Specificity of the interaction of RocR with the rocGrocA intergenic region in Bacillus subtilis

Naima Ould Ali1, Josette Jeusset2, Eric Larquet3,5, Eric Le Cam2, Boris Belitsky4, Abraham L. Sonenshein4, Tarek Msadek1 and Michel Débarbouillé1

1 Unité de Biochimie Microbienne, Institut Pasteur, URA 2172 du Centre National de la Recherche Scientifique, 25 rue du Docteur Roux, 75724 Paris Cedex 15, France
2 Laboratoire de Microscopie Cellulaire et Moléculaire, Institut Gustave Roussy, UMR 1598 du Centre National de la Recherche Scientifique, 94805 Villejuif Cedex, France
3 Groupe de Microscopie Structurale Moléculaire, Institut Pasteur, URA 2185 du Centre National de la Recherche Scientifique, 25 rue du Docteur Roux, 75724 Paris Cedex 15, France
4 Department of Molecular Biology and Microbiology, Tufts University, School of Medicine, Boston, MA 02111, USA
5 Laboratoire de Minéralogie Cristallographie, Université Paris 6, UMR 7590, IPGP, CNRS Case 115, Tour 16, 4 Place Jussieu, 75252 Paris Cedex 05, France

Correspondence
Michel Débarbouillé
mdebarbo{at}pasteur.fr


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
In Bacillus subtilis, expression of the rocG gene, encoding glutamate dehydrogenase, and the rocABC operon, involved in arginine catabolism, requires SigL ({sigma}54)-containing RNA polymerase as well as RocR, a positive regulator of the NtrC/NifA family. The RocR protein was purified and shown to bind specifically to the intergenic region located between rocG and the rocABC operon. DNaseI footprinting experiments were used to define the RocR-binding site as an 8 bp inverted repeat, separated by one base pair, forming an imperfect palindrome which is present twice within the rocG–rocABC intergenic region, acting as both a downstream activating sequence (DAS) and an upstream activating sequence (UAS). Point mutations in either of these two sequences significantly lowered expression of both rocG and rocABC. This bidirectional enhancer element retained partial activity even when moved 9 kb downstream of the rocA promoter. Electron microscopy experiments indicated that an intrinsically curved region is located between the UAS/DAS region and the promoter of the rocABC operon. This curvature could facilitate interaction of RocR with {sigma}54-RNA polymerase at the rocABC promoter.


Abbreviations: DAS, downstream activating sequence; poly-(dI-dC){bullet}poly-(dI-dC), poly-(deoxyinosinic-deoxycytidylic) acid; UAS, upstream activating sequence


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacillus subtilis can use ammonium or any one of several amino acids as sole nitrogen sources (Baumberg & Klingel, 1993; Cunin et al., 1986; Fisher, 1993; Fisher & Débarbouillé, 2002). Arginine, for instance, allows rapid growth in minimal medium. Bacteria use various strategies to degrade arginine and some bacteria, such as Pseudomonas putida, have multiple catabolic pathways (Tricot et al., 1991).

In B. subtilis, arginine is converted to glutamate by the three-step arginase pathway. The relevant enzymes are encoded by the rocABC and rocDEF operons (Calogero et al., 1994; Gardan et al., 1995). Arginine is cleaved by arginase (RocF) to give ornithine, which is then converted to glutamate semialdehyde by ornithine aminotransferase (RocD). Finally, the conversion of glutamate semialdehyde to glutamate is catalysed by a pyrroline-5-carboxylate dehydrogenase (RocA). This last reaction is shared with the proline catabolic pathway. RocG (glutamate dehydrogenase) catalyses the conversion of glutamate to 2-oxoglutarate, the final step in the utilization of arginine, ornithine and proline as carbon or nitrogen sources (Belitsky & Sonenshein, 1998). The RocR protein positively regulates the rocABC and rocDEF operons and the rocG gene (Belitsky & Sonenshein, 1999; Calogero et al., 1994; Gardan et al., 1995, 1997).

Expression of rocG, rocABC and rocDEF is {sigma}54-dependent (Belitsky & Sonenshein, 1998; Calogero et al., 1994; Gardan et al., 1995). Expression of these genes is fully induced by arginine or ornithine in the growth medium. An intermediate level of expression was observed with proline (Belitsky & Sonenshein, 1998; Gardan et al., 1997). RocR is a member of the NtrC/NifA family of proteins that bind to enhancer-like elements often referred to as upstream activating sequences (UASs). Members of this family of activators contain a region of 220–240 residues called the ‘central domain’, which is specifically required for the formation of open complexes between {sigma}54-containing RNA polymerase and {sigma}54-dependent promoters, also known as ‘-12, -24’ promoters (Kustu et al., 1989, 1991; Merrick, 1993; Morett & Segovia, 1993).

A DNA region essential for rocABC transcription was identified by deletion mapping near position -165 with respect to the transcription start site. This region includes a palindromic sequence of 17 bp that may serve as a UAS. A second copy of this sequence is present near position -122 of the promoter region. We designated these two sequences UAS1 and UAS2, respectively, and suggested that RocR acts by binding to these 17 bp UASs to stimulate transcription of rocABC (Calogero et al., 1994). Similar sequences appear twice within the rocDEF upstream region as well (Gardan et al., 1995). The rocG gene, located just upstream from rocABC, is transcribed by {sigma}54-containing RNA polymerase and also requires RocR for its transcription. The rocG gene has no UAS, however. Its expression depends instead on a sequence designated downstream activating sequence (DAS) located 1·5 kb downstream of the rocG -12, -24 promoter. The same DNA region includes the rocABC UAS1 and UAS2 (Belitsky & Sonenshein, 1999).

In the present work, RocR was purified and shown to bind specifically to the intergenic region between rocG and the rocABC operon. DNaseI footprinting experiments showed that RocR binds to two sites that correspond closely to UAS1 and UAS2. Point mutations in the palindromic sequences affected transcription of both rocG and rocABC. To function, these UAS/DAS sequences must be brought into close proximity with the -12, -24 promoters upstream from rocG and rocABC, suggesting that DNA conformation plays a role during the activation process. Electron microscopy analysis indicated that this region contains an intrinsic curvature centred between the UAS/DAS enhancer element and the rocABC promoter.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains and culture media.
The B. subtilis strains used in this work are listed in Table 1. Escherichia coli K12 strain TG1 [{Delta}(lac proAB) supE thi hsd{Delta}5 (F' traD36 proAB lacIq lacZ {Delta}M15)] (Gibson, 1984) was used for cloning experiments, and E. coli strain BL21{lambda}DE3 [F- ompT hsdSB () gal dcm (DE3)] was used for protein overexpression and purification (Studier & Moffatt, 1986). E. coli was grown in LB broth (Sambrook et al., 1989) and B. subtilis was grown in SP medium (8 g nutrient broth l-1, 1 mM MgSO4, 10 mM KCl, 0·5 mM CaCl2, 10 mM MnCl2, 2 µM FeSO4) or minimal medium [60 mM K2HPO4, 44 mM KH2PO4, 15 mM (NH4)2SO4, 3 mM trisodium citrate, 2 mM MgSO4, 2·2 mg ferric ammonium citrate l-1] supplemented with a carbon source (0·1 %) and auxotrophic requirements (at 100 mg l-1). TSS medium (Fouet & Sonenshein, 1990) was supplemented with 0·5 % glucose and 0·2 % glutamate.


View this table:
[in this window]
[in a new window]
 
Table 1. B. subtilis strains used in this study

 
Transformation and phenotype characterization.
Standard procedures were used to transform E. coli (Sambrook et al., 1989) and transformants were selected on LB broth plates containing 100 µg ampicillin ml-1. B. subtilis was transformed with plasmid or chromosomal DNA as described previously (Martin-Verstraete et al., 1990) and transformants were selected on SP medium plates containing 5 µg chloramphenicol ml-1 or 5 µg kanamycin ml-1 or 1 µg erythromycin ml-1 and 10 µg lincomycin ml-1.

DNA manipulation.
Standard procedures were used to extract plasmids from E. coli (Sambrook et al., 1989). Restriction enzymes, phage T4 DNA polymerase, phage T4 DNA ligase and T4 polynucleotide kinase were used as recommended by the manufacturers. DNA fragments were purified from the agarose gel with a Prep-A-Gene kit (Bio-Rad).

Plasmids constructions.
Plasmid pNO11 was constructed as follows. A 1·7 kbp DNA fragment containing rocG and the beginning of rocA was amplified by PCR using the Pfu polymerase and two oligonucleotides, OMD41b and OMD42b. Oligonucleotides used for PCR included mismatches allowing the creation of EcoRI and BamHI restriction sites. The mutagene M13 in vitro mutagenesis kit (Bio-Rad) was used, based on the procedure described by Kunkel et al. (1987). The 1 kb EcoRI–BamHI DNA fragment was cloned between the EcoRI and BamHI restriction sites of the replicative form of M13mp19 to give phage M13mp19 rocG rocA. Mutations were created as described previously (Martin-Verstraete et al., 1992) using oligonucleotides OMD49 (5'-GAG AAG CCT CCA CAA AAT AAT TTT GC-3'), OMD50 (5'-GCC TCC GCA AAA TAA TTT TAC ATT TG-3'), OMD59 (5'-GCC TCC ACA AAA TAA TTT TAC ATT TG-3'), OMD60 (5'-GCC TCC GAA AAA TAA TTT TGC ATT TG-3'), OMD61 (5'-GCC TCC GCA AAA TAA TTT TGA ATT TG-3'), OMD62 (5'-GCC TCC GCA AGG TAA TTT TGC ATT TG-3') and OMD63 (5'-GCA AAA AAC ACA AAAT AAA TTT ACG TTC AAG-3'), containing one or several mismatches. The presence of the mutations was verified by DNA sequencing and the EcoRI–BamHI DNA fragment was then cloned between the corresponding sites of vector pDIA5307 just upstream from a promoterless lacZ gene.

Plasmid pNO12 was used to introduce a deletion replacement of the UAS/DAS region in B. subtilis 168. This plasmid was constructed by cloning a 500 bp EcoRI–BamHI DNA fragment, a 1 kb BamHI–PstI DNA fragment carrying the kanamycin-resistance gene and a 500 bp PstI–HindIII DNA fragment between the EcoRI and HindIII sites of pHT181 (Lereclus & Arantès, 1992), to give pNO12. The 500 bp EcoRI–BamHI DNA fragment and the 500 bp PstI–HindIII DNA fragment correspond, respectively, to the chromosomal DNA region encoding the end of rocG and the beginning of rocA and were generated by PCR using oligonucleotides OMD51, OMD52 and OMD53, and OMD54, respectively. Plasmid pNO12 was used to transform B. subtilis 168. Kanamycin-resistant, erythromycin-sensitive integrants arose through a double crossover event in which the entire UAS/DAS region was deleted and replaced with the kanamycin-resistance gene. The corresponding mutant was designated QB7705.

The previously constructed plasmid pDIA5326 (Calogero et al., 1994) contains the promoterless lacZ gene fused to the EcoRI–SalI fragment of the rocG–rocA region containing 0·75 kb of the rocG sequence, the UAS, the rocA promoter and 0·3 kb of the rocA sequence. Plasmid pBB1085 was constructed by deleting the 0·74 kb SmaI–SacII fragment of pDIA5326 and contains the UAS and the rocA promoter but no rocG sequence. Plasmid pBB1087 was constructed by deleting the 0·85 kb SmaI–BclI fragment of pDIA5326 and contains the rocA promoter but no UAS or rocG sequence. Plasmid pET28a was used for protein overexpression and purification (Novagen).

RocR was overproduced using pETRocR. This plasmid was constructed by cloning a 1391 bp NcoI–XhoI DNA fragment corresponding to the rocR coding sequence between the NcoI and XhoI sites of pET28a. This DNA fragment was generated by PCR using oligonucleotides TM-268 (5'-CCACCATGGTCAAAGACAGCGAATTCCTCACATTG-3') and TM-269 (5'-CTCCTCGAGTTCGTTAGACGAGTGAGAAAAT-3'), thus replacing the original GTG translational initiation codon with an ATG codon, and the TAA stop codon by the XhoI restriction site. This allows the creation of a translational fusion adding six carboxy-terminal histidine residues to the RocR protein and placing expression of the gene under the control of a T7 bacteriophage promoter.

Overproduction and purification of RocR.
Plasmid pETRocR was used to transform E. coli BL21{lambda}DE3, in which the T7 RNA polymerase gene is under the control of the inducible lacUV5 promoter (Studier & Moffatt, 1986). The resulting strain was grown in 1 l LB medium at room temperature until the mid-exponential phase (OD600 value of approx. 0·8), IPTG was added (1 mM) and incubation pursued for 3 h. The cells were centrifuged at 10 810 g for 30 min, and resuspended in 1/50 of the culture volume of buffer I [50 mM NaH2PO4 (pH 8), 300 mM NaCl, 20 mM imidazole]. The cells were disrupted by sonication and cell debris was removed by two consecutive centrifugation steps at 17 210 g for 30 min. E. coli crude protein extracts were loaded onto a 0·15 ml Ni-NTA agarose (Qiagen) column previously equilibrated with buffer I. The column was washed with 10 volumes of buffer II [50 mM NaH2PO4 (pH 6), 300 mM NaCl, 30 mM imidazole] and RocR was eluted with an imidazole gradient (30–500 mM). Fractions were pooled and dialysed against buffer III [50 mM NaH2PO4 (pH 8), 300 mM NaCl, 50 % (v/v) glycerol] to remove the imidazole and concentrate the protein solution approximately fourfold.

SDS-PAGE on 12 % acrylamide gels was performed as described previously; gels were stained with Coomassie blue (Laemmli, 1970). Protein concentrations were determined using the Bio-Rad protein assay (Bradford, 1976).

Gel mobility shift and DNA-binding assay.
A 263 bp EcoRI–BamHI DNA fragment, corresponding to the promoter region of the rocABC operon (positions -267 to -25 with respect to the transcription initiation site), was generated by PCR using oligonucleotides RA1 (5'-GAAGAATTCGTGGATATGCGTTTGGCGGC-3') and RA2 (5'-CGCGGATCCGCCATTTTCTTGTCTTTTGC-3'). This DNA fragment was radioactively labelled by BamHI digestion and treatment with the Klenow fragment of DNA polymerase I (Roche) in the presence of a mixture of dGTP, dCTP, dTTP (0·08 mM) and [{alpha}-32P]dATP (40 µCi; 148 kBq). Binding of RocR to DNA was carried out in a 10 µl reaction mixture containing 104 c.p.m. [{alpha}-32P]-labelled DNA, 1 µg poly-(dI-dC){bullet}poly-(dI-dC) (Amersham-Pharmacia), 25 mM NaH2PO4 (pH 7), 150 mM NaCl, 0·1 mM EDTA, 2 mM MgSO4, 1 mM DTT and 10 % (v/v) glycerol. The DNA-binding reaction was initiated by the addition of RocR and was incubated at room temperature for 20 min. Samples were then directly loaded onto a 4 % polyacrylamide gel (89 mM Tris/borate, 2 mM EDTA) during electrophoresis (14 V cm-1). Electrophoresis was pursued for 1 h at room temperature and the gels were then dried and analysed by autoradiography.

DNaseI footprinting.
DNA fragments used for DNaseI footprinting were prepared by PCR, using the Pfu polymerase (Stratagene) and 20 pmol primers RA1 and RA2, one of which was previously labelled with T4 polynucleotide kinase (New England Biolabs) and [{gamma}-32P]dATP. Labelled PCR products were purified using the Qiaquick PCR purification kit (Qiagen). RocR binding to DNA was carried out as described above, with the addition of BSA (0·1 µg), and DNaseI footprinting experiments were then performed as described previously (Derré et al., 1999).

{beta}-Galactosidase and glutamate dehydrogenase assays.
B. subtilis cells containing lacZ fusions were grown to an OD600 value of 1. {beta}-Galactosidase specific activities were determined as described previously and are expressed as Miller units per milligram of protein or as Miller units per OD600 (Miller, 1972).

Glutamate dehydrogenase was assayed, as described previously, in strains carrying a gudB : : tet mutation, to avoid any contribution from the GudB glutamate dehydrogenase (Belitsky et al., 1998). The values reported represent a mean of at least two independent assays.

DNA-fragment preparation for electron microscopy.
All DNA fragments used in electron microscopy experiments were obtained by PCR using the Pwo polymerase and the following oligonucleotides: OMD145 (5'-GGAAGAACACTGAAAGAAACACAGATTTTAAACG-3'); OMD141 (5'-CCTAGCGAGTGCCGGCGGCGTCACGGTTTC-3'); OMD140 (5'-GCATCGCTTTTTCAGCAAGCTCTTGATCCGCTGTAG-3').

OMD140 was biotinylated at the 5' end. The oligonucleotides were purified by an anion exchange MonoQ column using a SMART system (Pharmacia). The 696 bp biotinylated fragment (containing the two UAS sites and rocABC promoter amplified using OMD140 and OMD141) was dimerized using streptavidin. This technique allowed us to obtain specifically oriented tail-to-tail DNA dimers. A similar strategy was used to synthesize the 2131 bp DNA fragment using oligonucleotides OMD140 and OMD145. This DNA fragment contains the promoter of rocG, rocG and the promoter of rocABC.

RocR–DNA complex formation.
Purified RocR was incubated for 10 min at 20 °C at different ratios with the 2131 bp DNA fragment in binding buffer containing 10 mM Tris/HCl (pH 7·5), 50 mM NaCl and 5 mM MgCl2. DNA was maintained at 2 nM and complexes were formed at 10, 20, 40 and 100 nM of RocR. Complexes were purified by Superose 6B chromatography (APbiotech).

Observation by electron microscopy.
After purification, 5 ml of a solution containing 0·5 mg DNA ml-1 were deposited onto a 600 mesh copper grid covered with a very thin carbon film activated by a glow discharge in the presence of pentylamine (Dubochet et al., 1971). Grids were washed with 2 % aqueous uranyl acetate, dried and observed in annular dark-field mode, using a Zeiss 902 electron microscope (Le Cam & Delain, 1995).

Acquisition and digitization of the molecules.
For each orientation, 200 images of individual dimer molecules showing unambiguous trajectories were taken and stored on a hard disk through the use of a camera coupled to the electron microscope. Pictures at a magnification of 140 000x times; were digitized on an image analyser (IBAS Kontron) by following the contour of the molecule and picking up 420–440 points.

Curvature analysis.
Experimental curvature maps of DNA fragment were obtained from the analysis of the set of 200 molecules. The co-ordinates (x, y) of each molecular pathway were smoothed and iso-segmented according to the real base-pair number of the DNA. Using the DNA ResCue program (Larquet et al., 1995), the curvature was quantified by the SD parameter, ratio of the curvilinear (S) to end-to-end (D) lengths of the fragment within a 60 bp window moving along the DNA by 10 bp (Muzard et al., 1990). This ratio gives the w parameter, mean dinucleotide wedge angle per base pair, which is an intrinsic parameter for DNA curvature, independent of the fragment length.

DNA trajectory modelling.
The spatial sequence path of the DNA helix axis was modelled with the DNA ResCue program. A local co-ordinate system is associated to each base pair. Its origin is on the helical axis. The x and y axes are, respectively, the transversal and longitudinal axis in the base-pair plane, and the z axis points to the next base pair. The local co-ordinate system linked to the (n+1)th base pair is deduced from the (n)th base-pair co-ordinate system by applying a unit z-translation, and three successive rotations: twist around the z axis, roll around the y axis so formed, and tilt around the x axis resulting from the two previous rotations. Twist angles (Kabsch et al., 1982) and roll and tilt angles were taken from the curvature model considered. Calculations were performed using transformation matrices that refer to the centre of the co-ordinate system of each base pair in the co-ordinate system associated with the first base pair. The three-dimensional co-ordinates of the centres of the successive base pairs determine the path of the DNA helix axis.


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Purified RocR binds specifically to the rocABC upstream region
To determine whether RocR controls expression by binding directly to the UAS1 and UAS2 elements of the rocABC promoter region or indirectly via other regulatory elements, the RocR protein was overproduced and purified. The rocR coding sequence was cloned in the vector pET28a, generating a carboxy-terminal translational fusion with six histidine residues, under the control of a T7 bacteriophage promoter (see Methods).

A comparison by SDS-PAGE of crude extracts from E. coli cells carrying the vector pET28a alone or the plasmid pETRocR showed an overproduced RocR-specific band of the expected molecular mass (approx. 53 kDa) (Fig. 1, lanes 2 and 3). His-tagged RocR protein was purified in a single step using a Ni-NTA agarose column (see Methods) and SDS-PAGE analysis revealed a high degree of purity (Fig. 1, lane 4).



View larger version (125K):
[in this window]
[in a new window]
 
Fig. 1. Overproduction and purification of RocR. Tris SDS-PAGE analysis of crude extracts from E. coli BL21{lambda}DE3 carrying pET28a (lane 2) or pETRocR (lane 3) and purified RocR after Ni-NTA affinity chromatography (29 pmol, lane 4). Molecular mass standards were loaded in lane 1.

 
Purified RocR was used in gel mobility shift and DNA-binding assays with DNA fragments corresponding to the promoter region of the rocABC operon. All DNA-binding assays were performed with an excess of non-specific competitor DNA [1 µg poly-(dI-dC){bullet}poly-(dI-dC)]. Radiolabelled PCR-generated probes corresponding to positions -267 to -25 of the rocABC promoter region were incubated with increasing amounts of purified RocR (Fig. 2). RocR bound to the DNA fragment, forming a single protein–DNA complex (Fig. 2, lane 3).



View larger version (52K):
[in this window]
[in a new window]
 
Fig. 2. Specific binding of RocR to the rocG–rocA intergenic region. Gel mobility shift experiments were performed by incubating purified RocR with radiolabelled DNA fragments (10 000 c.p.m.) corresponding to the promoter region of the rocABC operon (positions -267 to -25 with respect to the transcription initiation site). Lanes: 1, no protein; 2, 11 pmol RocR; 3, 22 pmol RocR. The arrow indicates the shifted DNA band.

 
RocR recognizes an octanucleotide inverted repeat sequence
To precisely determine the location and sequence of the RocR-binding site, DNaseI footprinting assays were performed on DNA fragments carrying the rocABC promoter region. As shown in Fig. 3, when the non-template strand was end-labelled, RocR protected two regions, extending from positions -180 to -157 and -136 to -108 (Fig. 3a). The template strand was protected from positions -172 to -149 and -141 to -112 (Fig. 3b). These results suggest two binding sites for RocR within the rocABC promoter region, corresponding to UAS1 and UAS2.



View larger version (42K):
[in this window]
[in a new window]
 
Fig. 3. DNaseI footprinting analysis of RocR binding to the rocG–rocA intergenic region. Each lane contains 50 000 c.p.m. radiolabelled DNA fragment corresponding to the non-template strand (a) or template strand (b). A+G Maxam and Gilbert reactions of the appropriate DNA fragments were loaded in lane 1. Fragments were incubated in 50 µl volumes with increasing amounts of purified RocR: lane 2, no protein; lane 3, 11 pmol; lane 4, 22 pmol. Regions protected by RocR are indicated by square brackets. (c) Extent of the DNaseI protected regions. The sequence of the rocG–rocA intergenic region is shown, with the DNaseI protected areas boxed and shaded. Positions are indicated relative to the {sigma}54-dependent transcription start site, indicated by +1. The -12 and -24 promoter sequences are boxed. The octanucleotide inverted repeat sequences are indicated by horizontal arrows. DNaseI hypersensitive cleavage sites are shown by vertical arrowheads.

 
Nucleotide sequence analysis of the protected regions (Fig. 3c) revealed an octanucleotide inverted repeat sequence, 5'-CGCAAAAT-3', forming an imperfect palindrome, which is present twice within the rocG–rocABC intergenic region. Numerous DNaseI hypersensitive sites appeared as enhanced cleavages (Fig. 3), suggesting that binding of RocR induces bending of the DNA within this region.

Mutations affecting rocABC and rocG expression
To define precisely the DNA sequence important for the activation of rocABC and rocG, point mutations were introduced into the dyad symmetry elements of UAS1 and UAS2. A DNA fragment containing the end of rocG, the entire rocA promoter region and the two UASs was cloned upstream from lacZ in plasmid pDIA5307. Point mutations were introduced by site-directed mutagenesis in each copy of the two UASs. The resulting plasmids were subsequently introduced by transformation into a B. subtilis strain containing a deletion of the two UASs and the rocABC promoter region. In the recombinant strains, the plasmids integrated by a single crossover event via homologous sequences at the end of the rocG gene. As a result, UAS1 and UAS2 were placed between the end of rocG and the promoter of the rocA'–lacZ transcriptional fusion (Fig. 4a). The two 17 bp palindromic sequences are centred at positions -168 (UAS1) and -124 (UAS2) with respect to the rocABC transcription start site, and each inverted repeat contains an invariant GC dinucleotide. As shown in Table 2, single mutations in the GC doublet of UAS1 caused a two- to fivefold decrease in {beta}-galactosidase levels compared to the parental strain. Mutants were constructed in which two mutations were introduced in a polyadenine region in the middle of UAS1 (strain QB7712) or both GC doublets of UAS1 were modified (strain QB7709). Expression of rocA'–lacZ in these two strains was also lowered two- to fivefold, respectively (Table 2). Mutations modifying the conserved GC doublets of UAS2 (strain QB7715) also led to a significant decrease in lacZ expression (Table 2), indicating that both UASs are required for full rocABC expression. To test if the two UASs act in transcriptional activation of rocABC, we constructed a mutant affected in all four GC doublets (strain QB7716). As shown in Table 2, the combination of these four mutations completely abolished rocABC expression.



View larger version (21K):
[in this window]
[in a new window]
 
Fig. 4. Genetic experiments affecting the UAS/DAS region. (a) Integration strategy for placing point mutations in the rocG–rocA intergenic region. pNO11 carries a DNA fragment containing the end of rocG, the two UASs and the entire rocA promoter region cloned upstream from lacZ. (b) Relocation experiments of the UAS/DAS region downstream from the rocA promoter region.

 

View this table:
[in this window]
[in a new window]
 
Table 2. Effect of mutations located in UAS1 or UAS2 on the expression of rocG or rocABC

 
To test the effect of point mutations in UAS1 and/or UAS2 on rocG expression, glutamate dehydrogenase activity was assayed in the mutant strains. As indicated in Table 2, RocG activities in the single and double mutants were lowered, although not to the extent observed for the rocA–lacZ fusion. Glutamate dehydrogenase activity was strongly decreased in strain QB7716 in which both UAS/DAS sequences were mutated. These results show that the two UAS/DAS elements are essential for induction of both rocABC and rocG.

Relocation of the UAS/DAS downstream from rocA
Unlike other {sigma}54-dependent promoters, rocG has no UAS and its expression depends on a sequence located 1·5 kb downstream from its promoter. This sequence still retains partial activity and can activate rocG if moved upstream from its promoter or 15 kb downstream (Belitsky & Sonenshein, 1999). To determine whether this activation at a distance is specific to the rocG promoter, relocation experiments were carried out to move the UAS/DAS region downstream from the integrated pBB1085. Plasmid pBB1085 contains the two UASs, the rocA promoter and the beginning of rocA. Strain BB1662 contains a chromosomal copy of the rocA promoter lacking UAS1 and UAS2. When introduced into strain BB1662, pBB1085 can only integrate into the chromosome downstream of the UAS deletion. As a result, the two UASs of pBB1085 are placed downstream from the rocA–lacZ fusion in strain BB1672 and are separated from the rocA promoter by the entire length of pBB1085 (9·1 kbp) (Fig. 4b). Results of {beta}-galactosidase assays are shown in Table 3. In strain BB1672, rocA'–lacZ expression levels represent 15 % of that of the wild-type strain (BB1648). This intermediate level is RocR-dependent, as shown in strain BB1673. As a control, plasmid pBB1087, identical to pBB1085 but lacking the two UAS/DAS, was integrated in the chromosome of strain BB1662, leading to strain BB1889, in which rocA'–lacZ expression is virtually abolished. Taken together, these experiments indicate that the two UASs relocated downstream from the rocA gene are still able to allow RocR-dependent activation of transcription from the -12, -24 promoter.


View this table:
[in this window]
[in a new window]
 
Table 3. Expression of rocA'–lacZ with a UAS/DAS relocated downstream

{beta}-Galactosidase activities were assayed during growth in TSS medium containing 0·5 % glucose, 0·2 % glutamate and 0·2 % ornithine as the inducer.

 
Theoretical models predict intrinsic DNA curvature of the rocABC promoter region
Electrophoretic mobilities of circularly permuted DNA fragments suggested that the rocA promoter region contains a stable bend (Miller et al., 1997). A computer-based analysis of the nucleotide sequence of the 696 bp UAS intergenic region (Fig. 5a) revealed the presence of a discrete intrinsically curved DNA segment. Three theoretical models of DNA curvature (Bolshoy et al., 1991; De Santis et al., 1988; Koo & Crothers, 1988; Ulanovsky et al., 1986) were considered. The three-dimensional trajectory of this fragment was simulated in relation to the DNA sequence (Fig. 5b). Each base pair (bp) is defined by the reference frame attached to it. The algorithm calculates the reference frame transformations from one base pair to the adjacent one along the DNA sequence. Local curvature is calculated by the SD parameter, the ratio of the curvilinear (S) to end-to-end (D) lengths for a 60 bp window moving along the DNA by 10 bp steps (Muzard et al., 1990). This ratio gives the w parameter, mean dinucleotide wedge angle per bp. Theoretical maps of the 696 bp fragment curvature predicted by these models are shown in Fig. 5(c). Positions are indicated relative to the length of the DNA fragment. These three models clearly predict an intrinsically curved region extending from positions 310 to 430 (Fig. 5a), containing a large number of A or T tracks in phase with the DNA helix pitch. The maximum curvature value within this peak is located at position 370 (between the two UAS sites and the promoter of the rocABC operon).



View larger version (29K):
[in this window]
[in a new window]
 
Fig. 5. Predicted intrinsic DNA curvature. (a) Schematic representation of the 696 bp and 2131 bp DNA fragments used in this work. The rocG and rocA genes are indicated by arrows and the -24, -12 promoter and UAS/DAS sites are underlined. (b) Curvature maps of the 696 bp DNA fragment are obtained from the Trifonov (broken line), de Santis (grey line) and Crothers (solid line) models. The mean dinucleotide wedge angle w (expressed in degrees per base pair) is calculated for a 60 bp window moving along the DNA sequence by 10 bp steps. (c) The DNA helix three-dimensional trajectory is simulated according to the previous three models. Positions are indicated relative to the length of the DNA fragment.

 
Experimental curvature maps obtained from electron microscopy data
The 696 bp fragment was chosen to determine an experimental curvature map by electron microscopy. Biotinylated DNA molecules were oriented by streptavidin dimerization as described in Methods. Experimental curvature maps of dimeric DNA fragments were obtained from the analysis of 200 digitized molecules (Fig. 6a) and are expressed in terms of mean dinucleotide wedge angles (w, in degrees per base pair) using 60 bp windows (Fig. 6b). Positions are indicated relative to the length of the DNA fragment. The curvature map exhibits a symmetrical profile where the region of higher curvature peaks corresponds to positions 310–430, with a maximal wedge value (1·4) at position 380. The experimental curvature profile is very similar in this region to the theoretical profiles in valleys and peaks of relative curvature (Fig. 6b). As previously observed (Muzard et al., 1990), the magnitude of the theoretical curvature is lower than that obtained from the experimental analysis. When superimposed two at a time, these curvature maps show many similarities. These results confirm the ability to determine the local curvature from the pathway of a population of molecules visualized by electron microscopy. The origin of curvature in this region is clearly due to the high number of phased A tracts.



View larger version (110K):
[in this window]
[in a new window]
 
Fig. 6. Visualization of DNA curvature by electron microscopy. (a) An electron micrograph of the monomeric (m) and dimeric (d) forms of the 696 bp DNA fragment is presented. Bar, 100 nm. (b) Curvature maps of 200 specifically orientated tail-to-tail DNA dimers are calculated for a 60 bp window and a 10 bp moving step (solid line) and are compared with theoretical curvature models (the Trifonov model is presented; broken line). Positions are indicated relative to the length of the DNA fragment.

 
Complex formation between RocR and the rocABC promoter region analysed by electron microscopy
To visualize the association of RocR with the rocABC promoter region, purified RocR was incubated at different ratios with a 2131 bp DNA fragment containing the rocG promoter, the two UAS/DAS and the promoter region of the rocABC operon. At a molecular ratio of 2 nM of DNA fragments for 25 nM of protein, complexes were analysed by electron microscopy and mapped. These complexes clearly resulted from polymerization of the protein along the DNA and bridging between two adjacent double strand DNA segments, leading to a large ‘hairpin-like’ structure. The polymerization can progress along the DNA, inducing local condensation (Fig. 7a). This process can induce the formation of a totally condensed DNA fragment. Higher concentrations of RocR induced high condensation of DNA molecules and aggregation of the complexes. Mapping of 50 complexes (Fig. 7b) shows that the protein binds to the promoter region containing the curved region. A histogram (Fig. 7b) was obtained by plotting the total number of interactions (in per cent) within a 20 bp window to take into account all regions covered by RocR. Fifteen to thirty-five per cent of the complexes were localized between positions 1660 to 1860 (this region includes the UASs and the promoter). Positions are indicated relative to the length of the DNA fragment.



View larger version (89K):
[in this window]
[in a new window]
 
Fig. 7. Analysis of RocR–DNA complexes by electron microscopy. (a) RocR–DNA complexes with the 2131 bp fragment visualized by electron microscopy. The five panels show representative RocR–DNA complexes. Different levels of polymerization of RocR can be visualized (arrows) associated with large hairpin structures resulting from bridges between two adjacent RocR–DNA complexes. Bar, 100 nm. (b) Mapping of RocR–DNA complexes within the 2131 bp fragment. The binding sites of 50 RocR–DNA complexes were mapped and the data presented in the histogram show the total number of interactions (%) within a 20 bp window to take polymerization of the protein along DNA into account. Positions are indicated relative to the length of the DNA fragment. The rocA promoter and the two UAS/DAS sequences are indicated.

 

   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
We have previously shown that a DNA sequence located downstream from the rocG gene and upstream from the rocABC promoter is necessary for full induction of transcription of these genes. This enhancer-like element, called UAS/DAS, was predicted to be a binding site for RocR, the activator of the roc regulon. By a combination of site-directed mutagenesis, relocation experiments and DNaseI footprinting experiments, we have shown that RocR interacts with two 17 bp sequences containing imperfect inverted repeats. The palindromic nature of the RocR-binding site relates each half of the site by a twofold axis of symmetry at right angles to the DNA helix. This suggests that RocR could act as a dimer, with its twofold axis of symmetry coinciding with that of the DNA-binding site. Each subunit would interact in the same manner with each half-site. This is typical of helix–turn–helix DNA-binding proteins, in agreement with the classical helix–turn–helix motif previously identified within the carboxy-terminal domain of RocR (amino acids 431–453) (Calogero et al., 1994).

The two octanucleotide inverted repeat sequences are separated by a single base pair, thus placing them on the same face of the DNA double helix. The inverted repeat sequence is present twice within the rocGrocABC intergenic sequence, constituting two UAS/DAS, both of which are required for full expression of rocG and the rocABC operon. Binding to both UASs appears to occur simultaneously, as no intermediate DNA–protein complexes are detected by gel mobility shift experiments (Fig. 2) as RocR concentrations are varied, and in DNaseI footprinting protection is not progressive but is instead immediate and simultaneous for both UAS elements. This suggests that RocR has the same affinity for each site (Fig. 3). This is in agreement with the mutational analysis, since single- or double-point mutations in either of the inverted repeats diminish rocABC and rocG expression, indicating that the UAS and the DAS are the same regulatory elements. Gel mobility shift and DNaseI footprinting experiments performed in the presence of 10 mM ornithine or arginine showed no difference in RocR binding, suggesting that the protein does not require the inducer to bind to DNA in vitro (data not shown). This confirms previous results indicating that RocR is able to repress its own expression in vivo in an inducer-independent fashion and suggests that the role of the inducer is to alter the conformation of RocR so that it can interact with {sigma}54 RNA polymerase (Gardan et al., 1995).

We have previously reported that deletion of half of UAS1 led to a partial inhibition of rocABC transcription (Calogero et al., 1994). A strong inhibition was only observed in strains containing a deletion of both UASs. These results led to the assumption that RocR multimerizes on both UAS/DAS and that mutations in both sites are required to abolish transcription activation by RocR. A similar situation was described for the interaction of NtrC with its cognate UAS, with dimers of the protein binding co-operatively to the enhancer element, which consists of two specific binding sites. Co-operative binding of NtrC results from interaction between NtrC dimers. Moreover, it was also shown that transcription activation by NtrC at the glnA promoter might depend on formation of an oligomer larger than a tetramer in which additional dimers are associated through protein–protein interactions with those directly bound to the enhancer (Wyman et al., 1997). Multimerization to higher forms is a common feature of all NtrC-like proteins tested so far (Farez-Vidal et al., 1996; Perez-Martin & De Lorenzo, 1996; Wikstrom et al., 2001; Wilson et al., 1994; Wyman et al., 1997).

{sigma}54-Dependent promoters are activated at a distance by specific regulators bound to UASs that must loop out for productive contacts with the {sigma}54-RNA polymerase holoenzyme. The function of the enhancer is to tether the regulator at a high local concentration near the promoter. Studies of activation of {sigma}54-dependent transcription have shown that looping can be effected by random or transient changes in DNA conformation or can be directed by a site-specific DNA-binding protein such as integration host factor (IHF). In Gram-negative bacteria, an IHF site is often found between the binding site for the activator and the {sigma}54-RNA polymerase-binding site. However, no equivalent of IHF has been found to be encoded in the B. subtilis genome.

DNA fragments encompassing the rocA promoter region migrate anomalously on polyacrylamide gels and it was suggested that this effect is caused by an intrinsic bend in the DNA (Miller et al., 1997). Electron microscopy experiments indicate unambiguously that a strongly curved region is indeed observed upstream from the -12, -24 promoter of rocABC. Binding of RocR also appears to induce bending of the DNA, since several enhanced DNaseI cleavage sites appear within the protected regions constituting UAS1 and UAS2 (Fig. 3). These results suggest a molecular basis for the contact between the RocR activator bound at the UAS/DAS elements and the RNA polymerase located at the rocABC promoter. However, the intrinsic curvature located between the UAS/DAS region and the -12, -24 promoter of rocABC does not explain the simultaneous activation at a distance of the rocG promoter. One possibility is that RocR binds with high affinity to strongly curved DNA as described for HNS (Thei Dame et al., 2001). Oligomerization of RocR along the DNA starting from its initial nucleation-binding site might bridge the adjacent UAS/DAS segment. Such a structure may facilitate activation of both promoters by RocR. The role of an additional regulatory protein, AhrC, remains unclear. AhrC is similar to the E. coli ArgR repressor and is a positive regulator of the synthesis of the enzymes of the arginine catabolic pathway (Harwood & Baumberg, 1977). It has been previously shown that AhrC makes contacts with the -12, -24 region of the rocABC promoter (Miller et al., 1997). Since AhrC is also required for the activation of rocG expression (Belitsky & Sonenshein, 1999), it could be involved in the stabilization of a putative DNA loop, including the rocG and the rocABC promoters.


   ACKNOWLEDGEMENTS
 
We are grateful to Georges Rapoport, in whose laboratory part of this work was carried out, for critical reading of the manuscript. We thank Joëlle Bignon and Annie Landier for excellent technical assistance and Christine Dugast for expert secretarial assistance. This work was supported by research funds from the Institut Pasteur, the Centre National de la Recherche Scientifique, the Ministère de la Recherche and the US Public Health Service (grant no. GM36718).


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Baumberg, S. & Klingel, U. (1993). Biosynthesis of arginine, proline and related compounds. In Bacillus subtilis and Other Gram-positive Bacteria: Biochemistry, Physiology and Molecular Genetics, pp. 229–306. Edited by A. L. Sonenshein, J. A. Hoch & R. Losick. Washington, DC: American Society for Microbiology.

Belitsky, B. & Sonenshein, A. L. (1998). Role and regulation of Bacillus subtilis glutamate dehydrogenase genes. J Bacteriol 180, 6298–6305.[Abstract/Free Full Text]

Belitsky, B. & Sonenshein, A. L. (1999). An enhancer element located downstream of the major glutamate dehydrogenase gene of Bacillus subtilis. Proc Natl Acad Sci U S A 96, 10290–10295.[Abstract/Free Full Text]

Bolshoy, A., McNamara, P., Harrington, R. E. & Trifonov, E. N. (1991). Curved DNA without A-A: experimental estimation of all 16 DNA wedge angles. Proc Natl Acad Sci U S A 88, 2312–2316.[Abstract]

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]

Calogero, S., Gardan, R., Glaser, P., Schweitzer, J., Rapoport, G. & Débarbouillé, M. (1994). RocR, a novel regulatory protein controlling arginine utilization in Bacillus subtilis, belongs to the NtrC/NifA family of transcriptional activators. J Bacteriol 176, 1234–1241.[Abstract]

Cunin, R., Glandsdorff, N., Pierard, A. & Stalon, V. (1986). Biosynthesis and metabolism of arginine in bacteria. Microbiol Rev 50, 314–352.

De Santis, P., Palleschi, A., Savino, M. & Scipioni, A. (1988). A theoretical model of DNA curvature. Biophys Chem 32, 305–317.[CrossRef][Medline]

Derré, I., Rapoport, G. & Msadek, T. (1999). CtsR, a novel regulator of stress and heat shock response, controls clp and molecular chaperone gene expression in Gram-positive bacteria. Mol Microbiol 31, 117–132.[CrossRef][Medline]

Dubochet, J., Ducommun, M., Zollinger, M. & Kellenberger, E. (1971). A new preparation method for dark-field electron microscopy of macromolecules. J Ultrastruct Res 35, 147–167.[Medline]

Farez-Vidal, M. E., Wilson, T. J., Davidson, B. E., Howlett, G. J., Austin, S. & Dixon, R. A. (1996). Effector-induced self association and conformation changes in the enhancer-binding protein NtrC. Mol Microbiol 22, 779–788.[Medline]

Fisher, S. (1993). Utilization of amino acids and other nitrogen containing compounds. In Bacillus subtilis and Other Gram-positive Bacteria: Biochemistry, Physiology and Molecular Genetics, pp. 221–228. Edited by A. L. Sonenshein, J. A. Hoch & R. Losick. Washington, DC: American Society for Microbiology.

Fisher, S. & Débarbouillé, M. (2002). Nitrogen source utilization and its regulation. In Bacillus subtilis and its Closest Relatives. From Genes to Cells, pp. 181–191. Edited by A. L. Sonenshein, J. A. Hoch & R. Losick. Washington, DC: American Society for Microbiology.

Fouet, A. & Sonenshein, A. L. (1990). A target for carbon source-dependent negative regulation of the citB promoter of Bacillus subtilis. J Bacteriol 172, 835–844.[Medline]

Gardan, R., Rapoport, G. & Débarbouillé, M. (1995). Expression of the rocDEF operon involved in arginine catabolism in Bacillus subtilis. J Mol Biol 249, 843–856.[CrossRef][Medline]

Gardan, R., Rapoport, G. & Débarbouillé, M. (1997). Role of the transcriptional activator RocR in the arginine-degradation pathway of Bacillus subtilis. Mol Microbiol 24, 825–837.[CrossRef][Medline]

Gibson, T. J. (1984). Studies on the Epstein–Barr Virus Genome. Cambridge: University of Cambridge.

Harwood, C. R. & Baumberg, S. (1977). Arginine hydroxamate resistant mutants of Bacillus subtilis with altered control of arginine metabolism. J Gen Microbiol 100, 177–188.[Medline]

Kabsch, W., Sander, S. & Trifonov, E. N. (1982). The ten helical twist angles of B-DNA. Nucleic Acids Res 10, 1097–1104.[Abstract]

Koo, H.-S. & Crothers, D. M. (1988). Calibration of DNA curvature and a unified description of sequence-directed bending. Proc Natl Acad Sci U S A 85, 1763–1767.[Abstract]

Kunkel, T. A., Roberts, J. D. & Zakour, R. A. (1987). Rapid and efficient site-specific mutagenesis without phenotypic selection. Methods Enzymol 154, 367–382.[Medline]

Kustu, S., Santero, E., Keener, J., Popham, D. & Weiss, D. (1989). Expression of sigma 54 (ntrA)-dependent genes is probably united by a common mechanism. Microbiol Rev 53, 367–376.

Kustu, S., North, A. K. & Weiss, D. (1991). Prokaryotic transcriptional enhancers and enhancer-binding proteins. Trends Biochem Sci 16, 397–402.[CrossRef][Medline]

Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680–685.[Medline]

Larquet, E., Furrer, P., Stasiak, A., Dubochet, J. & Revet, B. (1995). DNA ResCue: 3D reconstruction and simulation curvature of DNA from the analysis of electron micrographs. (Ninth Conversation in Biomolecular Stereodynamics, June 20–24, 1995). J Biomol Struct Dyn 12, A134.

Le Cam, E. & Delain, E. (1995). Nucleic acids–ligand interactions. In Visualization of Nucleic Acids, pp. 333–358. Edited by G. Morel. Boca Raton, FL: CRC Press.

Lereclus, D. & Arantès, O. (1992). spbA locus ensures the segregational stability of pHT1030, a novel type of Gram-positive replicon. Mol Microbiol 6, 35–46.[Medline]

Martin-Verstraete, I., Débarbouillé, M., Klier, A. & Rapoport, G. (1990). Levanase operon of Bacillus subtilis includes a fructose-specific phosphotransferase system regulating the expression of the operon. J Mol Biol 214, 657–671.[Medline]

Martin-Verstraete, I., Débarbouillé, M., Klier, A. & Rapoport, G. (1992). Mutagenesis of the Bacillus subtilis "-12, -24" promotor of the levanase operon and evidence for the existence of an upstream activating sequence. J Mol Biol 226, 85–99.[Medline]

Merrick, M. J. (1993). In a class of its own – the RNA polymerase sigma factor sigma 54 (sigma N). Mol Microbiol 10, 903–909.[Medline]

Miller, C., Baumberg, S. & Stockley, P. G. (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]

Miller, J. H. (1972). Experiments in Molecular Genetics. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Morett, E. & Segovia, L. (1993). The {sigma}54 bacterial enhancer-binding protein family: mechanism of action and phylogenetic relationship of their functional domains. J Bacteriol 175, 6067–6074.[Medline]

Muzard, G., Theveny, B. & Revet, B. (1990). Electron microscopy mapping of pBR322 DNA curvature. Comparison with theoretical models. EMBO J 9, 1289–1298.[Abstract]

Perez-Martin, J. & de Lorenzo, V. (1996). ATP binding to the sigma 54-dependent activator XylR triggers a protein multimerization cycle catalyzed by UAS DNA. Cell 86, 331–339.[Medline]

Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Studier, F. W. & Moffatt, B. A. (1986). Use of bacteriophage T7 RNA polymerase to direct selective high-level expression of cloned genes. J Mol Biol 189, 113–130.[Medline]

Thei Dame, R., Wyman, C. & Goosen, N. (2001). Structural basis for preferential binding of H-NS to curved DNA. Biochimie 83, 231–234.[CrossRef][Medline]

Tricot, C., Stalon, V. & Legrain, C. (1991). Isolation and characterization of Pseudomonas putida mutants affected in arginine, ornithine and citrulline catabolism: function of the arginine oxidase and arginine succinyltransferase pathways. J Gen Microbiol 137, 2911–2918.[Medline]

Ulanovsky, L., Bodner, M., Trifonov, E. N. & Choder, M. (1986). Curved DNA: design, synthesis and circularization. Proc Natl Acad Sci U S A 83, 862–866.[Abstract]

Wikstrom, P., O'Neil, E. & Shingler, V. (2001). The regulatory N-terminal region of the aromatic-responsive transcriptional activator DmpR constrains nucleotide-triggered multimerization. J Mol Biol 314, 971–984.[CrossRef][Medline]

Wilson, T. J., Maroudas, G., Howlett, G. J. & Davidson, B. D. (1994). Ligand-induced self association of the Escherichia coli regulatory protein TyrR. J Mol Biol 238, 309–318.[CrossRef][Medline]

Wyman, C., Rombel, I., North, A., Bustamante, C. & Kustu, S. (1997). Unusual oligomerization required for activity of NtrC, a bacterial enhancer-binding protein. Science 275, 1658–1661.[Abstract/Free Full Text]

Received 23 September 2002; revised 11 December 2002; accepted 19 December 2002.



This Article
Abstract
Full Text (PDF)
Alert me when this article is cited
Alert me if a correction is posted
Citation Map
Services
Email this article to a friend
Similar articles in this journal
Similar articles in PubMed
Alert me to new issues of the journal
Download to citation manager
Google Scholar
Articles by Ali, N. O.
Articles by Débarbouillé, M.
Articles citing this Article
PubMed
PubMed Citation
Articles by Ali, N. O.
Articles by Débarbouillé, M.
Agricola
Articles by Ali, N. O.
Articles by Débarbouillé, M.


HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
INT J SYST EVOL MICROBIOL MICROBIOLOGY J GEN VIROL
J MED MICROBIOL ALL SGM JOURNALS
Copyright © 2003 Society for General Microbiology.