Purification, crystallization and preliminary X-ray analysis of the effector domain of AlsR, an LysR-type transcriptional regulator from Bacillus subtilis

The cloning, purification, crystallization and preliminary crystallographic analysis of the effector-binding domain of the transcriptional regulator AlsR from B. subtilis are described.

Abstract

AlsR from Bacillus subtilis, a member of the LysR-type transcriptional regulator (LTTR) family, regulates the transcription of the alsSD operon encoding enzymes involved in acetoin biosynthesis. LTTRs represent the largest known family of transcriptional regulators in bacteria. In this study, AlsR_82–302S100A, representing the effector domain, was produced in Escherichia coli, purified and crystallized using the sitting-drop vapour-diffusion method in the presence of 2.1 M dl-malic acid pH 7.0 at 293 K. The crystals belonged to space group C2, with unit-cell parameters a = 142.91, b = 74.96, c = 94.39 Å, β = 110.543°. X-ray data extending to a resolution of 2.6 Å were collected.

1. Introduction  

Bacillus subtilis forms acetoin under anaerobic fermentative growth conditions and as a product of the aerobic carbon-overflow metabolism. This reaction requires the enzymes acetolactate synthase (AlsS) and acetolactate decarboxylase (AlsD) encoded by the alsSD operon. The alsSD promoter was found to be activated in response to acetate accumulation during fermentative growth or the aerobic stationary growth phase and low pH (Renna et al., 1993 ▶). The transcriptional regulator AlsR was found to be essential for alsS–lacZ reporter gene expression under all growth conditions tested. AlsR belongs to the family of LysR-type transcriptional regulators (LTTRs; Schell, 1993 ▶).

The LTTRs were first described by Henikoff et al. (1988 ▶). Members of the LTTRs are widely distributed within the bacterial and archaeal kingdoms. They are involved in the regulation of metabolic functions such as amino-acid synthesis, sugar catabolism, antibiotic resistance, aromatic compound degradation and virulence (Schell, 1993 ▶; Clark et al., 2004 ▶; Maddocks & Oyston, 2008 ▶; Quade et al., 2011 ▶). LTTRs typically have a molecular weight of between 30 and 36 kDa and consist of two domains: an N-terminal DNA-binding domain (∼65 residues) with a winged-helix–turn–helix motif and a C-terminal regulatory domain (∼200 residues) for effector binding and oligomerization. These two domains are connected via a large linker helix (∼30 residues) which is involved in oligomerization of the proteins. LTTRs have been shown to adopt a wide range of oligomerization states, such as dimers (Zhou et al., 2010 ▶), tetramers (Muraoka et al., 2003 ▶; Monferrer et al., 2010 ▶) and even octamers (Sainsbury et al., 2009 ▶).

LTTRs bind at promoter sequences composed of a high-affinity regulatory binding site (RBS) with a conserved palindromic sequence T-N₁₁-A and a lower affinity activator binding site (ABS) (Schell, 1993 ▶). Binding of the regulator at both sites and its oligomerization controls promoter activity. They can activate or repress the expression of target genes depending on the position of the LTTR box relative to the transcriptional start site (Maddocks & Oyston, 2008 ▶). Binding of LTTRs to target promoters is modulated by effectors. Effector binding at the C-terminal effector-binding domain (EBD) results in conformational changes that in turn control the affinity of the N-terminal DNA-binding domain for its target promoter (Schell, 1993 ▶). For AlsR, the metabolite acetate or an acidic pH value have been postulated to be the inducing agents.

Structural studies of full-length LTTRs have been hampered owing to their low solubility and their tendency to aggregate (Ezezika et al., 2007 ▶; Smirnova et al., 2004 ▶). To date, only a few full-length LTTRs have been crystallized. Some effector-domain structures have been described, but only the analyses of BenM and DntR resulted in structures which contained the effector (Ezezika et al., 2007 ▶; Devesse et al., 2011 ▶). The structure of the effector domain of BenM, one of the closest structural relatives of AlsR, with 36% amino-acid sequence identity, was first solved. The full-length structure of BenM was subsequently determined (Ruangprasert et al., 2010 ▶).

In previous studies, amino acids of B. subtilis AlsR located in the potential co-inducer binding site were mutated to analyse the functional role of AlsR. A mutation at position 100 from serine to alanine resulted in a complete loss of transcriptional activation in vivo and in vitro. Moreover, the binding ability of AlsR(S100A) was altered. Binding of the wild-type AlsR protein to an alsS promoter fragment containing the RBS and the ABS resulted in the formation of three complexes (complexes I, II and III) that were retarded in electrophoretic mobility shift assays (EMSA). In contrast, the transcription-incompetent AlsR(S100A) mutant failed to form the slowest migrating complex III, indicating its importance for the formation of the higher ordered transcription-competent complex at the DNA (Frädrich et al., 2012 ▶).

2. Materials and methods  

2.1. Limited proteolysis of AlsR  

Chymotrypsin treatment of purified AlsR protein was carried out in AlsR buffer [100 mM Tris–HCl pH 8.0, 150 mM NaCl, 10%(v/v) glycerol]. The AlsR protein concentration was adjusted to 1 mg ml⁻¹ and the enzyme concentration to 20 µg ml⁻¹. Samples were incubated at room temperature and the reaction was stopped at different time points by boiling for 10 min at 363 K in SDS sample buffer. Reaction products were separated using 15% SDS–PAGE stained with Coomassie Blue. N-terminal sequencing of an AlsR fragment with a molecular weight of 26 000 ± 3000 by the Edman degradation method indicated the presence of a fragment starting at residue 82. MALDI–TOF mass-spectrometric analyses determined a molecular mass of 25 015 Da for the C-terminal AlsR protein fragment (Fig. 1 ▶).

Figure 1.

Limited proteolysis of B. subtilis AlsR. A Coomassie Blue-stained gel of limited proteolysis of AlsR with chymotrypsin is shown. The arrows indicate the sizes of full-length AlsR and the stable fragments.

2.2. Cloning, expression and purification  

The corresponding alsR_82–302S100A gene was generated via PCR amplification using the primer pair EH336 (5′-TCCCCCCGGGG­CACAGCGGACGGCCCGC-3′) and EH236 (5′-GCGAGCTCTC­ATGTACCTGCATCACTC-3′) with pHRBalsRS100A as a template. The PCR product was digested with XmaI and SacI and subsequently ligated into vector pET52bTrx carrying an N-terminal thioredoxin-Strep (Trx-Strep) tag and an HRV-3C protease-cleavage site (Frädrich et al., 2012 ▶).

For production and purification of the B. subtilis AlsR_82–302S100A protein, a 6 l culture of Escherichia coli BL21(DE3) cells containing pET52bTrx-AlsR_82–302S100A was grown at 310 K in LB medium with the addition of appropriate antibiotics. At an optical density of 0.8 at 578 nm, expression of alsR was induced by the addition of 0.1 mM isopropyl β-d-1-thiogalactopyranoside (IPTG) followed by a temperature shift to 298 K. The Trx-Strep-AlsR_82–302 S100A fusion protein was expressed overnight. The cells were harvested by centrifugation at 3000g and resuspended in washing buffer (100 mM Tris–HCl pH 8.0, 150 mM NaCl, 2 mM dithiothreitol, 1 mM EDTA). Cells were disrupted via passage through a French Press at 132.4 MPa and were centrifuged for 1.5 h at 27 000g to remove cell debris. The clear supernatant was applied onto two 6 ml Strep-Tactin columns (IBA, Göttingen, Germany) equilibrated with washing buffer. The columns were washed with at least ten column volumes of washing buffer and the protein was eluted with three column volumes of elution buffer [washing buffer with the addition of 10%(v/v) glycerol, 10 mM NDSB-195 (Merck, Darmstadt, Germany) and 2.5 mM desthiobiotin]. The purification was monitored by separation of the Trx-Strep-AlsR_82–302 S100A fusion protein on an SDS–PAGE gel and visualization by Coomassie Blue R250 staining. HRV-3C protease (Merck, Darmstadt, Germany) was used to remove the Trx-Strep tag. Cleavage was performed according to the manufacturer’s instructions at 277 K and was monitored via SDS–PAGE. Subsequently, the His-tagged HRV-3C protease was removed via affinity chromatography on an Ni–IDA column (Machery & Nagel, Düren, Germany). The protein solution was concentrated using Vivaspin 15 10 kDa molecular-weight cutoff devices (Sartorius, Göttingen, Germany). Gel-permeation chromatography on a Superdex 75 HR 10/30 column (Amersham Biosciences, Piscataway, USA) pre-equilibrated with GPC buffer [50 mM Tris–HCl pH 8.0, 150 mM NaCl, 10%(v/v) glycerol] was used as the final purification step. Separation was performed at a flow rate of 0.5 ml min⁻¹ at 290 K. The following protein standards were used to calibrate the column: albumin (molecular weight 67 000), carboanhydrase (29 000) and lysozyme (14 000). AlsR_82–302S100A eluted at 10.30 ml, which corresponds to a molecular weight of 44 000 ± 5000 and suggests that AlsR_82–302S100A forms a dimer (Fig. 2 ▶). The AlsR_82–302S100A fractions were pooled based on purity and were subsequently concentrated to 11 mg ml⁻¹. The AlsR protein concentration was determined using a 2-D Quant kit (Amersham Bio­sciences, Piscataway, USA) according to the manufacturer’s instructions. The purified AlsR protein was stored at 277 K.

Figure 2.

Purification of B. subtilis AlsR via gel-permeation chromatography. (a) The chromatography of AlsR_82–302S100A resulted in three distinct peaks. The first peak contained aggregated AlsR_82–302S100A protein. In the second peak AlsR_82–302S100A eluted with a molecular weight of 44 000 ± 5000, indicating dimeric organization of the native protein. The third peak corresponded to elution of the Trx-Strep tag. (b) SDS–PAGE analysis of column fractions 6–16 revealed the presence of AlsR_82–302S100A in fractions 9–11 and of the Trx-Strep tag in fractions 12–13. PageRuler Prestained Protein Ladder (Thermo Fisher Scientific, Waltham, USA) was used as molecular-weight standards.

2.3. Crystallization of AlsR_82–302S100A  

Crystallization of AlsR_82–302S100A was performed at 293 K in 96-­well Intelli-Plates (Art Robbins Instruments, Sunnyvale, USA) using the sitting-drop vapour-diffusion method. Initial crystallization screening was performed using commercially available crystallization kits from Qiagen (Hilden, Germany). The plates were set up by mixing 0.2 µl AlsR_82–302S100A protein solution (8 mg ml⁻¹) with 0.2 µl crystallization solution using a HoneyBee crystallization robot (Zinsser Analytic GmbH, Frankfurt, Germany). A small rod-like crystal was noticed in The Classics II Suite condition B11 (2.1 M dl-­malic acid pH 7.0; Fig. 3 ▶). The crystal reached its final length of approximately 0.26 mm after 13 d. The crystal was flash-cooled in liquid nitrogen prior to data collection.

Figure 3.

Crystal of AlsR_82–302S100A protein obtained using 2.1 M dl-malic acid pH 7.0.

2.4. Data collection and analysis  

X-ray diffraction data were collected from a single crystal on beamline ID23-1 of the European Synchroton Radiation Facility (ESRF), Grenoble, France equipped with a PILATUS 6M detector. The oscillation range per image was 0.1° and a complete data set was recorded over a total of 180°. Data were processed in space group C2 to a high-resolution limit of 2.6 Å using the XDS package (Kabsch, 2010 ▶). Complete data-collection and processing statistics are shown in Table 1 ▶.

Table 1. Data-collection statistics for AlsR_82–302S100A.

Values in parentheses are for the outermost resolution shell.

Temperature (K)	100
Wavelength (Å)	0.85
Oscillation range (°)	0.1
Space group	C2
Unit-cell parameters
a (Å)	142.91
b (Å)	74.96
c (Å)	94.39
β (°)	110.543
Resolution (Å)	20.0–2.6 (2.67–2.60)
Observed reflections	98028 (5634)
Unique reflections	28684 (1988)
Multiplicity	3.42 (2.83)
Data completeness	0.99 (0.939)
R merge †	0.109 (0.409)
Average I/σ(I)	8.88 (2.23)

^† R _merge = , where I_i(hkl) is the intensity of an individual measurement and 〈I(hkl)〉 is the corresponding mean value.

3. Results and discussion  

Limited proteolysis of AlsR resulted in a protein fragment of approximately 26 kDa starting at residue 82 and extending into the C-­terminal effector-binding domain (Fig. 1 ▶). The mutated effector domain of AlsR, AlsR_82–302S100A, was successfully produced in E. coli with a Trx-Strep tag and purified via affinity chromatography and gel-permeation chromatography (Fig. 2 ▶). The AlsR_82–302S100A protein was crystallized using the sitting-drop vapour-diffusion method. After 13 d, diffracting crystals with a length of 0.26 mm were obtained (Fig. 3 ▶). A data set was collected from a single crystal on the ID23-1 beamline of the ESRF and resulted in 99% completeness to 2.6 Å resolution. The crystals belonged to the monoclinic space group C2, with unit-cell parameters a = 142.91, b = 74.39, c = 94.39 Å, β = 110.54°. Calculation of the Matthews parameter (Matthews, 1968 ▶) suggested the presence of four AlsR_82–302S100A molecules in the asymmetric unit (V _M = 2.37 ų Da⁻¹; solvent content = 0.4822).

The MOLREP self-rotation function (Vagin & Teplyakov, 2010 ▶) revealed three perpendicular twofold rotation axes, one of which coincides with the crystallographic axis. This indicated the arrangement of four AlsR_82–302S100A molecules in the asymmetric unit with D ₂ point-group symmetry (Fig. 4 ▶). From this result, we deduced a dimer-of-dimers oligomeric composition of AlsR in the crystal. This form was also found for BenM, the closest relative of AlsR within the LTTR family (Clark et al., 2004 ▶). The obtained data provide the basis for the thorough understanding of a central regulator of fermentation in the Gram-positive model bacterium B. subtilis at the atomic level.

Figure 4.

Stereographic projection of a MOLREP (Vagin & Teplyakov, 2010 ▶) self-rotation function calculated in a search for twofold rotation (χ = 180°). Strong contour lines indicate intersections of rotation axes. Lines of latitude as well as lines of longitude are 10° apart. The twofold axis along y (red) corresponds to the symmetry of space group C2. Two strong NCS axes (purple and blue) are perpendicular to each other and to a third axis which coincides with the crystallographic symmetry. Together, they indicate the presence of D ₂ point-group symmetry in the asymmetric unit, which is likely to arise from a dimer of dimers of AlsR_82–302S100A. The weaker axes (cyan) might indicate pseudo-twofold symmetries that relate the two domains of AlsR to each other.