SodA and Manganese Are Essential for Resistance to Oxidative Stress in Growing and Sporulating Cells of Bacillus subtilis

Abstract

We constructed a sodA-disrupted mutant of Bacillus subtilis 168, BK1, by homologous recombination. The mutant was not able to grow in minimal medium without Mn(II). The spore-forming ability of strain BK1 was significantly lower in Mn(II)-depleted medium than that of the wild-type strain. These deleterious effects caused by the sodA mutation were reversed when an excess of Mn(II) was used to supplement the medium. Moreover, the growth inhibition by superoxide generators in strain BK1 and its parent strain was also reversed by the supplementation with excess Mn(II). We therefore estimated the Mn-dependent superoxide-scavenging activity in BK1 cells. Whereas BK1 cells have no detectable superoxide dismutase (Sod) on native gel, the superoxide-scavenging activity in crude extracts of BK1 cells grown in Mn(II)-supplemented LB medium (10 g of tryptone, 5 g of yeast extract, and 5 g of NaCl per liter) was significantly detected by the modified Sod assay method without using EDTA. The results obtained suggest that Mn, as a free ion or a complex with some cellular component, can catalyze the elimination of superoxide and that both SodA and Mn(II) are involved not only in the superoxide resistance of vegetative cells but also in sporulation.

Reactive oxygen species, such as superoxide anion (O₂⁻) and hydrogen peroxide (H₂O₂), are formed as a consequence of partial reduction of oxygen in the cellular compartment. These reactive oxygen species have been known to cause severe damage to DNA, RNA, proteins, and lipids (10, 17, 27). Therefore, most aerobic organisms have potent systems to protect themselves from this damage. Superoxide dismutase (Sod; EC 1.15.1.1) catalyzes the dismutation of O₂⁻ to oxygen and H₂O₂ (13), and catalase and peroxidase prevent the intracellular accumulation of toxic H₂O₂ (15).

Bacillus subtilis is an aerobic bacterium and forms an endospore which is highly resistant to a variety of stresses, such as oxidants, heat, and UV light. We have recently reported that this bacterium possesses one detectable Sod (SodA), which is present in both vegetative cells and spores, and also that this enzyme is predicted to protect cells from oxidative stress during growth and sporulation (18). Casillas-Martinez and Setlow (6) have recently demonstrated that SodA, KatX, alkyl hydroperoxide reductase, and MrgA play no role in spore resistance to heat and oxidants. In this study, we characterized a B. subtilis sodA mutant and found that a SodA-independent defensive system against oxidative stress is mediated substantially by manganese in B. subtilis.

Construction of a sodA mutant of B. subtilis.

To investigate the function of SodA in B. subtilis, we constructed a sodA-disrupted mutant by the following method. A recombinant plasmid which contained the sodA gene disrupted by the insertion of a chloramphenicol acetyltransferase-encoding (cat) gene derived from pMSG-CAT was constructed, designated pBR-sodA::cat, and introduced into cells of B. subtilis 168 (trpC2). Some chloramphenicol-resistant colonies which appeared on an agar plate of LB medium (10 g of tryptone, 5 g of yeast extract, 5 g of NaCl per liter) containing 5-μg/ml chloramphenicol were selected. It was confirmed by Southern blotting analysis that sodA was expectably replaced with sodA::cat in the chromosomal DNAs of some chloramphenicol-resistant mutants (data not shown). One of these sodA::cat mutants was designated BK1.

Growth ability of the B. subtilis sodA mutant.

Although no difference in the growth rate at 37°C in LB medium was seen between the wild-type strain and strain BK1, the final cell mass of the latter reached only half of that of the former (Fig. 1C, open symbols). In Spizizen salts medium [14 g of K₂HPO₄, 6 g of KH₂PO₄, 2 g of (NH₄)₂SO₄, 1 g of sodium citrate, 0.2 g of MgSO₄ · 7H₂O, and 5 g of glucose per liter] supplemented with 20 mg of l-tryptophan per liter, the growth of strain BK1 was markedly depressed (Fig. 1A).

FIG. 1.

Manganese requirement for the growth of strain BK1. Overnight cultures of the wild-type and BK1 strains were grown in LB medium not supplemented with metal were diluted 1:50 in Spizizen salts medium containing different kinds of metal ion at 0.1 mM (A and B) or in LB medium containing 0.1 mM EDTA (C) or 1 μM CdCl₂ (D). Cell growth was monitored by measuring OD₆₅₀. (A) Effect of MnCl₂ (solid symbols) on the growth of the wild type (squares) and strain BK1 (circles). Control values (open symbols) are also shown. (B) Effects of MnCl₂ (□), FeSO₄ (○), Ca(NO₃)₂ (▵), CuSO₄ (◊), and ZnCl₂ (×) on the growth of strain BK1. (C and D) Effects of EDTA and CdCl₂ (solid symbols), respectively, on the growth of the wild type (squares) and strain BK1 (circles). Control values (open symbols) are also shown.

An Escherichia coli sodA sodB double mutant lacking cytoplasmic Sods has been reported to require branched-chain amino acids for growth in minimal medium (5). Unlike E. coli, however, strain BK1 was not able to recover the ability to grow, even in minimal medium supplemented with Casamino Acids at final concentrations of 0.02 to 0.2% (data not shown). But unexpectedly, this growth depression was found to be completely reversed by supplementation with MnCl₂ at concentrations higher than 1 μM (Fig. 1A). No such drastic restoring effect was observed with the other metals tested [FeSO₄, CuSO₄, ZnCl₂, and Ca(NO₃)₂], although Fe(II) at 0.1 mM had a slight effect (Fig. 1B). To confirm this manganese requirement of strain BK1, we compared the sensitivities of both strains of B. subtilis to EDTA as a divalent-cation chelator and Cd(II), which is a competitive inhibitor of Mn(II) uptake (21). Addition of EDTA at 0.1 mM or Cd(II) at 1 μM to the culture caused a remarkable delay in the growth of strain BK1 (Fig. 1C and D). No such phenomena were observed with the wild-type strain under identical conditions. These results indicate that manganese is necessary for the SodA-deficient strain to grow.

Effect of Mn(II) supplementation on oxidant sensitivity.

We examined the effects of oxidants on the cell growth (growth test) and viability (viability test) of strain BK1. In the growth test, cells were grown in LB medium until the optical density at 650 nm (OD₆₅₀) of the culture reached 0.3 and then paraquat, menadione, or H₂O₂ at various concentrations was added to the culture. The subsequent cell growth was monitored. In the viability test, cells grown in LB medium were harvested at the mid-log or stationary phase (cultivated for 8 h), washed with 50 mM Tris-HCl buffer (pH 7.0), and resuspended in the same buffer at a density of approximately 10⁸ cells/ml. The suspension was treated with paraquat or H₂O₂ at appropriate concentrations for 15 min at 37°C. The viable cells were counted on an LB agar plate after incubation for 12 to 18 h at 37°C.

Growing cells of strain BK1 were more sensitive to paraquat, menadione, and H₂O₂ than were those of the wild-type strain (Fig. 2). Similar results were obtained in the viability test (data not shown). Since the ability of strain BK1 to grow in minimal medium was restored by supplementation with MnCl₂, Mn(II) is suggested to protect B. subtilis cells against oxidative stress. We therefore tested whether Mn(II) influences cellular sensitivity to some oxidants or not. Indeed, supplementation of the growth medium with 0.1 mM MnCl₂ markedly relieved cells from growth inhibition by paraquat and menadione (Fig. 2A and B). In the case of H₂O₂ at 200 μM, on the other hand, MnCl₂ caused transient cell lysis after exposure. This phenomenon may result from the repression of katA, one of the peroxide regulon genes (8), by Mn(II).

FIG. 2.

Effect of manganese on sensitivity to various oxidants. Overnight cultures of the wild type and strain BK1 grown in LB medium were diluted 1:50 with fresh LB medium with (solid symbols) or without (open symbols) 0.1 mM MnCl₂. Cultures were incubated to an OD₆₅₀ of 0.3 (at 0 h), and then paraquat (A), menadione (B), or H₂O₂ (C) was added (0 h) to the medium at the following concentrations: 50 (circles), 100 (triangles), or 200 (diamonds) μM paraquat; 2 μM menadione (circles), and 200 μM H₂O₂ (circles) for the wild-type strain and 5 (circles), 20 (triangles), or 50 (diamonds) μM paraquat; 0.1 μM menadione (circles); and 200 μM H₂O₂ (circles) for strain BK1. Untreated control values (squares) are also shown.

We further examined the effect of Mn(II) on oxidant sensitivity by using the viability test. However, no supplementation effect of MnCl₂ at 0.1 mM was observed either in growth medium or during paraquat treatment. It should be noted that the effect of Mn(II) on oxidant sensitivity was observed only in growing cells.

Effects of SodA and Mn(II) on sporulation.

Cells of strain BK1 normally sporulated in Schaeffer’s sporulation medium consisting of 0.8% nutrient broth (Difco), 27 mM KCl, 2 mM MgSO₄ · 7H₂O, 1 mM Ca(NO₃)₂ · 4H₂O, 0.1 mM MnCl₂ · 4H₂O, and 1 μM FeSO₄ · 7H₂O (data not shown), suggesting that SodA is not essential for sporulation under conventional sporulation conditions. However, in Mn(II)-depleted Schaeffer’s sporulation medium in which MnCl₂ was replaced with a corresponding amount of FeSO₄, the sporulation frequency of strain BK1 was markedly lower than that of the wild-type strain (Table 1). This result suggests that strain BK1 requires Mn(II) not only for vegetative growth but also for sporulation at a concentration higher than that needed by the wild-type strain.

TABLE 1.

Sporulation frequency in Mn(II)-deficient medium^a

Strain and cultivation time (days)	Count (CFU/ml) of:		Sporulation frequency (%)
	Total cells	Sporesb
168
1	9.0 × 107	3.7 × 103	0.004
4	3.7 × 107	2.2 × 105	0.59
7	2.2 × 107	6.7 × 105	3.0
BK1
1	4.0 × 107	8.3 × 102	0.002
4	1.4 × 107	7.1 × 102	0.005
7	1.2 × 107	6.9 × 102	0.006

^aMn-depleted Schaeffer’s sporulation medium in which MnCl₂ was replaced with 0.1 mM FeSO₄ was used. 

^bSpores were counted on LB agar plates after heat treatment at 80°C for 15 min. 

Furthermore, even in Schaeffer’s sporulation medium originally containing 0.1 mM MnCl₂, the sporulation frequencies of both strain BK1 and the wild-type strain bearing vector plasmid pHY300PLK (Takara) were affected by the addition of paraquat at the time of entry into the sporulation phase (time zero) (Fig. 3A). Addition of paraquat at a concentration of 5 μM, which did not affect vegetative growth, inhibited the sporulation of the wild-type strain, and more importantly, the sporulation of strain BK1 was profoundly depressed at an even lower concentration (1 μM). Overproduction of SodA from pHY-sod, which was derived from pHY300PLK and contained the complete sodA gene, restored viability during sporulation and the ability to sporulate after the addition of paraquat at time zero in both strain BK1 and the wild-type strain (Fig. 3B). Addition of paraquat at 2 h of incubation or later did not affect the sporulation frequency (data not shown). These results suggest that an elevated level of O₂⁻ may inhibit an early event in the sporulation process and also that Mn(II), as well as SodA, may relieve the toxicity of such an oxidative milieu for sporulating cells.

FIG. 3.

Inhibition of sporulation by paraquat. The wild type (wt) and strain BK1 carrying pHY300PLK (A) or pHY-sod (B) were sporulated in Schaeffer’s sporulation medium containing 25 μg of tetracycline per ml. Paraquat was added at the concentrations indicated at the time of entry into the sporulation phase (0 h), and the culture was incubated for a further 48 h with shaking. Spore counts (solid bars) on LB agar plates obtained after heat treatment at 80°C for 15 min (left axis) and viable cell counts (open bars) are shown. The sporulation frequencies of the wild-type strain (squares) and strain BK1 (circles) are also plotted (right axis).

Manganese-dependent O₂⁻-scavenging activity in B. subtilis.

Manganese is known to have a Sod-like activity in vitro, and several Mn(II) complexes and manganous porphyrins catalyze the elimination of O₂⁻ (2, 4, 9, 11, 19). Furthermore, Mn(II) suppresses the methionine and lysine auxotrophic phenotypes of a Saccharomyces cerevisiae SOD1 mutant lacking intracellular Cu, Zn-Sod (7, 22) and functions as a substitute for Sod in Lactobacillus plantarum (1). We therefore measured the Mn-dependent O₂⁻-scavenging activity in crude extracts of BK1 cells. Although the Sod assay, which is based upon the inhibition of cytochrome c reduction (26), is usually carried out with 50 mM potassium phosphate buffer (pH 7.8) containing 0.1 mM EDTA as previously described (18), non-EDTA-supplemented 50 mM Tris-HCl buffer (pH 7.8) was used here for the measurement of Mn(II)-dependent O₂⁻-scavenging activity. Whereas BK1 cells had no detectable Sod protein on native gel, the O₂⁻-scavenging activity in crude extracts of BK1 cells grown in LB medium supplemented with 0.1 mM MnCl₂ was significantly detectable in this modified assay system and was approximately 60% of that of the wild-type strain (Table 2). No such detectable O₂⁻-scavenging activity was observed in crude extracts of BK1 cells grown in the absence of MnCl₂ (data not shown). The activity in BK1 cells was diminished by addition of EDTA or dialysis, in contrast to that in wild-type cells (Table 2). After boiling for 10 min, the activities remaining in the supernatants of BK1 and wild-type cells were approximately 85 and 55% of the total O₂⁻-scavenging activity, respectively (Table 2). These results suggest that Mn(II) itself or in a complex with some cellular component is involved in the cellular defense against O₂⁻ stress and therefore can substitute for Sod. However, the finding that this Mn effect was observed only with growing cells is a puzzle. Mn(II) alone or in a complex might not be recycled as an O₂⁻ scavenger in nongrowing cells. Alternatively, Mn(II) might form a functional complex with some cellular component depending upon cell growth.

TABLE 2.

SodA-independent superoxide-scavenging activity in crude cell extract

Treatment	Total superoxide-scavenging activity (U/mg of protein)a
	168	BK1
None	10.0	6.7
EDTA, 1 mM	6.5	—c
100°C, 10 minb	5.5	5.7
Dialysis, 2 h	9.2	3.7
Dialysis, 12 h	9.2	<0.1

^aThe activity in the crude extract of each strain grown in MnCl₂-supplemented LB medium for 8 h was measured in non-EDTA-supplemented buffer as described in the text. Each protein concentration was determined before treatment. 

^bSamples were boiled for 10 min and then centrifuged. The activity in the supernatant was measured. 

^c—, activity below the limit of detection. 

As other possibilities for the manganese effect, Mn(II) might function in vivo as a stabilizer of some critical O₂⁻-sensitive protein(s) such as glutamine synthetase, which contains Mn and is known to be sensitive to oxidative stress (23, 24), or as a cofactor of some regulator protein which is involved in oxidative stress resistance. To understand the physiological role of manganese, detailed chemical and genetic studies are necessary.

In E. coli, some [4Fe-4S] cluster-containing proteins are known to be to O₂⁻ generator or oxygen sensitive (12, 14, 20, 25). Supplementation with Fe has been reported to restore the ability of a Sod-deficient strain of E. coli to grow in minimal medium. This Fe effect is suggested to be the reactivation of such inactivated proteins (3). In this report, we demonstrate that Fe(II) partially restores the growth of a sodA mutant in minimal medium, and therefore, this Fe(II) effect may resemble that observed with a Sod-deficient strain of E. coli.

From the finding that the sporulation frequency of the mutant markedly decreased, compared with that of the wild-type strain, both under stress conditions and in Mn-depleted medium, it is suggested that cells in the early sporulation phase and/or premature spores are sensitive to O₂⁻ and also that both SodA and Mn(II) protect sporulating cells from oxidative stress. It is still unclear whether the resistance of mature spores to oxidative stress directly reflects the O₂⁻-scavenging ability of those spores. Henriques et al. (16) have recently reported that the spores of a sodA mutant of B. subtilis was not able to mature sufficiently. We presume that the spores of a Sod-deficient strain produced in the absence of Mn(II) may be sensitive to oxidative stress.

In conclusion, our findings indicate that not only SodA but also Mn(II) possesses a functional role in the cellular defense against O₂⁻ and that these two systems function in both growing and sporulating cells. This is the first paper reporting that Mn(II) is an additional key factor in the cellular defense against oxidative stress in B. subtilis.