1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Conjugal transfer of the Ti plasmid pTiC58 is regulated by a quorum-sensing system involving the transcriptional activator TraR and the acyl homoserine lactone autoinducer N-(3-oxo-octanoyl)-l-homoserine lactone (AAI). Activation of tra gene expression by TraR and AAI is inhibited by TraM, an 11 kDa protein also coded for by the Ti plasmid. Previous studies suggested that TraM interferes with TraR activity by directly interacting with the activator protein. Using the yeast two-hybrid system, constructs of Saccharomyces cerevisiae containing a fusion of traR to the B42 domain of the prey plasmid pJG4.5 and a fusion of traM to the lexA gene of the bait plasmid pEG202 produced β-galactosidase and grew on medium lacking leucine, both phenotypes indicative of an interaction between the two proteins. Early termination mutants and substitution mutants mapping to the C-terminus of TraM were isolated by screening for alleles unable to interfere with TraR activity in Agrobacterium tumefaciens. These mutants all failed to interact with the TraR fusion in the two-hybrid system. An N-terminal deletion mutant of TraM lacking the first 27 residues weakly interacted with TraR in the two-hybrid system whereas deletions of 48 amino acids or more abolished the interaction. As assessed by Western blot analysis, the mutant fusion proteins were produced at levels indistinguishable from that of the wild-type TraM in the yeast tester strain. Mutants of TraR that were not inhibited by TraM in A. tumefaciens were isolated and fell into two classes. In the first, the mutation resulted in increased expression of wild-type TraR. In the second, a proline residue at position 176 was changed to serine (P176 [RIGHTWARDS ARROW] S) or to leucine (P176 [RIGHTWARDS ARROW] L). The P176 [RIGHTWARDS ARROW] S mutant interacted with wild-type TraM, but at a detectably lower level, in the two-hybrid assay. Mutants of TraR with N-terminal deletions as large as 105 amino acids interfered with the ability of TraM to inhibit wild-type TraR in A. tumefaciens. Two-hybrid assays indicated that these mutants, as well as a C-terminal 49 residue fragment of TraR, can interact with TraM. We conclude that TraM and TraR interact in vivo and that this interaction is responsible for inhibition of TraR-mediated activation. We also conclude that the two proteins interact with each other through domains located at their respective C-termini.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Conjugal transfer of the nopaline/agrocinopine-type Ti plasmid pTiC58 is regulated by two hierarchical signalling systems. At the highest level, a subset of opines, produced by the crown gall tumours induced by the Agrobacterium tumefaciens strain harbouring this Ti plasmid, activate transfer. However, while directly regulating expression of the Ti plasmid genes involved in their transport and catabolism, the conjugal opines only indirectly control expression of the tra regulon. Rather, this set of genes, composed of the three operons traAFB, traCDG and trb, is controlled by the LuxR-like transcriptional activator TraR in concert with its acyl homoserine lactone (acyl-HSL) co-inducer AAI [N-(3-oxo-octanoyl)-l-homoserine lactone] (Piper et al., 1993; Zhang et al., 1993; reviewed in Farrand, 1998). Opines regulate conjugation by controlling the expression of traR. For example, transfer of pTiC58 is induced by the sugar phosphodiester opines agrocinopines A and B (Ellis et al., 1982). In this plasmid, traR is a member of a five-gene operon called arc (Piper et al., 1999) This operon is expressed from a promoter regulated by AccR, the transcriptional repressor responsive to these signal molecules (Beck von Bodman et al., 1992; Piper et al., 1999). In the absence of the conjugal opines, the expression of arc is repressed by AccR. When the opines are present, repression by AccR is relieved and the arc operon, including traR, is expressed.

LuxR homologues and their cognate acyl-HSLs serve to induce certain gene sets only when the bacterial population has reached some critical size. Called quorum sensing, the bacteria use the accumulation of the acyl-HSL signal to monitor their population density (Fuqua and Greenberg, 1998). During growth, each cell in the population produces the acyl-HSL at a basal level. The signal is released into the environment, either by passive diffusion or by a combination of diffusion and active efflux (Kaplan and Greenberg, 1985; Evans et al., 1998; Pearson et al., 1999), where it accumulates at ever-increasing amounts with growth of the bacterial population. In at least two of the systems, the signal freely diffuses back into the cells such that its intracellular concentration also rises as a function of the increase in the bacterial population (Kaplan and Greenberg, 1985; Evans et al., 1998; Pearson et al., 1999). At some target population size, the acyl-HSL reaches a concentration within the cells allowing it to interact with the transcriptional activator. This interaction converts the activator into its functional form thereby triggering expression of the target genes.

Although influenced by other global circuitry, regulation of the lux operon of Vibrio fischeri, the paradigm quorum-sensing system, apparently is composed of only two components, LuxR and its acyl-HSL co-inducer VAI [N-(3-oxo-hexanoyl)-l-HSL]. Quorum-dependent regulation of the tra regulon, however, contains a third element, TraM, which negatively modulates TraR/AAI-mediated induction of the Ti plasmid tra regulon (Fuqua et al., 1995; Hwang et al., 1995). Two lines of evidence indicate that TraM interferes with the regulatory process by directly inhibiting TraR. First, although mutations in traM confer a transfer-constitutive (Trac) phenotype, TraM does not directly affect transcription of traR or of the tra and trb operons (Hwang et al., 1995). Second, inhibition depends on the relative levels of expression of traM and traR ; raising the levels of TraR with respect to TraM overcomes the inhibitory effect. Furthermore, under normal conditions, inhibition by TraM cannot be overcome by addition of excess AAI (Zhang and Kerr, 1991; Hwang et al., 1995). From these observations, we proposed a model in which TraM interacts directly with TraR to sequester the activator from suboptimal levels of AAI, thus preventing premature conjugation when the bacterial population size is low (Hwang et al., 1995). In this report, we present evidence that supports this hypothesis. Using the yeast two-hybrid system (Gyuris et al., 1993), we show that TraR and TraM can interact in vivo. In addition, we identify mutants of TraM that are unable to inhibit TraR in A. tumefaciens and show that these mutants are impaired in their ability to interact with the activator. We also identify mutants of TraR refractory to inhibition by TraM. Some of these mutations map to the traR gene whereas others affect the level at which the wild-type activator is produced. We conclude that TraM is an antiactivator and that it modulates transcriptional activation of the Ti plasmid tra regulon by directly interacting with TraR.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

TraR interacts with TraM

We used the yeast two-hybrid system to determine whether TraR and TraM can interact with each other. For these tests, traR was fused to the Escherichia coli B42 activation domain of the prey plasmid pJG4.5 to produce pJGR and traM was fused to the lexA DNA-binding domain sequence of the bait plasmid pEG202 to produce pEGM. In the former construct, expression of the B42 fusion is repressed by glucose and induced by galactose. The two plasmids were introduced into Saccharomyces cerevisiae EGY48 and transformants were tested for production of β-galactosidase and leucine-independent growth, phenotypes indicative of interaction between the two fusion proteins. Strains harbouring both plasmids grew on medium lacking leucine, but only when galactose was provided to induce expression of the B42 fusion in the prey plasmid (Fig. 1). Similarly, the transformants produced detectable amounts of β-galactosidase activity, but only when grown on medium containing galactose (Table 1). Expression of either fusion protein in combination with the empty complementing plasmid resulted in a non-interactive phenotype even when the cells were cultured on medium containing galactose (Fig. 1 and Table 1[link]). Thus, the interactive phenotypes are dependent upon the presence of both fusion proteins.


Figure 1. . Interactions between TraR and TraM and their mutants. Strains of Saccharomyces cerevisiae EGY48 harbouring interaction constructs of the bait plasmid pEG202 with traM or its mutant derivatives and the prey plasmid pJG4.5 with traR or its P176 [RIGHTWARDS ARROW] S mutant traR7 were tested for growth on medium containing: A, galactose without leucine; B, glucose + leucine; C, glucose without leucine. Positions represent the growth patterns of strains expressing: 1, TraR and pEG202; 2, TraM and pJG4.5; 3, TraR and TraM; 4, TraR and TraM2; 5, TraR and TraM4; 6, TraR and TraM1; 7, TraR and TraM11; 8, TraR7 and TraM; 9, TraR and TraM28; 10, TraR and TraM48; 11, TraR and TraM72.

Download figure to PowerPoint

Table 1. . Interactions between TraR and TraM and their mutants. a. traR and its mutant allele were fused to the B42 domain of the prey plasmid pJG4.5.b.traM and its mutant alleles were fused to the lexA domain of the bait plasmid pEG202.c. Expressed as units of β-galactosidase activity (Ausubel et al., 1989). All assays were repeated at least twice. Data are from a representative experiment.d.Saccharomyces cerevisiae EGY48 was grown in complete minimal medium containing either glucose or a mixture of raffinose and galactose as sole source of carbon.Thumbnail image of

Isolation of traM mutants unable to inhibit TraR activity in Agrobacterium

Activation of a traG ::lacZ fusion by TraR is inhibited if TraM is co-expressed at sufficient levels (Hwang et al., 1995). We mutagenized pDCBP, which carries these two genes, with hydroxylamine in vitro, transformed the DNA into NT1(pH4I41), which harbours the traG ::lacZ reporter fusion, and screened for derivatives in which the traM gene no longer inhibited TraR-mediated expression of the reporter on medium containing AAI and Xgal. Ten blue colonies were isolated and purified. In each case, the BamHI/Pst I fragment containing traM was excised and recloned into pPLE33, which contains a wild-type allele of traR. Cloning a copy of wild-type traM into this vector results in inhibition of TraR activity. When recloned, each of the 10 traM genes obtained from the mutagenesis retained the non-inhibiting phenotype (Table 2 and data not shown).

Table 2. . TraM mutants do not inhibit TraR activation of traG in Agrobacterium. a. Alleles of traM were tested in A. tumefaciens NT1 harbouring pH4I41 which contains a copy of wild-type traR as well as the traG ::lacZ reporter fusion. AAI was added to each culture to a final concentration of 25 nM.b. WT, wild type; Δ, deletion, with extent shown in amino acid residues deleted; FS, frameshift mutation at the nucleotide position shown; V86 [RIGHTWARDS ARROW] G and P97 [RIGHTWARDS ARROW] S, amino acid substitution mutations at the residue position shown.c. Expressed as units of β-galactosidase activity per 108 colony forming units (cfu). All assays were repeated at least twice and data from a representative experiment are shown.Thumbnail image of

By sequence analysis, the 10 traM mutants fell into three groups. Six contained nonsense mutations, two contained missense mutations, and two contained frameshift mutations (Fig. 2 and Table 3[link]). TraM2, TraM4, TraM6 and TraM7 contain stop codons at positions 82, 67, 51 and 41 respectively. In TraM1, valine 86 is changed to glycine (V86 [RIGHTWARDS ARROW] G), whereas in TraM11 proline 97 is mutated to serine (P97 [RIGHTWARDS ARROW] S). traM9 contains a single nucleotide (A) insertion between codons 86 and 87 resulting in a polypeptide that contains 36 additional missense residues and thus is 20 amino acids longer than the wild-type protein. In traM5, a single nucleotide deletion at codon 67 results in a protein that is 82 residues in size and which contains 16 missense amino acids at its C-terminus. traM3 is allelic to traM2 and traM10 is allelic to traM4, although each came from an independent mutagenesis.

Table 3. . Mutants of TraR and TraM used in this study. a. HA, hydroxylamine treatment; PCR, gene constructions by polymerase chain reaction.b. M, missense mutation; N, nonsense mutation; FS, frameshift mutation; NTD, constructed N-terminal deletion; T, G to A transition mutation; IF, constructed internal fragment.c. Sites of mutations are given as nucleotide positions within traR or traM, with the A of the initiation codon as position 1. traR3 contains a mutation located 113 nucleotides upstream of the ATG.d. Sites of mutations or extents of deletions are given in amino acid residue numbers for each protein with the initiating methionine as position 1.Thumbnail image of

Figure 2. . Structure of traM and its mutant alleles. The upper section shows the genetic organization of traM from pTiC58. Position 1 represents the A of the translational initiation codon while 306 represents the third position base of the last codon of the gene. The lower section shows the structure of the 102-residue TraM protein with the shaded box representing the strongly hydrophobic domain at the C-terminus. The sites of each relevant hydroxylamine-induced mutation are shown along with the type of mutation and the allele name. FS indicates a frameshift mutation. Also shown are the C- and N-terminal deletion derivatives of TraM, each with its allele name.

Download figure to PowerPoint

We recloned several of these traM mutants behind the BAD promoter into pBBRC1MCS-3 and tested the constructs in NT1 harbouring pKPCMI41. This latter plasmid is a derivative of pTiC58 carrying a traG ::lacZ fusion and a null mutation in traM and consequently expresses the reporter (Hwang et al., 1995). However, because TraR is produced at low levels, only small amounts of TraM are required to inhibit the activator. Thus, the reporter fusion in this plasmid is a sensitive indicator of TraM activity. Consistent with this expectation, when expressed from the BAD promoter wild-type TraM inhibited activation of the reporter in this strain even when cultured without arabinose (Table 4). Such inhibition is consistent with the observation of Newman and Fuqua (1999) that the ara promoter is expressed at a relatively high basal level in A. tumefaciens. Constructs expressing TraM1, TraM2 and TraM4 each failed to inhibit TraR activation of the reporter, even when their expression was induced with arabinose (Table 4). TraM11 partially inhibited TraR activity when induced by arabinose, but exhibited no detectable activity in the absence of the inducer.

Table 4. . Overexpression of traM11, but not other mutant alleles of traM, results in weak inhibition of TraR activity. a. Mutations are as described in Table 3.b. Expressed as units of β-galactosidase activity per 108 cfu. Cells were grown in ABM minimal medium. Each assay was repeated three times. Data from a representative experiment are shown.c. Expression of traM and its mutant alleles was induced by addition of arabinose to a final concentration of 0.4%.Thumbnail image of

Western blot analysis using anti-TraM polyclonal antibody revealed that when induced by arabinose TraM11 was detectable at levels similar to that of wild-type TraM (data not shown). TraM1 was expressed at detectable but somewhat lower levels, whereas TraM2 and TraM4, the two early termination mutants tested, were not detected in the lysates.

N-terminal deletion mutants of TraM

The amino acid substitutions in TraM1 and TraM11 map to the hydrophobic C-terminus of the protein (Fig. 2). Moreover, we did not obtain mutants of TraM with alterations located in the N-terminal region, suggesting that the active site is located somewhere in the C-terminal portion of the protein. Using PCR, we constructed a series of 5′ deletion derivatives of traM in which the first codon of the deleted gene is fused to an ATG expressible from the BAD promoter of the vector. Clones pASM28 (traM28 ), pASM48 (traM48 ) and pASM72 (traM72 ), coding for N-terminal deletion derivatives consisting of residues 28–102, 48–102 and 72–102 (Table 3), were tested for their ability to inhibit TraR-mediated activation of the reporter fusion in pKPCMI41. Each failed to inhibit the activator even in cells grown with 0.4% arabinose (Table 4). However, we were unable to detect crossreacting protein in lysates from any of these mutants by Western blot analysis using the anti-TraM antiserum (data not shown).

Isolation of traR mutants not inhibited by TraM

Clones of NT1(pDCI41) containing pSVB33 mutagenized in vitro with hydroxylamine were screened on medium containing AAI and Xgal for mutants of TraR that are not inhibited by TraM. Thirteen blue colonies were isolated and further analysed. All 13 mutants retained a dependence on AAI for activation of the traG ::lacZ reporter (Table 5 and data not shown). By sequence analysis, nine of these mutants, including traR7, were identical, containing a C [RIGHTWARDS ARROW] T transition at bp position 528. This mutation results in a substitution of a serine residue for the proline at position 176 (P176 [RIGHTWARDS ARROW] S) (Table 3 and Fig. 3[link]). In two mutants, traR11 and traR28, this same proline is replaced by leucine (P176 [RIGHTWARDS ARROW] L) (Table 3 and Fig. 3[link]). The open reading frame of traR3 is unaltered, but the mutant contains a single G [RIGHTWARDS ARROW] A substitution 113 nucleotides upstream from the ATG initiation codon (Fig. 3). The sequence of the entire 1.8 kb EcoRI fragment containing the traR5 allele is identical to that of the wild type. When recloned into a new copy of the vector, the activator encoded by this fragment was inhibited by TraM (data not shown).

Table 5. . Mutants of TraR are refractory to inhibition by TraM but still require AAI. a. Mutants of traR are as described in Table 3.b. Expressed as units of β-galactosidase activity per 108 cfu. Cells were grown in ABM medium. Each assay was repeated twice and data are from a representative experiment.c. AAI was added to cultures to a final concentration of 25 nM.d. Tested using pH4I41 which contains the traG ::lacZ reporter but lacks traM (Hwang et al., 1995).Thumbnail image of

Figure 3. . Structure of traR and its mutant alleles. The upper section shows the genetic organization of EcoRI fragment 33 from pTiC58. Position 1 represents the A of the translational initiation codon of traR while 702 represents the third position base of the last codon of the gene. The arrow labelled splA indicates the position of the 3′ end of the splA gene located on the clone. The position and nature of the traR3 mutation, located in the 5′ untranslated region of traR, is indicated. The lower section shows the structure of the 234-residue TraR protein with the shaded box representing the helix–turn–helix domain. The location and nature of the two substitution mutations at position 176 and their allele names are shown. Also shown are the N-terminal deletion derivatives of TraR and one internal fragment of the protein, each with its allele name.

Download figure to PowerPoint

The mutation upstream of traR3 affects expression of the gene

The mutation in traR3 is located in the intergenic region of the arc operon between splA and traR (Fig. 3; Piper et al., 1999). Because inhibition of TraR activity by TraM is dependent upon the relative amounts of the two proteins present in the cell, we reasoned that this mutation results in an elevated level of expression of the gene. We examined this possibility by comparing the expression of an in frame fusion to lacZ with wild-type traR and with traR3 in two sets of strains. When present on clones in strain NT1, which lacks a Ti plasmid, the traR3 ::lacZ fusion was expressed at a level almost eight times higher than that of an identical fusion with the wild-type gene (Table 6). Similarly, the traR3 ::lacZ fusion expressed at a level three times higher than that of the wild type when the two were marker exchanged into pTiC58, the wild-type Ti plasmid. However, the two fusions expressed at similar levels when marker exchanged into the Trac Ti plasmid pTiC58ΔaccR, which expresses traR at constitutive levels from the far upstream arc promoter (Table 6) (Piper et al., 1999).

Table 6. . The mutation at −113 in the traR3 allele results in increased expression of the gene. a. Expressed as units of β-galactosidase activity per 108 cfu. Each assay was repeated twice. Data are from a representative experiment.b.traR ::lacZ and traR3 ::lacZ were marker exchanged (mx) into pKP24 to form pKP24–12 and pH3RZ respectively.c.traR ::lacZ and traR3 ::lacZ were marker exchanged (mx) into pTiC58 to form pKPC12 and pIHC3RZ respectively.d.traR ::lacZ and traR3 ::lacZ were marker exchanged (mx) into pTiC58ΔaccR to form pKPK12 and pIHK3RZ respectively.Thumbnail image of

Activity of the TraR mutants is partially inhibited by TraM

We quantified the effect of TraM on TraR and its mutants by assessing β-galactosidase activity from the traG ::lacZ fusion in the presence of increasing amounts of AAI. As expected, TraM completely inhibited activation of the reporter mediated by wild-type TraR, even at high concentrations of AAI (Fig. 4A). However, although each of the tested TraR mutants activated the reporter in the presence of TraM, levels of expression were two- to eightfold lower than those observed in the absence of the antiactivator at all concentrations of AAI tested (Fig. 4A and B). TraR3, the putative up-promoter mutant, exhibited greater activity than wild-type TraR at saturating concentrations of AAI (Fig. 4B), but was significantly inhibited by TraM (Fig. 4A). In contrast, the two substitution alleles TraR7 and TraR28 were less active than wild-type TraR at all concentrations of AAI tested, and, when based on percentage of activity lost, were significantly less inhibited by TraM compared with the wild-type activator (Fig. 4A and B).


Figure 4. . Effect of TraM on AAI-dependent tra gene activation by TraR and its mutants. Strains of A. tumefaciens NT1 harbouring a traG ::lacZ reporter clone (A) with traM (pDCI41) or (B) without traM (pH 4I41) and the traR clone pSVB33 or its mutant derivatives were cultured in medium with various amounts of AAI. The cells were harvested and assayed for β-galactosidase activity as described in the Experimental procedures. ▪, wild-type traR ; ●, traR3 ; \tf="PS6658"6, traR5 ; ▴, traR7 ; +, traR28.

Download figure to PowerPoint

N-terminal deletion mutants of TraR interfere with inhibition of wild-type TraR by TraM

We have observed that alleles of TraR with N-terminal deletions ranging from 5 to 185 residues abolished activation and also DNA-binding ability as measured in a repressor assay (Luo and Farrand, 1999a). However, these mutants are recessive to wild-type TraR. We reasoned that, although unable to activate transcription, these N-terminal deletion mutants still may interact with TraM. To assess this, we tested several deletion derivatives in a strain containing wild-type traR and traM along with a traG ::lacZ reporter. In the absence of a traR mutant, wild-type TraM inhibits activity of TraR and the reporter is silent (Table 7). However, strains co-expressing derivatives of TraR deleted of up to 90 N-terminal residues strongly activated the reporter fusion (Table 7). We interpret this to mean that the deletion derivatives of TraR are interfering with the interaction between TraM and the wild-type activator. TraRΔ2-104, which lacks 104 N-terminal residues, exerted modest interference, whereas mutants of TraR lacking 120 or more N-terminal residues had no effect on TraM-mediated inhibition of wild-type TraR activity (Table 7). However, we could not detect these shorter polypeptides of TraR in extracts of A. tumefaciens by Western blot analysis (data not shown).

Table 7. . Some N-terminal deletion mutants of TraR interfere with TraM-mediated inhibition of wild-type TraR activity. a. Mutants of traR are described in Table 3.b. Length of each N-terminal deletion derivative is given in amino acid residues.c. Expressed as units of β-galactosidase activity per 109 cfu. Each deletion allele of traR was tested in NT1 harbouring pRMLH4I41, which codes for traR, traM and the traG ::lacZ reporter fusion. Cells were grown in ABM containing AAI at a final concentration of 25 nM.Thumbnail image of

Interaction properties of TraM and TraR mutants in the yeast two-hybrid system

We tested each of the TraM mutants and the P176 [RIGHTWARDS ARROW] S substitution mutant of TraR for interaction with their wild-type counterparts using the yeast two-hybrid system. As judged by growth on medium lacking leucine and by β-galactosidase activity, none of the early termination mutants of TraM examined interact with wild-type TraR when tested against the B42 fusion (Fig. 1; Table 1[link]). Strains expressing the LexA::TraM1 (V86 [RIGHTWARDS ARROW] G) fusion also failed to produce β-galactosidase or grow in the absence of leucine when tested with the TraR fusion. However, a strain harbouring the LexA::TraM11 (P97 [RIGHTWARDS ARROW] S) fusion grew on medium lacking leucine (Fig. 1) and produced low but significant levels of β-galactosidase (Table 1). In both tests, the phenotype was dependent upon galactose and also upon the presence of the B42::TraR fusion. Thus, mutations affecting the C-terminus of TraM lessened or abolished interaction with TraR as judged by the two-hybrid assay.

Strains harbouring the B42::TraR fusion plasmid and fusions between LexA and TraM48 or TraM72, deleted for the first 47 and 71 residues respectively, failed to produce β-galactosidase or grow in the absence of leucine (Fig. 1; Table 1[link]). However, the tester strain expressing the LexA::TraM28 fusion, which is deleted for the first 27 amino acids, produced low but significant levels of β-galactosidase and grew, although poorly, on medium lacking leucine (Fig. 1; Table 1[link]).

Because several of the TraM mutants, including all of the N-terminal deletion derivatives, were not detectable in A. tumefaciens by Western blot analysis, we were concerned that the two-hybrid tests could be influenced by instability of the LexA fusions. Western blot analysis of extracts from yeast cells using anti-TraM antiserum detected all of the LexA::TraM fusions tested except TraM4 (1–67) and TraM72 (72–102), although TraM1, TraM48 and TraM28 did not produce strong signals (data not shown). However, using antibody directed against LexA, we could detect all of the TraM fusions (Fig. 5). Furthermore, all gave signals of approximately the same intensity and the fusion proteins electrophoresced with mobilities corresponding to their expected sizes.


Figure 5. . Western blot analysis of fusions between LexA and TraM and its mutants. Total protein was isolated from Saccharomyces cerevisiae EGY48 harbouring TraM fusion derivatives of pEG202, electrophoresed on SDS polyacrylamide gels, and transferred to nylon membranes. The blots were probed with anti-LexA antiserum and reactive bands were visualized by chemiluminescence, all as described in the Experimental procedures. In most cases, two or three independent clones of each fusion were tested. Lanes contain fusions of LexA to: 1, 2 and 20, wild-type TraM; 3 and 4, TraM4; 5 and 6, TraM10; 7 and 8, TraM11; 10, TraM72; 11–13, TraM28; 14–16, TraM48; 17–19, TraM72. Lane 9 contains total protein from EGY48 devoid of pEG202 or its derivatives.

Download figure to PowerPoint

The B42::TraR7 (P176 [RIGHTWARDS ARROW] S) fusion promoted growth of the yeast tester strain on medium lacking leucine, but only when the wild-type LexA::TraM fusion was present and when galactose was provided as inducer (Fig. 1). The strain also produced low but significant levels of β-galactosidase, but again only in medium containing galactose (Table 1).

We also tested selected N-terminal deletion derivatives of TraR for interaction with wild-type TraM. Fusions between the B42 domain and a C-terminal fragment containing residues 120–234 (traRΔ2–119 ), 130–234 (traRΔ2–129 ) and 185–234 (traRΔ2–184 ) (Fig. 6) promoted growth on medium lacking leucine (Fig. 6). Similarly, a strain expressing a B42 fusion with an internal fragment of TraR comprising residues 121–185 (traR121–185 ) grew on medium lacking leucine (Fig. 6). In all cases, the phenotypes were not dependent upon induction by galactose. However, expression required the presence of pEGM, the lexA ::traM fusion plasmid. Cells lacking this element, or harbouring the lexA bait plasmid without traM, failed to grow on medium lacking leucine. The results suggest that a region of TraR extending in both directions from residue 185 is responsible for interaction with TraM.


Figure 6. . Interactions between TraM and N-terminal deletion derivatives of TraR. Clones of pJG4.5 with fusions between the B42 domain and wild-type TraR or its N-terminal deletion derivatives were introduced into Saccharomyces cerevisiae EGY48 harbouring the LexA::TraM fusion plasmid pEGM or the LexA::TraM4 fusion plasmid pEGM4 and the resulting strains were tested for growth on medium containing: A, galactose without leucine; B, glucose + leucine; and C, glucose without leucine. Positions 1–4 show the growth patterns of strains expressing wild-type TraM and: 1, TraRΔ2–185; 2, TraRΔ121–185; 3, TraRΔ2–120; 4, TraRΔ2–130. Positions 6–9 show the growth patterns of strains expressing TraM4 and: 6, TraRΔ2–185; 7, TraRΔ121–185; 8, TraRΔ2–120; 9, TraRΔ2–130. Positions 11–14 show the growth patterns of strains expressing the empty LexA vector pEG202 and: 11, TraRΔ2–185; 12, TraRΔ121–185; 13, TraRΔ2–120; 14, TraRΔ2–130. Position 5 shows the growth pattern of the positive control strain expressing wild-type forms of TraR and TraM, position 10 shows the growth pattern of a strain expressing wild-type TraR and the non-interacting alleleTraM4, and position 15 shows the growth pattern of the control strain expressing TraR and the empty LexA plasmid pEG202.

Download figure to PowerPoint


  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Two lines of evidence from this work support our hypothesis that TraM inhibits activation of the tra regulon by directly interacting with TraR. First, as judged by two-hybrid analysis, the wild-type forms of TraM and TraR can interact with each other. Moreover, mutants of TraM that no longer inhibit TraR in A. tumefaciens generally fail to interact with the activator in the two-hybrid assay. Thus, there is a strong correlation between the in vivo activity of the antiactivator and its mutants in A. tumefaciens and their activity in the two-hybrid assay. TraR11 is particularly interesting in this respect. This mutant fails to inhibit TraR in A. tumefaciens when expressed at moderate levels from its own promoter. However, increasing expression of this allele from the BAD promoter resulted in a significant level of inhibition of TraR activity in A. tumefaciens, but only upon induction by arabinose. Consistent with this, TraM11 interacts weakly with TraR as assessed in the two-hybrid assay. Thus, the substitution of serine for proline at position 97 in TraM11 apparently lessens, but does not abolish, the interaction between this protein and TraR.

Second, as judged by their effect on TraM-mediated inhibition of wild-type TraR, the transcriptionally inactive N-terminal deletion mutants of TraR apparently can interact with the antiactivator in A. tumefaciens. This conclusion is supported by results from the two-hybrid assays; all such mutants of TraR gave a positive interaction phenotype with TraM in the yeast screens.

Analysis of deletion and substitution mutants suggests that the C-terminal domain of TraM is particularly important for its inhibitory activity. In this regard, TraM resembles NifL of Klebsiella pneumoniae (Sidoti et al., 1993). This sensor/antiactivator interacts with its target NifA through its C-terminus (Narberhaus et al., 1995). The C-terminal region of TraM is characterized by a strongly hydrophobic domain extending from residue 84 to residue 98. TraM1 and TraM11 both contain amino acid substitutions located in this region. Of the two mutant sites, only the proline at position 97 is conserved among the two known active alleles of TraM. However, the valine at position 84 of TraM from pTiC58 is replaced by a leucine in TraM from pTiR10 (Fuqua et al., 1995). Apparently, conservative substitutions can be tolerated at this position, while a non-conservative change such as glycine yields a defective protein. Furthermore, despite the robust nature of our screen, alleles with substitution mutations located elsewhere in TraM were not isolated. In contrast, we did isolate several early termination mutants. The longest of these, TraM2, lacks only 21 C-terminal residues, but this comprises the entire hydrophobic domain. Significantly, this polypeptide fails to interact with TraR in the two-hybrid screen.

One or more domains within the middle third of TraM also are required for activity. While mutants of TraM deleted for 48 or more residues at the N-terminus fail to interact in the two-hybrid assay, a mutant lacking only the first 27 residues exhibits weak interaction with TraR. Thus, one or more regions of TraM located in the interval beginning somewhere between residues 27 and 48 and extending through the hydrophobic region at the carboxy-terminus are required for interaction with TraR.

The protein products of several of the traM mutants, including all of the N-terminal deletion derivatives, were not detectable by Coomassie blue staining or Western blot analysis in lysates of A. tumefaciens. However, all of the LexA::TraM fusion proteins reacted strongly with anti-LexA antibody, but not with anti-TraM antibody, in lysates from S. cerevisiae, and each electrophoresed with the mobility expected for its size. These results suggest that mutants with large C-terminal deletions, such as TraM4 and TraM72, lack epitopes recognizable by our anti-TraM antibody. Similarly, TraM1, TraM28 and TraM48, which react less strongly with anti-TraM antibody, apparently lack at least some of the epitopes recognized by the anti-TraM antiserum. We conclude that although some of these mutant proteins may be rapidly degraded in A. tumefaciens all of the LexA–mutant fusions are as stable as the LexA fusion to wild-type TraM in S. cerevisiae. Thus, the failure of these mutants to interact with TraR in the two-hybrid tests is not likely to be due to degradation of the TraM component of the LexA fusion construct in the yeast tester strain.

Two lines of evidence indicate that TraM interacts with a region located at the carboxy end of TraR. First, although unable to initiate transcription, N-terminal deletion mutants of TraR missing as many as 90 residues apparently can interfere with interactions between TraM and wild-type TraR. We take this dominant-interfering phenotype to mean that these deletants interact with TraM, thereby titrating the available antiactivator. Second, such deletion derivatives yield a positive interaction phenotype when tested with TraM in the two-hybrid system. Moreover, the phenotype, while requiring the LexA::TraM fusion construct, is not dependent upon induction by galactose, suggesting that the interaction between these mutant peptides and TraM is very strong. This is consistent with the strongly dominant-interfering phenotype against TraM observed for these mutants in A. tumefaciens. The shortest polypeptide tested, which contains only the C-terminal 48 residues (residues 186–234), interacts strongly with TraM in the two-hybrid screen. However, a fusion between the B42 domain and a non-overlapping internal polypeptide containing residues 121–185 also gives a positive interaction phenotype. We conclude from these results that TraM interacts with one or more domains of TraR comprising a fairly large region beginning upstream of residue 185 and continuing perhaps as far as the C-terminus of the protein. In this regard, TraR resembles FliA, the sigma factor responsible for the expression of a subset of operons required for flagellar synthesis in Salmonella typhimurium. FliA is inhibited by the anti-sigma factor FlgM (Ohnishi et al., 1992). Furthermore, like their TraR counterparts, N-terminal deletion mutants of FliA relieve inhibition by titrating the available FlgM (Kutsukake et al., 1994). Consistent with this, purified FlgM interacts with these C-terminal fragments of FliA in an in vitro assay (Kutsukake et al., 1994).

The location and extent of this region may explain why we failed to isolate otherwise wild-type substitution mutants of TraR resistant to inhibition by TraM. The putative DNA-binding domain of the activator is located in this C-terminal region (Luo and Farrand, 1999a). Moreover, deletions as short as two residues from the C-terminus of TraR abolish transcriptional activation and DNA binding (Luo and Farrand, 1999a). These results suggest that much of the C-terminal domain of TraR is essential for its activity. We propose that residues of TraR important for interaction with TraM also are important for DNA binding and transcriptional activation. Thus, because our mutant search required that TraR maintain activity, it is likely that substitution mutants unable to interact with TraM simultaneously lost activator function, and therefore were excluded by the screen.

The two exceptions, the serine and leucine substitutions in TraR at proline 176, may define a region important to the structure of the activator. This residue is conserved in the products of all traR genes sequenced to date as well as a few other homologues, but not in LuxR or most other members of the LuxR family. Substituting serine or leucine at this position compromises but does not abolish transcriptional activation. Furthermore, as judged by the two-hybrid assay, TraM still interacts, albeit more weakly, with the P176 [RIGHTWARDS ARROW] S mutant. Thus, this proline is unlikely to directly interact with the antiactivator. The residue is located in the proposed hinge region between the acyl-HSL binding/multimerization domains and the DNA-binding domain of the LuxR-like transcriptional activators (Fuqua and Greenberg, 1998). We propose that substitutions in this region sterically alter the protein, thereby weakening the interaction with TraM. Such a structural alteration also may account for the diminished ability of the protein to activate transcription.

The observation that the traR3 allele contains a mutation in the 5′ non-coding region is consistent with our model that inhibition by TraM is dependent upon the relative amounts of the two proteins present in the cell. Clearly, the mutation, which produces a sequence more like the canonical −35 promoter element, results in increased expression of the wild-type traR gene. This, in turn, suggests that a weak native promoter lies directly upstream of traR. This interpretation is consistent with our proposal that in the context of the intact Ti plasmid a weak promoter located just upstream of the gene is responsible for the low level of expression of traR that occurs under conditions in which the arc operon is repressed by AccR (Piper et al., 1999).

Overall, these results are consistent with our model in which TraM serves to inactivate the low levels of TraR produced under non-inducing conditions. We propose that TraM accomplishes this by interacting directly with the C-terminal region of activator. In this regard, TraM resembles other antiactivators, including NifL of Klebsiella pnemoniae (Sidoti et al., 1993) and MecA of Bacillus subtilis (Turgay et al., 1998), as well as anti-sigma factors such as SpoIIAB of B. subtilis (Schmidt et al., 1990), MucA of Pseudomonas aeruginosa (Xie et al., 1996) and FlgM of Salmonella typhimurium (Ohnishi et al., 1992). Each of these inhibitors prevents transcriptional activation by directly interacting with its target protein. Furthermore, TraM is necessary for quorum-dependent regulation of Ti plasmid transfer; although tra gene induction in traM mutants still requires TraR and AAI (Hwang et al., 1995), transfer now occurs at very low population densities (K. R. Piper and S. K. Farrand, submitted for publication). Thus, the antiactivator plays an essential role in the regulatory coupling of conjugation to the cues that signal that environmental conditions are favourable for plasmid transfer.

Experimental procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Strains, media and chemicals

Agrobacterium tumefaciens NT1 (Watson et al., 1975), Escherichia coli DH5α (Sambrook et al., 1989), and DH10b (BRL) and Saccharomyces cerevisiae EGY48 (Gyuris et al., 1993) were used for all constructs. Luria broth (LB, BRL) was used as the rich liquid medium for both A. tumefaciens and E. coli strains. A. tumefaciens strains were grown on nutrient agar (NA, Difco), and in AB minimal medium (Chilton et al., 1974) supplemented with 0.2% mannitol as the sole carbon source. Yeast strains were grown in complete minimal (CM) dropout medium (Ausubel et al., 1989) supplemented with the appropriate amino acids and either 2% glucose (non-inducing conditions) or 2% galactose plus 1% raffinose (inducing conditions) as carbon source. When required, antibiotics were added to bacterial media at concentrations described previously (Farrand et al., 1996). E. coli strains were grown at 37°C, and A. tumefaciens and S. cerevisiae strains were grown at 28°C. Xgal (Sigma) was included in media at 40 μg ml−1. Synthetic N-(3-oxo-octanoyl)-l-HSL was a generous gift from David Lynn (University of Chicago).

DNA manipulations

Plasmids were isolated from E. coli and A. tumefaciens by the alkaline lysis method (Sambrook et al., 1989). Plasmid DNA was isolated from yeast cells as described by Hoffman and Winston (1987). Digestions with restriction endonucleases were carried out as described by the manufacturers of the enzymes. Recombinant DNA techniques and electrophoresis in agarose gels were performed as described by Sambrook et al. (1989). Plasmid DNA was introduced into E. coli by calcium chloride-mediated transformation (Sambrook et al., 1989), into yeast cells by lithium acetate-mediated transformation (Gietz et al., 1992), and into A. tumefaciens by electroporation (Cangelosi et al., 1991) or by biparental matings from E. coli S17-1 (Cook et al., 1997). Mutant genes were marker exchanged into pTiC58 and pTiC58ΔaccR as described by Ruvkun and Ausubel (1981), and the constructions were confirmed by restriction enzyme analysis of isolated Ti plasmid DNA.

Genetic constructions

The TraM coding sequence was amplified from pDCBP by polymerase chain reaction (PCR) using primers that introduced an in frame NdeI site at the 5′ end and a BamHI site at the 3′ end (the sequences of all primers used are available on request). The product obtained was digested with these enzymes and cloned into pET14-b (Novagene) to form pMA2. The resulting construct expresses TraM with a 6×His tag at the N-terminus.

A lacZ-Km cassette from pLKC482 (Tiedeman and Smith, 1988) was fused to traR3 at a NotI site located in the coding sequence of the gene in pTZR3. The resulting EcoRI fragment containing the traR3 ::lacZ–Km fusion was substituted for the wild-type EcoRI 33 fragment in pKP24 (Piper, 1999) to form the recombinant plasmid pH3RZ.

Construction of plasmids for analysis in the yeast two-hybrid system was carried out as follows. Wild type and mutant TraM coding sequences were amplified by PCR to give products with an in frame EcoRI site at the 5′ end and a BamHI site at the 3′ end. PCR products digested with the two enzymes were cloned into the yeast expression vector pEG202 (Gyuris et al., 1993), creating the pEGM series. These constructs express the wild-type and mutant TraM proteins translationally fused to the E. coli LexA protein. Constructs expressing C-terminal deletion mutants of TraM were made similarly, using forward primers containing in frame EcoRI sites and appropriate reverse primers. Wild-type traR and the traR7 mutant were amplified by PCR using primers that introduced EcoRI and XhoI sites at the 5′ and 3′ termini respectively. The PCR products were digested and cloned into the yeast expression vector pJG4.5 (Gyuris et al., 1993), creating pJGR and pJGR200. These clones express the TraR proteins fused to the chimeric polypeptide that contains the transcriptional activation domain of the E. coli B42 protein (Ma and Ptashne, 1987).

For expression in A. tumefaciens, the traM gene was recovered as an EcoRI to BamHI fragment from pEGM1 and cloned into pUC19 to form pUC::traM. The traM coding sequence was excised from pUC::traM by digestion with EcoRI and Sal I and cloned into pBAD22 (Guzman et al., 1995). The clone obtained was digested with ClaI and ScaI and the fragment containing the BAD promoter–traM fusion and the araC gene (Luo and Farrand, 1999b; Newman and Fuqua, 1999) was cloned into pBBR1MCS-3 (Kovach et al., 1995) digested with BstBI and SmaI. The resulting clone, pASM, expresses TraM under arabinose regulation in both E. coli and A. tumefaciens. Clones expressing the TraM mutants were prepared similarly.

Fragments of traR that were 5′ deleted were amplified by PCR using primers that introduced NcoI and HindIII sites at the 5′ and 3′ ends, respectively, and cloned into pBAD22 to generate in frame fusions with a translational start site provided by the vector. The clones obtained were co-integrated into pDSK519 (Keen et al., 1988) to allow expression in A. tumefaciens. The 5′ deletion mutants of traR were excised from pBAD22 by digestion with EcoRI and Sal I and cloned into EcoRI/XhoI-digested pJG4.5 for analysis in the yeast two-hybrid system.

Purification of TraM and preparation of polyclonal antibody

His-tagged TraM was induced by growing E. coli BL21(DE3)(pLysS)(pMA2) in the presence of 0.5 mM IPTG for 3 h. The cells were harvested, broken using a French pressure cell at 172 500 kPa, and cell debris was removed by centrifugation. The supernatant, which contained soluble 6×His-TraM, was loaded onto a nickel-sepharose 6B Fast Flow column (Novagene) and the column was developed according to the manufacturer's instructions. The 6×His-TraM protein obtained was greater than 95% pure as estimated by SDS–PAGE (Sambrook et al., 1989). Murine polyclonal antibodies against purified 6×His-TraM were produced by the Genetic Engineering facility of the University of Illinois.

Western blot analysis

Approximately equal amounts of total proteins were separated on 15% SDS polyacrylamide gels and were subjected to Western blot analysis (Sambrook et al., 1989) using either anti-TraM polyclonal antibody or anti-Lex A polyclonal antibody (a generous gift from Dr Erica Golemis, FCCC, Philadelphia, PA, USA). Reactive proteins were visualized using an enzyme-linked chromogenic method (Sambrook et al., 1989).

Isolation of traM and traR mutants

Plasmid DNA was mutagenized with hydroxylamine as described by Slock et al. (1990) and introduced into A. tumefaciens reporter strains as described in the text. Plasmids were isolated from potential mutants, subjected to restriction enzyme analysis and reintroduced into the respective test strains to confirm their mutant status. In the case of pDCBP mutants, the 1.1 kb BamHI/Pst I fragment containing traM was isolated and substituted into wild-type pDCBP to confirm that the mutation was in traM rather than in traR.

Nucleotide sequence analysis

Complete double-stranded nucleotide sequence was determined for all mutant alleles and gene fusions using automated sequencing and dye terminator chemistries by the University of Illinois Biotechnology Center.

β-Galactosidase assays

Production of β-galactosidase by A. tumefaciens strains and by yeast strains was quantified as described by Hwang et al. (1994) and by Ausubel et al. (1989) respectively.

  1. *Present address: Korea Research Institute of Bioscience and Biotechnology, Taejeon 305-600, South Korea

  2. †These authors contributed equally to this study


  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

We thank David Rivier and his group for an introduction into the mysteries of Saccharomyces cerevisiae, and Philippe Oger, Ping Gao and Yinping Qin for helpful discussions. This work was supported in part by grant No. R01 GM52465 from the NIH to S.K.F.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  • 1
    Ausubel, F.M., Brent, R., Kingston, R.E., Moore, D.D., Smith, J.A., Seidman, J.G., Struhl, K. (1989) Current Protocols in Molecular Biology. New York: John Wiley.
  • 2
    Beck von Bodman, S., Hayman, G.T., Farrand, S.K. (1992) Opine catabolism and conjugal transfer of the nopaline Ti plasmid pTIC58 are coordinately regulated by a single repressor. Proc Natl Acad Sci USA 89: 643647.
  • 3
    Cangelosi, G.A., Best, E.A., Martinetti, C., Nester, E.W. (1991) Genetic analysis of Agrobacterium tumefaciens. Methods Enzymol 145: 177181.
  • 4
    Chilton, M.-D., Currier, T.C., Farrand, S.K., Bendich, A.J., Gordon, M.P., Nester, E.W. (1974) Agrobacterium tumefaciens DNA and bacteriophage PS8 DNA not found in crown gall tumors. Proc Natl Acad Sci USA 71: 36723676.
  • 5
    Cook, D.M., Li, P.-L., Ruchaud, F., Padden, S., Farrand, S.K. (1997) Ti plasmid conjugation is independent of vir : reconstruction of the tra functions from pTiC58 as a binary system. J Bacteriol 179: 12911297.
  • 6
    Ellis, J.G., Kerr, A., Petit, A., Tempé, J. (1982) Conjugal transfer of nopaline and agropine Ti-plasmids — The role of agrocinopines. Mol Gen Genet 186: 269273.
  • 7
    Evans, K., Passador, L., Srikumar, R., Tsang, E., Nezezon, J., Poole, K. (1998) Influence of MexAB-OprM multidrug efflux system on quorum sensing in Pseudomonas aeruginosa. J Bacteriol 180: 54435447.
  • 8
    Farrand, S.K. (1998) Conjugation in the Rhizobiaceae. In The Rhizobiaceae Molecular Biology of Model Plant-Associated Bacteria. Spaink, H.P., Kondorosi, A., and Hooykaas, P.J.J. (eds). Dordrecht: Kluwer Academic Publishers, pp. 199233.
  • 9
    Farrand, S.K., Hwang, I., Cook, D.M. (1996) The tra region of the nopaline-type Ti plasmid is a chimera with elements related to the transfer systems of RSF1010, RP4 and F. J Bacteriol 196: 42334247.
  • 10
    Fuqua, W.C. & Greenberg, E.P. (1998) Self perception in bacteria: quorum sensing with acylated homoserine lactones. Curr Opin Microbiol 1: 183189.
  • 11
    Fuqua, W.C. & 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.
  • 12
    Fuqua, W.C., Burbea, M., 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.
  • 13
    Gietz, D., St Jean, A., Woods, R.A., Schiestl, R.H. (1992) Improved method for high-efficiency transformation of intact yeast cells. Nucleic Acids Res 20: 1425.
  • 14
    Guzman, L.M., Belin, D., Carson, M.J., Beckwith, J. (1995) Tight regulation, modulation, and high level expression by vectors containing the arabinose PBAD promoter. J Bacteriol 177: 41214130.
  • 15
    Gyuris, J., Golemis, E.A., Chertkov, H., Brent, R. (1993) Cdi1, a human G1- and S-phase protein phosphatase that associates with Cdk2. Cell 75: 791803.
  • 16
    Hoffman, C.S. & Winston, F. (1987) A ten minute DNA preparation from yeast efficiently releases autonomous plasmids for transformation of E. coli. Gene 57: 267272.
  • 17
    Hwang, I., Li, P.-L., Zhang, L., Piper, K.R., Cook, D.M., Tate, M.E., 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.
  • 18
    Hwang, I., Cook, D.M., Farrand, S.K. (1995) A new regulatory element modulates homoserine lactone-mediated autoinduction of Ti plasmid conjugal transfer. J Bacteriol 177: 449458.
  • 19
    Kaplan, H.B. & Greenberg, E.P. (1985) Diffusion of autoinducer is involved in regulation of the Vibrio fischeri luminescence system. J Bacteriol 163: 12101214.
  • 20
    Keen, N.T., Tamaki, S., Kobayashi, D., Trolinger, D. (1988) Improved broad-host-range plasmids for DNA cloning in Gram-negative bacteria. Gene 70: 191197.
  • 21
    Kovach, M.E., Elzer, P.H., Hill, D.S., Robertson, G.T., Farris, M.A., Roop, II, R.M., Peterson, K.M. (1995) Four new derivatives of the broad-host-range cloning vector pBBR1MCS, carrying different antibiotic genes. Gene 166: 175176.
  • 22
    Kutsukake, K., Iyoda, S., Ohnishi, K., Iino, T. (1994) Genetic and molecular analyses of the interaction between the flagellum-specific sigma and anti-sigma factors in Salmonella typhimurium. EMBO J 13: 45684576.
  • 23
    Luo, Z.-Q. & Farrand, S.K. (1999 a) Signal-dependent DNA binding and functional domains of the quorum-sensing activator TraR as identified by repressor activity. Proc Natl Acad Sci USA 96: 90099014.
  • 24
    Luo, Z.-Q. & Farrand, S.K. (1999 b) Cloning and characterization of a tetracycline resistance determinant present in Agrobacterium tumefaciens C58. J Bacteriol 181: 618626.
  • 25
    Ma, J. & Ptashne, M. (1987) A new class of yeast transcriptional activators. Cell 51: 113119.
  • 26
    Narberhaus, F., Lee, H.-S., Schmitz, R.A., He, L., Kustu, S. (1995) The C-terminal domain of NifL is sufficient to inhibit NifA activity. J Bacteriol 177: 50785087.
  • 27
    Newman, J.R. & Fuqua, C. (1999) Broad-host-range expression vectors that carry the L-arabinose-inducible Escherichia coli araBAD promoter and the araC regulator. Gene 227: 197203.
  • 28
    Ohnishi, K., Kutsukake, K., Suzuki, H., Iino, T. (1992) A novel transcriptional regulation mechanism in the flagellar regulon of Salmonella typhimurium : an anti-sigma factor inhibits the activity of the flagellum-specific sigma factor, σF. Mol Microbiol 6: 31493157.
  • 29
    Pearson, J.P., Van Delden, C., Iglewski, B.H. (1999) Active efflux and diffusion are involved in transport of Pseudomonas aeruginosa cell–cell signals. J Bacteriol 181: 12031210.
  • 30
    Piper, K.R. (1999) Molecular, genetic and physiological studies on the regulation of conjugal transfer of the Ti plasmid pTiC58 by Agrobacterium tumefaciens. PhD Dissertation. Illinois: University of Illinois at Urbana-Champaign.
  • 31
    Piper, K.R., Beck von Bodman, S., Farrand, S.K. (1993) Conjugation factor of Agrobacterium tumefaciens regulates Ti plasmid transfer by autoinduction. Nature 362: 448450.
  • 32
    Piper, K.R., Beck von Bodman, S., Hwang, I., Farrand, S.K. (1999). Hierarchical gene regulatory systems arising from fortuitous gene associations: regulating quorum sensing by the opine regulon in Agrobacterium. Mol Microbiol 32: 10771089.
  • 33
    Ruvkun, G. & Ausubel, F. (1981) A general method for site-directed mutagenesis in prokaryotes. Nature 289: 8588.
  • 34
    Sambrook, J., Fritsch, E.F., Maniatis, T. (1989) Molecular Cloning. A Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press.
  • 35
    Schmidt, R., Margolis, P., Duncan, L., Coppolecchia, R., Moran, C.P., Losick, R. (1990) Control of developmental transcription factor sigma F by sporulation regulatory proteins SpoIIAB in Bacillus subtilis. Proc Natl Acad Sci USA 87: 92219225.
  • 36
    Sidoti, C., Harwood, G., Ackerman, R., Coppard, J., Merrick, M. (1993) Characterization of mutations in the Klebsiella pneumoniae nitrogen fixation regulatory gene nifL, which impair oxygen regulation. Arch Microbiol 159: 276281.
  • 37
    Slock, J., VanRiet, D., Kolibachuk, D., Greenberg, E.P. (1990) Critical regions of the Vibrio fischeri LuxR protein defined by mutational analysis. J Bacteriol 172: 39743979.
  • 38
    Tiedeman, A.A. & Smith, J.M. (1988) lacZY gene fusion cassettes with KanR resistance. Nucleic Acids Res 16: 3587.
  • 39
    Turgay, K., Hahn, J., Burghoorn, J., Dubnau, D. (1998) Competence in Bacillus subtilis is controlled by regulated proteolysis of a transcription factor. EMBO J 17: 67306738.
  • 40
    Watson, B., Currier, T.C., Gordon, M.P., Chilton, M.-D., Nester, E.W. (1975) Plasmid required for virulence of Agrobacterium tumefaciens. J Bacteriol 123: 255264.
  • 41
    Xie, Z.-D., Hershberger, C.D., Shankar, S., Ye, R.W., Chakrabarty, A.M. (1996) Sigma factor–anti-sigma factor interaction in alginate synthesis: Inhibition of AlgT by MucA. J Bacteriol 178: 49904996.
  • 42
    Zhang, L. & Kerr, A. (1991) A diffusible compound can enhance conjugal transfer of the Ti plasmid in Agrobacterium tumefaciens. J Bacteriol 173: 18671872.
  • 43
    Zhang, L., Murphy, P.J., Kerr, A., Tate, M.E. (1993) Agrobacterium conjugation and gene regulation by N-acyl-L-homoserine lactones. Nature 362: 446448.