A Region in the Bacillus subtilis Transcription Factor Spo0A That Is Important for spoIIG Promoter Activation

Abstract

Spo0A is a DNA binding protein in Bacillus subtilis required for the activation of spoIIG and other promoters at the onset of endospore formation. Activation of some of these promoters may involve interaction of Spo0A and the ς^A subunit of RNA polymerase. Previous studies identified two single-amino-acid substitutions in ς^A, K356E and H359R, that specifically impaired Spo0A-dependent transcription in vivo. Here we report the identification of an amino acid substitution in Spo0A (S231F) that suppressed the sporulation deficiency due to the H359R substitution in ς^A. We also found that the S231F substitution partially restored use of the spoIIG promoter by the ς^A H359R RNA polymerase in vitro. Alanine substitutions in the 231 region of Spo0A revealed an additional amino acid residue important for spoIIG promoter activation, I229. This amino acid substitution in Spo0A did not affect repression of abrB transcription, indicating that the alanine-substituted Spo0A was not defective in DNA binding. Moreover, the alanine-substituted Spo0A protein activated the spoIIA promoter; therefore, this region of Spo0A is probably not required for Spo0A-dependent, ς^H-directed transcription. These and other results suggest that the region of Spo0A near position 229 is involved in ς^A-dependent promoter activation.

A growing body of evidence supports the model that promoter activation in bacteria involves direct contacts between RNA polymerase and transcriptional activators. Several sites on RNA polymerase, including its alpha, sigma, and β′ subunits, appear to be the targets for interaction with transcriptional activators (reviewed in reference 5). For example, FNR, PhoB, MalT, and cI from phage lambda bind to sites that overlap the −35 regions of promoters and probably contact the ς subunit of RNA polymerase (6, 13, 14, 18, 18a, 19). Spo0A from Bacillus subtilis is a DNA binding protein required for the activation of some promoters at the onset of sporulation (e.g., spoIIG, spoIIE, and spoIIA) (17, 23, 27, 29, 30, 32). Spo0A activates at least two types of promoters, those used by RNA polymerase containing ς^A, the primary ς factor in B. subtilis, and promoters used by RNA polymerase containing ς^H, a secondary sigma factor. The phosphorylated, active form of Spo0A binds to the spoIIG and spoIIE promoters at sites that overlap the −35 regions of these promoters and stimulates utilization of the promoters by RNA polymerase containing ς^A (2, 3, 16, 22, 23). Activation of these promoters may involve interaction of Spo0A with ς^A. Baldus et al. (1) found that spoIIG and spoIIE promoter activities were reduced in mutants of B. subtilis in which ς^A contained one of two single-amino-acid substitutions, replacement of lysine at position 356 by glutamate (K356E) or replacement of histidine at position 359 by arginine (H359R). However, these substitutions did not affect the utilization of Spo0A-independent promoters or of promoters used by RNA polymerase containing the secondary sigma factor, ς^H. Moreover, alanine substitutions at positions 356 and 359 in ς^A had similar effects (25). These observations led to the hypothesis that spoIIG and spoIIE promoter activation by Spo0A required the region near positions 356 to 359 of ς^A, possibly because Spo0A interacts with this region of ς^A at these promoters. This model predicts that a surface of Spo0A interacts with ς^A and that this interaction is prevented by amino acid substitutions at position 359 or 356 in ς^A. Here we report the identification of a single-amino-acid substitution in Spo0A (a substitution of phenylalanine for serine at position 231) that partially suppresses the effect of the H359R substitution in ς^A. To test the hypothesis that position 231 in Spo0A lies near a region of ς^A that is required for activation of ς^A-dependent promoters such as spoIIG, we examined the effects of single alanine substitutions at this and neighboring positions in Spo0A. The results support the hypothesis that this region of Spo0A is required for activation of ς^A-dependent, Spo0A-dependent promoters but not for Spo0A-dependent repression of abrB promoter activity or for Spo0A-dependent activation of the ς^H-dependent promoter spoIIA. Complementary studies by Hatt and Youngman (11) reported in an accompanying article support this hypothesis.

MATERIALS AND METHODS

Plasmids.

pCB2 was constructed to be a vector in which wild-type spo0A could be cloned and then introduced into the B. subtilis chromosome. pCB2 is a derivative of pAH256 (12) which contains 800 bp of spo0A, a spectinomycin resistance marker, and 700 bp of a region found downstream from the spo0A gene in the B. subtilis chromosome. The Spo0A gene was PCR amplified from the chromosome with the forward primer 0AUS (5′-AAGCAAGCTTACTGCCGGAGTTTCCGGA-3′) and the reverse primer 0ADS (5′-TCTAGGGTTGATCATGCTTCGTGATCC-3′), digested with BclI and HindIII, and cloned into the BamHI and HindIII sites of pAH256, creating pCB1. The downstream sequences of Spo0A were then amplified from the chromosome with the primers DS0AFOR (5′-TATAGGATCTCGAGGCATGATCGACCC-3′) and DS0AREV (5′-GTGCATTACTAGTCGGCTACCGCCTGTC-3′), digested with XhoI and SpeI, and cloned into the XhoI and SpeI sites of pCB1, creating pCB2wtspo0A. pCB3 was constructed in order to clone mutant alleles of spo0A and use them to replace the chromosomal allele. The downstream sequences of Spo0A were amplified and digested as described above and were cloned into the XhoI and SpeI sites of pAH256. The pCB2spo0A site-directed mutants I229A, D230A, S231A, S231F, I232A, and S233A and the Spo0A nonsense allele were constructed by digesting the PCR-mutagenized spo0A fragments with BclI and HindIII and cloning them between the BamHI and HindIII sites in pCB3. pGS2 was a Spo0A-expressing plasmid constructed from the N-terminal histidine tag carrying plasmid pET-16b (Novagen). The DNA encoding the carboxy-terminal amino acids of Spo0A (carboxy-terminal domain [CTD]) (see Fig. 1) was first amplified with the oligonucleotides 5′-AATCTCATATGGCCAGCAGTGTGACGC-3′ and 5′-GGCAAGCTTCCACTTAATAAGCTCAT-3′, used as forward and reverse primers, respectively, and was then inserted in frame with the His tag coding sequence between the NdeI and BamHI sites of pET-16b. The S231F mutant form of the CTD Spo0A was obtained by QuikChange site-directed mutagenesis (Stratagene) performed directly on the pGS2 plasmid.

FIG. 1.

Domain structure of Spo0A. Spo0A consists of two domains, an N-terminal domain, which undergoes specific aspartate phosphorylation at D56, indicated by the arrow, and a CTD, which is responsible for DNA binding. The position number for the identifying amino acid of each domain is indicated. A magnification of the 5-amino-acid region of Spo0A discussed in this study is shown, and the amino acids are numbered. The S231F mutation was found to suppress the sporulation defect of the ς^A mutation H359R.

Site-directed PCR mutagenesis.

Site-specific mutations were made in spo0A by a multiple-step PCR procedure (7). The first step was amplification with the primer 0AUS and a reverse primer that overlapped the region to be mutagenized and contained the appropriate base substitutions (I229AREV, etc.). The second step was amplification with the 0ADS primer and a forward primer that overlapped the reverse primer from the first step and also contained the appropriate base substitutions (I229AFOR, etc.). To construct the spo0A nonsense allele, spo0A195, mutagenic forward and reverse primers, 0ASTOP-FOR (5′-GGCAGCATTACATAAGTCCTCTAGCCGGACATCGCC-3′) and 0ASTOP-REV (5′-GGCGATGTCCGGCTAGAGGACTTATGTAATGCTGCC-3′), which mutated two codons at the end of the putative helix-turn-helix motif, at amino acids 195 and 198, to stop codons, were used. The PCR products from both steps were then used in a reaction with the two outside primers, 0AUS and 0ADS, so that the entire region was amplified, including the base substitutions. The presence of the mutation was confirmed by sequencing the spo0A allele in each plasmid with a Sequenase kit from Amersham. The DNA polymerase used in all of the above cloning reactions was the high-fidelity Pfu enzyme from Stratagene.

EMS mutagenesis.

The ethyl methane sulfonate (EMS) mutagenesis procedure was adapted from the procedure described by Green et al. (9). B. subtilis EUC9720 (Table 1) was grown in 50 ml of Luria broth (LB) (20) with 5 μg of kanamycin/ml and 50 μg of spectinomycin/ml until an optical density at 600 nm (OD₆₀₀) of 0.6 was reached. The cells were then spun down and resuspended in 1 ml of LB, and various dilutions were plated onto DSM agar (24) and allowed to dry. A paper disk to which 3 drops of EMS (1.7 g/ml; Sigma) were added was placed in the center of the plate, and the cells were incubated at 42°C for 2 days. At that time the plates were inverted over 400-μl pools of chloroform for 20 min, removed from the chloroform exposure for 20 min, and returned to grow at 42°C overnight. Sporulation-proficient (Spo⁺) survivors, which were able to form colonies by using nutrients released from lysed, nonsporulating cells, were then scraped off the plates, pooled, and used to inoculate 10 ml of LB, which was then grown to an OD₆₀₀ of 0.8, and chromosomal DNA was extracted by using the Quick Procedure (8). The chromosomal DNA was used to transform strain EUB9403 to spectinomycin resistance, and bacteria were plated on DSM agar to screen for a Spo⁺ phenotype. Twelve transformants that appeared Spo⁺ on the plates were picked and used to isolate chromosomal DNA. This DNA was used to transform EUB9403 to spectinomycin resistance in order to identify chromosomal DNA in which the spectinomycin resistance marker was >50% linked to a mutation that appeared to suppress the sporulation defect of EUB9403. Ten of the chromosomal DNAs produced >90% Spo⁺ transformants of EUB9403. We determined the nucleotide sequence of the spo0A region and found a single-base-pair substitution, a transition, that changed codon 231 (TCC), encoding serine, to TTC, which encodes phenylalanine. This allele was reconstructed in vitro by site-directed PCR mutagenesis, cloned into pCB3, and used to replace the wild-type allele of spo0A in the chromosomes of B. subtilis strains that express wild-type or mutant alleles of sigA (1, 25). The presence of the mutation was confirmed by cycle sequencing spo0A PCR products from the chromosome by using the fmol kit from Promega.

TABLE 1.

Bacterial strains and bacteriophages

Strain or phage	Derivation or genotype	Source or reference
Strains
JH642	trpC2 phe-1	J. A. Hoch
EUB9401	JH642 transformed with pJB2wtsigA	1
EUB9402	JH642 transformed with pJB2K356EsigA	1
EUB9403	JH642 transformed with pJB2H359RsigA	1
EUC9603	JH642 transformed with pJB2H359AsigA	25
EUC9762	EUB9401 transformed with pCB2wtspo0A	This work
EUC9766	EUB9401 transformed with pCB2spo0AS231F	This work
EUC9708	EUC9402 transformed with pCB2wtspo0A	This work
EUC9795	EUC9402 transformed with pCB2spo0AS231F	This work
EUC9720	EUB9403 transformed with pCB2wtspo0A	This work
EUC9722	EUB9403 transformed with pCB2spo0AS231F	This work
EUC9723	EUC9603 transformed with pCB2wtspo0A	This work
EUC9725	EUC9603 transformed with pCB2spo0AS231F	This work
EUC9763	EUB9401 transformed with pCB2spo0AI229A	This work
EUC9764	EUB9401 transformed with pCB2spo0AD230A	This work
EUC9765	EUB9401 transformed with pCB2spo0AS231A	This work
EUC9767	EUB9401 transformed with pCB2spo0AI232A	This work
EUC9768	EUB9401 transformed with pCB2spo0AS233A	This work
EUC9790	EUB9401 transformed with pCB2spo0A195	This work
Bacteriophages
SPβ spoIIG-lacZ		22
SPβ spoIIA-lacZ		31
SPβ abrB-lacZ		33

Sporulation assay.

The B. subtilis strains were grown in DSM liquid containing 5 μg of kanamycin/ml and 50 μg of spectinomycin/ml for 24 h. Aliquots from each culture (1 ml) were heated for 10 min at 80°C. The numbers of CFU in the heated samples were determined by plating 20 μl of undiluted cells and cells diluted by 10⁻², 10⁻⁴, and 10⁻⁶ onto LB agar containing 5 μg of kanamycin/ml and 50 μg of spectinomycin/ml.

RNA polymerases and CTD Spo0A purifications.

ς^A holoenzymes, wild-type Eς^A and Eς^A H359R, were isolated by a procedure consisting of a low-pressure affinity chromatography on heparin followed by anion-exchange fast protein liquid chromatography as described previously (28). The wild-type form of CTD Spo0A was prepared from a derivative of Escherichia coli BL21 (DE3) pLysS containing the pGS2 plasmid. BL21 cells (2L), grown at 37°C, were induced for 3 h to overproduce CTD Spo0A proteins (the wild-type and S231F mutant forms) by addition of 1 mM isopropyl-β-d-thiogalactopyranoside (IPTG) at an OD₆₀₀ of 0.6. The harvested cells were resuspended in 10 ml of 10 mM Tris HCl (pH 8.0)–0.1 M NaCl–5% glycerol–1 mM EDTA–1 mM β-mercaptoethanol–1 mM phenylmethylsulfonyl fluoride (PMSF) (buffer 1) containing 2.5 mM imidazole, then disrupted at 4°C by a single passage through a French pressure cell at 70,000 kPa, and finally centrifuged for 30 min at 10,000 rpm in an SS34 rotor. Proteins from the supernatant were then adsorbed at 4°C to 4 ml of a Ni²⁺-nitrilotriacetic acid matrix (Qiagen) previously equilibrated with buffer 1. After an hour, the matrix was packed into a disposable column and washed with 15 ml of buffer 1 containing 20 mM imidazole. The His-tagged proteins were eluted with 10 ml of buffer 1 with 300 mM imidazole. CTD Spo0A was then purified further through gel filtration chromatography on a Superdex 200 16/60 HR column (Pharmacia) equilibrated with a solution containing 10 mM Tris HCl (pH 8.0), 50 mM NaCl, 1 mM EDTA, 1 mM dithiothreitol (DTT), 10 mM MgCl₂, and 1 mM PMSF.

In vitro transcription.

Plasmid pJH101IIG-38C (22) was cut with BamHI and used as a template in in vitro transcription reactions. RNA polymerase (0.04 μM), CTD Spo0A (concentration, 250 or 500 nM), and the plasmid (5 nM) were preincubated at 37°C for 10 min in 50 μl (final volume) of a 33 mM Tris acetate (pH 7.9)–10 mM magnesium acetate–0.5 mM DTT–0.15 mg of bovine serum albumin/ml–66 mM potassium acetate buffer. Ribonucleotides (500 μM [each] ATP, GTP, and CTP [final concentration; Boehringer Mannheim] and 10 μCi of [α-³²P]UTP [800 Ci/mmol; Amersham]) were added for 1 min, before reinitiation was stopped by addition of 10 μg of heparin (Sigma). Five minutes later, unlabelled UTP (Boehringer Mannheim) (final concentration, 500 μM) was added for a further 5 min before addition of sodium acetate (final concentration, 0.3 M) and ethanol precipitation of the nucleic acids. Before being loaded on a 5% polyacrylamide–7 M urea sequencing gel, nucleic acids were resuspended in 10 μl of a sequencing formamide dye and heated at 95°C for 3 min. Transcripts were localized by autoradiography, then quantified by densitometry with ImageQuant software coupled to a PhosphorImager 445 SI (Molecular Dynamics). Specific transcription from the spoIIG promoter produced a 135-nucleotide transcript.

RESULTS

A mutant spo0A allele suppresses the sporulation-defective phenotype caused by sigA H359R.

We took advantage of the sporulation-defective phenotype caused by the H359R substitution in ς^A to seek suppressor mutations in Spo0A. We used localized EMS mutagenesis of strain EUC9720, which contains a spectinomycin resistance determinant linked to spo0A and the H359R allele of sigA, and isolated a sporulation-proficient survivor of chloroform treatment. The DNA sequence of the Spo0A region in this mutant revealed a single-base-pair substitution that resulted in a substitution of phenylalanine for serine at position 231 (S231F) in the carboxy terminus of Spo0A (Fig. 1). We examined the effects of the S231F substitution on the efficiency of endospore formation in strains containing the wild-type or mutant alleles of sigA (Fig. 2). Strain EUC9720 (sigAH359R) formed 2.5 × 10² heat-resistant spores per ml, whereas EUC9722 (sigAH359R spo0AS231F) formed 3.5 × 10⁶ spores per ml. The S231F allele of spo0A also partially suppressed the sporulation defects caused by sigA alleles H359A and K356E (Fig. 2). The S231F allele slightly reduced the sporulation efficiency of an otherwise isogenic strain containing a wild-type sigA allele (Fig. 2).

FIG. 2.

Spo0A S231F suppresses the sporulation phenotypes of sigA mutants. Shown are the numbers of heat-resistant spores (expressed as logs of the numbers shown along the y axis) produced by strains containing wild-type (wt) or mutant sigA alleles (shown along the x axis) in combination with wild-type or mutant (S231F) spo0A alleles (solid or shaded bars, respectively).

Spo0A S231F stimulates transcription of the spoIIG promoter in vitro by RNA polymerase containing ς^A H359R.

We wanted to determine whether the partial suppression of the sporulation-defective phenotype by Spo0A S231F was due to the partial restoration of Spo0A-dependent stimulation of ς^A-directed transcription and not to activation of transcription by another, possibly unknown, form of RNA polymerase. Therefore, we examined the effect of Spo0A S231F in an in vitro transcription assay using purified components (Fig. 3). In confirmation of our previous results (25), we found that the CTD of Spo0A stimulated in vitro transcription from the spoIIG promoter by RNA polymerase containing wild-type ς^A (Fig. 3, lanes 2 and 3). However, the CTD of Spo0A was unable to stimulate transcription of spoIIG by RNA polymerase containing ς^A H359R (Fig. 3, lanes 8 and 9). In contrast, the CTD of Spo0A containing the S231F substitution efficiently stimulated transcription of spoIIG by RNA polymerase containing ς^A H359R (Fig. 3, lanes 11 and 12). The S231F-substituted Spo0A CTD also stimulated transcription by wild-type ς^A RNA polymerase.

FIG. 3.

Spo0A S231F partially restores spoIIG promoter activation in vitro in the presence of RNA polymerase containing ς^A H359R. In vitro transcription experiments were performed at the spoIIG promoter by using B. subtilis RNA polymerase containing either wild-type (wt) ς^A (lanes 1 to 6) or ς^A H359R (lanes 7 to 12) in the presence of the wild-type Spo0A CTD (lanes 1 to 3 and 7 to 9) or the Spo0A S231F CTD (lanes 4 to 6 and 10 to 12). Increasing concentrations of the Spo0A CTD were used for activation: none (lanes 1, 4, 7, and 10), 250 nM (lanes 2, 5, 8, and 11), and 500 nM (lanes 3, 6, 9, and 12). The arrow indicates transcripts from the spoIIG promoter.

An alanine substitution at position 229 in Spo0A reduces spoIIG transcription.

The S231F-substituted form of Spo0A stimulated transcription of spoIIG in vitro by wild-type ς^A polymerase (Fig. 3, lanes 5 and 6) and supported efficient sporulation in a strain containing wild-type sigA (Fig. 2). Since the S231F substitution in Spo0A did not prevent transcription by wild-type ς^A polymerase, we could not conclude whether S231 or the region in Spo0A near position 231 is required for Spo0A-dependent activation of ς^A-dependent promoters in wild-type cells. To examine the importance of the amino acids of Spo0A at or near position 231 in ς^A-dependent spoIIG promoter activation, we made single-amino-acid substitutions at five positions, changing the serine at 231 and the two adjoining amino acids in each direction, upstream and downstream, to alanines (Fig. 1). None of these alanine-encoding spo0A alleles, when used to transform the wild-type ς^A parent strain, EUB9401, caused a severe decrease in spore production (all strains sporulated at or near 10⁸ spores per ml [data not shown]). However, sporulation efficiency is not decreased significantly unless spoIIG or spoIIE transcription is reduced to less than 10% of the level in wild-type cells (10, 15, 25). Therefore, to monitor the effects of each amino acid substitution on the activation of specific promoters, the strains expressing the alanine-substituted forms of Spo0A, the isogenic wild-type parent strain, and the isogenic nonsense mutant Spo0A strain (spo0A195) were lysogenized with specialized SPβ transducing phages that carried lacZ transcription fusions to one of three Spo0A-regulated promoters, spoIIG, spoIIA, and abrB (Table 2 and Fig. 4). The I229A form of Spo0A caused the most dramatic decrease in activity from the Spo0A-dependent, ς^A-dependent promoter spoIIG, reducing its activity to about 35% of the wild-type levels (Fig. 4a). The Spo0A mutants D230A and I232A also exhibited somewhat decreased spoIIG activity, about 50% of wild-type activity (Table 2).

TABLE 2.

Effects of amino acid substitutions in Spo0A on spoIIG, spoIIA, and abrB promoter activities^a

Strain (spo0A genotype)	β-Galactosidase activity (Miller units) from:
	spoIIG-lacZ		spoIIA-lacZ		abrB-lacZ
	T2	T3	T2	T3	T2	T3
EUC9762 (wt)	19	17	172	169	4	3
EUC9763 (I229A)	3	6	174	230	4	2
EUC9764 (D230A)	7	12	242	348	7	4
EUC9765 (S231A)	34	53	150	179	6	4
EUC9766 (S231F)	34	33	208	375	5	3
EUC9767 (I232A)	6	12	169	174	11	5
EUC9768 (S233A)	13	14	145	133	8	6
EUC9790 (195)	1	2	4	7	142	147

^aStrains were transduced with Spβ phages carrying the indicated promoter-lacZ fusions. Transductants were grown in DS medium, and samples were harvested at hourly intervals. β-Galactosidase activities are reported for samples harvested 2 (T₂) and 3 (T₃) h after the end of the exponentional-growth phase. 

FIG. 4.

Effects of amino acid substitutions in Spo0A on spoIIG, spoIIA, and abrB promoter activities. B. subtilis EUC9762 (wild-type spo0A) (circles), EUC9763 (spo0AI229A) (squares), and EUC9790 (spo0A195) (triangles) containing the spoIIG (a), spoIIA (b), or abrB (c) promoter-lacZ fusion were grown in DSM medium. Samples were taken from cultures growing at mid-log phase (T₋₁), at the end of exponential growth (T₀), and at 1-h intervals after the onset of stationary phase (T₁ to T₄) and were then assayed for β-galactosidase activity (22). Independent transductants of each of the above strains were assayed for β-galactosidase activity and were found to show essentially the same results (data not shown).

The abrB promoter is repressed at the onset of sporulation by Spo0A (26). The I229A amino acid substitution in Spo0A had no effect on the repression of the abrB promoter; however, the strain containing spo0A195 exhibited no repression of this promoter, and abrB activity remained at high levels throughout sporulation (Fig. 4c). We also assayed the activity of the Spo0A-dependent, ς^H-dependent promoter spoIIA (Fig. 4b). We found expression from the spoIIA promoter in the strain containing the alanine substitution at I229 was essentially the same as that measured in wild-type cells, whereas the strain containing the spo0A195 allele retained only 5% of wild-type spoIIA activity (Fig. 4b). These results suggest that I229 is important for wild-type levels of Spo0A-dependent, ς^A-dependent spoIIG promoter activity. It seems unlikely that the substitution in Spo0A has long-range effects on the overall folding of the protein or effects on its ability to bind DNA, since the repression of the Spo0A-dependent, ς^A-dependent promoter abrB is unaffected. Moreover, this amino acyl residue is not required for activity of the Spo0A-dependent, ς^H-dependent promoter spoIIA.

DISCUSSION

Here we report the identification of an amino acid substitution (S231F) in Spo0A that partially suppresses the phenotype of the H359R substitution in ς^A. The finding that a mutation in Spo0A can suppress the sporulation defect caused by the H359R substitution in ς^A adds support to our hypothesis that the H359R substitution in ς^A prevents its interaction with Spo0A. The S231F substitution in Spo0A restores Spo0A-dependent stimulation of ς^A H359R RNA polymerase transcription. However, it seems unlikely that this involves a direct interaction of the arginine residue at position 359 in the mutant ς^A and the phenylalanine residue at position 231 in the mutant Spo0A, because the S231F substitution also partially suppressed the effects of the H359A and K356E substitutions in ς^A, and the S231F Spo0A efficiently stimulated wild-type ς^A polymerase. It seems more likely that the S231F substitution in Spo0A establishes a new interaction with ς^A RNA polymerase.

This model, in which the S231F substitution in Spo0A establishes a new interaction with RNA polymerase, predicts that position 231 of Spo0A is likely to be located near ς^A RNA polymerase when Spo0A and ς^A RNA polymerase are bound to a promoter. Therefore, we tested whether the amino acyl residues in the 231 region of Spo0A were required for activation of Spo0A-dependent, ς^A-dependent transcription by examining the effects of single alanine substitutions in this region. We found that an alanine substitution at position 229 in Spo0A reduced spoIIG transcription. The I229A-substituted Spo0A repressed abrB transcription; therefore, its ability to bind DNA is probably not affected. spoIIA promoter activity requires phosphorylated Spo0A and ς^H RNA polymerase. The I229A-substituted Spo0A is probably phosphorylated, since it was able to activate spoIIA promoter activity. Evidently the I229A substitution in Spo0A specifically reduces Spo0A-dependent activation of ς^A-dependent transcription. It seems likely that the amino acid side chain of I229 is involved in interaction of Spo0A and ς^A RNA polymerase, although it is not known if this involves a direct contact. These results suggest that this region of Spo0A is required specifically for interaction with ς^A RNA polymerase. This interpretation is consistent with the results of Hatt and Youngman (11), in which they identified other amino acid substitutions in this region of Spo0A (at positions 227, 233, 236, and 240) that specifically prevented activation of ς^A-dependent promoters. Although it is not known if this activation region (AR) of Spo0A interacts directly with ς^A RNA polymerase, it is attractive to speculate that this AR of Spo0A may interact with ς^A, near positions 356 to 359 of ς^A, since this region of ς^A is required for Spo0A-dependent activation (1, 25). However, it remains a possibility that this AR of Spo0A interacts with another region of ς^A or even another subunit of ς^A RNA polymerase.

Spo0A is required for use of the spoIIA promoter by ς^H RNA polymerase. The 229 region of Spo0A does not appear to be directly required for this activity, since the alanine substitutions at positions 229 to 233 did not reduce spoIIA transcription. A different region of Spo0A may interact with ς^H RNA polymerase to stimulate its activity. An amino acid substitution in Spo0A (A257T) has been found to prevent spoIIA transcription but not abrB repression (21). The A257T substitution may define a region of Spo0A that is required for activation of ς^H-dependent promoters. It is not known whether this region of Spo0A is also required for activation of ς^A-dependent transcription. In other work (4), we have recently identified amino acid substitutions in ς^H that have effects analogous to those of the H359R substitution in ς^A. RNA polymerases containing the mutant ς^H proteins do not use the Spo0A-dependent promoter spoIIA but are able to use the Spo0A-independent spoVG and citGp2 promoters. The sporulation defects caused by these amino acid changes in ς^H are not suppressed by the S231F substitution in Spo0A (unpublished data). These results are consistent with the hypothesis that different amino acids in Spo0A are involved in the activation of ς^A and ς^H RNA polymerases. The versatility of Spo0A in activating transcription by two forms of RNA polymerase and the finding that a single-amino-acid substitution in Spo0A (S231F) apparently can provide a new productive interaction with ς^A RNA polymerase suggest that a wide variety of interactions between DNA binding proteins and RNA polymerase can be used to activate transcription.