The katX Gene, Which Codes for the Catalase in Spores of Bacillus subtilis, Is a Forespore-Specific Gene Controlled by ς^F, and KatX Is Essential for Hydrogen Peroxide Resistance of the Germinating Spore

Abstract

Previous work has shown that the katX gene encodes the major catalase in dormant spores of Bacillus subtilis but that this enzyme has no role in dormant spore resistance to hydrogen peroxide. Expression of a katX-lacZ fusion began at approximately h 2 of sporulation, and >75% of the katX-driven β-galactosidase was packaged into the mature spore. A mutation in the gene coding for the sporulation-specific RNA polymerase sigma factor ς^F abolished katX-lacZ expression, while mutations in genes encoding ς^E, ς^G, and ς^K did not. Induction of ς^F synthesis in vegetative cells also resulted in katX-lacZ expression, while induction of ς^G expression did not; the katX-lacZ fusion was also not induced by hydrogen peroxide. Upstream of the in vivo katX transcription start site there are sequences with good homology to those upstream of known ς^F-dependent start sites. These data indicate that katX is an additional member of the forespore-specific ς^F regulon. A mutant in the katA gene, encoding the major catalase in growing cells, was sensitive to hydrogen peroxide during sporulation, while a katX mutant was not. However, outgrowth of katX spores, but not katA spores, was sensitive to hydrogen peroxide. Consequently, a major function for KatX is to protect germinating spores from hydrogen peroxide.

Aerobic organisms must constantly deal with a variety of reactive oxygen species which can cause damage to DNA and proteins. One common reactive oxygen species is hydrogen peroxide, and a major defense against this molecule is its enzymatic destruction by catalase. Bacteria often contain multiple catalases (14, 17), and growing cells of Bacillus subtilis contain at least two major catalases, the products of the katA and katB genes (2, 8, 18). KatA is the major catalase in growing cells, its level is increased by exposure to hydrogen peroxide, and loss of this enzyme results in increased sensitivity of growing cells to hydrogen peroxide (2, 3). In contrast, the katB gene is under control of the stress-regulated RNA polymerase sigma factor ς^B, KatB levels are not increased specifically in response to hydrogen peroxide, and loss of this enzyme does not increase a growing cell’s sensitivity to hydrogen peroxide (8, 9). Recently, a third catalase, the product of the katX gene (42), was identified as the major if not only catalase in dormant spores (4). KatX was not found in growing cells, and loss of this enzyme had no effect on vegetative cell hydrogen peroxide resistance.

Dormant spores of B. subtilis are much more resistant than are growing cells to hydrogen peroxide (21, 30). Factors contributing to increased spore hydrogen peroxide resistance include the low permeability of spores to hydrophilic compounds (11) and the protection of spore DNA from damage by its saturation with a novel group of DNA binding proteins (30). It was possible that catalase was involved in spore resistance to hydrogen peroxide, and as noted above, spores contain their own catalase, KatX (4). However, loss of this enzyme (or of KatA or KatB) had no effect on dormant spore hydrogen peroxide resistance (4). Thus, the precise function of KatX is not clear.

While KatX was readily detected in the dormant spore, it was not found in growing cells, even in cells of katA and katB mutants (4), which suggests that KatX is a spore-specific enzyme. One characteristic of spore-specific proteins is that their genes are expressed only in the forespore compartment of the sporulating cell. Forespore-specific transcription in B. subtilis is directed by RNA polymerase carrying one of the two sigma factors ς^F and ς^G, which have similar but distinct promoter specificities (15, 31, 38). While a number of genes that are members of the ς^G regulon are known, there are many fewer genes which are known to be under ς^F control (13, 15, 31, 33, 38), and the characteristics of good ς^F-dependent promoters are based on analyses of relatively few genes. Consequently, elucidation of the factors controlling katX expression could be quite informative. In this work, we report that katX is under ς^F control; we also report that KatX plays an important role in the resistance of germinating spores to hydrogen peroxide.

MATERIALS AND METHODS

Bacteria and plasmids used, growth, sporulation, and spore germination.

The bacterial strains and plasmids used in this work are listed in Table 1. Escherichia coli and B. subtilis strains were routinely grown at 37°C in 2xYT medium (16 g of tryptone, 10 g of yeast extract, and 5 g of NaCl per liter), and E. coli TG1 (28) was routinely used for cloning purposes. Production of B. subtilis spores was at 37°C in 2xSG medium, and spores were purified and stored as described previously (25). The resuspension method (36) was used for analysis of gene expression during sporulation and for analysis of hydrogen peroxide sensitivity during sporulation. Spores were routinely heat shocked (30 min at 70°C) in water prior to initiation of spore germination, and spores were germinated at an optical density at 600 nm (OD₆₀₀) of 0.5 to 0.7 in 2xSG medium (25) without glucose and with 4 mM l-alanine to promote initiation of germination.

TABLE 1.

B. subtilis strains and plasmids used

Strain or plasmid	Genotype and phenotype	Source or reference
B. subtilis strains
PS832a	Trp+ revertant of strain 168	Laboratory stock
PS2488a	katA::cat Cmr	4
PS2558a	katX::cat Cmr	4
PS2573a,c	katA::katA-lacZ Cmr	YB2003→PS832d
PS2663a	katX::erm Emr	pCM::Erm→PS2558
PS2664a	katA::cat katX::erm Cmr Emr	PS2663→PS2488
PY79b	Wild type	43
RL831b	spoIIIG::kan Kmre	R. Losick
SC137b	spoIIGB::erm Emr	S. Cutting
SC500b	spoIIIGΔ1	S. Cutting
SC829b	spoIVCB23	S. Cutting
IB427a	katX::katX-lacZ Cmr	pIB426→PS832
IB429b	katX::katX-lacZ spoIIAC Cmr	IB427→SC1159
IB430b	katX::katX-lacZ spoIIGB Cmr Emr	IB427→SC137
IB431b	katX::katX-lacZ spoIIIG Cmr	IB427→SC500
IB432b	katX::katX-lacZ spoIVCB Cmr	IB427→SC829
IB434b	katX::katX-lacZ Cmr	IB427→PY79
IB435b	katX::katX-lacZ spoIIIG [pSDA4] Cmr Kmr	pSDA4→IB431
IB436b	katX::katX-lacZ spoIIIG [pDG298] Cmr Kmr	pDG298→IB431
SC1159b	spoIIAC1	S. Cutting
IB439a	amyE::katX-lacZ Cmr	pIB437→PS832
IB441a	amyE::katX-lacZ spoIIIG::kan Cmr Kmr	RL831→IB439
IB446a	katA::erm Emr	pCm::Erm→PS2488
IB447a	katA::erm katX::katX-lacZ Cmr Emr	IB427→IB446
YB2003c	katA::katA-lacZ Cmr	2
Plasmids
pCm::Erm	Amprf Emr	35
pCR2.1	Ampr Kmr	Invitrogen
pDG268	Ampr Cmr	37
pDG298	Kmr, carries spoIIIG (ςG) under Pspac control	40
pIB426	Ampr Cmr, translational katX-lacZ fusion in pJF751	This work
pIB437	Ampr Cmr, translational katX-lacZ fusion in pDG268	This work
pIB449	1,181 bp of katX coding and upstream sequence in pCR2.1	This work
pJF751	Ampr Cmr	10
pSDA4	Ampr Kmr, carries spoIIAC (ςF) under Pspac control	32

^aGenetic background is PS832. 

^bGenetic background is PY79. 

^ckatA mutant. 

^dThe source of the donor DNA in transformation is to the left of the arrow, and the recipient is to the right. 

^eKm^r, kanamycin resistance. 

^fAmp^r, ampicillin resistance. 

Construction of B. subtilis strains containing a translational katX-lacZ fusion.

A fragment from 193 bp upstream of to 25 bp into the katX open reading frame (ORF) (42) was amplified by PCR. The primers used were katX5′ (5′-GGAATTCGGGCAAGCTCAAGAGCGG-3′) and katX3′ (3′-CTTTCTACTAGTAGTTTTGTTCGCCTAGGGC-5′), with extra nucleotides including EcoRI or BamHI sites at the 5′ ends of the primers (underlined residues) to facilitate cloning. The PCR product was cut with EcoRI and BamHI and cloned in E. coli between the EcoRI and BamHI sites of plasmid pJF751 (10). The resulting plasmid was linearized with BamHI, treated with T4 DNA polymerase to fill the ends, ligated, and cloned in E. coli. The resulting plasmid (pIB426) contained a translational katX-lacZ fusion and was integrated at the katX locus on the chromosome of B. subtilis PS832 by a single crossover event with selection for resistance to chloramphenicol (5 μg/ml; Cm^r). Southern blot analysis of the resulting strain (IB427) showed the presence of only one copy of the katX-lacZ fusion at the katX locus (data not shown). Chromosomal DNA was isolated from cells of strain IB427 and used to transform B. subtilis PY79 (43) and its derivatives containing mutations in genes coding for various sporulation sigma factors to Cm^r. Since the katX promoter region is completely within the PCR fragment used to create the katX-lacZ fusion (see Results), these strains are not katX mutants.

To integrate the katX-lacZ fusion at the amyE locus, plasmid pIB426 was cut with EcoRI and ClaI, and the 1,050-bp fragment carrying the katX regulatory region fused to lacZ was isolated and cloned between the EcoRI and ClaI sites of plasmid pDG268 (37) in E. coli. The resulting plasmid, pIB437, was linearized with PstI and transformed into B. subtilis PS832 with selection for Cm^r, and the absence of amylase activity in the resulting strain (IB439) was determined as described previously (6). Strain IB439 was transformed to erythromycin resistance (Em^r) with chromosomal DNA from B. subtilis RL831 containing a mutation in the spoIIIG gene. A transformant containing the katX-lacZ fusion at the amyE locus as well as a mutation in the spoIIIG gene was called strain IB441.

For analysis of the katX transcription start site, we also cloned a fragment carrying a larger amount of DNA sequence upstream of the katX ORF. A 1,181-bp fragment from 1,114 bp upstream of to 67 bp into the katX ORF was amplified by PCR. The primers used were katX-45 (5′-GAGAAAACGCTTCCTCGCTCC-3′) and katX-4Rev (5′-CCGGATCCCCGCCATCAGCAACGCCAG-3′), with the latter primer containing extra 5′ residues (underlined). The PCR product was cloned into plasmid pCR2.1 by using a TA cloning kit (Invitrogen) according to the manufacturer’s instructions. The resulting plasmid was called pIB449.

Construction of katA and katX mutants.

To generate a katX mutant strain in which the Cm^r marker was exchanged for an Em^r marker, strain PS2558 (katX Cm^r) was transformed to Em^r with loss of Cm^r, using EcoRI-linearized plasmid pCm::Erm (35), generating strain PS2663. DNA from this strain was then used to transform strain PS2488 (katA Cm^r) to Em^r, generating the katA katX mutant strain PS2664. To generate a strain with a katX-lacZ fusion that was also a katA mutant, the Cm^r marker in strain PS2488 was exchanged for an Em^r marker by using plasmid pCM::Erm as described above, generating strain IB446. This strain was then transformed with DNA from strain IB427 (katX::katX-lacZ), generating strain IB447.

Determination of the katX transcriptional start site.

Total RNA was extracted from sporulating cells of B. subtilis IB439 or IB441 after 4 (IB439) or 5 (IB441) h of sporulation in 2xSG medium as described previously (24, 25). The RNA was then used in a primer extension analysis (24) using as the primer either katX-45 (complementary to nucleotides [nt] 67 to 46 in katX mRNA) or lacZ-70 (5′-AAGGCGATTAAGTTGGGTAACG-3′) complementary to nt 69 to 47 in katX-lacZ mRNA; note that the latter region of katX-lacZ mRNA has only lacZ sequence. Primer extension reactions were performed with avian myeloblastosis virus reverse transcriptase at 47°C and analyzed as described previously (24). Appropriate DNA size standards were produced by using the same primers in DNA sequencing reactions. The katX-45 primer was used with plasmid pIB449, and the lacZ-70 primer was used with plasmid pIB437.

Other methods.

Germinated spores, vegetative cells, and sporulating cells were permeabilized and assayed for β-galactosidase with o-nitrophenyl-β-d-galactopyranoside as described previously (25). Spores were decoated prior to rupture with lysozyme to allow assay of β-galactosidase (25). To test for induction of katX by hydrogen peroxide, B. subtilis strains with the katX-lacZ fusion were grown at 37°C in LB medium (10 g of tryptone, 5 g of yeast extract, 10 g of NaCl, and 1 ml of 1 N NaOH per liter) to an OD₆₀₀ of 0.07, hydrogen peroxide was added to 50 μM (3), and aliquots were taken subsequently for assay of β-galactosidase. All β-galactosidase specific activities are expressed in Miller units (22). Analysis of viability of cells or spores was by plating dilutions on LB plates with appropriate antibiotics as described previously (4). Dipicolinic acid (DPA) was extracted from sporulating cells and assayed as described elsewhere (27).

RESULTS

Expression of katX during growth and sporulation.

Previous work has shown that KatX is present in B. subtilis spores but not growing cells (4), suggesting that the katX gene is expressed only during sporulation. To analyze this possibility in detail, we constructed a B. subtilis strain containing a katX-lacZ fusion integrated at the katX locus and analyzed β-galactosidase expression during growth and sporulation. No expression of the katX-lacZ fusion was observed in a medium (LB) that does not support sporulation, either in vegetative growth or in early stationary phase (Fig. 1). In contrast, a katA-lacZ fusion was found to exhibit significant expression in vegetative cells and higher expression in stationary-phase cells (3) (Fig. 1). The katA-lacZ fusion was also induced significantly by sublethal hydrogen peroxide treatment (3), but katX-lacZ expression was not induced by 50 μM H₂O₂ in either a katA mutant or an otherwise wild-type strain (data not shown). These data indicate that katX is not a part of an oxidative stress regulon. However, the katX-lacZ fusion was expressed starting 1.5 to 2 h after induction of sporulation (Fig. 2). This timing coincided with the start of expression of the ς^F-dependent gene spoIIR (data not shown). The kinetics and levels of β-galactosidase expression from the translational katX-lacZ fusion incorporated at the katX or amyE locus were similar (data not shown), indicating that the 193-bp region upstream of katX contains the complete katX promoter. Analysis of the β-galactosidase level in spores containing the katX-lacZ fusion further showed that >75% of the β-galactosidase accumulated during sporulation was incorporated into the mature spore (data not shown). During the first 90 min of germination and outgrowth of spores of strain IB427, there was no significant (<15%) increase in the amount of β-galactosidase (data not shown), indicating that katX is not significantly transcribed during this period of development. However, low expression of the katA-lacZ fusion began ∼25 min after the initiation of spore germination (data not shown).

FIG. 1.

Expression of katA- and katX-lacZ fusions in growing and stationary-phase cells. Strains PS2573 (katA-lacZ) and IB427 (katX-lacZ) were grown at 37°C in LB medium, and samples were taken for assay of β-galactosidase as described in Materials and Methods. Symbols: ○ and •, strain PS2573; ▵ and ▴, strain IB427; ○ and ▵, OD₆₀₀; • and ▴, β-galactosidase specific activity.

FIG. 2.

Expression of the katX-lacZ fusion in spo mutants. Strains of the wild-type or spo strains carrying the katX-lacZ fusion were sporulated by the resuspension method, and samples were taken for assay of β-galactosidase as described in Materials and Methods. Time zero is the time of initiation of sporulation. Symbols: •, IB434 (wild type); ○, IB431 (spoIIIG); □, IB429 (spoIIAC); ◊, IB430 (spoIIGB); ▵, IB432 (spoIVCB) (ς^K).

Sigma factor dependence of katX expression.

The fact that katX-driven β-galactosidase first appeared 1.5 to 2 h after initiation of sporulation and was found in mature spores suggested that katX is transcribed by RNA polymerase containing either ς^F (Eς^F) or Eς^G or both enzymes. To obtain more evidence on this point, we analyzed the expression of the katX-lacZ fusion in strains with mutations in genes coding for sporulation-specific sigma factors (Fig. 2). Mutations in the spoIVCB and spoIIGB genes, coding for the late mother cell sigma factor ς^K and the early mother cell sigma factor ς^E, respectively (15), did not have a large effect on katX-lacZ expression. However, mutation of the spoIIAC gene, coding for ς^F, abolished katX-lacZ expression, consistent with katX being a ς^F-dependent gene. Interestingly, a mutation in the spoIIIG gene, coding for the late forespore sigma factor ς^G, resulted in 1.5-fold overexpression of katX-lacZ. The elevated expression of a ς^F-dependent gene (csfC) in a mutant lacking ς^G has been noted previously (7).

The data noted above suggested not only that katX is transcribed by Eς^F but also that this gene is not transcribed by Eς^G. To prove conclusively that Eς^G does not direct katX expression, we introduced plasmid pDG298 (40), containing the structural gene for ς^G (spoIIIG) under the control of a promoter (P_spac) inducible by isopropyl-β-d-thiogalactopyranoside (IPTG), into a strain containing the katX-lacZ fusion as well as a mutation in the chromosomal copy of spoIIIG. Upon induction of ς^G synthesis in vegetatively growing cells of this strain, we observed no increase in β-galactosidase activity (Fig. 3), showing that ς^G was unable to direct expression of katX-lacZ. However, vegetative cells containing plasmid pSDA4 (32) carrying an IPTG-inducible spoIIAC gene (coding for ς^F) rapidly accumulated katX-driven β-galactosidase upon induction with IPTG (Fig. 3). We therefore conclude that katX is transcribed exclusively by Eς^F.

FIG. 3.

Induction of the katX-lacZ fusion in vegetative cells producing ς^F or ς^G. Strains IB435 (producing ς^F under P_spac control) and IB436 (producing ς^G under P_spac control) were grown in 2xYT medium. At an OD₆₀₀ of 0.25, each culture was split in half, and one half was made 2 mM in IPTG. Samples were taken subsequently for assay of β-galactosidase as described in Materials and Methods. Symbols: ○ and •, strain IB435; ▵ and ▴, strain IB436; • and ▴, without IPTG; ○ and ▵, with IPTG.

Mapping of the katX promoter.

To precisely identify the katX promoter, we carried out primer extension analysis using RNA from sporulating cells of strains IB439 (amyE::katX-lacZ) and IB441 (amyE::katX-lacZ spoIIIG). Two different primers were used for this analysis; one annealed to the lacZ portion of katX-lacZ mRNA, while the other annealed to the katX mRNA. Both primers gave the same 5′ end with RNA from both strains (Fig. 4 and 5 and data not shown); presumably this 5′ end is the start site for katX transcription which began 23 nt upstream of the katX translational start codon (TTG) at a G residue (Fig. 4 and 5). Upstream of the katX transcription start site there are also sequences with good homology to those upstream of known ς^F-dependent promoters (see Discussion). The katX transcript was not observed in cells in which sporulation had just been initiated (data not shown).

FIG. 4.

Primer extension analysis of the katX transcription start site. RNA was isolated from sporulating cells, and primer extension analysis was carried out with primer katX-45 and products analyzed as described in Materials and Methods. Lanes A, G, C, and T, sequencing reactions with primer katX-45 and plasmid pIB449; lane 1, primer extension product with RNA from strain IB441 (amyE::katX-lacZ spoIIIG); lane 2, primer extension product with RNA from strain IB439 (amyE::katX-lacZ). Extension products are marked with an arrow, and the transcription start site in the katX upstream sequence to the left is marked with a dot.

FIG. 5.

DNA sequence of the amino-terminal coding and upstream regions of katX. Sequences corresponding to −10 and −35 promoter elements are labeled and underlined; the important two G residues upstream of the −10 region are in boldface; the likely ribosome binding site (RBS) is underlined; the nucleotide at the transcription initiation site is in boldface and labeled +1. The amino-terminal coding regions of both katX and a possible divergently transcribed gene (yxlJ) are also shown, with the encoded amino acid given beneath the second nucleotide of each codon. The DNA sequence in this region is from reference 42.

Effect of a katX mutation on sporulation, germination, and outgrowth.

The knowledge that katX was expressed only in the developing forespore during sporulation suggested that KatX might be involved in the hydrogen peroxide resistance of the developing forespore. To test this possibility, we first carried out control experiments, which determined that wild-type and katA, katX, and katA katX mutant strains all sporulated similarly in resuspension medium (data not shown); 0.2 mM hydrogen peroxide also had essentially no effect on the sporulation of a wild-type culture sporulating in resuspension medium, as measured by the OD₆₀₀ and DPA accumulation in the culture (data not shown). However, this amount of hydrogen peroxide caused a large inhibition of subsequent growth in katA or katA katX mutants when added 1 h after initiating sporulation, but there was no significant effect in a katX mutant strain (Fig. 6A). Addition of 0.2 mM hydrogen peroxide 1 h after initiating sporulation also reduced eventual (24 h) DPA accumulation in katA and katA katX strains by 60 to 75% but had no effect on the katX strain (data not shown); DPA accumulation normally began at ∼5 h in the wild-type culture. Addition of hydrogen peroxide to 0.2 mM even 3 h after initiating sporulation still had a significant effect on the subsequent growth of katA and katA katX strains but again had no effect on a katX strain (Fig. 6B). These results suggest that KatA plays a significant role in hydrogen peroxide resistance during sporulation but that KatX does not.

FIG. 6.

Effects of hydrogen peroxide on sporulating cells of various strains. Strains were sporulated by the resuspension method, hydrogen peroxide was added to 0.2 mM 1 h (A) or 3 h (B) after initiating sporulation (arrow), and the OD₆₀₀ of the culture was measured. Symbols: ○, PS832 (wild type); •, PS2488 (katA); ▵, PS2558 (katX); ▴, PS2664 (katA katX).

KatX is the only catalase detected in dormant spores, but katX spores as well as katA katX spores exhibit hydrogen peroxide resistance identical to that of wild-type spores (reference 4 and data not shown). The lack of effect of KatX on spore hydrogen peroxide resistance is presumably due to the general inactivity of enzymes in dormant spores (4). However, KatX and other spore enzymes become active in the first minutes of spore germination, and KatX could play an important role in this period of development. Analysis of the kinetics of spore germination, outgrowth, and resumption of vegetative growth showed that neither katA, katX, or katA katX mutations had any significant effect on these kinetics in the absence of hydrogen peroxide (data not shown). Hydrogen peroxide at up to 10 mM also did not interfere with the initiation of spore germination (as monitored by the fall in OD₆₀₀ of a spore culture), and katA, katX, and katA katX spores initiated spore germination as rapidly as wild-type spores in the presence of 10 mM hydrogen peroxide (data not shown). Hydrogen peroxide at 0.2 mM also had no significant effect on the outgrowth and resumption of vegetative growth of wild-type spores, but 2 and 10 mM hydrogen peroxide slowed the return to vegetative growth slightly and dramatically, respectively (Fig. 7). However, the return to vegetative growth of katX and katA katX spores was tremendously slowed by addition of 2 mM hydrogen peroxide 10 min after the initiation of spore germination, while the effect on spores of a katA mutant was identical to that on wild-type spores (Fig. 8). Addition of 2 mM hydrogen peroxide 10 min after initiation of spore germination also resulted in killing of >90% of katX spores after 2.5 h but had no effect on wild-type spores (data not shown). These effects of the katX mutation seem most likely to be due to loss of katX and not a polar effect on a downstream gene, since it appears that katX is a monocistronic gene (42). Consequently, these data indicate that KatX plays a major role in the resistance of the germinating spore to hydrogen peroxide.

FIG. 7.

Effects of hydrogen peroxide on spore outgrowth. Spores of strain PS832 (wild type) were germinated as described in Materials and Methods, various amounts of hydrogen peroxide were added 10 min after initiating germination (arrow), and the OD₆₀₀ of the culture was measured. Symbols: ○, no hydrogen peroxide; •, plus 0.2 mM hydrogen peroxide; ▵, plus 2 mM hydrogen peroxide; ▴, plus 10 mM hydrogen peroxide.

FIG. 8.

Effects of hydrogen peroxide on outgrowth of spores of catalase mutants. Spores of various strains were germinated as described in Materials and Methods, hydrogen peroxide was added to 2 mM 10 min after initiation of germination (arrow), and the OD₆₀₀ of the culture was determined. Symbols: ○, PS832 (wild type); •, PS2488 (katA); ▵, PS2588 (katX); ▴, PS2664 (katA katX).

DISCUSSION

The data presented in this report indicate that katX is expressed only in the forespore compartment during sporulation and only under the control of ς^F. This latter sigma factor is synthesized at a significant level only during sporulation and is made prior to the asymmetric septation that separates the mother cell and forespore (12). However, ς^F is held in an inactive state by a variety of mechanisms until after asymmetric septation, when it becomes active only in the forespore (1, 23, 29). Until now, the genes primarily or absolutely dependent on ς^F whose function is known included only spoIIR (16, 20) and spoIIO (19); katX is clearly an addition to this group. There are also a number of other genes which are dependent on ς^F whose function is not yet known (7), as well as genes which are transcribed by both Eς^F and Eς^G (13, 15, 26, 33, 38, 41). Indeed, the promoter specificity of Eς^F is not completely distinct from that of Eς^G, as the latter enzyme initiates transcription of several genes at the same nucleotide as does Eς^F (38, 39, 41). Comparison of the sequences upstream of the known promoters of genes transcribed well by Eς^F reveals significant homology in the −10 and −35 regions, allowing assignment of consensus sequences for these regions (Fig. 9) (15, 33, 39); all of these genes match the consensus −10 and −35 sequences in at least three positions (Fig. 9). The consensus −10 and −35 sequences for ς^G are very similar to those recognized by ς^F, but a feature which appears to distinguish promoters recognized only by ς^G from those recognized by ς^F (and sometimes ς^G) is the presence of two G residues just upstream of the −10 region of ς^F-dependent promoters (15, 39), with the spoIIR gene being the only exception (Fig. 9). Indeed, introduction of two G residues in these positions converts a ς^G promoter to a ς^F promoter (39). Comparison of the region upstream of the katX transcription start site with those of other genes recognized well by Eς^F (Fig. 9) reveals significant homology in the −10 and −35 regions and also the presence of two G residues just upstream of the −10 region. This finding provides further evidence for the importance of these residues in recognition by Eς^F.

FIG. 9.

Comparison of promoter regions of ς^F-dependent genes. Promoter sequences of ς^F genes were taken from primer extension analyses reported in references 13, 15, 16, 33, 39, and 41 and this work. The −10 and −35 regions are labeled and underlined; the G residues upstream of the −10 region are in boldface, as are the transcription initiating nucleotides. The consensus sequence shown is from the seven gene sequences shown. Positions with single residues in the consensus have these residues in at least five genes; where there are two residues shown, each is present in at least two genes.

While dormant spores lack KatA, the latter enzyme is present at the beginning of sporulation and can be present in sporulating cells at the time of DPA production (3, 5, 18). Since KatA is not in dormant spores, this enzyme is presumably not present in forespores at h 4 to 5 of sporulation and must therefore be confined to the mother cell compartment. However, the presence of KatA at the initiation of sporulation prior to asymmetric septation indicates that this enzyme must initially be present in the forespore compartment. Consequently, KatA must be lost from the forespore, presumably by proteolysis, as the forespore develops. In this regard, KatA is similar to a number of other enzymes (e.g., several enzymes of the tricarboxylic acid cycle) that are present prior to asymmetric septation but not found in the mature spore (34). The proteolytic system for removal of these enzymes from the developing spore has not been identified but seems likely not to act continuously during forespore development, as a number of extremely protease-sensitive proteins (i.e., small, acid-soluble spore proteins [31]) are accumulated by the developing spore several hours after asymmetric septation.

The fact that katX is under control of ς^F rather than ς^G indicates that KatX is synthesized earlier than most forespore-specific gene products. In this regard it is somewhat surprising that a katX mutation, which had a dramatic effect on the hydrogen peroxide sensitivity of a germinated spore, had no noticeable effect on the hydrogen peroxide resistance of the sporulating cell. There may be several explanations for this finding. First, KatA is present in the mother cell at the beginning of sporulation and clearly can provide the sporulating cell with protection against hydrogen peroxide. In the absence of KatA, the KatX in the developing spore may well protect the forespore from hydrogen peroxide, but the mother cell compartment may become so badly damaged that sporulation is blocked. Second, significant KatA may remain in the forespore compartment until late in forespore development and may work with KatX to provide hydrogen peroxide resistance to the developing forespore. Finally, the forespore is surrounded by mother cell cytoplasm, and this in itself may detoxify significant hydrogen peroxide. The dramatic changes in forespore permeability which take place late in sporulation (11) presumably also provide the forespore some hydrogen peroxide protection. It is, of course, also possible that KatX does play some subtle role in forespore hydrogen peroxide resistance that we have been unable to discern.

Where it has been examined, all genes recognized by Eς^F are transcribed only in the forespore, and at least in some cases the gene products are found within the dormant spore; this is also the case with KatX. Given that KatX is the only catalase detectable in the dormant spore, it is perhaps not surprising that katX mutants are hydrogen peroxide sensitive during spore germination. KatA, the major catalase in growing cells of B. subtilis, is not present in spores, and as shown here, katA is not transcribed until at least 20 min after the initiation of spore germination. Thus, KatX appears to be the only catalase that can detoxify hydrogen peroxide early in spore germination. Presumably, after synthesis of KatA, outgrowing spores of a katX mutant will become hydrogen peroxide resistant, but we have not yet tested this presumption directly.