Alkyl Hydroperoxide Reductase, Catalase, MrgA, and Superoxide ...

Download PDF
2 downloads 0 Views 1MB Size Report
Comment
experiments, to John Helmann for strains and advice, to Peter Zuber for information about ... D.C.. 18. Hassett, D. J., H. P. Schweizer, and D. E. Ohman. 1995.
JOURNAL OF BACTERIOLOGY, Dec. 1997, p. 7420–7425 0021-9193/97/$04.0010 Copyright © 1997, American Society for Microbiology

Vol. 179, No. 23

Alkyl Hydroperoxide Reductase, Catalase, MrgA, and Superoxide Dismutase Are Not Involved in Resistance of Bacillus subtilis Spores to Heat or Oxidizing Agents LILLIAM CASILLAS-MARTINEZ

AND

PETER SETLOW*

Department of Biochemistry, University of Connecticut Health Center, Farmington, Connecticut 06032 Received 21 July 1997/Accepted 1 October 1997

Only a single superoxide dismutase (SodA) was detected in Bacillus subtilis, and growing cells of a sodA mutant exhibited paraquat sensitivity as well as a growth defect and reduced survival at an elevated temperature. However, the sodA mutation had no effect on the heat or hydrogen peroxide resistance of wild-type spores or spores lacking the two major DNA protective a/b-type small, acid-soluble, spore proteins (termed a2b2 spores). Spores also had only a single catalase (KatX), as the two catalases found in growing cells (KatA and KatB) were absent. While a katA mutation greatly decreased the hydrogen peroxide resistance of growing cells, as found previously, katA, katB, and katX mutations had no effect on the heat or hydrogen peroxide resistance of wild-type or a2b2 spores. Inactivation of the mrgA gene, which codes for a DNA-binding protein that can protect growing cells against hydrogen peroxide, also had no effect on spore hydrogen peroxide resistance. Inactivation of genes coding for alkyl hydroperoxide reductase, which has been shown to decrease growing cell resistance to alkyl hydroperoxides, had no effect on spore resistance to such compounds or on spore resistance to heat and hydrogen peroxide. However, Western blot analysis showed that at least one alkyl hydroperoxide reductase subunit was present in spores. Together these results indicate that proteins that play a role in the resistance of growing cells to oxidizing agents play no role in spore resistance. A likely reason for this lack of a protective role for spore enzymes is the inactivity of enzymes within the dormant spore. inactivation of an alkyl hydroperoxide reductase results in decreased resistance to alkyl hydroperoxides (3). Although protective enzymes and MrgA clearly play a role in the resistance of growing cells of B. subtilis to oxidative stress, there are no data on the role of these proteins in spore resistance. Spores of Bacillus species are much more resistant than are their growing cell counterparts to a variety of treatments including oxidizing agents, heat, UV radiation, and dessication (17, 25, 34). There are generally multiple factors involved in spore resistance to any of these treatments. For example, sporulation temperature, spore dehydration and mineralization, and specialized DNA-binding proteins termed a/b-type small, acid-soluble, spore proteins (SASP) (not related to MrgA) all play a role in spore heat resistance (17, 34). Spore impermeability, the absence of NAD(P)H from the spore, and a/b-type SASP have all been identified as or suggested as being important factors in spore resistance to oxidative stress (25, 34). Given the important roles of protective enzymes and MrgA in the resistance of growing cells of B. subtilis to oxidizing agents, it seems possible that these proteins might be an additional factor in spore resistance. Consequently, in this communication we have examined the roles of MrgA and protective enzymes in spore resistance to oxidizing agents. In addition, since it has been suggested that heat killing of wildtype spores is due to oxidative damage caused by free radicals (25), we have also examined the roles of MrgA and protective enzymes in spore heat resistance.

Growing bacterial cells have a number of mechanisms to minimize the deleterious effects of oxidizing agents such as hydroperoxides and superoxides, compounds which are readily generated in vivo under aerobic conditions (14, 21). These mechanisms include the repair of damage caused by oxidizing agents, in particular damage to DNA, as well as the prevention of such damage. Mechanisms which prevent damage include protective DNA-binding proteins (1, 11) as well as enzymes such as alkyl hydroperoxide reductase (22, 35), catalase (8, 20), and superoxide dismutase (7), which can destroy the oxidizing agents alkyl hydroperoxides, hydrogen peroxide, and superoxide, respectively. The significance of these two mechanisms in bacterial resistance to oxidative stress has been most extensively studied in Escherichia coli. In this organism lack of the DNA-binding protein Dps results in increased sensitivity to hydrogen peroxide and reduced survival in stationary phase (1), while loss of catalase or alkyl hydroperoxide reductase results in increased sensitivity to hydrogen peroxide and alkyl hydroperoxides, respectively (20, 35). Inactivation of the genes (sod) coding for the two cytoplasmic superoxide dismutases also results in cells which withstand oxidative stress very poorly (7, 15). In addition, a double sod mutant is significantly more sensitive to elevated temperatures under aerobic conditions than is a wild-type cell (5, 6). This latter finding has also been made with the yeast Saccharomyces cerevisiae (12). Studies of the role of protective proteins in the resistance of Bacillus subtilis to oxidative stress have not been as extensive as have those with E. coli. However, work with growing cells of B. subtilis has shown that (i) lack of the major vegetative cell catalase (KatA) or a Dps homolog (called MrgA) results in decreased hydrogen peroxide resistance (8, 11, 13) and (ii)

MATERIALS AND METHODS Bacterial strains, growth, sporulation, and measurement of resistance. The strains of B. subtilis used in this work are listed in Table 1; all are derivatives of strain 168. B. subtilis cells were transformed to resistance to chloramphenicol (3 mg/ml; Cmr) or spectinomycin (100 mg/ml; Spr) as previously described (2). Growing cells of B. subtilis were prepared at 37°C in Luria-Bertani (LB) medium (per liter: 10 g of tryptone, 5 g of yeast extract, 10 g of NaCl, and 1 ml of 1 M

* Corresponding author. Phone: (860) 679-2607. Fax: (860) 6793408. E-mail: [email protected]. 7420

RESISTANCE OF B. SUBTILIS SPORES TO OXIDIZING AGENTS

VOL. 179, 1997 TABLE 1. B. subtilis strains Strain

PS356 PS832 PS2485 PS2488 PS2489 PS2493 PS2495 PS2499 PS2500 PS2504 PS2506 PS2507 PS2512 PS2513 PS2514 PS2515 PS2516 PS2517 PS2558 PS2559 HB1703 HB1730

Genotype and phenotype 2

2

a b Wild type mrgA Cmr katA Cmr katB Cmr ahpF Cmr sodA Cmr a2b2 katA Cmr a2b2 katB Cmr a2b2 ahpF Cmr a2b2 sodA Cmr a2b2 mrgA Cmr ahpC Spr ahpA Cmr ahpA ahpC Cmr Spr a2b2 ahpC Spr a2b2 ahpA Cmr a2b2 ahpA ahpC Cmr Spr katX Cmr a2b2 katX Cmr ahpC Spr ahpA Cmr

Reference or source

26 Laboratory stock This worka,b This worka,b This worka,b This worka,b This worka,b This worka,c This worka,c This worka,c This worka,c This worka,c This work (HB17033PS832)d This work (HB17303PS832)d This work (HB17033PS2513)d This worka (HB17033PS356)d This work (HB17303PS356)d This work (HB17033PS2516)d This worka,b This worka,c John Helmann John Helmann

a Plasmids constructed as described in Materials and Methods were used to create these insertion mutants. b The background for this strain is PS832. c The background for this strain is PS356. d Chromosomal DNA from the strain to the left of the arrow was used to transform the strain to the right.

NaOH). Spores were prepared at 37°C in 23 SG medium without antibiotics (27), cleaned by repeated washing with water (27), and stored in water at 10°C. All spores used in this work were free (. 98%) of growing cells and cell debris. To measure the resistance of growing cells to hydrogen peroxide or superoxide, cells were grown in LB medium to an optical density at 600 nm (OD600) of 1.0 or 3.0 and aliquots (50 to 100 ml) were mixed with 3 ml of soft agar and overlain on an LB plate without antibiotic. A 0.8-cm-diameter filter disc impregnated with 10 ml of 880 mM hydrogen peroxide or 300 mM paraquat was placed in the center of each plate. The plates were incubated for ;18 h at 37°C, and the diameter of the zone of inhibition around each disc was measured. All of these measurements were done at least in duplicate and each experiment was carried out at least twice. Determination of the resistance of spores to hydrogen peroxide, alkyl hydroperoxides, or heat was carried out as previously described (28, 30, 32). Again, all measurements of spore resistance were carried out at least twice and on at least two independent spore preparations. In all measurements of the resistance of mutant cells or spores, the resistance of cells or spores of the parental strain prepared at the same time was measured in parallel. Construction of mutant strains. Strains with mutations in genes coding for protective proteins were constructed by insertional mutagenesis with either previously constructed transposon mutants (ahpA and ahpC) or with plasmid pJH101 (16) carrying a fragment of the gene to be mutated. These fragments were generated by PCR, and all PCR primers that were used contained eight extra 59 nucleotides: two G or C residues at the 59 end followed by either a BamHI or an EcoRI restriction enzyme cleavage site. The locations of the primers used for the various genes (numbers indicate the nucleotides in the coding sequences) as well as the locations of the EcoRI (E) and BamHI (B) sites are as follows: (i) ahpF, 59-E-427–440-39 and 39-1081–1095-B-59; katA, 59-E-554– 568-39 and 39-1051–1066-B-59; katB, 59-E-320–332-39 and 39-859–872-B-59; mrgA, 59-E-134–144-39 and 39-301–314-B-59; sodA, 59-E-90–106-39 and 39-488–509-B-59. To isolate a fragment from within the katX gene the entire katX coding sequence plus significant upstream and downstream sequences were amplified by PCR by using as primers an oligonucleotide from residues 59-E-1114–1096-39 upstream of the A residue of the translation start codon (numbered 1) and a second oligonucleotide from residues 39-2288–2306-B-59 downstream of the A residue of the translation start codon. PCR was carried out in 100 ml of a mixture containing 100 ng of B. subtilis chromosomal DNA as template, 0.5 to 1 mM of each primer, 0.2 mM each deoxynucleoside triphosphate, 2 mM MgSO4, and 2 U of Vent or Taq DNA polymerase. Following incubation at 94°C for 3 min, 30 to 40 cycles were run under the following conditions: 1 min at 94°C, 1 min at 50 to 59°C (the specific temperature was determined by the annealing temperature of the primers), and 1 min at 72°C. The PCR fragments from the ahpF, katA, katB, mrgA, and sodA genes were digested with BamHI and EcoRI and ligated with plasmid pJH101, which had also been cut with BamHI and EcoRI, and the ligation mixture was used to transform E. coli JM83 to ampicillin resistance (50 mg/ml).

7421

Clones containing the desired plasmids were identified, and their identities were confirmed by restriction enzyme digestion. The 3.4-kb PCR fragment containing katX was cleaved with HindIII, and the 740-bp fragment from bp 295 to 1035 of the katX coding sequence was isolated and cloned into HindIII-cut pJH101 as described above. In all cases a plasmid containing the desired insert was used to transform B. subtilis PS356 and PS832 to chloramphenicol resistance. Southern blot analysis of appropriately digested chromosomal DNA of transformants confirmed that the genomic structure of the transformants was as expected. Integration of the pJH101 derivative resulted in the truncation of the open reading frame of the targeted gene by 25 to 40%. Isolation and analysis of cell and spore protein. Growth of cells for extraction of protein was done in LB medium. Cells in LB medium (10 ml) were grown to an OD600 of 1 to 3, harvested by centrifugation, and suspended in 200 ml of 10 mM Tris-HCl (pH 8)–150 mM NaCl. Lysozyme was added to a concentration of 0.2 mg/ml, and after ;15 min at 37°C the suspension was sonicated briefly to reduce the viscosity and centrifuged (10 min, 12,000 3 g), and the supernatant fluid was stored frozen. Spore protein was extracted as previously described (27) by first decoating spores with a urea-sodium dodecyl sulfate (SDS)-dithiothreitol mixture containing 10 mM EDTA and then washing the decoated spores and inducing spore lysis by adding lysozyme and the Tris-NaCl buffer described above. After a brief sonication to reduce the viscosity of the extract, the suspension was centrifuged as described above. Protein samples were run on 8.5 or 10% polyacrylamide gels, and the gels were stained for catalase or superoxide dismutase, respectively, as previously described (4, 23). Protein samples for Western blot analysis were run on an SDS– 15% polyacrylamide gel, the proteins were transferred to nitrocellulose paper, and the small subunit of alkyl hydroperoxide reductase was detected by using a 1/5,000 dilution of rabbit anti-Salmonella typhimurium AhpC, followed by alkaline phosphatase coupled to goat anti-rabbit immunoglobulin G and Lumi-Phos 530 (Boehringer Mannheim) as previously described (19). The anti-AhpC serum (35) was the generous gift of G. Storz. Protein was routinely determined by the Lowry method (24).

RESULTS Effects of loss of superoxide dismutase on cells and spores. Work with several types of bacteria including E. coli, Pseudomonas aeruginosa, and S. typhimurium has shown that these organisms have two different types of cytoplasmic superoxide dismutases, a Mn21 enzyme and an Fe21 enzyme (7, 18). While loss of either enzyme alone does not have a large phenotypic effect, loss of both enzymes results in cells that are extremely sensitive to oxidative stress (7, 15, 18). These double mutants are also more sensitive than wild-type cells to elevated temperatures (5, 6). Superoxide dismutases have not been studied in Bacillus species, but the B. subtilis genomic sequencing project has identified a gene (yqgD) coding for a protein whose sequence is ;60% identical to that of the Mn21 superoxide dismutase of E. coli. We have termed this gene sodA, and to date this is the only sod gene identified in B. subtilis. Analysis of superoxide dismutase activity on acrylamide gels of total B. subtilis vegetative cell protein gave only a single band of enzyme activity, and this band was absent with protein from growing cells of the sodA mutant (Fig. 1, lanes 1 and 2). Analysis of the sodA mutant cells showed that their growth rate at 37°C in an aerobic rich medium (LB) was identical to that of wild-type cells, but the sodA cells only reached ;60% of the OD600 achieved by the wild-type cells (data not shown). In addition, early in stationary phase sodA cells were significantly more temperature sensitive than were wild-type cells (Fig. 2). Analysis of the sensitivity of mutant and wild-type cells to paraquat, which generates superoxide in vivo (14), also showed that sodA cells were significantly more paraquat sensitive (Table 2). Since mutation of the sodA gene has a clear phenotype in growing cells of B. subtilis, we further examined the effect of the sodA mutation on spore properties. As found in growing cells, only a single band of superoxide dismutase activity was present in spores, and this band was absent from spores of the sodA mutant (Fig. 1, lanes 3 and 4). The sodA mutant sporulated normally (data not shown), and the sodA spores exhibited heat and hydrogen peroxide resistance identical to that of

7422

CASILLAS-MARTINEZ AND SETLOW

J. BACTERIOL. TABLE 2. Resistance of growing cells to various agentsa Strain (genotype)

Inhibitor

Diam of zone of inhibition (mm)

PS832 (wtb) PS2495 (sodA) PS832 (wt) PS2488 (katA) PS2489 (katB) PS2495 (sodA) PS2558 (katX)

Paraquat Paraquat H2O2 H2O2 H2O2 H2O2 H2O2

28 45 11 35 11 10 11

a Cells were grown and the inhibition of their growth by various compounds was tested as described in Materials and Methods. b wt, wild type.

FIG. 1. Analysis of superoxide dismutase activity in cells and spores of B. subtilis. Extracts of cells grown in LB medium and spores of wild type (wt) (PS832) and the sodA mutant (PS2495) were prepared, aliquots were run on a 10% polyacrylamide gel, and superoxide dismutase activity (arrow) was detected as described in Materials and Methods. The amounts of protein from cells or spores run on the various lanes are as follows: lane 1, 18 mg from growing cells of PS2495; lane 2, 6 mg from growing cells of PS832; lane 3, 6 mg from spores of PS2495; lane 4, 2 mg from spores of PS832. The dye front of the gel is just above the bottom of the figure.

wild-type spores (Fig. 3A, and data not shown). These properties were also examined with spores lacking the two major DNA protective a/b-type SASP (a2b2 spores). Although a2b2 spores are significantly more heat and hydrogen peroxide sensitive than are wild-type spores (26, 30), the resistance of a2b2 spores was not further lowered by the sodA mutation (Fig. 3A and data not shown). Effects of loss of catalase and MrgA on cells and spores. Previous work has identified two major catalases in B. subtilis, the products of the katA and katB genes (the latter has also been termed katE) (8, 12, 23). KatA appears to be the major catalase in growing cells and plays a significant role in the resistance of growing cells to hydrogen peroxide (8, 13, 23). KatB is generally found only in stationary-phase cells, and katB is under the control of the RNA polymerase sigma factor sB. In contrast to KatA, KatB appears to play no significant role in

FIG. 2. Temperature sensitivity of early stationary-phase cells of the wildtype PS832 and sodA (PS2495) strains. Cells of PS832 and PS2495 were grown at 37°C in LB medium without antibiotics. When the OD600 of the cultures reached 3.0, both were shifted to 50°C and the viable cells in each culture were determined by plate counts at various times after the shift. Symbols: E, PS832; F, PS2495.

the hydrogen peroxide resistance of growing or stationaryphase cells (13, 23). KatB but not KatA has been reported to be present in the spore (23). Recently, a third gene encoding a putative catalase, termed katX, has been identified in the course of the B. subtilis genome sequencing project (36). However, the katX gene product has not yet been identified. In order to definitively identify the catalase(s) present in spores we generated insertion mutations in the katA, -B, and -X genes. As found previously, katA mutants lacked catalase in vegetative cells (data not shown), and growing cells of katA mutants were quite sensitive to hydrogen peroxide (13, 23) (Table 2). In contrast, neither katB nor katX mutants exhibited any increase in hydrogen peroxide sensitivity relative to that of wild-type cells (Table 2). This has been seen previously for katB mutants (13). Analysis of spore extracts revealed only a single band of catalase activity on polyacrylamide gels (Fig. 4, lane 1). This band was also present in extracts of katA and katB spores (Fig. 4, lanes 2 and 3) but was absent in katX mutant spores (Fig. 4, lane 4). This indicates that the major, if not only, catalase present in spores is KatX, although previous work had suggested that KatB might be the major spore catalase (reference 23 but also see Discussion). Analysis of katX spores showed that their heat and hydrogen peroxide resistances were identical to those of wild-type spores (Fig. 3B and data not shown); the heat and hydrogen peroxide resistances of a2b2 katX spores were also identical to those of the a2b2 parent spores (data not shown). However, we did notice that introduction of the katX mutation in either a wild-type or an a2b2 background slowed sporulation significantly (.20%) (10). As expected, the heat and hydrogen peroxide resistances of katA and katB spores were identical to those of their parental counterparts in both wild-type and a2b2 backgrounds (data not shown). We also analyzed the effect of an mrgA mutation on spore hydrogen peroxide resistance, as previous work has shown that MrgA can provide some hydrogen peroxide resistance to growing cells (11). However, there was no effect of an mrgA mutation on hydrogen peroxide resistance of either wild-type or a2b2 spores (data not shown). Effect of inactivation of ahp genes on spore resistance. Additional oxidizing agents to which cells can be exposed are alkyl hydroperoxides; cells generally have an alkyl hydroperoxide reductase(s) which can detoxify such compounds (22, 35). These enzymes generally have two different subunits of ;20 to 25 kDa and 55 to 60 kDa and use either NADH or NADPH as the reductant. Growing cells lacking these enzymes display increased sensitivity to exogenous alkyl hydroperoxides (22, 35). In B. subtilis two different alkyl hydroperoxide reductases have been tentatively identified. One enzyme is encoded by the ahpC and ahpF genes, which are adjacent to each other and are

VOL. 179, 1997

RESISTANCE OF B. SUBTILIS SPORES TO OXIDIZING AGENTS

7423

FIG. 4. Catalase in spores of various strains. Protein (25 to 30 mg) from extracts of spores of various strains were run on an 8.5% polyacrylamide gel and stained for catalase activity (arrow) as described in Materials and Methods. Lanes: 1, PS832 (wild type); 2, PS2488 (katA); 3, PS2489 (katB); 4, PS2558 (katX). The dye front of the gel was just below the bottom of the figure.

FIG. 3. Hydrogen peroxide resistance of spores of strains with and without a sodA mutation (A) or with and without a katX mutation (B). Spores at an OD600 of ;1.0 were made to 5% in hydrogen peroxide at room temperature and the percentages of survivors were determined at various times. Symbols: E, PS832 (wild type); F, PS2495 (sodA); Ç, PS356 (a2b2); å, PS2506 (a2b2 sodA) (A); E, PS832 (wild type); F, PS2558 (katX) (B). The slight variation in resistance of the spores at 30 min between panels A and B is undoubtedly due to the fact that these were results from different spore preparations.

most likely cotranscribed (3, 9); B. subtilis AhpC exhibits 66% sequence identity to AhpC from S. typhimurium (3). The second enzyme is encoded by the ahpA and ahpT genes, which may also be cotranscribed, but the AhpC-like subunit from this enzyme (AhpA) has significantly less sequence identity to S. typhimurium AhpC than does B. subtilis AhpC (37). A mutation in the ahpC gene alone renders growing cells of B. subtilis sensitive to alkyl hydroperoxides (3), but the enzyme encoded by ahpA and ahpT has not been characterized functionally or biochemically. The products of the ahpC and ahpF genes are present in growing cells, and both genes can be induced by appropriate treatments or mutations (3, 9). However, there is no information on AhpC and -F levels in spores. Consequently we carried out Western blot analysis with extracts of growing cells and spores of several strains using an antiserum against S. typhimurium AhpC (Fig. 5). This antiserum detected a protein with a molecular mass of ;22 kDa (the expected molecular

mass for AhpA or AhpC is approximately 20,500 Da) in both cells and spores, and this protein was absent in extracts of the ahpC and ahpA ahpC mutants (Fig. 5). The Western blot data give no definitive information on the levels of AhpA in cells and spores, since the antiserum may not have detected this protein. However, AhpC and almost certainly AhpF, since ahpF is most likely cotranscribed with ahpC, are present in spores, although their levels in spores may be lower than those in growing cells. Despite the presence of AhpC in spores, inactivation of ahpA, ahpC, ahpF, or both ahpA and ahpC had no effect on the resistance of either wildtype or a2b2 spores to t-butyl hydroperoxide (Fig. 6 and data not shown), cumene hydroperoxide, or heat (data not shown). The spores of the various ahp mutants also exhibited no increased resistance to hydrogen peroxide (data not shown). Previous work has shown that an ahpC mutation results in increased hydrogen peroxide resistance in growing cells due to induction of the per regulon (3, 9).

FIG. 5. Levels of AhpC in growing cells and spores of various strains. Extracts of growing cells or spores were prepared, samples (15 mg of protein) were run on an SDS–15% polyacrylamide gel, proteins were transferred to paper, and the small subunit of alkyl hydroperoxide reductase was detected as described in Materials and Methods. The samples run in lanes 1 to 4 were from growing cells (OD600, ;1.0); the samples in lanes 5 to 8 were from spores. Lanes: 1 and 5, strain PS832 (wild type); 2 and 6, strain PS2513 (ahpA); 3 and 7, strain PS2512 (ahpC); 4 and 8, strain PS2514 (ahpA ahpC). Molecular mass markers (not shown) were run adjacent to lane 8. The arrow indicates the position of a band of 22 kDa that reacts with the antiserum but is absent in extracts from ahpC mutants; this band seems most likely to be B. subtilis AhpC. The bands labelled c are presumably non-Ahp proteins that react with the antisera used.

7424

CASILLAS-MARTINEZ AND SETLOW

FIG. 6. Alkyl hydroperoxide resistance of spores of various strains. Spores at an OD600 of 1.0 were made 0.73 M in t-butyl hydroperoxide and incubated at 60°C, and the percentages of survivors were determined at various times. Symbols: E, PS832 (wild type); F, PS2514 (ahpA ahpC); Ç, PS356 (a2b2); å, PS2517 (a2b2 ahpA ahpC).

DISCUSSION The results presented in this communication allow a number of conclusions. First, there appears to be only a single major superoxide dismutase, most likely a Mn21-dependent enzyme, in B. subtilis in contrast to the situation in gram-negative bacteria. Furthermore, mutation of the single sodA gene in B. subtilis gives a phenotype, including temperature sensitivity, which is similar in several respects to the phenotype of sodA sodB mutants in E. coli. While it is possible that a second superoxide dismutase exists in B. subtilis, the fact that we obtained a significant phenotype with only an sodA mutation is consistent with SodA being the predominant if not the only superoxide dismutase in B. subtilis. We also observed only a single product (from sodA) upon PCR amplification of potential sod genes in the B. subtilis genome with primers from regions highly conserved in sod genes and low annealing temperatures (10). A single major superoxide dismutase was also found in the gram-positive bacterium Lactobacillus lactis (29). This enzyme shows ;65% amino acid sequence identity with B. subtilis SodA, and mutation of L. lactis sodA results in cells that are impaired in aerobic growth. These findings suggest that gram-positive bacteria may have only a single predominant superoxide dismutase, in contrast to the situation in gramnegative bacteria. Analysis of the B. subtilis genomic sequence, when it is complete, will provide definitive evidence for or against the possibility of an additional enzyme in B. subtilis. One cautionary note about assignment of the phenotype of the B. subtilis sodA mutant solely to the effect of loss of SodA is that downstream of sodA (yqgD) is a second gene, yqgE. This gene shows significant sequence similarity to a gene downstream of E. coli sodA and may well be cotranscribed with sodA (yqgD). Since the insertional mutation in B. subtilis sodA is likely to be polar, it is possible that the phenotype of this mutant is due to the loss of yqgE and not of sodA. However, in L. lactis sodA appears to be a monocistronic gene (29). A second conclusion is that KatX is the major catalase present in spores. Previous work had suggested that KatB was the major spore catalase (23), although it was not rigorously shown that this enzyme was not loosely bound on the outside

J. BACTERIOL.

of the spore. In our preparation of spore extracts we treated spores with denaturing agents which would remove any surface-bound proteins prior to spore lysis. An obvious question then is whether KatX is only found in the spore. Since the katX gene’s sequence is known, it should be relatively straightforward to determine the timing and location of katX expression by analysis of an appropriate katX-reporter construct. To date we have shown that katX is not under control of the RNA polymerase sigma factor sB, in contrast to the situation for katB (12, 31). A second question concerns the possible role of KatX in spores, since it is not involved in dormant spore resistance. Two possibilities come to mind. One is that KatX is involved in the hydrogen peroxide resistance of the forespore during sporulation. A second possibility is that KatX is involved in hydrogen peroxide resistance of the germinated and outgrowing spore, although it is possible that KatA is synthesized during this period as well. Clearly, further studies of KatX function and the timing and location of katX expression are warranted. The third and major conclusion from this work is that MrgA and protective enzymes play no role in spore resistance to oxidative stress or elevated temperatures. This is in contrast to the situation in growing cells of a number of bacteria where proteins like MrgA and enzymes such as catalase and alkyl hydroperoxide reductase are important in resistance to oxidative stress (3, 8, 20, 22, 35), while superoxide dismutase is important in resistance not only to oxidative stress but also to elevated temperatures (5, 6, 7). It is possible that the reason that MrgA plays no role in spore resistance is that this protein is not present in spores, as we have no data on this point. However, SodA is present in spores, although it is possible that SodA has no protective effect because there is never any significant production of superoxide in spores due to their lack of metabolism (33). KatX and AhpC are also in spores, and hydrogen peroxide and alkyl hydroperoxides do enter spores as in some cases these reagents cause damage to spore DNA (30, 32). However, it is possible that levels of these enzymes in dormant spores are too low to play a significant role in hydroperoxide removal. Indeed, loss of the generally minor KatB activity has very little if any effect on the hydrogen peroxide resistance of growing cells (13). While the individual explanations given above as to why specific proteins exert no protective effect in bacterial spores could be correct, we favor a single simple explanation, i.e., enzymes in the spore core are inactive (25, 33). For an individual enzyme such as alkyl hydroperoxide reductase the absence of a required substrate (NADH or NADPH [33]) from the spore will result in a lack of enzyme activity. However, there is significant evidence that enzymes in the spore core, the site of most spore enzymes, are inactive due in large part to the relative dehydration and mineralization of the spore core (25, 33). If this is indeed the case, then all protective enzymes will be inactive and can play no role in spore resistance. Indeed, the importance of other mechanisms which spores have adopted for protection against oxidative stress, i.e., decreased permeability and the binding of a/b-type SASP to DNA, may be because of the conditions in the spore core which preclude enzyme action and thus a role for protective enzymes in spore protection. ACKNOWLEDGMENTS We are grateful to Barbara Setlow for assistance in a few of the experiments, to John Helmann for strains and advice, to Peter Zuber for information about ahpA and ahpT, and to Gisela Storz for the anti-AhpC serum. This work was supported by a grant from the Army Research Office.

RESISTANCE OF B. SUBTILIS SPORES TO OXIDIZING AGENTS

VOL. 179, 1997 REFERENCES

1. Almiron, M., A. J. Link, D. Furlong, and R. Kolter. 1992. A novel DNAbinding protein with regulatory and protective roles in starved Escherichia coli. Genes Dev. 6:2646–2654. 2. Anagnostopoulos, C., and J. Spizizen. 1961. Requirements for transformation in Bacillus subtilis. J. Bacteriol. 81:74–76. 3. Antelmann, H., S. Engelmann, R. Schmid, and M. Hecker. 1996. General and oxidative stress responses in Bacillus subtilis: cloning, expression, and mutation of the alkyl hydroperoxide reductase operon. J. Bacteriol. 178: 6571–6578. 4. Beauchamp, C., and I. Fridovich. 1971. Superoxide dismutase: improved assays and an assay applicable to acrylamide gels. Anal. Biochem. 44:276– 287. 5. Benov, L., and I. Fridovich. 1995. Superoxide dismutase protects against aerobic heat shock in Escherichia coli. J. Bacteriol. 177:3344–3346. 6. Benov, L., and I. Fridovich. 1995. A superoxide dismutase mimic protects sodA sodB Escherichia coli against aerobic heating and stationary phase death. Arch. Biochem. Biophys. 322:291–294. 7. Beyer, W., J. Imlay, and I. Fridovich. 1991. Superoxide dismutases. Prog. Nucleic Acid Res. 40:221–253. 8. Bol, D. K., and R. E. Yasbin. Characterization of an inducible oxidative stress system in Bacillus subtilis. J. Bacteriol. 172:3503–3506. 9. Bsat, N., L. Chen, and J. D. Helmann. 1996. Mutation of the Bacillus subtilis alkyl hydroperoxide reductase (ahpCF) operon reveals compensatory interactions among hydrogen peroxide stress genes. J. Bacteriol. 178:6579–6586. 10. Casillas-Martinez, L., and P. Setlow. 1997. Unpublished results. 11. Chen, L., and J. D. Helmann. 1995. Bacillus subtilis MrgA is a Dps (PexB) homologue: evidence for metalloregulation of an oxidative stress gene. Mol. Microbiol. 18:295–300. 12. Davidson, J. F., B. Whyte, P. H. Bissinger, and R. H. Schiestl. 1996. Oxidative stress is involved in heat-induced cell death in Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. USA 93:5116–5121. 13. Engelmann, S., and M. Hecker. 1996. Impaired oxidative stress resistance of Bacillus subtilis sigB mutants and the role of katA and katE. FEMS Microbiol. Lett. 145:63–69. 14. Farrs, S. F., and F. Kogoma. 1991. Oxidative stress responses in Escherichia coli and Salmonella typhimurium. Microbiol. Rev. 55:561–585. 15. Fan, S. B., R. D’Ari, and D. Tovati. 1986. Oxygen-dependent mutagenesis in Escherichia coli lacking superoxide dismutase. Proc. Natl. Acad. Sci. USA 83:8268–8272. 16. Ferrari, F. A., A. Nguyen, D. Long, and J. A. Hoch. 1983. Construction and properties of an integrable plasmid for Bacillus subtilis. J. Bacteriol. 154: 1513–1515. 17. Gerhardt, P., and R. E. Marquis. 1989. Spore thermoresistance mechanisms, p. 17–63. In I. Smith, R. Slepecky, and P. Setlow (ed.), Regulation of prokaryotic development. American Society for Microbiology, Washington, D.C. 18. Hassett, D. J., H. P. Schweizer, and D. E. Ohman. 1995. Pseudomonas aeruginosa sodA and sodB mutants defective in manganese- and iron-cofactored superoxide dismutase activity demonstrate the importance of the ironcofactored form in aerobic metabolism. J. Bacteriol. 177:6330–6337.

7425

19. Illades-Aguiar, B., and P. Setlow. 1994. Studies of the processing of the protease which initiates degradation of small, acid-soluble proteins during germination of spores of Bacillus species. J. Bacteriol. 176:2788–2795. 20. Imlay, J., and S. Linn. 1987. Mutagenesis and stress responses induced in Escherichia coli by hydrogen peroxide. J. Bacteriol. 169:2967–2976. 21. Imlay, J. A., and S. Linn. 1988. DNA damage and oxygen radical toxicity. Science 240:1302–1309. 22. Jacobson, F. S., R. B. Morgan, M. F. Christman, and B. N. Ames. 1989. An alkyl hydroperoxide reductase from Salmonella typhimurium involved in the defense of DNA against oxidative DNA damage. J. Biol. Chem. 264:1488– 1496. 23. Loewen, P. C., and J. Switala. 1987. Multiple catalases in Bacillus subtilis. J. Bacteriol. 169:3601–3607. 24. Lowry, O. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265–275. 25. Marquis, R. E., J. Sim, and S. Y. Shin. 1994. Molecular mechanisms of resistance to heat and oxidative damage. J. Appl. Bacteriol. 76:40S–48S. 26. Mason, J. M., and P. Setlow. 1986. Evidence for an essential role for small, acid-soluble, spore proteins in the resistance of Bacillus subtilis spores to ultraviolet light. J. Bacteriol. 167:174–178. 27. Nicholson, W. L., and P. Setlow. 1990. Sporulation, germination and outgrowth, p. 391–450. In C. R. Harwood and S. M. Cutting (ed.), Molecular biological methods for Bacillus, John Wiley and Sons, Chichester, United Kingdom. 28. Popham, D. L., S. Sengupta, and P. Setlow. 1995. Heat, hydrogen peroxide, and UV resistance of Bacillus subtilis spores with increased core water content and with or without major DNA binding proteins. Appl. Environ. Microbiol. 61:3633–3638. 29. Sanaders, J. W., K. J. Leenhouts, A. J. Haandrikman, G. Venema, and J. Kole. 1995. Stress response in Lactobacillus lactis: cloning, expression analysis, and mutation of the lactococcal superoxide dismutase gene. J. Bacteriol. 177:5254–5260. 30. Setlow, B., and P. Setlow. 1993. Binding of small, acid-soluble spore proteins to DNA plays a significant role in the resistance of Bacillus subtilis spores to hydrogen peroxide. Appl. Environ. Microbiol. 59:3418–3423. 31. Setlow, B., and P. Setlow. 1997. Unpublished results. 32. Setlow, B., C. A. Setlow, and P. Setlow. 1997. Killing bacterial spores by organic hydroperoxides. J. Indust. Microbiol. 18:384–388. 33. Setlow, P. 1994. Mechanisms which contribute to the long-term survival of spores of Bacillus species. J. Appl. Bacteriol. 176(Suppl.):49S–60S. 34. Setlow, P. 1995. Mechanisms for the prevention of damage to the DNA in spores of Bacillus species. Annu. Rev. Microbiol. 49:29–54. 35. Storz, G., F. S. Jacobson, L. A. Tartaglia, R. W. Morgan, L. A. Silveira, and B. N. Ames. 1989. An alkyl hydroperoxide reductase induced by oxidative stress in Salmonella typhimurium and Escherichia coli: genetic characteristics and cloning of ahp. J. Bacteriol. 171:2049–2055. 36. Yoshida, K., K. Shindo, H. Sano, S. Seki, M. Fujimura, N. Yanai, and Y. Fujita. 1996. Sequencing of a 65 kb region of the Bacillus subtilis genome containing the lic and cel loci, and creation of a 177 kb contig covering the gnt-sacXY region. Microbiology 142:3113–3123. 37. Zuber, P. Personal communication.