This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ghochikyan, A.
Right arrow Articles by Sakanyan, V.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ghochikyan, A.
Right arrow Articles by Sakanyan, V.

 Previous Article  |  Next Article 

Journal of Bacteriology, December 2002, p. 6602-6614, Vol. 184, No. 23
0021-9193/02/$04.00+0     DOI: 10.1128/JB.184.23.6602-6614.2002
Copyright © 2002, American Society for Microbiology. All Rights Reserved.

Arginine Operator Binding by Heterologous and Chimeric ArgR Repressors from Escherichia coli and Bacillus stearothermophilus

Anahit Ghochikyan,1 Iovka Miltcheva Karaivanova,1 Michèle Lecocq,1 Patricia Vusio,2 Marie-Claire Arnaud,1 Marina Snapyan,1 Pierre Weigel,1 Laetitia Guével,1 Malcolm Buckle,3,{dagger} and Vehary Sakanyan1,4*

Laboratoire de Biotechnologie, FRE CNRS 2230, Unité Biocatalyse, Faculté des Sciences et des Techniques, Université de Nantes, 44322 Nantes,1 IFR 26, INSERM, 44035 Nantes,2 Laboratoire d'Enzymologie et Cinétique Structurale, UMR 8532 du CNRS, LBPA, Ecole Normale Supérieure de Cachan, 94235 Cachan,3 ProtNeteomix, Université de Nantes, 44322 Nantes, France4

Received 23 May 2002/ Accepted 27 August 2002


arrow
ABSTRACT
 
Bacillus stearothermophilus ArgR binds efficiently to the Escherichia coli carAB operator, whereas the E. coli repressor binds very poorly to the argCo operator of B. stearothermophilus. In order to elucidate this contradictory behavior between ArgRs, we constructed chimeric proteins by swapping N-terminal DNA-binding and C-terminal oligomerization domains or by exchanging the linker peptide. Chimeras carrying the E. coli DNA-binding domain and the B. stearothermophilus oligomerization domain showed sequence-nonspecific rather than sequence-specific interactions with arg operators. Chimeras carrying the B. stearothermophilus DNA-binding domain and E. coli oligomerization domain exhibited a high DNA-binding affinity for the B. stearothermophilus argCo and E. coli carAB operators and repressed the reporter-gene transcription from the B. stearothermophilus PargCo control region in vitro; arginine had no effect on, and indeed even decreased, their DNA-binding affinity. With the protein array method, we showed that the wild-type B. stearothermophilus ArgR and derivatives of it containing only the exchanged linker from E. coli ArgR or carrying the B. stearothermophilus DNA-binding domain along with the linker and the {alpha}4 regions were able to bind argCo containing the single Arg box. This binding was weaker than binding to the two-box operator but was no longer arginine dependent. Several lines of observations indicate that the {alpha}4 helix in the oligomerization domain and the linker peptide can contribute to the recognition of single or double Arg boxes and therefore to the operator DNA-binding specificity in similar but not identical ArgR repressors from two distant bacteria.


arrow
INTRODUCTION
 
Recent genome comparative analysis has revealed a global vision of arg genes transcription regulation in microorganisms (3, 23, 26, 30). However, experimental studies on distant organisms are required to elucidate the molecular mechanisms and to understand the evolutionary principles established for arginine regulatory systems in various microbes.

A substantial amount of functional and structural information regarding transcription regulation has accumulated for the Escherichia coli arginine repressor, ArgR (14, 22, 40). In this organism, the repressor, in cooperation with L-arginine, governs expression of the arginine biosynthesis genes (arg) and carAB genes, coding for carbamoylphosphate synthetase (EC 6.3.5.5), providing the carbamoylphosphate required for the synthesis of both arginine and pyrimidine residues. The ArgR protein consists of N-terminal DNA-binding and C-terminal oligomerization domains connected by a short protease-sensitive linker (15), and a three-dimensional structure has been separately resolved for each domain (41, 45). A winged helix-turn-helix (wHTH) motif of the DNA-binding domain (41) recognizes a 40-bp sequence which comprises two adjacent imperfect 18-bp palindromes, known as Arg boxes, that are separated by a 3-bp spacer in the majority of cognate operators or by a 2-bp spacer in the argR operator (6, 44, 46). ArgR monomers associate spontaneously to form trimers, and arginine, as a ligand, binds distinct amino acids located within the oligomerization domain and provides the transition of two identical apo-trimers to a holohexameric molecule (45).

Transcription of the arginine biosynthesis genes is also repressed in the moderate thermophilic bacterium Bacillus stearothermophilus (32, 47). The ArgR repressor binds a 42-bp operator comprising two Arg box-like sequences separated by a 2-bp spacer and overlapping the PargC promoter located upstream of the argCJBD operon (10, 35). The three-dimensional structure of a full-length aporepressor and arginine-bound C-terminal domain of B. stearothermophilus ArgR has recently been resolved (28). The DNA-binding and oligomerization domains of the E. coli and B. stearothermophilus repressors adapt similar folds despite the fact that these proteins show only 27% amino acid sequence similarity. Furthermore, arginine binding to distinct amino acids (four of the six residues involved are conserved in the oligomerization domain of both ArgR proteins; Fig. 1) leads to allosteric changes in both hexameric repressors (28, 41, 45), thereby increasing their arg operator DNA-binding affinity (5, 6, 8, 10, 21, 42, 43, 46). It has been shown that binding six arginine molecules, in addition to reinforcing interactions between monomers within a trimer and between trimers within a hexamer, also provokes a 15° rotation of two of the trimers with respect to each other (28). Such an allosteric modification appears to expose four of the six wHTH modules properly with respect to four halves of both Arg box palindromes and therefore improve the operator-binding affinity (28).



View larger version (55K):
[in this window]
[in a new window]
  		FIG. 1. Amino acid sequences of wild-type and domain- and linker-exchanged chimeric proteins constructed from E. coli and B. stearothermophilus ArgR repressors (a) and nucleotide sequences of E. coli carAB and B. stearothermophilus argCo operators (b). The amino acid sequence for E. coli is shown in lightface, and that for B. stearothermophilus is shown in boldface. Secondary-structure features of the wild-type ArgR repressors from E. coli (41, 45) and B. stearothermophilus (28) are shown above and below the corresponding sequences, respectively. The DNA-binding (amino acids 1 to 64) and oligomerization (amino acids 73 to 149) domains of B. stearothermophilus ArgR are connected by an 8-amino-acid-long linker peptide (28). The structure between the ß2 and ß3 sheets has not yet been determined for E. coli ArgR. The amino acid residues involved in arginine binding are underlined in the wild-type repressor sequences. The C-terminal His tag is shown in lowercase letters. Dashed brackets indicate operator sequences protected against DNase I cleavage upon wild-type repressor binding (35, 46). Spacer nucleotides between two Arg boxes are shown in lowercase letters. Dashed-line arrows indicate the sites for amplification of the argCo operator DNAs with double (76-bp) and single (56-bp) Arg boxes.

The similarity of protein and operator organization suggests a common DNA recognition mechanism for the mesophilic E. coli and thermophilic B. stearothermophilus ArgR repressors. There is, however, a major difference between the DNA-binding properties of the two ArgR proteins. The E. coli repressor has a very low affinity for the argCo operator of B. stearothermophilus in vitro and is practically unable to repress transcription from the B. stearothermophilus PargCo promoter-operator region in E. coli host cells (32, 33). On the contrary, the B. stearothermophilus repressor binds E. coli operators with high efficiency (19, 32, 35). Furthermore, the overexpressed wild-type B. stearothermophilus ArgR repressor, unlike the E. coli protein (44), behaves as a superrepressor in E. coli host cells, which become auxotrophic for arginine due to repression of arginine and probably other biosynthesis genes, suggesting that it is unable to recognize a wider range of operator sequences (19).

In an effort to understand why two similar ArgR proteins possess divergent DNA-binding specificities, we constructed chimeric proteins comprising various regions of the E. coli and B. stearothermophilus repressors and analyzed their DNA-binding properties. We show that, depending on the presence of a substituted domain or a linker peptide, ArgR chimeric proteins differ in their ability to bind to cognate arg operators, in solubility, and in electrophoretic migration. Our data indicate that the B. stearothermophilus repressor is able to bind to the argCo operator comprising a single Arg box and that the {alpha}4 helix in the oligomerization domain is a major structural determinant contributing to the operator-binding specificity.


arrow
MATERIALS AND METHODS
 
Bacterial strains, plasmids, and growth conditions. The bacterial strains and plasmids used in this study are described in Table 1. The strain B. stearothermophilus NCIB 8224 was grown with aeration at 60°C in a liquid Luria-Bertani (LB) medium (24). E. coli strains were grown with aeration at 28°C or 37°C on LB medium with appropriate antibiotics: ampicillin, 100 µg/ml, or kanamycin, 25 µg/ml.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Bacterial strains and plasmids used in this study

To express His-tagged wild-type and chimeric ArgR proteins, the protease-deficient E. coli BL21(DE3) strain carrying a plasmid with a wild-type or chimeric argR gene was grown in LB with ampicillin at 28°C to an A600 of 0.8. Synthesis of T7 RNA polymerase in this strain was induced by addition of isopropylthiogalactopyranoside (IPTG) at a final concentration of 1 mM. The cells were harvested 3 h later by centrifugation and used for protein purification. Liquid M9 medium (24) was used to assess the arginine auxotrophy of E. coli strain Top10 harboring a pCR-Blunt vector derivative carrying a wild-type or chimeric argR gene by measuring the optical density of corresponding cultures after a 48-h incubation at 37°C.

Construction of recombinant DNAs and DNA labeling by PCR. The DNA-binding and oligomerization domains of ArgR are separated by an 8- and 5-amino-acid-long linker peptide in the B. stearothermophilus and E. coli repressors, respectively. Four domain-exchanged chimeric proteins were constructed in which the DNA-binding domain with the corresponding linker was taken from the same wild-type repressor or the linker peptide and oligomerization domain were taken from the same repressor (see Fig. 1).

The chimeric argR genes were constructed in two consecutive PCR steps by the overlap extension method (16). At the first PCR step, two separate DNA fragments corresponding to the N-terminal and C-terminal domains of the ArgR repressors were amplified on a chromosomal DNA template from the B. stearothermophilus NCIB 8224 and E. coli XA4 strains by the creation of overlapping sequences at expected junctions. The amplified DNA fragments were then combined in a subsequent fusion PCR, and the full-length argR hybrid DNAs were obtained with only two flanking primers, carrying in addition an NcoI or XhoI restriction site.

The oligonucleotide primer sequences used are described in Table 2. The amplified DNA fragments were treated with NcoI and XhoI and inserted into the pET21d(+) vector digested with the same enzymes in order to create a hybrid argR gene fused in frame with six histidine codons (C-terminal His tag). The resulting ArgR chimeric proteins (see Fig. 1) contained (i) ArgR-EN1, a 68-amino-acid N-terminal portion fused to the next C-terminal 64 to 149 amino acids from the E. coli and B. stearothermophilus repressors, respectively; (ii) ArgR-EN2, an 87-amino-acid N-terminal portion fused to the next C-terminal 86 to 149 amino acids from the E. coli and B. stearothermophilus repressors, respectively; (iii) ArgR-BN1, a 63-amino-acid N-terminal portion fused to the next C-terminal 69 to 156 amino acids from the B. stearothermophilus and E. coli repressors, respectively; and (iv) ArgR-BN2, an 85-amino-acid N-terminal portion fused to the next C-terminal 88 to 156 amino acids from the B. stearothermophilus and E. coli repressors, respectively.


View this table:
[in this window]
[in a new window]
  		TABLE 2. Oligonucleotide primers used for cloning or construction of wild-type and chimeric argR genes

Two other chimeric derivatives, ArgR-EbE and ArgR-BeB, were also constructed by mutually replacing the linker peptide in the wild-type E. coli and B. stearothermophilus His-tagged ArgR repressors (see Fig. 1). The pArgR-EN1 and pArgR-BN1 plasmids carrying chimeric argR genes and wild-type chromosomal argR genes were used as templates in these PCR amplifications with the corresponding oligonucleotides (Table 3).


View this table:
[in this window]
[in a new window]
  		TABLE 3. Rate and equilibrium constants of wild-type ArgR repressors and domain-exchanged and linker-exchanged chimeras for carAB and argCo operatorsa

For the mobility shift assay, two DNA fragments were synthesized: a 137-bp DNA fragment carrying the argC operator from B. stearothermophilus NCIB 8224 was amplified with nonlabeled 5'-GGCTGCCGGGACAAATCGG and digoxigenin-labeled 5'-CCCGTATGCCTCATGTAG oligonucleotide primers, and a 141-bp DNA fragment carrying the carAB operator (46) from E. coli XA4 was amplified with nonlabeled 5'-ACGTCATCATTGTGAATTAA and digoxigenin-labeled 5'-CAACCGCCGAACCTGTTG primers. The same operator DNA was labeled at a downstream terminus with biotin for surface plasmon resonance studies.

For testing protein-DNA interactions by the protein chip method, we used two probes, a 76-bp DNA carrying the entire two-Arg-box operator region and a 56-bp DNA carrying only the downstream Arg box of the B. stearothermophilus argCo operator region (see Fig. 1). Both DNA probes were amplified by PCR with 5'-CCTCGAAAATTATTAAAT or 5'-ACATTTGATTTTATTTTTATAC upstream primers with a 5'-CCCGTATGCCTCATGTAG downstream primer labeled at the first position with IRDye800. Concentrations of labeled DNAs were measured with a UV/VIS spectrometer (Perkin Elmer) and by comparison of fluorescent DNA bands in an agarose gel.

Overlap extension was carried out with Pfu DNA polymerase (Stratagene), whereas other PCR amplifications were performed with Taq DNA polymerase (Qiagen). The nucleotide sequences of domain- and linker-exchanged ArgR chimeras were verified by automatic sequencing.

Oligonucleotide primers were purchased from MWG Biotech.

Purification and molecular mass determination of His-tagged proteins. His-tagged wild-type and chimeric ArgR proteins were purified with minor modifications as follows. Bacterial cells were suspended in a buffer (50 mM NaH2PO4 [pH 8.0], 300 mM NaCl, 10 mM imidazole) and sonicated, and cell extracts were subjected to affinity chromatography on a nickel-nitrilotriacetic acid column (Qiagen). The column was equilibrated and washed with the buffer defined above, and ArgR proteins were eluted with the same buffer containing 250 mM imidazole. Protein samples of an E. coli wild-type ArgR, ArgR-BN1, and ArgR-BN2 were dialyzed against a low-salt-concentration buffer containing 20 mM Tris-HCl (pH 7.5), 10 mM MgCl2, and 10 mM ß-mercaptoethanol, whereas those of a B. stearothermophilus wild-type ArgR, ArgR-EN1, ArgR-EN2, ArgR-BeB, and ArgR-EbE were dialyzed against a high-salt-concentration buffer of the same composition except that 1 M NaCl was used.

Molecular masses of His-tagged ArgR proteins were determined by size exclusion chromatography on a Sephacryl S200 column (Amersham Pharmacia). Columns were calibrated with molecular mass markers (Amersham Pharmacia), consisting of bovine serum albumin (67 kDa), ovalbumin (43 kDa), chymotrypsin A (25 kDa), and RNase A (13.7 kDa). Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) of proteins was performed as described by Ausubel et al. (2). Protein concentration was measured by the described method (4) with bovine serum albumin as the standard.

Computer-assisted characterization of proteins was conducted with the MacDNAsis V3.6 program (Hitachi Software).

Mobility shift assay. Binding reactions were performed in 10 mM Tris-HCl (pH 7.5)-250 mM KCl-5 mM MgCl2-2.5 mM CaCl2-2.5% glycerol-0.5 mM dithiothreitol-10 mM L-arginine for the wild-type B. stearothermophilus ArgR, ArgR-EN1, and ArgR-EN2 chimeric proteins and in 10 mM Tris-HCl (pH 7.5)-100 mM KCl-10 mM MgCl2-2.5% glycerol-0.5 mM dithiothreitol-10 mM L-arginine for the wild-type E. coli ArgR, ArgR-BN1, and ArgR-BN2 chimeric proteins. Alternatively, the binding reaction was carried out in 20 mM Tris-HCl (pH 7.9)-50 mM NaCl-50 mM KCl-0.1 mM dithiothreitol-0.005% surfactant P20 as described previously (27). Binding buffer contained ~0.03 pmol of digoxigenin-labeled operator DNA and a 100-fold excess of unlabeled sonicated herring sperm DNA. The incubation was performed at 37 or 55°C for 30 min.

Samples were loaded on a 2% agarose gel prepared in TAE buffer (40 mM Tris-base, 10 mM sodium acetate, 1 mM EDTA [pH 8.0]) with 10 mM L-arginine and migrated by electrophoresis in the same buffer at room temperature at 12 V.cm-1 for 1 h. The DNA-protein complexes were transferred onto nylon membranes by the capillary method (36), and the immunological detection of bound DNA was carried out with CSPD (disodium 3-(4-methoxyspiro{1,2-dioxetane-3,2'-(5'-chloro)tricyclo[3,3,1,13,7]decan}-4-yl)phenyl phosphate)-mediated luminescence (Boehringer Mannheim, Mannheim, Germany). Quantification of free and retarded DNA in gels was done by scanning densitometry of chemiluminograms (Molecular Analyst; Bio-Rad). A weaker chemiluminescent signal for protein-DNA complexes relative to free DNA could be due to differences in the efficiency of transfer and/or fixation to the membrane. An apparent dissociation constant (Kd monomer equivalent) of protein-DNA complexes formed was estimated by eye as the point at which 50% of the probed DNA remained free (average data from graphic representations of three experiments). Other details were described previously (19).

Surface plasmon resonance. The surface plasmon resonance method was used to study the real-time interaction of wild-type and chimeric ArgR proteins with operator DNAs. The uniquely end-biotinylated DNA fragments corresponding to the B. stearothermophilus argCo and E. coli carAB operators were purified from an excess of primers and deoxynucleoside triphosphates by passage across a spin cartridge containing a silica-based membrane (Life Technologies, Gibco-BRL), and their immobilization was carried out on streptavidin-captured biosensor chips (Biacore AB). Biotinylated operator DNA at concentrations of 5 µg/ml in the binding buffer described above at pH 7.9 was injected over the sensor chip at a flow rate of 5 µl/min at 25°C. Immobilization was controlled manually to about 150 resonance units (RU). Binding assays were carried out in the same buffer with and without 10 mM L-arginine by injection of a wild-type or chimeric ArgR protein at 10 to 100 nM (monomer equivalent) for 2 to 8 min (association phase) at a flow rate of 20 µl/min at 25°C, followed by the injection of a proteinless buffer for 15 min (dissociation phase).

Surface plasmon resonance measurements were conducted in parallel channels with a Biacore 2000 or Biacore 3000 (Biacore AB). The sensor chip was regenerated by washing with 1 M NaCl at a flow rate of 10 µl/min. The channel without immobilized DNA was used as a reference signal for each cycle. Data for binding of ArgR repressors and chimeric proteins were evaluated from sensorgrams with the 1:1 binding model (BIAevaluation Software Handbook, 1999), and Kd values were estimated for the ArgR monomer equivalents. The surface plasmon resonance signal is directly proportional to the mass changes at the sensor chip surface and is expressed in resonance units: 1,000 RU correspond to a surface concentration of approximately 1 ng mm-2 for a globular protein.

Coupled transcription-translation in vitro. The argC gene (coding for N-acetylglutamate-5-semialdehyde dehydrogenase, EC 1.2.1.38) from B. stearothermophilus NCIB 8224 (31) is transcribed from a strong PargC promoter which largely overlaps the argCo operator (32, 33). In preliminary experiments, ArgC protein synthesis from this promoter was found to be rather high in a coupled transcription-translation system of E. coli with a corresponding kit purchased from Promega. However, further experiments were carried out in S30 extracts prepared as described by Chen and Zubay (7) with minor modifications.

Plasmid pHAV2 carrying the PargC-argCo DNA region from B. stearothermophilus (32) was used as a circular DNA template. Protein synthesis was performed at 37°C for 90 min with 10 µCi of L-[35S]methionine (specific activity, 1,000 Ci/mmol, 37 TBq/mmol; Amersham-Pharmacia Biotech). The synthesized products were separated by SDS-PAGE, and the gels were electroblotted on polyvinylidene difluoride membranes (Bio-Rad) or fixed on 3MM paper. Autoradiography was performed after treatment of gels with an Amplifyer solution (Amersham Pharmacia Biotech). Analysis of the synthesized ArgC protein was carried out by densitometric analysis of autoradiograms (BioMax MR film; Kodak) with Molecular Analyst software (Bio-Rad).

Preparation of protein array and fluorescence detection of protein-DNA interactions. Purified His-tagged proteins were dissolved in phosphate-buffered saline with 130 mM NaCl and 10% glycerol and serially fourfold diluted in the same buffer, and 50 to 500 pl was printed with a GMS 417 microarrayer (Affymetrix) onto a BA83 nitrocellulose membrane (Schleicher & Schuell) attached manually to a glass slide. Membranes were incubated in a DNA-binding buffer at pH 7.9 containing 25 µg of sonicated salmon DNA per ml at room temperature for 30 min. Then, an IRDye-labeled DNA probe (10 ng/ml) was added to the solution and incubated for 12 h with slow rotation at room temperature overnight. If necessary, 10 mM L-arginine was included in all buffers. The membranes were washed three times with phosphate-buffered saline, and fluorescent signals from bound DNA-protein complexes were detected at 800 nm with an Odyssey Imager (LI-COR, Inc.). Fluorescent signals were analyzed with GenePix Pro4.0 (Axon Instruments) software.


arrow
RESULTS
 
Rationale for construction of domain-exchanged ArgR chimeras. We were first interested in examining whether the exchanged domains of the E. coli and B. stearothermophilus ArgR repressors might have a similar effect on the function of the chimeric proteins. The availability of three-dimensional structural data on domain boundaries of the B. stearothermophilus arginine repressor (28) allowed us to design an accurate reciprocal domain exchange between the E. coli and B. stearothermophilus proteins by swapping the oligomerization domain along with a corresponding linker peptide from one ArgR to the last amino acid in the DNA-binding domain of the other ArgR. The amino acid sequences of the two resulting ArgR-EN1 and ArgR-BN1 chimeras, carrying the N-terminal DNA-binding domains from E. coli and B. stearothermophilus, respectively, are shown in Fig. 1a.

The linker peptide in E. coli ArgR appears to be 3 amino acid residues shorter than that in the B. stearothermophilus repressor. Therefore, in order to conserve the integrity of an {alpha}-helix sequence with respect to the upstream-located DNA-binding domain sequence, we constructed two more chimeras, ArgR-EN2 and ArgR-BN2, in which the N-terminal regions were taken from the E. coli and B. stearothermophilus repressors, respectively, but extended to the Leu87 and Ile85 residues, respectively, located in a ß3 sheet of the C-terminal oligomerization domain (see Fig. 1a).

Recently, two other ArgR chimeras were constructed by mutual swapping of the DNA-binding and oligomerization domains from the E. coli and B. subtilis arginine repressors (17) with junction sites close to those of the ArgR-EN2 and ArgR-BN2 chimeras described in this study (see Fig. 1). The B. subtilis protein also repressed E. coli arg operators, but not vice versa (34). However, since essential differences exist between B. subtilis and B. stearothermophilus regulatory systems and since different DNA targets have been used (10, 17, 19, 25, 32, 39; this study), it is impossible to extrapolate data obtained from these two studies.

Domain-exchanged chimera binding to operator DNAs. Previously it was shown that the wild-type E. coli ArgR is soluble in a low-salt-concentration buffer (21), whereas the wild-type B. stearothermophilus ArgR is soluble in a high-salt-concentration buffer (10). We observed that both ArgR-BN1 and, to a lesser extent, ArgR-BN2 were optimally soluble in low-salt buffer, as opposed to ArgR-EN1 and ArgR-EN2, which were more soluble in high-salt buffer. Consequently, DNA-binding reactions were performed for each protein in the optimum buffer (see Materials and Methods).

The DNA-binding properties of the four ArgR chimeras constructed and two wild-type repressors were first studied by the mobility shift assay. In the presence of arginine, the wild-type B. stearothermophilus ArgR repressor bound the B. stearothermophilus argCo (Fig. 2b), and the E. coli carAB (Fig. 3d) operators at 37°C with similar efficiencies (an apparent dissociation constant [Kd] was ~100 nM, close to the value observed for the untagged protein). The wild-type E. coli repressor bound the carAB operator with an apparent Kd of close to 120 nM (Fig. 3a), whereas it bound the B. stearothermophilus argCo operator with an apparent Kd in excess of 1 µM (Fig. 2a). Thus, in the presence of arginine, the DNA affinity of the wild-type E. coli repressor was more than 50 times lower for the heterologous argCo operator than for its homologous carAB operator under the conditions tested.



View larger version (87K):
[in this window]
[in a new window]
  		FIG. 2. Mobility shift assay of wild-type ArgR from E. coli (a) and B. stearothermophilus (b) and chimeric proteins ArgR-BN1 (c) and ArgR-BN2 (d) bound to B. stearothermophilus argCo operator DNA in the presence of arginine. argCo operator DNA was incubated with increasing amounts of His-tagged proteins at 37°C for 30 min and electrophoresed in a 2% agarose gel in the presence of arginine. (a) Lanes 2 to 5 contained 170 nM, 340 nM, 680 nM, and 1,360 nM E. coli ArgR, respectively; (b) lanes 2 to 5 contained 86 nM, 102 nM, 118 nM, and 130 nM B. stearothermophilus ArgR, respectively; (c) lanes 2 to 4 contained 15 nM, 25 nM, and 40 nM ArgR-BN1, respectively; (d) lanes 1 to 4 contained 12 nM, 24 nM, 48 nM, and 96 nM ArgR-BN2, respectively. Lane 1 in a, b, and c and lane 5 in d contained only operator DNA.



View larger version (63K):
[in this window]
[in a new window]
  		FIG. 3. Mobility shift assay of wild-type ArgRs of E. coli (a) and B. stearothermophilus (d) and domain-exchanged chimeric proteins ArgR-BN1 (b), ArgR-BN2 (c), and ArgR-EN2 (e) bound to E. coli carAB operator DNA in the presence of arginine. carAB operator DNA was incubated with increasing amounts of His-tagged proteins at 37°C for 30 min and electrophoresed in a 2% agarose gel in the presence of arginine. (a) Lanes 2 to 4 contained 12.4 nM, 24.8 nM, and 37.2 nM of E. coli ArgR protein, respectively; (b) lanes 2 to 5 contained 2.1 nM, 5.3 nM, 7.4 nM, and 9.5 nM ArgR-BN1, respectively; (c) lanes 2 to 5 contained 6 nM, 12 nM, 18 nM, 24 nM, and 48 nM ArgR-BN2, respectively; (d) lanes 2 to 5 contained 26 nM, 52 nM, 104 nM, and 208 nM B. stearothermophilus ArgR, respectively; (e) lanes 2 to 4 contained 600 nM, 1,200 nM, and 1,900 nM ArgR-EN2, respectively. Lane 1 in a, b, c, d, and e contained only operator DNA.

No binding was detected for the ArgR-EN1 chimeric protein to either of the two operators. Binding was also not detected between ArgR-EN2 and the argCo operator, and binding to the carAB operator was very weak (Fig. 3e). Two other chimeras, ArgR-BN1 and ArgR-BN2, both carrying the N-terminal domain from B. stearothermophilus ArgR, bound efficiently to the argCo (Fig. 2c and 2d) and carAB operators (Fig. 3b and 3c). Apparent Kd values were calculated to be 29 nM and 5 nM, respectively, for ArgR-BN1 and 34 nM and 11 nM, respectively, for ArgR-BN2 for these operators. However, when binding reactions were carried out at 55°C, no retarded bands could be detected for ArgR-BN1 or ArgR-BN2, indicating that these chimeras did not gain in thermostability compared to the wild-type B. stearothermophilus ArgR (10).

Karaivanova et al. previously described that the wild-type B. stearothermophilus ArgR repressor and the argCo operator DNA-bound complexes barely enter a nondenaturing agarose gel (19) (Fig. 2b). A similar behavior was observed for wild-type B. stearothermophilus ArgR-carAB complexes but not for E. coli ArgR-carAB complexes. ArgR-BN1 and ArgR-BN2 bound to carAB or argCo complexes also migrated as compact retarded bands (see Fig. 2 and 3). Therefore, different migration of chimeric protein-DNA complexes appeared to be related to the oligomerization domain of the corresponding parent repressors rather than to target DNA particularities. Indeed, the deduced isoelectric points of His-tagged proteins were found to be 5.6, 5.2, and 5.5 for wild-type E. coli ArgR, ArgR-BN1, and ArgR-BN2, respectively, whereas they were 8.0, 8.2, and 7.3 for wild-type B. stearothermophilus ArgR, ArgR-EN1, and ArgR-EN2, respectively. Consequently, their electrophoretic migration in a 6% nondenaturing polyacrylamide gel was in accordance with their deduced isoelectric point values, i.e., proteins possessing the basic oligomerization domain of B. stearothermophilus ArgR migrated slowly (data not shown).

The mobility shift assay detects DNA-protein interactions at equilibrium, although this analysis may be restricted by desalting and cage effects that modify off and on rates in an unpredictable fashion. The use of surface plasmon resonance allows access to rate constants for the on and off processes under more constant buffer conditions and also allows a comparison between apparent Kd values obtained either by the ratio of apparent kinetic constants or by the half-saturation concentrations at steady state (29). Therefore, we used the latter method to compare apparent kinetic constants characteristic of wild-type repressor and chimeric protein interactions to operator DNAs. Binding reactions were carried out in a buffer at pH 7.9.

Both the E. coli and B. stearothermophilus wild-type repressors exhibited arginine dependence for carAB operator binding, but the shapes of the corresponding sensorgrams differed significantly. As can be seen in Fig. 4, the association phase for E. coli ArgR stabilized rapidly after a 2-min injection, whereas the B. stearothermophilus ArgR association peak was rather high for both the carAB and argCo operators after short injections but decreased gradually during longer injections. A similar picture was observed for B. stearothermophilus ArgR interactions in the absence of arginine. Therefore, assuming a possible stabilizing effect of the prolonged injection on the formation of the B. stearothermophilus ArgR protein-DNA complexes, we performed further binding reactions for 8 min (association phase), followed by the injection of a proteinless buffer for 15 min (dissociation phase).



View larger version (20K):
[in this window]
[in a new window]
  		FIG.4. Sensorgrams of binding of wild-type E. coli and B. stearothermophilus ArgR repressors to operator DNAs. The curves represent binding of 50 nM wild-type (wt) ArgR to the 141-bp carAB operator DNA of E. coli (a and b) and the 137-bp argCo operator DNA of B. stearothermophilus (c) 2 min, 4 min, 6 min, and 8 min after injection of the protein. The signals were monitored for E. coli ArgR in the presence of arginine (a) and for B. stearothermophilus in the presence of arginine (thick lines) and in the absence of arginine (thin lines) in b and c. No binding of E. coli ArgR to carAB was detected in the absence of arginine or to argCo irrespective of the presence of arginine.

As expected, the wild-type E. coli ArgR displayed strong binding to the carAB operator in the presence of arginine, whereas no binding could be detected in the absence of arginine, and the protein appeared to be unable to bind the argCo operator irrespective of the presence of arginine (all surface plasmon resonance sensorgrams are available from the authors upon request). The wild-type B. stearothermophilus ArgR repressor bound the carAB and argCo operators in the absence of arginine, but arginine improved the binding to both operators. Quantitative analysis of sensorgrams showed that the wild-type B. stearothermophilus ArgR repressor bound the carAB and argCo operators in the absence of arginine, but arginine improved its affinity nearly fourfold (Table 3). On the other hand, wild-type E. coli ArgR repressor binding to the carAB operator was found to be strongly dependent on arginine; even in the presence of arginine, the E. coli protein bound to the heterologous argCo operator with a 1,000-fold lower affinity than to the homologous operator, a greater relative difference than was detected by the mobility shift assay.

Resonance signals were detected for all domain-exchanged chimeric proteins when passed over sensor chips with coupled carAB or argCo operator DNAs. In the absence of arginine, the ArgR-BN1 and ArgR-BN2 chimeras bound the carAB and argCo operators with a higher affinity than wild-type B. stearothermophilus ArgR did under the same conditions. When chimeric proteins were passed over the immobilized carAB or argCo operator DNA in the presence of arginine, ArgR-BN1 and especially ArgR-BN2 binding affinities were lower for both operators compared to those in the absence of arginine (see Table 3). Sensorgrams with injected ArgR-EN1 and ArgR-EN2 chimeras indicated a possible interaction with both operators in the presence or absence of arginine. However, fits with a simple single-binding model used to calculate apparent rate constants were poor (for example, a standard {chi}2 value was in excess of 1,000 in two independent experiments). Therefore, we concluded that sensorgrams of these chimeras reflect a more complex binding mode that could not be accessed for meaningful and comparable rate constants.

Thus, the surface plasmon resonance method showed that wild-type and domain-exchanged chimeras have different kinetics of interactions with operator DNAs and that arginine differentially affects these interactions.

To clarify the oligomeric nature of the chimeras, we determined their molecular masses by gel filtration chromatography. In the absence of arginine, two peaks were eluted for ArgR-BN1 and ArgR-BN2. A major peak was close to 41 ± 5 kDa, and another peak was approximately 200 kDa. On the basis of these experiments, no clear distinction could be made between dimeric and trimeric molecules for the major peak, whereas the second peak probably reflects aggregated proteins under these conditions. In the presence of arginine, a peak of 100 ± 8 kDa was additionally detected for purified ArgR-BN1 and ArgR-BN2 protein samples, which is consistent with the appearance of hexameric molecules. No clear peaks could be detected for ArgR-EN1 and ArgR-EN2, suggesting their tendency to aggregate and indicating that the resolution of the gel filtration technique is insufficient to determine the apparent molecular masses of these chimeric proteins.

ArgR-BN1 and ArgR-BN2 chimeras provide gene repression. To assess whether the wild-type ArgR and domain-exchanged chimeric proteins were able to mediate gene repression, we evaluated argC reporter gene expression from the PargCo promoter-operator region in a coupled transcription-translation system with S30 extracts of E. coli (arginine was present in the reaction mixture). An abundant band of the ArgC protein could still be detected after addition of 10 pmol of wild-type E. coli ArgR (Fig. 5a), confirming a weak repression effect of this protein on the heterologous promoter. No repression of ArgC synthesis was detected after addition of 10 pmol of the ArgR-EN1 or ArgR-EN2 chimeric protein to the reaction mixture. However, according to expectations, ArgC protein synthesis strongly decreased when smaller quantities of wild-type B. stearothermophilus ArgR were added to the reaction mixture (decreasing nearly fourfold with 2.5 pmol of protein). Moreover, the two other chimeras, ArgR-BN1 and ArgR-BN2, exhibited stronger repression of ArgC synthesis than wild-type B. stearothermophilus ArgR. The most pronounced repression was detected for ArgR-BN1; 0.5 pmol of this chimeric protein almost completely abolished reporter argC gene expression in vitro.



View larger version (27K):
[in this window]
[in a new window]
  		FIG. 5. Autoradiogram of B. stearothermophilus ArgC protein synthesis in a coupled transcription-translation system with E. coli. pHAV2 plasmid DNA (10 µg/ml) was used as the template for protein synthesis from a PargC promoter, and 1 µl of wild-type ArgR or chimeric proteins at different concentrations (indicated for the hexameric molecule) was added to the mixture containing a premixed solution, unlabeled amino acids, and L-[35S] methionine, and the reaction was initiated by adding S30 extracts (a 30-µl total volume). (a) Lanes 1 and 10, without addition of exogenous ArgR protein to the reaction mixture. Other samples show ArgC synthesis after addition to the reaction mixture as follows: lanes 2 and 3, 1 pmol and 2.5 pmol, respectively, of wild-type B. stearothermophilus ArgR; lanes 4 and 5, 0.1 pmol and 0.5 pmol, respectively, of ArgR-BN1; lanes 6 and 7, 1 pmol and 2.5 pmol, respectively, of ArgR-BN2; lanes 8 and 9, 2.5 pmol and 10 pmol, respectively, of ArgR-EN1; lanes 11 and 12, 5 pmol and 10 pmol, respectively, of wild-type E. coli ArgR; lanes 13, 14, and 15, 2.5 pmol, 5 pmol, and 10 pmol, respectively, of ArgR-EN2. (b) Lane 1, without ArgR; lanes 2 and 3, 1 pmol and 5 pmol, respectively, of wild-type B. stearothermophilus ArgR; lanes 4 and 5, 1 pmol and 5 pmol, respectively, of ArgR-BeB; lane 6, 20 pmol of wild-type E. coli ArgR; lanes 7 and 8, 5 pmol and 20 pmol, respectively, of ArgR-EbE.

We also found that the ArgR-BN1 and ArgR-BN2 chimeras inhibited the growth of E. coli Top10 host cells in minimal medium irrespective of the presence of arginine. Neither ArgR-EN1 nor ArgR-EN2 displayed such a property. Consequently, the ArgR-BN1 and ArgR-BN2 chimeras, as described previously for wild-type B. stearothermophilus ArgR (19), behave as superrepressors in the heterologous E. coli host.

Therefore, in summary, the ArgR-BN1 and ArgR-BN2 chimeras but not ArgR-EN1 or ArgR-EN2 act as strong repressors both in vitro and in vivo.

Analysis of linker-exchanged ArgR chimeras. Taking into consideration the positions of junction sites between domains in the four chimeras studied, it was reasonable to assume that, along with the {alpha}4 structure, the interdomain-linker peptide might affect the DNA-binding properties of ArgRs. Therefore, we designed two new derivatives, ArgR-EbE and Arg-BeB, in which only the linker sequence was replaced in wild-type ArgR repressors (see Fig. 1a).

The purified ArgR-BeB protein was found to be soluble in high-salt buffer and hardly entered nondenaturing polyacrylamide gels, properties similar to those of the wild-type B. stearothermophilus ArgR. ArgR-EbE was somewhat more soluble in high-salt buffer, but its migration was closer to that of the wild-type E. coli repressor (data not shown). The mobility shift assay was performed in the pH 7.9 buffer used for the surface plasmon resonance studies. Under these conditions, wild-type B. stearothermophilus protein-DNA complexes entered a nondenaturing agarose gel more easily, forming compact retarded bands, and moreover, the affinity of the repressor was almost twice that seen in the previous buffer. In the presence of arginine, ArgR-EbE bound only the carAB operator with an affinity similar to the wild-type E. coli repressor in the presence of arginine (Fig. 6a and 6b), whereas ArgR-BeB bound both carAB and argCo operator DNAs with an affinity higher than that of the wild-type B. stearothermophilus ArgR repressor (Fig. 6c and 6d). The ArgR-BeB binding was accompanied by the appearance of an additional band(s) that might reflect trimeric or dimeric protein-DNA-bound complexes.



View larger version (48K):
[in this window]
[in a new window]
  		FIG. 6. Mobility shift assay of wild-type E. coli and B. stearothermophilus ArgR repressors and linker-exchanged ArgR-BeB and ArgR-EbE chimeras with respect to the E. coli carAB (a and c) and B. stearothermophilus argCo (b and d) operator DNAs. Operator DNAs were incubated with increasing amounts of His-tagged proteins at 37°C for 30 min in 20 mM Tris-HCl (pH 7.9)-50 mM NaCl-50 mM KCl-0.1 mM dithiothreitol-0.005% surfactant P20 in the presence of arginine and electrophoresed in a 2% agarose gel with arginine. (a) Lane 1 contained carAB DNA only; lanes 2 and 3 contained 3.2 nM and 6.4 nM E. coli ArgR, respectively; lanes 4 and 5 contained 3.2 nM and 6.4 nM ArgR-EbE, respectively. (b) Lane 1 contained argCo DNA only; lanes 2 to 6 contained 416 nM, 832 nM, 1,664 nM, 3,328 nM, and 6,656 nM E. coli ArgR, respectively; lanes 7 to 12 contained 416 nM, 832 nM, 1,664 nM, 3,328 nM, 6,656 nM, and 13,312 nM ArgR-EbE, respectively. (c and d) Lanes 1 contained carAB and argCo DNAs only, respectively; lanes 2 to 6 contained 6.5 nM, 13 nM, 26 nM, 52 nM, and 104 nM B. stearothermophilus ArgR, respectively; lanes 7 to 12 contained 6.5 nM, 13 nM, 26 nM, 52 nM, 104 nM, and 208 nM ArgR-BeB, respectively.

As further shown by surface plasmon resonance, the ArgR-EbE protein bound perfectly to carAB in the presence of arginine but was practically unable to bind the carAB or argCo operator in the absence of arginine (see Table 3). The ArgR-BeB protein was able to bind both carAB and argCo operator DNAs with affinities higher than that of the wild-type B. stearothermophilus ArgR repressor, and the binding was less affected by arginine.

The linker-exchanged chimeric proteins were tested for their ability to repress B. stearothermophilus argC gene transcription from the PargCo promoter-operator region in vitro (see Fig. 5b). The same quantity of added ArgR-EbE protein appeared to decrease ArgC synthesis more than the wild-type protein from E. coli, whereas the ArgR-BeB protein appeared to be slightly weaker compared to the wild-type B. stearothermophilus protein.

The linker-exchanged chimeras ArgR-EbE and ArgR-BeB were thus found to be reminiscent of the wild-type E. coli and B. stearothermophilus repressors, respectively. Nevertheless, minor differences detected in their properties indicated that the linker peptide could modulate the DNA-binding properties of the chimeric proteins.

Wild-type B. stearothermophilus ArgR protein can recognize a single Arg box. Since the stimulating effect of arginine on DNA binding was less noticeable for the wild-type B. stearothermophilus ArgR, absent for ArgR-BN1 and, moreover, decreased for the ArgR-BN2 chimeric protein, we assumed that target DNA recognition by these proteins is less dependent on the double Arg box organization of operators. To address this question, we applied the protein array method, which permits simultaneous analysis of numerous proteins (11).

Eight wild-type ArgR and chimeric proteins were serially fourfold diluted and spotted onto a nitrocellulose membrane. Two IRDye800-labeled DNA probes, a 76-bp fragment carrying the entire B. stearothermophilus argCo operator and a shorter 56-bp fragment lacking the upstream Arg box of the operator (see Fig. 1b), were incubated with duplicate membranes in the presence and absence of arginine. Membranes were washed thoroughly, and protein-DNA binding was monitored by detection of a fluorescent signal.

As expected, no response was detected from the wild-type E. coli ArgR and chimeric ArgR-EbE, ArgR-EN1, and ArgR-EN2 proteins with either of the two probes (Fig. 7). A clear signal was detected from the wild-type B. stearothermophilus ArgR and chimeric ArgR-BeB, ArgR-BN1, and ArgR-BN2 proteins with respect to a 76-bp DNA probe. The linker-exchanged ArgR-BeB chimera exhibited better affinity for the argCo operator than the wild-type protein under the conditions used. A signal was still detectable for 0.93 pg of spotted ArgR-BeB or wild-type ArgR, indicating the extreme sensitivity of the detection method (see Fig. 7a). Arginine increased the binding affinity of the wild-type B. stearothermophilus repressor (except for a 238-pg protein spot) and probably the ArgR-BN1 chimera for the operator DNA. However, this amino acid decreased the DNA-binding efficiency of the ArgR-BN2 chimera and probably the ArgR-BeB chimera under the conditions used (see Fig. 7b).



View larger version (40K):
[in this window]
[in a new window]
  		FIG. 7. Fluorescence detection of DNA-protein interactions with protein arrays (a to d) and relative intensity of fluorescence signals (e). His-tagged purified proteins were serially fourfold diluted and spotted on duplicate nitrocellulose membranes. The nondiluted sample corresponds to 238 pg of spotted protein. Binding reactions were carried out with equal quantities of IRDye800-labeled probes, a 76-bp DNA (a and b) and a 56-bp DNA (c and d), for 12 h at 18°C. L-Arginine (10 mM) was included in the protein dilution and DNA-binding buffers (b and d). Membranes were washed and scanned with an Odyssey Imager (LI-COR).

A weaker signal was detected from the wild-type B. stearothermophilus ArgR and the ArgR-BeB and ArgR-BN2 chimeras bound to a 56-bp DNA probe, and no signal was detectable from ArgR-BN1 spots (see Fig. 7c and 7d). It was clearly visible that the presence of arginine had no more stimulating effect on the binding of wild-type B. stearothermophilus ArgR, suggesting that allosteric modifications of the wild-type repressor are not necessary to recognize the single-Arg-box operator DNA.

Analysis of fluorescent spots with the GenePix Pro4.0 program confirmed the visual observations (see Fig. 7e). Though it is impossible quantify binding constants by the protein chip method, the data obtained from titrated protein samples are, in general, in agreement with the results of the mobility shift assay and surface plasmon resonance data. The difference in the physicochemical characteristics of the chimeras, especially in solubility, can explain variations in binding affinities detected by various methods, as optimal conditions differed for each protein. Consequently, comparative study of such proteins is a laborious task. Therefore, the ability to detect DNA-binding responses from numerous immobilized proteins with high sensitivity and in parallel experiments with various probes might give a great advantage to the protein array method compared to other approaches.


arrow
DISCUSSION
 
Arginine metabolism in two bacterial species, the gram-negative mesophilic E. coli and the gram-positive thermophilic B. stearothermophilus, is governed by a similar transcription-regulatory system. However, the data presented in this study show that the wild-type E. coli arginine-bound or unbound repressor is practically unable to recognize the heterologous argCo operator, whereas the B. stearothermophilus repressor binds both the argCo and the carAB operators irrespective of the presence of arginine (see also the introduction).

By applying different methodological approaches, we have shown that DNA-binding and gene repression functions have been affected in domain-exchanged and linker-exchanged ArgR chimeras. We have found that, due to structural perturbations, the ArgR-EN1 and ArgR-EN2 chimeras, both conserving the DNA-binding domain of E. coli ArgR, were unable to achieve sequence-specific recognition, even with respect to carAB in the presence of arginine. Regardless of expectations, ArgR-BN2 and probably ArgR-BN1, both harboring the oligomerization domain from the E. coli repressor, bound more weakly to the carAB and argCo operators in the presence of arginine than in its absence, indicating that their hexameric molecules have lower affinity for a target than other oligomeric forms, and in this respect both chimeras resembled the wild-type ArgR protein of the hyperthermophilic bacterium Thermotoga neapolitana (9). Furthermore, both chimeras display stronger repression of the argC reporter gene than the wild-type B. stearothermophilus repressor did, indicating that they may differ in other protein-protein interactions involved in transcriptional regulation from the B. stearothermophilus argC promoter-operator region (M. Snapyan, M. Lecocq, L. Guével, M.-C. Arnaud, A. Ghochikyan, and V. Sakanyan, unpublished data). Finally, the replacement of the linker peptide in the wild-type E. coli and B. stearothermophilus repressors slightly changed the DNA-binding parameters of the corresponding chimeras.

In agreement with observed modulations in DNA-binding properties, the six chimeras constructed also possessed various solubility, electrophoretic migration, and thermostability properties depending on the region substituted. Given the junction sites between substituted domains and the observed differences in solubility and velocity within the pairs ArgR-BN1 and ArgR-BN2, ArgR-EN1 and ArgR-EN2, and ArgR-BeB and ArgR-Ebe, we attribute the particularities of the chimeric proteins to their oligomerization domain, in particular to the region covering the {alpha}4 helix and, to a lesser extent, the linker peptide. It has already been shown that mutations in the linker peptide of the E. coli CytR and LacI repressors can affect the operator-binding affinity (12, 18).

The three-dimensional structure of the region covering the linker peptide and {alpha}4 helix is not yet available for E. coli ArgR. The linker peptide sequences in the E. coli and B. stearothermophilus repressors are completely different, and therefore it is difficult to obtain a model structure even for ArgR-EbE in order to make confident interpretations of apparent functional modifications in linker-substituted chimeras. However, a model simulation of B. stearothermophilus ArgR clearly predicts that the {alpha}4 helix substitution within the oligomerization domain can affect interdomain interactions in chimeric proteins (V. Duyne, personal communication).

In the wild-type B. stearothermophilus hexameric protein, each DNA-binding domain of one trimer is in contact with the oligomerization domain of the neighboring trimer, i.e., amino acids of the {alpha}1 helix interact with residues of the {alpha}4 helix (28). The {alpha}4 helix structure in E. coli ArgR appears to be about one turn shorter than its B. stearothermophilus homologue, or this structure is of the same length but kinked due to a Pro76 residue that might make a loop shorter by 3 to 4 residues. Therefore, the {alpha}4 helix substitution should change the spacing between the DNA-binding and oligomerization domains in trimers of domain-exchanged chimeras and thereby affect wHTH module positioning with respect to the Arg boxes in the operator sequence. Moreover, arginine-mediated allosteric modification in the {alpha}4 helix-substituted ArgR-BN2 and probably ArgR-BN1 chimeras can lead to worsening of the cooperative compliance between the four DNA-binding motifs in the hexamers and the four patches in the two Arg boxes of the operator and thereby to decreasing the DNA-binding affinity. Though these structural predictions need to be confirmed, it follows from our data that the oligomerization domain of E. coli and B. stearothermophilus ArgRs have a different impact on the physicochemical and DNA-binding properties of the corresponding repressors.

DNA-protein interactions are highly dynamic processes that depend on both partners, and local folding transition of a transcription-regulatory protein can be coupled to site-specific DNA binding (37). DNA-binding affinity varies for different arg-specific operators and is considerably reduced for a single Arg box of E. coli operators (6, 8, 42, 43). A single Arg box has been identified for several arginine-related and nonrelated genes (13, 20, 25, 26, 30, 38), though binding of ArgR to these targets has not yet been proven. We confirmed that arginine-stimulated allosteric transition is essential in the molecular recognition of operators carrying two adjacent Arg boxes by the wild-type B. stearothermophilus repressor. In addition, we show that, contrary to E. coli ArgR, the wild-type B. stearothermophilus repressor is able to bind a single Arg box of the B. stearothermophilus argCo operator with a rather high efficiency, and arginine-mediated allosteric modification appears not to be necessary for this recognition, since arginine has no activating effect on the binding of such a target.

In the B. stearothermophilus repressor, the oligomerization domain appears to contribute, via interdomain contacts, to the adaptation of the DNA-binding domain to various geometries in target DNAs. In particular, protein contacts between residues of the oligomerization domain and the DNA-binding domain in hexamers or trimers of the thermophilic ArgR protein may modulate the position of wHTH with respect to a single or a double Arg-box. Consequently, the binding of wild-type B. stearothermophilus ArgR repressor to a single Arg-box can reflect its ability to bind to a wider range of operators, whereas binding of the wild-type E. coli repressor to the two-Arg-box-associated operator reflects its restricted ability to recognize DNA targets.

It is worth mentioning that the ArgR repressor of T. neapolitana displays similar affinity for single- and double-Arg-box operators (M.-C. Arnaud and V. Sakanyan, unpublished data), and its own argRo operator can be efficiently recognized by the B. stearothermophilus but not by the E. coli repressor (35). Therefore, the capacity of bacterial ArgR repressors to distinguish and bind a single Arg box appears to be a basic condition for interactions with a wide range of arg-specific and other operator sequences.

Maas has postulated that the arginine repressor might play the role of a global regulatory protein in bacterial cells (22), which is supported by the presence of many single-Arg-box-like sequences in the E. coli genome (30). Therefore, we explain the superrepressor behavior of B. stearothermophilus ArgR and some chimeras in E. coli as the result of repression of single-box-carrying operators for metabolic genes encoding proteins involved in synthesis of a growth factor(s) other than arginine growth factors and that this repression can arrest host cell reproduction in synthetic medium supplied with this amino acid.

Since thermophilic bacteria branched from the 16S rRNA tree earlier than mesophilic enterobacteria (1), we put forward the hypothesis that the ArgR-mediated regulatory mechanism evolved from a less-specific single-box to a highly gene-specific double-box operator organization in microbial genomes and that the oligomerization domain {alpha}4 structure-determined interactions with the DNA-binding domain might reflect a molecular adaptation mechanism of various ArgR regulators to the corresponding targets.


arrow
ACKNOWLEDGMENTS
 
We thank G. Van Duyne for modeling ArgR chimeras and helpful comments. H. Chebrou is acknowledged for contributing to construction of the domain-exchanged chimeras. We also thank the staff of Biacore-France for the possibility of using Biacore 3000.

A.G. was supported by a postdoctoral fellowship from the Région des Pays de la Loire, and this work was supported by grants from the Région des Pays de la Loire (Contrat de Plan Etat-Région) and the Ministère de l'Education et de la Recherche.


arrow
FOOTNOTES
 
* Corresponding author. Mailing address: Laboratoire de Biotechnologie, FRE CNRS 2230 Unité Biocatalyse, Faculté des Sciences et des Techniques, Université de Nantes, 2 rue de la Houssinière, 44322 Nantes, France. Phone: (33) 251125620. Fax: (33) 251125637. E-mail: Vehary.Sakanyan{at}chimbio.univ-nantes.fr. Back

{dagger} Present address: Institut Gustave Roussy UMR 8532 du CNRS PR II, 94805 Villejuif, France. Back


arrow
REFERENCES
 
    1
  1. Achenbach-Richter, L., R. Gupta, K. O. Stetter, and C. R. Woese. 1987. Were the original eubacteria thermophiles? Syst. Appl. Microbiol. 9:34-39.[Medline]
    2. 2
  2. Ausubel, F. M., R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A. Smith, and K. Struhl. 1993. Current protocols in molecular biology. John Wiley & Sons, New York, N.Y.
    3. 3
  3. Bentez-Bellon, E., G. Moreno-Hagelsieb, and J. Colado-Vides. 2002. Evaluation of thresholds for the detection of binding sites for regulatory proteins in Escherichia coli K12 DNA. Genome Biol. 3:0013.1-0013.16. [Online.]
    4. 4
  4. Bradford, M. M. 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72:248-254.[CrossRef][Medline]
    5. 5
  5. Burke, M., A. M. Merican, and D. J. Sheratt. 1994. Mutant Escherichia coli arginine repressor proteins that fail to bind L-arginine, yet retain the ability to bind their normal DNA-binding sites. Mol. Microbiol. 13:609-618.[CrossRef][Medline]
    6. 6
  6. Charlier, D., M. Roovers, F. Van Vliet, A. Boyen, R. Cunin, Y. Nakamura, N. Glansdorff, and A. Pierard. 1992. Arginine regulon of Escherichia coli K-12. A study of repressor-operator interactions and of in vitro binding affinities versus in vivo repression. J. Mol. Biol. 226:367-386.[CrossRef][Medline]
    7. 7
  7. Chen, H. Z., and G. Zubay. 1983. Prokaryotic coupled transcription-translation. Methods Enzymol. 101:674-690.[Medline]
    8. 8
  8. Chen, S. H., A. F. Merican, and D. J. Sherratt. 1997. DNA-binding of Escherichia coli arginine repressor mutants altered in oligomeric state. Mol. Microbiol. 24:1143-1156.[CrossRef][Medline]
    9. 9
  9. Dimova, D., P. Weigel, M. Takahashi, F. Marc, G. Van Duyne, and V. Sakanyan. 2000. Thermostability, oligomerization and DNA-binding properties of the regulatory protein ArgR from the hyperthermophilic bacterium Thermotoga neapolitana. Mol. Gen. Genet. 263:119-130.[Medline]
    10. 10
  10. Dion, M., D. Charlier, H. Wang, D. Gigot, A. Savchenko, J. N. Hallet, N. Glansdorff, and V. Sakanyan. 1997. The highly thermostable arginine repressor of Bacillus stearothermophilus: gene cloning and repressor-operator interactions. Mol. Microbiol. 25:385-398.[CrossRef][Medline]
    11. 11
  11. Emili, A. Q., and G. Cagney. 2000. Large-scale functional analysis with peptide or protein arrays. Nat. Biotechnol. 18:393-397.[CrossRef][Medline]
    12. 12
  12. Falcon, C. M., and K. S. Matthews. 1999. Glycine insertion in the hinge region of lactose repressor protein alters DNA-binding. J. Biol. Chem. 274:30849-30857.[Abstract/Free Full Text]
    13. 13
  13. Gardan, R., G. Rapoport, and M. Débarbouillé. 1995. Expression of the rocDEF operon involved in arginine catabolism in Bacillus subtilis. J. Mol. Biol. 249:843-856.[CrossRef][Medline]
    14. 14
  14. Glansdorff, N. 1996. Biosynthesis of arginine and polyamines, p. 408-433. In F. C. Neidhardt et al. (ed.), Escherichia coli and Salmonella: cellular and molecular biology. American Society for Microbiology, Washington, D.C.
    15. 15
  15. Grandori, R., T. A. Lavoie, M. Pflumm, G. Tian, H. Niersbach, W. K. Maas, R. Fairman, and J. Carey. 1995. The DNA-binding domain of the hexameric arginine repressor. J. Mol. Biol. 254:150-162.[CrossRef][Medline]
    16. 16
  16. Ho, N. S., D. H. Hunt, M. R. Horton, K. J. Pullen, and R. L. Pease. 1989. Site directed mutagenesis by overlap extension with the polymerase chain reaction. Gene 77:51-59.[CrossRef][Medline]
    17. 17
  17. Holtham, C. A., K. Jumel, C. M. Miller, S. E. Harding, S. Baumberg, and P. G. Stockley. 1999. Probing activation of the prokaryotic arginine transcriptional regulator with chimeric proteins. J. Mol. Biol. 289:707-727.[CrossRef][Medline]
    18. 18
  18. Jorgensen, C. I., B. H. Kallipolitis, and P. Valentin-Hansen. 1998. DNA-binding characteristics of the Escherichia coli CytR regulator: a relaxed spacing requirement between operator half-sites is provided by a flexible, unstructured interdomain linker. Mol. Microbiol. 27:41-50.[CrossRef][Medline]
    19. 19
  19. Karaivanova, I. M., P. Weigel, M. Takahashi, C. Fort, A. Versavaud, G. Van Duyne, D. Charlier, J. N. Hallet, N. Glansdorff, and V. Sakanyan. 1999. Mutational analysis of the thermostable arginine repressor from Bacillus stearothermophilus: dissecting residues involved in DNA-binding properties. J. Mol. Biol. 291:843-855.[CrossRef][Medline]
    20. 20
  20. Klingel, U., C. M. Miller, A. K. North, P. G. Stockley, and S. Baumberg. 1995. A binding site for activation by the Bacillus subtilis AhrC protein, a repressor/activator of arginine metabolism. Mol. Gen. Genet. 248:329-340.[CrossRef][Medline]
    21. 21
  21. Lim, D., J. D. Oppenheim, T. Eckhardt, and W. K. Maas. 1987. Nucleotide sequence of the argR gene of Escherichia coli K12 and isolation of its product, the arginine repressor. Proc. Natl. Acad. Sci. USA 84:6697-6701.[Abstract/Free Full Text]
    22. 22
  22. Maas, W. 1994. The arginine repressor of Escherichia coli. Microbiol. Rev. 58:631-640.[Abstract/Free Full Text]
    23. 23
  23. Makarova, K. S., A. A. Mironov, and M. S. Gelfand. 2001. Conservation of the binding site for the arginine repressor in all bacterial lineages. Genome Biol. 2:0013.1-0013.8. (Online.)
    24. 24
  24. Miller, J. H. 1992. A short course in bacterial genetics. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
    25. 25
  25. Miller, C. M., S. Baumberg, and P. G. Stockley. 1997. Operator interactions by the Bacillus subtilis arginine repressor/activator, AhrC: novel positioning and DNA-mediated assembly of a transcriptional activator at catabolic sites. Mol. Microbiol. 26:37-48.[CrossRef][Medline]
    26. 26
  26. Mironov, A. A., E. V. Koonin, M. A. Roytberg, and M. S. Gelfand. 1999. Computer analysis of transcription regulatory patterns in completely sequenced bacterial genomes. Nucleic Acids Res. 27:2981-2989.[Abstract/Free Full Text]
    27. 27
  27. Muskhelishvili, G., M. Buckle, H. Heumann, R. Kahmann, and A. A. Travers. 1997. FIS activates sequential steps during transcription initiation at a stable RNA promoter. EMBO J. 16:3655-3665.[CrossRef][Medline]
    28. 28
  28. Ni, J.-P., V. Sakanyan, D. Charlier, N. Glansdorff, and G. D. Van Duyne. 1999. Structure of the arginine repressor from Bacillus stearothermophilus. Nature Struct. Biol. 6:427-432.[CrossRef][Medline]
    29. 29
  29. Persson, B., M. Buckle, and P. G. Stockley. 2000. Kinetics of DNA interactions studied by surface plasmon resonance spectroscopy, p. 257-279. In A. Travers and M. Buckle, (ed.), DNA-protein interactions: a practical approach. Oxford University Press, Oxford, United Kingdom.
    30. 30
  30. Robison, K., A. M. McGuire, and G. M. Church. 1998. A comprehensive library of DNA-binding site matrices for 55 proteins applied to the complete Escherichia coli K-12 genome. J. Mol. Biol. 284:241-254.[CrossRef][Medline]
    31. 31
  31. Sakanyan, V. A., A. S. Hovsepyan, I. L. Mett, A. V. Kochikyan, and P. K. Petrosyan. 1990. Molecular cloning and structural-functional analysis of arginine biosynthesis genes of the thermophilic bacterium Bacillus stearothermophilus. Genetika (USSR) 26:1915-1925.[Medline]
    32. 32
  32. Savchenko, A., D. Charlier, M. Dion, P. Weigel, J. N. Hallet, C. Holtham, S. Baumberg, N. Glansdorff, and V. Sakanyan. 1996. The arginine operon of Bacillus stearothermophilus: characterization of the control region and its interaction with the heterologous B. subtilis arginine repressor. Mol. Gen. Genet. 252:69-78.[CrossRef][Medline]
    33. 33
  33. Savchenko, A., P. Weigel, D. Dimova, M. Lecocq, and V. Sakanyan. 1998. The Bacillus stearothermophilus argCJBD operon harbours a strong promoter as evaluated in Escherichia coli cells. Gene 212:167-177.[CrossRef][Medline]
    34. 34
  34. Smith, M. C. M., L. Czaplewski, A. K. North, S. Baumberg, and P. G. Stockley. 1989. Sequence required for regulation of arginine biosynthesis promoters are conserved between Bacillus subtilis and Escherichia coli. Mol. Microbiol. 3:23-28.[CrossRef][Medline]
    35. 35
  35. Song, H., H. Wang, D. Gigot, D. Dimova, V. Sakanyan, N. Glansdorff, and D. Charlier. 2002. Transcription regulation in thermophilic bacteria: high resolution contact probing of Bacillus stearothermophilus and Thermotoga neapolitana arginine repressor-operator interactions. J. Mol. Biol. 315:255-274.[CrossRef][Medline]
    36. 36
  36. Southern, E. M. 1975. Detection of specific sequences among DNA fragments separated by gel electrophoresis. J. Mol. Biol. 98:503-517.[CrossRef][Medline]
    37. 37
  37. Spolar, R., and T. Record, Jr. 1994. Coupling of local folding to site-specific binding of proteins to DNA. Science 263:777-784.[Abstract/Free Full Text]
    38. 38
  38. Stirling, C. J., G. Szatmari, G. Stewart, M. C. M. Smith, and D. Sherratt. 1988. The arginine repressor is essential for plasmid-stabilizing site-specific recombination at the ColE1 cer locus. EMBO J. 7:4389-4395.[Medline]
    39. 39
  39. Stockley, P. G., A. J. Baron, C. M. Wild, I. D. Parsons, C. M. Miller, C. A. M. Holtham, and S. Baumberg. 1998. Dissecting the molecular details of prokaryotic transcriptional control by surface plasmon resonance: the methionine and arginine repressor proteins. Biosensors Bioelectronics 13:637-650.[CrossRef][Medline]
    40. 40
  40. Summers, D. 1998. Timing, self-control and a sense of direction are the secrets of multicopy plasmid stability. Mol. Microbiol. 29:1137-1145.[CrossRef][Medline]
    41. 41
  41. Sunnerhagen, M., M. Nilges, G. Otting, and J. Carey. 1997. Solution structure of the DNA-binding domain and model for the complex multifunctional hexameric arginine repressor with DNA. Nat. Struct. Biol. 4:819-826.[CrossRef][Medline]
    42. 42
  42. Szwajkajzer, D., L. Dai, J. W. Fukayama, B. Abramszyk, R. Fairman, and J. Carey. 2001. Quantitative analysis of DNA-binding by the Escherichia coli arginine repressor. J. Mol. Biol. 312:949-962.[CrossRef][Medline]
    43. 43
  43. Tian, G., D. Lim, J. Carey, and W. K. Maas. 1992. Binding of the arginine repressor of Escherichia coli K12 to its operator sites. J. Mol. Biol. 226:387-397.[CrossRef][Medline]
    44. 44
  44. Tian, G., and W. K. Maas. 1994. Mutational analysis of the arginine repressor of Escherichia coli. Mol. Microbiol. 13:599-608.[Medline]
    45. 45
  45. Van Duyne, G. D., G. Ghosh, W. K. Maas, and P. B. Sigler. 1996. Structure of the oligomerization and L-arginine binding domain of the arginine repressor of Escherichia coli. J. Mol. Biol. 256:377-391.[CrossRef][Medline]
    46. 46
  46. Wang, H., N. Glansdorff, and D. Charlier. 1998. The arginine repressor of Escherichia coli K-12 makes direct contacts to minor and major groove determinants of the operators. J. Mol. Biol. 277:805-824.[CrossRef][Medline]
    47. 47
  47. Yang, H. C., S. N. Park, W. G. Nolan, C. D. Lu, and A. T. Abdelal. 1997. Cloning and characterization of the arginine-specific carbamoylphosphate synthetase from Bacillus stearothermophilus. Eur. J. Biochem. 249:443-449.[Medline]

Journal of Bacteriology, December 2002, p. 6602-6614, Vol. 184, No. 23
0021-9193/02/$04.00+0     DOI: 10.1128/JB.184.23.6602-6614.2002
Copyright © 2002, American Society for Microbiology. All Rights Reserved.




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ghochikyan, A.
Right arrow Articles by Sakanyan, V.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ghochikyan, A.
Right arrow Articles by Sakanyan, V.

Copyright © 2009 by the American Society for Microbiology.   For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»