![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Previous Article | Next Article 
Journal of Bacteriology, February 2008, p. 894-904, Vol. 190, No. 3
0021-9193/08/$08.00+0 doi:10.1128/JB.01759-07
Copyright © 2008, American Society for Microbiology. All Rights Reserved.
U Activity in Streptomyces coelicolor
,
Department of Chemistry, Williams College, Williamstown, Massachusetts 01267
Received 3 November 2007/ Accepted 21 November 2007
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
U blocks the process of aerial mycelium formation in this organism. Using a green fluorescent protein transcriptional reporter, we have demonstrated that sigU transcription is autoregulated. We have defined a U-dependent promoter sequence and used this to identify 22 likely U regulon members in the S. coelicolor genome. Since many of these genes encode probable secreted proteins, we characterized the extracellular proteome of a mutant with high U activity caused by disruption of rsuA, the presumed cognate anti-sigma factor of U. This mutant secreted a much greater quantity and diversity of proteins than the wild-type strain. Peptide mass fingerprinting was used to identify 79 proteins from the rsuA mutant culture supernatant. The most abundant species, SCO2217, SCO0930, and SCO2207, corresponded to secreted proteins or lipoproteins of unknown functions whose genes are in the proposed U regulon. Several unique proteases were also detected in the extracellular proteome of the mutant, and the levels of the protease inhibitor SCO0762 were much reduced compared to those of the wild type. Consequently, extracellular protease activity was elevated about fourfold in the rsuA mutant. The functions of the proteins secreted as a result of U activity may be important for combating cell envelope stress and modulating morphological differentiation in S. coelicolor. |
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A growing body of evidence suggests that pathways mediating environmental stress responses and morphological development are linked in S. coelicolor (9, 19, 31, 35, 36, 62). The functions of alternative sigma factors appear to play an important role in connecting these processes. The S. coelicolor genome encodes 65 sigma factors (4), 10 of which are in the group 3 family, similar to Bacillus subtilis B, the master regulator of the general stress response in this organism (36), and 51 of which are in the group 4, or extracytoplasmic function (ECF), family, which are typically activated by an external environmental stimulus and regulate cell surface functions (23, 41). Relatively few of these S. coelicolor alternative sigma factors have been characterized in detail (21, 41); however, several have been shown to be required for aerial mycelium formation by phenotypic analysis of mutant strains. The ECF sigma factor BldN is required for aerial mycelium formation and is directly involved in transcription of the bldM gene, encoding a response regulator (5). S. coelicolor B is necessary for protection from osmotic stress, but it is also needed for normal differentiation, as a sigB mutant is bald and overproduces the polyketide antibiotic actinorhodin on rich medium (11). SigB also appears to control a cascade of other group 3 sigma factors (L and M) important for downstream events in sporulation (36). Yet, another B homolog, N, is required for aerial mycelium formation on minimal medium containing glucose (13).
In addition to these cases in which the activity of an alternative sigma factor is necessary for aerial mycelium formation, there are also examples in S. coelicolor in which proper negative regulation of sigma factor activity is important for development. Mutation of the S. coelicolor anti-sigma factor gene prsH (ushX), whose product has been shown to bind to H (51, 61), yielded a strain that was conditionally bald on mannitol-containing medium and also overproduced actinorhodin (58). Disruption of the orthologous anti-sigma factor gene rshA in Streptomyces griseus also gave a bald phenotype that was independent of carbon source (55). Supporting the importance of proper control of H levels and activity during development, it has been shown that BldD represses transcription from the sigHp2 promoter in vegetative hyphae, which is deregulated in a bldD mutant (31).
The appropriate negative regulation of the ECF sigma factor U, the subject of this study, is also important for the development of an aerial mycelium in S. coelicolor. RsuA, a member of the ZAS (zinc anti-sigma factor) family (41), is believed to reside in the membrane and negatively regulate the activity of U by binding to and sequestering this sigma factor away from core RNA polymerase. Mutation of the rsuA gene blocks aerial mycelium formation and severely delays actinorhodin production in S. coelicolor (19). Since the sigU gene itself is not required for proper differentiation, it appears that the unregulated activity of U gives rise to the observed developmental defects in the rsuA bald mutant (19).
In order to understand the relationship between U activity and aerial mycelium formation in S. coelicolor, we sought to identify genes regulated by this ECF sigma factor since their products likely contribute to the observed developmental block in the rsuA mutant. Herein, we propose and provide evidence for a consensus sequence for U-dependent promoters. Since several of the U regulon members encode probable secreted proteins, we performed a proteomic characterization of extracellular proteins from cultures of the rsuA bald mutant compared to that for wild-type S. coelicolor. Activity of U leads to the secretion of large quantities of unique proteins, including those with protease activity.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
sigU-rsuA::tsr mutant were as described previously (19). Strains were cultured in yeast extract-malt extract (YEME) liquid medium or on R2YE solid medium at 30°C (32). Cultures for time courses were prepared in YEME (30 ml) inoculated with 108 M145 spores or a dense mash of NY415 mycelia. At each time point indicated over a 5-day period, 2 ml of culture was removed for determination of optical density at 450 nm, pH, protein concentration, and protease activity (see below). The A450 of the cell-free culture supernatant was subtracted from the culture optical density to correct for pigment production. In particular, the NY415 mutant secreted a pink pigment beginning at 3 days of growth (absorbance maximum, 
490 nm). YEME cultures for the two-dimensional (2D) gel electrophoresis and proteomics experiments were prepared in the same manner but harvested after 3 days of growth.
Construction of GFP reporter strains.
Candidate U-dependent promoters were PCR amplified from M145 genomic DNA and subsequently cloned into the BamHI and XbaI restriction enzyme sites of the green fluorescent protein (GFP) reporter plasmid pIJ8660S by using standard protocols (see Table S1 in the supplemental material for the PCR primer sequences). This plasmid is a modified version of pIJ8660 (52) in which the aac(3)IV gene, conferring apramycin resistance, was removed by digestion with SacI and replaced by the aadA gene, conferring spectinomycin/streptomycin resistance (see reference 19 for PCR primers). The cloned promoter regions included about 200 bp upstream of the start codon of each gene. The fidelity of each promoter insert was confirmed by DNA sequencing (Gene Gateway, Hayward, CA).
The GFP transcriptional reporter plasmids were introduced into S. coelicolor M145 and NY415 and the sigU-rsuA::tsr mutant by conjugation from the methylation-deficient Escherichia coli strain ET12567 (pUB307) (18). Exconjugants were selected by flooding the plates with spectinomycin (200-µg/ml final concentration).
Fluorescence microscopy.
The GFP reporter strains were grown in YEME liquid medium (25 ml) supplemented with spectinomycin (100 µg/ml) for 2 days at 30°C. Cultures were inoculated with 107 spores or a dense mycelial mash for the NY415 bald mutant. Mycelia were harvested by centrifugation, washed once with 20% glycerol, and then resuspended in 20% glycerol prior to fluorescence microscopy. Microscopy was performed with a Nikon Eclipse E600W microscope with a Y-FL epifluorescence attachment and a mercury lamp. A CFI Plan Apo 100x oil objective was used, and GFP was visualized with an Endow GFP HYQ filter cube (Nikon). Exposure time was 1 s in all cases. Images were captured with a Spot RT monochrome camera (Diagnostic Instruments), and fluorescence intensity was assessed using ImageJ software (NIH) (1). Images shown were cropped using Adobe Photoshop.
Determination of extracellular protein. Cell-free culture supernatants were prepared by centrifugation of liquid cultures (2,000 x g). Extracellular protein concentration was determined using Coomassie (Bradford) protein assay reagent and bovine serum albumin (BSA) standards according to the manufacturer's instructions (Pierce). Samples were analyzed in duplicate, and replicate cultures yielded similar results.
Extracellular protein samples were prepared for sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) by mixing 300 µl of culture supernatant with 300 µl 20% trichloroacetic acid (TCA) and incubating the mixture on ice for 30 min. The protein pellet, obtained by centrifugation, was dissolved in 15 µl 1 M Tris base and 6 µl SDS sample buffer (6x) (3). The samples, along with a broad-range protein marker (NEB), were boiled for 3 min prior to being loaded on an 8-to-16% linear gradient Tris-HCl Criterion gel (Bio-Rad) for electrophoresis. The gel was stained with Coomassie brilliant blue R-250, destained, and scanned on a GS-800 densitometer (Bio-Rad).
2D gel electrophoresis. Samples were prepared for 2D electrophoresis using a ReadyPrep 2D starter kit according to the manufacturer's instructions (Bio-Rad). Cell-free culture supernatants were prepared as described above and filtered using a Millex-GV PVDF low-protein binding filter (Millipore). For the gels shown in Fig. 4, equal volumes (1 ml) of cell-free culture supernatant and 20% TCA were mixed and incubated on ice for 30 min. The precipitated protein was retrieved by centrifugation, washed twice with acetone, and air dried. The protein pellet (110 µg for NY415 and 23 µg for M145) was dissolved in 185 µl ReadyPrep rehydration/sample buffer (Bio-Rad). In the first dimension, IPG strips (11 cm), pH 4 to 7, were used for isoelectric focusing with a PROTEAN IEF Cell system (Bio-Rad) with the following conditions: 250 V, 20 min, linear ramp; 8,000 V, 2.5 h, linear ramp; and 8,000 V, 20,000 V·h, rapid ramp. In the second dimension, the IPG strips were electrophoresed on 8-to-16% Tris-HCl Criterion gels (Bio-Rad). Gels were stained overnight with Coomassie brilliant blue R-250 stain and destained. Gels were scanned with a GS-800 densitometer (Bio-Rad), and proteins of interest were removed from the gel with a OneTouch manual spot picker (1.5 mm; Gel Company). The densities of the protein spots were analyzed with the densitometer's Quantity One software. Seven pairs of gels loaded with mutant and wild-type extracellular protein were prepared and analyzed in this manner with samples from three biological replicate cultures. Gels were loaded either with equal masses of protein or with protein isolated from equal volumes of cell-free culture supernatant (as described above). The pattern of protein spots observed was highly reproducible among all of the replicate gels, regardless of the loading method.
|
-cyano-4-hydroxycinnamic acid in 33% acetonitrile, 0.1% trifluoroacetic acid) prior to being spotted on the target plate. These were analyzed on a Microflex MALDI-TOF mass spectrometer (Bruker Daltonics). The peak mass list for each sample was submitted to the Mascot search engine (Matrix Science) (43) and compared to that in the database of S. coelicolor proteins (peptide tolerance, 0.2 Da). Mowse scores above 52 were indicative of significant matches. For those matches with Mowse scores from 52 to 59, postsource-decay analysis of a selected tryptic peptide was used to confirm the protein identification. Most of the proteins identified had Mowse scores much greater then 60, and the predicted molecular weights and pI values were in good agreement with those experimentally observed in the 2D electrophoresis experiments. Many proteins were identified more than once between replicate 2D gels. Assay for extracellular protease activity. An azocasein hydrolysis assay was used to assess protease activity in the cell-free culture supernatants (26). A solution of 2% azocasein (dissolved in 50 mM potassium phosphate, pH 7.0 [125 µl]) was incubated with the culture supernatant (75 µl) for 1.5 h at 37°C. The reaction was stopped with 20% TCA (600 µl), and the precipitated protein was pelleted by centrifugation. The amount of azo dye released by proteolytic activity was determined by adding 1 M NaOH (700 µl) to the postcentrifugation supernatant (600 µl) and measuring the absorbance at 440 nm. The absorbance of a control reaction mixture, in which no culture supernatant was added, as well as the absorbance of a mock reaction mixture, which contained no azocasein to eliminate the contribution to the A440 from the pigmented culture supernatant itself, was subtracted from each measurement. Samples were analyzed in duplicate, and replicate cultures yielded similar results.
Sequence analysis.
The candidate U-dependent promoters were identified in the S. coelicolor genome (GenBank accession no. AL645882) (4) and the Streptomyces avermitilis genome (GenBank accession no. BA000030) (28) by manual searches of the genome sequences, using the EditSeq module of DNASTAR Lasergene software or using the pattern search function of xBase 2 (http://xbase.bham.ac.uk/) (10). Appropriate sequences found within 200 bp upstream of the start codon of a gene were considered U-dependent promoters. The predicted functions of the products of the proposed U regulon genes and the proteins identified in the proteomics study were from ScoDB (http://streptomyces.org.uk/), the S. avermitilis genome project (http://avermitilis.ls.kitasato-u.ac.jp/), or the Pfam search of the protein sequence (17). The SignalP 3.0 server was used for the prediction of signal peptide cleavage sites in the protein sequences (Hidden Markov model) (15).
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
U-dependent promoters by consensus search and transcriptional reporter experiments.
Since the ECF sigma factors are typically autoregulated (23), the U dependence of sigU expression was tested using a GFP transcriptional reporter. The DNA region upstream of sigU was cloned in front of a promoterless copy of egfp, and this construct was introduced into a strain with presumed high, constitutive U activity due to disruption of the gene encoding its cognate anti-sigma factor (NY415; rsuA::Tn5apr), into a sigU deletion mutant (sigU-rsuA::tsr), and into wild-type S. coelicolor (M145). Fluorescence microscopy was used to detect GFP production after growth in liquid or solid media. As expected for a U-dependent gene, bright GFP fluorescence, which was absent in the sigU deletion mutant and the wild type, was observed in the rsuA mutant (Fig. 1A).
|
U-dependent promoter sequence. The sequence TGAACA(N16)CGTA was found 45 bp upstream of the sigU start codon and was also identified upstream of sig45, the sigU ortholog in the recently completed S. avermitilis genome (28). This sequence is quite similar to the promoter consensus recently defined for Bacillus subtilis W (7) and for Mycobacterium tuberculosis L (22), both ECF sigma factors regulated by a ZAS family anti-sigma factor similar to RsuA (12, 19, 40, 50). The S. coelicolor genome was searched with the candidate U-dependent promoter sequence to identify other potential members of the U regulon (Table 1) . Six genes in addition to sigU had this sequence appropriately positioned (upstream, within 200 bp of the start codon) to serve as a U-dependent promoter, including four encoding predicted secreted proteins. The same sequence was found upstream of the S. avermitilis ortholog in all cases, except that only one serine protease gene (SAV7497) was identified instead of two (SCO0732 and SCO0752), and assuming that the SCO5104 ortholog, SAV3185, has a start codon downstream of the one annotated (data not shown). The S. avermitilis gene SAV5765 also had an exact match, and the upstream region of its S. coelicolor ortholog, SCO2408, had one mismatch relative to the sigU promoter sequence (TGAACG in the –35 region). One gene with a putative U-dependent promoter with an exact match but an altered 17-bp spacing (SCO0930) was identified in the S. coelicolor genome. The U dependence of these eight U regulon candidate genes (SCO0732, SCO0752, SCO1356, SCO2217, SCO5104, SCO6650, SCO0930, and SCO2408) was tested using GFP transcriptional reporter strains and fluorescence microscopy as previously described for sigU.
|
U-dependent genes, fluorescence observed from the GFP reporter was expected to be high in the rsuA mutant compared to that in the sigU deletion mutant. Representative results with the reporters for SCO1356 (a perfect match relative to the sigU promoter sequence), SCO0930 (17-bp spacing), and SCO2408 (one mismatch relative to the –35 region) are shown in Fig. 1. In many cases, at least under these experimental conditions, the genes seemed to be solely under the control of U, with significant fluorescence observed in NY415 but little if any fluorescence observed in the sigU deletion mutant or the wild type (SCO0732, SCO0930, SCO1356, SCO2408, and SCO6650). In the cases of SCO0752 and SCO2217, some fluorescence was observed in the sigU deletion mutant and the wild type, but the fluorescence intensity was greatly increased in the NY415 strain (data not shown). In only one case was the U dependence of expression unable to be confirmed. The SCO5104 reporter exhibited high levels of fluorescence in the wild-type background, which were not convincingly elevated in NY415 (data not shown). Perhaps this gene is subject to additional regulation that obviates the effect of unregulated U activity. The list of confirmed or likely U-dependent genes given in Table 1 includes the genes whose promoters were tested here (group A plus SCO2408), additional genes having the same sequence as SCO2408 appropriately positioned upstream of these genes (group B), and genes with a similar sequence whose use as a U-dependent promoter was suggested by the results of the proteomics experiments described below (group C; TGAGCA in the –35 region [mismatch underlined]).
Effect of U activity on extracellular protein.
Given that many of the confirmed U-dependent genes were predicted to encode extracellular proteins (secreted proteins or lipoproteins), we assayed the protein content of the medium of liquid cultures of NY415 (high U activity) and wild-type S. coelicolor (Fig. 2A). Over a 5-day time course, the protein content of the wild-type culture medium was relatively low and constant (about 17 µg/ml). In contrast, by 3 days of growth, NY415 had generated almost an order of magnitude more protein in the culture medium (150 µg/ml), and these protein levels continued to increase in samples taken at later time points (300 µg/ml at 5 days).
|
|
5.0 by 41 h of growth (corresponding to entry into stationary phase) (Fig. 2A). By contrast, the growth of the NY415 mutant increased the culture pH to 8.0 by the end of the 5-day time course. Extracellular proteome of the anti-sigma factor mutant compared to that of the wild type. The identities of the major species of extracellular proteins generated by the NY415 anti-sigma factor mutant and wild-type S. coelicolor were determined by 2D electrophoresis and peptide mass fingerprinting using MALDI-TOF mass spectrometry. Cultures of both strains were grown for approximately 3 days in liquid medium prior to protein isolation by TCA precipitation. At this time, the NY415 strain had begun to release a large quantity of proteins into the medium, yet the emergence of suppressor mutants (19) was still at low levels in the culture. The profile of proteins found in the mutant sample was different and much more diverse than that seen for the wild type (Fig. 4).
Twenty-three proteins were identified in the extracellular proteome of the wild-type bacterium (Table 2). Eighteen of the 23 proteins were expected to be secreted based on the prediction of an N-terminal signal peptide (15). The two major species secreted were SCO1968, a putative glycerophosphoryl diester phosphodiesterase recently shown to be regulated by PhoP (47), and SCO0762, an experimentally characterized secreted protease inhibitor (29, 33). Eight of these 23 proteins were also identified in the extracellular proteome of the M600 wild-type strain of S. coelicolor by Kim et al. (33), including SCO1968 and SCO0762.
|
|
Fifty-four of the 79 extracellular proteins identified from the NY415 culture were predicted to have cytoplasmic locations (see Table S2 in the supplemental material) (15). Over half of these 54 proteins are likely involved in the primary metabolism of the cell, and particularly well represented were enzymes of glycolysis, the TCA cycle, and the pentose phosphate pathway. Chaperones with roles in protein folding were detected (GroEL, GroES, and DnaK), as well as several other stress response proteins with possible roles in cold shock, oxidative stress, and tellurium resistance. The identification of these presumed cytoplasmic proteins in the NY415 extracellular proteome suggests some lysis or leakiness of the mutant cell wall. Microscopic examination of the mutant hyphae at this growth stage did not indicate elevated levels of cell lysis (data not shown), and the most abundant species in the NY415 cytoplasmic proteome, SCO6282 (K. Larabee, unpublished data), was not detected in the extracellular proteome.
The results of this proteomic study are consistent with our assignment of genes to the U regulon based on the promoter consensus searches and GFP transcriptional reporter experiments (Table 1 and Fig. 1). Many of the predicted secreted proteins/lipoproteins encoded by genes in the U regulon were identified in the extracellular proteome of NY415, including SCO0752, SCO2217, and SCO0930. In addition, the predicted cytoplasmic protein SCO6650 was found extracellularly (see Table S2 in the supplemental material). The gene for SCO2207, one of the major extracellular proteins in the rsuA mutant, has an appropriately positioned sequence with only one mismatch relative to the sigU promoter (TGAGCA in the –35 region), so we have included genes with this upstream sequence in Table 1 as likely members of the U regulon. The isoelectric points for the other predicted secreted proteins encoded by likely U regulon genes (SCO0732, SCO0986, and SCO1573) are basic, and thus, these proteins were not expected to be resolved in the 2D gel experiments, given the isoelectric focusing conditions used. Thus, only SCO1356 was expected extracellularly but not detected, perhaps because this small protein (about 15 kDa) is present in relatively small amounts or was not well resolved in the electrophoresis experiments.
Secreted protease activity in the anti-sigma factor mutant. Since several of the extracellular proteins unique to the rsuA mutant were predicted to function as proteases, the cell culture supernatants from NY415 and the wild type were assayed for protease activity over the time course of growth. While the wild type exhibited only low and relatively constant proteolytic activity in an azocasein hydrolysis assay, protease activity in the mutant culture supernatants began to increase at 2 days of growth, reaching a plateau at a level about fourfold higher than that of the wild type at 4 days of growth (Fig. 2B). The medium from a 3-day-old NY415 culture was also able to promote the time-dependent degradation of BSA, as assessed by gel electrophoresis, while that from a comparable wild-type culture did not (data not shown). No cytoplasmic extracts from either the wild type or NY415 caused BSA proteolysis (data not shown).
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
U, given the dramatic effect that uncontrolled activity of this sigma factor has on halting the developmental progress of S. coelicolor. With direct evidence from GFP reporter studies and supporting evidence from proteomic studies, we propose TGA(A/G)C(A/G)(N16-17)CGTA as a provisional promoter consensus sequence for U-dependent genes and have used this sequence to identify likely U regulon members. This consensus sequence is entirely consistent with the results of a recent microarray study that identified a group of 18 genes, including sigU and 6 of the regulon members proposed here, as having similar expression profiles during growth of M600 S. coelicolor on a maltose-based solid medium (see reference 25, additional data file 6, QT cluster 11). This group identified a common DNA sequence upstream of a subset of these similarly expressed genes that is the same as the U-dependent promoter consensus sequence reported here, with the exception that more variability was allowed at the fifth position of the –35 element (25). It will be important to characterize the full range of DNA sequences recognized by U to fully define this regulon. We also note that Table 1 does not include any downstream genes that may exist in an operon with a U regulon gene. Few of the genes reported here have been annotated as being in a likely operon (only SCO1356, SCO6650, SCO1935, and SCO2575) (ScoCyc database; http://scocyc.jic.bbsrc.ac.uk:1555/); however, this is another avenue for future study.
The ECF sigma factors often regulate stress response or transport functions at the cell envelope (23). Consistent with a general role in cell surface events, over half of the proposed U regulon genes encode probable secreted proteins, lipoproteins, or membrane proteins. This led us to examine the S. coelicolor extracellular proteome under conditions where U was highly active (in this case in an rsuA mutant). The direct and indirect consequences of U activity had a dramatic impact on the extracellular proteome in the rsuA bald mutant, elevating the concentration of extracellular protein over an order of magnitude above that seen in a wild-type culture. The three most abundant protein species observed in the extracellular proteome of the rsuA mutant were encoded by confirmed or likely U-dependent genes and appeared to be absent from the wild-type culture supernatant (SCO2217, SCO2207, and SCO0930). SCO2217 was clearly the most abundant secreted protein produced by this mutant. BLAST analysis (49) showed that SCO2217 shares over 50% identity with SCO0644, another product of a U regulon gene that was detected in the extracellular proteome of the rsuA mutant. These proteins have not been characterized, and they contain a conserved domain of an unknown function (DUF1996; Pfam accession number PF09362) (17) that is most common in fungal proteins, although there are other bacterial examples. In some of the fungal homologs, the proteins possess a WSC domain(s) in addition to the DUF1996 domain; the WSC domain is a putative carbohydrate binding domain believed to play a role in maintaining cell wall integrity under stress conditions (57). We note that large amounts of SCO2217 were recently observed in a proteomic study of exported cell wall-associated proteins when S. coelicolor was grown on mannitol soya solid medium (63) (see the supplemental material).
It is striking that many of the other secreted proteins that were observed solely in the rsuA mutant culture supernatant have hypothesized hydrolytic activities or functions in cell wall metabolism (Table 3). The novel secreted hydrolytic enzymes include a potential nuclease (SCO1908), phosphatase (SCO2068), and esterase (SCO3053) and potential peptidases (SCO0752 and SCO6131). Four of the proteins likely bind to cell wall peptidoglycan, including the penicillin binding proteins SCO1875 and SCO3901, the low-molecular-mass penicillin binding protein SCO6131 (with probable D-alanyl-D-alanine carboxypeptidase activity) (48), and SCO3097, an uncharacterized protein in the resuscitation-promoting factor (Rpf) family whose members have muralytic activity and stimulate cell growth after periods of dormancy (30, 38, 46). In addition to its putative esterase domain, SCO3053 possesses a D-mannose binding lectin domain (Pfam accession no. PF01453) (17) and is homologous to the potent bacteriocin putidacin from Pseudomonas sp. strain BW11M1 (42). It is tempting to speculate that U, whose activity elicits the production of hydrolases, probable peptidoglycan remodeling enzymes, and even a potential bacteriocin, plays a role in enhancing the competitiveness of S. coelicolor in its complex soil environment in analogy to the B. subtilis ECF sigma factor W (6, 14, 24).
Consistent with the observation of novel extracellular proteases and a decrease in levels of the known protease inhibitor SCO0762 (33), extracellular protease activity was found to be considerably elevated in an rsuA mutant culture compared to that in the wild type. It has been proposed that protease activity promotes aerial mycelium formation by facilitating recycling of nutrients from the degraded substrate mycelium for use in construction of an aerial mycelium. Several studies have found that application of serine protease inhibitors, including p-tosyl-L-lysine chloromethyl ketone and SgiA, the S. griseus ortholog of SCO0762, can inhibit aerial mycelium formation in Streptomyces species (26, 34, 39). Conversely, in Streptomyces albogriseolus S-3253, a mutant that failed to produce a SCO0762 ortholog, SSI, exhibited enhanced extracellular protease activity accompanied by decreased sporulation (54). A mutant with enhanced extracellular protease activity and sporulation defects has also been isolated from Streptomyces fradiae (45). It may be that developmental events in the streptomycetes require the proper balance of extracellular protease and protease-inhibitory activities, which is disrupted in an rsuA mutant. In addition to the predicted secreted proteases identified in Table 3, we also detected several proteases of likely cytoplasmic origin in the extracellular proteome of the rsuA mutant (see Table S2 in the supplemental material). The contribution of these proteases to the observed extracellular protease activity remains to be determined; however, a cytoplasmic extract of the rsuA mutant did not support proteolysis in a BSA degradation assay.
Misregulation of carbon metabolism is another possible factor contributing to the bald phenotype of the rsuA mutant. We have shown that the bald mutant does not acidify a glucose-based culture medium as the wild-type strain does. It has been commonly observed that Streptomyces species acidify the growth medium by excreting organic acid products of glucose metabolism (pyruvate, and -ketoglutarate) (2, 27, 37, 53). The pH differences between the wild-type and rsuA mutant cultures suggest that U activity affects the normal pathways of glucose metabolism in S. coelicolor. The dependence of aerial mycelium formation on the proper control of carbon utilization has been previously demonstrated, with the finding that many bald mutants are defective in carbon catabolite repression (44) and the observation of bald phenotypes for citrate synthase (citA) and aconitase (acoA) mutants of S. coelicolor (59, 60). Further elucidation of the links between the regulation of carbon metabolism, extracellular protease activity, and U may illuminate the molecular basis for the bald phenotype of the rsuA mutant.
In addition to proteins with predicted N-terminal signal sequences (15), we also observed many proteins without typical export signals in the culture supernatant of the rsuA bald mutant (see Table S2 in the supplemental material). A large majority of the wild-type extracellular proteins identified here were also predicted to be secreted (18 of 23) (Table 2). The origin of the presumed cytoplasmic proteins in the extracellular milieu of the mutant is unclear at his point. Perhaps activities of the unique hydrolytic and/or cell wall-modifying proteins cause leakiness of the mutant's cell wall. Others have also observed the presence of many presumed cytoplasmic proteins in the S. coelicolor extracellular proteome. In a recent characterization of wild-type S. coelicolor M600, 36 of 47 extracellular proteins identified were predicted to be cytoplasmic (33), and in a study of surface-grown M145, 61 of 98 extracellular proteins found were expected to be cytoplasmic (63).
In summary, the unregulated activity of U has a multitude of effects in S. coelicolor, including (i) eliciting the secretion of large quantities of novel proteins, (ii) increasing extracellular protease activity, and (iii) causing the basification rather than acidification of the culture supernatant when S. coelicolor is grown in a glucose-based medium. We suggest that upon U activation, the cell mounts a vigorous response at its surface, perhaps remodeling its own cell envelope and/or attacking that of a microbial competitor(s). Further studies of this ECF sigma factor should yield fruitful insight into the relationship between stress response, morphological differentiation, and antibiotic production in the streptomycetes.
|
ACKNOWLEDGMENTS |
|---|
This work was supported by Williams College and by an AREA grant from the National Institutes of Health (1 R15 GM076028-01) to A.M.G.
|
FOOTNOTES |
|---|
Published ahead of print on 7 December 2007.
Supplemental material for this article may be found at http://jb.asm.org/.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-keto acids by strains of Streptomyces venezuelae. Can. J. Microbiol. 30:1014-1021.[Medline]BldN, an extracytoplasmic function RNA polymerase sigma factor required for aerial mycelium formation in Streptomyces coelicolor A3(2). J. Bacteriol. 182:4606-4616.W-dependent genes that provide intrinsic resistance to antimicrobial compounds produced by Bacilli. Mol. Microbiol. 60:765-782.[CrossRef][Medline]W regulon: a comparative analysis of promoter consensus search, run-off transcription/macroarray analysis (ROMA), and transcriptional profiling approaches. J. Mol. Biol. 315:443-457.L and roles in virulence and in global regulation of gene expression. Infect. Immun. 74:2457-2461.W protein and intrinsic resistance to antimicrobial compounds. Sci. Prog. 89:243-266.[Medline]H controls the expression of ssgB, a homologue of the sporulation-specific cell division gene ssgA, in Streptomyces coelicolor A3(2). Mol. Genet. Genomics 267:536-543.[CrossRef][Medline]B governs osmotic and oxidative response as well as differentiation via a network of sigma factors in Streptomyces coelicolor. Mol. Microbiol. 57:1252-1264.[CrossRef][Medline]phoP mutant. Proteomics 7:2410-2429.[CrossRef][Medline]W anti-sigma factor RsiW is degraded by intramembrane proteolysis through YluC. Mol. Microbiol. 52:1091-1105.[CrossRef][Medline]H is regulated by an anti-sigma factor. FEMS Microbiol. Lett. 209:229-235.[Medline]H and related sigma factors in glucose-dependent initiation of morphological and physiological development of Streptomyces griseus. Gene 320:127-135.[CrossRef][Medline]
| |||||||||||||||||||||||||||||||||
