradient ultracentrifugation with 4 mL of 10% sucrose in Buffer I. Thereafter, 4 mL of 35% sucrose in Buffer I was added to the bottom of the tube using a syringe. The upper lid of the tube was covered with parafilm, and the tube was allowed to stand tilted to form a sucrose density gradient at 20°C for 2 h, after which the tube was left at 4°C for 1 h. Crude ribosomal cell extract prepared as described above was layered on top of the sucrose density gradient. An OptimaTML-80XP Ultracentrifuge (Beckman Coulter) was used for ultracentrifugation (222,000 × g, 4°C, 3 h) using an SW40Ti rotor (6 × 14 mL; Beckman Coulter). Fractions of 200 μL each were taken from the upper layer of the tube (45 fractions), and the A₂₆₀ of each fraction was measured using a Thermo Nano Drop200C (Thermo Fisher Scientific K.K., Tokyo, Japan). The sucrose concentration was measured using an Atago handheld refractometer (MASTER-PM, ATAGO Co., Ltd., Tokyo, Japan). After the measurements, a separation profile diagram was constructed.

For B. subtilis, 4.5 mL of 10% sucrose in Buffer I was poured into a 14-mL Ultra-Clear^™ centrifuge tube, after which the 4.5 mL of 35% sucrose in Buffer I was added to the bottom of the tube using a syringe. Subsequent experiments were performed as described above for the TS-1 to create a ribosome profile.

Using B. subtilis, S. aureus, E. coli, P. aeruginosa, and S. cerevisiae as representative microorganisms, the effect of Mg²⁺ addition on Cs⁺ resistance was investigated. Cs⁺ resistance improved upon the addition of Mg²⁺ to the culture media for all microorganisms, although there was a difference in the degree of Cs⁺ resistance observed between them. B. subtilis showed only a slight increase in Cs⁺ tolerance with the addition of Mg²⁺ when the pH of LB medium was 7.0 (data not shown). Therefore, the experiment was performed at pH 7.5, where a more pronounced effect was observed. In B. subtilis, Cs⁺ resistance improved to 300 mM (Figure 3A) when 50 mM MgCl₂ was added to the medium compared with 100 mM without Mg²⁺. Regarding the Cs⁺ resistance of S. aureus, growth inhibition was observed at 300 mM CsCl without Mg²⁺. However, growth was observed even with 600 mM CsCl when 100 mM MgCl₂ was added (Figure 3B). E. coli cultures were resistant to 400 mM CsCl when 50 mM MgCl₂ was added compared with 300 mM without Mg²⁺ (Figure 3C). P. aeruginosa grew in the presence of 200 mM CsCl when 125 mM MgCl₂ was added, compared with only 100 mM CsCl without Mg²⁺ (Figure 3D). S. cerevisiae grew in the presence of 200 mM CsCl when 150 mM MgCl₂ was added, compared to its growth in the absence of Mg²⁺ (Figure 3E).

Intracellular Cs⁺ concentrations were measured in TS-1 wild-type, Cs⁺-sensitive mutants (Mut4 and Mut5), revertants (Mut4R and Mut5R), and B. subtilis with and without CsCl treatment (Figure 1A). In B. subtilis, intracellular Cs⁺ concentration increased with increasing CsCl concentration. Conversely, the intracellular Cs⁺ concentration was maintained at <200 mM in the TS-1 wild-type. In Mut4, the intracellular Cs⁺ concentration increased with increasing CsCl concentration, similar to the trend seen in B. subtilis. Mut4R had a lower Cs⁺ concentration than that of Mut4, and this was comparable to the wild-type. Strains Mut5 and Mut5R exhibited intracellular Cs⁺ concentrations intermediate to those of TS-1 wild-type and B. subtilis.

Intracellular K⁺ concentrations were measured in TS-1 wild-type, Cs⁺-sensitive mutants, revertants, and B. subtilis with and without Cs⁺ treatment (Figure 1B). The addition of CsCl to B. subtilis significantly decreased intracellular K⁺ concentration. The TS-1 wild-type showed a more gradual decrease in intracellular K⁺ concentration than that of B. subtilis following CsCl addition, showing a significant difference (p < 0.05). Similar to B. subtilis, Mut4 showed a drastic reduction in the intracellular K⁺ concentration when CsCl was added. Mut4R, Mut5, and Mut5R exhibited approximately similar behaviors as the wild-type strain.

Intracellular Mg²⁺ concentrations were measured in TS-1 wild-type, Cs⁺-sensitive mutants, revertants, and B. subtilis with and without Cs⁺ treatment (Figure 1C). There was no significant difference in intracellular Mg²⁺ concentrations between TS-1 wild-type, Mut5, and Mut5R. In contrast, Mut4 and Mut4R showed high intracellular Mg²⁺ concentrations.

Tukey test raw data for post hoc analysis of the results in Figure 1 are presented in Supplementary Table S1.

Intracellular Cs⁺ concentrations in B. subtilis, representing common microorganisms, were measured with and without CsCl and MgCl₂ treatments (Figure 2A). At CsCl concentrations of 0–300 mM, adding 50 mM MgCl₂ did not change the intracellular Cs⁺ concentration, but at 400 mM CsCl, intracellular Cs⁺ concentration decreased by approximately 50%.

Similarly, the intracellular K⁺ concentration in B. subtilis was also measured (Figure 2B). There was no significant difference in the intracellular K⁺ concentrations with the addition of 50 mM MgCl₂ at CsCl concentrations of 0, 100, 300, and 400 mM. However, intracellular K⁺ concentrations significantly increased at a CsCl concentration of 200 mM.

Likewise, intracellular Mg²⁺ concentrations were also measured in B. subtilis (Figure 2C). Without CsCl treatment, adding 50 mM MgCl₂ increased the intracellular Mg²⁺ concentration by approximately 40 mM. As the amount of CsCl increased, the intracellular Mg²⁺ concentration increased slightly, and a significant difference (p < 0.05) was observed at a CsCl concentration of 400 mM.

Tukey test raw data for post hoc analysis of the results in Figure 2 are presented in Supplementary Table S2.

TS-1 wild-type, Mut4, and Mut5 were cultured, treated with CsCl after 4 h of culturing, and harvested after 1 h to prepare ribosomes. The doubling times of TS-1 wild-type, Mut4, and Mut5 up to 4 h of culture were approximately 2.6, 2.5, and 2.5 h, respectively. The growth of the TS-1 wild-type continued regardless of the presence or absence of CsCl treatment (Figure 4A). In contrast, growth inhibition of both Mut4 and Mut5 was observed after the addition of CsCl (Figures 4B,C). Next, B. subtilis was cultured, CsCl treatment was performed after 2.5 h of culture, and the cells were harvested after 1 h to prepare ribosomes. The doubling time of B. subtilis up to 2.5 h before culture was approximately 37 min, and growth inhibition was observed after the addition of CsCl (Figure 4D).

To investigate the effect of Cs⁺ treatment on ribosome complex formation in the TS-1 wild-type and its mutants, crude cell extracts were prepared from cultured cells, and the ribosome complexes were separated using sucrose density gradient ultracentrifugation (Figure 5A). In the TS-1 wild-type, no effect was observed on the formation of ribosomal complexes, regardless of the presence or absence of CsCl treatment. Next, the ribosome complexes were analyzed for the Cs⁺-sensitive mutants, Mut4 and Mut5, in the same manner as the wild-type (Figures 5B,C). Both mutants were treated with CsCl at 200 and 400 mM, upon which the 70S ribosomes collapsed, and a finer peak than that of 30S ribosomes was observed.

The effect of CsCl treatment on ribosome complex formation was investigated in B. subtilis. We also investigated whether ribosome complex formation was affected by the addition of MgCl₂. Ribosome complexes were analyzed using sucrose density gradient ultracentrifugation in the same manner as that described for the TS-1 strain (Figure 5D). Even when B. subtilis was treated with CsCl, 70S ribosomes tended to decrease slightly; however, ribosomes were not significantly affected. In addition, no effect on the ribosome complex was observed when MgCl₂ was added to the culture medium.

Examining the effect of Mg²⁺ addition on Cs⁺ resistance in B. subtilis, P. aeruginosa, E. coli, S. aureus, and S. cerevisiae as representative microorganisms showed that Cs⁺ resistance was improved by adding Mg²⁺ to the medium for all the five strains tested in this study. Furthermore, it was effective not only for TS-1 but also for microorganisms in general.

Mg²⁺ plays various essential roles in many organisms, including maintaining ribosome structure, genome stabilization, and enzyme activation (Akanuma et al., 2018). It has been reported that increasing the intracellular Mg²⁺ concentration stabilizes ribosomes in B. subtilis, in which the structure of the ribosome is destabilized due to the deletion of a part of the ribosomal protein (Akanuma et al., 2018). Therefore, we hypothesized that Mg²⁺ enhanced Cs⁺ tolerance. Cs⁺ uptake by cells affected ribosome complex formation. As Mg²⁺ stabilized the structure of ribosomes, the upper limit of Cs⁺ concentrations tolerated by the strain was considered to be improved.

Differences were found in the improvement in Cs⁺ tolerance and the amount of Mg²⁺ required for the outcome obtained by adding Mg²⁺ to each microbial species culture. It was speculated that the increase in intracellular Mg²⁺ could counteract the drastic decrease in intracellular K⁺ concentration to some extent or that the stability of ribosomes might differ among microorganisms or both.

Changes in intracellular Cs⁺, K⁺, and Mg²⁺ concentrations in strains TS-1 and B. subtilis with and without CsCl treatment were measured.

In B. subtilis, intracellular Cs⁺ concentration increased with increasing CsCl concentration in the medium, and the K⁺ concentration decreased sharply. From the Cs⁺-tolerant growth test results, the intracellular K⁺ concentration decreased to approximately 13% of that in the absence of Cs⁺ at 200 mM CsCl, at which B. subtilis could not grow. It was speculated that the extreme decrease in intracellular K⁺ concentration in B. subtilis impeded the maintenance of vital functions. This was consistent with results reported for E. coli (Avery, 1995). B. subtilis and E. coli do not harbor a Cs⁺ efflux mechanism. Thus, it was inferred that a decrease in intracellular K⁺ concentration was the leading cause of growth inhibition in these bacteria. CsCl treatment slightly increased intracellular Mg²⁺ but did not make a significant difference. Contrastingly, in strain TS-1, the intracellular Cs⁺ concentration remained lower than that in B. subtilis even when the CsCl concentration increased, and the concentration was maintained at approximately <200 mM. Intracellular K⁺ concentration was moderately decreased in TS-1 following Cs⁺ treatment compared to that in B. subtilis, and intracellular K⁺ concentration was high compared to that in B. subtilis. This may be because TS-1 harbors a Cs⁺ efflux system, CshA (Koretsune et al., 2022). The apparent K_m value of CshA at pH 8.0 was approximately 250 mM, indicating that TS-1 prevented the intracellular Cs⁺ concentration from increasing and K⁺ deficiency.

Analysis of the changes in intracellular Cs⁺, K⁺, and Mg²⁺ concentrations with and without Cs⁺ treatment in the TS-1 Cs⁺-sensitive mutants and its revertants revealed that CsCl treatment of Mut4 resulted in an influx of Cs⁺ into cells, and the K⁺ concentration was significantly reduced. This was similar to the results obtained for B. subtilis, which does not possess a Cs⁺ efflux system. This suggested that Cs⁺ excretion by CshA was essential for Cs⁺ resistance in TS-1 cells. In addition, the intracellular Mg²⁺ concentration in Mut4 was maintained at higher levels than those in the wild-type strain. This suggested that Mg²⁺ was incorporated into cells to resist the toxicity caused by the influx of a large amount of Cs⁺ into the cells. Since CshA in the Cs⁺ efflux system is active in Mut5, which is a MgtE mutant, Cs⁺ and K⁺ concentrations were similar to those of the wild-type even after Cs⁺ treatment. No significant difference was observed in the intracellular Mg²⁺ concentration. This suggested the possibility that another Mg²⁺ transporter performed the uptake of Mg²⁺ into the cell and functioned when Cs⁺ entered the cells of Mut5.

Mg²⁺ has been shown to improve Cs⁺ resistance in several microorganisms. We investigated changes in intracellular Cs⁺, K⁺, and Mg²⁺ concentrations when MgCl₂ was added simultaneously with CsCl in B. subtilis. The addition of 50 mM of MgCl₂ decreased the intracellular Cs⁺ concentration at a CsCl concentration of 400 mM. The E. coli Cs⁺ uptake system, Kup, becomes less active when the cell is filled with K⁺ (Dosch et al., 1991). Although the details of the Cs⁺ uptake system of B. subtilis have not been elucidated, the K⁺ uptake system is expected to take up Cs⁺. A significant increase in the intracellular K⁺ concentration was observed when Mg²⁺ was added at a CsCl concentration of 200 mM (p = 0.03). Although there was no significant difference in this experiment, the intracellular K⁺ concentration increased at CsCl concentrations of 300 and 400 mM at a level close to the CsCl concentration of 200 mM (300 mM: p = 0.06, 400 mM: p = 0.08). This was presumed to be because the addition of high concentrations of CsCl and MgCl₂ enhanced the uptake of K⁺ to adapt to the high osmotic pressure environment.

To elucidate the mechanism of Cs⁺ resistance through Mg²⁺ uptake in TS-1, we investigated whether the presence or absence of CsCl treatment in TS-1, its Cs⁺-sensitive mutants, and B. subtilis affected the formation of ribosomal complexes. In addition, we investigated whether the ribosomal complex was affected when Mg²⁺ was added to B. subtilis cultures to improve Cs⁺ tolerance. We observed that Cs⁺ treatment did not affect the ribosomal complex in the TS-1 wild-type. However, in the Cs⁺-sensitive strains Mut4 and Mut5, the 70S ribosomes collapsed following CsCl treatment. However, CsCl treatment did not affect the ribosomal complex of B. subtilis.

Mut4 is a CshA mutant, and a large amount of Cs⁺ accumulated in the cells based on the measurement results of intracellular Cs⁺ concentration. Therefore, it is suggested that high concentrations of intracellular Cs⁺ destabilize ribosomes. Strain Mut5 is an MgtE mutant, and although it has no issue with Mg²⁺ uptake, it is presumed that Mg²⁺ uptake cannot be enhanced when Mg²⁺ is required. Therefore, it was speculated that ribosomes destabilized by Cs⁺ could be stabilized owing to the lack of Mg²⁺. High concentrations of K⁺ inhibit ribosome complex formation by competing with Mg²⁺ (Nierhaus, 2014; Senbayram et al., 2015). Because Cs⁺ has physicochemical properties similar to those of K⁺, it was speculated that Cs⁺ destabilized ribosomes by competing with Mg²⁺.

TS-1 wild-type suppressed intracellular Cs⁺ to ≤200 mM in a high-concentration Cs⁺ environment by excreting intracellular Cs⁺ using CshA. In a Cs⁺ environment of ≤200 mM, MgtE is considered to enhance Mg²⁺ uptake and stabilize the ribosomal complex, thereby adapting to the Cs⁺ environment.

Bacillus subtilis did not exhibit Cs⁺ resistance, and when treated with CsCl, Cs⁺ accumulated in the cells, similar to Mut4. Thus, the ribosome complex is expected to be similarly destroyed by CsCl treatment. However, CsCl treatment did not have a significant effect on ribosomal complexes. The impact of improving Cs⁺ tolerance by adding Mg²⁺ differed depending on the microbial species. Hence, the stability of ribosomes to Cs⁺ was expected to vary depending on the microbial species. Certain ribosomal proteins function to replace Mg²⁺ for ribosomal stabilization (Hsiao et al., 2009; Akanuma et al., 2014). Although Cs⁺ competes with Mg²⁺ to destabilize ribosomes, it does not affect ribosome stabilization by ribosomal proteins, suggesting that B. subtilis ribosomes are not affected by Cs⁺. However, it has been suggested that the influx of Cs⁺ into the cells of B. subtilis leads to an extreme decrease in K⁺ and that a shortage of K⁺ interferes with life support.