γ-Butyrolactone-Dependent Expression of the Streptomyces Antibiotic Regulatory Protein Gene srrY Plays a Central Role in the Regulatory Cascade Leading to Lankacidin and Lankamycin Production in Streptomyces rochei

Abstract

Our previous studies revealed that the srrX and srrA genes carried on the large linear plasmid pSLA2-L constitute a γ-butyrolactone-receptor system in Streptomyces rochei. Extensive transcriptional analysis has now showed that the Streptomyces antibiotic regulatory protein gene srrY, which is also carried on pSLA2-L, is a target of the receptor/repressor SrrA and plays a central role in lankacidin and lankamycin production. The srrY gene was expressed in a growth-dependent manner, slightly preceding antibiotic production. The expression of srrY was undetectable in the srrX mutant but was restored in the srrX srrA double mutant. In addition, SrrA was bound specifically to the promoter region of srrY, and this binding was prevented by the addition of the S. rochei γ-butyrolactone fraction, while the W119A mutant receptor SrrA was kept bound even in the presence of S. rochei γ-butyrolactone. Furthermore, the introduction of an intact srrY gene under the control of a foreign promoter into the srrX or srrA(W119A) mutant restored antibiotic production. All of these results confirmed the signaling pathway from srrX through srrA to srrY, leading to lankacidin and lankamycin production.

Streptomycetes produce many secondary metabolites, including antibiotics, and also undergo a complex process of morphological development. To regulate these complex cellular processes, the bacteria have evolved an intricate hierarchic regulatory system. γ-Butyrolactone, a diffusible signaling molecule, often governs antibiotic production and/or morphogenesis in streptomycetes (3, 30). Pioneering studies to understand the γ-butyrolactone signaling system have been carried out with Streptomyces griseus, a streptomycin producer. In this bacterium, the γ-butyrolactone known as A-factor (2-isocapryloyl-3R-hydroxymethyl-γ-butyrolactone) (6, 12) is synthesized by the afsA gene product (7, 10). In the absence of A-factor, the A-factor receptor protein, ArpA (23), binds to the promoter region of the global transcriptional activator gene, adpA, and represses its transcription (22). When A-factor reaches a critical concentration, it binds to ArpA, dissociates ArpA from the promoter of adpA, and thereby relieves the repression of the gene, which in turn triggers streptomycin production and morphological development. ArpA belongs to the TetR family of receptor proteins and contains a helix-turn-helix motif in its N-terminal region and a tryptophan residue at the 119 position. The former has DNA-binding activity, while the latter is necessary for A-factor binding (27). The ArpA-binding site was found within the −35 and −10 regions of the adpA promoter and contains a 22-bp palindromic sequence, suggesting that ArpA represses adpA transcription by preventing the binding of RNA polymerase (22, 23). Similar regulatory systems comprising a γ-butyrolactone and its cognate receptor protein have been found in other streptomycetes, for example, the barX-barA system for virginiamycin production in Streptomyces virginiae (11), farX-farA for showdomycin and minimycin in Streptomyces lavendulae (17), and scbA-scbR for actinorhodin and undecylprodigiosin in Streptomyces coelicolor A3(2) (31). However, their effects on antibiotic production and morphogenesis are different from species to species.

The γ-butyrolactone-receptor systems regulate many regulatory genes, including the Streptomyces antibiotic regulatory protein (SARP) family genes (33). Members of this family are characterized by the presence of an OmpR-like DNA-binding domain (19) and regulate the biosynthesis of many antibiotics. They are exemplified by actII-orf4 for actinorhodin (2), dnrI for daunorubicin (18), papR1 for pristinamycin (5), tylS for tylosin (4, 25), and kasO for a hypothetical type I polyketide (32).

Streptomyces rochei strain 7434AN4 has three linear plasmids (pSLA2-L, pSLA2-M, and pSLA2-S) and produces two different polyketide antibiotics, lankacidin (LC) and lankamycin (LM) (14, 15). We have shown previously that the biosynthetic gene clusters lkc (orf4 to orf18) for LC and lkm (orf24 to orf53) for LM are located on the largest plasmid, pSLA2-L (210,614 bp) (20, 29). This plasmid contains two additional biosynthetic gene clusters, one for a hypothetical type II polyketide (roc; orf62 to orf70) and one for carotenoid (crt; orf104 to orf110). Furthermore, numerous regulatory genes have been found on pSLA2-L, including homologues of the A-factor regulatory genes in S. griseus. They are the γ-butyrolactone biosynthetic gene srrX (orf85), similar to afsA; six tetR family receptor genes, srrA (orf82), srrB (orf79), srrC (orf74), orf92, orf99, and orf126; three SARP genes, srrY (orf75), srrW (orf55), and srrZ (orf71); and two transcriptional activator genes, orf116 and orf3, similar to adpA and strR, respectively. Thus, these genes may form a more complex regulatory cascade than those found in other streptomycetes.

Our previous studies revealed that srrX and srrA constitute a γ-butyrolactone-receptor system in S. rochei (1, 20). srrX has a positive effect on antibiotic production and a negative effect on spore formation, whereas srrA reverses both effects of srrX. The exogenous addition of the culture extract of the parent strain 51252 to the srrX mutant restores the production of both LC and LM, suggesting that the γ-butyrolactone is enough for restoration without the SrrX protein. Until now, the S. rochei γ-butyrolactone has not been isolated. In addition to srrA, the srrB gene has a negative effect on LC and LM production, whereas srrC has a positive effect on spore formation. Furthermore, the SARP gene srrY shows positive effects on both antibiotic production and sporulation (though we have revised our conclusions about the latter effect; see below). Thus, together with several regulatory genes, including srrB, srrC, and srrY, the srrX-srrA γ-butyrolactone-receptor system may regulate antibiotic biosynthesis and sporulation. Therefore, the identification of regulators functioning downstream of the srrX-srrA system is of interest and may facilitate an understanding of the entire picture of the regulatory cascade in S. rochei.

Extensive transcriptional analysis in this study has revealed that srrY is a primary target of the receptor/repressor SrrA and that the signaling pathway from srrX through srrA to srrY plays a central role in the regulation of antibiotic production. Based on all the results, including preliminary ones, we propose a possible regulatory cascade leading to LC and LM production in S. rochei.

MATERIALS AND METHODS

Bacterial strains, plasmids, media, and DNA manipulation.

S. rochei strain 51252 carrying only pSLA2-L (14) was used as the parent strain, and mutant strains were constructed as described below and listed in Table 1. Plasmids used in this study are listed in Table 2, and their construction is described below. S. rochei strains were grown in YM medium (0.4% yeast extract, 1.0% malt extract, and 0.4% glucose, pH 7.3) or tryptic soy broth medium, and Escherichia coli strains were grown in Luria-Bertani medium. Antibiotics were used at the following concentrations: ampicillin, 100 μg/ml; apramycin, 25 μg/ml; chloramphenicol, 10 μg/ml; kanamycin, 10 μg/ml; and thiostrepton, 10 μg/ml. DNA manipulations for Streptomyces (13) and E. coli (24) were carried out according to standard procedures. PCR amplification was carried out with a 2720 thermal cycler (Applied Biosystems) using Thermococcus kodakarensis DNA polymerase (KOD Plus; Toyobo). Nucleotide sequences of DNA primers are given in Table 3.

TABLE 1.

S. rochei strains used in this study

Strain	Description	Production of:		Reference(s)
		LC	LM
51252	pSLA2-L; does not carry pSLA2-M or pSLA2-S	+	+	14
KY85	51252 srrX::Kan	−	−	20
KA12	51252 ΔsrrA	+	+	1
KA21	KY85 ΔsrrA	+	+	1
KU82	51252 srrA(W119A)	−	−	This study
KY75	51252 srrY::Kan	−	−	20; this study
KA61	51252 ΔsrrY	−	−	This study
KY71	51252 srrZ::Kan	+	−	This study
TN01	51252 srrW::Kan	+	+	This study

TABLE 2.

Plasmids used in this study

Plasmid	Description	Source or reference
pUC4-KIXX	pUC derivative vector containing Kan cassette; Aprr	Pharmacia
pUC19	Cloning vector; Ampr
pKY75-1	pUC19 1.8-kb BamHI fragment containing srrY	This study
pKY75-2	pKY75-1 srrY::Kan	This study
pKAR4002	pUC19 9.2-kb PstI fragment containing srrY	This study
pKAR3012	pUC19 srrA	1
pKAU8202	pUC19 srrA(W119A)	This study
pRES18	E. coli-Streptomyces shuttle vector; Ampr Tsrr	8
pKY75-3	pRES18 srrY::Kan	This study
pKAR3055	pRES18 1.5-kb EcoRI-PstI fragment from pKAR3054	This study
pKY71-1	pRES18 2.2-kb SphI fragment containing srrZ	This study
pKY71-2	pKY71-1 srrZ::Kan	This study
pTN03	pRES18 srrW::Kan	This study
pKAU8203	pRES18 srrA(W119A)	This study
Litmus28i	Cloning vector; Ampr	New England Biolabs
pKAR3023	Litmus28i 1·5-kb AgeI fragment containing srrA from pKAR3012	This study
pBluescript SK-plus	Cloning vector; Ampr	Stratagene
pKAR3001	pBluescript SK-plus 1.7-kb Eco47III fragment containing srrW	This study
pRSET-B	Cloning vector; Ampr	Invitrogen
pKAR3053	pRSET-B 1.8-kb EcoRI-PstI fragment from pKY75-1	This study
pKAR3054	pKAR3053 267-bp PvuII fragment deleted from srrY	This study
pTN01	pRSET-B 1.7-kb EcoRI-BamHI fragment containing srrW from pKAR3001	This study
pTN02	pTN01 srrW::Kan	This study
pAlter-1	Cloning vector for site-directed mutagenesis; Tetr	Promega
pKAR3025	pAlter-1 KpnI and HindIII fragment of srrA from pKAR3023	This study
pKAU8201	pAlter-1 srrA(W119A)	This study
pET32b(+)	T7 expression vector for His10 tagging; Ampr	Novagen
pKAR3035	pET32b(+) srrA	This study
pET-Amt	pET32b(+) srrA(W119A)	This study
pIJ8600	Integrative E. coli-Streptomyces shuttle vector for PtipA expression; Aprr Tsrr	28
pKAR3049	pIJ8600 srrY	This study

TABLE 3.

Primers used in this study

Primer name	Sequence (5′ to 3′)
Ampicillin repair
oligonucleotide	GTTGCCATTGCTGCAGGCATCGTGGTG
KAR-75OE01	GCGCATATGGACATCGACGTACTGGGCAC
KAR-75OE03	CCAGCGGATCCTCGCGCAGC
SRRYf2	GGCGTCGTCTGCCTGCTGCC
SRRYr2	ATATCCGCCGGGGGCGGTGG
SRRYr4	GCGCCCGCGGCGTCACCGAGA
KAR8201OE	CTAGGATCCGCATATGGCACAGCAGGAAC
KAR8201SDM	GCCTTCCCCACCGCGATCGCCTTCTCG
KAR8202OE	GAAGAATTCGGCGCGCCGCCCATGAC

Targeted mutagenesis. (i) srrY (orf75) disruption.

A 1.8-kb BamHI fragment containing the srrY gene was cloned into pUC19 (pKY75-1), and a 1.2-kb SmaI fragment of pUC4-KIXX (Pharmacia) bearing a kanamycin resistance cassette was inserted into the NruI site in the center of srrY to obtain pKY75-2. The vector part of this plasmid was replaced by pRES18, an E. coli-Streptomyces shuttle vector (8), to give a targeting plasmid, pKY75-3 (see Fig. S1 in the supplemental material).

(ii) In-frame deletion of srrY.

A 1.8-kb EcoRI-PstI fragment of pKY75-1 was cloned into pRSET-B (Invitrogen) to obtain pKAR3053. A 267-bp PvuII fragment of pKAR3053 corresponding to the central part of srrY was eliminated to give pKAR3054, the vector part of which was replaced by pRES18 to give a targeting plasmid, pKAR3055 (see Fig. S2 in the supplemental material).

(iii) srrZ (orf71) disruption.

A 2.2-kb SphI fragment containing the srrZ gene was cloned into pRES18 to obtain pKY71-1. A 0.4-kb BglII fragment in the middle of srrZ in pKY71-1 was replaced by a 1.6-kb BamHI fragment of pUC4-KIXX to give a targeting plasmid, pKY71-2 (see Fig. S3 in the supplemental material).

(iv) srrW (orf55) disruption.

A 1.7-kb Eco47III fragment containing the srrW gene was cloned into the SmaI site of pBluescript SK-plus to obtain pKAR3001. Its 1.7-kb EcoRI-BamHI fragment was recloned into pRSET-B (pTN01), and then a 1.2-kb SmaI fragment of pUC4-KIXX was inserted into the PvuII site in the middle of srrW to give pTN02. The vector part of pTN02 was replaced by pRES18 to obtain a targeting plasmid, pTN03 (see Fig. S4 in the supplemental material).

(v) Introduction of the W119A point mutation into srrA.

A 1.5-kb AgeI fragment of pKAR3012, which contained the portion of srrA encoding the C-terminal region of the gene product, was inserted into the corresponding site of Litmus28i (New England Biolabs) to obtain pKAR3023. This plasmid was digested with KpnI and HindIII, and the vector part was replaced by pAlter-1 (Promega) to obtain pKAR3025. Site-directed mutagenesis was carried out with the Altered Sites II in vitro mutagenesis system (Promega) using two oligonucleotides, KAR8201SDM and the ampicillin repair oligonucleotide, to obtain pKAU8201. This plasmid carried a gene encoding a mutated SrrA protein with alanine at the 119 position in place of tryptophan (SrrA^W119A). A 1.5-kb AgeI fragment of pKAU8201 was substituted for that of pKAR3012 to obtain pKAU8202. The vector part of pKAU8202 was replaced by pRES18 to obtain a targeting plasmid, pKAU8203 (see Fig. S5 in the supplemental material).

Targeted mutagenesis was performed as follows. The parent strain 51252 was transformed with each of the targeting plasmids constructed as described above, and thiostrepton-resistant transformants were obtained. Among these transformants, single-crossover strains (those with integrated plasmids) were selected by Southern hybridization analysis. Selected colonies were serially grown in liquid YEME medium (13) containing kanamycin to facilitate a second crossover. Finally, kanamycin-resistant and thiostrepton-sensitive colonies were selected as double-crossover mutants. When the targeting plasmid did not contain a kanamycin cassette, the second crossover was induced by serial culturing in YEME medium without kanamycin.

Extraction and detection of antibiotics.

S. rochei strains were grown in 100 ml of YM liquid medium in Sakaguchi flasks at 28°C for various periods. The broth filtrate was extracted with ethyl acetate, concentrated, and subjected to thin-layer chromatography (TLC) with chloroform-methanol (15:1). Antibiotic activity was detected by bioautography as described previously (14). The crude extract was dissolved in 1 ml of methanol, 1 μl of which was used as the S. rochei γ-butyrolactone fraction for SrrA-binding experiments.

srrY expression from a foreign promoter.

The srrY gene was PCR amplified using pKAR4002 as a template and primers KAR-75OE01 and KAR-75OE03. The amplified product was digested with NdeI and BamHI and inserted into the corresponding sites of pIJ8600 (28), an E. coli-Streptomyces shuttle vector containing a thiostrepton-inducible tipA promoter, to obtain pKAR3049. Strains KY75, KY85, and KU82 were transformed with this plasmid. The transformants were grown at 28°C for 24 h in YM liquid medium containing apramycin, and then thiostrepton was added to induce srrY expression. After a further 48-h cultivation, the ethyl acetate extracts were analyzed by TLC.

RNA preparation.

S. rochei strains were cultured in 100 ml of YM liquid medium in Sakaguchi flasks at 28°C for various periods. Cells were harvested from 1 ml of growing cultures by centrifugation at 4°C, homogenized in 1 ml of TRI reagent (Invitrogen), and stored at room temperature for 10 min. The cell suspension was mixed vigorously with 200 μl of chloroform, and the mixture was centrifuged at 4°C. A 500-μl aliquot of the aqueous fraction containing RNA was mixed with 500 μl of isopropyl alcohol, and the mixture was stored at room temperature for 10 min and centrifuged at 4°C. The pellet containing RNA was washed with 80% ethanol, vacuum dried, and dissolved in 100 μl of diethylpyrocarbonate-treated H₂O. The purified RNA was quantified by UV absorbance at 260 nm.

S1 nuclease protection analysis.

A uniquely end-labeled DNA probe for S1 analysis was generated by PCR as follows. The primer SRRYr4 was 5′ end labeled with [γ-³²P]ATP (GE Healthcare) and T4 polynucleotide kinase (Toyobo) and used for PCR with the unlabeled primer SRRYf2 and the template pKAR4002, which generated a 702-bp product containing the upstream region of srrY (positions −602 to +100 relative to the transcriptional start site of srrY). Thirty micrograms of total RNA was denatured at 75°C for 10 min and hybridized to the ³²P-labeled probe at 30°C for 3 h in 50 μl of S1 hybridization buffer (80% formamide, 0.4 M NaCl, 1 mM EDTA, 20 mM HEPES [pH 6.5]). The S1 nuclease reaction mixture (300 μl) contained the hybridization reaction mixture, 500 U of S1 nuclease (Takara), 30 mM sodium acetate (pH 4.6), 280 mM NaCl, and 1 mM ZnSO₄. After incubation for 20 min at 25°C, the reaction was terminated by the addition of 300 μl of phenol-chloroform. After centrifugation at 4°C, the aqueous fraction containing DNA was precipitated by ethanol and separated on a 5% polyacrylamide gel containing 6 M urea. The labeled DNA was detected by autoradiography. Sequencing ladders were generated by the T7 sequencing kit (USB Corporation) using the ³²P-labeled primer SRRYr4 and the template pKAR4002.

Preparation of SrrA and SrrA^W119A.

By using pKAR3012 as a template, the srrA gene was PCR amplified with primers KAR8201OE and KAR8202OE. The amplified product was digested with BamHI and SalI and inserted into the corresponding sites of pET32b to obtain pKAR3035. This plasmid expresses SrrA as a His₁₀-tagged form (His-SrrA). Next, by using pKAU-8203 as a template, the mutated srrA(W119A) gene was PCR amplified with primers KAR8201OE and KAR8202OE. The amplified product was digested with BamHI and SalI and inserted into the corresponding sites of pET32b to obtain pET-Amt. This plasmid also expresses SrrA^W119A as a His₁₀-tagged form (His-SrrA^W119A).

The His-SrrA (or His-SrrA^W119A) protein synthesized in E. coli strain BL21(DE3) harboring both pLysS and pKAR3035 (or pET-Amt) was affinity purified on Ni⁺-nitrilotriacetic acid agarose (Qiagen) according to the method described previously (34). The purified proteins exhibited a single band of 43.5 kDa after separation by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (see Fig. S6 in the supplemental material). These proteins were digested with recombinant enterokinase (Novagen), and their N-terminal portions containing the His₁₀ tag (16 kDa) were removed by using Ni⁺-nitrilotriacetic acid agarose (see Fig. S6 in the supplemental material). The protein concentration was determined by using a protein assay kit (Bio-Rad) with bovine serum albumin as a standard.

Gel shift assay.

A 935-bp DNA fragment containing the upstream region of srrY (positions −602 to +333) was PCR amplified from pKAR4002 with primers SRRYf2 and SRRYr2 and digested with NaeI and BanI. The resulting 548-bp DNA fragment (probe Y1, positions −452 to +100) and 233-bp DNA fragment (probe Y2, positions +101 to +333) were gel purified, and the former was labeled at the 3′ end of the nontemplate strand with [α-³²P]dCTP (GE Healthcare) by using the Klenow fragment (Toyobo). The binding reaction mixture (20 μl) contained 1 nM labeled DNA and various concentrations of SrrA (or SrrA^W119A) in the binding buffer (20 mM Tris-HCl [pH 8.0], 100 mM NaCl, 1 mM dithiothreitol, 0.1 mg of bovine serum albumin/ml, 5% glycerol). When necessary, 1 ml of the culture extract of strain 51252 or KY85 was added as the S. rochei γ-butyrolactone fraction. In the competition assay, unlabeled DNA competitors were added to the reaction mixture at a final concentration of 200 nM. The reaction mixture was incubated for 30 min at 25°C and subjected to electrophoresis at 4°C on a native 5% polyacrylamide gel in 0.5× TBE buffer (46 mM Tris base, 46 mM boric acid, 1 mM EDTA). Labeled DNA was detected by autoradiography.

DNase I footprinting.

DNase I footprinting was performed using probe Y1 end labeled on either the nontemplate or the template strand. The nontemplate strand-labeled probe was generated as described for the gel shift assay, and the template strand-labeled probe was constructed as follows. The ³²P-labeled DNA probe used for the S1 analysis was digested with NaeI, and the resulting 552-bp DNA fragment was gel purified. The binding reaction mixture (50 μl) contained 2 nM labeled DNA and various concentrations of SrrA in the binding buffer described above. After incubation for 30 min at 25°C, 50 μl of 5 mM MgCl₂-5 mM CaCl₂ solution containing 0.6 μg of DNase I (Roche) was added, and the mixture was incubated for 2 min at 25°C. The reaction was terminated by 300 μl of phenol-chloroform. After centrifugation at 4°C, the aqueous fraction containing DNA was precipitated by ethanol and separated on a 5% polyacrylamide gel containing 6 M urea. Labeled DNA was detected by autoradiography. Sequencing ladders were generated by the same method used for the S1 mapping (for the template strand) or Maxam-Gilbert sequencing of the labeled probe (for the nontemplate strand).

RESULTS

Two SARP genes, srrY and srrZ, are involved in antibiotic production.

Phenotypic studies of single and double mutant forms of the afsA homologue srrX and three arpA homologues, srrA, srrB, and srrC, revealed that srrX and srrA make a γ-butyrolactone-receptor system in S. rochei similar to the A-factor regulatory cascade in S. griseus (1). Additional homologues of S. griseus genes, an adpA homologue, orf116, and an strR homologue, orf3, are also located on the linear plasmid pSLA2-L. However, the disruption of the latter two genes gave no effect on antibiotic production or morphological differentiation (data not shown). Instead, the disruption of the SARP gene srrY in mutant KY75 (all the mutants used in this study are listed in Table 1) ceased both antibiotic production and spore formation (Fig. 1A) (20) (in this study, we focused on antibiotic production and therefore do not show the results for spore formation).

FIG. 1.

(A) Effects of regulatory mutations on antibiotic production. The extracts from cultures of S. rochei strains were assayed by bioautography. The strains used were 51252 (parent), KY75 (srrY mutant), KA61 (srrY mutant), KY71 (srrZ mutant), TN01 (srrW mutant), and KU82 [srrA(W119A) mutant]. (B) Restoration of antibiotic production in the regulatory mutants by introduction of an intact srrY gene. Three regulatory mutants, KY75, KY85 (srrX mutant), and KU82, carrying pKAR3049 (intact srrY gene) or pIJ8600 (control) were analyzed. The culture extracts were separated by TLC and detected by baking with anisaldehyde-H₂SO₄.

To confirm the function of srrY in antibiotic production and spore formation, complementation experiments were carried out. When an intact srrY gene was cloned into plasmid pIJ8600 and expressed from a thiostrepton-inducible tipA promoter in mutant KY75, the production of both LC and LM was restored (Fig. 1B) but spore formation was not. This result suggested that SrrY actually acts as an activator of antibiotic production, and we suspected that the defect in spore formation in mutant KY75 might be caused by a polar effect on the downstream srrC (orf74) gene, encoding a positive regulator of morphological differentiation (1), because mutant KY75 was constructed by the insertion of a kanamycin resistance gene cassette into srrY (20). This speculation was confirmed by the construction of another mutant, KA61, with an in-frame deletion in srrY, which did not produce either antibiotic (Fig. 1A) but sporulated normally.

To reveal the function of two additional SARP genes located on pSLA2-L, srrW (orf55) and srrZ (orf71), we disrupted these genes also. As shown in Fig. 1A, the srrZ disruptant KY71 produced LC but did not produce LM and sporulated normally. On the other hand, the disruption of srrW (mutant TN01) gave no effect on antibiotic production or spore formation. These results indicated that two SARP genes, srrY and srrZ, are involved in antibiotic production, the former possibly being located at a level above the latter in the regulatory hierarchy. Thus, we focused at first on the function of srrY in antibiotic production, and detailed results of our analysis are described below.

Growth-dependent expression of srrY.

The production of antibiotics in streptomycetes occurs at a specific stage of growth. Similarly, the antibiotic activity in S. rochei 51252 was growth dependent, detected at 24 h after inoculation and stably maintained until at least 48 h (Fig. 2A and B). To know the correlation between antibiotic production and srrY expression, RNA was extracted at different growth stages and analyzed by a high-resolution S1 nuclease protection assay. The analysis showed that srrY mRNA increased markedly at 12 to 24 h and disappeared after further cultivation (Fig. 2C). Thus, the expression of srrY also occurred in a growth-dependent manner, slightly preceding antibiotic production. In addition, we identified a transcriptional start site 33 nucleotides upstream of the translational start codon of srrY (Fig. 2C and 3A). Immediately upstream of this site, −35 and −10 sequences similar to the σ^hrdB- and σ^hrdD-type promoters of S. coelicolor A3(2) (9) were also found (Fig. 3A).

FIG. 2.

Growth-dependent antibiotic production and srrY expression in S. rochei 51252. (A) Growth curve of strain 51252. The optical densities at 600 nm (OD600) were obtained from triplicate cultures. (B) Bioautography of the culture extracts. (C) High-resolution S1 nuclease protection analysis of srrY mRNA (upper panel). Lanes G, A, T, and C are sequencing ladders derived from the same labeled primer used for probe preparation. The transcriptional start site (TSS) of srrY is indicated. Lane P contains only the probe. The lower panel shows the ethidium bromide-stained total RNA pattern on 1% agarose gel.

FIG. 3.

(A) Characterization of the upstream region of srrY. The srrY promoter (−35 and −10), transcriptional start site (TSS), Shine-Dalgarno sequence (SD), and initiation codon are boxed. The SrrA-binding site deduced from the gels shown in Fig. 6 is indicated by boldface letters. (B) Comparison of the binding sequences for SrrA and typical γ-butyrolactone receptor proteins. Possible SrrA-binding sites upstream of srrB and srrW were deduced from sequence data. Bases identical to those of the first SrrA-binding sequence are indicated by boldface letters. Complementary bases in the top three palindromes are marked by asterisks. The center of the palindromes is shown by a vertical dashed line.

The γ-butyrolactone-receptor system regulates srrY expression.

To know whether the srrX-srrA system regulates srrY expression, we analyzed the effects of the srrX and srrA mutations on srrY mRNA. Total RNAs were isolated at 24 h, and srrY mRNA was quantified by a low-resolution S1 nuclease protection assay. As shown in Fig. 4, srrY mRNA was almost undetectable in the srrX mutant KY85, while its level in the srrA mutant KA12 was not significantly changed relative to that in the parent strain. The level of srrY mRNA in mutant KY85 was recovered to near normal by the introduction of a second srrA mutation (in mutant KA21). These results suggested that the srrX and srrA genes have positive and negative effects, respectively, on srrY expression. The effects of the srrX and srrA mutations on srrY expression were similar to those on antibiotic production (1, 20). It is noteworthy that the srrA mutation (or the srrX srrA double mutation) did not lead to the overproduction of srrY mRNA (Fig. 4) or antibiotics (1), which suggests a possible compensative mechanism in the regulatory system (see Discussion).

FIG. 4.

Effects of regulatory mutations on srrY expression. srrY mRNAs from various strains were analyzed by a low-resolution S1 nuclease protection assay (upper panel). The lower panel indicates the ethidium bromide-stained total RNA pattern on 1% agarose gel. Strains used were 51252 (parent), KY85 (srrX mutant), KA12 (srrA mutant), KA21 (srrX srrA double mutant), and KU82 [srrA(W119A) mutant].

Specific binding of SrrA to the upstream region of srrY and its inhibition by S. rochei γ-butyrolactone.

To test if the SrrA protein binds to the srrY region, a gel shift assay was performed using ³²P-labeled probe Y1 that contained the upstream and N terminus-encoding regions of srrY (Fig. 5A) (see Fig. S5 in the supplemental material for SrrA preparation). The addition of SrrA to the binding reaction mixture gave a shifted band in a concentration-dependent manner (Fig. 5B, lanes 2 to 5). Competition experiments showed that unlabeled probe Y1 behaved as an effective competitor (Fig. 5C, lane 3), whereas unlabeled probe Y2 containing the srrY coding region (Fig. 5A) did not (Fig. 5C, lane 4), indicating high-level binding specificity of SrrA for the upstream region of srrY. Moreover, the shifted band disappeared with the addition of the culture extract of the parent strain 51252 but not with that of the srrX mutant KY85 (Fig. 5D, lanes 3 and 4), which indicates that SrrA was dissociated from the DNA by the binding of S. rochei γ-butyrolactone. These results, together with gene expression in the mutants (Fig. 4), suggest that the upstream region of srrY is a target of the repressor SrrA and that srrY transcription is controlled by SrrA together with S. rochei γ-butyrolactone.

FIG. 5.

Gel shift assay of the binding of SrrA and SrrA^W119A. (A) Location of the two probes Y1 and Y2. The transcriptional start site of srrY is numbered +1. (B) Concentration-dependent binding of SrrA or SrrA^W119A to the upstream region of srrY. Labeled probe Y1 (1 nM) was mixed with various concentrations of SrrA (lanes 2 to 5) or SrrA^W119A (lanes 6 to 9). Positions of bound (B) and free (F) DNA are shown by arrows on the right. −, no protein. (C) Specific binding of SrrA to the upstream region of srrY. Each reaction mixture contained 1 nM labeled probe Y1 and 500 nM protein (lanes 2 to 7). Then, 200 nM unlabeled probe Y1 (lanes 3 and 6) or unlabeled probe Y2 (lanes 4 and 7) was added. +, present; −, absent. (D) Effect of the S. rochei γ-butyrolactone fraction on the binding of SrrA or SrrA^W119A. To the same reaction mixture described for panel B, the culture extract from strain 51252 (lanes 3 and 6) or KY85 (lanes 4 and 7) was added.

The SrrA-binding site is located upstream of srrY.

DNase I footprinting was carried out to identify an SrrA-binding site(s) in the upstream region of srrY. As shown in Fig. 6, positions −59 to −32 of the nontemplate strand and positions −61 to −35 of the template strand were protected by SrrA. The protected region slightly overlapped with the −35 sequence of the srrY promoter (Fig. 3A) and contained a 26-bp palindromic sequence (Fig. 3B), which is similar to the binding sequences for typical γ-butyrolactone receptors (4, 16, 22, 23, 32). Similar palindromic sequences were also found in the upstream regions of srrB and srrW (Fig. 3B), which suggests their possible function as a target for SrrA binding (see Discussion). In the latter two cases, the palindromic sequences completely overlapped with the putative promoter regions.

FIG. 6.

DNase I footprinting analysis of SrrA-binding site. Probe Y1 was end labeled on either the nontemplate or the template strand. Each reaction mixture contained 2 nM labeled DNA and 0 to 2 μM SrrA. Sequencing ladders were generated by Maxam-Gilbert sequencing of the labeled probe Y1 (A) or cycle sequencing using the labeled primer for probe preparation (B). The sequences protected from DNase I digestion are indicated on the right. Lanes G/A and G, A, T, and C are sequencing ladders.

Trp119 of SrrA is involved in S. rochei γ-butyrolactone binding.

The tryptophan residue at the 119 position of ArpA is necessary for A-factor binding, and this residue is conserved in all of the γ-butyrolactone receptor proteins, including SrrA. To analyze the function of Trp119 of SrrA, we constructed an srrA gene with a point mutation [srrA(W119A)] encoding a protein in which the Trp residue was replaced by Ala. The srrA(W119A) mutant KU82 failed to produce either antibiotic (Fig. 1A) or express srrY (Fig. 4), indicating that SrrA^W119A represses srrY expression even in the presence of S. rochei γ-butyrolactone. To further characterize this relationship in vitro, the binding of SrrA^W119A was analyzed by a gel shift assay. SrrA^W119A was bound to the labeled probe Y1 in a concentration-dependent manner (Fig. 5B, lanes 6 to 9). Competition experiments showed that unlabeled probe Y1 behaved as an effective competitor (Fig. 5C, lane 6), whereas unlabeled probe Y2 did not (Fig. 5C, lane 7). These results were identical to those obtained for the intact SrrA, suggesting that the Trp119Ala mutation did not affect the DNA-binding affinity or specificity of SrrA. In contrast, SrrA^W119A was kept bound to the DNA even in the presence of S. rochei γ-butyrolactone (Fig. 5D, lane 6), indicating that Trp119 of SrrA is crucial for the binding of S. rochei γ-butyrolactone.

srrY is a target of SrrA essential for antibiotic production.

Concerning the antibiotic regulatory cascade in S. rochei, the question of whether SrrA controls other genes in addition to srrY was raised. To answer this question, we expressed srrY under a foreign promoter in the srrX (KY85) and srrA(W119A) (KU82) mutants. If SrrA regulates additional genes essential for antibiotic production, the expression of srrY under a constitutive promoter would not be enough to restore antibiotic production. When srrY was expressed from a thiostrepton-inducible tipA promoter, the production of LC and LM occurred in both mutants (Fig. 1B). These results suggested that srrY may be the sole gene that is repressed by SrrA and also is indispensable for antibiotic production.

DISCUSSION

Extensive transcriptional analysis in this study demonstrated that the srrX-srrA γ-butyrolactone-receptor system in S. rochei regulates LC and LM production through an SARP gene (srrY)-dependent pathway. The srrY gene was expressed in a growth-dependent manner, slightly preceding antibiotic production (Fig. 2). srrY expression was not observed in the srrX mutant and was restored in the srrX srrA double mutant (Fig. 4), which suggested that the transcriptional repression of srrY by SrrA is relieved by the binding of S. rochei γ-butyrolactone synthesized by SrrX. Consistent with this possibility, SrrA was bound specifically to the upstream region of srrY and released by the addition of the S. rochei γ-butyrolactone fraction (Fig. 5). Furthermore, the expression of srrY was always repressed in the SrrA^W119A mutant, and the binding of SrrA^W119A occurred even in the presence of S. rochei γ-butyrolactone (Fig. 4 and 5). All of these results confirmed the signaling pathway from srrX through srrA to srrY, where SrrA functions as a receptor/repressor of S. rochei γ-butyrolactone and srrY. The conserved Trp119 residue in SrrA may form a ligand-binding pocket for a γ-butyrolactone molecule, as shown by X-ray crystallography for CprB, an ArpA homologue in S. coelicolor A3(2) (21). Thus, SrrA^W119A lost S. rochei γ-butyrolactone-binding activity but still kept DNA-binding activity (Fig. 5), as observed for ArpA^W119A in S. griseus (27).

γ-Butyrolactone receptor proteins are bound to a palindromic sequence within the promoter regions of the target genes and repress the transcription of these genes (4, 16, 22, 23, 32). The binding site of SrrA overlaps with the −35 sequence of the srrY promoter and contains a 26-bp palindromic sequence with high similarity to the binding sequences for other γ-butyrolactone receptors (Fig. 3B). Therefore, the binding of SrrA to this site may play a critical role in repressing the transcription of srrY, possibly by preventing the binding of RNA polymerase.

In Streptomyces fradiae, the γ-butyrolactone receptor TylP represses the expression of the SARP gene tylS (4, 25), which in turn results in the repression of tylR, a gene encoding a pathway-specific activator in tylosin biosynthesis (26). In “Streptomyces pristinaespiralis,” the receptor SpbR binds to the promoter of the SARP gene papR1 and regulates pristinamycin synthesis (5). The receptor ScbR in S. coelicolor A3(2) binds directly to two upstream regions of kasO, a pathway-specific SARP gene for a cryptic type I polyketide gene cluster, and represses kasO transcription (32). Thus, accumulated data together with the results of this study suggest that the SARP family regulatory genes are often targets of γ-butyrolactone receptors, except in the case of A-factor, whose target is adpA, an araC family transcriptional regulator gene.

The introduction of an intact srrY gene under the control of a foreign promoter into the srrX or srrA(W119) mutant restored antibiotic production (Fig. 1B), indicating that the derepression of srrY is sufficient for restoration. However, this result did not exclude the possibility that SrrA may repress another gene(s) (such as srrB and/or srrW) dispensable for antibiotic production. Both srrB and srrW contain a consensus palindromic sequence in their upstream regions (Fig. 3B). These regions were actually isolated by a nitrocellulose filter binding assay using the SrrA protein; nevertheless, the promoter region of srrY has not been obtained until now (unpublished results). Furthermore, the disruption of srrB caused an increase in the production of LC and LM (1), whereas the disruption of srrW did not show any effects (Fig. 1A). Our preliminary data showed that SrrB binds to the upstream region of srrY and negatively regulates its expression (unpublished results). These results suggest a possibility that SrrA may also positively regulate srrY through the transcriptional repression of srrB, which may explain why the srrA mutation did not significantly affect srrY expression or antibiotic production (Fig. 4) (1).

Different from the srrY mutant, which did not produce LC or LM, the srrZ mutant produced only LC (Fig. 1A), suggesting a lower location for srrZ than for srrY in the regulatory cascade. Based on all the results hitherto obtained, including preliminary ones, we could depict a possible regulatory cascade leading to LC and LM production in S. rochei (Fig. 7). In this scheme, the signaling pathway from srrX through srrA to srrY has been confirmed by extensive transcriptional analysis. Similar analysis of aspects of the cascade around srrB and under srrY is necessary to confirm this scheme and reveal the entire picture of the complex regulatory cascade in S. rochei.

FIG. 7.

Possible γ-butyrolactone-dependent regulatory cascade leading to LC and LM production in S. rochei. The signaling pathway from srrX via srrA to srrY (solid lines) has been confirmed by extensive transcriptional analysis in this study. Additional pathways (dashed lines) were suggested based on various data, including unpublished results. →, activation; ⊥, inhibition.