SEARCH

SEARCH BY CITATION

Summary

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

The symbiosis island ICEMlSymR7A of Mesorhizobium loti R7A is an integrative and conjugative element (ICE) that carries genes required for a nitrogen-fixing symbiosis with Lotus species. ICEMlSymR7A encodes homologues (TraR, TraI1 and TraI2) of proteins that regulate plasmid transfer by quorum sensing in rhizobia and agrobacteria. Introduction of traR cloned on a plasmid induced excision of ICEMlSymR7A in all cells, a 1000-fold increase in the production of 3-oxo-C6-homoserine lactone (3-oxo-C6-HSL) and a 40-fold increase in conjugative transfer. These effects were dependent on traI1 but not traI2. Induction of expression from the traI1 and traI2 promoters required the presence of plasmid-borne traR and either traI1 or 100 pM 3-oxo-C6-HSL, suggesting that traR expression or TraR activity is repressed in wild-type cells by a mechanism that can be overcome by additional copies of traR. The traI2 gene formed an operon with hypothetical genes msi172 and msi171 that were essential for ICEMlSymR7A excision and transfer. Our data suggest that derepressed TraR in conjunction with TraI1-synthesized 3-oxo-C6-HSL regulates excision and transfer of ICEMlSymR7A through expression of msi172 and msi171. Homologues of msi172 and msi171 were present on putative ICEs in several α-proteobacteria, indicating a conserved role in ICE excision and transfer.


Introduction

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

Mobile genetic elements that mediate the horizontal transfer of DNA between bacteria represent a major driving force in prokaryotic evolution (Ochman et al., 2000; Frost et al., 2005). Available genome sequence data suggest that genomic islands represent a significant portion of horizontally acquired DNA (Dobrindt et al., 2004; Binnewies et al., 2006). For the majority of genomic islands little is known about their origin or mobilization and as such they are usually identified in silico by the presence of atypical DNA sequence composition, integration adjacent to tRNA genes and by their absence in related isolates (Mantri and Williams, 2004; Yoon et al., 2005; Guy, 2006; Ou et al., 2006). However, a recently defined subset of mobile genomic islands termed ‘integrative and conjugative elements’ or ICEs (Burrus et al., 2002; Burrus and Waldor, 2004) are able to excise from the chromosome, transfer to recipient cells via conjugation and (re)-integrate at a specific site in both recipient and donor. A better understanding of the selective forces and mechanisms that govern transfer of these ICEs will aid our understanding of prokaryotic evolution.

The symbiosis island ICEMlSymR7A of Mesorhizobium loti strain R7A was discovered through the ability of the strain to transfer genes required for a nitrogen-fixing symbiosis with Lotus corniculatus to non-symbiotic mesorhizobia (Sullivan et al., 1995). The 502 kb ICE is inserted downstream of a phe-tRNA gene in the M. loti chromosome and is flanked by a 17 bp direct repeat that represents the core sequence of its attachment site (Sullivan and Ronson, 1998). ICEMlSymR7A contains over 400 open reading frames including a gene at its left end intS that encodes a P4 phage-like integrase required for the island's integration and excision (Sullivan et al., 2002; Ramsay et al., 2006). Like other ICEs, ICEMlSymR7A excises from the genome to form a circular intermediate prior to conjugative transfer. Efficient excision and conjugative transfer require the ICEMlSymR7A-encoded recombination directionality factor RdfS as well as IntS (Ramsay et al., 2006). Genes encoding homologues of RdfS are present in a conserved gene cluster together with genes encoding homologues of TraF (the TrbC protease that is essential for mating pore formation) and the ICEMlSymR7A relaxase RlxS on several putative genomic islands in a range of α- and β-proteobacteria (Ramsay et al., 2006). Many of these islands also contain a trb gene cluster of the same organization as the trb cluster found on ICEMlSymR7A (Sullivan et al., 2002; Toussaint et al., 2003; Ramsay et al., 2006) that is required for conjugative transfer.

The proportion of cells in which ICEMlSymR7A is excised during culture in a laboratory medium varies depending on the growth phase of the culture, ranging from approximately 0.06% of cells in exponentially growing cultures to approximately 6% of cells in stationary phase (Ramsay et al., 2006). This observation suggests that excision may be regulated in response to growth phase or cell population density. Many Gram-negative bacteria encode N-acyl-homoserine lactone (AHL)-dependent quorum sensing (QS) systems that regulate gene expression in response to cell population density (Waters and Bassler, 2005; Williams et al., 2007). ICEMlSymR7A encodes a putative LuxR-family transcriptional regulator TraR and two putative AHL synthases TraI1 and TraI2 (Sullivan et al., 2002), which show homology to the TraR and TraI proteins of Agrobacterium tumefaciens and Rhizobium leguminosarum bv. viciae that regulate plasmid transfer via QS (Piper et al., 1993; Fuqua and Winans, 1994; Hwang et al., 1994; Danino et al., 2003; Sanchez-Contreras et al., 2007; White and Winans, 2007). The traR and traI2 genes are located upstream of and in the same orientation as two hypothetical open reading frames (ORFs) msi172 and msi171 while the traI1 gene is located 162 kb away in a single gene operon (Sullivan et al., 2002). Introduction of a plasmid containing cloned traR and traI2 genes into M. loti strain R7A resulted in excision of ICEMlSymR7A in 100% of cells and increased AHL production, strongly suggesting that excision of ICEMlSymR7A is regulated by QS (Ramsay et al., 2006).

In this study we further investigated the roles of the ICEMlSymR7A-encoded traR, traI1 and traI2 genes in the regulation of excision. We show that introduction of traR alone on a plasmid resulted in 100% excision and a 1000-fold increase in the production of 3-oxo-C6-homoserine lactone (3-oxo-C6-HSL). Both phenotypes were dependent on the presence of a functional copy of traI1. We also show that when active, TraR in conjunction with AHLs synthesized by TraI1 induces ICEMlSymR7A excision and transfer through the expression of two conserved hypothetical genes msi172 and msi171, which are encoded on a TraR-regulated polycistronic mRNA initiating upstream of traI2.

Results

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

Excision of ICEMlSymR7A and AHL production are induced by plasmid-borne traR

We previously developed a quantitative PCR (QPCR) assay that determines the proportion of cells containing ICEMlSymR7A in excised form (Ramsay et al., 2006). The assay measures the proportion of attP ICE attachment sites and attB chromosomal attachment sites that are formed upon ICEMlSymR7A excision, relative to the chromosomal gene melR in a cell population. Use of this assay showed that plasmid pJR206 carrying the ICEMlSymR7AtraR and traI2 genes induces excision of ICEMlSymR7A in 100% of M. loti strain R7A cells; in the absence of pJR206, ICEMlSymR7A is excised in at most 6% of cells. In addition, strain R7A(pJR206) causes strong induction of violacein production in the AHL indicator strain Chromobacterium violaceum CV026 whereas strain R7A shows weak or no induction of CV026 (Ramsay et al., 2006).

To determine whether the effects of pJR206 on excision and AHL production required traI2, a derivative pJRtraR containing only traR with its native promoter was constructed. Strain R7A(pJRtraR) induced a strong response in CV026 (Fig. 1A) identical to that previously seen with R7A(pJR206), and QPCR assays of DNA extracted from exponential and stationary-phase cultures showed that the ICEMlSymR7A was excised in 100% of cells (Fig. 2A). In contrast, introduction of pJRtraR or pJR206 into strain R7ANS, a derivative of R7A cured of ICEMlSymR7A, did not enable the strain to induce violacein production in CV026. These results indicate that the introduction of a plasmid-borne copy of traR is sufficient to induce AHL production and excision of ICEMlSymR7A in strain R7A and that AHL production requires additional ICEMlSymR7A genes.

image

Figure 1. Induction of violacein production in C. violaceum CV026 by various M. loti strains. Stationary phase (64 h) M. loti cultures were spotted on a TY plate (indicated by a white circle) and incubated for 24 h, after which a CV026 overlay was applied and the plate incubated overnight. A dark halo indicates where M. loti cultures have induced strain CV026 to produce violacein through production of AHLs. A. Strains R7A, R7AΔtraI1 and R7AΔtraR with or without pJRtraR. Strain R7AΔtraI1 complemented with pJRtraI1 is also shown. B. Strains R7A, R7AΔmsi171 or R7AΔmsi172 with or without pJRtraR.

Download figure to PowerPoint

image

Figure 2. Excision of ICEMlSymR7A in R7A and tra gene mutants. QPCR of DNA templates attP and attB was used as a measure of ICEMlSymR7A excision relative to the chromosome population detected by QPCR of melR. Percentage attP and attB at 24 h (exponential growth) are shown by black and white bars respectively, while percentage attP and attB at 64 h (stationary phase) are shown by black bars/white hatching and white bars/black hatching respectively. Cultures of strain R7A(pJRtraI1E) were induced with 0.1 mM IPTG at the time of inoculation. Data presented are the average of at least two independent experiments (error bars represent ± 1 standard deviation). Individual QPCR measurements are the mean of triplicate reactions for each amplicon. Asterisks indicate samples with significant deviation from the wild-type data (*P < 0.05, **P < 0.01).

Download figure to PowerPoint

traI1 but not traI2 is required for induction of violacein production in CV026

To investigate if either of the putative AHL synthases TraI1 or TraI2 were required for AHL synthesis, mutant strains with markerless in-frame deletions in traI1 (strain R7AΔtraI1) and traI2 (strains R7AΔtraI2* and R7AΔtraI2) were constructed. Strain R7AΔtraI2* contained a deletion beginning 22 bp upstream of the traI2 start codon, due to the inadvertent use of an incorrect translational start codon during design of the deletion, whereas the deletion in strain R7AΔtraI2 began at the 41st codon of the gene. The deletion end-point in both mutants was 36 bp before the stop codon of traI2. The ability of the mutants, in the presence or absence of pJRtraR, to induce violacein production in CV026 was then assayed. Surprisingly, strain R7AΔtraI2* showed weak induction whereas strains R7AΔtraI1 and R7AΔtraI2, like R7A, showed no induction (Fig. 1A and data not shown). Strains R7AΔtraI2*(pJRtraR) and R7AΔtraI2(pJRtraR) showed a similar level of induction to that of R7A(pJRtraR) (data not shown), whereas R7AΔtraI1(pJRtraR) did not induce detectable violacein production (Fig. 1A).

To further investigate the roles of traI1 and traI2 in AHL synthesis, each gene was cloned downstream of the IPTG-inducible tac promoter in an expression vector pTH1227 and the resulting plasmids pJRtraI1E and pJRtraI2E were introduced into Escherichia coli strain DH5α and M. loti strains R7A and R7ANS. Both M. loti derivatives and DH5α containing pJRtraI1E strongly induced violacein production in CV026 when cultured in the presence of IPTG. However, strains containing pJRtraI2E showed no violacein induction.

AHLs produced by strain R7A and its derivatives

Reversed-phase thin-layer chromatography of dichloromethane extracts from M. loti supernatants revealed that strain R7A(pJRtraR) produced at least four AHLs capable of inducing pigment production in CV026, while no AHLs were detected in extracts from R7A (data not shown). To identify and quantify the AHL species present in culture supernatants of strain R7A and its derivatives, extracts were analysed using liquid chromatography coupled to hybrid quadrupole-linear ion trap mass spectrometry (LC-MS/MS). The extracts were screened for molecules with 4, 6, 8, 10, 12 or 14 carbons with or without 3-oxo- or 3-hydroxy- substitutions (Table 1). Strain R7A produced low levels of C4-HSL, 3-oxo-C6-HSL and 3-oxo-C12-HSL. Strain R7A(pJRtraR) extracts contained over 1000-fold more 3-oxo-C6-HSL than R7A and substantial amounts of several other C6 and C8 AHLs. The amount of 3-oxo-C12-HSL in R7A(pJRtraR) was similar to that in R7A. Strains R7AΔtraI1, R7AΔtraR and R7ANS produced reduced levels of 3-oxo-C6-HSL compared with wild type and wild-type levels of 3-oxo-C12-HSL. The AHL profile of R7AΔtraI2*(pJRtraR) was similar to that of R7A(pJRtraR), except that the amounts of most AHL species were three- to eightfold increased and 3-oxo-C12-HSL was not detected. Strain R7AΔtraI2* showed an increased amount of 3-oxo-C6-HSL compared with R7A and low amounts of several AHLs that were likely present (given their presence in R7A/pJRtraR) but below the threshold of detection in R7A (Table 1). As 3-oxo-C12-HSL was near the detection limit of the assay in all strains where it was observed including R7AΔtraI2*, it is likely that it was present but below the detection limit in R7AΔtraI2*(pJRtraR).

Table 1.  Abundance of AHLs in M. loti R7A and derivative strains.
Source of extractPicomoles of AHL per sample ± standard deviationa
Acyl-chain length and 3C substituent
4-H4-O4-OH6-H6-O6-OH8-H8-O8-OH10-O10-OH12-O
  • a.

    The data presented are the average and standard deviation of results from two independent experiments. For each experiment, sample data were calculated by averaging LC-MS/MS data from three extractions, each from 8 ml of culture, and normalizing to individual reference standards after taking into account extraction efficiencies.

R7A5.9 ± 8.419 ± 5.822 ± 2.3
R7A(pJRtraR)91 ± 484.9 × 102  ± 2.5 × 1024.1 × 103  ± 1.3 × 1032.5 × 104  ± 7.4 × 1034.6 × 102  ± 1.9 × 1023.1 × 103  ± 4.0 × 1023.0 × 103  ± 2.4 × 10354 ± 6.4 × 10−11.3 × 102  ± 1.8 × 1026.3 ± 9.011 ± 9.7
R7AΔtraR5.1 ± 7.330 ± 10
R7AΔtraI13.1 ± 4.421 ± 2.7
R7AΔtraI1(pJRtraR)45 ± 2315 ± 15
R7AΔtraI2*7.3 ± 1.01.9 × 102  ± 2.6 × 1022.5 × 102  ± 3.3 × 1027.5 ± 1.1 × 10221 ± 3025 ± 1.4
R7AΔtraI2*(pJRtraR)1.8 × 102  ± 273.7 × 103  ± 1.3 × 1031.2 × 102  ± 126.1 × 103  ± 1.1 × 1034.8 × 104  ± 3.0 × 1033.8 × 103  ± 2.4 × 1037.2 × 103  ± 2.0 × 1031.0 × 104  ± 2.6 × 1034.3 × 102  ± 1.0 × 1022.8 × 102  ± 3.3 × 10224 ± 23
R7ANS2.3 ± 3.22.4 ± 3.413 ± 18

To test if the increased AHL production in R7AΔtraI2* required a functional traI1, a double mutant R7AΔtraI2traI1 was constructed and pJRtraR or pJR206 (traR traI2) introduced into it. Strains R7AΔtraI2traI1(pJRtraR) and R7AΔtraI2traI1 (pJR206) did not induce violacein production in CV026.

Taken together, these results suggest that TraI1 was responsible for the substantial increase in synthesis of 3-oxo-C6-HSL and concomitant increases in several other C6 and C8 AHLs observed in R7A(pJRtraR) and R7AΔtraI2*(pJRtraR), while additional AHL synthase(s) encoded outside ICEMlSymR7A may be responsible for the production of very low amounts of 3-oxo-C6-HSL and 3-oxo-C12-HSL (Table 1).

QS mutants apart from R7AΔtraI2* show similar excision levels to wild type

It seemed likely that traR and traI1 were involved in both AHL production and regulation of ICEMlSymR7A excision, given that pJRtraR induced excision in 100% of R7A cells and traI1 was required for the pJRtraR-dependent induction of AHL production. However, excision of ICEMlSymR7A was only slightly reduced in strains R7AΔtraR and R7AΔtraI1 and, surprisingly, an increase in excision was still observed in stationary phase cultures (Fig. 2A). Like R7A(pJRtraR), strain R7AΔtraR(pJRtraR) induced higher levels of violacein production in CV026 (Fig. 1A) and excision of ICEMlSymR7A (Fig. 2A) than R7A. Consistent with its inability to produce AHLs above R7A levels, strain R7AΔtraI1(pJRtraR) showed similar excision levels to R7A (Fig. 2A). To complement the ΔtraI1 mutation, plasmid pJRtraI1 that contains traI1 and its upstream sequence was introduced into R7AΔtraI1. Strain R7AΔtraI1(pJRtraI1) showed weak or no induction of violacein production in CV026 (Fig. 1A) and moderately increased excision relative to that observed in R7A (Fig. 2A), indicating that pJRtraI1 complemented and likely overcompensated for the effect of the traI1 mutation on excision. The above results indicate that, while the traR gene when present on pJRtraR induced ICEMlSymR7A excision in 100% of cells in a traI1-dependent manner, other factors contributed to the dependence of excision on growth phase in the wild-type R7A strain.

Next we investigated excision of ICEMlSymR7A in the two traI2 mutants (Fig. 2B). R7AΔtraI2 showed a similar phenotype to R7A, but R7AΔtraI2* showed markedly reduced levels of excision, a phenotype that was not corrected by addition of a cloned copy of traI2 in pJRtraI2 (Fig. 2B). Derivatives of R7AΔtraI2* carrying pJRtraR or pJR206 (traRtraI2) also had excision levels (Fig. 2B) that were markedly reduced in comparison with those observed in R7A carrying the same plasmids (Fig. 2A and Ramsay et al., 2006).

TraR activates transcription of traI1 and traI2

To determine whether expression of traI1 and traI2 was affected by pJRtraR, traI1 or traI2 mutant strains were constructed by insertion of the suicide vector pFUS2. The mutants contained a transcriptional fusion between the 5′ end of the respective traI gene, which was inactivated by the insertion, and lacZ. The plasmid pJRtraR or its parent vector pFAJ1700 were then introduced into each mutant strain. The R7AtraI2::lacZ mutant containing pJRtraR induced violacein production in CV026 but the R7AtraI1::lacZ strain containing pJRtraR did not, consistent with the phenotypes of the previously constructed in-frame deletion mutants. The traI2::lacZ fusion was not active (< 10 Miller units) in the absence of pJRtraR, even when 50 nM 3-oxo-C6-HSL was added. However, traI2::lacZ was strongly expressed (437 ± 22 Miller units) when pJRtraR was present. As expected given that R7AΔtraI2(pJRtraR) makes large quantities of AHLs, the addition of 50 nM 3-oxo-C6-HSL had no effect on expression (412 ± 20 Miller units). The traI1::lacZ fusion was not expressed in the absence of 3-oxo-C6-HSL even in the presence of pJRtraR. The addition of 3-oxo-C6-HSL at a concentration of 100 pM induced strong expression of the traI1::lacZ fusion only in the presence of pJRtraR (Fig. 3). Several of the other AHLs (C6-HSL, 3-hydroxy-C6-HSL, 3-hydroxy-C8-HSL and 3-oxo-C8-HSL) produced by R7A(pJRtraR) were also assayed for their ability to induce β-galactosidase activity in strain R7AtraI1::lacZ(pJRtraR) when added at 20 nM. All induced the fusion to a similar level as 3-oxo-C6-HSL, indicating that TraR is able to recognize several AHLs and may not have a single cognate AHL. The results also showed that transcription of both traI1 and traI2 was activated by pJRtraR and that, for the traI1 promoter at least, this was dependent on the presence of AHL species produced by TraI1. However, in the wild-type strain the genes were poorly expressed even in the presence of high levels of exogenous AHL.

image

Figure 3. Transcriptional response of the traI1 promoter to 3-oxo-C6-HSL in the presence or absence of plasmid-borne traR. Filled circles, R7AtraI1::lacZ(pJRtraR); filled triangles, R7AtraI1::lacZ(pFAJ1700). Cultures were grown in TY medium in the presence of the indicated concentration of 3-oxo-C6-HSL for 48 h and washed cells then assayed for β-galactosidase activity, expressed as Miller units. The mean values of two to four replicates together with the standard errors are shown.

Download figure to PowerPoint

AHL-dependent activation of excision by TraR is subject to an additional level of regulation in strain R7A

To test whether increased traI1 expression induced excision of ICEMlSymR7A in the absence of pJRtraR, the effect of the traI1 expression plasmid pJRtraI1E was tested. Strain R7A (pJRtraI1E) induced a similar violacein production response in CV026 as R7A(pJRtraR) (data not shown), indicating increased production of AHLs as expected. However, excision only increased to approximately 1% in exponential phase and to less than 50% in stationary phase, compared with the 100% excision induced by pJRtraR in both growth phases (Fig. 2A). The analogous plasmid containing traI2, pJRtraI2E, had no effect on excision (data not shown). These results suggest that an additional barrier must be overcome before AHL-dependent activation of excision by TraR occurs. As pJRtraR induced 100% excision of ICEMlSymR7A, the presence of extra copies of traR and its upstream sequence must overcome this barrier.

Mapping of the traI2 promoter

Directly downstream of traI2 are two ORFs msi172 and msi171 in the same orientation that both encode conserved hypothetical proteins (Fig. 4B). As there are few nucleotides between each of the genes, it seemed likely that traI2-msi172-msi171 formed an operon. To test this, RNA was extracted from R7AΔmsi170[a strain with an identical phenotype to R7A(pJRtraR) with respect to excision and AHL production; our unpublished data] and cDNA synthesized using a primer (msi171SP1, Table S1) specific for the 3′ end of msi171. PCR was carried out on the resulting cDNA using primers specific for the 5′ end of traI2 and a nested primer within the 3′ end of msi171 (Table S1, I2IPTGF and msi171SP2 respectively). Sequence analysis of the resulting PCR product confirmed that it contained all three ORFs and hence that the three genes formed an operon.

image

Figure 4. Genetic organization and transcription of traI2, msi172 and msi171. A. A manual alignment of the DNA sequence upstream of the traI1 and traI2 coding sequences (start codon shown in bold italics). An inverted repeat resembling the tra-box binding motif is indicated by inward facing arrows and nucleotides matching the Ti plasmid tra-box are shown in bold (note that the match to the Ti plasmid consensus is offset by one nucleotide). Also indicated are the transcriptional start site (+1) and the start point of the deletion in strain R7AΔtraI2*. B. Genetic organization of the traR-msi171 gene cluster.

Download figure to PowerPoint

An alignment of the DNA sequences upstream of traI1 and traI2 revealed the presence of an imperfect inverted repeat upstream of each gene, which partially aligned with the Ti plasmid tra-box half-site ‘ATGTGCAGA’ (Fuqua and Winans, 1996; Pappas and Winans, 2003), offset by one nucleotide (Fig. 4A). No other potential tra-boxes were identified on ICEMlSymR7A. The tra-box elements that bind TraR on the Ti plasmid are centred approximately 63 or 43 bp upstream of the transcriptional start site. To map the transcriptional start site of the traI2-msi172-msi171 operon, the cDNA from the previous experiment (extended from primer msi171SP1) was used in a 5′RACE experiment, using a primer complementary to traI2. Sequencing of the PCR product revealed that the transcriptional start site was located in the intergenic region between traR and traI2, 44 bp downstream of the centre of tra-box2 (Fig. 4A), consistent with this DNA motif being a tra-box-like element. These and the data described above strongly suggest that TraR complexed with a C6 or C8 AHL species activates transcription of the traI2-msi172-msi171 operon by binding to tra-box2.

TraR-activated excision requires msi171 and msi172

The above results suggested that expression of the traI2-msi172-msi171 operon was regulated by TraR but did not require TraI2. Furthermore, the traI2 transcriptional start site was mapped to 1 bp upstream of the start of the deletion in strain R7AΔtraI2* (Fig. 4A), suggesting that the deletion may affect expression of the operon. These observations, together with the lack of complementation by traR and/or traI2 of the excision phenotypes in R7AΔtraI2* and the wild-type phenotypes of R7AΔtraI2, suggested that the R7AΔtraI2* phenotypes may be due to the loss or reduction of expression of the msi171 or msi172 genes rather than deletion of the traI2 gene. To test this idea, a cosmid pUT11G that contains the entire traR-msi171 region plus flanking DNA was introduced into R7A and R7AΔtraI2*. Both strains R7A and R7AΔtraI2* containing pUT11G showed close to 100% ICEMlSymR7A excision, indicating that the cosmid restored excision in R7AΔtraI2* (Fig. 5). The high levels of excision presumably reflect the presence of traR on pUT11G.

image

Figure 5. Excision of ICEMlSymR7A in R7AΔmsi171 and strains containing pUT11G. QPCR of DNA templates attP and attB was used as a measure of ICEMlSymR7A excision relative to the chromosome population detected by QPCR of melR. Percentage attP and attB at 24 h (exponential growth) are shown by black and white bars respectively, while percentage attP and attB at 64 h (stationary phase) are shown by black bars/white hatching and white bars/black hatching respectively. Data presented are the average of at least two independent experiments (error bars represent ± 1 standard deviation). Individual QPCR measurements are the mean of triplicate reactions for each amplicon. Asterisks indicate samples with significant deviation from the wild-type data (*P < 0.05, **P < 0.01). Amplicons from R7AΔmsi171 and R7AΔmsi171(pJRtraR) samples were only detected once or not at all.

Download figure to PowerPoint

In-frame deletion mutants of msi171 and msi172 were therefore constructed and analysed for excision of ICEMlSymR7A. The attP/attB amplicons were only detected sporadically (once in six independent experiments) at less than 0.02% in DNA extracted from either R7AΔmsi172 or R7AΔmsi171 in exponential or stationary-phase cultures (Fig. 5 and data not shown), demonstrating that deletion of either gene nearly abolished excision. Next, pJRtraR was introduced into these mutants and the resulting strains analysed by QPCR and CV026 bioassay. Both mutants containing pJRtraR induced violacein production in CV026 similarly to R7A(pJRtraR) (Fig. 1B), but ICEMlSymR7A excision products were again only present sporadically (once from three experiments) below 0.06% (Fig. 5 and data not shown).

Cosmid pUT11G was then introduced into strains R7AΔmsi172 and R7AΔmsi171. Strains R7AΔmsi172(pUT11G) (data not shown) and R7AΔmsi171(pUT11G) showed close to 100% excision of ICEMlSymR7A as found for R7A(pUT11G) (Fig. 5), indicating that the cosmid complemented the defect in excision. Plasmids pNJmsi172E and pNJmsi171E containing msi172 and msi171 downstream of a constitutive nptII promoter were then introduced into both mutants. QPCR analysis of these strains revealed that pNJmsi171E complemented the mutation in strain R7AΔmsi171 (Fig. 5); however, pNJmsi172E did not complement the msi172 mutation (data not shown), suggesting that this mutation may have a polar effect on msi171 or that Msi171 and Msi172 were produced in a non-functional stoichiometry in the strain. A plasmid expressing both msi172 and msi171 from the nptII promoter was also constructed, but this plasmid could not be introduced into R7A or the mutants using either electroporation or biparental mating. In summary, these data suggested that mutation of msi171 and likely also msi172 prevents excision irrespective of AHL production, indicating that TraR controls excision through activation of msi172-msi171 expression.

Effects of pJRtraR and various mutations on conjugative transfer of ICEMlSymR7A

The abilities of R7A and the various mutant strains, with or without pJRtraR, to transfer ICEMlSymR7A to the non-symbiotic M. loti strain N18 were determined (Table 2). Strain R7A transferred ICEMlSymR7A at a frequency per donor of about 4 × 10−5, while transfer from R7A(pJRtraR) occurred at an approximately 40-fold higher frequency. Transfer of ICEMlSymR7A from R7AΔtraR to strain N18 was below the detection limit of 1 × 10−8 transconjugants per donor, indicating that traR was required for conjugative transfer despite the mutant showing some excision. Transfer from R7AΔtraI1 was detected at a low frequency that was not increased by the presence of pJRtraR. Strains R7AΔtraI2 and R7AΔtraI2(pJRtraR) transferred ICEMlSymR7A at similar levels to R7A and R7A(pJRtraR), respectively, whereas transfer from R7AΔtraI2* was not detected irrespective of the presence of pJRtraR. Neither R7AΔmsi171 nor R7AΔmsi172 transferred ICEMlSymR7A in the absence of pJRtraR while a very low level of transfer was detected from R7AΔmsi172 but not R7AΔmsi171 in the presence of pJRtraR. With the exception of the result with R7AΔtraR, these results are consistent with those obtained using the excision assay, confirming that conjugative transfer of ICEMlSymR7A is inherently linked with excision.

Table 2.  Transfer frequency of ICEMlSymR7A from R7A and QS mutant strains with or without pJRtraR.
Donor strainRecipient strainTransfer frequencya
  • a.

    Expressed as the average number of transconjugants per donor from two experiments.

R7AN18 (pFAJ1700)4.45 × 10−5
R7AΔmsi171N18 (pFAJ1700)< 1 × 10−8
R7AΔmsi172N18 (pFAJ1700)< 1 × 10−8
R7AΔtraI1N18 (pFAJ1700)1.63 × 10−6
R7AΔtraI2N18 (pFAJ1700)1.06 × 10−5
R7AΔtraI2*N18 (pFAJ1700)< 1 × 10−8
R7AΔtraRN18 (pFAJ1700)< 1 × 10−8
R7AN18 (pPROBE-KT)3.14 × 10−5
R7A(pJRtraR)N18 (pPROBE-KT)1.12 × 10−3
R7AΔmsi171(pJRtraR)N18 (pPROBE-KT)< 1 × 10−8
R7AΔmsi172(pJRtraR)N18 (pPROBE-KT)5.05 × 10−7
R7AΔtraI1(pJRtraR)N18 (pPROBE-KT)1.04 × 10−7
R7AΔtraI2(pJRtraR)N18 (pPROBE-KT)6.96 × 10−3
R7AΔtraI2*(pJRtraR)N18 (pPROBE-KT)< 1 × 10−8
R7AΔtraR(pJRtraR)N18 (pPROBE-KT)9.49 × 10−4

Msi172 and Msi171 homologues are encoded on a large family of putative ICEs

The msi172 and msi171 genes encode 82- and 179-amino-acid proteins respectively, and blastp analyses indicated that they showed similarity to proteins of unknown function, with Msi171 belonging to COG5419. Previously, it was shown that several genes on ICEMlSymR7A encode proteins with similarity to those conserved on the genomic island Tn4371 and related elements (Toussaint et al., 2003; Ramsay et al., 2006). A search for proteins resembling Msi172 and Msi171 on Tn4371 identified a single protein RO0034, which resembled a fused product of Msi172 and Msi171 in that order. The RO0034 protein is conserved on a number of elements closely related to Tn4371 (Toussaint et al., 2003). Further blastp analyses of Msi172 and Msi171 revealed the presence of homologues of Msi172 and/or Msi171 encoded as single or separate ORFs in a large number other proteobacterial species, including most of the species identified by Ramsay et al. (2006) as carrying homologues of RdfS. Interestingly, TraR/TraI homologues were only found upstream of msi172/171 in M. loti MAFF303099 (ICEMlSymMAFF; NC_002678) and on a putative ICE in Sphingomonas sp. SKA56 (NZ_AAQG01000007.1).

Discussion

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

In this study, we showed that excision of ICEMlSymR7A in M. loti was induced in 100% of cells upon introduction of pJRtraR, a low-copy-number plasmid that carries a cloned copy of the traR gene including its native promoter region. The plasmid also caused a 1000-fold increase in the production of 3-oxo-C6-HSL and concomitant increases in several other C6 and C8 AHLs, and a 40-fold increase in conjugation frequency. The increases in excision, AHL production and conjugation were dependent on the ICEMlSymR7A-encoded AHL synthase gene traI1. Excision and conjugal transfer of ICEMlSymR7A required the conserved hypothetical genes msi172 and msi171 that are part of a traI2-msi172-msi171 operon that is transcribed from a TraR-regulated promoter, thus placing excision and conjugation under N-acyl homoserine lactone-mediated regulation. Homologues of msi172 and msi171 are present on a large family of putative genomic islands but in most cases are not associated with traR or traI homologues.

Expression of traI1 in E. coli and M. loti R7ANS conferred the ability to induce violacein production in strain CV026 on these strains, indicating that traI1 encodes a functional AHL synthase that is consistent with its requirement for both pJRtraR-mediated phenotypes. In contrast, our experiments did not reveal a role for the second putative AHL synthase-encoding gene traI2. Strain R7AΔtraI2 had the same phenotype as strain R7A with respect to excision and AHL production. It also responded in the same manner as R7A to the presence of plasmid-borne traR. Finally constitutive expression of traI2 in either E. coli or M. loti R7A or R7ANS did not lead to CV026-detectable AHL production. Similar results with respect to production of 3-oxo-C6-HSL production by TraI1 and no detectable AHL production by TraI2 both in M. loti and when overexpressed in E. coli were recently found for orthologous proteins in M. loti strain NZP2213 (Yang et al., 2009). Nevertheless, the predicted traI1 and traI2 gene products share 65% identity and are more similar to each other than to any homologues from other organisms. Hence, it seems possible that traI1 and traI2 arose from a gene duplication event or independent acquisition from a common source and that traI2 subsequently became a pseudogene. However, CDD analysis (Marchler-Bauer et al., 2007) indicates that TraI2 contains all 16 residues that are invariant in the 13 members of pfam00765 (autoinducer synthetases). Hence, it remains possible that traI2 plays an as yet undetected role in AHL production and the regulation of excision. Recent studies have revealed that the variety of acyl-HSL signals synthesized by LuxI homologues is greater than previously appreciated (Krick et al., 2007; Schaefer et al., 2008). It is therefore possible that TraI2 produces molecules that were not detected by the assays used in this study.

The two traI-containing operons on ICEMlSymR7A were both positively regulated by TraR as shown by reporter gene assays, and both contained an inverted repeat in their promoter regions that resembled the consensus of tra-boxes found in promoter regions of QS-regulated operons on the A. tumefaciens Ti plasmid and pRL1JI of R. leguminosarum that are binding sites for TraR (Fuqua and Winans, 1996; Pappas and Winans, 2003; McAnulla et al., 2007). The sequence upstream of traI2 (tra-box2) was centred 45 bp upstream of the determined transcriptional start site of the traI2-msi172-msi171 operon, an almost identical positioning to the Ti plasmid and pRL1JI tra-boxes. It is likely that the transcriptional start site of traI1 is in an analogous position relative to tra-box1. As no other tra-box matches were found on ICEMlSymR7A, TraR may directly mediate activation of only traI1 (autoinduction) and the traI2 operon. The increased production of AHLs in the ΔtraI2* mutant (in the presence and absence of pJRtraR) compared with analogous R7A strains was dependent on traI1 and may therefore reflect increased traI1 transcription. This may reflect a greater availability of TraR in the ΔtraI2* mutant cells due to it not being sequestered for traI2 operon transcription, implying that the amount of TraR in the cell is limiting for transcription.

Our previous finding that excision of ICEMlSymR7A is increased in stationary-phase broth cultures led us to propose a model in which excision is upregulated in response to increased population density through QS (Ramsay et al., 2006). However, our current results show that deletion of either traR or traI1 only slightly reduced the proportion of cells containing excised ICEMlSymR7A and that excision still appeared to be upregulated in stationary phase cultures. Despite this, strain R7AΔtraR showed extremely reduced levels of conjugative transfer compared with R7A. Furthermore, the traI2-msi172-msi171 and traI1 operons were expressed at most at a very low level (< 8 Miller units) in the absence of pJRtraR, despite msi172 and msi171 being required for excision. These results suggest that ICEMlSymR7A excision may be under dual regulation with a mechanism other than QS being at least partly responsible for the growth phase response and the excision seen in cultures of R7A. The abolition of ICEMlSymR7A transfer in the ΔtraR mutant suggests that the level of msi171/172 expression in these cells is insufficient to induce all operons required for transfer, although it is sufficient to allow some excision. The low basal level of transfer observed with R7A under the conditions used suggests that TraR may be activated in only a small percentage of cells, the activation possibly being an effect of stochastic processes.

A striking result was that expression from the traI1 promoter was fully induced in a traI1 mutant background by supplying cells with exogenous 3-oxo-C6-HSL but only when the strain carried a plasmid with a cloned copy of traR. In the presence of pJRtraR, close to full induction was obtained with as little as 100 pM 3-oxo-C6-HSL, which equates to about one molecule of AHL per cell (Su et al., 2008), indicating maximal signal sensitivity and hence that very few molecules of TraR-3-oxo-C6-HSL are required for full activation. In contrast, in the absence of pJRtraR only weak induction of traI1 expression, to a level about 5% of that in the presence of pJRtraR, was observed in the presence of 1 nM 3-oxo-C6-HSL and no further induction was observed even with 1 μM 3-oxo-C6-HSL. These results strongly suggest that in the absence of additional copies of traR, the amount of TraR in the cell is very low and limits the transcriptional response irrespective of the quantity of AHL signal present. Consistent with this, only a moderate increase in excision of ICEMlSymR7A was observed in strain R7A(pJRtraI1E) that carries traI1 fused to an IPTG-inducible promoter. This strain exhibited an elevated level of AHL production, similar to or greater than that produced by R7A(pJRtraR) that maintains ICEMlSymR7A in excised form in 100% of cells. The effects of pJRtraR must be manifested by increased TraR production either at the level of traR transcription or TraR translation or function. It is unlikely that the cloned traR gene in pJRtraR is expressed from a vector promoter, resulting in deregulated expression, as the pFAJ1700 vector multiple cloning site is protected against read-through transcriptional activity of vector sequences by the trpA transcriptional terminator (Dombrecht et al., 2001), and similar results were obtained with the complementing cosmid pUT11G. Taken together, the above results indicate that there are additional as yet unidentified factors regulating excision, both with respect to the response of TraR to AHLs and with respect to growth phase.

Whether the traR gene on ICEMlSymR7A is regulated at the transcriptional level is currently unknown. The expression of traR on plasmid pRL1JI of R. leguminosarum requires activation by a second LuxR-family protein BisR, which is activated by 3-OH-7-cis-C14:1-HSL (Danino et al., 2003), whereas the expression of traR on the Agrobacterium Ti plasmids requires activation or derepression by a transcriptional regulator in response to opines produced by crown gall tumours (Fuqua and Winans, 1994; Piper et al., 1999; White and Winans, 2007). Several LuxR and LuxI homologues are present in the sequenced M. loti strain MAFF303099 (Kaneko et al., 2000), which is closely related to R7A. It is possible that homologues of these proteins in strain R7A interact with the ICEMlSymR7A QS system. Indeed, the LC-MS/MS data showed that M. loti strain R7A produces very low amounts of 3-oxo-C12-HSL and Yang et al. (2009) have shown that this production is due to a conserved AHL synthase encoded on the M. loti chromosome.

Both Ti plasmids and pRL1JI also encode the QS antiactivator TraM (Fuqua et al., 1995; Hwang et al., 1995; Danino et al., 2003), which directly inhibits TraR through protein–protein interactions (Chen et al., 2007; Qin et al., 2007). In A. tumefaciens, the low level of TraR present due to basal traR expression in the absence of opines is prevented from prematurely activating Ti plasmid transfer by TraM (Khan et al., 2008; Su et al., 2008). In the absence of TraM, Ti plasmid conjugative transfer is maximally induced by 100 pM acyl-HSL signal, the theoretical maximum sensitivity given that each TraR dimer requires two signal molecules, and hence transfer is not responsive to population cell density (Su et al., 2008). ICEMlSymR7A does not contain a homologue of TraM nor is a homologue present in the MAFF303099 genome. However, it is possible that another protein may be an antiactivator of QS and that the extra copies of traR on pJRtraR result in sufficient TraR protein to overcome this inactivation. As noted above, only very few molecules of activated TraR and 3-oxo-C6-HSL are required for the transcriptional response and so whether a particular cell is able to activate ICEMlSymR7A excision and transfer may depend on the relative levels of TraR and an antiactivator in that cell. Further investigations into transfer events at the single-cell level are required to fully understand both the transfer mechanisms and evolutionary strategy of ICEMlSymR7A and related mobile elements.

Our data showed that msi172 and msi171 are required for excision of ICEMlSymR7A. The translated products of these genes have no matches to other proteins in the databases apart from hypothetical proteins. In each case, the Msi172/171 homologues are encoded near genes encoding homologues of the ICEMlSymR7A recombination directionality factor RdfS. We have previously shown that the presence of RdfS defines a large family of genomic islands that are likely to have a conserved transfer mechanism (Ramsay et al., 2006). In some genomic islands such as Tn4371 (Toussaint et al., 2003), the Msi172/171 homologue is present as a single protein, indicating that Msi172 and Msi171 are likely to function cooperatively on a common target or targets. This is also consistent with the phenotype of the mutated strains. They are unlikely to be part of the nucleoprotein complex required for excision as only the integrase IntS is required along with RdfS for efficient excision of a mini-ICE in the non-symbiotic strain R7ANS (Ramsay et al., 2006). Neither protein has a known DNA-binding domain and hence it seems likely that Msi172 and Msi171 exert a regulatory effect through interaction with other proteins conserved between the ICEMlSym genomic island family members or possibly as RNA-binding proteins.

Most of the putative genomic islands containing Msi172/171 homologues do not contain TraR/TraI homologues, suggesting that the QS regulatory system has been superimposed on an existing regulatory system in a recent evolutionary event to allow regulation of ICEMlSymR7A transfer in response to autoinducer molecules. As noted above, traI1 and traI2 probably arose from a common source. We propose that traR-traI2 were acquired upstream of msi172, and traI1 elsewhere on ICEMlSymR7A either through a duplication event followed by translocation or in a separate acquisition event. This event would have allowed divergent evolution of the traI1 and traI2 loci and therefore permitted fine-tuning of AHL production by TraI1 independently of TraR-activated excision of ICEMlSymR7A through expression of traI2-msi172-msi171.

In summary, our data suggest the following model (Fig. 6). Excision and transfer of ICEMlSymR7A are absolutely dependent on Msi171 and Msi172 and hence on expression of the traI2-msi172-msi171 operon. This operon is normally expressed at a low basal level that is insufficient to allow expression of the conjugative transfer apparatus. Under some set of unknown conditions, the level of TraR in at least some cells in the population rises above a threshold level. This leads to activation of TraR followed by autoinduction of traI1-dependent AHL production and maximally induced expression of msi172-171. Production of Msi172 and Msi171 leads to stable expression of the conjugative apparatus as well as excision.

image

Figure 6. Model for the regulation of ICEMlSymR7A excision and transfer. See text for details.

Download figure to PowerPoint

Future work will focus on determining the mechanism of action of Msi172/Msi171 in activating excision as well as identifying the factors that regulate traR expression or TraR function, and the factors that lead to the growth phase dependence of excision in cultures of the wild-type strain and traR and traI1 mutants.

Experimental procedures

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

Bacteria, plasmids and growth conditions

The bacterial strains and plasmids used in this study are listed in Table 3. Mesorhizobium strains were cultured at 28°C in either TY or RDM media with 10 mM glucose (G/RDM) as described (Ramsay et al., 2006). E. coli strains were cultured at 37°C in either 2YT medium or TY. C. violaceum strain CV026 was cultured at 28°C in either LB or TY medium. Media were supplemented with antibiotics where required at the following concentrations: for E. coli 50 μg ml−1 kanamycin, 12.5 μg ml−1 tetracycline, 100 μg ml−1 ampicillin, 25 μg ml−1 gentamicin; for mesorhizobia 2.0 μg ml−1 tetracycline, 200 μg ml−1 neomycin, 50 μg ml−1 gentamicin, 200 μg ml−1 streptomycin. When required, media were supplemented with 0.1 mM or 1 mM IPTG for M. loti or E. coli cultures respectively.

Table 3.  Bacterial strains and plasmids.
Strain/plasmidDescriptionSource or reference
Mesorhizobium
 N18Nonsymbiotic Mesorhizobium strain; field isolateRamsay et al. (2006)
 R7AField reisolate of ICMP 3153; wild-type symbiotic strainSullivan et al. (1995)
 R7ANSNon-symbiotic derivative of R7A; lacks ICEMlSymR7ARamsay et al. (2006)
 R7AΔmsi171Δmsi171; markerless in-frame deletion mutantThis study
 R7AΔmsi172Δmsi172; markerless in-frame deletion mutantThis study
 R7AΔtraI1ΔtraI1msi039); markerless in-frame deletion mutantThis study
 R7AΔtraI2ΔtraI2msi173); markerless in-frame deletion mutantThis study
 R7AΔtraI2*ΔtraI2msi173) mutant containing a markerless deletion of traI2 and 22 bp of upstream sequenceThis study
 R7AΔtraI2traI1ΔtraI2traI1 double mutantThis study
 R7AΔtraRΔtraRmsi174); markerless in-frame deletion mutantThis study
 R7AtraI1::lacZpFUS2 insertion duplication mutant; traI1 transcriptionally fused to lacZThis study
 R7AtraI2::lacZpFUS2 insertion duplication mutant; traI2 transcriptionally fused to lacZThis study
E. coli
 DH10BF-, mcrAΔ(mrr-hsd RMS-mrcBC) (Φ80lacZΔM15) ΔlacX74 deoR endA1 araD139Δ(ara, leu)7697 galU galKλ-rspL nupGGrant et al. (1990)
 S17-1 λpirTpR SmRrecA thi pro hsdR-M+recA::RP4-2-Tc::Mu Km::Tn7 λpirde Lorenzo et al. (1993)
C. violaceum
 CV026cviI::mini-Tn5 derivative of ATCC 31532, KmR, AHL-McClean et al. (1997)
Plasmids
 pFAJ1700Broad-host-range vector, oriVRK2 TcRDombrecht et al. (2001)
 pFAJ1708Broad-host-range expression vector, PnptII oriVRK2 TcRDombrecht et al. (2001)
 pFUS2oriVColE1oriTRK2lacZ transcriptional reporter; suicide vector, GmRAntoine et al. (2000)
 pJQ200SKSuicide vector containing sacB gene, GmRQuandt and Hynes (1993)
 pTH1227Broad-host-range plasmid containing lacIq-Ptac cassette, TcRCheng et al. (2007)
 pJRtraI1pFAJ1700 containing traI1 (XbaI fragment) and upstream intergenic sequenceThis study
 pJRtraI1EpTH1227 carrying traI1 cloned downstream of lacIq-Ptac promoterThis study
 pJRtraI2pFAJ1700 containing the overlap-extension PCR product used to create R7AΔtraR cloned as an XbaI fragment. Contains traI2 and DNA upstream of traI2 and traRThis study
 pJRtraI2EpTH1227 carrying traI2 cloned downstream of lacIq-Ptac promoterThis study
 pJRtraRpFAJ1700 containing traR (BamHI fragment) amplified by PCR from pJR206This study
 pJR201pUC8 containing PCR products spanning attP, attB and a region of melR, linearized and used as a positive template control for QPCR reactions and standard curvesRamsay et al. (2006)
 pJR206pFAJ1700 containing traRtraI2 and upstream intergenic DNARamsay et al. (2006)
 pNJmsi171EpFAJ1708 with msi171 cloned downstream of nptII promoterThis study
 pNJmsi172EpFAJ1708 with msi172 cloned downstream of nptII promoterThis study
 pPROBE-KTBroad-host-range vector, oriVpVS1oriVp15a NmRMiller et al. (2000)
 pUT11GBroad-host-range pIJ3200-based cosmid from R7A library; contains DNA corresponding to nucleotides 198895 to 221662 of ICEMlSymR7A (msi163 to msi184)U. Sharma, this lab

Strain and plasmid constructions

Primers used in this study are described in Table S1. Plasmids were introduced into E. coli and M. loti strains either by biparental matings using E. coli S17-1 donor strains or by electroporation, as previously described (Ramsay et al., 2006). To construct the markerless in-frame deletion mutants, overlap extension PCR was first used to create the deletions. For each gene to be deleted, approximately 1 kb of DNA flanking each side of the deletion was amplified by PCR using a 5′ flanking primer and a reverse overlap primer or a forward overlap primer and a 3′ flanking primer (see Table S2 for arrangement of the primers). The two purified PCR products were then used as templates in a PCR reaction containing the 5′ and 3′ flanking primers. The purified PCR product from this reaction was cloned into pJQ200SK as an XbaI fragment, and the M. loti mutants then constructed using the allelic replacement strategy previously described for the construction of R7AΔrdfS (Ramsay et al., 2006). The deletions in each mutant strain were confirmed by PCR and sequence analysis.

Strains R7AtraI1::lacZ and R7AtraI2::lacZ were constructed by insertion duplication mutagenesis using the suicide plasmid pFUS2. An internal region of each gene was amplified by PCR using primers traiI1pfusL and traiIpfusR for traI1 and trai2pfusL and trai2pfusR for traI2 (Table S1). The PCR products were cloned into pFUS2 as HindIII/Asp718 (traI1) and HindIII/BamHI (traI2) fragments and the resultant plasmids transferred into R7A by conjugation. Transconjugants were confirmed via Southern hybridization by probing successively with pFUS2 and the PCR products used to construct the mutants.

For construction of pJRtraI1, a 1.4 kb region containing traI1 and upstream intergenic DNA was amplified by PCR (primers traI1clone5 and traI1clone3) and cloned into pFAJ1700 as an XbaI fragment. pJRtraR was constructed by amplifying an 1.4 kb region containing traR and upstream intergenic DNA by PCR (primers traRI2BamHI5 and traRBamHI3) and cloning into pFAJ1700 as a BamHI fragment. The plasmid pJRtraI2 was constructed by subcloning the overlap extension PCR product used to construct R7AΔtraR from pJQ200SK as an XbaI fragment into pFAJ1700. To create plasmids expressing the traI1 or traI2 gene from an IPTG-inducible promoter, each ORF was amplified by PCR using primer pairs I1IPTGF and I1IPTGR or I2IPTGF and I2IPTGR respectively, and cloned into pTH1227 as a XbaI-HindIII or XbaI-PstI fragment, producing plasmids pJRtraI1E and pJRtraI2E. To construct plasmids that expressed msi171 or msi172 from the constitutive nptII promoter, the msi171 and msi172 genes were amplified by PCR using primers msi171FAJ08L and msi171FAJ08R (msi171) or msi172FAJ08L and msi172FAJ08R (msi172). Primers msi171FAJ08L, and msi172FAJ08L which anneal to the 5′ ends of the genes, contained stops in all three frames preceding a synthetic ribosome-binding site. The PCR products were digested with XbaI and Asp718 and then cloned into pFAJ1708 adjacent to the nptII promoter. All plasmid constructs were confirmed by DNA sequencing.

Cosmid pUT11G was isolated from a pIJ3200 library of R7A genomic DNA (Sullivan and Ronson, 1998) by PCR screening and end-sequenced using T3 and T7 primers to determine its content (U. Sharma, pers. comm.).

DNA manipulations and sequence analysis

Mesorhizobium DNA was prepared as described previously (Ramsay et al., 2006). PCR products were amplified using the Phusion High-Fidelity PCR kit (Finnzymes) and purified using the High-Pure PCR product purification kit (Roche). Southern hybridizations and DNA sequencing were carried out as described (Ramsay et al., 2006). Sequence comparisons with the NCBI nr database or RhizoBase (http://bacteria.kazusa.or.jp/rhizobase/) were performed using blast tools (Altschul et al., 1997). Sequence alignments were carried out as described (Chen et al., 2007). The tra-box1 and tra-box2 motifs were initially identified by visual inspection, after which the sequences were used to create a motif pattern using the online MEME tool (Bailey et al., 2006), which was then used in a MAST search (Bailey and Gribskov, 1998) against the ICEMlSymR7A sequence to search for additional tra-boxes. No additional matches were identified.

Quantitative PCR assays for excision

To quantify the attP and attB products of ICEMlSymR7A excision, DNA was extracted from cell cultures at 24 h (exponential phase) or 64 h (stationary phase) and analysed by QPCR as described previously (Ramsay et al., 2006). All data presented are the average of at least two biological replicates and three QPCR replicates each sample.

RNA isolation, reverse transcriptase-PCR and 5-prime RACE analysis

Mesorhizobium loti G/RDM cultures were grown to an OD600 of 0.4–0.6 and 8 ml culture was then added to 10 ml of boiling lysis buffer [2% SDS, 30 mM NaAc (pH 5.5), 3 mM EDTA], mixed thoroughly and incubated at 100°C for 3 min. Protein, genomic DNA and other material were then removed from the aqueous phase by thorough mixing with two 16 ml volumes of acidified phenol (65°C), one 16 ml volume of Tris-equilibrated phenol and one 16 ml volume of chloroform. Nucleic acids were then precipitated with 2 vols of ethanol, after which pellets were washed with 10 ml 70% ethanol and suspended in 1 ml of DEPC-treated H2O containing 400 U of Invitrogen RNaseOUT. Samples were then treated with Ambion TURBO DNase and applied to Qiagen RNeasy columns as per manufacturer's instructions. DNA was still present in these samples (detected by PCR using primers I2IPTGF and msi171SP1) and so samples were treated a second time with DNase and applied to a second RNeasy column that produced DNA-free RNA samples.

A Roche 5′/3′ RACE second-generation kit was used to map the transcriptional start site of traI2-msi172-msi171 as per the manufacturer's instructions. The primers msi171SP1, traI2SP1 and traI2SP2 correspond to specific primers SP1, SP2 and SP3, respectively, as referred to in the kit instructions. A single PCR product was obtained in both the first and second rounds of PCR amplification and so this PCR product was sequenced directly without cloning, using primer traI2SP2.

Extraction of N-acyl homoserine lactones

Broth cultures of M. loti strains were grown in 50 ml of G/RDM for 64 h. Aliquots (8 ml) of supernatant from each culture were passed through 0.45 μm Millipore filters and extracted twice with equal volumes of dichloromethane (McClean et al., 1997). Extracts were then evaporated to dryness in a vacuum centrifuge and resuspended in 50 μl methanol.

CV026 bioassay

For analysis of M. loti AHL production, the AHL-sensitive bioassay strain C. violaceum CV026 (McClean et al., 1997) was used either in agar overlays or streaked adjacent to M. loti cultures on TY plates. TY broth cultures of M. loti were grown for 64 h at 28°C, after which 10–20 μl of culture was spotted onto a 20-cm-diameter TY agar plate and incubated for either 24 or 48 h. For overlays, a 100 ml LB broth was inoculated from an overnight culture of CV026 and incubated overnight at 28°C. One hundred millilitres of molten LB agar cooled to 40°C was mixed with the CV026 culture and quickly applied to the agar plate. For streak-plate assays, a loopful of CV026 from an overnight LB agar plate culture was streaked adjacent to the M. loti culture. The resulting plate was then incubated overnight at 28°C.

LC-MS/MS of M. loti AHLs

Liquid chromatography was carried out on a Shimadzu series 10AD VP LC system fitted with a Phenomenex Gemini C18 150 mm × 2 mm (5 μm particle size) column that was used at 45°C. Mass spectrometry was conducted using a 4000 QTRAP hybrid triple-quadrupole-linear ion trap mass spectrometer (Applied Biosystems), with a TurboIon ion source operating in positive ion mode. AHL molecules were identified by comparison with spectra generated from synthetic AHL standards using precursor ion triggered enhanced product ion spectra. Detailed synthetic and analytical methods and instrument settings are given in Chhabra et al. (1993; 2003) and Ortori et al. (2007). Quantification relative to AHL calibration standards was performed using multiple reaction monitoring (Gould et al., 2006; C. Ortori, et al., in preparation). The lower limit of quantification for each AHL molecule was 5 pmol per sample.

β-Galactosidase assays

β-Galactosidase assays were performed on washed cells from M. loti TY broth cultures grown to stationary phase, as described (Miller, 1972). AHLs were added to broths at the time of inoculation where required.

Conjugative transfer of ICEMlSymR7A

Matings for transfer of ICEMlSymR7A to non-symbiotic M. loti strain N18 were carried out as previously described (Sullivan and Ronson, 1998), except that the recipient strain contained either pFAJ1700 (Dombrecht et al., 2001) or pPROBE-KT (Miller et al., 2000) and either tetracycline or neomycin was included in the medium to counterselect the donor.

Acknowledgements

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

J.P.R. thanks the University of Otago for a PhD scholarship and for an Elman Poole Travelling Scholarship. We thank Gabriella Stuart for carrying out some of the conjugation experiments, Utsav Sharma for providing pUT11G, and Alex Truman and Siri Ram Chhabra for AHL synthesis. This work was supported by a grant from the Marsden Fund administered by the Royal Society of New Zealand.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Altschul, S.F., Madden, T.L., Schäffer, 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.
  • Antoine, R., Alonso, S., Raze, D., Coutte, L., Lesjean, S., Willery, E., et al. (2000) New virulence-activated and virulence-repressed genes identified by systematic gene inactivation and generation of transcriptional fusions in Bordetella pertussis. J Bacteriol 182: 59025905.
  • Bailey, T.L., and Gribskov, M. (1998) Combining evidence using p-values: application to sequence homology searches. Bioinformatics 14: 4854.
  • Bailey, T.L., Williams, N., Misleh, C., and Li, W.W. (2006) MEME: discovering and analysing DNA and protein sequence motifs. Nucleic Acids Res 34: W369W373.
  • Binnewies, T.T., Motro, Y., Hallin, P.F., Lund, O., Dunn, D., La, T., et al. (2006) Ten years of bacterial genome sequencing: comparative-genomics-based discoveries. Funct Integr Genomics 6: 165185.
  • Burrus, V., and Waldor, M.K. (2004) Shaping bacterial genomes with integrative and conjugative elements. Res Microbiol 155: 376386.
  • Burrus, V., Pavlovic, G., Decaris, B., and Guédon, G. (2002) Conjugative transposons: the tip of the iceberg. Mol Microbiol 46: 601610.
  • Chen, G., Jeffrey, P.D., Fuqua, C., Shi, Y., and Chen, L. (2007) Structural basis for antiactivation in bacterial quorum sensing. Proc Natl Acad Sci USA 104: 1647416479.
  • Cheng, J., Sibley, C.D., Zaheer, R., and Finan, T.M. (2007) A Sinorhizobium meliloti minE mutant has an altered morphology and exhibits defects in legume symbiosis. Microbiology 153: 375387.
  • Chhabra, S.R., Stead, P., Bainton, N.J., Salmond, G.P., Stewart, G.S., Williams, P., and Bycroft, B.W. (1993) Autoregulation of carbapenem biosynthesis in Erwinia carotovora by analogues of N-(3-oxohexanoyl) -L-homoserine lactone. J Antibiot (Tokyo) 46: 441454.
  • Chhabra, S.R., Harty, C., Hooi, D.S., Daykin, M., Williams, P., Telford, G., et al. (2003) Synthetic analogues of the bacterial signal (quorum sensing) molecule N-(3-oxododecanoyl)-L-homoserine lactone as immune modulators. J Med Chem 46: 97104.
  • Danino, V.E., Wilkinson, A., Edwards, A., and Downie, J.A. (2003) Recipient-induced transfer of the symbiotic plasmid pRL1JI in Rhizobium leguminosarum bv. viciae is regulated by a quorum-sensing relay. Mol Microbiol 50: 511525.
  • Dobrindt, U., Hochhut, B., Hentschel, U., and Hacker, J. (2004) Genomic islands in pathogenic and environmental microorganisms. Nat Rev Microbiol 2: 414424.
  • Dombrecht, B., Vanderleyden, J., and Michiels, J. (2001) Stable RK2-derived cloning vectors for the analysis of gene expression and gene function in gram-negative bacteria. Mol Plant Microbe Interact 14: 426430.
  • Frost, L.S., Leplae, R., Summers, A.O., and Toussaint, A. (2005) Mobile genetic elements: the agents of open source evolution. Nat Rev Microbiol 3: 722732.
  • Fuqua, C., and Winans, S.C. (1996) Conserved cis-acting promoter elements are required for density-dependent transcription of Agrobacterium tumefaciens conjugal transfer genes. J Bacteriol 178: 435440.
  • Fuqua, C., Burbea, M., and Winans, S.C. (1995) Activity of the Agrobacterium Ti plasmid conjugal transfer regulator TraR is inhibited by the product of the traM gene. J Bacteriol 177: 13671373.
  • Fuqua, W.C., and Winans, S.C. (1994) A LuxR-LuxI type regulatory system activates Agrobacterium Ti plasmid conjugal transfer in the presence of a plant tumor metabolite. J Bacteriol 176: 27962806.
  • Gould, T.A., Herman, J., Krank, J., Murphy, R.C., and Churchill, M.E. (2006) Specificity of acyl-homoserine lactone synthases examined by mass spectrometry. J Bacteriol 188: 773783.
  • Grant, G.S.N., Jessee, F.R., Bloom, F.R., and Hanahan, D. (1990) Differential plasmid rescue from transgenic mouse DNAs into Escherichia coli methylation-restriction mutants. Proc Natl Acad Sci USA 87: 46454649.
  • Guy, L. (2006) Identification and characterization of pathogenicity and other genomic islands using base composition analyses. Future Microbiol 1: 309316.
  • Hwang, I., Li, P.L., Zhang, L., Piper, K.R., Cook, D.M., Tate, M.E., and Farrand, S.K. (1994) TraI, a LuxI homologue, is responsible for production of conjugation factor, the Ti plasmid N-acylhomoserine lactone autoinducer. Proc Natl Acad Sci USA 91: 46394643.
  • Hwang, I., Cook, D.M., and Farrand, S.K. (1995) A new regulatory element modulates homoserine lactone-mediated autoinduction of Ti plasmid conjugal transfer. J Bacteriol 177: 449458.
  • Kaneko, T., Nakamura, Y., Sato, S., Asamizu, E., Kato, T., Sasamoto, S., et al. (2000) Complete genome structure of the nitrogen-fixing symbiotic bacterium Mesorhizobium loti. DNA Res 7: 331338.
  • Khan, S.R., Gaines, J., Roop, R.M., and Farrand, S.K. (2008) Broad-host-range expression vectors with tightly regulated promoters and their use to examine the influence of TraR and TraM expression on Ti plasmid quorum sensing. Appl Environ Microbiol 74: 50535062.
  • Krick, A., Kehraus, S., Eberl, L., Riedel, K., Anke, H., Kaesler, I., et al. (2007) A marine Mesorhizobium sp. produces structurally novel long-chain N-acyl-L-homoserine lactones. Appl Environ Microbiol 73: 35873594.
  • De Lorenzo, V., Eltis, L., Kessler, B., and Timmis, K.N. (1993) Analysis of Pseudomonas gene products using lacIq/Ptrp-lac plasmids and transposons that confer conditional phenotypes. Gene 123: 1724.
  • McAnulla, C., Edwards, A., Sanchez-Contreras, M., Sawers, R.G., and Downie, J.A. (2007) Quorum-sensing-regulated transcriptional initiation of plasmid transfer and replication genes in Rhizobium leguminosarum biovar Viciae. Microbiology 153: 20742082.
  • McClean, K.H., Winson, M.K., Fish, L., Taylor, A., Chhabra, S.R., Cámara, M., et al. (1997) Quorum sensing and Chromobacterium violaceum: exploitation of violacein production and inhibition for the detection of N-acylhomoserine lactones. Microbiology 143: 37033711.
  • Mantri, Y., and Williams, K.P. (2004) Islander: a database of integrative islands in prokaryotic genomes, the associated integrases and their DNA site specificities. Nucleic Acids Res 32: D55D58.
  • Marchler-Bauer, A., Anderson, J.B., Derbyshire, M.K., DeWeese-Scott, C., Gonzales, N.R., Gwadz, M., et al. (2007) CDD: a conserved domain database for interactive domain family analysis. Nucleic Acids Res 35: D237D240.
  • Miller, J.H. (1972) Experiments in Molecular Genetics. Cold Spring Harbor: Cold Spring Harbor Laboratory Press.
  • Miller, W.G., Leveau, J.H., and Lindow, S.E. (2000) Improved gfp and inaZ broad-host-range promoter-probe vectors. Mol Plant Microbe Interact 13: 12431250.
  • Ochman, H., Lawrence, J.G., and Groisman, E.A. (2000) Lateral gene transfer and the nature of bacterial innovation. Nature 405: 299304.
  • Ortori, C.A., Atkinson, S., Chhabra, S.R., Cámara, M., Williams, P., and Barrett, D.A. (2007) Comprehensive profiling of N-acylhomoserine lactones produced by Yersinia pseudotuberculosis using liquid chromatography coupled to hybrid quadrupole-linear ion trap mass spectrometry. Anal Bioanal Chem 387: 497511.
  • Ou, H.Y., Chen, L.L., Lonnen, J., Chaudhuri, R.R., Thani, A.B., Smith, R., et al. (2006) A novel strategy for the identification of genomic islands by comparative analysis of the contents and contexts of tRNA sites in closely related bacteria. Nucleic Acids Res 34: e3.
  • Pappas, K.M., and Winans, S.C. (2003) A LuxR-type regulator from Agrobacterium tumefaciens elevates Ti plasmid copy number by activating transcription of plasmid replication genes. Mol Microbiol 48: 10591073.
  • Piper, K.R., Beck von Bodman, S., and Farrand, S.K. (1993) Conjugation factor of Agrobacterium tumefaciens regulates Ti plasmid transfer by autoinduction. Nature 362: 448450.
  • Piper, K.R., Beck Von Bodman, S., Hwang, I., and Farrand, S.K. (1999) Hierarchical gene regulatory systems arising from fortuitous gene associations: controlling quorum sensing by the opine regulon in Agrobacterium. Mol Microbiol 32: 10771089.
  • Qin, Y.P., Su, S.C., and Farrand, S.K. (2007) Molecular basis of transcriptional antiactivation – TraM disrupts the TraR-DNA complex through stepwise interactions. J Biol Chem 282: 1997919991.
  • Quandt, J., and Hynes, M.F. (1993) Versatile suicide vectors which allow direct selection for gene replacement in gram-negative bacteria. Gene 127: 1521.
  • Ramsay, J.P., Sullivan, J.T., Stuart, G.S., Lamont, I.L., and Ronson, C.W. (2006) Excision and transfer of the Mesorhizobium loti R7A symbiosis island requires an integrase IntS, a novel recombination directionality factor RdfS, and a putative relaxase RlxS. Mol Microbiol 62: 723734.
  • Sanchez-Contreras, M., Bauer, W.D., Gao, M., Robinson, J.B., and Downie, J.A. (2007) Quorum-sensing regulation in rhizobia and its role in symbiotic interactions with legumes. Philos Trans R Soc Lond B 362: 11491163.
  • Schaefer, A.L., Greenberg, E.P., Oliver, C.M., Oda, Y., Huang, J.J., Bittan-Banin, G., et al. (2008) A new class of homoserine lactone quorum-sensing signals. Nature 454: 595599.
  • Su, S.C., Khan, S.R., and Farrand, S.K. (2008) Induction and loss of Ti plasmid conjugative competence in response to the acyl-homoserine lactone quorum-sensing signal. J Bacteriol 190: 43984407.
  • Sullivan, J.T., and Ronson, C.W. (1998) Evolution of rhizobia by acquisition of a 500-kb symbiosis island that integrates into a phe-tRNA gene. Proc Natl Acad Sci USA 95: 51455149.
  • Sullivan, J.T., Patrick, H.N., Lowther, W.L., Scott, D.B., and Ronson, C.W. (1995) Nodulating strains of Rhizobium loti arise through chromosomal symbiotic gene transfer in the environment. Proc Natl Acad Sci USA 92: 89858989.
  • Sullivan, J.T., Trzebiatowski, J.R., Cruickshank, R.W., Gouzy, J., Brown, S.D., Elliot, R.M., et al. (2002) Comparative sequence analysis of the symbiosis island of Mesorhizobium loti strain R7A. J Bacteriol 184: 30863095.
  • Toussaint, A., Merlin, C., Monchy, S., Benotmane, M.A., Leplae, R., Mergeay, M., and Springael, D. (2003) The biphenyl- and 4-chlorobiphenyl-catabolic transposon Tn4371, a member of a new family of genomic islands related to IncP and Ti plasmids. Appl Environ Microbiol 69: 48374845.
  • Waters, C.M., and Bassler, B.L. (2005) Quorum sensing: cell-to-cell communication in bacteria. Annu Rev Cell Dev Biol 21: 319346.
  • White, C.E., and Winans, S.C. (2007) Cell-cell communication in the plant pathogen Agrobacterium tumefaciens. Philos Trans R Soc Lond B 362: 11351148.
  • Williams, P., Winzer, K., Chan, W.C., and Cámara, M. (2007) Look who's talking: communication and quorum sensing in the bacterial world. Philos Trans R Soc Lond B Biol Sci 362: 11191134.
  • Yang, M., Sun, K., Zhou, L., Yang, R., Zhong, Z., and Zhu, J. (2009) Functional analysis of three AHL autoinducer synthase genes in Mesorhizobium loti reveals the important role of quorum sensing in symbiotic nodulation. Can J Microbiol 55: 210214.
  • Yoon, S.H., Hur, C.G., Kang, H.Y., Kim, Y.H., Oh, T.K., and Kim, J.F. (2005) A computational approach for identifying pathogenicity islands in prokaryotic genomes. BMC Bioinformatics 6: 184.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information
FilenameFormatSizeDescription
MMI_6843_sm_Tables_S1-S2.pdf66KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.