Research in Microbiology 158 (2007) 150e158 www.elsevier.com/locate/resmic
Characterization of a luxI/luxR-type quorum sensing system and N-acyl-homoserine lactone-dependent regulation of exo-enzyme and antibacterial component production in Serratia plymuthica RVH1 Rob Van Houdt a,1, Pieter Moons a, Abram Aertsen a, An Jansen a,2, Kristof Vanoirbeek a, Mavis Daykin b, Paul Williams b, Chris W. Michiels a,* a
b
Laboratory of Food Microbiology, Katholieke Universiteit Leuven, Kasteelpark Arenberg 22, B-3001 Leuven, Belgium Institute of Infection, Immunity and Inflammation, Centre for Biomolecular Sciences, University of Nottingham, Nottingham, NG7 2RD, UK Received 31 July 2006; accepted 25 November 2006 Available online 29 December 2006
Abstract Quorum sensing by means of N-acyl-L-homoserine lactones (AHLs) is widespread in Gram-negative bacteria, where diverse AHLs influence a wide variety of functions, even in a single genus such as Serratia. Here we report the identification and characterization of the quorum sensing system of Serratia plymuthica strain RVH1. This strain isolated from a raw vegetable processing line produces at least three AHLs which were identified as N-butanoyl- (C4-HSL), N-hexanoyl- (C6-HSL) and N-(3-oxo-hexanoyl)-homoserine lactone (3-oxo-C6-HSL). The identified LuxI homolog SplI synthesizes 3-oxo-C6-HSL, and influences the production of C4-HSL and C6-HSL, as splI gene inactivation resulted in loss of 3oxo-C6-HSL production and smaller amounts of C4-HSL and C6-HSL produced. SplI-dependent quorum sensing controls 2,3-butanediol fermentation (previously reported) and the production of an extracellular chitinase, nuclease, protease and antibacterial compound. The identity of the latter is not yet elucidated, but appears to be different from the known antibacterial compounds produced by Serratia strains. SplR, the homolog of the LuxR regulator, appears to act as a repressor of synthesis of extracellular enzymes and antibacterial compound and to autorepress its own expression, probably by binding to a 21 bp lux box sequence. Ó 2006 Elsevier Masson SAS. All rights reserved. Keywords: Quorum sensing; Serratia; N-acyl-L-homoserine lactone
1. Introduction Quorum sensing is a widespread signalling mechanism that allows bacteria to activate or repress specific target genes in * Corresponding author. Laboratory of Food Microbiology, Katholieke Universiteit Leuven, Kasteelpark Arenberg 22, B-3001 Leuven, Belgium. Tel.: þ32 16 321578; fax: þ32 16 321960. E-mail addresses:
[email protected] (R. Van Houdt), Chris.
[email protected] (C.W. Michiels). 1 Present address: Molecular and Cellular Biology, Institute for Health, Environmental and Safety, Belgian Nuclear Research Centre (SCK*CEN), Boeretang 200, B-2400 Mol, Belgium. 2 Present address: Whitehead Institute for Biomedical Research, Massachusetts Institute of Technology, Nine Cambridge Center, Cambridge, Massachusetts 02142, USA. 0923-2508/$ - see front matter Ó 2006 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.resmic.2006.11.008
response to population density. Several signalling systems have been characterized, but the predominant system in Gram-negative bacteria is based on N-acyl-L-homoserine lactones (AHLs). AHLs are a group of small diffusible signalling molecules which can differ in the length or substitution of their acyl side chains. AHL-dependent communication was first described in the marine bacterium Vibrio fischeri, in which the transcriptional regulator LuxR, upon binding of the signalling molecule N-(3-oxohexanoyl)-L-homoserine lactone synthesized by the LuxI protein, activates the lux operon, resulting in a bioluminescent phenotype when the bacterial population exceeds a certain threshold [6]. Since its discovery, AHL-mediated cell-to-cell signalling has been observed in Gram-negative bacteria in many different environments and has been shown to play a role in diverse processes such as
R. Van Houdt et al. / Research in Microbiology 158 (2007) 150e158
pathogenesis, biofilm development, food spoilage and resistance to antimicrobial compounds (for review see Refs. [20,33]). In Serratia spp. different AHL production profiles and target genes have been described, and these studies illustrate the diversity and specificity of quorum sensing signalling molecules and of the regulatory pathways involved, even in a single genus. Phenotypes reported to be regulated in an AHL-dependent manner are a lipB-encoded secretion system required in the production of extracellular lipolytic and proteolytic activities for Serratia proteamaculans B5a [5], swarmer cell differentiation [7], the lipB-secretion system [21], formation of cell chains and differentiated long filamentous cells implicated in biofilm formation for Serratia marcescens MG1 (previously Serratia liquefaciens MG1) [16], production of the red pigment prodigiosin, nuclease NucA, a biosurfactant for S. marcescens SS-1 [13], production of the b-lactam antibiotic 1-carbapen-2-em-3-carboxylic acid, prodigiosin and the exo-enzymes pectate lyase and cellulase for Serratia sp. ATCC 30096 [23]. Recently, we reported that 2,3-butanediol fermentation is dependent on AHL quorum sensing in S. plymuthica RVH1, and although 2,3-butanediol fermentation has been well studied, its regulation by quorum sensing is unprecedented [30]. This interesting AHL-dependent phenotype, together with the observed production of a seemingly novel antibacterial compound, motivated us to isolate and sequence the luxI and luxR homologs and to characterize the AHLs produced by this bacterial strain which was isolated from a raw vegetable processing line as previously described [28,29]. Furthermore, this is the first report describing the production and quorum-dependent regulation of an extracellular protease, chitinase, nuclease and antibacterial compound in S. plymuthica.
2. Materials and methods 2.1. Strains, plasmids, culture media, chemicals and oligonucleotides Strains, plasmids and cosmids used in this study are listed in Table 1. Escherichia coli strains were cultured at 37 C, S. plymuthica and Chromobacterium violaceum CV026 at 30 C in Luria-Bertani (LB) broth or agar (1.5% agar), or in M9 minimal medium. The following concentrations of antibiotics were used as needed: ampicillin, 100 mg/ml; kanamycin, 50 mg/ml; tetracycline, 20 mg/ml; chloramphenicol, 30 mg/ml; gentamicin, 20 mg/ml. Carbenicillin (200 mg/ml) was used instead of ampicillin for S. plymuthica strains. The synthetic AHLs, N-butanoyl-L-homoserine lactone (C4-HSL), N-hexanoyl-L-homoserine lactone (C6-HSL) and N-(3-oxohexanoyl)-L-homoserine lactone (3-oxo-C6-HSL), were either synthesized as described before [4] or purchased from Sigma (Bornem, Belgium). The oligonucleotides used in this study are listed in Table 2 and were synthesized by Eurogentec (Seraing, Belgium). Sequencing was performed by MWGBiotech AG (Ebersberg, Germany).
151
2.2. Construction and screening of S. plymuthica RVH1 cosmid library for luxI/R homologs A pWEB cosmid library (Epicentre, BIOzymTC, Landgraaf, The Netherlands) of S. plymuthica RVH1 was constructed in host strain E. coli MT102 carrying the AHL reporter plasmid pJBA132. Seven hundred independent clones were screened for functional luxI homologs by measuring green fluorescent protein (Gfp) production in a fluorescence microplate reader. Cosmids able to induce pJBA132 were subcloned into pUC18 and DNA fragments capable of inducing Gfp production by pJBA132 sequenced with M13 rev (29) and M13 uni (21) sequencing primers (see Table 2). 2.3. Construction of S. plymuthica RVH1 splI and splR insertion mutants The splI gene (S. plymuthica RVH1 luxI homolog) was PCR-amplified using primers FW-I and REV-I, and cloned into pUC18, resulting in pRVH10. Next, splI was mutagenized by insertion of Tn10 using lNK1324 [15], or of a gentamicinresistance (aacC1) cassette from pMS255, resulting in, respectively, pRVH11 and pRVH12. Subsequently, these mutated alleles were transferred to the suicide plasmid pSF100, resulting, respectively, in pRVH13 and pRVH14. For allelic exchange, pRVH13 and pRVH14 were conjugated from E. coli S17-1 lpir into S. plymuthica RVH1 and exconjugants were selected, confirmed by bioassays and PCR to have the desired chromosomal mutation, and designated S. plymuthica RVH1-1 (splI::aacC1) and RVH1-11 (splI::Tn10). For construction of the splR knockout, splR was amplified with the primers FW-R and REV-R and cloned into pUC18, resulting in pRVH15. Next, a gentamicin-resistance cassette from pMS266 was inserted into pRVH15, resulting in pRVH16. Finally, splR::aacC1 was transferred to pSF100 (pRVH17), and exchanged into S. plymuthica RVH1 as described above. The resulting splR::aacC1 mutant was designated S. plymuthica RVH1-2. A splI::Tn10esplR::aacC1 double mutant (RVH1-3) was created by changing the splR mutant allele (pRVH17) into S. plymuthica RVH1-11 as described above. 2.4. Construction of splI and splR promoter fusions The splR gene together with its upstream lux box sequence were amplified with the primers FW-R and REV-Rþ and ligated into the low copy vector pFPV25, resulting in pRVH18. Reporter fusions of splR lacking the lux box were constructed by amplifying the splR gene with primers FW-R and REV-R and cloning the product into pFPV25, resulting in pRVH19, and into a high copy vector, resulting in pRVH20. A reporter fusion of splI was created by cloning the amplification product obtained with FW-I and REV-I into pFPV25, resulting in pRVH21. 2.5. AHL bioassays, isolation and identification of AHLs AHL production was evaluated with biosensor strains C. violaceum CV026 [18] E. coli MG1655 (pJBA89) [1], E. coli
152
R. Van Houdt et al. / Research in Microbiology 158 (2007) 150e158
Table 1 Strains and plasmids used in this study Strain or plasmid
Relevant genotype or description
Reference
Chromobacterium violaceum CV026
cviI::mini-Tn5 derivative of ATCC 31532, KmR, AHL
[18]
þ F F80d lacZDM15 D(lacZYAeargF )U169 deoR recA1 endA1 hsdR17(r k mk ) supE44 l thi-1 girA96 relA1 recA1 endA1 gyrA96 thi hsdR17 supE44 relA1 D(laceproAB) mcrA [F0 traD36 proAB lacIq lacZ DM15] F l ilvG rfb-50 rph-1 F thi araD139 ara-leuD7679 D(lacIOPZY ) galU gal0 K r mþ, SmR recA, thi, pro, hsdRMþRP4: 2-Tc:Mu: Km Tn7 l pir, SmR
L.C.
[12] [1] L.C.
Serratia plymuthica RVH1 RVH1-1 RVH1-11 RVH1-2 RVH1-3
Natural isolate RVH1 splI::aacC1, GmR RVH1 splI::Tn10, CmR RVH1 splR::aacC1, GmR RVH1 splI::Tn10, splR::aacC1, CmR, GmR
[29] This This This This
Plasmids pFPV25 pJBA132 pJBA89 pMS255 pMS266 pRVH1 pRVH2 pRVH10 pRVH11 pRVH12 pRVH13 pRVH14 pRVH15 pRVH16 pRVH17 pRVH18 pRVH19 pRVH20 pRVH21 pSB1075 pSB403 pSB536 pSF100 pUC18 pUT mini-Tn5 luxCDABE pWEB
colE1 mob, promoterless gfp, ApR luxR-PluxI::gfp, p15A ori, TcR luxR-PluxI::gfp, ColE1 ori, ApR Gentamicin-resistance cassette, ApR, GmR Gentamicin-resistance cassette, ApR, GmR Cosmid, which induces pJBA132, ApR Cosmid, which induces pJBA132, ApR splI cloned into pUC18, ApR splI::Tn10 in pUC18, ApR, CmR splI::aacC1 in pUC18, ApR, GmR splI::Tn10 in pSF100, ApR, KmR, CmR splI::aacC1 in pSF100, ApR, KmR, GmR splR cloned into pUC18, ApR splR::aacC1 in pUC18, ApR, GmR splR::aacC1 in pSF100, ApR, KmR, GmR PsplR::gfp, ApR PsplR::gfp (no PsplR lux box), ApR PsplR::luxCDABE (no PsplR lux box), ApR, KmR PsplI::gfp, ApR lasR-PlasI::luxCDABE, ApR luxR-PluxI::luxCDABE, TcR ahyR-PahyI::luxCDABE, ApR oriR6K mobRP4 ApR, KmR Cloning vector, ApR oriR6K mobRP4, mini-Tn5 luxCDABE Km2 transposon, ApR, KmR Cosmid vector, ApR
[27] [1] [1] [3] [3] This study This study This study This study This study This study This study This study This study This study This study This study This study This study [34] [34] [24] [22] L.C. [35] Epicentre
Escherichia coli DH5a JM109 MG1655 MT102 S17-1 lpir
[1]
study study study study
wt, Wild-type; Ap, ampicillin; Cm, chloramphenicol; Gm, gentamicin; Km, kanamycin; Sm, streptomycin; Tc, tetracycline; L.C., laboratory collection.
JM109 (pSB403) [34], E. coli JM109 (pSB536) [24] and E. coli JM109 (pSB1075) [34]. The most general test was a cross-feeding assay in which the strains to be tested were stabbed into LB soft agar (0.7%) seeded with C. violaceum CV026 and purple pigment production was followed during incubation at 30 C. To obtain preliminary information on the chemical nature and spectrum of AHLs produced by S. plymuthica, thin-layer chromatography (TLC) of organic solvent extracts of cellfree culture supernatants from stationary-phase cultures of S. plymuthica was undertaken as described by McClean et al. [18] for short-chain AHLs and by Yates et al. [36] for long chain AHLs. After chromatography, TLC plates were
dried and overlaid with a thin film of one of the biosensor strains in 1.4% LB agar, and monitoring the appearance of spots indicating induction of the purple pigment violacein, of fluorescence or of bioluminescence, depending on the biosensor strain used. Bioluminescence was observed with a Berthold NightOWL LB 981 (EG&G Wallac Berthold, Wildbad, Germany) or a Luminograph LB980 (Berthold) photon video camera. Green fluorescent protein (Gfp) was observed with an illuminator equipped with appropriate emission and excitation filters. For full chemical characterization, cell-free culture supernatants (3 l) from stationary-phase cultures of S. plymuthica RVH1 grown in M9 minimal medium were extracted with
R. Van Houdt et al. / Research in Microbiology 158 (2007) 150e158 Table 2 Primers used in this study
153
249 amino acids with a predicted molecular mass of 28.01 kD (GenBank accession no. AAR32909).
Primer name
Primer sequence (from 50 to 30 )
FW-I REV-I SEQ-R FW-R REV-R REV-Rþ M13 uni (21) M13 rev (29)
TTGGCTGCAGTGTGTTCGCATGACCG CCTCTCTAGAACGGACGCAGACAAACCCA TGGCCCAATACAGCACCTCAa TGTTGAGCTCTCGCTGCCGGTGTAATAAGT GGGCTCTAGACGGGCTATAATTCGTAAG TATAGAATTCAAGTGTCCCGGCCAGCAGC TGTAAAACGACGGCCAGT CAGGAAACAGCTATGACC
Restriction sites introduced for cloning are italicized. a Primer complementary to 30 end of luxR used for sequencing only.
dichloromethane and the dried residue dissolved in acetonitrile (100 ml). The HPLC and MS (LC-MS) methodology described by Laue et al. [17] was used to identify unequivocally the AHLs produced by S. plymuthica. Retention times and spectral profiles were compared with those of synthetic AHL standards subjected to the same conditions. 2.6. Assay of chitinase, protease, nuclease and antibacterial compound For detection of chitinase activity, a semi-minimal medium (0.5 g NaCl, 1.62 g nutrient broth, 6 g M9 salts l1) supplemented with 0.2% colloidal chitin was used. Nuclease and protease activities were analyzed by spotting 10 ml from an overnight LB broth culture, respectively, on DNase agar (Oxoid, Drongen, Belgium) and on LB agar supplemented with 10% skimmed milk. Enzyme activity was reported as the diameter of the clearing zone minus the colony diameter after 24 h (protease and nuclease) or 48 h (chitinase) incubation at 30 C. The production of antibacterial compounds was analyzed by stab inoculating in LB agar plates (0.7% agar) seeded with E. coli MG1655 as an indicator and comparing zones of inhibition after 24 h incubation at 30 C. 3. Results 3.1. Cloning and sequencing of the luxI and luxR homologs from S. plymuthica RVH1 Two cosmids, pRVH1 and pRVH2, from an S. plymuthica RVH1 library (consisting of 700 clones with an average insert size of 40 kbp, thus covering a 5 Mbp genome with 99.6% probability), induced the reporter plasmid pJBA132. After subcloning, sequence analysis of a 2203 bp DNA fragment revealed two open reading frames (ORFs) corresponding to the presumed AHL synthase, splI, and the response regulator, splR (DDBJ/EMBL/GenBank accession number AY394723). The splI and splR ORFs are oriented convergently, overlapping by 23 bp in their 30 regions, and encoding putative proteins of, respectively, 210 amino acids with a predicted molecular mass of 24.21 kD (GenBank accession no. AAR32908) and
3.2. AHLs produced by S. plymuthica RVH1 and its splI and splR mutants TLC chromatograms of extracts from S. plymuthica RVH1 cell-free culture supernatants revealed the presence of several different AHLs. In an overlay with C. violaceum CV026, two spots were apparent, and the retention factor (Rf) of the slowest migrating spot corresponded to that of synthetic C6-HSL (TLC data not shown; see Fig. 1A, third spectrum). The AHL biosensors E. coli JM109 (pSB403) and E. coli MG1655 (pJBA89) revealed a large spot with a long tail, characteristically produced by 3-oxo derivatives and probably overlapping the spot corresponding to C6-HSL. Judged by its migration, which is further than for C6-HSL, this spot could be 3-oxo-C6-HSL (see Fig. 1A, second spectrum). The unmigrated spot at the loading point could indicate the presence of an AHL with a long acyl chain, but such an AHL should not be able to elicit violacein production. In addition, the E. coli JM109 (pSB536) biosensor produced a single spot (see Fig. 1A, first spectrum), which is likely to correspond to C4-HSL since this biosensor is highly specific for this AHL [24]. Due to the very small difference in Rf between C4-HSL and 3-oxo-C6-HSL, both compounds appear as one spot on the C. violaceum CV026 overlay. From these preliminary data, it is likely that S. plymuthica RVH1 produces C4-HSL, C6-HSL, and 3-oxo-C6-HSL. To unequivocally identify the AHLs produced, the crude supernatant extract was separated into seven fractions (F1eF7) by preparative reverse-phase HPLC using a linear gradient of acetonitrile in water, and the presence of AHLs in these fractions was analyzed by TLC and overlay with C. violaceum CV026 or E. coli (pSB536). Fraction F2 contained two compounds which co-eluted with synthetic C4-HSL and 3oxo-C6-HSL in HPLC (retention times of 5.3 and 6.1 min, respectively). After LC-MS, the spectrum of the first compound revealed the presence of a molecular ion [M þ H] of 172 and a profile of breakdown products including the [M þ H] 102 fragment characteristic of the homoserine lactone (HSL) moiety, indistinguishable from the synthetic C4-HSL standard (Fig. 1). The second compound had a molecular ion [M þ H] of 214 and also the [M þ H] 102 breakdown product (Fig. 1). In fraction F3 [11e17 min], a single active compound with a retention time of 9.8 min was identified, the spectrum of which revealed a molecular ion [M þ H] of 200 and the [M þ H] 102 breakdown product, indistinguishable from the synthetic C6-HSL standard (Fig. 1). Thus, S. plymuthica produces at least three different AHLs, C4-HSL, C6-HSL and 3-oxo-C6-HSL. A comparison of AHL production in S. plymuthica RVH1, RVH1-1 (splI::aacC1) and RVH1-2 (splR::aacC1) by crossfeeding assays showed that overall production of AHLs was reduced by knock-out of splI but remained unaffected by knock-out of splR (data not shown). When, for a more sensitive analysis, the total amount of AHLs in 10 ml overnight
154
R. Van Houdt et al. / Research in Microbiology 158 (2007) 150e158
Fig. 1. ES-MS spectrum obtained for (A) compounds purified from spent culture supernatant of S. plymuthica RVH1 and (B) synthetic AHL molecules.
culture in LB was extracted with ethyl acetate, separated by TLC, and subsequently overlaid with C. violaceum CV026, two spots at the same position as the wild-type spots could still be detected for the splI mutant, but clearly of lower intensity, while the splR mutant produced similar levels to the parent strain (data not shown). TLC separation combined with E. coli MG1655 (pJBA89) detection revealed only one untailed spot corresponding to C6-HSL for the splI mutant and a tailing spot for the splR mutant, similar to wild-type. The splI splR double mutant was also analyzed in this way with both reporter strains and produced the same AHL profiles as the splI mutant (data not shown). Thus, our results are consistent with a direct involvement of SplI in the synthesis of 3-oxo-C6HSL since its inactivation resulted in loss of 3-oxo-C6-HSL production. On the other hand, the residual levels of C4HSL and C6-HSL in a splI mutant suggest the possible existence of an additional AHL synthase.
3.3. Phenotypes regulated by quorum sensing Next, the role of the splIR system in S. plymuthica RVH1 was investigated. The production of extracellular chitinase, protease and nuclease in RVH1 was reduced in the splI mutant, and could be restored by the addition of 10 mM C6-HSL or 3oxo-C6-HSL (Fig. 2), but not C4-HSL (data not shown). No difference or a slightly increased production was observed in S. plymuthica RVH1-2 (splR::aacC1) and RVH1-3 (splI::Tn10 splR::aacC1) compared to strain RVH1, suggesting that SplR may act as a repressor rather than as an activator. The minimum amount of inducer needed to elicit a marginal increase in protease activity in the splI mutant was 0.5 mM for C6-HSL and 0.05 mM for 3-oxo-C6-HSL, confirming that 3-oxo-C6-HSL is probably the cognate AHL for the identified splIR system. The production of an antibacterial compound by S. plymuthica RVH1 was demonstrated by the formation of an
R. Van Houdt et al. / Research in Microbiology 158 (2007) 150e158
clearing zone (mm)
7 6 5 4 3 2 1 ND
0 RVH1
RVH1-1
RVH1-1 C4-HSL
RVH1-1 RVH1-1 RVH1-2 C6-HSL 3-oxo-C6HSL
RVH1-3
Fig. 2. Results of colony spot test for nuclease (white), protease (light grey) and chitinase (dark grey) activity in S. plymuthica RVH1, RVH1-1 (splI::aacC1), RVH1-2 (splR::aacC1) and RVH1-3 (splI::Tn10 splR::aacC1) strains. C4-HSL, C6-HSL and 3-oxo-C6-HSL were added at 10 mM (ND, not determined).
inhibition zone when this strain was stabbed into LB soft agar seeded with E. coli MG1655. Furthermore, production of the compound was regulated by quorum sensing, since it was abolished in RVH1-1 (splI::aacC1) and could be restored by the addition of 10 mM 3-oxo-C6-HSL or C6-HSL but not by C4-HSL (data not shown). As was the case for enzyme production, no difference or a slightly increased production of antibacterial compound was observed in S. plymuthica RVH1-2 (splR::aacC1) and RVH1-3 (splI::Tn10 splR::aacC1) compared to strain RVH1, and 0.5 mM C6-HSL or 0.05 mM 3oxo-C6-HSL was sufficient to observe increased activity in the splI mutant. 3.4. SplR autorepresses its own expression in an AHL-dependent manner but does not control AHL production Many genes regulated via AHL-dependent quorum sensing contain a lux box-like sequence in their promoter regions. A 21-bp region of dyad symmetry was found between position 139 and 159 upstream of the splR start codon, which matches 19 of the 20 nucleotides of the lux box sequences found in V. fischeri and S. marcescens (Table 3). This suggested that the expression of splR is AHL-dependent. Fig. 3A shows that this is indeed the case, since a plasmid-borne splR::gfp fusion was poorly expressed in a splI mutant when compared with a splR mutant, splI splR double mutant or parent strain. However, splR expression in the splI mutant could be fully restored by providing either 3-oxo-C6-HSL at 10 mM or
Table 3 Comparison of the lux box sequence upstream of splR with known lux box sequences Species
Gene
lux Box sequence
S. plymuthica RVH1 S. marcescens SS-1 V. fischeri P. stewartii
splR spnR luxI esaR
ACCTGACCGTACAGGTCAGGT ACCTGACCG-AAAGGTCAGGT ACCTGTAGG-ATCGTACAGGT GCCTGTAC TATAGTGCAGGT
Conserved sequences are highlighted.
155
C6-HSL at 100 mM, but not C4-HSL (Fig. 3B). Fig. 3A also indicates that splR expression is affected by the SplR protein, since higher expression is observed in the splR and splI splR double mutant than in the parent strain. Growth analysis in LB broth showed no differences for the different strains carrying a plasmid-based fusion and therefore it was not necessary to correct the relative fluorescence units and relative light units in Fig. 3 for optical density. In addition, the splR::gfp reporter fusion lacking the lux box (pRVH19) was found to be expressed poorly (comparable with pRVH18 in RVH1-1 see Fig. 3A) and at the same levels in the parent strain, splR mutant, splI splR double mutant and splI mutant with or without addition of 3-oxo-C6HSL, C6-HSL or C4-HSL. To increase the signal output of the reporter and thus allow more sensitive analysis, we also used a high copy plasmid-based splR::luxCDABE reporter fusion lacking the lux box (pRVH20). Also, this construct was found to be expressed at the same levels in the parent, splR mutant and splI mutant. Although no differences in splR expression were observed when the splR reporter lacked the lux box, the expression level may be biased, since promoter prediction (analyzed by Prokaryotic Promoter Prediction [http://bioinformatics. biol.rug.nl] and Virtual Footprint [http://prodoric.tu-bs.de/vfp]) indicated that primer REV-R used for these constructs only included 5 bp upstream from the putative 10 element. The promoter prediction is shown in Fig. 4 for splR and spnR of S. marcescens SS-1 and include the putative lux box (bold), promoter (boxed) and 10 element (overlined). These data clearly show that SplR represses splR expression in the absence of SplI (AHLs), probably by interacting with a lux box, since there is little expression of splR in the splI mutant and splR expression is only induced in a splI mutant when exogenous AHLs are added. At least up to position 554 relative to the start codon, no obvious lux box was apparent upstream of splI. The splI-gfp reporter fusion (pRVH21) exhibited comparable expression levels throughout growth in the S. plymuthica RVH1 parent strain, RVH1-1 (splI::aacC1) and RVH1-2 (splR::aacC1). In addition, splI expression in the splI mutant was unaffected by 3-oxo-C6-HSL, C6-HSL or C4-HSL when provided at 10 mM (data not shown). Therefore, the expression of splI and production of AHLs is independent of splR. 4. Discussion In this study, we have conducted a detailed genetic and functional analysis of the quorum-sensing system in S. plymuthica RVH1. We identified the quorum-sensing regulon using a method based on that described by Swift et al. [25], adapted for a cosmid-based instead of a plasmid-based library. Construction of the cosmid library in the host strain E. coli MT102 carrying pJBA132, provided a simple screening tool for genes responsible for AHL synthesis due to the easy detection of green fluorescent protein and the relatively small number of cosmid clones needed to cover the complete genome. In S. plymuthica RVH1, the luxI and luxR homologs are organized convergently as in other Serratia species and in most g-proteobacteria [9], but different from the situation in V. fischeri, where these genes are
R. Van Houdt et al. / Research in Microbiology 158 (2007) 150e158
156 14
A
B
12
RFU
10 8 6 4 2 5
10
15
20
25
time (h)
30
5
10
15
20
25
30
time (h)
Fig. 3. Activity of the splR::gfp reporter on pRVH18 (comprising the splR promotor lux box) over a 30-h growth period in various backgrounds: (A) in S. plymuthica RVH1 [solid circles], RVH1-1 (splI::aacC1) [open circles], RVH1-2 (splR::aacC1) [solid squares] and RVH1-3 (splI::Tn10 splR::aacC1) [open squares]; (B) in RVH1-1 (splI::aacC1) in the presence of 100 mM C4-HSL [open circles], 100 mM C6-HSL [black squares] and 10 mM 3-oxo-C6-HSL [solid circles], and in RVH13 (splI::Tn10 splR::aacC1) in the presence of 100 mM 3-oxo-C6-HSL [open squares]. Average results for at least three independent experiments are shown.
transcribed divergently [8]. In bacteria where this convergent organization occurs, luxI expression tends not to be controlled by luxR [2], and this was also confirmed in our work. The SplI protein is shown in this work to direct the synthesis of 3-oxo-C6-HSL. Its closest relatives in the genus Serratia are SplI (S. plymuthica HRO-C48), SprI (S. proteamaculans B5a) and SpnI (S. marcescens SS-1) and this group is more closely related to EsaI from Pantoea stewartii than to SmaI and SwrI, two other Serratia LuxI representatives (data not shown). This EsaI group has a threonine residue at position 140 (EsaI numbering), believed to be characteristic for LuxI homologs synthesizing 3-oxo substituted AHLs [32]. In contrast, SmaI and SwrI have an alanine residue at this position and produce AHLs without 3-oxo substituent. Inactivation of the splI gene resulted in the loss of 3-oxo-C6-HSL production, while C4-HSL and C6-HSL production was reduced, indicating that there may be another AHL synthase of the other type present. A similar situation has been reported in S. marcescens SS-1 [13]. Addition of synthetic 3-oxo-C6-HSL could not restore parental levels of C4-HSL and C6-HSL in the splI mutant (data not shown), suggesting that 3-oxo-C6-HSL does not induce this putative second AHL synthase and that C4HSL and C6-HSL are also synthesized by SplI. To the best of our knowledge, RVH1 is the first S. plymuthica strain in which AHL quorum sensing has been studied in
any detail. At least five phenotypes are AHL-dependent, including 2,3-butanediol fermentation [30], the production of extracellular chitinase, nuclease and protease and the production of an unidentified antibacterial compound. Extracellular enzymes are under control of quorum sensing in many bacteria including some Serratia species, and have been implicated in pathogenesis of both plant and animal hosts [10], and in food spoilage [5]. Also the production of antibacterial compounds has been documented in Serratia strains. Examples are bacteriocin L and 28b produced by most S. marcescens biotypes [11]; serracin P, a phage-tail-like bacteriocin, produced by S. plymuthica J7 [14]; and the quorum sensing-regulated carbapenem, 1-carbapen-2-em-3-carboxylic acid, identified in Serratia sp. ATCC 30096 [26]. Preliminary characterization of antibacterial compound produced by S. plymuthica RVH1 indicated that it is likely to be a protein with a molecular weight of at least 30,000 Da, but with different antibacterial spectrum and properties than the aforementioned bacteriocins. Therefore, we have currently begun a more thorough identification. Our data strongly suggest that SplR acts as a repressor, since the reduced activity of the extracellular enzymes and antibacterial compound in the absence of 3-oxo-C6-HSL (splI mutant) could be fully or partly restored by inactivation of SplR (splI splR mutant). In addition, SplR itself was found to be a target gene of the splIR system and represses its own
Fig. 4. Alignment of the S. plymuthica RVH1 and S. marcescens SS-1 regions upstream of the start codons of the luxR homolog (italic) including promoter prediction (boxed), putative lux box (bold) and 10 element (overlined).
R. Van Houdt et al. / Research in Microbiology 158 (2007) 150e158
expression, probably by interacting with a 21 bp region of dyad symmetry upstream of the splR start codon, which is highly similar to previous reported lux box sequences. Contrary to LuxR-type signal-responsive transcriptional activators, a subfamily of LuxR-type regulators including EsaR of P. stewartii, CarR and ExpR of Erwinia carotovora ssp. carotovora (Ecc) bind DNA in the absence of AHLs [33]. EsaR has a well characterized activity in P. stewartii and represses capsular polysaccharide synthesis as well as its own expression, until a sufficient level of 3-oxo-N-hexanoylL-homoserine lactone accumulates to release it from its target DNA binding site [19,31]. LuxR proteins that act as a repressor have also been demonstrated in Serratia species, such as SpnR of S. marcescens SS-1 [13], SmaR of Serratia sp. ATCC 30096 [23] and SprR of S. proteamaculans B5a [5]. However, only in S. plymuthica RVH1 and S. marcescens SS-1 has a lux box sequence upstream of, respectively, splR and spnR been demonstrated. Remarkably, SpnR autoactivates its own transcription [13], whereas SplR autorepresses its own transcription, perhaps as a result of a different position of the lux box relative to the 10 element. These data underscore the diversity of quorum sensing signalling molecules, regulatory proteins and target genes in the Serratia genus and emphasize that negative control of AHLregulated phenotypes is perhaps a more general trait in the genus Serratia. Acknowledgments Rob Van Houdt worked on this project as research assistant from the Fund for Scientific Research-Flanders (F.W.O.Vlaanderen) and as postdoctoral fellow from the K.U. Leuven Research Fund. Abram Aertsen is a postdoctoral researcher from the F.W.O.-Vlaanderen, and Pieter Moons has a scholarship from the Institute for the Promotion of Innovation through Science and Technology in Flanders (IWT-Vlaanderen). References [1] J.B. Andersen, A. Heydorn, M. Hentzer, L. Eberl, O. Geisenberger, B.B. Christensen, S. Molin, M. Givskov, gfp-based N-acyl homoserine lactone sensor systems for detection of bacterial communication, Appl. Environ. Microbiol. 67 (2001) 575e585. [2] S. Atkinson, J.P. Throup, G.S. Stewart, P. Williams, A hierarchical quorum-sensing system in Yersinia pseudotuberculosis is involved in the regulation of motility and clumping, Mol. Microbiol. 33 (1999) 1267e1277. [3] A. Becker, M. Schmidt, W. Ja¨ger, A. Pu¨hler, New gentamicin-resistance and lacZ promoter-probe cassettes suitable for insertion mutagenesis and generation of transcriptional fusions, Gene 162 (1995) 37e39. [4] S.R. Chhabra, C. Harty, D.S.W. Hooi, M. Daykin, P. Williams, D.L. Pritchard, B.W. Bycroft, Synthetic analogues of bacterial quorum sensing molecules as immune modulators, J. Med. Chem. 46 (2003) 97e104. [5] A.B. Christensen, K. Riedel, L. Eberl, L.R. Flodgaard, S. Molin, L. Gram, M. Givskov, Quorum-sensing-directed protein expression in Serratia proteamaculans B5a, Microbiology 149 (2003) 471e483. [6] A. Eberhard, A.L. Burlingame, C. Eberhard, G.L. Kenyon, K.H. Nealson, J. Oppenheimer, Structural identification of autoinducer of Photobacterium fischeri luciferase, Biochemistry 20 (1981) 2444e2449.
157
[7] L. Eberl, M.K. Winson, C. Sternberg, G.S.A.B. Stewart, G. Christiansen, S.R. Chhabra, B. Bycroft, P. Williams, S. Molin, M. Givskov, Involvement of N-acyl-L-homoserine lactone autoinducers in controlling the multicellular behaviour of Serratia liquefaciens, Mol. Microbiol. 20 (1996) 127e136. [8] J. Engebrecht, M. Silverman, Identification of genes and gene products necessary for bacterial bioluminescence, Proc. Natl. Acad. Sci. USA 81 (1984) 4154e4158. [9] K.M. Gray, J.R. Garey, The evolution of bacterial LuxI and LuxR quorum sensing regulators, Microbiology 147 (2001) 2379e2387. [10] P.A. Grimont, F. Grimont, The genus Serratia, Annu. Rev. Microbiol. 32 (1978) 221e248. [11] J.F. Guasch, J. Enfedaque, S. Ferrer, D. Gargallo, M. Regue, Bacteriocin 28b, a chromosomally encoded bacteriocin produced by most Serratia marcescens biotypes, Res. Microbiol. 146 (1995) 477e483. [12] M.S. Guyer, R.R. Reed, J.A. Steitz, K.B. Low, Identification of a sex-factor-affinity site in E. coli as gd, Cold Spring Harb. Symp. Quant. Biol. 45 (1981) 135e140. [13] Y.T. Horng, S.C. Deng, M. Daykin, P.C. Soo, J.R. Wei, K.T. Luh, S.W. Ho, S. Swift, H.C. Lai, P. Williams, The LuxR family protein SpnR functions as a negative regulator of N-acylhomoserine lactonedependent quorum sensing in Serratia marcescens, Mol. Microbiol. 45 (2002) 1655e1671. [14] A. Jabrane, A. Sabri, P. Compere, P. Jacques, I. Vandenberghe, J. Van Beeumen, P. Thonart, Characterization of serracin P, a phage-tail-like bacteriocin, and its activity against Erwinia amylovora, the fire blight pathogen, Appl. Environ. Microbiol. 68 (2002) 5704e5710. [15] N. Kleckner, J. Bender, S. Gottesman, Use of transposons with emphasis on Tn10, Method. Enzymol 204 (1991) 139e180. [16] M. Labbate, S.Y. Queck, K.S. Koh, S.A. Rice, M. Givskov, S. Kjelleberg, Quorum sensing-controlled biofilm development in Serratia liquefaciens MG1, J. Bacteriol. 186 (2004) 692e698. [17] B.E. Laue, Y. Jiang, S.R. Chhabra, S. Jacob, G.S.A.B. Stewart, A. Hardman, J.A. Downie, F. O’Gara, P. Williams, The biocontrol strain Pseudomonas fluorescens F113 produces the Rhizobium small bacteriocin, N-(3-hydroxy-7-cis tetradecenoyl)-homoserine lactone via HdtS, a putative novel N-acylhomoserine lactone synthase, Microbiology 146 (2000) 2469e2480. [18] K.H. McClean, M.K. Winson, L. Fish, A. Taylor, S.R. Chhabra, M. Camara, M. Daykin, J.H. Lamb, S. Swift, B.W. Bycroft, G.S. Stewart, P. Williams, Quorum sensing and Chromobacterium violaceum: exploitation of violacein production and inhibition for the detection of N-acylhomoserine lactones, Microbiology 143 (1997) 3703e3711. [19] T.D. Minogue, M. Wehland-von Trebra, F. Bernhard, S.B. von Bodman, The autoregulatory role of EsaR, a quorum-sensing regulator in Pantoea stewartii ssp. stewartii: evidence for a repressor function, Mol. Microbiol. 44 (2002) 1625e1635. [20] N.C. Reading, V. Sperandio, Quorum sensing: the many languages of bacteria, FEMS Microbiol. Lett. 254 (2006) 1e11. [21] K. Riedel, T. Ohnesorg, K.A. Krogfelt, T.S. Hansen, K. Omori, M. Givskov, L. Eberl, N-acyl-L-homoserine lactone-mediated regulation of the lip secretion system in Serratia liquefaciens MG1, J. Bacteriol. 183 (2001) 1805e1809. [22] X. Rubires, F. Saigi, N. Pique, N. Climent, S. Merino, S. Alberti, J.M. Tomas, M. Regue, A gene (wbbL) from Serratia marcescens N28b (O4) complements the rfb-50 mutation of Escherichia coli K-12 derivatives, J. Bacteriol. 179 (1997) 7581e7586. [23] H. Slater, M. Crow, L. Everson, G.P.C. Salmond, Phosphate availability regulates biosynthesis of two antibiotics, prodigiosin and carbapenem, in Serratia via both quorum-sensing-dependent and eindependent pathways, Mol. Microbiol. 47 (2003) 303e320. [24] S. Swift, A.V. Karlyshev, L. Fish, E.L. Durant, M.K. Winson, S.R. Chhabra, P. Williams, S. Macintyre, G.S. Stewart, Quorum sensing in Aeromonas hydrophila and Aeromonas salmonicida: identification of the LuxRI homologs AhyRI and AsaRI and their cognate N-acylhomoserine lactone signal molecules, J. Bacteriol. 179 (1997) 5271e5281. [25] S. Swift, M.K. Winson, P.F. Chan, N.J. Bainton, M. Birdsall, P.J. Reeves, C.E.D. Rees, S.R. Chhabra, P.J. Hill, J.P. Throup, B.W. Bycroft,
158
[26]
[27]
[28]
[29]
[30]
[31]
R. Van Houdt et al. / Research in Microbiology 158 (2007) 150e158 G.P.C. Salmond, P. Williams, G.S.A.B. Stewart, A novel strategy for the isolation of luxI homologs: evidence for the widespread distribution of a LuxR:LuxI superfamily in enteric bacteria, Mol. Microbiol. 10 (1993) 511e520. N.R. Thomson, M.A. Crow, S.J. McGowan, A. Cox, G.P. Salmond, Biosynthesis of carbapenem antibiotic and prodigiosin pigment in Serratia is under quorum sensing control, Mol. Microbiol. 36 (2000) 539e556. R.H. Valdivia, S. Falkow, Bacterial genetics by flow cytometry: rapid isolation of Salmonella typhimurium acid-inducible promoters by differential fluorescence induction, Mol. Microbiol. 22 (1996) 367e378. R. Van Houdt, A. Aertsen, A. Jansen, A.L. Quintana, C.W. Michiels, Biofilm formation and cell-to-cell signalling in Gram-negative bacteria isolated from a food processing environment, J. Appl. Microbiol. 96 (2004) 177e184. R. Van Houdt, P. Moons, A. Jansen, C.W. Michiels, Genotypic and phenotypic characterization of a biofilm-forming Serratia plymuthica isolate from a raw vegetable processing line, FEMS Microbiol. Lett. 246 (2005) 265e272. R. Van Houdt, P. Moons, M. Hueso Buj, C.W. Michiels, N-acyl-L-homoserine lactone quorum sensing controls butanediol fermentation in Serratia plymuthica RVH1 and Serratia marcescens MG1, J. Bacteriol. 188 (2006) 4570e4572. S.B. von Bodman, D.R. Majerczak, D.L. Coplin, A negative regulator mediates quorum-sensing control of exopolysaccharide production in
[32]
[33]
[34]
[35]
[36]
Pantoea stewartii subsp. Stewartii, Proc. Natl. Acad. Sci. USA 95 (1998) 7687e7692. W.T. Watson, T.D. Minogue, D.L. Val, S.B. von Bodman, M.E. Churchill, Structural basis and specificity of acyl-homoserine lactone signal production in bacterial quorum sensing, Mol. Cell. 9 (2002) 685e694. N.A. Whitehead, A.M. Barnard, H. Slater, N.J. Simpson, G.P. Salmond, Quorum-sensing in Gram-negative bacteria, FEMS Microbiol. Rev. 25 (2001) 365e404. M.K. Winson, S. Swift, L. Fish, J.P. Throup, F. Jørgensen, S.R. Chhabra, B.W. Bycroft, P. Williams, G.S. Stewart, Construction and analysis of luxCDABE-based plasmid sensors for investigating N-acyl homoserine lactone-mediated quorum sensing, FEMS Microbiol. Lett. 163 (1998a) 185e192. M.K. Winson, S. Swift, P.J. Hill, C.M. Sims, G. Griesmayr, B.W. Bycroft, P. Williams, G.S. Stewart, Engineering the luxCDABE genes from Photorhabdus luminescens to provide a bioluminescent reporter for constitutive and promoter probe plasmids and mini-Tn5 constructs, FEMS Microbiol. Lett. 163 (1998b) 193e202. E.A. Yates, B. Philipp, C. Buckley, S. Atkinson, S.R. Chhabra, R.E. Sockett, M. Goldner, Y. Dessaux, M. Camara, H. Smith, P. Williams, N-acylhomoserine lactones undergo lactonolysis in a pH-, temperature-, and acyl chain length-dependent manner during growth of Yersinia pseudotuberculosis and Pseudomonas aeruginosa, Infect. Immun. 70 (2002) 5635e5646.
