SEARCH

SEARCH BY CITATION

Keywords:

  • Activator;
  • Gene expression;
  • Heavy metal;
  • Metal induction;
  • DNA distortion

Abstract

  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Regulation by MerR
  5. 3Other MerR family members
  6. 4Structural studies
  7. 5Metal-dependent regulators
  8. 6Distribution and evolution of MerR-like regulators
  9. 7Conclusions
  10. Acknowledgements
  11. References

The MerR family is a group of transcriptional activators with similar N-terminal helix-turn-helix DNA binding regions and C-terminal effector binding regions that are specific to the effector recognised. The signature of the family is amino acid similarity in the first 100 amino acids, including a helix-turn-helix motif followed by a coiled-coil region. With increasing recognition of members of this class over the last decade, particularly with the advent of rapid bacterial genome sequencing, MerR-like regulators have been found in a wide range of bacterial genera, but not yet in archaea or eukaryotes. The few MerR-like regulators that have been studied experimentally have been shown to activate suboptimal σ70-dependent promoters, in which the spacing between the −35 and −10 elements recognised by the σ factor is greater than the optimal 17±1 bp. Activation of transcription is through protein-dependent DNA distortion. The majority of regulators in the family respond to environmental stimuli, such as oxidative stress, heavy metals or antibiotics. A subgroup of the family activates transcription in response to metal ions. This subgroup shows sequence similarity in the C-terminal effector binding region as well as in the N-terminal region, but it is not yet clear how metal discrimination occurs. This subgroup of MerR family regulators includes MerR itself and may have evolved to generate a variety of specific metal-responsive regulators by fine-tuning the sites of metal recognition.


1Introduction

  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Regulation by MerR
  5. 3Other MerR family members
  6. 4Structural studies
  7. 5Metal-dependent regulators
  8. 6Distribution and evolution of MerR-like regulators
  9. 7Conclusions
  10. Acknowledgements
  11. References

The archetype of the MerR family of transcriptional activators is the regulator of Gram-negative mercury resistance (mer) operons found on transposable elements Tn21 and Tn501[1–5]. This regulator was shown to be an activator of the mer genes in the presence of Hg(II) salts and a weak repressor in the absence of Hg(II) [6,7]. MerR also regulated its own synthesis. A remarkable and unprecedented observation at that time was that both activation and repression of the mer genes occurred with the regulator bound at the same site on DNA, lying between the −35 and −10 regions of the major promoter, PmerTPAD. Activation involved distortion of the DNA at the centre of the operator to realign the −35 and −10 sequences, which were separated by a longer spacer than normal [7–11].

Subsequent data obtained for the regulators TipAL from Streptomyces lividans, responsible for thiostrepton-dependent gene activation [12], and SoxR from Escherichia coli, responsible for part of the oxidative stress response [13,14], indicated that these were also members of the MerR family and that they activated gene expression in similar ways. The N-terminal similarity between these and other proteins led to the idea of a ‘MerR family’ of regulators which activate gene expression by distorting the operator DNA sequence and cause RNA polymerase to initiate transcription at an otherwise suboptimal promoter. Recently, significant data on how these systems work have been obtained from studies of Tn21 MerR [15,16] and the dye-responsive activators of drug efflux transporters in Bacillus subtilis, BltR and BmrR [17,18]. In particular, the crystal structures of BmrR–DNA complexed with an activator molecule [19] and of the N-terminal domain of the Mta regulator, also from Bacillus[20], indicate the essential structural elements of the family, and confirmed predictions made from MerR [15,16].

The initially recognisable characteristic of a member of the MerR family is the high degree of sequence similarity in the N-terminal DNA binding region. C-terminal similarity may be non-existent. In the last four years a subset of MerR family regulators has been identified which responds to metal ions (e.g. [21–29]). Some members of this subgroup show appreciable C-terminal sequence similarity, particularly in potential metal-coordinating residues, but respond to different metals.

This review puts together for the first time the characteristics of MerR family regulators and explores the characteristics of the metal-responsive members of the family. We also propose that there is a structural subset of this family that comprises simple metal-dependent regulators. At the time of writing no direct three-dimensional structural data are available for members of this subfamily.

Table 1 shows some of the better known members of the MerR family and gives references to example papers describing their roles and mechanism of action. The increasing availability of genome sequences reveals a plethora of regulators and a snapshot of the distribution of some of these is given in Section 6. The last three years or so have demonstrated that open reading frames with sequence similarity to MerR family regulators are common and are present in a variety of bacterial genera, although very few have been studied in any detail.

Table 1.  Examples of members of the MerR family
GeneOrganismInducerRegulated genesMolecular details knownExample references
merRvarious mobile elementsHg(II)mercury resistance (mer) genesactivation mechanism known, mutants identified[33,69]
soxRE. coli and othersoxidative stresssoxSactivation mechanism known[13,76,117]
tipAStreptomycesthiostreptonthiostrepton-responsive genesactivation mechanism known[12,78,89]
nolABradyrhizobiumgenistein (or another plant product)nodulation genesthree gene products of nolA, autoregulated by NolA1[113,118,119]
bmrRB. subtilisvarious dyesbmr multidrug resistancestructure determined and C-terminal inducer binding[17–19]
bltRB. subtilisvarious dyesblt multidrug resistanceC-terminal inducer binding[17,18]
mtaB. subtilisvarious dyesblt and bmr multidrug resistancepartial structure available[20,37]
cueRE. coliCu(I), Ag(I), Au(I)copA, cueO genesdata on mechanism and mutants[24,26,94]
zntRE. coliZn(II), Cd(II), Pb(II)zntA transportdata on mechanism and mutants[22,80,81]
pbrRR. metalliduransPb(II)pbrA transportsome data on mechanism[25]

The best studied members of the MerR family (Table 1) respond to a variety of co-effectors (in all cases studied so far, these are inducers) and the differences in their primary structures reflecting this are in the C-terminal regions. The N-terminal regions are very similar and contain predicted helix-turn-helix motifs. Some of the overall amino acid similarities with MerR are shown in Fig. 1, in which significant amino acid identity or conservative substitution can be seen to approximately amino acid 110. This was thought to define the DNA binding region. The best early evidence that there were separate N-terminal DNA binding and C-terminal inducer binding regions came from mutagenesis of the Tn21 and Tn501 MerR proteins [30–32] in which a discrete N-terminal DNA binding region was identified, separate from the more C-terminal Hg(II) binding cysteines [33]. A clear indication of the fact that the DNA binding and inducer binding regions were separate domains came from the TipAL regulator [12], the longer of two products of the tipA gene of S. lividans (Section 3.2). Later studies of MerR [34,35], BltR, BmrR [17,19,36] and Mta [20,37] also provided direct evidence for separate DNA binding and inducer binding domains (Sections 2 and 3) from mutagenesis and structural experiments. Attempts to obtain direct structural information from MerR have so far proved elusive (unpublished data from five separate laboratories) and most information on structure has been obtained indirectly. However, comparisons with the structures of BmrR and MtaN [19,20] (Section 4) suggest that the regions of amino acid similarity between the MerR family members extend through the DNA binding region and an antiparallel coiled-coil dimerisation region and diverge only in the inducer binding region, where there are no protein–DNA or protein–subunit interactions common to the family as a whole.

image

Figure 1. Diagram of the alignment of MerR family regulators with MerR from Tn501, showing the overall similarity extends for ca. 100 amino acids from the N-terminus (see text for references to proteins).

Download figure to PowerPoint

2Regulation by MerR

  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Regulation by MerR
  5. 3Other MerR family members
  6. 4Structural studies
  7. 5Metal-dependent regulators
  8. 6Distribution and evolution of MerR-like regulators
  9. 7Conclusions
  10. Acknowledgements
  11. References

The merR genes of transposons Tn501 from Pseudomonas aeruginosa[38] and Tn21 from the Shigella flexneri R100 plasmid [4] were the first studied members of this gene family. They have been treated interchangeably for both genetic analysis and biochemical analysis of the gene products. This is justified as early data showed that they, and a large number of other merR genes from Gram-negative bacteria, could complement a Tn21 merR mutant [39]. Sequence analysis of different MerR proteins and the promoters on which they act, suggest that information from the Gram-negative systems and the Gram-positive MerR activators applies to each of the other systems [40]. An exception is the MerR repressor of S. lividans which is a member of the SmtB/ArsR regulator family [41,42]. The MerR regulator was reviewed in detail in 1992 [33], at which time many of the basics of mercury-dependent regulation had been established, but the detailed mechanism was far from clear. These early data will be summarised briefly.

MerR regulates expression of the mercury resistance (mer) operon and, in Gram-negative systems, is divergently transcribed from the major regulated promoter (Fig. 2). The major promoter transcribes the genes merTP(C/F)AD(E) which encode the transporter MerT, a periplasmic protein MerP, additional transporters MerC (in Tn21) [43] or MerF (in pMER327/419 and Tn5053) [44,45], the mercuric reductase enzyme MerA, a probable repressor MerD [46] and a possible further transporter MerE [40,47]. Further details of the mercury resistance systems are described elsewhere in this volume [48].

image

Figure 2. Structure of the promoter/operator region of the mer operon of Tn501. The −35 and −10 sequences of the promoters PmerTPAD and PmerR are boxed and labelled against the ‘sense’ strands of the appropriate promoters. The MerR binding sequence is boxed according to DNA footprinting data and the dyad symmetrical sequence is indicated by arrows.

Download figure to PowerPoint

2.1The mer promoter and its mutants

Some of the first studies on the mer regulatory system were to map and identify the transcripts [6] and to select and identify up-promoter and down-promoter mutants. The detail of the mer operator/promoter region is shown in Fig. 2. The unusually high spacing of 19 bp between the −35 and −10 sequences suggested that the promoter structure was important to the regulatory mechanism. Up-promoter mutants of the mer promoter were small deletions in this 19 bp spacer [11]. Deletion mutants showed that the −35 and −10 sequences correctly separated by 19 bp were absolutely required for normal promoter activation by MerR [7,49]. Spacer deletion mutations outside the MerR binding sequence increased constitutive promoter activity, which was repressed by MerR even in the presence of Hg(II) salts; insertions of 1 or 2 bp in the spacer removed promoter activity. Mutation of the dyad symmetrical region in the spacer severely disrupted MerR binding and, although the effect of such mutations on constitutive promoter activity was negligible, the effect on MerR-dependent regulation was profound [34,50].

2.2The MerR protein

The MerR proteins of Tn501 and Tn21 are 144 amino acids long and differ in nine residues, three of which are conservative substitutions. The regulators can be considered identical for most purposes. Alignment of MerR regulators from a wide variety of Gram-negative bacterial genera shows that the proteins are conserved in about 90% of these 144 amino acids, but there is variation in the very C-terminal region and considerable difference with MerR proteins from Gram-positive sources (Fig. 3). However, three cysteine residues are conserved in all MerR proteins. These were originally suggested as the site of Hg(II) binding, and this was confirmed by a variety of methods including mutagenesis of the cysteines to alanine or serine in which the Hg(II)-dependent activation is lost [32,51] and spectroscopy showing S–Hg(II) bonding [52–54]. In proteins from Gram-negative bacteria, the very C-terminal region is responsible for determining whether the protein responds to organomercurials, as removal of this region from the organomercurial-responsive MerR of plasmid pDU1358 abolished the organomercurial response, but the protein could still respond to inorganic mercuric salts [55]. Other genetic and biochemical studies have helped elucidate how the MerR protein functions and these will be described below.

image

Figure 3. Alignment of MerR proteins from a variety of Gram-negative and Gram-positive sources showing the consensus sequences. The Hg(II) binding cysteines are marked with arrows. The Serratia sequence (pDU1358) is from a ‘broad spectrum’ MerR, responding to organomercurials as well as inorganic Hg(II). Note that the MerR genes from Gram-positive sources (Bacillus, Staphylococcus) differ from those from Gram-negative sources.

Download figure to PowerPoint

The purification of the MerR protein [56] and its footprinting on the mer regulatory region in vitro [10], together with the equivalent in vivo analysis of binding [34,35] have shown that MerR and RNA polymerase are bound simultaneously to the mer promoter. O'Halloran [10] has argued that there is little direct contact between them, as the footprints of MerR and RNA polymerase are additive to give the footprint of the ternary complex. No interaction was identified with the α subunit of RNA polymerase where many transcriptional regulators interact [57], but Summers’ group showed allele-specific effects of α and σ subunit mutations on MerR-dependent modulation of transcription [58] and were able to cross-link the MerR to the β subunit as well as α and σ[59]. This occurred in the absence of operator DNA, but interaction was increased by its presence, and indicates a close association between MerR and RNA polymerase. The main effect of RNA polymerase mutations thus far detected has been in suppression of the effect of suboptimal spacer length by mutations in the σ subunit [58]. These data are compatible with lack of direct and specific interaction between MerR and RNA polymerase, but with their being in very close proximity and interacting primarily through binding to the same region of DNA.

Cross-linking studies [59] showed that MerR altered the interactions between RNA polymerase subunits, suggesting that there was a direct interaction between MerR and RNA polymerase which distorted the normal structure of the polymerase. Addition of Hg(II) did not alter the number of MerR–RNA polymerase cross-links, but did increase self-cross-linking of MerR, providing direct evidence that a conformational change does occur in MerR on binding Hg(II). This increase in MerR–MerR cross-links was not dependent on DNA binding, but was increased in the presence of RNA polymerase [59].

2.3MerR mutants

Earlier work [30] had shown that mutants of the merR gene of Tn21 could be isolated with phenotypes representative of loss of either activation or repression or both, and that mutation of three of the four cysteines caused loss of Hg(II)-inducible activation. Mutagenesis also showed that enhanced activation and enhanced repression mutants can be identified and isolated in which activation occurs in the absence of Hg(II) or in which repression occurs even though Hg(II) is present [31,35]. Mutants which affected DNA binding (E22K and R25H) helped define the DNA binding region [31,60], and indicated that the N-terminal helix-turn-helix motif, rather than a similar motif more centrally in the protein, was responsible for DNA binding [30]. With the exception of mutations in the helix-turn-helix motif, or those at or very close to the mercury binding cysteine residues, MerR mutations have weak (or leaky) phenotypic effects [30]. When mutations are combined strong phenotypes are seen, such as the strong MerR constitutive activator mutations, in which promoter activation occurs even in the absence of Hg(II) salts [60,61].

Caguiat et al. [16] investigated the metal response of MerR by seeking mutants which responded to Cd(II). Eleven mutants were obtained which had an elevated response to Cd(II). In each case the constitutive activation of the promoter (i.e. without metal) was elevated, and an Hg(II) response was maintained. Five of the mutants were within the dimer interface region, and none affected the essential cysteines, Cys82, Cys117 and Cys126. Although the responses to Zn(II) of these mutants were weak, they were greater than that of MerR. These data suggest that the mutations cause reduction in metal specificity of MerR rather than a change in specificity per se. Certainly, two of the mutations (S131L and A89V) had previously been identified as being responsible for enhanced activation phenotypes [30,61]. It has been possible to relax the metal specificity of MerR ([16] and unpublished data), but complete alteration in the metal specificity of MerR has only been achieved by replacing the C-terminal region with the equivalent region from a zinc-dependent regulator [22].

Mutations in which an activator and repressor phenotype could be genetically separated (e.g. A89V–S131L for activation in the absence of Hg(II) [61] and L128Q–Q61R for repression with Hg(II) binding (S.P. Kidd, B. Lawley and N.L. Brown, unpublished)) favour a model for MerR action in which the protein adopts a repressing conformation in the absence of the metal and an activating conformation in its presence. Both activation and repression are exerted when bound to a single operator sequence on DNA [10,50]. Mutational studies and the comparison of different regulators (Section 4) indicate that the response of MerR to Hg(II) is highly specific and relies not simply on the presence of metal binding amino acids, but on their correct positioning and orientation within the protein. This statement may be trite, but there is often an assumption that the mere presence of metal-coordinating amino acids in approximately the correct positions will enable a metal response.

2.4Physical studies of MerR

In spite of considerable efforts, it has not yet been possible to crystallise the MerR protein, and physical studies have given only partial structural information. Current attempts to crystallise subdomains of MerR and increasingly powerful NMR machines may yield direct structural information on the protein. Structures from other MerR family members (Section 4; [19,20]) help predict the structure of MerR. However, considerable information has been revealed about the structure of MerR from other biophysical and chemical studies.

Spectroscopic techniques have been used to examine the environment in which Hg(II) is bound on MerR. Extended X-ray absorption fine structure (EXAFS) spectroscopy showed that Hg–MerR has a three-coordinate Hg(S-cys)3 environment, with an average Hg–S distance of 2.43 Å[53]. Hg-199 NMR studies confirmed that the Hg(II) was complexed in a planar trigonal complex containing cysteine residues [62,63]. Extensive description of the spectroscopy of Hg–protein complexes and model biomimetic compounds is beyond the scope of this review, but the reader is referred to reviews from the O'Halloran laboratory [52,62].

Partial proteolysis of the MerR protein revealed that there was a trypsin-sensitive site at position 44 in the sequence [22], separating the amino-terminal helix-turn-helix region from that predicted to be a coiled-coil motif. Summers’ group [15] created MerR deletion mutants and demonstrated that only residues 80–128 were required for stable dimer formation and retained a high affinity for Hg(II). They showed by circular dichroism that MerR had significant α-helical content which was retained in the deletion derivatives. X-ray absorption spectroscopy showed that the wild-type and the maximally deleted protein bound Hg(II) into similar chemical environments, suggesting that, in addition to containing the metal binding cysteines at positions 82, 117 and 126, residues 80–128 contained sufficient structural information to form the Hg(II) binding site. They speculated on the basis of model three-helix bundle studies [64,65] that there are two Hg(II) binding sites on MerR, which are symmetrical and equivalent in the absence of bound Hg(II). The binding of a single Hg(II) ion to one site causes an allosteric change that renders the other site less able to bind Hg(II) in competition with other thiols. This agrees with data showing that a MerR dimer binds only one ion of Hg(II) [51].

2.5Model for MerR-dependent regulation – a hypersensitive switch

The mer promoter responds across a very narrow range of mercuric ion concentration. The promoter responds from 10 to 90% of fully induced activity across a seven-fold increase in Hg(II) concentration at ca. 10−8 M Hg(II) in the presence of thiols [66,67]. The mer promoter therefore responds not only to a very low concentration of mercuric salts, but the response is also hypersensitive, with a Hill coefficient of ca. 2.6 [66,68]. Studies on the Tn501 mer promoter in vivo show that it is fully induced across a narrow range of Hg(II) concentration below the threshold at which Hg(II) adversely affects the growth rate of bacterial cells [68]. The details of this induction have been elucidated in a series of elegant experiments by Ansari et al. [69,70]. The current model for action of MerR at the merOP region is shown in Fig. 4.

image

Figure 4. The model for MerR action showing the distortion of the merOP region. A: MerR bound to DNA and RNA polymerase recruited but not able to form an open complex. B: The conformational change induced by Hg(II) binding to the ternary complex and initiation of transcription. C: The hypothetical repression of the promoter by displacement of Hg–MerR2 with MerD at low Hg(II) (or non-metallated MerR could bind, giving the structure shown in A). MerR is shown in green and MerD in orange; the −10 and −35 sequences are indicated by red boxes. The separate RNA polymerase holoenzymes are: α=pink, β,β′=blue, σ=green.

Download figure to PowerPoint

  • 1
    In the absence of Hg(II) and MerR, RNA polymerase preferentially transcribes from the merR promoter, increasing the amount of MerR present in the cell.
  • 2
    Once MerR binds to merOP, transcription of the merR promoter is repressed and the DNA becomes bent and unwound at the operator sequence. RNA polymerase is recruited to the mer promoter, forming a ternary complex of DNA, MerR and RNA polymerase.
  • 3
    In the absence of Hg(II), the MerR protein is bound to DNA in the repressor conformation (R) maintaining repression of the promoter.
  • 4
    Binding of Hg(II) to one of two binding sites on the MerR causes a conformational change to put MerR in the activating conformation (A). Due to the tight binding of MerR to the operator, this causes DNA distortion at the centre of the operator, giving a ca. 33° unwinding of DNA and straightening of helix backbone.
  • 5
    The reorientation of the −35 and −10 sequences so caused, allows them to interact productively with the RNA polymerase σ70 subunit to form an open transcriptional complex and transcription is initiated.

This model explains the known properties of the mer regulatory system if one assumes that the interaction of Hg(II) with the MerR–RNA polymerase–DNA ternary complex is not diffusion limited. The promoter can then act with pseudo-zero-order kinetics as outlined by Koshland [71–73]. The properties of known MerR [15,30,31,51,60,61,74] and promoter [11,49,50] mutants are explained by this model, and indeed several were isolated to test it. The properties of RNA polymerase mutants and cross-linking [57,58] can be explained by the proximity of MerR and RNA polymerase on the DNA, without having to invoke specific MerR–RNA polymerase activating contacts.

Other members of the MerR family are assumed to act by a very similar mechanism, and this has been demonstrated in part for SoxR [75,76], TipAL[12,77,78], CueR [24,26,79] and ZntR [22,80,81]. Where identified, promoters regulated by MerR family members have extended spacing between the −35 and −10 recognition elements. In each case tested, the regulators are activators and repression is limited or non-existent. The repressor phenotype of MerR is probably a consequence of the relatively high constitutive activity of the promoter from the −10 sequence, the −35 sequence not being required for constitutive activity [7]. In the presence of MerR, the −35 sequence may become more important for binding, and subsequent open complex formation at the −10 sequence only occurs following the MerR-dependent change in the conformation of the promoter.

An important, and frequently ignored, aspect of gene regulation is how expression is switched off once the stimulus is removed. In the case of MerR, it is unlikely that Hg(II) will be readily released by MerR and the protein returned to the repressor conformation, as the stability constant of the Hg(II)–sulfhydryl adduct is extremely high. It is more likely that the Hg(II)–MerR2 complex will dissociate from the DNA and be replaced by non-metallated MerR2. MerR2 has a higher affinity for DNA than Hg(II)–MerR2[10,61], and may simply displace the activating form; alternatively, MerD may displace the Hg(II)–MerR2 complex. MerD is a product of the mer operon [82] and has been shown to repress the merTPAD promoter [83], with DNA footprinting studies indicating that MerD and MerR bind to the same operator site [46]. There is sequence similarity in the N-terminal domains of MerR and MerD but the relative affinities of MerD and metallated MerR for the merOP region have not been compared.

3Other MerR family members

  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Regulation by MerR
  5. 3Other MerR family members
  6. 4Structural studies
  7. 5Metal-dependent regulators
  8. 6Distribution and evolution of MerR-like regulators
  9. 7Conclusions
  10. Acknowledgements
  11. References

3.1SoxR

The SoxR regulator responds to oxidative stress and is part of the SoxR–SoxS regulon [84,85]. SoxR homologues are found in a number of Gram-negative and Gram-positive genera, but has been best studied in E. coli. SoxR acts at the soxS promoter to induce expression of the regulator SoxS, which in turn induces expression of approximately 12 genes in the SoxRS regulon. The soxS promoter has a suboptimal spacing and experiments similar to those previously performed with MerR [49] have shown that the 19 bp spacing is essential for normal SoxR-dependent activation [75].

The active form of SoxR is a dimer containing a [2Fe–2S] cluster in each 154 amino acid subunit. Superoxide (O2) or nitric oxide is sensed by oxidation of the cluster to the Fe3+–Fe3+ form. This oxidised form of SoxR causes transcriptional activation of the soxS promoter, and this is associated with DNA distortion in the centre of a dyad symmetrical sequence in the spacer region [86,87]. The mechanism of activation is therefore very similar to that of MerR. SoxR (apo- or Fe-form) autoregulates its own synthesis and represses soxS; oxidised Fe–SoxR activates transcription, repression and activation being through binding at a single site [76].

The binding of Fe by apo-SoxR is to cysteine residues in the motif Cx2CxCx5C to form the [2Fe–2S] cluster. Although this cysteine motif is unique, the C-terminal region of the protein, including the approximate location of the cysteines is similar to that of MerR. Mutation of the cysteine residues to alanine prevents formation of the [2Fe–2S] clusters. The mutant proteins still bind to soxS promoter DNA, but can no longer activate the promoter [88].

3.2TipAL

The tipA gene of S. lividans and other streptomycetes, is autoregulated by its minor product, the TipAL protein, and produces a more abundant product, TipAS[12]. TipAS is a 144 amino acid thiostrepton binding protein [89], probably produced from an internal translation start within the tipA transcript. TipAL is 254 amino acids and contains two functionally separate regions – an N-terminal DNA binding region showing sequence similarity to MerR, and a C-terminal region equivalent to TipAS. The tipA promoter has a 19 bp spacer between the presumed −35 and −10 regions containing a dyad symmetrical sequence to which TipAL binds and activation required thiostrepton [12]. The activation by thiostrepton (and other related compounds) involves irreversible modification of the TipAL protein [78]. Again, the mechanism of activation is similar to that described for MerR and there is direct evidence for different functions of the N- and C-terminal regions of the protein.

Recent data have shown a conformational change of TipAL on thiostrepton binding, and enhancement of RNA polymerase binding to the tipA promoter by thiostrepton–TipAL[77]. However, there are some differences in detail to the MerR–mer promoter interactions as deletion mutants of the tipA promoter do not show constitutive activity and the TipAL–thiostrepton adduct binds more tightly to the tipA promoter than does the unmodified protein, in contrast to MerR–Hg(II) which has a lower affinity for the mer promoter than MerR alone.

3.3Bacillus multidrug regulators

B. subtilis contains at least three regulators of the MerR family – BltR, BmrR and Mta – that are involved in activation of drug transport proteins. At present these are the only members of the MerR family for which crystal structures are available (Section 4); as such, they are important to understanding how this family of regulators works.

BltR and BmrR respectively regulate expression of the multidrug transporters Blt and Bmr, and have closely related N-terminal DNA binding domains, but their C-terminal sequences differ and they respond to different inducers [17]. The N-terminal domain of Mta (MtaN) also activates expression of both transporters in vitro, and Mta is presumed to cause expression of both in vivo as well as autoregulating its own synthesis [37]. BmrR binds to a dyad symmetrical sequence in the 19 bp spacer region of the bmr promoter and affinity of BmrR for this binding site is increased by addition of the co-activators, rhodamine or tetraphenylphosphonium (TPP). The C-terminal domain of BmrR expressed individually will bind rhodamine and TPP [36] and closely related structural compounds will also induce bmr expression [90].

3.4MerR-like regulators in E. coli

A homologue (YhdM) more closely related to MerR than SoxR and TipAL was first identified as the open reading frame of 141 amino acids at 74 min on the E. coli genome sequence [91]. Comparison of YhdM and MerR sequences showed strong identity between the amino acid sequences at the putative helix-turn-helix motif, and conservation of the cysteine residues involved in mercury binding, with an overall amino acid similarity of 52%, compared to 41–42% similarity between SoxR or TipAL and MerR. Subsequently the full genome sequence of E. coli K12 has revealed that it contains five MerR homologues: YhdM, YbbI, SoxR, YcgE and YehV. Of these five, SoxR has been extensively studied and is described above. The two sequences most closely related to MerR at the amino acid level are now known to be metal-dependent regulators: YhdM (renamed ZntR [22,80]) and YbbI (renamed CueR [24,26]). ZntR and CueR are discussed in more detail in Section 5. YehV (now MlrA [92]) is the regulator of aggregative fimbriae and extracellular matrix synthesis, and YcgE is of unknown function, but is closely related in amino acid sequence to MlrA. For neither MlrA nor YcgE is the nature of the inducing signal known.

4Structural studies

  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Regulation by MerR
  5. 3Other MerR family members
  6. 4Structural studies
  7. 5Metal-dependent regulators
  8. 6Distribution and evolution of MerR-like regulators
  9. 7Conclusions
  10. Acknowledgements
  11. References

4.1The activated BmrR–DNA structure

The crystal structure of the BmrR–DNA–TPP complex, using a 22 bp synthetic operator, has been solved at 3.0 Å resolution [19] (Fig. 5). This was the first published structure of a member of the MerR family and gave new insights into the mechanism of DNA binding and activation in the MerR family. The structure reveals a mechanism of activation involving localised base pair breaking, base sliding and realignment of the −35 and −10 operator elements.

image

Figure 5. A: Structure of the activated form of the BmrR dimer bound to DNA. B: Amino-terminal region of a BmrR monomer showing the helix-turn-helix motif and the a5 helix in the activated form. Reprinted by permission from Nature ([19] copyright (2001), Macmillan Publishers Ltd.).

Download figure to PowerPoint

The BmrR monomer contains three domains: (1) the N-terminal DNA binding domain (residues 1–75); (2) the linker and an 11 turn α-helix (residues 76–119) connecting the N- and C-terminal domains; and (3) the C-terminal drug binding domain (residues 120–278). The topology of the DNA binding domain is β1-α1-α2-β2-β3-α3-α4, containing four helix bundles and a three-stranded antiparallel β-sheet. BmrR uses a helix-turn-helix motif and two ‘wings’ to bind to the bmr promoter (Fig. 5B). The recognition helices (residues 19–28) contact two consecutive major grooves and each helical axis is almost perpendicular to the local DNA helical axis. No base-specific contacts can be identified, but this may be due to the resolution.

Most helix-turn-helix contacts are made by residues from helix α2 (Fig. 5B). Hydrogen bonds occur between the amide of Ala21 and guanine 4 phosphate, the NH2 and Nε of Arg23 and the cytosine 8′ phosphate and guanine 8′ phosphate, and the Tyr25 hydroxyl group and guanine 3 phosphate. Van der Waals contacts occur between the NH2 of Arg23 and the deoxyribose ring of cytosine 9′ and C7 atom of thymine 7′; the NH1 of Arg23 and C7 atom of thymine 7′; and the Tyr24 phenyl ring and C8 atom of adenine 2 and deoxyribose rings of guanine 3 and adenine 2.

The second DNA binding element, wing W1, includes strands β2 and β3 and the connecting loop (residues 35–46), of which Ser41, Tyr42, Arg43 and Asp26 make hydrogen bonds and van der Waals interactions with cytosine 8′, cytosine 9′, cytosine 10′ and thymine 7′. The third DNA binding element of BmrR, wing W2, is composed of helices α3 and α4 and their connecting turn. These helices are less crossed than the major groove binding helix-turn-helix motif. The protein–DNA contacts are made with Nε of Lys60 contacting guanine 3 phosphate and NH of Leu66 and the adenine 2 phosphate.

The structure of BmrR–TPP–DNA shows bending of the promoter by ∼50° at its middle towards the major groove and away from the protein. Unique base pair breaking and sliding of the central base pair is shown. The A–T base pairs that surround the pseudo-dyad of the promoter break and the unpaired adenine and thymine slide from each other in the 3′ direction. The operator ‘bunches-up’ in the middle. This distorted DNA conformation is stabilised by interactions of the phosphate backbone with residues Tyr24, Tyr25 and Lys60 and the N-terminus of helix α4. Similar structures are predicted for all activated MerR family regulators because each has a hydrogen bond donor (Lys, Arg, Gln) at position 60 and Tyr25 is conserved (Fig. 1).

4.2MtaN

MtaN is a 109 residue truncation mutant which contains the DNA binding and dimerisation regions of the protein Mta. It is a global activator of the bmr, blt and ydfK genes, as well as of its own gene. The crystal structure of the apo-protein has been determined at 2.75 Å resolution [20]. The structure is a winged helix-turn-helix, with a protruding eight-turn helix (α5) which forms an antiparallel coiled-coil in the dimer. The major difference between the structure of the MtaN apo-protein and the BmrR–TPP–DNA complex is the orientation of α5 relative to the α1-α4 DNA binding region. The differences include a reduction in spacing of the α2 recognition helices from 33.3 Å (close to the 34 Å repeat of B-form DNA) in MtaN to 30.6 Å in activated BmrR. Rotation between the subunits also occurs due to conformational changes in the coiled-coil region. Although MtaN is a constitutive activator, docking simulations showed that it could not bind the activated bmr operator structure without further conformational changes in MtaN [20]. These may occur on DNA binding. Further structural studies will be required to identify the structural changes occurring during activation by MerR family regulators.

4.3Simulation of the MerR structure

Alignment of the BmrR N-terminal sequence with those of MerR and related proteins has allowed the sites of significant residues in MerR and related proteins to be mapped (e.g. [81]). Such simulations are probably accurate for the DNA binding domain and the antiparallel coiled-coil region (which includes the first of the metal binding cysteines), but are less convincing for the more C-terminal metal binding residues. Direct experimental verification of the structure will be required in order to fully understand the processes of metal recognition and transcriptional activation.

5Metal-dependent regulators

  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Regulation by MerR
  5. 3Other MerR family members
  6. 4Structural studies
  7. 5Metal-dependent regulators
  8. 6Distribution and evolution of MerR-like regulators
  9. 7Conclusions
  10. Acknowledgements
  11. References

Although a number of MerR family proteins have been studied that respond to environmental stimuli other than metals (Section 3) their similarity to MerR has been primarily in the N-terminal DNA binding region, rather than in the C-terminal ‘sensing’ region of the protein.

In the last few years several different metal-dependent regulators belonging to the MerR family have been characterised. The explosion of prokaryotic genome sequence data has revealed many potential metal-dependent MerR-like open reading frames in bacterial genomes (Section 6). At first sight, the identity levels of these regulators may appear to be quite low (see Fig. 1), but to put these identity levels in context, the amino acid identities between MerR proteins from Gram-negative and Gram-positive sources are 37%, [91], yet they are functionally homologous, and to some extent interchangeable [93].

However, cysteine residues are essential in Hg(II) binding and activation in MerR [32,51] and have also been found to be essential for recognition of the cognate inducers in two MerR homologues, SoxR [88] and TipAL[89]. More critically, cysteine residues have been shown to be involved in the recognition of metal by ZntR [81] and CueR [94].

5.1Zinc

5.1.1ZntR

ZntR (originally YhdM) [22,80] is the regulator of the ZntA zinc/cadmium/lead ATPase [95]. It was the first E. coli metal-dependent MerR-like regulator to be described, and its function was initially determined using ZntR expressed in trans on a plasmid, regulating a plasmid-borne PzntA–lux transcriptional fusion [22]. ZntR in this assay system has been reported to be induced mainly by Zn(II). Mini Mu LacZ insertions into zntA show that under conditions in which cells are continually induced throughout growth with metal ions, Cd(II) causes maximum activation of PzntA, with Pb(II) and Zn(II) also activating [96].

The zntR gene is physically separate on the E. coli genome from PzntA, which has a 20 bp DNA sequence between the −35 and −10 promoter motifs. Overexpressed and purified ZntR was shown to bind PzntA DNA and protect the promoter from DNase I [22,80]. ZntR is a 141 amino acid protein and is 34% identical (52% similar) to Tn501 MerR [91]. It exists as a dimer and also displays a MerR-like DNA distortion mechanism for transcriptional activation in the presence of Zn(II) [80]. ZntR contains five cysteine residues, three of which are in identical positions to the metal binding Cys82, Cys117 and Cys126 in MerR. Recent mutagenesis studies on ZntR have shown that Cys79, Cys114, and Cys124, the residues equivalent to those in MerR, are required for activation by Zn(II), Pb(II) and Cd(II) [81]. Cys115 is also required for Zn(II) activation. The C-terminal Cys141 may control access of the metals to the binding site(s) on the protein, as mutagenesis has markedly different effects depending on the nature of the mutation and of the metal. His29, His53, His119 are required for a response to Zn(II), but not for Pb(II) or Cd(II). ZntR can bind two zinc ions per dimer [80], and has a very high affinity for Zn(II), which upon binding to ZntR causes a change in protein conformation [97]. The coordination of these metals by ZntR and the mechanism of metal recognition are not yet understood.

A hybrid protein was constructed from the N-terminal 44 amino acids of Tn501 MerR and the C-terminal 103 amino acids of ZntR, and was shown to act on a promoter containing the dyad symmetrical DNA region from PmerTPAD within a 20 bp spacer separating the −35 and −10 sequences. The MerR/ZntR hybrid protein responded to Zn(II), but not Hg(II) at this modified promoter [22]. These data demonstrated that domain swaps between MerR-like regulators can produce functional proteins with altered promoter recognition. This further strengthens the case for believing that there is a common mechanism amongst this subfamily of proteins for ‘sensing’ a metal and for transducing the ‘signal’ into expression from their cognate promoters. Activation of PzntA by Zn(II) and ZntR also occurs with a high Hill coefficient (3.2 [22]) indicating that this is also a hypersensitive biological switch.

5.1.2PmtR

The 135 amino acid protein, PmtR (Proteus mirabilis transcription regulator) was the first MerR-like metal-responsive regulator to which a function was ascribed [21]. This was done through heterologous expression in E. coli, where plasmids carrying the pmtR gene specifically increased Zn(II) tolerance of the cells, which also accumulated a 12 kDa protein in direct proportion to the cells’ ability to grow on increasing levels of zinc. This protein was identified as ZraP, a putative periplasmic zinc binding protein, whose expression was induced in response to Zn2+, but not Ni2+, Co2+, Cd2+, Mn2+ or Fe2+[21]. It now seems unlikely that PMTR regulated zraP expression directly [98]. The promoter activated by PMTR is not known, but one can speculate that it may have been the zntA promoter.

5.2Copper

5.2.1CueR from E. coli

The function of CueR from E. coli (formerly YbbI) was independently identified by three groups [24,26,99]. CueR is the regulator of the copper/silver ATPase, CopA, and the multicopper oxidase, CueO [24,100]. E. coli CueR is 37% identical to ZntR, 28% identical to Tn501 MerR, and contains four cysteine residues, two of which are in the corresponding positions to the Cys117 and Cys126 residues of MerR involved in binding mercury. There is no cysteine at a position corresponding to Cys82. Overexpressed and purified CueR binds to PcopA in gel retardation experiments and the affinity of CueR to the DNA decreases upon addition of Cu(I) [26]. CueR induces expression of PcopA on a low copy number lacZ fusion plasmid in trans in response to both Cu(II) and Ag(I) ions added to the external medium [26], although it is likely that the Cu(I) ion is the activator in vivo [24,26]. In this assay system PcopA responds in a hypersensitive manner to Ag(I) at much lower concentrations of Ag(I) than it does to Cu(II). The Hill coefficient for the response to Ag(I) is estimated at 3.4 [26]. The response of the promoter to Cu(II) is linear rather than hypersensitive, which may be a feature of the rate of conversion of Cu(II) to Cu(I) in vivo, rather than the response of CueR to copper [26]. Recently it has been shown that CueR will activate PcopA in response to gold salts [94], suggesting the effective nuclear charge (common between Cu(I), Ag(I) and Au(I)) is important in metal recognition. We have also shown that Cys112 and Cys120 (equivalent to Cys117 and Cys126 of MerR) are important in the recognition of metals by CueR [94].

5.2.2CueR from Pseudomonas putida

This is predicted to be a 137 amino acid protein that is activated by Cu(II), and is 43% identical to CueR from E. coli. The P. putida cueR gene is part of a two-gene operon, cueAR, where cueA encodes a P-type copper efflux ATPase 36% identical to CopA from E. coli[101]. In the cueAR operon the putative translational termination signal of cueA (TGA) appears to be part of the initiation codon for cueR (ATG). The PcueAR promoter shares similarities to the PcopA promoter, particularly in the DNA spacer region between the predicted −35 and −10 sequences, but also contains inverted repeat DNA sequences upstream and downstream of the presumptive promoter site [101]. In the region surrounding the cueAR operon there is a second candidate PcueA-like promoter, named PcopP, which is thought to regulate the expression of copP, a homologue of the CopZ copper chaperone from Enterococcus hirae[102,103]. Little is yet known of the regulation of these promoters.

5.2.3SctR

Salmonella enterica serovar Typhimurium encodes a 138 amino acid protein, SctR, which has been shown to regulate expression of cuiD, a multicopper oxidase [28]. Transposon disruption of sctR resulted in a copper-sensitive phenotype and loss of cuiD expression. Total soluble cell extracts from E. coli overexpressing SctR retarded radioactively labelled PcuiD DNA, showing that SctR has an affinity for PcuiD DNA. An amino acid identity of 92% between SctR and CueR indicates that SctR is the homologue of CueR in S. enterica serovar Typhimurium.

5.2.4HmrR

Both Rhizobium leguminosarum bv. viciae, and Sinorhizobium meliloti encode related P-type ATPases of the CPx subfamily [104] that are involved in copper homeostasis and acid tolerance in these bacteria [29]. These proteins, named ActP, presumably act in both copper and acid tolerance because higher free copper concentrations occur at lower pH [29]. Copper ions regulate the transcriptional activation of actP in both organisms in a pH-dependent manner, due to regulation of actP expression by the MerR family regulator HmrR (heavy metal-responsive regulator) [29]. The predicted amino acid sequences of HmrR from both R. leguminosarum and S. meliloti show about 43% identity to CueR from E. coli, and include residues equivalent to Cys112 and Cys120 of CueR, shown to be required for copper induction [94].

5.3Other metal

5.3.1PbrR – lead resistance

Ralstonia metallidurans CH34 confers resistance to at least seven toxic metals, the determinants of which are located on one of the two endogenous megaplasmids, pMOL28 and pMOL30 [105]. A cosmid library of Sau3A partially digested pMOL30 DNA contained a number of Pb2+ resistant clones, all of which had identical restriction endonuclease digestion patterns. DNA sequence analysis of these clones [25] revealed a merR homologue, named pbrR, which regulates expression of a gene (pbrA) encoding a CPx P-type ATPase and several other genes involved in Pb resistance. The promoter, PpbrA, has a 19 bp spacer between the −35 and −10 sequences. Comparison of the amino acid sequence of PbrR with other metal-responsive MerR-like regulators (see Fig. 6) reveals conservation of the three cysteine residues that are known to be important in metal binding in MerR. Primer extension experiments show that transcription of PpbrA is induced by Pb(II) ions. Experiments in vivo indicate that PbrR responds to Pb(II) but not significantly to other metals [106]. This is the first protein whose biological function is specifically to bind lead salts.

image

Figure 6. Alignment of metal-responsive MerR-like regulators with the N-terminal region of BmrR.

Download figure to PowerPoint

5.3.2CadR

CadR from P. putida 06909 is a predicted 147 amino acid protein which regulates expression of CadA, a cadmium efflux ATPase [27]. It is 38% identical to ZntR and 34% identical to MerR. The arrangement of the cadR and cadA genes, and the bidirectional promoter between them is similar to the arrangement of merR and the structural genes for mercury resistance in Gram-negative mer operons such as Tn501/Tn21 and pbrR and pbrA from the pbr operon of R. metallidurans CH34. The promoter for CadA (PcadA) is responsive to Cd(II), Zn(II) and Pb(II), whilst the promoter for cadR (PcadR) is responsive only to Cd(II) [27]. CadA and CadR confer cadmium resistance and partially confer zinc resistance in P. putida 06909, but do not confer resistance to significant levels of lead ions.

5.3.3CoaR

Open reading frame sII0794 from the cyanobacterium Synechocystis sp. strain PCC 6803 has been identified by two groups as a Co2+-dependent regulator [23,107] and named CoaR (or CorR). CoaR is cobalt-responsive and the first 80 amino acids or so of this protein align well with MerR from Tn501, with some similarity over the first 130 amino acids (Fig. 1). The remaining amino acid sequence of CoaR shows intermittent alignment with CobH (precorrin isomerase) from Pseudomonas denitrificans[23]. The very C-terminus of CoaR contains a sequence, Cys–His–Cys, beyond the region of similarity with CobH, which has been shown by mutagenesis to be required for cobalt sensing. CoaR may therefore respond to cobalt and to hydrogenobyrinic acid (the product of precorrin isomerase).

5.4Metal specificity of induction

The MerR-like metal-dependent regulators are highly discriminatory in their response to metal ions. MerR regulates the expression of the PmerTPAD promoter in vitro in response to Hg(II) ions at 102-fold lower concentration than Cd(II) ions, and 103-fold lower than Zn(II) ions [66]. For the remaining promoters and regulators, only in vivo data are available, which comprise access of the metal to the regulator, metal binding and then transcriptional activation. ZntR regulates the promoter activity of PzntA in response to Zn(II), Cd(II) and to a lesser extent Pb(II) ions [22,96], whilst CueR from E. coli regulates transcription from PcopA in response to copper, silver and gold ions [26,94]. CadR/PcadA from P. putida 06909 responds to Cd(II), Zn(II) and Pb(II) ions, yet the CadR/A system fully confers resistance to Cd(II), partially confers resistance to Zn(II), and does not significantly confer resistance to Pb(II) [27]. PbrR activates PpbrA only in response to Pb(II) ions ([106] and B. Borremans, unpublished data).

The promoters regulated by MerR, ZntR or CueR show a hypersensitive response to the regulatory metals [22,26,66–68]. Each of these promoters regulates transcription of one or more genes responsible for protecting the cell against toxic concentrations of metal ions. The hypersensitive biological switches controlling these promoters may allow them to be expressed at high subtoxic concentrations of metal, without unnecessarily expressing the promoter at lower metal concentrations. Having the necessary macromolecular components pre-assembled at the promoter, with activation dependent on a low molecular mass ligand (the metal ion) may assist in this [71–73]. We have argued previously [68] that it is advantageous for mechanisms protecting against purely toxic metals (e.g. mercury, cadmium, lead salts) to be induced in a hypersensitive manner, whereas mechanisms protecting against high concentrations of micronutrient metals (e.g. copper, zinc and cobalt salts) should be induced more gradually to maintain careful control over homeostasis. The fact that ZntR confers hypersensitive induction in response to Zn(II) argues against this. Data from CueR are inconclusive as the response has only been measured to Cu(II) and Ag(I) in whole cells and not to the likely inducer, Cu(I). Our modified hypothesis is that the MerR family regulators are responsible for hypersensitive induction and will regulate promoters expressing resistance genes. Genes responding to high metal ion concentrations within the normal homeostatic range would be regulated by non-MerR family regulators. These arguments and experiments are complicated by the fact that estimated free metal ion concentrations in the cytoplasm are extremely low [108].

5.5Promoter sequences

All the MerR family promoters so far identified have an elongated spacer region between the −10 and −35 sequences. Those in Tn501 and Tn21 are 19 bp; those in the Bacillus sp. RC607 [93] and S. aureus[109]mer promoters are 20 bp which may explain why Tn501 MerR will bind tightly to and repress expression of the Bacillus promoter, but will not activate it in the presence of Hg(II) ions [93]. The difference in orientation of the −10 and −35 sequences between a 17 bp spacing and a 19 bp spacing on B-form DNA is 6.8 Å of translational separation and 72° of rotation around the helix axis; the equivalent differences for a 20 bp spacer are 10.2 Å and 108°. Therefore it is not surprising that the MerR proteins from Gram-negative sources will not activate these mer promoters from Gram-positive bacteria.

The majority of metal-responsive promoters regulated by MerR-like proteins have 19 bp spacers; PzntA is an exception, having a 20 bp spacer. Interestingly, a MerR–ZntR hybrid regulator, formed by fusion of the first 44 amino acids of Tn501 MerR with the C-terminal 103 amino acids of ZntR, would activate a hybrid promoter containing the dyad symmetrical sequence recognised by MerR inside the 20 bp spacer of ZntR [22]. This suggests that the C-terminal region may dictate the degree of twist imparted to the protein, which in turn distorts the DNA and allows promoter recognition by RNA polymerase.

Each of the metal-dependent promoters has a dyad symmetrical sequence within the spacer region, which in many cases has been shown to be the regulator binding site. The mer promoters from Gram-negative or Gram-positive bacteria share a GTAC sequence on the inner edge of a hyphenated dyad symmetrical sequence. Table 2 shows the sequences of some metal-regulated promoters.

Table 2.  Promoter regions regulated by MerR-like proteins
  1. aThe transcript start, where known, is indicated by /; −35 and −10 sequences have double underlines and symmetrical sequences are shown in bold.

  2. bThe transcription start point has not been determined for the cadA promoter and alternative −10 sequences overlap such that the spacer could be 19 or 20 bp.

PromoterMetalRegulatorSequenceaSpacer (bp)Reference
Tn501/Tn21 merHgMerRCGCinline imageCCGTACATGAGTACGGAAGinline image TACGCT/ATCCA19[4,120]
Bacillus merHgMerRATAinline image CTGTACTAAGGTACGTGGTTinline image GTAAGTGAGG20[93]
Staphylococcus merHgMerRGACinline image GTGTACTATGGTACAGGGTTinline image TTTTATTGAG20[109]
E. coli zntAZnZntRAACinline imageCTGGAGTCGACTCCAGAGTGinline image TCGGTT/AATG20[22]
E. coli copACuCueRTTCinline imageTTCCCCTTGCTGGAAGGTTinline image TTATCAC/AGCC19[24,26]
E. coli cueOCuCueRGGCinline imageTTCCCGTAAGGGGAAGGACinline image CAACGTTTGAT19[24]
Synechocystis coaTCoCoaRACCinline imageTTGACACTAATGTTAAGGTTinline image GAGAAGGTAA20[23]
Ralstonia pbrAPbPbrRGTCinline imageCTATAGTAACTAGAGGGTGinline image CGGCAA/CGCGA19[25]
Pseudomonas cadACdCadRGGCinline imageCTATAGTGGCTACAGGGTGinline image GGCAACAGGC19/20b[27] and acc. no. AF333961
E. coli soxSSoxRCGCinline imageTCAAGTTAACTTGAGGAATinline image CCCCAAC/AGAT19[121]
S. lividans tipATipALGGCinline imageCTCACGTCACGTGAGGAGGinline image GGACGGC/GTCA19[12]
Bacillus bmr BmrRCCGinline image CTCCCCTAGGAGGAGGTCTinline image ATAAGGGATAC19[18]

Expression from a promoter regulated by a MerR-like regulator is dependent upon:

  • 1
    recognition by the regulator of the correct promoter sequence for binding;
  • 2
    the correct coordination of the metal by the regulatory protein;
  • 3
    the correct conformational change in the protein structure upon binding the metal; and
  • 4
    the correct degree of distortion of the DNA by the protein to align the −35 and −10 sequences for recognition by RNA polymerase holoenzyme.

In addition to metal discrimination by direct recognition and binding of the metal by the regulatory protein, a further level of discrimination between metals in induction of gene expression may arise through differences in the spacer length of the regulated promoters. A regulator acting upon a 19 bp spacer would not be able to change the conformation of a 20 bp spacer sufficiently to activate transcription, and a regulator that normally worked on a 20 bp spacer would over-distort a 19 bp spacer so that it would not be aligned correctly for contact with σ70.

6Distribution and evolution of MerR-like regulators

  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Regulation by MerR
  5. 3Other MerR family members
  6. 4Structural studies
  7. 5Metal-dependent regulators
  8. 6Distribution and evolution of MerR-like regulators
  9. 7Conclusions
  10. Acknowledgements
  11. References

We have looked for MerR-like proteins within translated sequences from the rapidly-growing number of bacterial genome sequences. Obviously any such study is out-of-date as soon as it is completed, but one may identify particularly interesting proteins from a previously little studied species. We used simple BLAST searches on sequences from both complete and incomplete bacterial genome sequences at the National Center for Biotechnology Information (NCBI) website and at the specific sites at sequencing facilities (e.g. The Sanger Institute, The US Department of Energy, or The Institute for Genomic Research). Because of the rapidly-changing nature of the data, we report here only the overview of our analysis. Anyone interested in a full analysis of the distribution of MerR-like sequences in bacterial genomes will need to perform this themselves (and at frequent intervals).

6.1Frequencies of MerR family regulators in bacterial genomes

There are features in addition to straightforward similarity analyses which are useful in categorising MerR family regulators. These are based on the biology of the regulators studied to date and discussed in this review. Thus, we have classified the MerR family into three subsets of presumptive regulators:

  • 1
    MerR– MerR in association with a mer operon and showing significant (>40%) identity to MerR of Tn501 or MerR of Bacillus sp. RC607, including three cysteines equivalent to Cys86, Cys117 and Cys126 of Tn501 MerR; usually associated with mercuric reductase.
  • 2
    Metal-responsive MerR-like regulators– likely metal-responding MerR-like regulator in association with metal uptake or export genes or containing two of three cysteines equivalent to Cys86, Cys117 or Cys126 of MerR.
  • 3
    MerR-associated regulators– MerR family regulators lacking similarity in the C-terminal region to known metal-responsive regulators. Some of these may well respond to metals, but cannot easily be identified as such by inspection.

Of the total number of microbial genomes examined, both complete and incomplete, 30% did not have a MerR-like regulator, 3% possessed a merR gene associated with mer operons, and 38% had at least one MerR-like regulator that was likely to be metal-responsive. 45% of the total genomes had a MerR family regulator in the ‘MerR-associated’ class. Many genomes had more than one type of MerR family regulator, hence the above figures exceed 100%. The MerR family of regulators is therefore widespread in eubacteria. Of the 312 MerR family regulators provisionally identified in 156 bacterial genomes, 5% were MerR, 44% were presumptive metal-responsive MerR-like regulators, and 51% were MerR-associated regulators.

The presence of multiple copies of MerR regulators within a genome was a striking feature of the analysis. Previous work has revealed five MerR family regulators within E. coli (SoxR, ZntR, CueR, MlrA and YcgE – see previous sections), but several other bacterial genomes had many MerR-like regulatory genes. Some of these are listed in Table 3 as examples. R. metallidurans CH34 had 15 MerR family genes, of which 13 are possibly metal-responsive; this is not surprising, given that R. metallidurans contains two megaplasmids with nine different metal resistance determinants, and has two mer determinants [105]. R. solanacearum contains five MerR family genes. There are several examples of different species within a genera possessing different numbers of MerR family regulators. This may be due to their particular environmental niche and the need for some bacteria to harbour intricate regulatory systems to respond to particular environmental stimuli.

Table 3.  Number of MerR family genes in a selection of bacterial species
  1. aStrains for which the genome sequences had been completed. In some cases this does not mean the annotation or ‘finishing’ of the sequences had been completed, it may have simply been the shotgun phase of the sequencing.

SpeciesNo.SpeciesNo.SpeciesNo.
Acidithiobacillus ferrooxidans4Desulfitobacterium hafniense7P. putida10
Agrobacterium tumefaciens6Enterococcus faecalis V583a4Pseudomonas syringae pv tomoato5
Bacillus halodurans C-125a6E. coli K12a5R. metalliduransa9
Bacillus stearothermophilus6E. coli 0157:H7 ELD933a4Rhodobacter sphaeroides6
B. subtilisa10E. coli 0157:H7 VT2-Sakai4S. enterica subsp. enterica serovar Dublin3
Bordetella bronchiseptica5Haemophilus influenzae KW20a4  
Bordetella parapertussis5Lactococuus lactisa4S. enterica Typhimurium LT2a7
B. pertussisa4Magnetospirillum magnetotacticum9Salmonella paratyphi4
Brucella melitensis biovar suis4Mesorhizobium lotia6Shewanella putrefaciens3
Burkholderia cepacia12Mycobacterium avium4S. melilotia6
Burkholderia mallei4Mycobacterium smegmatis6S. coelicolora23
Burkholderia pseudomallei3Novosphingobium aromaticivorans5Thermobifida fusca8
Clostridium acetobutylicum3P. aeruginosaa6Vibrio cholerae EI Tor N16961a5
Deinococcus radioduransa7Pseudomonas fluorescens6  

6.2Prediction of MerR regulator function

Inspection of a genome sequence can lead to an indication of function of an open reading frame. There are several examples where this has been demonstrated for the MerR family. Both ZntR and CueR were identified from the E. coli K12 genome sequence and their properties determined subsequently [22,24,26,91,99]. This was done, even though both regulators are separate from their cognate regulated genes. Other regulators have been identified in similar fashion. Some metal-dependent MerR family regulators can be identified from their position relative to other genes, their overall length (130–150 amino acids) and the presence of cysteines equivalent to those of MerR. Two examples are given below:

6.2.1The CadR regulator of P. putida

This was identified in the genome sequence as it was oriented divergently from a gene similar to cadA, encoding an ATP-dependent cadmium transporter [27]. Induction by heavy metals was examined and shown to be Cd(II)>Pb(II)>Zn(II) [27]. A gene designated CadR has also been identified in the P. aeruginosa genome sequence; again this is oriented divergently from a gene similar to cadA, but this was induced by metals in the order Cd(II)>Zn(II)>Hg(II)>Cu(II)>Ni(II)>Co(II)≫Pb(II) (K.R. Brocklehurst, S.J. Megit and A.P. Morby, personal communication).

6.2.2The ZccR regulator of Bordetella pertussis

An open reading frame encoding a presumptive MerR family protein was identified in the B. pertussis genome sequence. This reading frame would be transcribed divergently from an open reading frame encoding a potential CPx P-type ATPase [104] with an N-terminal region containing an unusual histidine-rich repeat. The presumptive promoter for the ATPase had 19 bp spacing between the optimum −10 and −35 sequences. The B. pertussis regulatory gene and the promoter for the ATPase were isolated from the cosmid bank used for sequencing, and cloned upstream of a promoterless lacZ gene. The regulator was shown to bind to a dyad symmetrical sequence in the promoter and to be activated by zinc, cadmium and cobalt [110].

6.3Thoughts on the evolution of MerR-like regulators

The MerR family is distinguished by the very clear divisions between an N-terminal DNA binding region and a C-terminal inducer binding region in members of the family. There is high amino acid similarity in the DNA binding region and low similarity in the inducer binding region; both regions are believed to be separate protein domains. The two structures available to date show that these domains are separate and joined by a coiled-coil region. Phylogenetic analysis of the helix-turn-helix proteins, including MerR, suggested that all such helix-turn-helix motifs arose from a common ancestor and further showed that the sequence in MerR was closely related to those in σ70 and σ54[111]– proteins which also positively regulate the RNA polymerase.

The separate protein products (TipAL and TipAS) of the tipA gene of S. lividans first led to the idea that MerR family regulators may have arisen from the fusion of N-terminal DNA binding and subunit interaction domains with a separate effector binding protein. In this case the TipAS protein is a thiostrepton binding protein from an internal translation start within the tipA reading frame. Our hypothesis is that a DNA sequence encoding a DNA binding domain and a coiled-coil region for dimer interaction became fused with a hypothetical ‘tipAS gene’ to form a gene equivalent to tipA. The new full-length protein had an extra 110 amino acids at its N-terminus which provided protein–DNA and protein–protein interaction; the protein–protein interactions would be dependent on the presence or absence of the bound inducer, thiostrepton.

Similarly, the nolA gene of Bradyrhizobium japonicum gives multiple gene products from internal translational starts [112]. Three proteins are produced, only the first of which, NolA1, contains the helix-turn-helix DNA binding motif and it regulates the expression of NolA2 and NolA3 as well as acting at the nolD promoter [113].

There are other examples in which gene product may be a fusion between N-terminal DNA binding regulatory regions and C-terminal enzymes or binding proteins. CoaR from Synechocystis PCC 6803 (see Fig. 1) consists of a region of ca. 135 amino acids similar to MerR, then 210 amino acids showing intermittent similarity to the P. denitrificans cobH gene product, precorrin isomerase, and finally a short (15 amino acids) sequence containing a Cys–His–Cys motif, which is required for cobalt recognition and response [23]. An evolutionary mechanism could be the initial generation of a regulatory gene by fusion of the merR-like sequence with an ancestral cobH gene, the product of which would respond to cobalamin precursors. Subsequent mutation generated a cobalt-responsive sequence at the very C-terminus of the regulator. The C-terminal amino acids of MerR are also important in inducer recognition, as this is where the specificity for the organomercurial response of MerR from ‘broad spectum’ mercury resistance determinants resides [55].

Evolution by gene fusion may also apply to the CarA protein of Myxococcus xanthus, which has N-terminal similarity to the MerR protein of Bacillus cereus over about 75 amino acids. The C-terminal region is similar to the cobalamin binding regions of methionine synthetase (MetH) of E. coli and methylmalonylCoA mutase (MutB) from Propionibacterium shermanii, and the regulator responds to cobalamin derivatives [114].

Evidence that the N-terminal region may in some cases exist as an independent domain has come from recent work on the bldC mutant of Streptomyces coelicolor. This mutant [115] is defective in producing aerial mycelium at an early stage of differentiation to form aerial spores and in antibiotic production. Rescue and sequencing of the bldC gene showed that it encoded a 68-residue MerR-like protein (which aligned with positions 1–65 of Tn501 MerR) and that the mutation was a single amino acid change in the sequence ([116] and A.C. Hunt and M.J. Buttner, personal communication). These data are the first evidence that the N-terminal DNA binding/subunit interaction domains of MerR family regulators can function independently of a C-terminal effector recognition domain.

One can envisage, therefore, the evolution of a new MerR family regulator by the fusion of a ‘BldC-like’ domain with an effector binding protein such that the binding of effector transmits a conformational change to the N-terminal DNA binding domain, thereby putting a pre-existing promoter under novel regulation. Subsequent mutation of the regulator and promoter would allow divergent evolution of the novel function. In all cases so far identified, the effector is an inducer; in principle, the new regulator might activate gene expression and require a co-repressor to regulate expression.

The metal-responsive MerR family regulators, MerR, ZntR, CueR, PbrR, PmtR, CadR, ZccR (see Section 5 and Fig. 1) are all closely related. Indeed, domain swaps can be done in this subfamily of regulators [22]. Evolution of this subfamily was probably through mutation of an ancestral metal-responsive regulator. Subsequent gain and loss of coordinating amino acid residues and small deletions and insertions which alter the position of coordinating functional groups, might effect changes to the metal binding specificity. It is not obvious which of the metals would have been the ancestral co-regulator, but the remaining metal specificities could have arisen from a single example.

7Conclusions

  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Regulation by MerR
  5. 3Other MerR family members
  6. 4Structural studies
  7. 5Metal-dependent regulators
  8. 6Distribution and evolution of MerR-like regulators
  9. 7Conclusions
  10. Acknowledgements
  11. References

We define the MerR family of transcriptional regulators as dimeric proteins with an N-terminal helix-turn-helix DNA binding region, followed by an antiparallel coiled-coil subunit interaction region, and usually by a C-terminal effector binding region. We predict that most, if not all, members of the family are activators, and act at promoters with long spacer regions and respond to the binding of inducers by distorting the promoter DNA to allow open complex formation and transcriptional activation. Testing this prediction requires the characterisation of more members of the family.

Some MerR-like regulators, including MerR, show distinct similarities between their inducer binding regions and respond to metals. These form a subfamily of MerR-like regulators (compare Figs. 6 and 1) differentiated by size and sequence from the separate subfamilies that bind large aromatic compounds, and we await experimental evidence to reveal the mechanisms of metal discrimination between different members of this subfamily. There may be a separate subfamily consisting of the independent DNA binding and subunit interaction region, with no C-terminal inducer binding region; this currently contains only one member, BldC.

Bacterial genome sequencing will identify increasing numbers of this family of regulators and we await elucidation of their roles.

Acknowledgements

  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Regulation by MerR
  5. 3Other MerR family members
  6. 4Structural studies
  7. 5Metal-dependent regulators
  8. 6Distribution and evolution of MerR-like regulators
  9. 7Conclusions
  10. Acknowledgements
  11. References

We are grateful to many colleagues (collaborators and competitors) over the years for discussion, to Mark Buttner, Alison Hunt and Andy Morby for communicating data prior to publication, and to an anonymous referee who made valuable suggestions to improve this article. Recent work in our laboratory was supported by grants from the Biotechnology and Biological Sciences Research Council and a Fellowship to N.L.B. from the Leverhulme Trust. J.V.S. was supported by the Darwin Trust of Edinburgh. Bioinformatics facilities were provided by Medical Research Council Infrastructure Award G.4500017.

References

  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Regulation by MerR
  5. 3Other MerR family members
  6. 4Structural studies
  7. 5Metal-dependent regulators
  8. 6Distribution and evolution of MerR-like regulators
  9. 7Conclusions
  10. Acknowledgements
  11. References
  • [1]
    Brown, N.L, Ford, S.J, Pridmore, R.D, Fritzinger, D.C (1983) Nucleotide sequence of a gene from the Pseudomonas transposon Tn501 encoding mercuric reductase. Biochemistry 22, 40894095.
  • [2]
    Brown, N.L, Pridmore, R.D, Fritzinger, D.C (1984) The mercury-resistance genes of transposon Tn501– nucleotide sequence of the mer operon and a possible mechanism for mercury detoxification. Biochem. Soc. Trans. 12, 276277.
  • [3]
    Brown, N.L, Misra, T.K, Winnie, J.N, Schmidt, A, Seiff, M, Silver, S (1986) The nucleotide sequence of the mercuric resistance operons of plasmid R100 and transposon Tn501– further evidence for mer genes which enhance the activity of the mercuric ion detoxification system. Mol. Gen. Genet. 202, 143151.
  • [4]
    Barrineau, P, Gilbert, P, Jackson, W.J, Jones, C.S, Summers, A.O, Wisdom, S (1984) The DNA sequence of the mercury resistance operon of the IncFII plasmid NR1. J. Mol. Appl. Genet. 2, 601619.
  • [5]
    Barrineau, P, Gilbert, P, Jackson, W.J, Jones, C.S, Summers, A.O, Wisdom, S (1985) The structure of the mer operon. Basic Life Sci. 30, 707718.
  • [6]
    Lund, P.A, Ford, S.J, Brown, N.L (1986) Transcriptional regulation of the mercury resistance genes of transposon Tn501. J. Gen. Microbiol. 132, 465480.
  • [7]
    Lund, P.A, Brown, N.L (1989) Regulation of transcription in Escherichia coli from the mer and merR promoters in the transposon Tn501. J. Mol. Biol. 205, 343353.
  • [8]
    O'Halloran, T.V, Walsh, C.T. Positive and negative control of prokaryotic gene-expression by a metalloprotein – purification and characterization of the MerR regulatory protein. J. Cell Biochem. Suppl. 10D, 1986. 104
  • [9]
    Heltzel, A, Gambill, D, Jackson, W.J, Totis, P.A, Summers, A.O (1987) Overexpression and DNA-binding properties of the mer-encoded regulatory protein from plasmid NR1 (Tn21). J. Bacteriol. 169, 33793384.
  • [10]
    O'Halloran, T.V, Frantz, B, Shin, M.K, Ralston, D.M, Wright, J.G (1989) The MerR heavy-metal receptor mediates positive activation in a topologically novel transcription complex. Cell 56, 119129.
  • [11]
    Lund, P, Brown, N (1989) Up-promoter mutations in the positively-regulated mer promoter of Tn501. Nucleic Acids Res. 17, 55175527.
  • [12]
    Holmes, D.J, Caso, J.L, Thompson, C.J (1993) Autogenous transcriptional activation of a thiostrepton-induced gene in Streptomyces lividans. EMBO J. 12, 31833191.
  • [13]
    Amabile-Cuevas, C.F, Demple, B (1991) Molecular characterization of the soxRS genes of Escherichia coli– 2 genes control a superoxide stress regulon. Nucleic Acids Res. 19, 44794484.
  • [14]
    Nunoshiba, T, Hidalgo, E, Cuevas, C.F.A, Demple, B (1992) 2-Stage control of an oxidative stress regulon – the Escherichia coli SoxR protein triggers redox-inducible expression of the soxS regulatory gene. J. Bacteriol. 174, 60546060.
  • [15]
    Zeng, Q.D, Stalhandske, C, Anderson, M.C, Scott, R.A, Summers, A.O (1998) The core metal-recognition domain of MerR. Biochemistry 37, 1588515895.
  • [16]
    Caguiat, J.J, Watson, A.L, Summers, A.O (1999) Cd(II)-responsive and constitutive mutants implicate a novel domain in MerR. J. Bacteriol. 181, 34623471.
  • [17]
    Ahmed, M, Lyass, L, Markham, P.N, Taylor, S.S, Vazquezlaslop, N, Neyfakh, A.A (1995) Two highly similar multidrug transporters of Bacillus subtilis whose expression is differentially regulated. J. Bacteriol. 177, 39043910.
  • [18]
    Ahmed, M, Borsch, C.M, Taylor, S.S, Vazquezlaslop, N, Neyfakh, A.A (1994) A protein that activates expression of a multidrug efflux transporter upon binding the transporter substrates. J. Biol. Chem. 269, 2850628513.
  • [19]
    Heldwein, E.E.Z, Brennan, R.G (2001) Crystal structure of the transcription activator BmrR bound to DNA and a drug. Nature 409, 378382.
  • [20]
    Godsey, M.H, Baranova, N.N, Neyfakh, A.A, Brennan, R.G (2001) Crystal structure of MtaN, a global multidrug transporter gene activator. J. Biol. Chem. 276, 4717847184.
  • [21]
    Noll, M, Petrukhin, K, Lutsenko, S (1998) Identification of a novel transcription regulator from Proteus mirabilis, PMTR, revealed a possible role of YJAI protein in balancing zinc in Escherichia coli. J. Biol. Chem. 273, 2139321401.
  • [22]
    Brocklehurst, K.R, Hobman, J.L, Lawley, B, Blank, L, Marshall, S.J, Brown, N.L, Morby, A.P (1999) ZntR is a Zn(II)-responsive MerR-like transcriptional regulator of zntA in Escherichia coli. Mol. Microbiol. 31, 893902.
  • [23]
    Rutherford, J.C, Cavet, J.S, Robinson, N.J (1999) Cobalt-dependent transcriptional switching by a dual-effector MerR-like protein regulates a cobalt-exporting variant CPx-type ATPase. J. Biol. Chem. 274, 2582725832.
  • [24]
    Outten, F.W, Outten, C.E, Hale, J, O'Halloran, T.V (2000) Transcriptional activation of an Escherichia coli copper efflux regulon by the chromosomal MerR homologue, CueR. J. Biol. Chem. 275, 3102431029.
  • [25]
    Borremans, B, Hobman, J.L, Provoost, A, Brown, N.L, van der Lelie, D (2001) Cloning and functional analysis of the pbr lead resistance determinant of Ralstonia metallidurans CH34. J. Bacteriol. 183, 56515658.
  • [26]
    Stoyanov, J.V, Hobman, J.L, Brown, N.L (2001) CueR (YbbI) of Escherichia coli is a MerR family regulator controlling expression of the copper exporter CopA. Mol. Microbiol. 39, 502511.
  • [27]
    Lee, S.W, Glickmann, E, Cooksey, D.A (2001) Chromosomal locus for cadmium resistance in Pseudomonas putida consisting of a cadmium-transporting ATPase and a MerR family response regulator. Appl. Environ. Microbiol. 67, 14371444.
  • [28]
    Kim, J.S, Kim, M.H, Joe, M.H, Song, S.S, Lee, I.S, Choi, S.Y (2002) The sctR of Salmonella enterica serovar Typhimurium encoding a homologue of MerR protein is involved in the copper-responsive regulation of cuiD. FEMS Microbiol. Lett. 210, 99103.
  • [29]
    Reeve, W.G, Tiwari, R.P, Kale, N.B, Dilworth, M.J, Glenn, A.R (2002) ActP controls copper homeostasis in Rhizobium leguminosarum bv. viciae and Sinorhizobium meliloti preventing low pH-induced copper toxicity. Mol. Microbiol. 43, 981991.
  • [30]
    Ross, W, Park, S.J, Summers, A.O (1989) Genetic analysis of transcriptional activation and repression in the Tn21 mer operon. J. Bacteriol. 171, 40094018.
  • [31]
    Shewchuk, L.M, Helmann, J.D, Ross, W, Park, S.J, Summers, A.O, Walsh, C.T (1989) Transcriptional switching by the MerR protein – activation and repression mutants implicate distinct DNA and mercury(II) binding domains. Biochemistry 28, 23402344.
  • [32]
    Helmann, J.D, Ballard, B.T, Walsh, C.T (1990) The MerR metalloregulatory protein binds mercuric ion as a tricoordinate, metal-bridged dimer. Science 247, 946948.
  • [33]
    Summers, A.O (1992) Untwist and shout – a heavy metal-responsive transcriptional regulator. J. Bacteriol. 174, 30973101.
  • [34]
    Lee, I.W, Livrelli, V, Park, S.J, Totis, P.A, Summers, A.O (1993) In vivo DNA-protein interactions at the divergent mercury resistance (mer) promoters. 2. Repressor activator (MerR)-RNA polymerase interaction with merOP mutants. J. Biol. Chem. 268, 26322639.
  • [35]
    Livrelli, V, Lee, I.W, Summers, A.O (1993) In vivo DNA-protein interactions at the divergent mercury resistance (mer) promoters. 1. Metalloregulatory protein MerR mutants. J. Biol. Chem. 268, 26232631.
  • [36]
    Markham, P.N, Ahmed, M, Neyfakh, A.A (1996) The drug-binding activity of the multidrug-responding transcriptional regulator BmrR resides in its C-terminal domain. J. Bacteriol. 178, 14731475.
  • [37]
    Baranova, N.N, Danchin, A, Neyfakh, A.A (1999) Mta, a global MerR-type regulator of the Bacillus subtilis multidrug-efflux transporters. Mol. Microbiol. 31, 15491559.
  • [38]
    Stanisich, V.A, Bennett, P.M, Richmond, M.H (1977) Characterisation of a translocation unit encoding resistance to mercuric ions that occurs on a nonconjugative plasmid in Pseudomonas aeruginosa. J. Bacteriol. 129, 12271233.
  • [39]
    Foster, T.J, Ginnity, F (1985) Some mercurial resistance plasmids from different incompatibility groups specify merR regulatory functions that both repress and induce the mer operon of plasmid-R100. J. Bacteriol. 162, 773776.
  • [40]
    Hobman, J.L. and Brown, N.L. (1997) Bacterial mercury resistance genes. In: Metal Ions in Biological Systems, Vol. 34, pp. 527–568. Marcel Dekker Inc.
  • [41]
    Sedlmeier, R, Altenbuchner, J (1992) Cloning and DNA sequence analysis of the mercury resistance genes of Streptomyces lividans. Mol. Gen. Genet. 236, 7685.
  • [42]
    Busenlehner, L.S, Pennella, M.A, Giedroc, D.P (2003) The SmtB/ArsR family of metalloregulatory transcriptional repressors: structural insights into prokaryotic metal resistance. FEMS Microbiol. Rev. 27, 131143.
  • [43]
    Liebert, C.A, Watson, A.L, Summers, A.O (2000) The quality of merC, a module of the mer mosaic. J. Mol. Evol. 51, 607622.
  • [44]
    Hobman, J, Kholodii, G, Nikiforov, V, Ritchie, D.A, Strike, P, Yurieva, O (1994) The sequence of the mer operon of Pmer327/419 and transposon ends of Pmer327/419, Pmer330 and Pmer05. Gene 146, 7378.
  • [45]
    Mindlin, S, Kholodii, G, Gorlenko, Z, Minakhina, S, Minakhin, L, Kalyaeva, E, Kopteva, A, Petrova, M, Yurieva, O, Nikiforov, V (2001) Mercury resistance transposons of Gram-negative environmental bacteria and their classification. Res. Microbiol. 152, 811822.
  • [46]
    Mukhopadhyay, D, Yu, H.R, Nucifora, G, Misra, T.K (1991) Purification and functional characterization of MerD – a coregulator of the mercury resistance operon in Gram-negative bacteria. J. Biol. Chem. 266, 1853818542.
  • [47]
    Liebert, C.A, Hall, R.M, Summers, A.O (1999) Transposon Tn21, flagship of the floating genome. Microbiol. Mol. Biol. Rev. 63, 507522.
  • [48]
    Barkay, T, Miller, S.M, Summers, A.O (2003) Bacterial mercury resistance from atoms to ecosystems. FEMS Microbiol. Rev. 27, 355384.
  • [49]
    Parkhill, J, Brown, N.L (1990) Site-specific insertion and deletion mutants in the mer promoter-operator region of Tn501– the 19 base-pair spacer is essential for normal induction of the promoter by merR. Nucleic Acids Res. 18, 51575162.
  • [50]
    Park, S.J, Wireman, J, Summers, A.O (1992) Genetic analysis of the Tn21 mer operator-promoter. J. Bacteriol. 174, 21602171.
  • [51]
    Shewchuk, L.M, Verdine, G.L, Nash, H, Walsh, C.T (1989) Mutagenesis of the cysteines in the metalloregulatory protein MerR indicates that a metal-bridged dimer activates transcription. Biochemistry 28, 61406145.
  • [52]
    Wright, J.G, Natan, M.J, Macdonnell, F.M, Ralston, D.M, O'Halloran, T.V (1990) Mercury(II) thiolate chemistry and the mechanism of the heavy-metal biosensor MerR. Prog. Inorg. Chem. 38, 323412.
  • [53]
    Wright, J.G, Tsang, H.T, Penner-Hahn, J.E, O'Halloran, T.V (1990) Coordination chemistry of the Hg-MerR metalloregulatory protein – evidence for a novel tridentate Hg-cysteine receptor-site. J. Am. Chem. Soc. 112, 24342435.
  • [54]
    Watton, S.P, Wright, J.G, Macdonnell, F.M, Bryson, J.W, Sabat, M, O'Halloran, T.V (1990) Trigonal mercuric complex of an aliphatic thiolate – a spectroscopic and structural model for the receptor-site in the Hg(II) biosensor MerR. J. Am. Chem. Soc. 112, 28242826.
  • [55]
    Nucifora, G, Chu, L, Silver, S, Misra, T.K (1989) Mercury operon regulation by the merR gene of the organomercurial resistance system of plasmid pDU1358. J. Bacteriol. 171, 42414247.
  • [56]
    O'Halloran, T, Walsh, C (1987) Metalloregulatory DNA-binding protein encoded by the merR gene – isolation and characterization. Science 235, 211214.
  • [57]
    Ishihama, A (1993) Protein–protein communication within the transcription apparatus. J. Bacteriol. 175, 24832489.
  • [58]
    Caslake, L.F, Ashraf, S.I, Summers, A.O (1997) Mutations in the alpha and sigma-70 subunits of RNA polymerase affect expression of the mer operon. J. Bacteriol. 179, 17871795.
  • [59]
    Kulkarni, R.D, Summers, A.O (1999) MerR cross-links to the α, β and σ70 subunits of RNA polymerase in the preinitiation complex at the merTPCAD promoter. Biochemistry 38, 33623368.
  • [60]
    Parkhill, J, Lawley, B, Hobman, J.L, Brown, N.L (1998) Selection and characterization of mercury-independent activation mutants of the Tn501 transcriptional regulator, MerR. Microbiology 144, 28552864.
  • [61]
    Parkhill, J, Ansari, A.Z, Wright, J.G, Brown, N.L, O'Halloran, T.V (1993) Construction and characterization of a mercury-independent MerR activator (MerRAC) – transcriptional activation in the absence of Hg(II) is accompanied by DNA distortion. EMBO J. 12, 413421.
  • [62]
    Huffman, D.L., Utschig, L.M. and O'Halloran, T.V. (1997) Mercury-responsive gene regulation and mercury-199 as a probe of protein structure. In: Metal Ions in Biological Systems, Vol. 34, pp. 503–526. Marcel Dekker Inc.
  • [63]
    Utschig, L.M, Bryson, J.W, O'Halloran, T.V (1995) Hg-199 NMR of the metal receptor site in MerR and its protein-DNA complex. Science 268, 380385.
  • [64]
    Dieckmann, G.R, McRorie, D.K, Tierney, D.L, Utschig, L.M, Singer, C.P, O'Halloran, T.V, Penner-Hahn, J.E, DeGrado, W.F, Pecoraro, V.L (1997) De novo design of mercury-binding two- and three-helical bundles. J. Am. Chem. Soc. 119, 61956196.
  • [65]
    Dieckmann, G.R, McRorie, D.K, Lear, J.D, Sharp, K.A, DeGrado, W.F, Pecoraro, V.L (1998) The role of protonation and metal chelation preferences in defining the properties of mercury-binding coiled coils. J. Mol. Biol. 280, 897912.
  • [66]
    Ralston, D.M, O'Halloran, T.V (1990) Ultrasensitivity and heavy-metal selectivity of the allosterically modulated MerR transcription complex. Proc. Natl. Acad. Sci. USA 87, 38463850.
  • [67]
    Condee, C.W, Summers, A.O (1992) A mer-lux transcriptional fusion for real-time examination of in vivo gene-expression kinetics and promoter response to altered superhelicity. J. Bacteriol. 174, 80948101.
  • [68]
    Rouch, D.A, Parkhill, J, Brown, N.L (1995) Induction of bacterial mercury- and copper-responsive promoters: functional differences between inducible systems and implications for their use in gene-fusions for in vivo metal biosensors. J. Indust. Microbiol. 14, 249253.
  • [69]
    Ansari, A.Z, Bradner, J.E, O'Halloran, T.V (1995) DNA-bend modulation in a repressor-to-activator switching mechanism. Nature 374, 371375.
  • [70]
    Ansari, A.Z, Chael, M.L, O'Halloran, T.V (1992) Allosteric underwinding of DNA is a critical step in positive control of transcription by Hg-MerR. Nature 355, 8789.
  • [71]
    Goldbeter, A, Koshland, D.E (1981) An amplified sensitivity arising from covalent modification in biological systems. Proc. Natl. Acad. Sci. USA 78, 68406844.
  • [72]
    Koshland, D.E (1987) Switches, thresholds and ultrasensitivity. Trends Biochem. Sci. 12, 225229.
  • [73]
    Goldbeter, A, Koshland, D.E (1984) Ultrasensitivity in biochemical systems controlled by covalent modification – interplay between zero-order and multistep effects. J. Biol. Chem. 259, 14411447.
  • [74]
    Comess, K.M, Shewchuk, L.M, Ivanetich, K, Walsh, C.T (1994) Construction of a synthetic gene for the metalloregulatory protein MerR and analysis of regionally mutated proteins on transcriptional regulation. Biochemistry 33, 41754186.
  • [75]
    Hidalgo, E, Demple, B (1997) Spacing of promoter elements regulates the basal expression of the soxS gene and converts SoxR from a transcriptional activator into a repressor. EMBO J. 16, 10561065.
  • [76]
    Hidalgo, E, Leautaud, V, Demple, B (1998) The redox-regulated SoxR protein acts from a single DNA site as a repressor and an allosteric activator. EMBO J. 17, 26292636.
  • [77]
    Chiu, M.L, Viollier, P.H, Katoh, T, Ramsden, J.J, Thompson, C.J (2001) Ligand-induced changes in the Streptomyces lividans TipAL protein imply an alternative mechanism of transcriptional activation for MerR-like proteins. Biochemistry 40, 1295012958.
  • [78]
    Chiu, M.L, Folcher, M, Katoh, T, Puglia, A.M, Vohradsky, J, Yun, B.S, Seto, H, Thompson, C.J (1999) Broad spectrum thiopeptide recognition specificity of the Streptomyces lividans TipAL protein and its role in regulating gene expression. J. Biol. Chem. 274, 2057820586.
  • [79]
    Chen, K, Outten, C.E, O'Halloran, T.V (2001) DNA distortion mechanism for transcriptional activation by a copper-responsive gene-regulatory protein CueR in E. coli. J. Inorg. Biochem. 86, 179179.
  • [80]
    Outten, C.E, Outten, F.W, O'Halloran, T.V (1999) DNA distortion mechanism for transcriptional activation by ZntR, a Zn(II)-responsive MerR homologue in Escherichia coli. J. Biol. Chem. 274, 3751737524.
  • [81]
    Khan, S, Brocklehurst, K.R, Jones, G.W, Morby, A.P (2002) The functional analysis of directed amino acid alterations in ZntR from Escherichia coli. Biochem. Biophys. Res. Commun. 299, 438455.
  • [82]
    Lee, I.W, Gambill, B.D, Summers, A.O (1989) Translation of MerD in Tn21. J. Bacteriol. 171, 22222225.
  • [83]
    Nucifora, G, Silver, S, Misra, T.K (1989) Down regulation of the mercury resistance operon by the merD gene. Mol. Gen. Genet. 220, 6972.
  • [84]
    Demple, B (1996) Redox signaling and gene control in the Escherichia coli soxRS oxidative stress regulon – A review. Gene 179, 5357.
  • [85]
    Pomposiello, P.J, Demple, B (2001) Redox-operated genetic switches: the SoxR and OxyR transcription factors. Trends Biotechnol. 19, 109114.
  • [86]
    Hidalgo, E, Bollinger, J.M, Bradley, T.M, Walsh, C.T, Demple, B (1995) Binuclear [2Fe-2S] clusters in the Escherichia coli SoxR protein and role of the metal centers in transcription. J. Biol. Chem. 270, 2090820914.
  • [87]
    Hidalgo, E, Demple, B. The iron-sulfur clusters present in SoxR regulate its activity as a transcriptional factor. J. Cell Biochem. Suppl. 21A, 1995. 245
  • [88]
    Bradley, T.M, Hidalgo, E, Leautaud, V, Ding, H, Demple, B (1997) Cysteine-to-alanine replacements in the Escherichia coli SoxR protein and the role of the [2Fe-2S] centers in transcriptional activation. Nucleic Acids Res. 25, 14691475.
  • [89]
    Chiu, M.L, Folcher, M, Griffin, P, Holt, T, Klatt, T, Thompson, C.J (1996) Characterization of the covalent binding of thiostrepton to a thiostrepton-induced protein from Streptomyces lividans. Biochemistry 35, 23322341.
  • [90]
    Markham, P.N, LoGuidice, J, Neyfakh, A.A (1997) Broad ligand specificity of the transcriptional regulator of the Bacillus subtilis multidrug transporter Bmr. Biochem. Biophys. Res. Commun. 239, 269272.
  • [91]
    Christie, G.E, White, T.J, Goodwin, T.S (1994) A MerR homolog at 74 minutes on the Escherichia coli genome. Gene 146, 131132.
  • [92]
    Brown, P.K, Dozois, C.M, Nickerson, C.A, Zuppardo, A, Terlonge, J, Curtiss, R (2001) MlrA, a novel regulator of curli (AgF) and extracellular matrix synthesis by Escherichia coli and Salmonella enterica serovar Typhimurium. Mol. Microbiol. 41, 349363.
  • [93]
    Helmann, J.D, Wang, Y, Mahler, I, Walsh, C.T (1989) Homologous metalloregulatory proteins from both Gram-positive and Gram-negative bacteria control transcription of mercury resistance operons. J. Bacteriol. 171, 222229.
  • [94]
    Stoyanov, J.V, Brown, N.L (2003) The Escherichia coli copper-responsive copA promoter is activated by gold. J. Biol. Chem. 278, 14071410.
  • [95]
    Beard, S.J, Hashim, R, MembrilloHernandez, J, Hughes, M.N, Poole, R.K (1997) Zinc(II) tolerance in Escherichia coli K-12: evidence that the zntA gene (o732) encodes a cation transport ATPase. Mol. Microbiol. 25, 883891.
  • [96]
    Binet, M.R.B, Poole, R.K (2000) Cd(II), Pb(II) and Zn(II) ions regulate expression of the metal-transporting P-type ATPase ZntA in Escherichia coli. FEBS Lett. 473, 6770.
  • [97]
    Hitomi, Y, Outten, C.E, O'Halloran, T.V (2001) Extreme zinc-binding thermodynamics of the metal sensor/regulator protein, ZntR. J. Am. Chem. Soc. 123, 86148615.
  • [98]
    Leonhartsberger, S, Huber, A, Lottspeich, F, Bock, A (2001) The hydH/G genes from Escherichia coli code for a zinc and lead responsive two-component regulatory system. J. Mol. Biol. 307, 93105.
  • [99]
    Petersen, C, Moller, L.B (2000) Control of copper homeostasis in Escherichia coli by a P-type ATPase, CopA, and a MerR-like transcriptional activator, CopR. Gene 261, 289298.
  • [100]
    Grass, G, Rensing, C (2001) CueO is a multi-copper oxidase that confers copper tolerance in Escherichia coli. Biochem. Biophys. Res. Commun. 286, 902908.
  • [101]
    Adaikkalam, V, Swarup, S (2002) Molecular characterization of an operon, cueAR, encoding a putative P1-type ATPase and a MerR-type regulatory protein involved in copper homeostasis in Pseudomonas putida. Microbiology 148, 28572867.
  • [102]
    Odermatt, A, Solioz, M (1995) Two trans-acting metalloregulatory proteins controlling expression of the copper ATPases of Enterococcus hirae. J. Biol. Chem. 270, 43494354.
  • [103]
    Solioz, M, Stoyanov, J.V (2003) Copper homeostasis in Enteroccocus hirae. FEMS Microbiol. Rev. 27, 183195.
  • [104]
    Solioz, M, Vulpe, C (1996) CPx-type ATPases: a class of P-type ATPases that pump heavy metals. Trends Biochem. Sci. 21, 237241.
  • [105]
    Mergeay, M, Monchy, S, Vallaeys, T, Auquier, V, Benotmane, A, Bertin, P, Taghavi, S, Dunn, J, van der Lelie, D, Wattiez, R (2003) Ralstonia metallidurans, a bacterium specifically adapted to toxic metals: towards a tentative catalogue of metal-responsive genes. FEMS Microbiol. Rev. 27, 385410.
  • [106]
    Corbisier, P, van der Lelie, D, Borremans, B, Provoost, A, De Lorenzo, V, Brown, N.L, Lloyd, J.R, Hobman, J.L, Csoregi, E, Johansson, G, Mattiasson, B (1999) Whole cell- and protein-based biosensors for the detection of bioavailable heavy metals in environmental samples. Anal. Chim. Acta 387, 235244.
  • [107]
    Garcia-Dominguez, M, Lopez-Maury, L, Florencio, F.J, Reyes, J.C (2000) A gene cluster involved in metal homeostasis in the cyanobacterium Synechocystis sp strain PCC 6803. J. Bacteriol. 182, 15071514.
  • [108]
    Outten, C.E, O'Halloran, T.V (2001) Femtomolar sensitivity of metalloregulatory proteins controlling zinc homeostasis. Science 292, 24882492.
  • [109]
    Laddaga, R.A, Chu, L, Misra, T.K, Silver, S (1987) Nucleotide sequence and expression of the mercurial resistance operon from Staphylococcus aureus plasmid pI258. Proc. Natl. Acad. Sci. USA 84, 51065110.
  • [110]
    Kidd, S.P, Brown, N.L (2003) ZccR – a MerR-like regulator from Bordetella pertussis which responds to zinc, cadmium and cobalt. Biochem. Biophys. Res. Commun. 302, 697702.
  • [111]
    Rosinski, J.A, Atchley, W.R (1999) Molecular evolution of helix-turn-helix proteins. J. Mol. Evol. 49, 301309.
  • [112]
    Loh, J, Stacey, M.G, Sadowsky, M.J, Stacey, G (1999) The Bradyrhizobium japonicum nolA gene encodes three functionally distinct proteins. J. Bacteriol. 181, 15441554.
  • [113]
    Loh, J, Stacey, G (2003) Nodulation gene regulation in Bradyrhizobium japonicum: a unique integration of global regulatory circuits. Appl. Environ. Microbiol. 69, 1017.
  • [114]
    Cervantes, M, Murillo, F.J (2002) Role for vitamin B-12 in light induction of gene expression in the bacterium Myxococcus xanthus. J. Bacteriol. 184, 22152224.
  • [115]
    Merrick, M.J (1976) A morphological and genetic mapping study of bald colony mutants of Streptomyces coelicolor A3. J. Gen. Microbiol. 96, 299315.
  • [116]
    Hunt, A.C. (2003) The bldC developmental locus of Streptomyces coelicolor. PhD Thesis. The University of East Anglia, Norwich.
  • [117]
    Fuentes, A.M, Diaz-Mejia, J.J, Maldonado-Rodriguez, R, Amabile-Cuevas, C.F (2001) Differential activities of the SoxR protein of Escherichia coli: SoxS is not required for gene activation under iron deprivation. FEMS Microbiol. Lett. 201, 271275.
  • [118]
    Sadowsky, M.J, Cregan, P.B, Gottfert, M, Sharma, A, Gerhold, D, Rodriguezquinones, F, Keyser, H.H, Hennecke, H, Stacey, G (1991) The Bradyrhizobium japonicum nolA gene and its involvement in the genotype-specific nodulation of soybeans. Proc. Natl. Acad. Sci. USA 88, 637641.
  • [119]
    Garcia, M, Dunlap, J, Loh, J, Stacey, G (1996) Phenotypic characterization and regulation of the nolA gene of Bradyrhizobium japonicum. Mol. Plant-Microbe Interact. 9, 625636.
  • [120]
    Misra, T.K, Brown, N.L, Fritzinger, D.C, Pridmore, R.D, Barnes, W.M, Haberstroh, L, Silver, S (1984) Mercuric ion-resistance operons of plasmid R100 and transposon Tn501: the beginning of the operon including the regulatory region and the first two structural genes. Proc. Natl. Acad. Sci. USA 81, 59755979.
  • [121]
    Hidalgo, E, Demple, B (1994) An iron-sulfur center essential for transcriptional activation by the redox-sensing SoxR protein. EMBO J. 13, 138146.