SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Concluding remarks
  7. Experimental procedures
  8. Acknowledgements
  9. References

Serratia sp. ATCC 39006 (39006) uses a complex hierarchical regulatory network allowing multiple inputs to be assessed before genes involved in secondary metabolite biosynthesis are expressed. This taxonomically ill-defined Serratia sp. produces a carbapenem antibiotic (Car; a β-lactam) and a red pigmented antibiotic, prodigiosin (Pig; a tripyrrole), which are controlled by the smaIR quorum sensing (QS) locus. SmaR is a repressor of Pig and Car when levels of N-acyl- l-homoserine lactones, produced by SmaI, are low. In this study, we demonstrate direct DNA binding of purified SmaR to the promoter of the Car biosynthetic genes and abolition of this binding by the QS ligand. We have also identified multiple new secondary metabolite regulators. QS controls production of secondary metabolites, at least in part, by modulating transcription of three genes encoding regulatory proteins, including a putative response regulator of the GacAS two-component signalling system family, a novel putative adenylate cyclase and Rap (regulator of antibiotic and pigment). Mutations in another gene encoding a novel predicted global regulator, pigP, are highly pleiotropic; PigP has a significant ‘master’ regulatory role in 39006 where it controls the transcription of six other regulators. The PigP protein and its homologues define a new family of regulators and are predicted to bind DNA via a helix-turn-helix domain. There are regulatory overlaps between the QS and PigP regulons that enable the information from different physiological cues to be funnelled into the control of secondary metabolite production.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Concluding remarks
  7. Experimental procedures
  8. Acknowledgements
  9. References

Serratia sp. ATCC 39006 (39006) is a member of the Enterobacteriaceae and, like S. marcescens, synthesizes the red, linear tripyrrole antibiotic, prodigiosin (Pig; 2-methyl-3-pentyl-6-methoxyprodigiosin) (Harris et al., 2004). Prodigiosins display immunosuppressive, proapoptotic and anti-cancer properties and are therefore potentially interesting candidates for drug development (reviewed in Manderville, 2001; Perez-Tomas et al., 2003). However, the exact physiological role of prodigiosins in the producer organisms is still debated (Rosenberg et al., 1986; Paruchuri and Harshey, 1987; Trutko and Akimenko, 1989; Hood et al., 1992). In addition to Pig, 39006 produces another secondary metabolite, a carbapenem (Car; 1-carbapen-2-em-3-carboxylic acid), a broad spectrum β-lactam antibiotic (Parker et al., 1982; Bycroft et al., 1987). Car is also produced by Erwinia carotovora ssp. carotovora (Ecc) and Photorhabdus luminescens ssp. laumondii strain TT01 (Parker et al., 1982; Bycroft et al., 1987; Derzelle et al., 2002).

Previously, the Pig biosynthetic gene cluster from 39006 was cloned (Thomson et al., 2000). The operon contains the genes pigA-O, which are transcribed as a single polycistronic mRNA (Slater et al., 2003). A biosynthetic pathway has been proposed for Pig based on sequence similarities with genes involved in production of the related compound undecylprodiginine from Streptomyces coelicolor A3(2) (Cerdeño et al., 2001; Harris et al., 2004). Recently, we performed a detailed analysis of the biosynthesis of Pig in 39006 and have significantly revised the pathway (Williamson et al., 2005). Two separate pathways utilize proline, alanine, acetate, serine and methionine to form the intermediates, MAP (2-methyl-3-amylpyrrole) and MBC (4-methoxy-2,2′bipyrrole-5-carboxyaldehyde), which then converge in a condensation reaction catalysed by a terminal enzyme to produce Pig (Morrison, 1966; Wasserman et al., 1973; Harris et al., 2004; Williamson et al., 2005).

Eight genes are present in the predicted carbapenem operon (carABCDEFGH) in 39006 (Cox et al., 1998; Thomson et al., 2000). Sequence comparison with the Ecc carbapenem cluster predicted that genes carA-E encode proteins required for Car biosynthesis whereas carF and carG encode a novel β-lactam resistance mechanism. The function of carH is unknown (McGowan et al., 1996; McGowan et al., 1997; McGowan et al., 1999; Thomson et al., 2000).

Quorum sensing (QS) is a mechanism whereby bacteria can regulate gene expression in response to the population cell density via detection of a diffusible signalling molecule (Whitehead et al., 2001). In 39006, an N-acyl homoserine lactone (N-AHL) QS system controls the production of Pig, Car and the exoenzymes pectate lyase (Pel) and cellulase (Cel) (Thomson et al., 2000; Crow, 2001; Slater et al., 2003). The luxIR homologues smaI and smaR (secondary metabolite activator) constitute the QS locus in 39006 and are transcribed convergently (Thomson et al., 2000; Crow, 2001). The N-AHL synthase, SmaI, produces two signalling molecules, N-butanoyl- l-homoserine lactone (BHL) and N-hexanoyl- l-homoserine lactone (HHL), with BHL being the major product (Thomson et al., 2000). A smaI mutant is phenotypically Pig and Car, and transcription of pigA-O and carA-H is reduced in this mutant. Furthermore, inactivation of smaR in a smaI mutant fully restored Pig and Car production, indicating that a functional SmaR is required for the observed phenotype of a smaI mutation (Slater et al., 2003). Genetic analysis has led to a model of QS regulation of Pig, Car and exoenzymes in 39006, in which, at low cell density, in the absence of BHL/HHL, SmaR binds DNA and represses transcription. At high cell density, BHL/HHL levels increase. These N-AHLs are thought to bind to SmaR, leading to a reduction in DNA-binding affinity with a consequent alleviation of repression (Slater et al., 2003). However, prior to this study, direct binding of SmaR to DNA had not been demonstrated.

Although QS is an important input into Pig and Car production, other regulatory mechanisms are also present, some of which are outlined below (Slater et al., 2003).

Upstream of carA-H is carR, which encodes a LuxR homologue essential for activation of the carbapenem operon in 39006 (Cox et al., 1998; Thomson et al., 2000). Unlike most LuxR family proteins, the activity of CarR does not appear to be directly modulated in response to the concentration of N-AHL (Cox et al., 1998). However, transcription of carR is repressed by the SmaIR QS system (Slater et al., 2003). In this way, the production of CarR, but not its activity, is controlled by the concentration of BHL/HHL. It is also possible that SmaR may directly repress carA transcription in 39006.

Mutations in rap (regulator of antibiotic and pigment) abolish Pig and Car production (Thomson et al., 1997). The rap gene is predicted to encode a protein homologous to the virulence factor SlyA from Salmonella enterica ssp. enterica serovar Typhimurium (Thomson et al., 1997). Rap controls Pig and Car, at least in part, at the level of transcription of the biosynthetic clusters, but has no effect on smaI transcription (Slater et al., 2003).

In this article, biochemical evidence is presented that demonstrates specific DNA binding of SmaR to the carA promoter and the inhibition of this interaction in the presence of the cognate signal, BHL. We examine further the regulation of secondary metabolite production in 39006 and identify a new suite of Pig regulators, some of which also act pleiotropically to regulate Car and/or exoenzymes. The transcriptional hierarchy of the newly identified regulators and the QS system is investigated with respect to Pig and Car production. We have identified a highly pleiotropic regulator (PigP), which is a novel master regulator with homologues in a very limited group of Enterobacteriaceae. Three regulators under control of PigP are also in the SmaIR QS regulon. One of these is Rap, the other is a putative novel adenylate cyclase and the third is a putative response regulator of the GacAS family of two-component systems. In addition, the putative partner histidine kinase of the response regulator was also identified and presumably responds to an, as yet unidentified, physiological signal. Therefore, the genes under dual control of QS and PigP represent hubs where information from different cues is integrated into secondary metabolite regulation.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Concluding remarks
  7. Experimental procedures
  8. Acknowledgements
  9. References

Promoter analysis of carA and carR

To understand the transcriptional control of QS regulated genes, the transcriptional start sites of carA and carR were mapped by primer extension (Fig. 1). Two putative transcriptional start sites were identified for the carA promoter (Fig. 1A, B and E). One of these start sites (denoted +1 in Fig. 1B) is the most probable candidate, based on the locations of possible −10 and −35 sequences. It is not known whether the two start sites represent transcripts produced from different promoters, processing of the 5′ end of the mRNA, or an experimental artefact resulting from secondary structure in the mRNA. A single transcriptional start site was observed for the carR mRNA (Fig. 1C–E). Adjacent to the predicted 5′ end of the mRNA is a putative extended −10 region (Barne et al., 1997) and a −35 sequence (Fig. 1D). The genetic organization of the Car cluster is shown in Fig. 1E. The transcriptional start site of pigA (the first gene in the prodigiosin biosynthetic cluster) was previously analysed by primer extension and putative promoter elements were identified (Slater et al., 2003).

image

Figure 1. Primer extension analysis of the carA and carR transcripts. Sequencing gel for the transcriptional start sites of carA (A) and carR (C). Two potential 5′ ends of the carA mRNA were identified (boxed). Promoter regions upstream of the carA (B) and carR (D) translational start sites. Potential −10 and −35 elements are underlined and in bold and the putative transcriptional start sites are in bold, boxed and denoted +1. Locations of the predicted transcriptional start sites are noted in relation to the start codon for translation of the corresponding protein. (E) Genetic organization of the carRABCDEFGH locus.

Download figure to PowerPoint

SmaR binding to the carA promoter is inhibited by BHL

Genetic and phenotypic evidence predict that SmaR represses Pig and Car production by binding to the promoter regions of the respective biosynthetic gene clusters (pig and car) and to the promoter of carR, which encodes the N-AHL independent LuxR-type activator of Car production. To test this hypothesis, labelled DNA fragments containing the promoter regions of carR, carA and pigA were combined with varying concentrations of purified N-terminally hexahistidine-tagged SmaR (His6-SmaR) and assayed for binding as described in the Experimental procedures. The DNA-binding band-shift assay demonstrated that increasing concentrations of His6-SmaR caused a decrease in mobility of a 250 bp DNA fragment containing the carA promoter, the majority of which is shifted by 30 nM His6-SmaR (Fig. 2A). Binding could be out-competed by a 50-fold excess of unlabelled carA promoter but not by a non-specific control, showing the binding to be specific. Furthermore, when His6-SmaR was combined with the labelled non-specific DNA fragment in binding assays under identical conditions, no shift was observed (data not shown). A variety of conditions were tested but specific binding of His6-SmaR to the carR and pigA promoters could not be demonstrated.

image

Figure 2. His6-SmaR binding to the carA promoter is inhibited by BHL. A. Binding of His6-SmaR to a 250 bp DNA fragment containing the carA promoter. DNA band-shift assays were performed as described in Experimental procedures. Each lane contains 1.5 nM of DIG-labelled carA promoter (carApro) DNA and the concentrations of His6-SmaR as indicated in the figure. In addition, competition experiments were performed, in which a 50-fold excess (75 nM) of either the unlabelled carA promoter fragment or a non-specific fragment were added. B. Binding of His6-SmaR to a DNA fragment containing the carA promoter in the presence of exogenous BHL. Each lane contains 1.5 nM of DIG labelled carA promoter DNA, 30 nM His6-SmaR where indicated on the figure and increasing concentrations of BHL as indicated. In addition, the final lane contains 15 µM OdDHL as a non-specific N-AHL control.

Download figure to PowerPoint

The proposed model of SmaIR action implies that the presence of BHL/HHL causes SmaR to lose the ability to bind DNA. To verify this model, BHL was incubated with the DNA-binding assays at a range of concentrations (Fig. 2B). The DNA binding ability of 30 nM His6-SmaR was completely inhibited on addition of 6 µM BHL. Furthermore, addition of a non-cognate N-AHL (15 µM N-(3-oxododecanoyl)- l-homoserine lactone; OdDHL) did not inhibit binding of His6-SmaR to DNA.

Identification and characterization of prodigiosin regulators

Demonstration that SmaR is able to bind to the carA promoter, but not to that of carR or pigA suggested that there are other regulatory proteins involved that control Pig and Car expression in response to SmaR activity. Indeed, the wealth of phenomenological data in the literature, showing that Pig production is influenced by multiple environmental and physiological stimuli, predicts that there are likely to be several regulatory systems involved (Williams et al., 1971a; 1971b; Silverman and Munoz, 1973; Witney et al., 1977; Rjazantseva et al., 1994; Sole et al., 1997; Cang et al., 2000). A mutagenesis approach was utilized to identify these additional predicted regulators of Pig. Mutants with reduced Pig phenotypes were visually identified and isolated from a population of mini-Tn5lacZ1 insertion mutants of Serratia 39006 strain LacA. The sites of the mini-Tn5lacZ1 insertions were mapped, by Southern blot and hybridization analysis, to within the pig biosynthetic gene cluster or outside of the cluster (potential regulators of Pig).

DNA sequence analysis revealed that the insertions carried by the ‘regulatory’ mutants mapped to eight loci. Most of the predicted products shared similarity with known regulators, as summarized in Table 1. More information about the regulators is introduced in the relevant sections.

Table 1.  Genetic and phenotypic analysis of mutants affected in prodigiosin production.
Mutant(s)Site(s) of insertionHomology (amino acid identity/similarity percentage)aSignificant motifs/notesProdigiosin production (%)b
  • a

    . References for the homologues are provided in the relevant sections of the article.

  • b

    . Measurements shown are the maximal levels of Pig attained by each strain throughout growth and expressed as a percentage of 39006 wild-type Pig levels. The mean values of triplicate experiments are shown ± SD.

  • c

    . The transposon insertion responsible for the hyper-Pig phenotype of strain ROP4 (a derivative of strain SO4 (Slater et al., 2003) was located downstream from that in HSPIG66 and within the gene, which shared similarity with yhdA. In this article we utilize ROP4 to examine regulation of pigX.

Wild type 39006   100 inline image
Pig regulators
 HSPIG17 (pigQ)0.657 kb ORFExpA, Erwinia carotovora (90/95)FixJ family of response regulators17 ± 5 inline image
 HSPIG23 (pigR)0.540 kb ORFCyaB, Aeromonas hydophila (42/56)Adenylate cyclase of unknown function72 ± 12 inline image
 HSPIG26 (pigS) 0.357 kb ORF (located 2.2 kb 5′ of pigA-O operon)HylU activator of haemolysin production, Vibrio cholerae (37/54)SmtB/ArsR family of regulatory proteins; also include repressors, e.g. NolR, Rhizobium. HTH46 ± 9 inline image
 HSPIG46 (pigV)0.393 kb ORFYgfX hypothetical protein, E. coli (45/60)Unknown function21 ± 9 inline image
 HSPIG62 (pigW)2.763 kb ORFExpS, Erwinia carotovora (71/81)Hybrid sensor/kinase-response regulator. Putative partner of ExpA30 ± 9 inline image
 HSPIG66 (pigX) (ROP4)c0.160 kb intergenic regionAdjacent gene products; YhdA (32/54, of 193 aa aligned) and YhdH, E. coli (71/84, of 198 aa aligned)possible diguanylate cyclase/phosphodiesterase and putative quinone oxidoreductase respectively46 ± 8 inline image
Pig and Car regulators
 HSPIG67 (pigP)0.615 kb ORFHypothetical protein in Serratiamarcescens Db11 (77/69)N-terminal HTH (λ repressor family)5 ± 3 inline image
 HSPIG43 (pigU)1.356 kb intergenic regionAdjacent to HexA repressor of virulence factors (96/97) and aspartate amino transferase (97/99), Erwinia carotovoraHexA belongs to LysR family, N-terminal HTH7 ± 7 inline image

Phenotypic analysis of Pig regulators

The level of Pig produced by each mutant was measured throughout growth in Luria–Bertani (LB). Although the timing of Pig production was not altered dramatically between mutants, the maximal levels produced varied greatly, ranging from very low pigmentation (pigP) to only a subtle change (pigR) when compared with the WT (WT refers to use of the LacA parent strain throughout this article) (Table 1).

Further phenotypes were assayed to determine whether the regulators identified were dedicated solely to Pig production. Of the phenotypes tested using plate assays (Car, exoenzymes (Cel and Pel), N-AHL production and swimming motility) four mutants were found to be pleiotropic. Other mutants may have subtle effects on these phenotypes, which were not detected using these plate assays.

Two strains, HSPIG67 (pigP) and HSPIG43 (insertion upstream of pigU), were defective in Car production and were translucent in appearance rather than the porcelain white colour observed for the Pig biosynthetic mutants (Fig. 3A). Furthermore, HSPIG67 (pigP) showed reduced levels of Cel and a small reproducible decrease in Pel, while also exhibiting increased swimming motility (Fig. 3B–D). HSPIG43 displayed a reduction in Cel activity (Fig. 3B). HSPIG17 (pigQ) and HSPIG62 (pigW) showed reductions in exoenzyme production and motility (Fig. 3B–D). In addition, HSPIG17 (pigQ) and HSPIG62 (pigW) showed subtle reductions in Car production in plate assays (Fig. 3A). However, Car production and carA::lacZ and carR::lacZ activity was identical to WT at all phases of growth, indicating that pigQ and pigW do not regulate carbapenem production (data not shown).

image

Figure 3. Phenotypes of secondary metabolite regulators. (A) Carbapenem, (B) cellulase and (C) pectate lyase production by LacA (WT), HSPIG67 (pigP), HSPIG43 (upstream of pigU), HSPIG17 (pigQ) and HSPIG62 (pigW). (D) Swimming motility of strains LacA (WT), HSPIG67 (pigP), HSPIG17 (pigQ) and HSPIG62 (pigW).

Download figure to PowerPoint

Production of BHL/HHL by all the Pig mutants was shown to be indistinguishable from the WT, when measured throughout growth (data not shown). Transcription of smaI was unaffected by mutations in pigP, pigQ or pigW verifying that these Pig regulators do not control BHL/HHL production (data not shown).

Pig regulatory mutants exhibit reduced pigA transcription

Many of the genes identified in the mutagenesis screen were predicted to act as transcriptional regulators (Table 1). Primer extension to qualitatively examine the pigA-O transcript was performed on RNA extracted from late exponential phase cultures of four mutants (Fig. 4). The pigA-O transcript was reduced in all mutants examined (pigP, pigQ, pigW and pigS) and was reflected in Pig levels in these mutants at the same stage of growth (Fig. 4). Importantly, the transcriptional start site of pigA-O was unaffected in the mutants examined.

image

Figure 4. Pig production and levels of the pigA-O transcript are reduced in four mutants. A. Pig levels of LacA (WT) and the four mutants HSPIG67 (pigP), HSPIG17 (pigQ), HSPIG62 (pigW) and HSPIG26 (pigS) after 9 h growth in LB. B. Primer extension analysis of the pigA-O transcript from the same four mutants in (A) examined after 9 h growth in LB. A negative control containing no RNA resulted in no primer extension product (data not shown).

Download figure to PowerPoint

QS regulates Pig production by controlling pigQ, pigR and rap expression

We were interested in investigating any interactions between the Pig regulators and the QS system, and the overall transcriptional hierarchy involved in regulation of secondary metabolite production in 39006. This, and the following section outline results that clarify the transcriptional network of regulators that control Pig production.

The possibility that QS was controlling the transcription of some of the Pig regulators and hence exerting its effect indirectly on Pig and Car was investigated. We examined the effects of QS on all Pig regulators that possessed active lacZ fusions, which included pigP, pigQ, pigR, pigS, pigV and pigX. In addition, we investigated whether the Pig and Car regulator, Rap, was modulated by QS. A rap::mini-Tn5lacZ1 strain (RAPL) was generated by transposon exchange mutagenesis. It was not possible to investigate the regulation of pigW or pigU because the mini-Tn5lacZ1 insertion in strain HSPIG62 was in the wrong orientation to generate a transcriptional fusion and the insertion in strain HSPIG43 was in an intergenic region near pigU. To characterize any impact of the smaI mutation on the Pig regulators, β-galactosidase activity from each of these ‘regulator’::lacZ fusion strains was assessed in the presence or absence of a functional smaI gene in early log, mid log, late log/transition and stationary phase (data not shown). Of the seven regulators examined three (pigR, rap and pigQ) had decreased transcription when smaI was disrupted (Fig. 5A–C) and the other four were unaffected (data not shown). The genes regulated by smaI included the novel adenylate cyclase pigR, the pleiotropic regulator rap and the two-component response regulator pigQ. Addition of 0.5 µM BHL at the beginning of growth ‘physiologically complemented’ the smaI mutation in these three strains by restoring transcription of the respective regulators (Fig. 5A–C). Therefore, the effect of the smaI mutation on the transcription of these regulatory genes results from the absence of BHL/HHL.

image

Figure 5. The smaIR locus regulates transcription of pigR, rap and pigQ throughout growth in LB. β-Galactosidase activity was measured from chromosomal pigR::lacZ (A), rap::lacZ (B) and pigQ::lacZ(C) fusions in an otherwise WT background (diamonds) or in a strain containing a chromosomal smaI::mini-Tn5Sm/Sp insertion (squares). BHL (0.5 µM final concentration) was also added at the beginning of growth to smaI mutant cultures (triangles). β-Galactosidase activity was measured from chromosomal pigR::lacZ (D), rap::lacZ (E) and pigQ::lacZ (F) fusions in strains carrying a WT smaIR locus (diamonds), a smaI mutation (squares), a smaR mutation (triangles) or mutations in both smaI and smaR (circles). Solid symbols and lines represent β-galactosidase assays whereas the open symbols and dashed lines represent the growth curves of the corresponding strains. Data shown are the means ± SD of at least three independent experiments.

Download figure to PowerPoint

SmaR is a repressor, the binding ability of which is inhibited by the presence of BHL/HHL (Fig. 2) (Slater et al., 2003). The role of SmaR in the QS regulation of pigR, rap and pigQ was determined by examining the effects of a smaR mutation or smaIR mutations on transcription of pigR, rap and pigQ. β-Galactosidase activity from lacZ fusions in pigR, rap and pigQ was measured throughout growth and compared in smaI, smaR and smaIR mutant backgrounds (Fig. 5D–F). Disruption of smaR in addition to a smaI mutation abolished the effect of the smaI mutation on transcription of pigQ, pigR and rap. Therefore, SmaR is required for the negative effect of a smaI mutation on transcription of pigQ, pigR and rap. This predicts that SmaR is directly or indirectly negatively regulating these genes and that smaI (via BHL/HHL production) inhibits and/or relieves this repression. This mechanism of smaIR transcriptional control is also observed for carA-H, carR and the pigA-O operon (Slater et al., 2003).

Therefore, QS controls Pig production, at least in part, by modulating transcription of pigQ, pigR and rap and can similarly affect Car production via control of rap (Thomson et al., 1997; Slater et al., 2003).

PigP regulates transcription of six Pig regulators

To examine the effect of the different Pig regulators on transcription of each other, alternative resistance markers were required to allow generation of the necessary double mutants. Using transposon exchange mutagenesis, mini-Tn5Sm/Sp insertions were generated in pigP, pigQ, pigW and pigX. A rap::mini-Tn5Sm/Sp mutant was already available (Crow, 2001). These mutations enabled an analysis of Pig regulators below pigP, pigQ, pigW, pigX and rap in the transcriptional hierarchy of Pig/Car production. β-Galactosidase activity from lacZ fusions in pigP, pigQ, pigR, pigS, pigV, pigX and rap was examined in early log, mid log, late log/transition and stationary phase in the presence or absence of pigP, pigQ, pigW, pigX and rap mutations (data not shown). Remarkably, a pigP mutation caused a decrease in transcription of three genes (pigQ, pigS and rap) (Fig. 6A–C) and an increase in expression of another three (pigR, pigV and pigX) (Fig. 6D–F). However, pigQ, pigW, pigX and rap did not control the transcription of any Pig regulators examined and therefore may be controlling pigA-O directly or via regulators not examined in this study (data not shown).

image

Figure 6. pigP regulates transcription of six other Pig regulators. β-Galactosidase activity was measured from chromosomal pigQ::lacZ (A), pigS::lacZ (B), rap::lacZ (C), pigR::lacZ (D), pigV::lacZ (E) and pigX::lacZ (F) fusions in an otherwise WT background (open bars) or in a strain containing a chromosomal pigP::mini-Tn5Sm/Sp disruption (solid bars). Solid symbols and lines represent the growth curves whereas the bars represent β-galactosidase activity. Data shown are the means ± SD of two independent experiments.

Download figure to PowerPoint

To determine if rap activates carA expression by regulating levels of CarR, transcription of a carR::lacZ fusion was analysed in the presence or absence of the rap::mini-Tn5Sm/Sp mutation. Transcription of carR was reduced by up to 3.5-fold throughout growth in the rap mutant (data not shown).

The putative pigQW two-component system was modulated by both QS and pigP and did not affect Car biosynthesis, and therefore it was picked for further study. Finally, the transcriptional hierarchy studies prompted further investigation of the pigP mutant because of the importance of PigP in control of other Pig regulators and the pleiotropic nature of the pigP mutant.

The PigQ/PigW two-component system regulates Pig by activating pigA-O transcription

Mutants HSPIG17 and HSPIG62 contain insertions in open reading frames (ORFs) predicted to encode proteins similar to ExpA and ExpS, respectively, from Ecc strain SCC3193, which constitute a GacAS family two-component system (Eriksson et al., 1998; Heeb and Haas, 2001). ExpS belongs to the class I family of hybrid histidine kinases, containing a response regulator receiver domain and a histidine phosphotransfer domain responsible for phosphorylation of the partner response regulator, ExpA (Dutta et al., 1999). Hence, it is likely that the genes disrupted in HSPIG17 and HSPIG62 encode a two-component signal transduction system, which regulates Pig; the genes were designated pigQ and pigW respectively.

Initial examination of the pigQ and pigW mutants demonstrated that both had reduced Pig, Pel, Cel and motility (Fig. 3). Pig production was measured throughout growth in LB for pigQ, pigW and a pigQW double mutant. For all the mutants, a decrease in Pig was observed when compared with the WT control (Fig. 7A). However, Pig production by the pigQ mutant was slightly lower than in the pigW mutant, which may be the result of phosphorylation of PigQ in the absence of PigW by non-cognate sensor kinases, or low molecular weight phosphodonors in a process known as ‘cross-talk’ (Bijlsma and Groisman, 2003). Furthermore, the pigQW double mutant had similar Pig levels to the pigW mutant, i.e. the effects of the mutations were not additive, consistent with both genes being partners of the same two-component system (Fig. 7A).

image

Figure 7. Pig production and pigA-O transcription are reduced by mutations in either gene, or both genes of the putative PigQW two-component system. (A) Pig production by WT (circles), HSPIG17 (pigQ) (diamonds), HSPIG62 (pigW) (triangles), 17S62L (pigQW) (squares) throughout growth in LB. β-Galactosidase activity was measured from a chromosomal pigA::lacZ fusion in an otherwise WT background (squares) or in a strains containing a chromosomal insertion in either pigQ::mini-Tn5Sm/Sp (B) or pigW::mini-Tn5Sm/Sp (C) (triangles). In all graphs solid symbols and lines represent either Pig or β-galactosidase assays whereas the open symbols and dashed lines represent the growth curves of the corresponding strains. Data shown are the means ± SD of at least three independent experiments.

Download figure to PowerPoint

To analyse whether Pig was controlled transcriptionally by pigQ and pigW, expression of a pigA::lacZ fusion was assessed in pigQ and pigW mutants. Mutation of pigQ or pigW caused a reduction in pigA::lacZ expression to approximately 75% compared with that of the WT, indicating that regulation of Pig is exerted at the transcriptional level (Fig. 7B and C). Primer extension studies (Fig. 4) indicated that levels of the pigA-O transcript were reduced in both pigQ and pigW mutants, supporting the results from the β-galactosidase assays. However, pigA-O mRNA levels were lower than predicted from the β-galactosidase assays, indicating the possibility of some degree of post-transcriptional regulation and/or altered transcript stability.

Sequence and genomic context of pigP, a novel pleiotropic regulator of secondary metabolites

Of the various regulatory mutants, the pigP mutant (HSPIG67) was of particular interest because of its pleiotropic nature and the lack of any functional information for similar proteins in the databases. In addition, transcription of six other Pig regulators was altered in a pigP mutant. DNA sequencing around pigP was performed to determine the genetic context of the pigP region, which is depicted schematically in Fig. 8A.

image

Figure 8. A. Genomic contexts of pigP homologues in Serratia marcescens Db11, Serratia sp. ATCC 39006 and Photorhabdus luminescens ssp. laumondii strain TT01. Predicted ORFs potentially encode the following proteins in Db11; orf1– similarities to acetyl transferase GNAT family, orf2– transcriptional regulator containing GerE-like DNA binding motif, orf3– homologue of cytosine permease codB, orf4pigP and plu1490 homologue, orf5– homologue of plu1489 and orf9, orf6E. coli yqcD and orf10 homologue, orf7– non-haemolytic phospholipase C; and in 39006 partial orf8– hypothetical protein (incomplete sequence), rhiE– homologue of rhamnogalacturonate lyase (rhiE) from Erwinia chrysanthemi, orf9– homologue of plu1489 and orf5, orf10E. coli yqcD and orf6 homologue and orf11– potential LysR-type regulator (incomplete sequence). ORF designations for TT01 refer to the annotated genome, the sequence shown covers positions 1 776 856–1 787 342. The putative ORF annotations for Db11 were based on sequence 994 125–1 001 669 from the unannotated genome assembly produced by the Serratia marcescens Sequencing Group at the Sanger Institute, available at http://www.sanger.ac.uk/Projects/S_marcescens/. B. Alignment of predicted proteins for PigP and its homologues identifying the putative helix-turn-helix (HTH) DNA binding domain at the N-terminus of the proteins. Asterisks denote amino acid conservation between the three predicted proteins. The alignment was performed using the program pileup in GCG.

Download figure to PowerPoint

The pigP ORF is predicted to encode a protein of 204 amino acids with an N-terminal XRE-like helix-turn-helix (HTH) DNA binding domain, which belongs to the xenobiotic response element family of transcriptional regulators (Fig. 8B). Homologues of PigP were identified in S. marcescens Db11 and P. luminescens ssp. laumondii strain TT01 (orf4 and plu1490, respectively) and in both cases the predicted product and location of the gene directly 3′ of pigP in 39006 (orf9) is also conserved (orf5 and plu1489 from S. marcescens Db11 and P. luminescens ssp. laumondii strain TT01 respectively) (Fig. 8A). In all strains, the homologues of pigP and orf9 in 39006 are in the same reading frame, such that the ATG initiation codon of orf9 directly follows on from the TAA stop codon of pigP. This indicates that pigP and the downstream gene might be translationally coupled. To confirm that the Pig phenotype of a pigP disruption was caused by the loss of functional PigP, and not a polar effect of the transposon, complementation was performed. Plasmid pTA10 containing WT pigP was able to restore Pig production in HSPIG67 (pigP::mini-Tn5lacZ1), whereas the vector control did not (data not shown). This indicated that pigP was responsible for the observed reduction in Pig.

The genomic contexts of the cognate genes encoding the PigP homologues demonstrated very little conservation of surrounding sequence except the genes homologous to orf9. Additionally, orf6 and orf10 from S. marcescens Db11 and 39006, respectively, are both predicted to encode GTP cyclohydrolase I enzymes potentially involved in folate biosynthesis. Excluding the pigP homologues in S. marcescens Db11 and P. luminescens ssp. laumondii strain TT01, the closest homologues of PigP are in a few Bordetella pertussis phages. Therefore, it is plausible that the pigP–orf9 region may have been acquired by horizontal gene transfer and that it originated from phage.

The predicted product of the gene immediately 5′ of pigP is homologous to RhiE from Erwinia chrysanthemi (Laatu and Condemine, 2003). RhiE has been shown to specifically breakdown the rhamnogalacturonan I form of the pectin backbone and is important for virulence of E. chrysanthemi in chicory leaves (Laatu and Condemine, 2003).

PigP regulates transcription of Car and Pig biosynthetic genes

The initial phenotypic analysis revealed that a pigP mutation results in a reduction in Pig (Table 1), and production of Car is abolished (Fig. 3). Levels of these antibiotics were assessed throughout growth in LB. There was no detectable Car at any stage of growth (Fig. 9A) and Pig was reduced to approximately 10% of WT levels (Fig. 9B). To examine the transcription of the Car and Pig biosynthetic genes, expression of chromosomal carA::lacZ and pigA::lacZ fusions was analysed in the presence or absence of a pigP::mini-Tn5Sm/Sp mutation (Fig. 9C and D). These lacZ fusion studies indicated that carA-H transcription was almost abolished in the pigP mutant and that pigA-O transcription was reduced in a pigP mutant to 75% of that in the WT (Fig. 9C and D). The differential reduction in pigA-O levels as assessed by β-galactosidase assays (Fig. 9D) and primer extension studies (Fig. 4) suggests that there may be issues of transcript stability and/or post-transcriptional regulation of the pigA-O transcript. To determine if PigP was regulating carR transcription and hence exerting control over the Car biosynthetic genes, β-galactosidase activity from a carR::lacZ fusion was assayed throughout growth in the presence of absence of a functional pigP gene. Transcription of carR was reduced to approximately 50% of WT levels in the pigP mutant (Fig. 9E). Therefore, pigP is regulating Car and Pig production by controlling the transcription of carR, carA and pigA.

image

Figure 9. pigP regulates Car and Pig production by controlling the expression of the biosynthetic genes, carA-H and pigA-O, and the activator carR. (A) Car and (B) Pig production by WT (diamonds) and PIG13S (pigP) (triangles) throughout growth in LB. β-Galactosidase activity was measured from chromosomal carA::lacZ (C) and pigA::lacZ (D) and carR::lacZ (E) fusions in otherwise WT backgrounds (diamonds) or in strains containing chromosomal pigP::mini-Tn5Sm/Sp disruptions (triangles). In all graphs solid symbols and lines represent Car, Pig or β-galactosidase assays whereas the open symbols and dashed lines represent the growth curves of the corresponding strains. Data shown are the means ± SD of at least three independent experiments.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Concluding remarks
  7. Experimental procedures
  8. Acknowledgements
  9. References

This study has significantly extended our previous investigations of the role of QS in the regulation of secondary metabolite production in Serratia sp. ATCC 39006 (39006). Specifically, biochemical evidence was obtained that supported our model of derepression of SmaR-repressed target genes in the presence of the cognate lactones BHL/HHL (Slater et al., 2003). In addition, a mutagenesis screen identified multiple new regulatory factors involved in Pig production. The transcriptional hierarchy of these putative Pig regulators and their interactions with the QS system have also been investigated.

In the absence of N-AHL, SmaR binds to, and represses expression from, the carA promoter

The current investigation demonstrated specific concentration-dependent DNA-binding of His6-SmaR, in the absence of N-AHLs, to a fragment of the carR-carA intergenic region containing the carA promoter (Fig. 2A). Addition of BHL completely abolished specific DNA binding of His6-SmaR to the carA promoter (Fig. 2B). Therefore, this supports our model that SmaR represses carA-H transcription by binding DNA upstream of the carA transcriptional start site. This repression is then relieved by the presence of the QS signal(s) BHL/HHL.

In contrast to our proposed mechanism, most examples of QS regulation involve activation of transcription of target genes by LuxR homologues in the presence of a threshold concentration of their cognate N-AHL. TraR from Agrobacterium tumefaciens and LuxR from Vibrio fisheri are known to function in this manner (Whitehead et al., 2001). There are only a small number of reported cases demonstrating LuxR homologues functioning via a derepression mechanism, such as EsaR from Pantoea stewartii ssp. stewartii (Minogue et al., 2002). Surface plasmon resonance studies, but not DNA-binding band shift assays, suggested a loss of DNA-binding affinity for dimerized EsaR in the presence of the cognate signal, 3-oxo-N-hexanoyl- l-homoserine lactone (OHHL) (Minogue et al., 2002). In addition, genetic data suggest that the SpnR protein from S. marcescens SS-1 may bind DNA and repress transcription in the absence of signal (Horng et al., 2002). A single shifted product is observed upon binding of SmaR to DNA containing the carA promoter (Fig. 2), which, based on the activity and structure of other LuxR homologues (such as TraR from A. tumefaciens (Vannini et al., 2002; Zhang et al., 2002), most likely represents dimerized SmaR.

In 39006, the carA promoter is regulated by two different LuxR homologues, SmaR and CarR (CarR39006), which act as a repressor and an activator respectively (Cox et al., 1998; Thomson et al., 2000; Slater et al., 2003). In the presence of BHL/HHL, the SmaR-mediated repression of carA and carR is relieved. This results in carR transcription, allowing CarR39006 to activate Car production. Because SmaR and CarR39006 are both LuxR homologues, it is conceivable that they might compete for the same binding site upstream of carA-H. However, it is  also  feasible  that  the  SmaR  binding  site  overlaps  the −35 and −10 promoter elements, which would result in direct exclusion of RNA polymerase (RNAP). In both cases, SmaR-mediated control of carA would be double edged, with SmaR controlling Car by regulating the quantity of CarR39006 and by limiting the access of essential factors (CarR39006 or RNAP) to the carA promoter. However, the exact location of the SmaR or CarR39006 binding sites at the carA promoter of 39006 remains to be determined.

Like 39006, Ecc also possesses a carA-H operon, and produces Car (Bainton et al., 1992; McGowan et al., 1996). In contrast to the N-AHL independence of CarR39006, Ecc CarR (CarREcc) binding upstream of carA-H and activation of transcription requires the presence of OHHL (McGowan et al., 1995; Welch et al., 2000). It is interesting when comparing the regulation of Car in 39006 with that in Ecc to note that, in 39006, all the regulators of carA-H examined to date regulate transcription of carR (SmaIR, PigP and Rap). Whereas, for example in Ecc, hor (homologue of rap) affects transcription of carA-H but does not control carR expression. CarREcc activity can be regulated post-translationally by binding OHHL. However, CarR39006 is N-AHL independent, so cannot be regulated in this manner. Thus, transcriptional control of CarR39006 may be more important than transcriptional control of CarREcc, which may be reflected in the number of regulatory factors that control its expression.

SmaIR regulates Pig and Car biosynthesis by controlling transcription of pigR, rap and pigQ

We could not demonstrate direct physical binding of SmaR to the carR or pigA promoters, although genetic evidence shows that the SmaIR QS system affects expression of these genes (Slater et al., 2003). One explanation is that SmaR may regulate expression of another Pig or Car regulatory gene, which prompted a search for other regulators involved in Pig, and potentially Car, production. A random mutagenesis identified eight putative regulators of Pig (Table 1), many of which were predicted to encode proteins showing similarities to known regulators or proteins containing putative DNA-binding domains. Indeed, for the four mutants examined, transcription of the Pig biosynthetic operon, pigA-O was reduced (Fig. 4).

Some of these Pig regulators could be involved in QS-mediated regulation of Pig, Car and exoenzymes. Mutation of smaI resulted in a SmaR-dependent reduction in transcription of three Pig regulators; pigR, rap and pigQ (Fig. 5). Based on sequence similarities, the predicted products of these genes are a putative novel adenylyl cyclase, a SlyA/MarR family transcriptional activator, and a putative response regulator of a GacAS family two-component system respectively. QS regulation of pigR, rap and pigQ is proposed to follow the same derepressive mechanism as other SmaIR-controlled genes such as carA-H, carR and pigA-O as discussed earlier (Slater et al., 2003). It is conceivable that SmaR repression of these three Pig regulators may be hierarchical, whereby QS controls expression of one gene, which is responsible for activation of other regulators. However, our studies on the transcriptional hierarchy of Pig regulation suggest that SmaR is either individually controlling expression of each of the three Pig regulators, or may be repressing an unidentified regulator that activates pigR, rap and pigQ (data not shown). PigQ and Rap affect Pig production by regulating transcription of pigA-O (Fig. 7) (Slater et al., 2003). However, it is less clear how PigR affects Pig levels. PigR is similar to the class IV adenylyl cyclase CyaB from Aeromonas hydrophila (Sismeiro et al., 1998; Baker and Kelly, 2004). CyaB is capable of producing cAMP, which is a common signalling molecule in both prokaryotes and eukaryotes. However, despite exhibiting adenylyl cyclase activity, the biological role of CyaB homologues is not transparent (Iyer and Aravind, 2002). The exact mechanism of PigR-mediated control of Pig production requires further study.

We demonstrated that none of the Pig regulators examined in this study controlled secondary metabolism by modulating the levels of BHL/HHL. It is interesting that HSPIG43 was not detectably affected in BHL/HHL production as mutations in the Ecc pigU homologue, hexA, caused slightly increased OHHL levels throughout growth, and hexA in trans decreased OHHL levels in WT Ecc (Harris et al., 1998). The GacA homologues of PigQ in some Pseudomonas species and S. plymuthica IC1270 have also been shown to upregulate synthesis of N-AHLs (Reimmann et al., 1997; Kitten et al., 1998; Chancey et al., 1999; Ovadis et al., 2004). However, N-AHL levels and smaI transcription were unaffected by mutations in pigQ or pigW, demonstrating that this putative two-component system does not regulate N-AHL production in 39006. Therefore, PigQW differs from other GacAS systems, because it does not control N-AHL production, and is itself regulated by QS (via control of pigQ transcription).

Our model for the QS-mediated control of secondary metabolite production and its interactions with the newly identified regulators is summarized in Fig. 10A. Interestingly, transcription of most SmaIR-regulated genes examined is not elevated in smaR mutants. Therefore, other growth phase-dependent factors may be required for the timing and level of expression in addition to the QS system (e.g. PigP; see below). This would allow multiple inputs to be assessed before the cell commits to production of secondary metabolites, or other phenotypes.

image

Figure 10. Proposed mechanistic model of regulation of Pig and Car production. A. At low cell density/early growth phase, transcription of carA-H, pigA-O and the regulators carR, pigR, pigQ and rap is low, presumably by either direct (binding to carA) or indirect repression of these genes by SmaR. As cell numbers increase, the concomitant increase in BHL/HHL, causes SmaR-mediated repression to be alleviated. This results in production of CarR, PigQ, Rap and PigR. PigQ is potentially phosphorylated by its sensor kinase partner (PigW) in response to other unknown signals. Active PigQ then activates pigA-O transcription either directly or indirectly. Rap activates pigA-O, carR and carA-H transcription. Rap indirectly activates carA-H via modulation of carR expression, but may also act directly to control carA-H transcription. Production of PigR affects production of Pig but it is not known if this control is transcriptional (dashed line). B. Proposed hierarchical model of regulation of Pig and Car production in 39006 combining results from previously published work and the current study. PigP differentially modulates six regulators of Pig (pigQ, pigR, pigS, pigV, pigX and rap) and the Pig biosynthetic genes, pigA-O. In addition, PigP regulates transcription of rap, carR and carA-H, all involved in Car production. Any signal(s) that regulate production or activity of PigP are currently unknown. PigP activates pigA-O indirectly via control of other Pig regulators, but may also directly activate the biosynthetic cluster. Likewise, PigP activation of carA-H may be direct in addition to the indirect regulation of both rap and carR expression. SmaI produces BHL/HHL, which accumulates at high cell density and inactivates SmaR, which is a repressor of pigQ, pigR, rap, carR, carA-H and pigA-O (also see panel A). Transcription of pigQ, pigR and rap is controlled by both PigP and QS therefore the relative levels of these regulators reflect multiple signals including cell density. The Rap protein represents a hub where information from both QS and PigP is channelled into regulation of both Pig and Car production. PigU is predicted to be a transcriptional repressor of Pig and Car production. Secondary metabolite biosynthesis also responds to phosphate limitation ([Pi]) via PstSCAB resulting in constitutive activity of the PhoBR two-component system. PhoB is predicted to directly activate smaI and pigA-O transcription, and therefore affects Pig by both QS-dependent and QS-independent mechanisms (Slater et al., 2003). Transcriptional activation is represented by bold arrow heads and solid lines and repression is depicted with flat arrow heads. Dashed lines indicate cases where activation or repression has not been demonstrated transcriptionally. Open arrow heads and dashed lines represent-non-transcriptional interactions.

Download figure to PowerPoint

A novel master regulator, PigP, controls secondary metabolite production by modulating expression of multiple newly identified regulatory genes

A pleiotropic regulator, PigP, was identified and found to be responsible for modulating transcription of six other Pig regulators (Fig. 6). The predicted PigP protein and its homologues define a new family of regulators containing putative N-terminal HTH DNA-binding domains, suggesting that they regulate transcription by binding DNA (Fig. 8B). Indeed, PigP was shown to be a transcriptional regulator, which affects Pig and Car production by activating (both indirectly and possibly directly) transcription of carA-H and pigA-O (Fig. 9). P. luminescens ssp. laumondii strain TT01 also produces Car so it would be interesting to determine whether it is regulated by the TT01 PigP homologue. In addition, transcription of the N-AHL-independent LuxR homologue, CarR39006, is activated by PigP (Fig. 9C), which would lead to increased carA-H expression. PigP activates rap transcription, which in turn promotes carR expression. Therefore, the effect of PigP on carR and carA expression may be indirect, via the activation of rap transcription. Interestingly, mutations in rap or pigP caused 70% and 50% reductions, respectively, in carR transcription while almost abolishing carA expression. It is not clear whether these observed reductions in carR transcription are sufficient to explain the decreased expression of carA, or whether other regulators are involved.

PigP-mediated regulation of pigA-O involved control of at least six other Pig regulators (pigQ, pigR, pigS, pigV, pigX and rap). PigP activated transcription of pigA-O by increasing expression of rap, pigQ and pigS, the products of which activate transcription of the biosynthetic genes (Figs 4 and 7) (Slater et al., 2003). The regulation of secondary metabolites by Rap and the two-component response regulator, PigQ, is discussed later. The gene pigS, is linked to the 5′ end of the pigA-O cluster (2.2 kb from pigA) and its predicted product is similar to HlyU from Vibrio cholerae (Williams and Manning, 1991; Williams et al., 1993) and NolR from Rhizobium meliloti (Kondorosi et al., 1991; Cren et al., 1995). PigS contains a HTH domain and is similar to proteins of the ArsR/SmtB family, the DNA binding activities of which respond to specific divalent metal ions (Busenlehner et al., 2003). In addition, PigP repressed expression of pigX, which is a repressor of pigA-O transcription (L. Everson, unpubl. data) and is predicted to encode a homologue of YhdA from Escherichia coli (a putative diguanylate cyclase/phosphodiesterase). Therefore, Pig biosynthesis is regulated by PigP-mediated repression of at least one repressor (pigX) or activation of other transcriptional activators (rap, pigQ and pigS) of Pig. Finally, PigP repressed expression of the putative novel adenylate cyclase, pigR and pigV, both of which appear to positively affect Pig production. PigV is similar to a conserved hypothetical inner membrane protein in E. coli denoted YgfX. From the sequence data available, pigV appears to be the final gene in an operon, which also encodes a protein similar to YgfY from E. coli located 5′ of the pigV ORF. It is not known how PigR and PigV control Pig levels.

We hypothesize that the Pig regulators respond to different physiological cues and therefore, by differentially modulating the levels of these regulators, PigP has some control over which signals are funnelled into Pig biosynthesis. The position of PigP in the regulatory hierarchy of Car and Pig is represented in Fig. 10B, demonstrating both the overlaps with, and independence from, the QS regulon.

PigQW, a two-component system of the GacAS family, regulates Pig production

Sequence, phenotypic and transcriptional data indicated that PigQ and PigW are putative partners in a two-component signal transduction system and members of the GacAS family, which regulate numerous phenotypes involved in pathogenesis and symbiosis in bacteria (Heeb and Haas, 2001). This study has demonstrated that transcription of pigQ responds to cell density (via SmaIR) (Fig. 5), and to unknown physiological cues via PigP (Fig. 6). Presumably, PigW (the putative sensor kinase) regulates the phosphorylation state of PigQ (the putative response regulator), and therefore the activity of PigQ will be responsive to environmental signals that are detected by PigW. There is evidence of a solvent-extractable signalling molecule in Pseudomonas fluorescens CHA0 that is detected by the PigW homologue, GacS (Heeb et al., 2002; Zuber et al., 2003).

PigQW regulates Pig production by activating transcription of pigA-O (Fig. 7) but does not affect expression of any of the other Pig regulators examined. Therefore, PigQ might activate transcription of pigA-O directly. However, this two-component system may still be activating Pig biosynthesis indirectly via an, as yet unidentified, regulator. The ExpAS two-component system in Ecc has been shown to regulate gene expression via the RsmA/rsmB system (Heeb and Haas, 2001; Hyytiainen et al., 2001), so it is possible that a regulatory RNA similar to rsmB is an intermediate in the PigQ-mediated Pig regulation in 39006. The position of PigQW in Pig regulation with regard to the QS regulon and the overall regulatory hierarchy is summarized schematically in Fig. 10.

Rap, regulator of antibiotic and pigment, deciphers multiple regulatory inputs in the control of secondary metabolite production

This study has shown that transcription of rap was controlled by both SmaIR and PigP. In addition, rap controlled production of Car by activating expression of carR. Previously, transcription of smaI was shown to be unaffected in a rap mutant (Slater et al., 2003). Regulation of rap by the QS system contrasts with that of hor (homologue of rap) in Ecc, which is unaffected by mutations in carI (N-AHL synthase) (McGowan et al., 2005). Furthermore, inactivation of hor had no effect on transcription of carI or carR (Thomson et al., 1997). Rap and Hor belong to the MarR/SlyA family of transcriptional regulators, some of which have been structurally characterized (Alekshun et al., 2001; Wu et al., 2003). This family of regulators associate as dimers, with each subunit containing a winged helix (WH) DNA-binding motif. MarR can be inhibited by salicylic acid, which binds near the WH domain. The position of Rap in the regulation of secondary metabolites with regard to both the QS regulon and the overall regulatory hierarchy is summarized schematically in Fig. 10.

Concluding remarks

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Concluding remarks
  7. Experimental procedures
  8. Acknowledgements
  9. References

This study has increased greatly our understanding of the regulation of secondary metabolite production in 39006. It has revealed a complex overlapping network of regulators involved in biosynthesis of Pig and Car. Based on information acquired in this study and from our previous reports (Thomson et al., 1997; Cox et al., 1998; Thomson et al., 2000; Slater et al., 2003), we propose a new integrated model of how Pig and Car are regulated in response to cell density (via QS), phosphate limitation (Slater et al., 2003) and other environmental inputs presumed to be sensed by PigP and the histidine kinase (PigW) of the two-component system (Fig. 10). Further functional analysis of the component parts of the regulatory cascades involved and, ultimately, a holistic integration of ‘parallel pathways’ into a neural network of dual antibiotic regulation are the aims of future studies.

We have shown Pig biosynthesis requires the activity of 14 or 15 enzymes and therefore, the pigment is assumed to be bioenergetically ‘expensive’ to produce (Harris et al., 2004; Williamson et al., 2005). We have now demonstrated that there are at least as many additional genes encoding regulators involved in its control (Fig. 10). This begs the question of why 39006 produces Pig and ‘invests’ so much in its regulatory control. Presumably, the answer lies in the ‘real’ physiological role of this secondary metabolite in 39006, which, at the current time, remains mysterious.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Concluding remarks
  7. Experimental procedures
  8. Acknowledgements
  9. References

Bacterial strains, plasmids, phage and culture conditions

Bacterial strains and plasmids are listed in Table 2. Serratia sp. ATCC 39006 strains were grown at 30°C and E. coli strains were grown at 37°C in LB (5 g l−1 yeast extract, 10 g l−1 bacto tryptone and 5 g l−1 NaCl) or minimal media [0.1% w/v (NH4)2SO4, 0.41 mM MgSO4, 0.2% w/v glucose, 40 mM K2HPO4, 14.7 mM KH2PO4, pH 6.9–7.1] at 300 r.p.m., or on LB agar supplemented with 1.5% (w/v) agar (LBA) (Miller, 1972). Bacterial growth (OD600) was measured in a Unicam Heλios spectrophotometer at 600 nm. When required, LB and minimal media were supplemented with antibiotics at the following concentrations; kanamycin 50 µg ml−1 (Km), spectinomycin 50 µg ml−1 (Sp), streptomycin 50 µg ml−1 (Sm), ampicillin 50 µg ml−1 (Ap) and chloramphenicol 25 µg ml−1 (Cm). The generalized transducing phage φOT8 was used for transduction of chromosomal mutations as previously described (Thomson et al., 2000).

Table 2.  Bacterial strains, plasmids and phages used in this study.
Strain/plasmidGenotype/phenotypeReference
Escherichia coli
 DH5αF, φ80ΔdlacZM15, Δ(lacZYA–argF)U169, endA1, recA1, hsdR17 (rKmK+), deoR, thi-1, supE44, λ, gyrA96, relA1Gibco/BRL
 ESSβ-lactam super sensitive indicator strain(Bainton et al., 1992)
 SM10 λpirthi-1, thr, leu, tonA, lacY, supE, recA::RP4-2-Tc::Mu, λpir, KmR (de Lorenzo et al., 1990)
 S17-1 λpir recA, pro,hsdR, recA::RP4-2-Tc::Mu, λpir, TmpR, SpR, SmR de Lorenzo et al. (1990)
Serratia
 ATCC 39006Wild type (Car+, Pig+)(Bycroft et al., 1987)
 LacA (WT)Lac derivative of ATCC 39006, made by EMS mutagenesis(Thomson et al., 2000)
Mutants derived from LacA
 13S17L pigP::mini-Tn5Sm/Sp, pigQ::mini-Tn5lacZ1, SpR, KmR; derivative of HSPIG17 following transduction using φOT8 grown on strain PIG13SThis study
 13S23L pigP::mini-Tn5Sm/Sp, pigR::mini-Tn5lacZ1, SpR, KmR; derivative of HSPIG23 following transduction using φOT8 grown on strain PIG13SThis study
 13S26L pigP::mini-Tn5Sm/Sp, pigS::mini-Tn5lacZ1, SpR, KmR; derivative of HSPIG26 following transduction using φOT8 grown on strain PIG13SThis study
 13S46L pigP::mini-Tn5Sm/Sp, pigV::mini-Tn5lacZ1, SpR, KmR; derivative of HSPIG46 following transduction using φOT8 grown on strain PIG13SThis study
 13SCAL pigP::mini-Tn5Sm/Sp, carA::mini-Tn5lacZ1, SpR, KmR; derivative of MCA54 following transduction using φOT8 grown on strain PIG13SThis study
 13SCRL pigP::mini-Tn5Sm/Sp, carR::mini-Tn5lacZ1, SpR, KmR; derivative of MCR14 following transduction using φOT8 grown on strain PIG13SThis study
 13SRAPL pigP::mini-Tn5Sm/Sp, rap::mini-Tn5lacZ1, SpR, KmR; derivative of RAPL following transduction using φOT8 grown on strain PIG13SThis study
 13SSIL pigP::mini-Tn5Sm/Sp, smaI::mini-Tn5lacZ1, SpR, KmR; derivative of LC13 following transduction using φOT8 grown on strain PIG13SThis study
 13SSO4L pigP::mini-Tn5Sm/Sp, pigX::mini-Tn5lacZ1, SpR, KmR; derivative of ROP4 following transduction using φOT8 grown on strain PIG13SThis study
 17S23L pigQ::mini-Tn5Sm/Sp, pigR::mini-Tn5lacZ1, SpR, KmR; derivative of HSPIG23 following transduction using φOT8 grown on strain PIG17SThis study
 17S26L pigQ::mini-Tn5Sm/Sp, pigS::mini-Tn5lacZ1, SpR, KmR; derivative of HSPIG26 following transduction using φOT8 grown on strain PIG17SThis study
 17S46L pigQ::mini-Tn5Sm/Sp, pigV::mini-Tn5lacZ1, SpR, KmR; derivative of HSPIG46 following transduction using φOT8 grown on strain PIG17SThis study
 17S62L pigQ::mini-Tn5Sm/Sp, pigW::mini-Tn5lacZ1, SpR, KmR; derivative of HSPIG62 following transduction using φOT8 grown on strain PIG17SThis study
 17S67L pigQ::mini-Tn5Sm/Sp, pigP::mini-Tn5lacZ1, SpR, KmR; derivative of HSPIG67 following transduction using φOT8 grown on strain PIG17SThis study
 17SCAL pigQ::mini-Tn5Sm/Sp, carA::mini-Tn5lacZ1, SpR, KmR; derivative of MCA54 following transduction using φOT8 grown on strain PIG17SThis study
 17SCRL pigQ::mini-Tn5Sm/Sp, carR::mini-Tn5lacZ1, SpR, KmR; derivative of MCR14 following transduction using φOT8 grown on strain PIG17SThis study
 17SRAPL pigQ::mini-Tn5Sm/Sp, rap::mini-Tn5lacZ1, SpR, KmR; derivative of RAPL following transduction using φOT8 grown on strain PIG17SThis study
 17SSIL pigQ::mini-Tn5Sm/Sp, smaI::mini-Tn5lacZ1, SpR, KmR; derivative of LC13 following transduction using φOT8 grown on strain PIG17SThis study
 17SSO4L pigQ::mini-Tn5Sm/Sp, pigX::mini-Tn5lacZ1, SpR, KmR; derivative of ROP4 following transduction using φOT8 grown on strain PIG17SThis study
 62S17L pigW::mini-Tn5Sm/Sp, pigQ::mini-Tn5lacZ1, SpR, KmR; derivative of HSPIG17 following transduction using φOT8 grown on strain PIG62SThis study
 62S23L pigW::mini-Tn5Sm/Sp, pigR::mini-Tn5lacZ1, SpR, KmR; derivative of HSPIG23 following transduction using φOT8 grown on strain PIG62SThis study
 62S26L pigW::mini-Tn5Sm/Sp, pigS::mini-Tn5lacZ1, SpR, KmR; derivative of HSPIG26 following transduction using φOT8 grown on strain PIG62SThis study
 62S46L pigW::mini-Tn5Sm/Sp, pigV::mini-Tn5lacZ1, SpR, KmR; derivative of HSPIG46 following transduction using φOT8 grown on strain PIG62SThis study
 62S67L pigW::mini-Tn5Sm/Sp, pigP::mini-Tn5lacZ1, SpR, KmR; derivative of HSPIG67 following transduction using φOT8 grown on strain PIG62SThis study
 62SCAL pigW::mini-Tn5Sm/Sp, carA::mini-Tn5lacZ1, SpR, KmR; derivative of MCA54 following transduction using φOT8 grown on strain PIG62SThis study
 62SCRL pigW::mini-Tn5Sm/Sp, carR::mini-Tn5lacZ1, SpR, KmR; derivative of MCR14 following transduction using φOT8 grown on strain PIG62SThis study
 62SPAL pigW::mini-Tn5Sm/Sp, pigA::mini-Tn5lacZ1, SpR, KmR; derivative of MCP2L following transduction using φOT8 grown on strain PIG62SThis study
 62SRAPL pigW::mini-Tn5Sm/Sp, rap::mini-Tn5lacZ1, SpR, KmR; derivative of MCP2L following transduction using φOT8 grown on strain PIG62SThis study
 62SSIL pigW::mini-Tn5Sm/Sp, smaI::mini-Tn5lacZ1, SpR, KmR; derivative of LC13 following transduction using φOT8 grown on strain PIG62SThis study
 62SSO4L pigW::mini-Tn5Sm/Sp, pigX::mini-Tn5lacZ1, SpR, KmR; derivative of ROP4 following transduction using φOT8 grown on strain PIG62SThis study
 HSPIG17 pigQ::mini-Tn5lacZ1, KmR; random transposon mutantThis study
 HSPIG23 pigR::mini-Tn5lacZ1, KmR; random transposon mutantThis study
 HSPIG26 pigS::mini-Tn5lacZ1, KmR; random transposon mutantThis study
 HSPIG43 pigUpro::mini-Tn5lacZ1, KmR; random transposon mutantThis study
 HSPIG46 pigV::mini-Tn5lacZ1, KmR; random transposon mutantThis study
 HSPIG62 pigW::mini-Tn5lacZ1, KmR (not an active fusion); random transposon mutantThis study
 HSPIG66 pigXpro::mini-Tn5lacZ1, KmR; random transposon mutantThis study
 HSPIG67 pigP::mini-Tn5lacZ1, KmR; random transposon mutantThis study
 HSPIG17::R4 pigX::mini-Tn5Sm/Sp, pigQ::mini-Tn5lacZ1, SpR, KmR; derivative of HSPIG17 following transduction using φOT8 grown on strain ROP4SThis study
 HSPIG23::R4 pigX::mini-Tn5Sm/Sp, pigR::mini-Tn5lacZ1, SpR, KmR; derivative of HSPIG23 following transduction using φOT8 grown on strain ROP4SThis study
 HSPIG26::R4 pigX::mini-Tn5Sm/Sp, pigS::mini-Tn5lacZ1, SpR, KmR; derivative of HSPIG26 following transduction using φOT8 grown on strain ROP4SThis study
 HSPIG46::R4 pigX::mini-Tn5Sm/Sp, pigV::mini-Tn5lacZ1, SpR, KmR; derivative of HSPIG46 following transduction using φOT8 grown on strain ROP4SThis study
 HSPIG67::R4 pigX::mini-Tn5Sm/Sp, pigP::mini-Tn5lacZ1, SpR, KmR; derivative of HSPIG67 following transduction using φOT8 grown on strain ROP4SThis study
 IR17L smaR::cat, smaI::mini-Tn5Sm/Sp, pigQ::mini-Tn5lacZ1, KmR, CmR, SpR; derivative of QSL2 following transduction using φOT8 grown on strain HSPIG17This study
 IR23L smaR::cat, smaI::mini-Tn5Sm/Sp, pigR::mini-Tn5lacZ1, KmR, CmR, SpR; derivative of QSL2 following transduction using φOT8 grown on strain HSPIG23This study
 IRRAPL smaR::cat, smaI::mini-Tn5Sm/Sp, rap::mini-Tn5lacZ1, KmR, CmR, SpR; derivative of QSL2 following transduction using φOT8 grown on strain RAPLThis study
 IS17L smaI::mini-Tn5Sm/Sp, pigQ::mini-Tn5lacZ1, SpR, KmR; derivative of HSPIG17 following transduction using φOT8 grown on strain LISThis study
 IS23L smaI::mini-Tn5Sm/Sp, pigR::mini-Tn5lacZ1, SpR, KmR; derivative of HSPIG23 following transduction using φOT8 grown on strain LISThis study
 IS26L smaI::mini-Tn5Sm/Sp, pigS::mini-Tn5lacZ1, SpR, KmR; derivative of HSPIG26 following transduction using φOT8 grown on strain LISThis study
 IS46L smaI::mini-Tn5Sm/Sp, pigV::mini-Tn5lacZ1, SpR, KmR; derivative of HSPIG46 following transduction using φOT8 grown on strain LISThis study
 IS67L smaI::mini-Tn5Sm/Sp, pigP::mini-Tn5lacZ1, SpR, KmR; derivative of HSPIG67 following transduction using φOT8 grown on strain LISThis study
 ISRAPL smaI::mini-Tn5Sm/Sp, rap::mini-Tn5lacZ1, SpR, KmR; derivative of RAPL following transduction using φOT8 grown on strain LISThis study
 ISTSO4 smaI::mini-Tn5Sm/Sp, pigX::mini-Tn5lacZ1, SpR, KmR; derivative of ROP4 following transduction using φOT8 grown on strain LISThis study
 LC13 smaI::mini-Tn5lacZ1, KmR(Thomson et al., 2000)
 LIS smaI::mini-Tn5Sm/Sp, SpR(Thomson et al., 2000)
 MCA54 carA::mini-Tn5lacZ1, KmR(Thomson et al., 2000)
 MCP2L pigA::mini-Tn5lacZ1, KmR(Slater et al., 2003)
 MCR14 carR::mini-Tn5lacZ1, KmR(Slater et al., 2003)
 MCR2000 smaR::cat, CmR(Slater et al., 2003)
 PIG13S pigP::mini-Tn5Sm/Sp, SpR; derivative of HSPIG67, constructed by transposon exchangeThis study
 PIG13SA pigP::mini-Tn5Sm/Sp, pigA::mini-Tn5lacZ1, SpR, KmR; derivative of MCP2L following transduction using φOT8 grown on strain PIG13SThis study
 PIG17S pigQ::mini-Tn5Sm/Sp, SpR; derivative of HSPIG17, constructed by transposon exchangeThis study
 PIG17SA pigQ::mini-Tn5Sm/Sp, pigA::mini-Tn5lacZ1, SpR, KmR; derivative of MCP2L following transduction using φOT8 grown on strain PIG13SThis study
 PIG62S pigW::mini-Tn5Sm/Sp, SpR; derivative of HSPIG62, constructed by transposon exchangeThis study
 QSL2 smaR::cat, smaI::mini-Tn5Sm/Sp, CmR, SpR(Slater et al., 2003)
 R17L smaR::cat, pigQ::mini-Tn5lacZ1, KmR, CmR; derivative of MCR2000 following transduction using φOT8 grown on strain HSPIG17This study
 R23L smaR::cat, pigR::mini-Tn5lacZ1, KmR, CmR; derivative of MCR2000 following transduction using φOT8 grown on strain HSPIG23This study
 RAP17L rap::mini-Tn5Sm/Sp, pigQ::mini-Tn5lacZ1, SpR, KmR; derivative of HSPIG17 following transduction using φOT8 grown on strain RAPSThis study
 RAP23L rap::mini-Tn5Sm/Sp, pigR::mini-Tn5lacZ1, SpR, KmR; derivative of HSPIG23 following transduction using φOT8 grown on strain RAPSThis study
 RAP26L rap::mini-Tn5Sm/Sp, pigS::mini-Tn5lacZ1, SpR, KmR; derivative of HSPIG26 following transduction using φOT8 grown on strain RAPSThis study
 RAP46L rap::mini-Tn5Sm/Sp, pigV::mini-Tn5lacZ1, SpR, KmR; derivative of HSPIG46 following transduction using φOT8 grown on strain RAPSThis study
 RAP67L rap::mini-Tn5Sm/Sp, pigP::mini-Tn5lacZ1, SpR, KmR; derivative of HSPIG67 following transduction using φOT8 grown on strain RAPSThis study
 RAPCRL rap::mini-Tn5Sm/Sp, carR::mini-Tn5lacZ1, SpR, KmR; derivative of MCR14 following transduction using φOT8 grown on strain RAPSThis study
 RAPL rap::mini-Tn5lacZ1, KmR; derivative of RAPS, constructed by transposon exchangeThis study
 RAPS rap::mini-Tn5Sm/Sp, SpR; random transposon mutantThis study
 ROP4 pigX::mini-Tn5lacZ1, KmR; random transposon mutantThis study
 ROP4S pigX::mini-Tn5Sm/Sp, SpR; derivative of ROP4, constructed by transposon exchangeThis study
 RRAPL smaR::cat, rap::mini-Tn5lacZ1, KmR, CmR; derivative of MCR2000 following transduction using φOT8 grown on strain RAPLThis study
 RTSO4L rap::mini-Tn5Sm/Sp, pigX::mini-Tn5lacZ1, SpR, KmR; derivative of ROP4 following transduction using φOT8 grown on strain RAPSThis study
Phage
 φOT8 Serratia generalized transducing phage(Crow et al. unpubl.)
Plasmids
 pBluescript II KS +Cloning vector, ColE1 replicon, ApRStratagene
 pQE-80 LCloning vector for N-terminal hexahistidine proteins, ApRQIAGEN
 pQE80SmaR smaR in pQE80L for His6-SmaR expressionThis study
 pTA10 pigP cloned into pBluescript II KS + for complementation, ApRThis study
 pTON245pHCP19 vector containing 24 kb including the Pig cluster(Harris et al., 2004)
 pTON28pSF6 (pNRT1 carR subclone)(Cox et al., 1998)
 pUC18Cloning vector, ColE1 replicon, ApR(Yanisch-Perron et al., 1985)
 pUC19Cloning vector, ColE1 replicon, ApR(Yanisch-Perron et al., 1985)
 pUC19-28pUC19 containing a 2.8 kb BamHI fragment from pTON28 containing carR and part of carAThis study
 pUCcarAproContains carA promoter on BamHI/HindIII fragment in pUC18, ApRThis study
 pUCcarRproContains carR promoter on BamHI/HindIII fragment in pUC18, ApRThis study
 pUCpigAproContains pigA promoter on BamHI/HindIII fragment in pUC18, ApRThis study
 pUTmini-Tn5lacZ1Delivery plasmid for mini-Tn5lacZ1, ApR, KmR de Lorenzo et al. (1990)
 pUTmini-Tn5Sm/SpDelivery plasmid for mini-Tn5Sm/Sp, ApR, SpR de Lorenzo et al. (1990)

DNA manipulations and genetic constructions

All molecular biology techniques, unless stated otherwise, were performed by standard methods (Sambrook et al., 1989). Oligonucleotide primers were obtained from Sigma Genosys. DNA sequencing was performed at the DNA Sequencing Facility, Department of Biochemistry, University of Cambridge. Nucleotide sequence data were analysed using GCG (Genetics Computer Group, University of Wisconsin) and compared with GenBank DNA or non-redundant protein sequence databases using BLAST (Altschul et al., 1997).

Random transposon mutagenesis of Serratia 39006 strain LacA was performed by conjugation with either E. coli S17-1 λpir harbouring plasmid pUTmini-Tn5lacZ1 or E. coli SM10 λpir harbouring plasmid pUTmini-Tn5Sm/Sp as described previously (Thomson et al., 2000). Mutants defective in Pig production were identified visually and their mutations moved by φOT8 transduction into a ‘clean’ LacA background to ensure that the phenotypes were associated with single transposition events. Genomic DNA from each mutant was prepared and subjected to Southern blot and hybridization analysis using the pig cluster as a probe (pTON245 labelled with DIG using the Random Primed DNA Labelling Kit, Roche). The precise sites of transposon insertions were determined by ‘single primer site PCR’ followed by DNA sequence analysis across the transposon/chromosomal junction. Briefly, genomic DNA was digested with up to three different combinations of restriction endonucleases (EcoRV/HindIII, SacII/PstI and EcoRI/BamHI) and after de-phosphorylation with shrimp alkaline phosphatase was ligated to pBluescript® II KS + (Stratagene) cut with the same enzymes. One microlitre of each ligation was then used as a template in PCR with pBluescript® II KS + primers (T3 and T7) in combination with primers that hybridized to the ends of mini-Tn5lacZ1 (LL-68 and LR-68) (all primers used in this study are listed in Table S1; Supplementary material). Specific PCR products were identified by gel electrophoresis, excised and purified by QIAquick® extraction (Qiagen®) and the sequence determined using primers LL-68 and LR-68 (which face out of the left and right hand-sides of mini-Tn5lacZ1, respectively) and T3 and T7 where appropriate. At least 600 bp of genomic DNA sequence flanking the insertions was determined for each mutant.

To enable further sequencing the Pig regulatory genes were cloned. Clones carrying wild-type copies of the regulatory genes identified above were selected by colony hybridization from one of three subgenomic libraries of 39006 DNA as outlined below. Southern blot and hybridization analysis with a probe homologous to mini-Tn5lacZ1 had previously revealed that the disrupted genes of mutants resided on 2–4 kb SphI fragments (HSPIG17, HSPIG26), 4–7 kb SphI fragments (HSPIG62) and on SphI fragments larger than 7 kb (HSPIG13, HSPIG23, HSPIG46, HSPIG66). Three subgenomic libraries of 39006 DNA digested with SphI were prepared (fragment sizes: 2–4 kb, 4–7 kb and 7 + kb) in pUC19 and maintained in E. coli DH5a. DIG-dUTP probes were generated by PCR using the following primer pairs HS1/HS2 (HSPIG13), HS18/HS6 (HSPIG17), HS7/HS8 (HSPIG23), HS9/HS10 (HSPIG26), HS19/HS12 (HSPIG43), HS20/HS14 (HSPIG46), HS21/HS16 (HSPIG66), and HS26/HS27 (HSPIG62) according to the manufacturer's instructions (Roche). Positive clones were identified, from colony blots using chemiluminescent detection, for strains HSPIG13, HSPIG17, HSPIG26, HSPIG46 and HSPIG62.

To enable complementation of HSPIG67 (pigP::mini-Tn5lacZ1), plasmid pTA10 was constructed. The pigP gene and approximately 300 bp of sequence 5′ of the translation initiation codon was amplified by PCR using primers PF24 and PF3, cloned in to the XhoI/PstI sites in pBluescript® II KS + and confirmed by DNA sequencing.

For generation of constructs carrying promoters for band-shift assays the promoter regions of carA, carR and pigA were amplified by PCR, using the primers shown in Table S1 (Supplementary material), and cloned into the BamHI/HindIII sites of pUC18. The resulting constructs, pUCcarApro, pUCcarRpro and pUCpigApro, respectively, were confirmed by DNA sequence analysis.

To construct a SmaR expression plasmid the wild-type smaR gene was amplified by PCR using the primers HS70 and HS71 and cloned into the BamHI/PstI sites of the His6 tag expression vector pQE-80L (Qiagen®) yielding construct pQE80SmaR. As confirmed by DNA sequencing, this construction maintained the wild-type smaR gene but incorporated a His6 tag such that, when expressed, the SmaR protein (His6-SmaR) carried the 11 amino acid N-terminal extension MRGSHHHHHHGS. The functionality of His6-SmaR was checked by phenotypic analysis of the QSL2 (smaIR) strain carrying pQE80SmaR. QSL2/pQE80SmaR cells were non-pigmented as expected if a functional SmaR protein was present, effectively producing a smaI phenocopy strain.

Expression and purification of His6-SmaR

His6-SmaR protein was purified from E. coli strain DH5α/pQE80SmaR. A 10 ml culture of DH5α/pQE80SmaR which had been grown overnight at 37°C in LB medium containing 100 µg ml−1 ampicillin was used to inoculate 500 ml of the same medium, pre-equilibrated at 37°C, which was then grown at 300 r.p.m. Upon reaching an optical density between 0.5 and 0.7 the cells were induced by the addition of isopropyl-β- d-thiogalactopyranoside (IPTG) to a final concentration of 1 mM, and the culture was grown for a further 4 h. The culture was then divided into two aliquots and the cells were harvested by centrifugation at 4°C and the pellets stored at −80°C until required. A cell pellet (equivalent to 250 ml culture) was thawed on ice and resuspended in 50 ml of ice-cold binding buffer (0.05 M NaH2PO4, 0.3 M NaCl and 0.01 M imidazole) containing an EDTA-free protease inhibitor cocktail tablet (Roche), and sonicated to completion on ice. The sonicated sample was centrifuged at 20 000 g for 20 min to separate the cell debris from the soluble fraction. The soluble supernatant was loaded at 1 ml min−1 onto a 0.3 ml column of Ni-NTA resin (Qiagen®), pre-equilibrated with binding buffer at 4°C, and the column was washed at the same flow rate with ∼300 ml of high salt buffer (0.05 M NaH2PO4, 0.3 M NaCl, 0.02 M imadazole and 20% glycerol; pH 8). The column was then washed with 10–20 ml of low salt buffer (0.05 M NaH2PO4, 0.1 M NaCl, 0.02 M imadazole and 20% glycerol; pH 8) before eluting the bound protein in 2 ml elution buffer (0.05 M NaH2PO4, 0.25 M imadazole, 0.1 M NaCl and 20% glycerol; pH 8). The eluted protein was dialysed overnight against 2 l of buffer (0.05 M Tris pH 8, 0.1 M NaCl and 10% glycerol) and stored at 4°C for a maximum of 7 days before use. SDS-PAGE analysis revealed that the eluted protein was relatively pure and N-terminal amino acid sequencing confirmed that the eluted protein was indeed His6-SmaR (data not shown).

DNA band-shift assays

The promoter regions of carA, carR and pigA were excised as BamHI/HindIII fragments from pUCcarApro, pUCcarRpro and pUCpigApro and then 3′-end labelled with DIG-11-ddUTP using terminal transferase according to the manufacturer's instructions (DIG Gel Shift Kit, Roche). A 380 bp non-specific control fragment was amplified by PCR, using primers HS3 and HS4, from a region internal to an ORF (pigT), and labelled as above. DNA band-shift reactions (10 µl) contained the indicated amounts of protein, DNA and BHL in the presence of binding buffer (20 mM HEPES, pH 7.6, 1 mM EDTA, 10 mM (NH4)2SO4, 1 mM DTT, Tween, 0.02% (w/v) and 30 mM (KCl), 10 ng ml−1 poly l-lysine and 50 µg ml−1 uncut λ phage genomic DNA. Routinely, the reactions were allowed to proceed for 20 min at 30°C before loading on 6% polyacrylamide gels pre-equilibriated in 0.5× TBE buffer at 4°C. Electrophoresis was performed at ∼5 V cm−1 at 4°C until the tracking dye was three-quarters of the way down the gel, before transferring the DNA onto nylon membranes by contact blotting. The blots were developed using chemiluminescent detection and exposure to X-ray film as described in the DIG Gel Shift Kit protocol (Roche).

Bioassays of prodigiosin, carbapenem, N-AHLs, exoenzymes and motility

The assays for Pig and Car were performed as described previously (Slater et al., 2003). Pig production was plotted as [(A534 ml−1 OD600 −1) × 50] and Car production represented as halo area (cm2/OD600). Detection of N-AHLs was performed using the Serratia LIS bioassay described in Thomson et al. (2000). Activity of pectate lyase and cellulase were analysed on agar plates containing the substrates as described previously (Andro et al., 1984). Motility was assessed on tryptone swarm agar plates (10 g l−1 bacto tryptone, 5 g l−1 NaCl and 3 g l−1 agar). Overnight bacterial cultures were adjusted to an OD600 of 0.2 and 3 µl was spotted onto the plates and halo size was examined after growth for 16 h at 30°C.

β-Galactosidase assays

β-Galactosidase activity in bacterial cultures grown in liquid media was determined using ο-nitrophenol-β-galactosidase (ONPG) as the substrate as described previously (Miller, 1972). Enzyme activity was expressed as the initial reaction rate per millilitre of sample per OD600 of the bacterial culture tested (ΔA420 min−1 ml−1/OD600). Results presented are the mean ± the standard deviation of three independent experiments, unless stated otherwise.

Transposon exchange mutagenesis

An efficient method was developed to allow exchange of a chromosomal transposon insertion for an alternative transposon. Random transposon mutagenesis of 39006 was performed as described earlier (de Lorenzo et al., 1990; Thomson et al., 2000). Following mutagenesis, potential mutants were isolated by selecting mutants with either decreased or increased Pig levels.

The transposon mutants isolated were used to produce mixed φOT8 lysates enriched for either reduced or increased Pig phenotypes. φOT8 lysates were produced as described previously (Thomson et al., 2000; Crow, 2001), but mixed cultures of mutants were infected with phage, instead of cultures grown from single colonies. Heterogeneous cultures were prepared by growing each mutant on LBA, pooling these in LB and diluting to an OD600 of 1.5. Phage lysates of φOT8 were purified from this mutant pool and referred to as φOT8 ‘heterolysates’. The φOT8 ‘heterolysates’ were used to transduce the mutation pool into strains that were to have their transposons exchanged. Transductants that had lost the parental resistance but gained the ‘heterolysate’ resistance were analysed further. Switching resistances indicates that homologous recombination has occurred in the region of interest and that the chromosomal ‘regulator’::transposon insertion has been replaced with an alternative transposon insertion. The method required the appropriate mutant to be represented in the original ‘heterolysate’.

The location of the new transposon was determined by PCR amplification across the insertion site using primers facing out of the transposon (LER1 and SP2 for mini-Tn5Sm/Sp and LL-68 and LR-68 for mini-Tn5lacZ1) and primers specific to the regulator of interest. PCR products in the expected size range were analysed further by sequencing and the point of insertion determined. Mutations were transduced into a ‘clean’ LacA background using φOT8 and confirmed by phenotype, PCR and in some cases by Southern blotting. Using this method, mutants with alternative markers were produced in pigP, pigQ, pigW pigX and rap. The resulting strains were PIG13S (pigP::mini-Tn5Sm/Sp), PIG17S (pigQ::mini-Tn5Sm/Sp), PIG62S (pigW::mini-Tn5Sm/Sp), ROP4S (pigX::mini-Tn5Sm/Sp) and RAPL (rap::mini-Tn5lacZ1).

Primer extension and RNA studies

A hot-acidic phenol method was used to extract total RNA from 39006 (Aiba et al., 1981). Primer extension analysis for the pigA transcript was performed as described previously (Slater et al., 2003). All primer extension reactions were performed with 25 µg of total RNA and 0.2 pmol of the appropriate 32P-labelled primer. Oligonucleotide primers HS60 and HS61 were used in primer extension reactions for carA and carR, respectively, and hybridize 31 bp and 36 bp downstream of the translational start sites. Primers HS61 and HS60 were also used to determine the transcriptional start sites of carA and carR, respectively, using the T7 sequencing kit (USB). Template DNA used for the sequencing reactions for both carR and carA was pUC19-28. Plasmid pUC19-28 was constructed by subcloning a 2.8 kb BamHI fragment containing carR and part of carA from pTON28 (Cox et al., 1998) into pUC19.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Concluding remarks
  7. Experimental procedures
  8. Acknowledgements
  9. References

The authors thank all members of the Salmond group, Martin Welch for helpful discussions, Ian Foulds for technical assistance, Anne Barnard for critically reading the manuscript and Barry Bycroft and Paul Williams for the generous supply of BHL. This work was supported by the BBSRC, UK and P.F. was supported by a Bright Futures Top Achiever Doctoral Scholarship from the Tertiary Education Commission of New Zealand.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Concluding remarks
  7. Experimental procedures
  8. Acknowledgements
  9. References
  • Aiba, H., Adhya, S., and de Crombrugghe, B. (1981) Evidence for two functional gal promoters in intact Escherichia coli cells. J Biol Chem 256: 1190511910.
  • Alekshun, M.N., Levy, S.B., Mealy, T.R., Seaton, B.A., and Head, J.F. (2001) The crystal structure of MarR, a regulator of multiple antibiotic resistance, at 2.3 Å Resolution. Nat Struct Biol 8: 710714.
  • Altschul, S.F., Madden, T.L., Schaffer, A.A., Zhang, J., Zhang, Z., Miller, W., and Lipman, D.J. (1997) Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res 25: 33893402.
  • Andro, T., Chambost, J.P., Kotoujansky, A., Cattaneo, J., Bertheau, Y., Barras, F., et al. (1984) Mutants of Erwinia chrysanthemi defective in secretion of pectinase and cellulase. J Bacteriol 160: 11991203.
  • Bainton, N.J., Stead, P., Chhabra, S.R., Bycroft, B.W., Salmond, G.P., Stewart, G.S., and Williams, P. (1992) N-(3-oxohexanoyl)-l-homoserine lactone regulates carbapenem antibiotic production in Erwinia carotovora. Biochem J 288: 9971004.
  • Baker, D.A., and Kelly, J.M. (2004) Structure, function and evolution of microbial adenylyl and guanylyl cyclases. Mol Microbiol 52: 12291242.
  • Barne, K.A., Bown, J.A., Busby, S.J., and Minchin, S.D. (1997) Region 2.5 of the Escherichia coli RNA polymerase sigma70 subunit is responsible for the recognition of the ‘extended-10’ motif at promoters. EMBO J 16: 40344040.
  • Bijlsma, J.J., and Groisman, E.A. (2003) Making informed decisions: regulatory interactions between two-component systems. Trends Microbiol 11: 359366.
  • Busenlehner, L.S., Pennella, M.A., and Giedroc, D.P. (2003) The SmtB/ArsR family of metalloregulatory transcriptional repressors: structural insights into prokaryotic metal resistance. FEMS Microbiol Rev 27: 131143.
  • Bycroft, B.W., Maslen, C., Box, S.J., Brown, A., and Tyler, J.W. (1987) The isolation and characterisation of (3R,5R)- and (3S,5R)-carbapenem-3-carboxylic acid from Serratia and Erwinia species and their putative biosynthetic role. J Chem Soc Chem Comm 21: 16231625.
  • Cang, S., Sanada, M., Johdo, O., Ohta, S., Nagamatsu, Y., and Yoshimoto, A. (2000) High production of prodigiosin by Serratia marcescens grown on ethanol. Biotechnol Lett 22: 17611765.
  • Cerdeño, A.M., Bibb, M.J., and Challis, G.L. (2001) Analysis of the prodiginine biosynthesis gene cluster of Streptomyces coelicolor A3 (2): new mechanisms for chain initiation and termination in modular multienzymes. Chem Biol 8: 817829.
  • Chancey, S.T., Wood, D.W., and Pierson, L.S., (1999) 3rd Two-component transcriptional regulation of N-acyl-homoserine lactone production in Pseudomonas aureofaciens. Appl Environ Microbiol 65: 22942299.
  • Cox, A.R., Thomson, N.R., Bycroft, B., Stewart, G.S., Williams, P., and Salmond, G.P. (1998) A pheromone-independent CarR protein controls carbapenem antibiotic synthesis in the opportunistic human pathogen Serratia marcescens. Microbiology 144: 201209.
  • Cren, M., Kondorosi, A., and Kondorosi, E. (1995) NolR controls expression of the Rhizobium meliloti nodulation genes involved in the core Nod factor synthesis. Mol Microbiol 15: 733747.
  • Crow, M.A. (2001) The Genetic Regulation of Pigment and Antibiotic Biosynthesis in Serratia sp. PhD Thesis. Cambridge: Department of Biochemistry, The University of Cambridge.
  • Derzelle, S., Duchaud, E., Kunst, F., Danchin, A., and Bertin, P. (2002) Identification, characterization, and regulation of a cluster of genes involved in carbapenem biosynthesis in Photorhabdus luminescens. Appl Environ Microbiol 68: 37803789.
  • Dutta, R., Qin, L., and Inouye, M. (1999) Histidine kinases: diversity of domain organization. Mol Microbiol 34: 633640.
  • Eriksson, A.R., Andersson, R.A., Pirhonen, M., and Palva, E.T. (1998) Two-component regulators involved in the global control of virulence in Erwinia carotovora subsp. carotovora. Mol Plant Microbe Interact 11: 743752.
  • Harris, S.J., Shih, Y.L., Bentley, S.D., and Salmond, G.P. (1998) The hexA gene of Erwinia carotovora encodes a LysR homologue and regulates motility and the expression of multiple virulence determinants. Mol Microbiol 28: 705717.
  • Harris, A.K., Williamson, N.R., Slater, H., Cox, A., Abbasi, S., Foulds, I., et al. (2004) The Serratia gene cluster encoding biosynthesis of the red antibiotic, prodigiosin, shows species- and strain-dependent genome context variation. Microbiology 150: 35473560.
  • Heeb, S., and Haas, D. (2001) Regulatory roles of the GacS/GacA two-component system in plant-associated and other Gram-negative bacteria. Mol Plant Microbe Interact 14: 13511363.
  • Heeb, S., Blumer, C., and Haas, D. (2002) Regulatory RNA as mediator in GacA/RsmA-dependent global control of exoproduct formation in Pseudomonas fluorescens CHA0. J Bacteriol 184: 10461056.
  • Hood, D.W., Heidstra, R., Swoboda, U.K., and Hodgson, D.A. (1992) Molecular genetic analysis of proline and tryptophan biosynthesis in Streptomyces coelicolor A3 (2): interaction between primary and secondary metabolism – a review. Gene 115: 512.
  • Horng, Y.T., Deng, S.C., Daykin, M., Soo, P.C., Wei, J.R., Luh, K.T., et al. (2002) The LuxR family protein SpnR functions as a negative regulator of N-acylhomoserine lactone-dependent quorum sensing in Serratia marcescens. Mol Microbiol 45: 16551671.
  • Hyytiainen, H., Montesano, M., and Palva, E.T. (2001) Global regulators ExpA (GacA) and KdgR modulate extracellular enzyme gene expression through the RsmA-rsmB system in Erwinia carotovora subsp. carotovora. Mol Plant Microbe Interact 14: 931938.
  • Iyer, L.M., and Aravind, L. (2002) The catalytic domains of thiamine triphosphatase and CyaB-like adenylyl cyclase define a novel superfamily of domains that bind organic phosphates. BMC Genomics 3: 33.
  • Kitten, T., Kinscherf, T.G., McEvoy, J.L., and Willis, D.K. (1998) A newly identified regulator is required for virulence and toxin production in Pseudomonas syringae. Mol Microbiol 28: 917929.
  • Kondorosi, E., Pierre, M., Cren, M., Haumann, U., Buire, M., Hoffmann, B., et al. (1991) Identification of NolR, a negative transacting factor controlling the nod regulon in Rhizobium meliloti. J Mol Biol 222: 885896.
  • Laatu, M., and Condemine, G. (2003) Rhamnogalacturonate lyase RhiE is secreted by the out system in Erwinia chrysanthemi. J Bacteriol 185: 16421649.
  • de Lorenzo, V., Herrero, M., Jakubzik, U., and Timmis, K.N. (1990) Mini-Tn5 transposon derivatives for insertion mutagenesis, promoter probing, and chromosomal insertion of cloned DNA in gram-negative eubacteria. J Bacteriol 172: 65686572.
  • McGowan, S., Sebaihia, M., Jones, S., Yu, B., Bainton, N., Chan, P.F., et al. (1995) Carbapenem antibiotic production in Erwinia carotovora is regulated by CarR, a homologue of the LuxR transcriptional activator. Microbiology 141: 541550.
  • McGowan, S.J., Sebaihia, M., Porter, L.E., Stewart, G.S., Williams, P., Bycroft, B.W., and Salmond, G.P. (1996) Analysis of bacterial carbapenem antibiotic production genes reveals a novel beta-lactam biosynthesis pathway. Mol Microbiol 22: 415426.
  • McGowan, S.J., Sebaihia, M., O'Leary, S., Hardie, K.R., Williams, P., Stewart, G.S., et al. (1997) Analysis of the carbapenem gene cluster of Erwinia carotovora: definition of the antibiotic biosynthetic genes and evidence for a novel beta-lactam resistance mechanism. Mol Microbiol 26: 545556.
  • McGowan, S.J., Holden, M.T., Bycroft, B.W., and Salmond, G.P. (1999) Molecular genetics of carbapenem antibiotic biosynthesis. Antonie Van Leeuwenhoek 75: 135141.
  • McGowan, S.J., Barnard, A.M.L., Bosgelmez, G., Sebaihia, M., Simpson, N.J.L., Thomson, N.R., et al. (2005) Carbapenem antibiotic biosynthesis in Erwinia carotovora is regulated by physiological and genetic factors modulating the quorum sensing-dependent control pathway. Mol Microbiol 55: 526545.
  • Manderville, R.A. (2001) Synthesis, proton-affinity and anti-cancer properties of the prodigiosin-group natural products. Curr Med Chem Anti-Canc Agents 1: 195218.
  • Miller, J. (1972) Experiments in Molecular Genetics. New York, NY: Cold Spring Harbour Laboratory Press.
  • Minogue, T.D., Trebra, M.W., Bernhard, F., and Bodman, S.B. (2002) The autoregulatory role of EsaR, a quorum-sensing regulator in Pantoea stewartii ssp. stewartii: evidence for a repressor function. Mol Microbiol 44: 16251635.
  • Morrison, D.A. (1966) Prodigiosin synthesis in mutants of Serratia marcesens. J Bacteriol 91: 15991604.
  • Ovadis, M., Liu, X., Gavriel, S., Ismailov, Z., Chet, I., and Chernin, L. (2004) The global regulator genes from biocontrol strain Serratia plymuthica IC1270: cloning, sequencing, and functional studies. J Bacteriol 186: 49864993.
  • Parker, W.L., Rathnum, M.L., Wells, J.S., Trejo, W.H., Principe, P.A., and Sykes, R.B. (1982) SQ 27, 860, A simple carbapenem produced by species of Serratia and Erwinia. J Antibiot 35: 653660.
  • Paruchuri, D.K., and Harshey, R.M. (1987) Flagellar variation in Serratia marcescens is associated with color variation. J Bacteriol 169: 6165.
  • Perez-Tomas, R., Montaner, B., Llagostera, E., and Soto-Cerrato, V. (2003) The prodigiosins, proapoptotic drugs with anticancer properties. Biochem Pharmacol 66: 14471452.
  • Reimmann, C., Beyeler, M., Latifi, A., Winteler, H., Foglino, M., Lazdunski, A., and Haas, D. (1997) The global activator GacA of Pseudomonas aeruginosa PAO positively controls the production of the autoinducer N-butyryl-homoserine lactone and the formation of the virulence factors pyocyanin, cyanide, and lipase. Mol Microbiol 24: 309319.
  • Rjazantseva, I.N., Andreeva, I.N., and Ogorodnikova, T.I. (1994) Effect of various growth conditions on pigmentation of Serratia marcescens. Microbios 79: 155161.
  • Rosenberg, M., Blumberger, Y., Judes, H., Bar-Ness, R., Rubinstein, E., and Mazor, Y. (1986) Cell surface hydrophobicity of pigmented and nonpigmented clinical Serratia marcescens strains. Infect Immun 51: 932935.
  • Sambrook, J., Fritsch, E.F., and Maniatis, T. (1989) Molecular Cloning: a Laboratory Manual, 2nd edn. New York, NY: Cold Spring Harbour Laboratory Press.
  • Silverman, M.P., and Munoz, E.F. (1973) Effect of iron and salt on prodigiosin synthesis in Serratia marcescens. J Bacteriol 114: 9991006.
  • Sismeiro, O., Trotot, P., Biville, F., Vivares, C., and Danchin, A. (1998) Aeromonas hydrophila adenylyl cyclase 2: a new class of adenylyl cyclases with thermophilic properties and sequence similarities to proteins from hyperthermophilic archaebacteria. J Bacteriol 180: 33393344.
  • Slater, H., Crow, M., Everson, L., and Salmond, G.P. (2003) Phosphate availability regulates biosynthesis of two antibiotics, prodigiosin and carbapenem, in Serratia via both quorum-sensing-dependent and – independent pathways. Mol Microbiol 47: 303320.
  • Sole, M., Rius, N., and Loren, J.G. (1997) The role of pH in the ‘glucose effect’ on prodigiosin production by non-proliferating cells of Serratia marcescens. Lett Appl Microbiol 25: 8184.
  • Thomson, N.R., Cox, A., Bycroft, B.W., Stewart, G.S., Williams, P., and Salmond, G.P. (1997) The rap and hor proteins of Erwinia, Serratia and Yersinia: a novel subgroup in a growing superfamily of proteins regulating diverse physiological processes in bacterial pathogens. Mol Microbiol 26: 531544.
  • Thomson, N.R., Crow, M.A., McGowan, S.J., Cox, A., and Salmond, G.P. (2000) Biosynthesis of carbapenem antibiotic and prodigiosin pigment in Serratia is under quorum sensing control. Mol Microbiol 36: 539556.
  • Trutko, S., and Akimenko, V. (1989) The Role of Prodigiosin Biosynthesis in the Regulation of Oxidative Metabolism of the Producer Serratia marcescens. Mikrobiologiya 58: 723729.
  • Vannini, A., Volpari, C., Gargioli, C., Muraglia, E., Cortese, R., De Francesco, R., et al. (2002) The crystal structure of the quorum sensing protein TraR bound to its autoinducer and target DNA. EMBO J 21: 43934401.
  • Wasserman, H.H., Skles, R.J., Peverada, P., Shaw, C.K., Cushley, R.J., and Lipsky, C.R. (1973) Biosynthesis of prodigiosin. Incorporation patterns of C-labeled alanine, proline, glycine, and serine elucidated by fourier transform nuclear magnetic resonance. J Am Chem Soc 95: 68746875.
  • Welch, M., Todd, D.E., Whitehead, N.A., McGowan, S.J., Bycroft, B.W., and Salmond, G.P. (2000) N-acyl homoserine lactone binding to the CarR receptor determines quorum-sensing specificity in Erwinia. EMBO J 19: 631641.
  • Whitehead, N.A., Barnard, A.M., Slater, H., Simpson, N.J., and Salmond, G.P. (2001) Quorum-sensing in Gram-negative bacteria. FEMS Microbiol Rev 25: 365404.
  • Williams, S.G., and Manning, P.A. (1991) Transcription of the Vibrio cholerae haemolysin gene, hlyA, and cloning of a positive regulatory locus, hlyU. Mol Microbiol 5: 20312038.
  • Williams, R.P., Gott, C.L., and Qadri, S.M. (1971a) Induction of pigmentation in nonproliferating cells of Serratia marcescens by addition of single amino acids. J Bacteriol 106: 444448.
  • Williams, R.P., Gott, C.L., Qadri, S.M., and Scott, R.H. (1971b) Influence of temperature of incubation and type of growth medium on pigmentation in Serratia marcescens. J Bacteriol 106: 438443.
  • Williams, S.G., Attridge, S.R., and Manning, P.A. (1993) The transcriptional activator HlyU of Vibrio cholerae: nucleotide sequence and role in virulence gene expression. Mol Microbiol 9: 751760.
  • Williamson, N.R., Simonsen, H.T., Ahmed, R.A.A., Goldet, G., Slater, H., Woodley, L., et al. (2005) Biosynthesis of the red antibiotic, prodigiosin, in Serratia: identification of a novel 2-methyl-3- n-amyl-pyrrole (MAP) assembly pathway, definition of the terminal condensing enzyme, and implications for undecylprodigiosin biosynthesis in Streptomyces. Mol Microbiol 56: 971989.
  • Witney, F.R., Failla, M.L., and Weinberg, E.D. (1977) Phosphate inhibition of secondary metabolism in Serratia marcescens. Appl Environ Microbiol 33: 10421046.
  • Wu, R.Y., Zhang, R.G., Zagnitko, O., Dementieva, I., Maltzev, N., Watson, J.D., et al. (2003) Crystal structure of Enterococcus faecalis SlyA-like transcriptional factor. J Biol Chem 278: 2024020244.
  • Yanisch-Perron, C., Vieira, J., and Messing, J. (1985) Improved M13 phage cloning vectors and host strains: nucleotide sequences of the M13mp18 and pUC19 vectors. Gene 33: 103119.
  • Zhang, R.G., Pappas, T., Brace, J.L., Miller, P.C., Oulmassov, T., Molyneaux, J.M., et al. (2002) Structure of a bacterial quorum-sensing transcription factor complexed with pheromone and DNA. Nature 417: 971974.
  • Zuber, S., Carruthers, F., Keel, C., Mattart, A., Blumer, C., Pessi, G., et al. (2003) GacS sensor domains pertinent to the regulation of exoproduct formation and to the biocontrol potential of Pseudomonas fluorescens CHA0. Mol Plant Microbe Interact 16: 634644.