RssA two-component system regulates biosynthesis of the tripyrrole antibiotic, prodigiosin, in Serratia marcescens

ARTICLE IN PRESS International Journal of Medical Microbiology 300 (2010) 304–312

Contents lists available at ScienceDirect

International Journal of Medical Microbiology journal homepage: www.elsevier.de/ijmm

The RssB/RssA two-component system regulates biosynthesis of the tripyrrole antibiotic, prodigiosin, in Serratia marcescens Yu-Tze Horng a,b, Kai-Chih Chang a, Yen-Ni Liu c, Hsin-Chih Lai b,n, Po-Chi Soo a,n,n a

Department of Laboratory Medicine and Biotechnology, Tzu Chi University, College of Medicine, 701 Section 3, Zhongyang Road, Hualien 97004, Taiwan, ROC Department of Medical Biotechnology and Laboratory Science, Chang Gung University, 259 Wen-Hwa 1st Road, Kwei-Shan, Tao-Yuan 333, Taiwan, ROC c Graduate School of Biotechnology and Bioengineering, Yuan Ze University, 135 Yuan-Tung Road, Chung-Li 320, Taiwan, ROC b

a r t i c l e in f o

a b s t r a c t

Article history: Received 3 November 2009 Received in revised form 15 January 2010 Accepted 31 January 2010

Serratia marcescens CH-1 produces a red, cell-associated pigment, prodigiosin, synthesized by enzymes encoded in the pig operon. The underlying regulatory mechanism, especially its relationship with the RssAB two-component system signaling, remained uncharacterized. Here, we show that phosphorylated RssB (RssB-P) directly binds to the promoter region of the pig operon (pigA promoter), as observed using an electrophoretic mobility shift assay. Furthermore, we identify the RssB-P binding site located downstream of the –10 and –35 regions in pigA using a DNase I footprinting assay. A compilation of the RssB-P binding sites in flhDC, rssB and pigA promoter regions reveals the presence of a conserved core sequence, GAGATTTTAGCTAAATTAATBTTT (B ¼C, G, or T), which we believe is the RssB binding sequence. Site-specific mutation of conserved nucleotides within the conserved RssB binding sequence in the pigA promoter region leads to absence of retardation in the presence of RssB-P in vitro and elevated transcription of pigA in vivo. These data suggest that RssAB signaling negatively regulates prodigiosin production, and such inhibition is mediated through direct and specific repression of transcriptional activity of the pig operon. & 2010 Elsevier GmbH. All rights reserved.

Keywords: Serratia marcescens RssAB Two-component system Prodigiosin

Introduction Serratia marcescens is a ubiquitous Gram-negative, rod-shaped bacterium found in soil, water, plant surfaces, and insects (Traub, 2000; Stock et al., 2003). The organism is an important nosocomial pathogen capable of causing pneumonia, urinary tract infections, sepsis, conjunctivitis, osteomyelitis, and endocarditis (von Graevenitz, 1980; Eisenstein, 1990; Hejazi and Falkiner, 1997; Parment, 1997). Besides, multiple antibiotic resistance also develops upon drug treatment (Arakawa et al., 2000; Knowles et al., 2000; Traub, 2000). S. marcescens is characterized by expression of the red pigment, prodigiosin (2-methyl-3-pentyl-6methoxyprodiginine), a secondary metabolite (Williamson et al., 2005, 2006). While the function of prodigiosin in S. marcescens remains to be defined, prodigiosin is attracting renewed scientific interest because of its therapeutic potential including immunosuppressive, proapoptotic, and anticancer activities (Rosenberg et al., 1986; Paruchuri and Harshey, 1987; Perez-Tomas et al., 2003; Williamson et al., 2006; Haddix et al., 2008).

n

Corresponding author. Corresponding author. Tel.: + 886 3 8565301x2347; fax: + 886 3 8571917. E-mail address: [email protected] (P.-C. Soo).

nn

1438-4221/$ - see front matter & 2010 Elsevier GmbH. All rights reserved. doi:10.1016/j.ijmm.2010.01.003

A prodigiosin biosynthesis cluster, designated pig, has been identified in S. marcescens ATCC 274 (Sma 274) and Serratia sp. ATCC 39006 (Serratia 39006) (Slater et al., 2003; Harris et al., 2004). The pig cluster was also found in other Serratia isolates (Harris et al., 2004). The cluster is transcribed as a polycistronic message from the pigA promoter. It has been reported that prodigiosin production in S. marcescens is regulated by multiple factors including quorum sensing signals (N-acyl-homoserine lactones or AI-2), several transcriptional regulators, and a twocomponent system PigQW (Whitehead et al., 2001; Fineran et al., 2005; Tanikawa et al., 2006; Williamson et al., 2006). Environmental factors, such as nutrient deprivation, stress, temperature, oxygen, pH, light, ionic strength, and phosphate availability also affect prodigiosin generation (Slater et al., 2003; Williamson et al., 2006). However, it is unclear how the bacteria sense the environmental factors and change the prodigiosin synthesis. The bacterial two-component systems are required for innumerable adaptive response in bacteria (Stock et al., 2000). It was shown that prodigiosin is positively modulated by a two-component phosphorelay system involving PigW (a sensor kinase) and PigQ (a response regulator) in Serratia 39006 (Fineran et al., 2005). The RssB/A two-component phosphorelay system was initially identified in S. marcescens CH-1 during screening for mutants displaying a precocious swarming phenotype (Lai et al., 2005). Swarming is a mode of surface translocation that is dependent

ARTICLE IN PRESS Y.-T. Horng et al. / International Journal of Medical Microbiology 300 (2010) 304–312

upon flagellar motility (Chilcott and Hughes, 2000; Berg, 2005). The genes in the flagellar system are organized into 3 hierarchical transcriptional classes. The class I master operon, flhDC, encodes the FlhD2C2 complex which is an activator of s70-dependent transcription from class II promoters (Chilcott and Hughes, 2000). We have previously shown that synthesis of flagella and flhDC expression were significantly increased in the rssA mutant (S. marcescens CH-1 DrssA) (Soo et al., 2008). The RssA sensor kinase phosphorylates the RssB response regulator (Wei et al., 2005a). Phosphorylated RssB (RssB-P) binds to the flhDC promoter (Soo et al., 2008). Missense mutation on histidine 248 of RssA and aspartate 51 of RssB led to abnormal swarming phenotype, unphosphotransfer from RssA to RssB, and unbinding to the flhDC and rssB promoters by RssB (Wei et al., 2005a; Soo et al., 2008). Taken together, the previous studies indicate the RssB/A twocomponent system negatively regulates the swarming phenotype in S. marcescens CH-1. A number of two-component systems are pleiotropic regulators that exert their effects on more than one phenotype (Adamidis et al., 1990; Adamidis and Champness, 1992; Anderson et al., 2001; Ryding et al., 2002). In this study, we set out to see whether there is any regulatory relationship between RssAB signaling and prodigiosin synthesis. While PigQW positively regulates pig cluster expression (Fineran et al., 2005), we presented evidence to show that the RssB/A two-component system acts as a negative regulator of prodigiosin production, and not only of swarming, in S. marcescens CH-1. Increased prodigiosin production is observed in rssA and rssBA mutants. In vitro studies suggest that this regulation is achieved through direct binding of RssB-P to the pigA promoter, which at least partly suppresses transcription of the pig cluster. We further identify the specific RssB-P binding site in the pigA promoter. A conserved RssB binding sequence was subsequently identified in promoter regions of pigA, flhDC, and rssB which were all negatively controlled by RssAB signaling (Wei et al., 2005a; Soo et al., 2008). The potential RssB binding sequence is newly defined as a conserved bacterial operator.

305

strains in which rssA and rssBA are disrupted by a gentamicinresistant cassette (Lai et al., 2005). Escherichia coli DH5a (Invitrogen, USA), a host strain for the maintenance of recombinant DNA plasmids, was cultured at 37 1C. S. marcescens was cultured at 30 1C in 5 mL of Luria-Bertani (LB) supplemented with the adequate antibiotics at the final concentration of gentamicin 20 mg/mL, ampicillin 50 mg/mL, and streptomycin 50 mg/mL, where necessary (Lai et al., 2005). All bacteria strains and plasmids used in this study are shown in Table 1. The bacterial growth was measured by detecting the optical density of bacterial culture (OD600 nm). Enzymes, chemicals, and primers DNA restriction and modification enzymes were purchased from Roche (Germany). Pfu polymerase and PCR-related products were obtained from Stratagene (USA) and Perkin-Elmer (USA). Other laboratory-grade chemicals were purchased from Sigma (USA) and Merck (Germany). The primers used in this study were acquired from MD Bio (Taiwan) and the sequences of the primers were listed in Table 2. Construction of transcriptional fusion plasmids The promoterless lacZ and streptomycin-resistant genes were cloned into pUK21 (Vieira and Messing, 1991) to create the pYN0 construct. The 500-bp wild type (Pa, Table 2) or mutated pigA promoter regions were PCR-amplified from S. marcescens CH-1 using the primer pairs, PpF1 and PpR2 or PpF1 and PpRm, respectively (Table 2). An EcoRV restriction site (underlined) was incorporated in the mutated RssB binding sequence to facilitate sequence confirmation. PCR fragments were cloned into pGEMs-T Easy Vector (Promega, USA), followed by transformation into E. coli DH5a. Ampicillin-resistant colonies were selected, and inserts were verified by restriction enzyme (EcoRV) digestion, PCR, and sequencing. The 500-bp pigA promoter with wild-type or mutated RssB binding sequence was excised using the BglII and XbaI restriction sites and subsequently cloned into pYN0 to yield pYN1 or pYNm, respectively.

Materials and methods

b-galactosidase activity assay Bacterial strains, mutants, and culture conditions S. marcescens CH-1, the clinical isolate used in this study, is wild-type. The S. marcescens CH-1DA and CH-1DBA are mutant

Bacterial cells harboring low-copy number of pYN0, pYN1, or pYNm were grown in LB broth containing 50 mg/ml of streptomycin at 30 1C. Overnight cultures were diluted 100-fold and

Table 1 Bacterial strains and plasmids used in this study. Strain or plasmid

Relevant characteristic(s)

Source or reference

Serratia marcescens strains CH-1 CH-1DA CH-1DBA

Clinical isolate rssA deleted mutant strain rssBA deleted mutant strain

(Wei et al., 2005a) (Wei et al., 2005a) This study

(Prentki, 1984)

DH5a

l-pir lysogen of CC118 [D(ara-leu) araD DlacX 74 galE galK phoA20 thi-1rpsE rpoB argE(Am) recA1]; permissive host for suicide plasmids requiring the Pir protein l-pir lysogen of S17-1 [thi pro hsdR– hsdM + recA RP4 2-Tc::Mu-Km::Tn7 (Tpr; Smr)]; permissive host able to transfer suicide plasmids requiring the Pir protein by conjugation to recipient cells F- F80lacZDM15D(lacZYA-argF)U169 deoR recA1 endA1 hsdR17(r-k, mk+ ) phoA supE44 thi-1 gyrA96 relA1 l-

Plasmids pYN0 pYN1 pYNm prssA prssBA

pUK21 (promoterless lacZ); Smr pUK21 (wild-type PpigA::lacZ); Smr pUK21 (mutated RssB binding sequence in PpigA::lacZ); Smr pBAD33-Cm::rssA; Cmr pBAD33-Cm::rssBA; Cmr

This This This This This

E. coli strains CC118(lpir) S17-1(lpir)

(Prentki, 1984) Invitrogene

study study study study study

ARTICLE IN PRESS 306

Y.-T. Horng et al. / International Journal of Medical Microbiology 300 (2010) 304–312

Table 2 Oligonucleotide primers used in this study. Primers

Sequence

Primer information

PpF1

5’-AGATCTCCTCGTAAAAACGAATCGTC-3’

PpR2 PpRm

5’-TCTAGACGAACTCCGCCATTGGGTT-3’ 5’-TCTAGACGAACTCCGCCATTGGGTTGcGg GATatcATTActTAATATTTCTAGTTTGGAGGTG-3’

Pb-F Pb-R

5’-GAGAAACCGATAGCCACATCC-3’ 5’-DIG-CTGCTGCTCACTTGATAAG-3’

wpigF mpigF

Pcd-F

5’-CTCCAAACTAGAAATA-3’ 5’-CTCCAAACTAGAAATATTATATAATGGGATCAC GCAACCCAATGGCGGAGTTCGTCATGGATTTT AACTTATCAAGTGAG-3’ 5’-GAGAAACCGATAGCCACATCC-3’

Forward primer for amplification of 500-bp Pa fragment or 500-bp pigA promoter containing mutated RssB binding sequence. The PCR product was subsequently cloned to pYN1 or pYNm, respectively. Reverse primer for 500-bp Pa fragment amplified. The PCR product was subsequently cloned to pYN1. Reverse primer for 500-bp pigA promoter containing mutated RssB binding sequence. The PCR product was subsequently cloned to pYNm. An EcoRV restriction site (in bold) was incorporated in the mutated RssB binding sequence to facilitate sequence confirmation. Substituted nucleotides are presented in lowercase letters. Forward primer for 347-bp Pb fragment amplified. The PCR product was used in EMSA. Reverse primer for Pb, pigA_wt or pigA_mut7 fragment amplified with DIG modification. The PCR products were used in EMSA. Forward primer for 86-bp pigA_wt fragment amplified. The PCR product was used in EMSA. Forward primer for 86-bp pigA_mut7 fragment amplified. The PCR product was used in EMSA.

Pcd-R

5’-GCATCCTGATCCAGCACCT-3’

Forward primer for 400-bp Pc or Pd amplified. If Pcd-F was labeled with FAM, Pc fragment was amplified. The PCR product was used in DNase I footprinting assay. Reverse primer for 400-bp Pc or Pd amplified. If Pcd-R was labeled with FAM, Pd fragment was amplified. The PCR product was used in DNase I footprinting assay.

grown for a further 8 hours. Cultures were assayed for b-galactosidase activity using o-nitrophenol-b-galactoside (ONPG), and activity was expressed as Miller units (Miller, 1972). Results are presented as means7one standard deviation of triplicate independent experiments. Quantification of prodigiosin production Prodigiosin production was quantified by modification of the procedure described by Wei et al. (2005b). Briefly, 0.5 ml suspension of the bacterial culture was mixed with an equal volume of 2% (w/v) aluminum sulfate potassium dodecahydrate. Methanol (4 ml) was added, followed by vigorous mixing. After centrifugation at 5000  g for 10 min, 800 ml of the supernatant was mixed with 200 ml solvent of 0.05 N HCl/methanol (4:1 [v/v]). Subsequently, the mixture containing the red pigment was subjected to high performance liquid chromatography (HPLC) with absorption detected at 535 nm (OD535). The retention time of prodigiosin was 16.5 min. The peak area was measured and converted to mass concentrations calibrated using purified prodigiosin as the standard. Electrophoretic mobility shift assay (EMSA) The genomic DNA of S. marcescens CH-1 was used as a template for PCR. The DNA fragment, Pb, was amplified using the Pb-F and Pb-R primers (Table 2). The 86-bp DNA fragment, pigA_wt, was amplified using primers wpigF and Pb-R. The 86-bp pigA_mut7 fragment containing the 7 mutated nucleotides and encompassing –58 to + 28 nucleotides relative to the translational start site of pigA was amplified using mpigF and Pb-R (Table 2). All the DNA fragments, Pb, pigA_wt, and pigA_mut7, were labeled with digoxigenin (DIG) by using DIGlabeled primer, Pb-R (Table 2). RssB was purified according to the procedure described by Wei et al. (2005a) and phosphorylated with acetylphosphate (Wei et al., 2005a; Soo et al., 2008). The EMSA was performed according to the procedure of Soo et al. (2008) with some modifications. Briefly, DNA–protein complexes were electroblotted onto a positively charged Hybond-N nylon membrane (Amersham, United Kingdom) and detected using alkaline phosphatase-conjugated anti-DIG antibodies (Roche, Germany). CSPDs was added as the substrate, as described by the manufacturer (Roche, Germany). Images of DIG-labeled

DNA on the membranes were detected with Chemicapt (VILBER LOURMAT, Germany) at room temperature for 30–60 min. DNase I footprinting DNase I footprinting was performed using the protocol of Soo et al. (2008). Briefly, the 400-bp pigA promoter region which covered from 80 bp downstream to 319 bp upstream of the translation start site of pigA was amplified from genomic DNA. The primer pairs, Pcd-F and Pcd-R, were used for PCR (Table 2). Depending on the strand analyzed, one primer was labeled with 6-carboxyfluorescein (FAM) (MD Bio, Taiwan) at the 5’ terminus. The PCR product was denoted as Pc or Pd, depending on whether the coding strand or the template strand was labeled, respectively (Fig. 2). The FAM-labeled DNA fragment was incubated with 28 mM phosphorylated RssB in 50 ml solution containing 11 mM Tris-HCl (pH 7.4), 0.1 mM EDTA, 5.5 mM MgCl2, 20 mg/ml of poly(dI-dC) (Pharmacia, Sweden), and 0.2 mg/ml bovine serum albumin. After 20 min incubation at room temperature, 50 ml of DNase I (Promega, USA) (1  10–3 U/ml, freshly prepared by diluting the stock with the D buffer consisting of 10 mM Tris-HCl [pH 7.5], 10 mM MgCl2, and 5 mM CaCl2) was added, and the mixture was further incubated at 26 1C for 3 min. Digestion was terminated by adding 100 ml of stop solution containing 0.2 M NaCl, 40 mM EDTA, 1% sodium dodecyl sulfate, and 125 mg/ml of tRNA. After incubation at 37 1C for 20 min, samples were extracted with a phenol chloroform-isoamyl alcohol solution (25:24:1), precipitated with absolute ethanol, washed with 70% ethanol, and dissolved in 10 ml deionized formamide. Following the addition of 0.5 ml ROX-400 molecular size standard (Applied Biosystems, USA), samples were denatured at 95 1C for 5 min and rapidly chilled on ice. Electrophoresis was performed using the ABI Prism 3100 capillary DNA genetic analyzer, and data were examined with GeneMapper software, version 3.5 (Applied Biosystems, USA), according to the instructions supplied by the manufacturer. As resolution of the data from the GeneMapper software is better when the digested DNA was longer than 50 nucleotides, the Pc and Pd DNA fragments were designed to be longer than the Pb fragment. Analysis of DNA sequences DNA sequences were compiled using the comparison program, ClustalW+ , provided by SeqWeb v. 3.1.2 (http://v8803.nhri.org. tw:8003/mgr.shtml).

ARTICLE IN PRESS Y.-T. Horng et al. / International Journal of Medical Microbiology 300 (2010) 304–312

Results RssAB negatively regulates prodigiosin synthesis The rssA or rssBA mutation in S. marcescens CH-1 results in a phenotype of precocious swarming, enhanced flagella production, and increased flhDC transcription (Soo et al., 2008). In addition, more red pigment was produced in the rssA (CH-1DA) and rssBA

CH-1

/prssA

/prssBA

307

(CH-1DBA) mutants, compared to wild-type CH-1, after incubation at 30 1C for 24 h (Fig. 1A). It is reported that the red pigment is a type of prodiginine, a family of tripyrrole red pigments, and a bacterial secondary metabolite. To quantify the amount of prodigiosin produced by S. marcescens CH-1, CH-1DA, and CH1DBA, red pigment was extracted and quantified by HPLC. Compared with wild-type CH-1, the prodigiosin levels produced by CH-1DA were increased by approximately 27-fold and 4.6-fold after incubation for 24 and 48 h, respectively. In CH-1DBA, prodigiosin levels even increased approximately 67.5-fold and 5.7-fold after incubation for 24 and 48 h, respectively (Fig. 1B) (Wei et al., 2005b). Complementation using plasmids containing rssA or rssBA led to decrease of prodigiosin levels in CH-1DA and CH-1DBA, respectively (Fig. 1B). These results propose that the turnover of prodigiosin may be activated by the RssA/RssB twocomponent system.

Amount of prodigiosin (µg/ml)

RssAB inhibits pig operon expression

CH-1 /prssA

/prssBA

Fig. 1. The red pigment produced by S. marcescens CH-1, CH-1DA, CH-1DA/prssA, CH-1DBA, and CH-1DBA/prssBA were directly observed (A) and quantified using HPLC (B) after incubation at 30 1C for 24 h (A and black bars in B) or 48 h (white bars in B).

The pig cluster is transcribed as a polycistronic message from the pigA promoter, but the intergenic regions upstream of pigA differ between S. marcescens ATCC 274 (Sma 274, accession number AJ833002) and Serratia sp. ATCC 39006 (Serratia 39006, accession number AJ833001) (Slater et al., 2003; Harris et al., 2004). A 407-bp intergenic region is flanked by cueR and pigA in Sma 274, whereas a 922-bp intergenic region is flanked by orfY and pigA in Serratia 39006 (Harris et al., 2004). Using primers specifically designed from the pigA upstream region of Sma 274, a 500-bp DNA fragment was amplified from S. marcescens CH-1 by PCR. In comparison, no DNA fragment was obtained using Serratia 39006 primers. The DNA sequence of the amplified fragment was subsequently identified to be 100% identity to that in Sma 274 (Supplementary Fig. 1). To investigate the effects of rssA or rssBA mutation on the pig cluster in S. marcescens CH-1, the transcriptional activity of pigA was analyzed, as a representative of the pig operon. The amplified upstream region of pigA from CH-1, designated Pa, was ligated with promoterless lacZ to generate a transcriptional fusion (Pa-lacZ) plasmid, pYN1 (Fig. 2). Pa-lacZ expression was measured in the S. marcescens CH-1, CH-1DA, and CH-1DBA backgrounds after incubation for 8 h at 30 1C. The bgalactosidase activity of Pa-lacZ was increased by approximately 49% and 67% in CH-1DA and CH-1DBA, respectively, compared with the parental (CH-1) background (Fig. 3). No significant differences in b-galactosidase activity were evident among the 3 S. marcescens strains carrying the pYN0 control vector (p value o0.05) (data not shown). No differences in growth dynamics were observed among the strains CH-1, CH-1DA, or CH-1DBA

100bp -472

-319

+1 +28 +81

pigA promoter

pigA lacZ Dig

FAM FAM

length Pa

500 bp

Pb

347 bp

Pc

400 bp

Pd

400 bp

Fig. 2. Diagram of DNA fragments containing the pigA promoter. Pa was fused with lacZ to generate the transcriptional fusion construct, pYN1. Pb was labeled with digoxigenin (Dig) (circles), and analyzed with RssB-P using EMSA. The Pc or Pd DNA fragment was labeled with 6 carboxyfluorescein (FAM) (rhombus) at the 5’-end of the coding or template strand, respectively. Numbers indicate the nucleotide positions relative to the translational start site. Hatched box, putative RssB-P binding site analyzed by DNA footprinting.

ARTICLE IN PRESS Y.-T. Horng et al. / International Journal of Medical Microbiology 300 (2010) 304–312

by data analysis using GeneMapper software (Soo et al., 2008). The results showed that the DNA region spanning the 21st to 46th bp upstream of the translation start site of pigA was protected by RssB-P (Fig. 5). In contrast, no regions within the Pc or Pd fragment were protected by RssB(D51E) (Supplementary Fig. 2) , indicating that phosphorylated RssB is essential for such an interaction. The promoter of pigA in Sma274 corresponding to the –35 and –10 regions has been proposed by Harris et al. (2004) (Fig. 5E). The sequence of this region in S. marcescens CH-1 is identical to that in Sma274. Our DNase I footprinting data show that the binding region of RssB-P is located downstream of the pigA promoter (Fig. 5E). Binding of RssB-P may thus obstruct transcription of the pig cluster by RNA polymerase.

160 140

20 0

CH-1/ pYNm

120 100 80 60 40

CH-1/pYN1

Specific pigA activity (Miller unit)

308

Identification of conserved RssB binding sequence

Fig. 3. Effects of rssA and rssB on promoter activity of pigA in S. marcescens CH-1. bgalactosidase activities were measured from lacZ in a plasmid within a wild-type background (CH-1), rssA (CH-1DA), or rssBA (CH-1DBA) knockout mutant strains after incubation of bacteria at 30 1C for 8 h. pYN1, plasmid containing fusion of the pigA promoter and lacZ; pYNm, plasmid with fusion of mutant pigA promoter and lacZ.

containing pYN1 or pYN0 (data not shown). We hypothesis that the hyperpigmentation observed in CH-1DBA was possibly attributable to higher transcription of the prodigiosin biosynthetic genes pig operon in S. marcescens CH-1.

propose a CH-1DA or levels of within the

RssB binds to the region upstream of pigA Previous studies had shown that the RssB which is the cognate response regulator of RssA binds directly to the rssB and flhDC promoter regions at its phosphorylated form (RssB-P) (Wei et al., 2005a; Soo et al., 2008). Whether the RssAB two-component system regulates the transcription of pigA via direct binding of RssB-P to the pigA promoter was next addressed. For comparison, the promoter regions of ygfF (370 bp) and rssB (104 bp) were used as negative and positive controls, respectively (Soo et al., 2008). The 347-bp Pb DNA fragment (0.5 ng) was retarded by RssB-P (Fig. 4A). In a competition assay, addition of the unlabeled Pb DNA fragment (1–50 ng), but not a non-specific control (an unlabeled ygfF fragment), inhibited retardation of Dig-Pb by RssB-P, indicating specific binding of RssB-P to the Pb DNA fragment (Fig. 4C). Earlier studies showed that Asp51 of RssB is an important site for the phosphorelay reaction during RssA–RssB signal transduction (Wei et al., 2005a). Mutated RssB, in which Asp51 was replaced with glutamate, denoted RssB(D51E), could not bind the flhDC promoter (Soo et al., 2008). It is assumed that Asp51 is important for phosphorylation or protein folding of RssB. To confirm that Asp51 in RssB is necessary for binding to the Pb fragment, EMSA was performed with RssB (D51E) (Wei et al., 2005a). The Pb fragment was not retarded by RssB(D51E) (Fig. 4B), clearly indicating that Asp51 in RssB is essential for interaction with the Pb fragment. Thus, it appears that RssB-P specifically binds to the upstream promoter region of pigA. Definition of the specific RssB-P binding site within the pigA upstream region For precise determination of the binding site(s) of RssB-P, DNase I footprinting experiments were performed using an automated capillary DNA sequencer (ABI Prism 3100), followed

Previous investigations had identified that the RssB-P specifically binds to the 63-bp and 24-bp regions of the rssB and flhDC promoter, respectively (Wei et al., 2005a; Soo et al., 2008). The DNase I footprinting analysis of the current study confirms that a 26 bp pigA promoter region is essential for RssB-P binding. To identify the consensus sequence for RssB-P recognition, DNA sequences of the 3 RssB-P binding regions were compiled using the statistical mechanics search program, ClustalW+. Several highly conserved positions were revealed within the 24-bp core sequence, which we hypothesized to be the conserved RssB binding sequence (Table 3). This theoretical assumption was subsequently validated using EMSA. The 86-bp PCR fragment containing the region between positions –58 and + 28, relative to the translational start site of pigA (pigA_wt), was retarded by RssB-P (Table 3, Fig. 6). The DNA fragment, pigA_mut7, containing the 7 mutated nucleotides within the potential RssB-binding consensus sequence were further analyzed (Table 3). In comparison, no DNA shifts were observed in the mixture of RssB-P and pigA_mut7 containing 7 mutated nucleotides, indicating that these conserved nucleotides are important for RssB-P recognition (Table 3, Fig. 6). As ygfF was too large (370 bp) to run EMSA together with pigA_mut7 on the same gel, the promoter regions of csrB (117 bp, accession number is U00096) and rssB (104 bp) were used as negative and positive controls, respectively (Fig. 6). To confirm that the conserved RssB binding sequence was important in vivo, a pigA promoter containing residues mutated in the RssB binding sequence was constructed by site-directed mutagenesis (Table 3). After PCR amplification using the primers PpF1 and PpRm, the 500-bp mutant promoter of pigA was subsequently fused with promoterless lacZ in the pYNm plasmid. The b-galactosidase activity from pYNm in S. marcescens CH-1 was increased by 70% relative to that from pYN1 in S. marcescens CH-1 (Fig. 3). One possible explanation is that RssB-P does not recognize the mutant RssB binding sequence, resulting in elevated transcriptional activity of the pig cluster. Collectively, the conserved RssB binding sequence is important for the regulation of prodigiosin synthesis in S. marcescens CH-1.

Discussion Here, we analyze the mechanism of RssA–RssB-mediated regulation of prodigiosin production. Disruption of the rssA and rssB genes leads to a significant increase in prodigiosin synthesis and transcription of the prodigiosin-synthesizing pig cluster, supporting the theory that the RssB/A two-component system is a negative regulator of prodigiosin production in S. marcescens CH-1 (Figs. 1 and 3). Our results show that one of the RssB target promoters is pigA, the first open reading frame of polycistronic mRNA transcribed from the pig cluster. Phosphorylated RssB

ARTICLE IN PRESS Y.-T. Horng et al. / International Journal of Medical Microbiology 300 (2010) 304–312

RssB-P

RssB-P

Conc.(µM) 0

2

rssB promoter

ygfF promoter

Pb fragment

4

RssB-P

RssB-P

0

2

4

0

309

Pb fragment

RssB RssB-P (D51E)

N

2

Pb fragment

Pb fragment

RssB-P

RssB-P

Unlabeled Pb

Unlabeled ygfF

Fig. 4. Retardation of the S. marcescens CH-1 pigA promoter by RssB. (A) EMSA analysis of the interactions between RssB-P and DNA fragments, such as Pb fragment (Fig. 2), ygfF promoter (negative control), or rssB promoter region (positive control). (B) An aspartate residue (D51) is essential for RssB binding to the Pb fragment. EMSA was performed using the Pb fragment alone or together with 4 mM RssB-P or RssB(D51E) (right to left). (C) Competition assay. A Pb fragment (0.5 ng) without RssB-P or with 4 mM RssB-P was mixed with unlabeled DNA. –, no unlabeled DNA; + , 1 ng unlabeled DNA; ++, 5 ng unlabeled DNA; +++ , 50 ng unlabeled DNA.

binds downstream of the putative –35 and –10 regions of pigA (Figs. 4 and 5). It is envisaged that RssB-P directly represses pigA transcription by interfering with RNA polymerase activity. The consensus RssB binding sequence, GAGATTTTAGCTAAATTAATBTTT, was identified in promoters of rssB, flhDC, and pigA in S. marcescens CH-1. Site-directed mutation of conserved nucleotides within the potential RssB binding sequence of pigA resulted in low binding affinity to RssB-P (Fig. 6) and higher transcriptional activity of pigA (Fig. 3). In addition to the RssB/A two-component system, the PigQ/W two-component signaling system regulates prodigiosin production in Serratia 39006 (Fineran et al., 2005). PigW is predicted as a sensor kinase and PigQ is a response regulator. Because both pigQ and pigW mutants have reduced levels of prodigiosin and pigA transcripts, it was suggested that phosphorylated PigQ activates the transcriptional activity of the pig cluster by an unknown mechanism, resulting in an elevation of the prodigiosin production in Serratia 39006. Besides prodigiosin, the PigQ/W twocomponent signaling system has a positive effect on cellulose synthesis, pectate lyase expression, and motility. However, the RssB/A two-component system negatively regulates not only prodigiosin production but also swarming motility and itself in S. marcescens CH-1 (Wei et al., 2005a; Soo et al., 2008). Phosphorylated RssB directly represses transcription of the pig cluster, flhDC and rssB (Fig. 7).

A marked increase in prodigiosin synthesis was evident in mutants containing deletions in the RssB/A-coding genes (27- and 67.5-fold increases in prodigiosin content in rssA or rssBA mutants, respectively, after 24 h incubation, compared to the wild-type strain; Fig. 1). However, only a slight increase in pig operon transcription was observed in these mutants (49% or 67% with respect to wild-type; Fig. 3). It is possible that the RssB/A two-component system is also involved in post-transcriptional control, or RssB/A indirectly regulates expression of genes involved in pigA transcription. There are considerable variations in the mode of prodigiosin regulation between Serratia 39006 and Sma 274. In Serratia 39006, the N-acyl homoserine lactone (N-AHL) quorum sensing (QS) system controls prodigiosin synthesis via indirect repression of the pig cluster by SmaR, which suppresses expression of the pigment regulators, pigR, rap, and pigQ (Williamson et al., 2006). In contrast, the AI-2 signal synthesized by LuxS is required for maximum prodigiosin production in Sma 274 (Williamson et al., 2006). These distinct regulatory mechanisms in S. marcescens CH-1 and SS-1 were also observed in our laboratory. In SS-1, prodigiosin production is regulated by the N-AHL (3-oxo-C6homoserine lactone and C6-homoserine lactone) QS system, SpnI/R. SpnR represses prodigiosin production during the absence of N-AHL signals, which are synthesized by SpnI at low cell density. At high cell density, accumulating N-AHL signal de-represses the

ARTICLE IN PRESS Y.-T. Horng et al. / International Journal of Medical Microbiology 300 (2010) 304–312

Template strand

Coding strand

Fragment Size (nt)

Fragment Size (nt) 250

50

300

Pd+RssB-P Pd

Fragment Size (nt) 120

Pd+RssB-P Pd

-35

150

200

250

300

Fragment Size (nt) 220

140

Pc+RssB-P

100

Pc

80

Fluorescence intensity (Arbitrary Units)

60

100

Fluorescence intensity (Arbitrary Units)

200

Pc+RssB-P

150

Pc

100

Fluorescence intensity (Arbitrary Units)

50

240

260

280

300 Fluorescence intensity (Arbitrary Units)

310

-10 coding strand template strand coding strand template strand

Fig. 5. Identification of the RssB-P binding site in the pigA promoter region by DNase I footprinting. FAM was employed to label the template strand of Pd (A and C) or coding strand of Pc (B and D) (see Fig. 2). Pd or Pc were left untreated or incubated with RssB-P prior to DNase I treatment. Fluorescence intensity of the FAM-labeled DNA fragment (ordinate) was plotted against fragment length. (C and D) Expanded views of A and B. The horizontal lines in C and D signify the protected regions located from nucleotides (nt) 104–118 of the DNA fragment in C and nt 272–294 of the DNA fragment in D. (E) A partial sequence of the pigA upstream region is shown. The start codon, ATG, of pigA is double-underlined. Nucleotides protected by RssB-P are presented in shadow. The promoter region, including the –10 and –35 regions predicted by Harris et al. (2004), are underlined.

effects of SpnR on prodigiosin production (Horng et al., 2002). Data obtained with biosensors and polymerase chain reaction (PCR) show that S. marcescens CH-1 does not synthesize N-AHL or contain the luxI/R homolog (data not shown). Therefore, it appears that prodigiosin biosynthesis is not regulated by the SpnI/R quorum system in CH-1. In contrast, the RssB/A two-component system appears to inhibit prodigiosin biosynthesis via direct control transcription of the pig cluster in S. marcescens CH-1. Furthermore, significant differences are observed in the regions upstream of pigA between Sma 274 and Serratia 39006. Sma 274 contains a 407-bp DNA fragment flanked by cueR and pigA, whereas a 922-bp DNA fragment is flanked by orfY and pigA in Serratia 39006 (Slater et al., 2003; Harris et al., 2004). PCR and DNA sequencing show that the region upstream of pigA in S. marcescens CH-1 is identical to that in Sma 274. A computational comparison reveals no potential RssB-binding sequence

upstream of pigA in Serratia 39006 (data not shown). Thus, variable regulation modes seem to be operative in distinct Serratia species, even in different S. marcescens isolates. In the further study, it will be checked whether the QS system intersects with RssA/B two-component system or whether it is completely independent. Whereas RssA is the cognate sensor kinase of RssB, prodigiosin production and transcriptional activity of the pig cluster in CH-1DBA were only slightly lower than seen in CH-1DA (Figs. 1 and 3). It is suggested that RssB is phosphorylated in the absence of RssA by a non-cognate sensor kinase or low molecular weight phosphodonors via a process known as ‘‘cross-talk’’ (Bijlsma and Groisman, 2003). In Serratia 39006, PhoR phosphorylates PhoB when signals are relayed by the PstSCABPhoU-Pi transport complex, presumably when Pi is limited, directly binding to Pho boxes in promoters to stimulate transcription of the smaI and pig

ARTICLE IN PRESS Y.-T. Horng et al. / International Journal of Medical Microbiology 300 (2010) 304–312

311

Table 3 Definition of the conserved RssB binding sequence. Upstream of genes

The sequence of the RssB binding region in the upstream of genesa

flhDC rssB pigA Consensusb Potential RssB binding sequencec pigA_mut7d pYNme

——————————————————————GTATGTGTATTTGGCTTAATTATTTTTTATTTTTG—————347—ATG CTGAAATTCTTCCGCGCTCCCTTAACGCGAGATTAAAGCCAGATTAAGGTTTGCTTAAGAACG—106—ATG ——————————————————————ACTAGAAATATTAGCTAATTTAATCTCTCAACCCA——————22—ATG GagattttaGCtaaaTTAat . TtT GAGATTTTAGCTAAATTAATBTTT ACTAGAAATATTATATAATGGGATCACGCAACCCA ACTAGAAATATTAAGTAATGATATCCCGCAACCCA

a The sequence alignment was performed by the program, ClustalW + , through SeqWeb v 3.1.2. The conserved RssB binding sequence is in bold. Numbers indicate nucleotide positions relative to the translational start site of each gene. b The degree of conservation of consensus of 24 represented nucleotides: capital letter ¼ 100%, lowercase letter 466%, dot o 33%. c B¼ C, G or T d The reconstituted 86-bp DNA fragment, pigA_mut7, spanning nucleotide positions –58 to +28 relative to the translational start site of pigA and containing 7 point mutations are in bold. e The plasmid, pYNm, contained lacZ reporter gene and 500-bp pigA promoter region with the mutated potential RssB binding sequence (Fig. 3). The 7 point mutations in te RssB box are in bold.

pigA_wt

pigA_mut7

rssB

csrB

Fig. 6. In vitro analysis of the RssB-P binding site in the pigA promoter. EMSAs were performed with pigA_wt (an 86-bp DNA fragment spanning the sequence from nucleotide positions –56 to + 30, relative to the translational start site of pigA); pigA_mut7 (reconstituted 86-bp pigA fragment with mutation; Table 3); csrB, csrB promoter (negative control); and rssB, rssB promoter (positive control). + , 4 mM RssB-P; –, no RssB-P.

Fig. 7. Schematic representation of RssB/A two-component system in S. marcescens CH-1. RssB-P directly represses transcription of the rssB, flhDC, and pig cluster promoters, respectively. The repression on flhDC leads to reduced production of flagellin, thereby inhibition of swarming phenomenon. The repression of pig cluster, pigA-N, results in preventing the prodigiosin production.

clusters (Williamson et al., 2006). A Pho box is also predicted between the –35 and –10 regions of the pigA promoter in Sma 274 and S. marcescens CH-1 (Harris et al., 2004). Intriguingly, the RssB binding site is located downstream of and close to the –35 and –10 regions of the pigA promoter in S. marcescens CH-1 (Sma 274). The interactions between the PhoB/R and RssB/A two-component system in S. marcescens CH-1 require clarification. A well-characterized phenomenon is the influence of temperature on prodigiosin production in S. marcescens. Prodigiosin is synthesized when bacteria are incubated at a low temperature (27 1C, in minimal or complete medium). However, no prodigiosin is produced at high temperature (38 1C) (Williams et al., 1971). Moreover, S. marcescens CH-1 produces the red pigment at 30 1C, but not at 37 1C, in LB broth upon overnight incubation. We have reported that S. marcescens CH-1 swarms at 30 1C and not at 37 1C. Moreover, RssB/A negatively regulates flagellin synthesis in S. marcescens CH-1 (Soo et al., 2008). A recent report shows that the inhibitory effect of RssB/A at 37 1C is significantly greater than that at 30 1C (Soo et al., 2008). This finding indicates that the effect of temperature on swarming is partly regulated by RssB/A signaling in S. marcescens CH-1. However, in the present study, the S. marcescens mutants CH-1DA and CH-1DBA did not synthesize red pigment at 37 1C in LB broth. It is suggested that at high temperature, prodigiosin production is inhibited via an unknown mechanism, not the RssB/A two-component signaling system. In the footprinting experiment, RssB-P protected a large region of the coding strand, but a shorter region was protected in the template strand (Fig. 5). Comparable results were obtained with footprinting assays of the flhDC promoter (Soo et al., 2008), whereas no footprinting data on the rssB promoter are available. This may be because of different binding conformations between RssB and these promoters. The protein structures of RssB alone and bound to DNA are currently under investigation. The conserved RssB binding site has been proposed in this study. More new RssB target genes and RssB regulon could be characterized by using the consensus sequence to perform transcriptome analysis or bioinformatic search. The relationship between pathogenesis and prodigiosin in Serratia is not clear. However, it is reported that an outbreak of S. marcescens in a neonatal unit at City Hospital, Birmingham, was surveyed. The isolates were divided into 2 clones according to molecular typing. One group of isolates caused invasive clinical infection and was non-pigmented. The other was noninvasive and displayed pink-red pigmentation (David et al., 2006). It is interesting to investigate whether S. marcescens isolates have the RssB/A two-component system. It is possible that RssB

ARTICLE IN PRESS 312

Y.-T. Horng et al. / International Journal of Medical Microbiology 300 (2010) 304–312

negatively regulates not only pig operon but also other virulence genes.

Acknowledgments This work was supported by the grants from the National Science Council (NSC-97-2320-B-320-012-MY3) and Chang Gung Memorial Hospital (CMRP160273) which were really appreciated.

Appendix A. Supplementary material Supplementary data associated with this article can be found in the online version at doi:10.1016/j.ijmm.2010.01.003.

References Adamidis, T., Champness, W., 1992. Genetic analysis of absB, a Streptomyces coelicolor locus involved in global antibiotic regulation. J. Bacteriol. 174, 4622–4628. Adamidis, T., Riggle, P., Champness, W., 1990. Mutations in a new Streptomyces coelicolor locus which globally block antibiotic biosynthesis but not sporulation. J. Bacteriol. 172, 2962–2969. Anderson, T.B., Brian, P., Champness, W.C., 2001. Genetic and transcriptional analysis of absA, an antibiotic gene cluster-linked two-component system that regulates multiple antibiotics in Streptomyces coelicolor. Mol. Microbiol. 39, 553–566. Arakawa, Y., Ike, Y., Nagasawa, M., Shibata, N., Doi, Y., Shibayama, K., Yagi, T., Kurata, T., 2000. Trends in antimicrobial-drug resistance in Japan. Emerg. Infect. Dis. 6, 572–575. Berg, H.C., 2005. Swarming motility: it better be wet. Curr. Biol. 15, R599–R600. Bijlsma, J.J., Groisman, E.A., 2003. Making informed decisions: regulatory interactions between two-component systems. Trends Microbiol. 11, 359–366. Chilcott, G.S., Hughes, K.T., 2000. Coupling of flagellar gene expression to flagellar assembly in Salmonella enterica serovar typhimurium and Escherichia coli. Microbiol. Mol. Biol. Rev. 64, 694–708. David, M.D., Weller, T.M., Lambert, P., Fraise, A.P., 2006. An outbreak of Serratia marcescens on the neonatal unit: a tale of two clones. J. Hosp. Infect. 63, 27–33. Eisenstein, B., 1990. Enterobacteriaceae. In: Mandell, G.L., Douglas, R.G., Bennett, J.E. (Eds.), Principles and Practice of Infectious Disease. Churchill Livingstone, New York. Fineran, P.C., Slater, H., Everson, L., Hughes, K., Salmond, G.P., 2005. Biosynthesis of tripyrrole and beta-lactam secondary metabolites in Serratia: integration of quorum sensing with multiple new regulatory components in the control of prodigiosin and carbapenem antibiotic production. Mol. Microbiol. 56, 1495–1517. Haddix, P.L., Jones, S., Patel, P., Burnham, S., Knights, K., Powell, J.N., LaForm, A., 2008. Kinetic analysis of growth rate, ATP, and pigmentation suggests an energy-spilling function for the pigment prodigiosin of Serratia marcescens. J. Bacteriol. 190, 7453–7463. Harris, A.K., Williamson, N.R., Slater, H., Cox, A., Abbasi, S., Foulds, I., Simonsen, H.T., Leeper, F.J., Salmond, G.P., 2004. The Serratia gene cluster encoding biosynthesis of the red antibiotic, prodigiosin, shows species- and straindependent genome context variation. Microbiology (Reading, England) 150, 3547–3560. Hejazi, A., Falkiner, F.R., 1997. Serratia marcescens. J. Med. Microbiol. 46, 903–912. Horng, Y.T., Deng, S.C., Daykin, M., Soo, P.C., Wei, J.R., Luh, K.T., Ho, S.W., Swift, S., Lai, H.C., Williams, P., 2002. The LuxR family protein SpnR functions as a negative regulator of N-acylhomoserine lactone-dependent quorum sensing in Serratia marcescens. Mol. Microbiol. 45, 1655–1671.

Knowles, S., Herra, C., Devitt, E., O’Brien, A., Mulvihill, E., McCann, S.R., Browne, P., Kennedy, M.J., Keane, C.T., 2000. An outbreak of multiply resistant Serratia marcescens: the importance of persistent carriage. Bone Marrow Transplant. 25, 873–877. Lai, H.C., Soo, P.C., Wei, J.R., Yi, W.C., Liaw, S.J., Horng, Y.T., Lin, S.M., Ho, S.W., Swift, S., Williams, P., 2005. The RssAB two-component signal transduction system in Serratia marcescens regulates swarming motility and cell envelope architecture in response to exogenous saturated fatty acids. J. Bacteriol. 187, 3407–3414. Miller, J.F., 1972. Experiments in Molecular Genetics. Cold Spring Harbor Laboratory Press, New York. Parment, P.A., 1997. The role of Serratia marcescens in soft contact lens associated ocular infections. A review. Acta Ophthalmol. Scand. 75, 67–71. Paruchuri, D.K., Harshey, R.M., 1987. Flagellar variation in Serratia marcescens is associated with color variation. J. Bacteriol. 169, 61–65. Perez-Tomas, R., Montaner, B., Llagostera, E., Soto-Cerrato, V., 2003. The prodigiosins, proapoptotic drugs with anticancer properties. Biochem. Pharmacol. 66, 1447–1452. Rosenberg, M., Blumberger, Y., Judes, H., Bar-Ness, R., Rubinstein, E., Mazor, Y., 1986. Cell surface hydrophobicity of pigmented and nonpigmented clinical Serratia marcescens strains. Infect. Immun. 51, 932–935. Ryding, N.J., Anderson, T.B., Champness, W.C., 2002. Regulation of the Streptomyces coelicolor calcium-dependent antibiotic by absA, encoding a cluster-linked two-component system. J. Bacteriol. 184, 794–805. Slater, H., Crow, M., Everson, L., 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, 303–320. Soo, P.C., Horng, Y.T., Wei, J.R., Shu, J.C., Lu, C.C., Lai, H.C., 2008. Regulation of swarming motility and flhDC(Sm) expression by RssAB signaling in Serratia marcescens. J. Bacteriol. 190, 2496–2504. Stock, A.M., Robinson, V.L., Goudreau, P.N., 2000. Two-component signal transduction. Annu. Rev. Biochem. 69, 183–215. Stock, I., Grueger, T., Wiedemann, B., 2003. Natural antibiotic susceptibility of strains of Serratia marcescens and the S. liquefaciens complex: S. liquefaciens sensu stricto, S. proteamaculans and S. grimesii. Int. J. Antimicrob. Agents 22, 35–47. Tanikawa, T., Nakagawa, Y., Matsuyama, T., 2006. Transcriptional downregulator hexS controlling prodigiosin and serrawettin W1 biosynthesis in Serratia marcescens. Microbiol. Immunol. 50, 587–596. Traub, W.H., 2000. Antibiotic susceptibility of Serratia marcescens and Serratia liquefaciens. Chemotherapy 46, 315–321. Vieira, J., Messing, J., 1991. New pUC-derived cloning vectors with different selectable markers and DNA replication origins. Gene 100, 189–194. von Graevenitz, A., 1980. Infection and colonization with Serratia. In: von Graevenitz, A., Rubin, S.J. (Eds.), The Genus Serratia.. CRC Press, Boca Raton, FL, pp. 167–185. Wei, J.R., Tsai, Y.H., Soo, P.C., Horng, Y.T., Hsieh, S.C., Ho, S.W., Lai, H.C., 2005a. Biochemical characterization of RssA–RssB: a two-component signal transduction system regulating swarming behavior in Serratia marcescens. J. Bacteriol. 187, 5683–5690. Wei, Y.H., Yu, W.J., Chen, W.C., 2005b. Enhanced undecylprodigiosin production from Serratia marcescens SS-1 by medium formulation and amino-acid supplementation. J. Biosci. Bioeng. 100, 466–471. Whitehead, N.A., Barnard, A.M., Slater, H., Simpson, N.J., Salmond, G.P., 2001. Quorum-sensing in Gram-negative bacteria. FEMS Microbiol. Rev. 25, 365–404. Williams, R.P., Gott, C.L., Qadri, S.M., Scott, R.H., 1971. Influence of temperature of incubation and type of growth medium on pigmentation in Serratia marcescens. J. Bacteriol. 106, 438–443. Williamson, N.R., Fineran, P.C., Leeper, F.J., Salmond, G.P., 2006. The biosynthesis and regulation of bacterial prodiginines. Nat. Rev. Microbiol. 4, 887–899. Williamson, N.R., Simonsen, H.T., Ahmed, R.A., Goldet, G., Slater, H., Woodley, L., Leeper, F.J., Salmond, G.P., 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, 971–989.