Effects of systematic variation of the minimal Escherichia coli met consensus operator site: in vivo and in vitro met repressor binding

Methionine

This review focuses on the steps unique to methionine biosynthesis, namely the conversion of homoserine to methionine. The past decade has provided a wealth of information concerning the details of methionine metabolism and the review focuses on providing a comprehensive overview of the field, emphasizing more recent findings. Details of methionine biosynthesis are addressed along with key cellular aspects, including regulation, uptake, utilization, AdoMet, the methyl cycle, and growing evidence that inhibition of methionine biosynthesis occurs under stressful cellular conditions. The first unique step in methionine biosynthesis is catalyzed by the metA gene product, homoserine transsuccinylase (HTS, or homoserine O-succinyltransferase). Recent experiments suggest that transcription of these genes is indeed regulated by MetJ, although the repressor-binding sites have not yet been verified. Methionine also serves as the precursor of S-adenosylmethionine, which is an essential molecule employed in numerous biological processes. S-adenosylhomocysteine is produced as a consequence of the numerous AdoMet-dependent methyl transfer reactions that occur within the cell. In E. coli and Salmonella, this molecule is recycled in two discrete steps to complete the methyl cycle. Cultures challenged by oxidative stress appear to experience a growth limitation that depends on methionine levels. E. coli that are deficient for the manganese and iron superoxide dismutases (the sodA and sodB gene products, respectively) require the addition of methionine or cysteine for aerobic growth. Modulation of methionine levels in response to stressful conditions further increases the complexity of its regulation.

Binding of MetJ repressor to specific and nonspecific DNA and effect of S-adenosylmethionine on these interactions

We have used analytical ultracentrifugation to characterize the binding of the methionine repressor protein, MetJ, to synthetic oligonucleotides containing zero to five specific recognition sites, called metboxes. For all lengths of DNA studied, MetJ binds more tightly to repeats of the consensus sequence than to naturally occurring metboxes, which exhibit a variable number of deviations from the consensus. Strong cooperative binding occurs only in the presence of two or more tandem metboxes, which facilitate protein-protein contacts between adjacent MetJ dimers, but weak affinity is detected even with DNA containing zero or one metbox. The affinity of MetJ for all of the DNA sequences studied is enhanced by the addition of SAM, the known cofactor for MetJ in the cell. This effect extends to oligos containing zero or one metbox, both of which bind two MetJ dimers. In the presence of a large excess concentration of metbox DNA, the effect of cooperativity is to favor populations of DNA oligos bound by two or more MetJ dimers rather than a stochastic redistribution of the repressor onto all available metboxes. These results illustrate the dynamic range of binding affinity and repressor assembly that MetJ can exhibit with DNA and the effect of the corepressor SAM on binding to both specific and nonspecific DNA.

Signatures of protein-DNA recognition in free DNA binding sites

One obstacle to achieving complete understanding of the principles underlying sequence-dependent recognition of DNA is the paucity of structural data for DNA recognition sequences in their free (unbound) state. Here, we carried out crystallization screening of 50 DNA duplexes containing cognate protein binding sites and obtained new crystal structures of free DNA binding sites for three distinct modes of DNA recognition: anti-parallel beta strands (MetR), helix-turn-helix motif + hinge helices (PurR), and zinc fingers (Zif268). Structural changes between free and protein-bound DNA are manifested differently in each case. The new DNA structures reveal that distinctive sequence-dependent DNA geometry dominates recognition by MetR, protein-induced bending of DNA dictates recognition by PurR, and deformability of DNA along the A-B continuum is important in recognition by Zif268. Together, our findings show that crystal structures of free DNA binding sites provide new information about the nature of protein-DNA interactions and thus lend insights towards a structural code for DNA recognition.

Structural basis for the differential regulation of DNA by the methionine repressor MetJ

The Met regulon in Escherichia coli encodes several proteins responsible for the biosynthesis of methionine. Regulation of the expression of most of these proteins is governed by the methionine repressor protein MetJ and its co-repressor, the methionine derivative S-adenosylmethionine. Genes controlled by MetJ contain from two to five sequential copies of a homologous 8-bp sequence called the metbox. A crystal structure for one of the complexes, the repressor tetramer bound to two metboxes, has been reported (Somers, W. S., and S. E. Phillips (1992) Nature 359, 387-393), but little structural work on the larger assemblies has been done presumably because of the difficulties in crystallization and the variability in the number and sequences of metboxes for the various genes. Small angle neutron scattering was used to study complexes of MetJ and S-adenosylmethionine with double-stranded DNA containing two, three, and five metboxes. Our results demonstrate that the crystal structure of the two-metbox complex is not the native solution conformation of the complex. Instead, the system adopts a less compact conformation in which there is decreased interaction between the adjacent MetJ dimers. Models built of the higher order complexes from the scattering data show that the three-metbox complex is organized much like the two-metbox complex. However, the five-metbox complex differs significantly from the smaller complexes, providing much closer packing of the adjacent MetJ dimers and allowing additional contacts not available in the crystal structure. The results suggest that there is a structural basis for the differences observed in the regulatory effectiveness of MetJ for the various genes of the Met regulon.

Kinetic analysis of operator binding by the E. coli methionine repressor highlights the role(s) of electrostatic interactions

MetJ is a member of the ribbon-helix-helix class of DNA-binding proteins whose affinity for operators is apparently controlled by an unprecedented long-range electrostatic effect from the tertiary sulphur atom of its co-repressor, S-adenosyl methionine. We report here the results of kinetic assays of DNA binding with MetJ mutant proteins having altered net charges. The results (a) suggest that MetJ locates its operators via a sliding mechanism, (b) support the idea that electrostatic steering is important in the initial DNA binding event and (c) highlight the sensitivity of this system to electrostatic effects.

Structural and functional studies of an intermediate on the pathway to operator binding by Escherichia coli MetJ

We present the results of in vitro DNA-binding assays for a mutant protein (Q44K) of the E. coli methionine repressor, MetJ, as well as the crystal structure at 2.2 A resolution of the apo-mutant bound to a 10-mer oligonucleotide encompassing an 8 bp met-box sequence. The wild-type protein binds natural operators co-operatively with respect to protein concentration forming at least a dimer of repressor dimers along operator DNAs. The minimum operator length is thus 16 bp, each MetJ dimer interacting with a single met-box site. In contrast, the Q44K mutant protein can also bind stably as a single dimer to 8 bp target sites, in part due to additional contacts made to the phosphodiester backbone outside the 8 bp target via the K44 side-chains. Protein-protein co-operativity in the mutant is reduced relative to the wild-type allowing the properties of an intermediate on the pathway to operator site saturation to be examined for the first time. The crystal structure of the decamer complex shows a unique conformation for the protein bound to the single met-box site, possibly explaining the reduced protein-protein co-operativity. In both the extended and minimal DNA complexes formed, the mutant protein makes slightly different contacts to the edges of DNA base-pairs than the wild-type, even though the site of amino acid substitution is distal from the DNA-binding motif. Quantitative binding assays suggest that this is not due to artefacts caused by the crystallisation conditions but is most likely due to the relatively small contribution of such direct contacts to the overall binding energy of DNA-protein complex formation, which is dominated by sequence-dependent distortions of the DNA duplex and by the protein-protein contact between dimers.

Direct and indirect readout in mutant Met repressor-operator complexes

BACKGROUND

The methionine repressor, MetJ, represses the transcription of genes involved in methionine biosynthesis by binding to arrays of two to five adjacent copies of an eight base-pair 'metbox' sequence. Naturally occurring operators differ from the consensus sequence to a greater extent as the number of metboxes increases. MetJ, while accommodating this sequence variation in natural operators, is very sensitive to particular base changes, even where bases are not directly contacted in the crystal structure of a complex formed between the repressor and consensus operator.

RESULTS

Here we report the high-resolution structure of a MetJ mutant, Q44K, bound to the consensus operator sequence (Q44Kwt19) and two related sequences containing mutations at sites believed to be important for indirect readout at non-contacted bases. The overall structure of the Q44Kwt19 complex is very similar to the wild-type complex, but there are small variations in sugar-phosphate backbone conformation and direct contacts to the DNA bases. The mutant complexes show a mixture of direct and indirect readout of sequence variations, with differences in direct contacts and DNA conformation.

CONCLUSIONS

Comparison of the wild-type and mutant repressor-operator complexes shows that the repressor makes sufficiently strong interactions with the sugar-phosphate backbone to accommodate some variation in operator sequence with minor changes in direct bases contacts. The reduction in repressor affinity for the two mutant repressor complexes can be partially attributed to a loss in direct contacts to the DNA. In one case, however, the replacement of a flexible TA base-step leads to an unfavourable DNA conformation that reduces the stability of the repressor-operator complex.

The structure of plasmid-encoded transcriptional repressor CopG unliganded and bound to its operator

The structure of the 45 amino acid transcriptional repressor, CopG, has been solved unliganded and bound to its target operator DNA. The protein, encoded by the promiscuous streptococcal plasmid pMV158, is involved in the control of plasmid copy number. The structure of this protein repressor, which is the shortest reported to date and the first isolated from a plasmid, has a homodimeric ribbon-helix-helix arrangement. It is the prototype for a family of homologous plasmid repressors. CopG cooperatively associates, completely protecting several turns on one face of the double helix in both directions from a 13-bp pseudosymmetric primary DNA recognition element. In the complex structure, one protein tetramer binds at one face of a 19-bp oligonucleotide, containing the pseudosymmetric element, with two beta-ribbons inserted into the major groove. The DNA is bent 60 degrees by compression of both major and minor grooves. The protein dimer displays topological similarity to Arc and MetJ repressors. Nevertheless, the functional tetramer has a unique structure with the two vicinal recognition ribbon elements at a short distance, thus inducing strong DNA bend. Further structural resemblance is found with helix-turn-helix regions of unrelated DNA-binding proteins. In contrast to these, however, the bihelical region of CopG has a role in oligomerization instead of DNA recognition. This observation unveils an evolutionary link between ribbon-helix-helix and helix-turn-helix proteins.

A comprehensive library of DNA-binding site matrices for 55 proteins applied to the complete Escherichia coli K-12 genome

A major mode of gene regulation occurs via the binding of specific proteins to specific DNA sequences. The availability of complete bacterial genome sequences offers an unprecedented opportunity to describe networks of such interactions by correlating existing experimental data with computational predictions. Of the 240 candidate Escherichia coli DNA-binding proteins, about 55 have DNA-binding sites identified by DNA footprinting. We used these sites to construct recognition matrices, which we used to search for additional binding sites in the E. coli genomic sequence. Many of these matrices show a strong preference for non-coding DNA. Discrepancies are identified between matrices derived from natural sites and those derived from SELEX (Systematic Evolution of Ligands by Exponential enrichment) experiments. We have constructed a database of these proteins and binding sites, called DPInteract (available at http://arep.med.harvard.edu/dpinteract).

Dissecting the molecular details of prokaryotic transcriptional control by surface plasmon resonance: the methionine and arginine repressor proteins

The commercial surface plasmon resonance (SPR) biosensors, BIACORE, have been used to investigate the molecular details of macromolecular interactions at prokaryotic promoter-operators. For the Escherichia coli methionine repressor, MetJ, we have quantitated the interaction of the protein with synthetic and natural operator sites and shown that the SPR response is directly related to the stoichiometry of the complexes being formed. The utility of a continuous flow system has also been exploited to investigate transcription from an immobilised promoter-operator fragment; with transcripts collected and subsequently characterised by RT-PCR. This technique has enabled us to investigate how repressor binding affects (i) the interaction of the RNA polymerase (RNAP) with the promoter and (ii) the ability of RNAP to initiate transcription. Remarkably, the repression complex appears to stabilise binding of RNAP, whilst having the expected effects on the levels of transcripts produced. This may well be a general mechanism allowing rapid transcription initiation to occur as soon as the repression complex dissociates. These techniques have also been used to examine protein-DNA interactions in the E. coli and Bacillus subtilis arginine repressor systems. The repressors are the products of the argR and ahrC genes, respectively. Both proteins form hexamers in rapid equilibrium with smaller subunits believed to be trimers. There are three types of operator in these systems, autoregulatory, biosynthetic and catabolic (B. subtilis only). Sensorgrams show that each protein recognises the three types of immobilised operator differently and that binding is stimulated over 100-fold by the presence of L-arginine.

Quantitation of the Escherichia coli methionine repressor-operator interaction by surface plasmon resonance is not affected by the presence of a dextran matrix

The effect of a dextran matrix on the apparent rate constants measured for the interaction of the Escherichia coli methionine repressor, MetJ, with its immobilized consensus operator has been studied using surface plasmon resonance (SPR) in a commercial biosensor, BIACORE (Biacore AB). Based on the results of computer simulations, it has been proposed that such data can deviate from the expected simple kinetic behavior due to effects generated by the dextran matrix, used at the biosensor surface to anchor one of the interacting molecules. We have tested this possibility experimentally by measuring the apparent rate constants for the interaction of MetJ with its operator DNA on sensor chip surfaces with no dextran matrix or having matrices 30 or 100 nm thick. The data show that for the MetJ-operator interaction, the dextran matrix has no significant effect on the apparent rate constants measured and that comparative measurements using this technique are informative.