Overexpression, purification, crystallization and preliminary X-ray diffraction analysis of the C-­terminal domain of Ss-LrpB, a transcription regulator from Sulfolobus solfataricus

The C-terminal domain of the transcriptional regulator Ss-LrpB from S. solfataricus was purified by affinity chromatography and crystallized. Crystals belong to space group P2₁2₁2. A complete data set was collected to a resolution of 2 Å.

Abstract

Ss-LrpB from Sulfolobus solfataricus P2 belongs to the bacterial/archaeal superfamily of Lrp-like (leucine-responsive regulatory protein-like) transcription regulators. The N-terminal DNA-binding domain contains a HTH motif and the C-terminal domain contains an αβ-sandwich (βαββαβ fold). The C-­terminal domain was overexpressed in Escherichia coli, purified and crystallized using the hanging-drop vapour-diffusion method. The crystals belong to space group P2₁2₁2, with unit-cell parameters a = 59.35, b = 74.86, c = 38.08 Å and a data set was collected to 2.0 Å resolution. Structure determination using a selenomethionine derivative is under way.

1. Introduction

Archaea are microorganisms constituting a domain of life that is distinct from the bacteria and eukarya (Woese et al., 1990 ▶). Most characterized archaeal species are extremophiles, thriving at extremes of temperature, pH, salt concentration etc. Sulfolobus solfataricus P2 is a hyperthermoacidophilic archaeon, growing optimally at 353 K and pH 2–4 (Zillig et al., 1980 ▶). Transcription in archaea appears to be typified by a eukaryal-like basal transcription apparatus, regulated mainly by bacteria-like helix–turn–helix (HTH) transcription regulators (Geiduschek & Ouhammouch, 2005 ▶).

Most of the characterized archaeal transcription regulators belong to the leucine-responsive regulatory protein (Lrp) family (Brinkman et al., 2003 ▶). This family is composed of both global and specific regulators. To date, few archaeal regulators with known physiological function have been described (Brinkman et al., 2002 ▶; Ouhammouch et al., 2003 ▶). Although it has been claimed that all specific Lrp-like regulators have a role in the regulation of amino-acid metabolism (Brinkman et al., 2003 ▶), this was refuted by the discovery that Ptr2 from Methanocaldococcus jannaschii activates a ferredoxin and a rubredoxin gene (Ouhammouch et al., 2003 ▶).

All Lrp-like proteins are small basic proteins composed of two domains connected by an extended flexible hinge. Crystal structures of two Lrp-like proteins, both archaeal, have been determined: LrpA from Pyrococcus furiosus (Leonard et al., 2001 ▶) and FL11 from Pyrococcus OT3 (Koike et al., 2004 ▶). The N-terminal DNA-binding domain contains an HTH motif and the C-terminal domain contains an αβ-sandwich (βαββαβ fold). This C-terminal domain is also called the regulation of amino-acid metabolism (RAM) domain and resembles the ACT domain, an allosteric regulatory domain of many metabolic enzymes (Ettema et al., 2002 ▶). It has been shown to be responsible for oligomerization (Leonard et al., 2001 ▶; Chen et al., 2001 ▶) and in the case of Escherichia coli Lrp for ligand binding and activation of transcription (Platko & Calvo, 1993 ▶).

Ss-LrpB from S. solfataricus binds its own control region (Peeters et al., 2004 ▶), as do most Lrp-like proteins. This binding occurs cooperatively at three regularly spaced 15 bp binding sites with as consensus the palindromic sequence 5′-TTGYAWWWWWTRCAA-3′ (Y = pyrimidine, R = purine, W = weak bp). This is in contrast to most other Lrp-like proteins, which recognize clusters of strongly degenerate binding sites (Wang & Calvo, 1993 ▶; Ouhammouch & Geiduschek, 2001 ▶). It has also been suggested that a secondary DNA structure might be recognized rather than a consensus sequence (for example, a pattern of repeating weak base pairs; Calvo & Matthews, 1994 ▶; Beloin et al., 2000 ▶). Ss-LrpB is therefore an interesting candidate for function prediction: the consensus binding sequence is being used to scan all promoter/operator regions in the S. solfataricus P2 genome. This might also provide useful information regarding a potential ligand.

Here, we report the cloning, overexpression, purification and crystallization of the C-terminal domain of Ss-LrpB comprising residues 69–147 of the full-length protein, omitting the last seven residues. This approach was chosen because full-length Ss-LrpB has a strong tendency to aggregate in solution. We aim towards a better understanding of the molecular regulation mechanism, protein oligomerization, potential cofactor binding and thermostability of this transcription regulator.

2. Material and methods

2.1. Cloning and overexpression

A DNA fragment containing residues 205–441 of the Ss-lrpB gene was amplified by PCR with pET24aSs-lrpB as template (Peeters et al., 2004 ▶). The forward primer 5′-GGAATTCCATATGAATGTCTCATTAGTAATGG-3′ introduced an NdeI restriction site (indicated in italics) and placed an ATG start codon in front of the first codon that translates into the C-terminal domain of Ss-LrpB. The reverse primer (5′-CCGCTCGAGCTTATATCTCTCAATTATCCTTGAAATC-3′) introduced an XhoI restriction site (indicated in italics). The restricted PCR fragment was ligated into the expression vector pET24a, previously digested with NdeI and XhoI. Using this cloning site, the protein was expressed including an additional C-terminal His-tag sequence, LEHHHHHH, allowing Ni²⁺ ion affinity purification (see below). This construct was designated pET24aSs-lrpBC-term. The DNA sequence of the insert was confirmed by sequence analysis with an ABI PRISM 3100 Genetic Analyzer (Applied Biosciences). The recombinant plasmid was transformed into E. coli BL21(DE3) in order to overexpress the protein, named Ss-LrpBC-term.

A 2.5 l culture was grown at 303 K in rich medium complemented with 30 µg ml⁻¹ kanamycin. Overexpression of Ss-LrpBC-term was induced at a cell density of 6 × 10⁸ cells ml⁻¹ by adding 1.0 mM IPTG followed by overnight growth. Cells were collected by centrifugation, resuspended in 35 ml extraction buffer (20 mM sodium phosphate buffer, 0.2 M NaCl, 40 mM imidazole pH 7.4) and disrupted by sonication for 1 h with a VibraCell sonicator (Bioblock Scientific) at 20% of the maximal amplitude in a cell cooled at 277 K. Cell debris was removed by ultracentrifugation at 23 000 rev min⁻¹ for 1 h (Beckman L8-70 ultracentrifuge, Ti-60 rotor).

2.2. Protein purification

The crude extract was incubated at 328 K for 5 min followed by centrifugation of the denatured E. coli proteins at 10 000 rev min⁻¹ for 10 min (microcentrifuge). In contrast to full-length Ss-LrpB, Ss-­LrpBC-term proved to be sensitive to temperatures exceeding 333 K. The supernatant was loaded onto a HisTrap HP column (1 ml, Amersham Biosciences) allowing Ni²⁺ ion affinity chromatography purification of the His-tagged protein. Equilibration of the column was performed with 20 mM sodium phosphate buffer, 0.2 M NaCl, 20 mM imidazole pH 7.4. Elution of Ss-LrpBC-term was obtained by applying a linear gradient from 20 to 500 mM imidazole. The purest fractions were pooled.

2.3. Crystallization

Ss-LrpBC-term was dialyzed against 20 mM sodium phospate buffer pH 7.4 and concentrated to a final concentration of 4 mg ml⁻¹ as determined by a MicroBCA Protein assay (Pierce). Higher concentrations resulted in protein precipitation. The initial crystallization conditions were screened using the hanging-drop vapour-diffusion method with sparse-matrix crystallizations kits (Crystal Screen II, Hampton Research, Riverside, CA, USA). Immediately prior to setting up crystallization, the protein was centrifuged at 13 000 rev min⁻¹ for 10 min to remove precipitates (microcentrifuge). 2 µl of protein (4 mg ml⁻¹) was mixed with an equal amount of reservoir solution and equilibrated against 500 µl reservoir solution. The crystallization trays were kept at 293 K. Crystals appeared after 2 d in two different conditions (Table 1 ▶; Figs. 1 ▶ a and 1 ▶ b). Crystallization under these conditions was repeated and optimized. In many drops too much nucleation occurred, resulting in a large number of small crystals. In order to increase the crystal size these drops were hung above a reservoir of water for 2 d. After dissolution of the crystals the drops were equilibrated again against the same reservoir solution as before to which 50 µl water was added.

Table 1. Crystallization conditions.

Condition	I	II
Salt	100 mM NaH2PO4, 100 mM KH2PO4	0.5 M (NH4)2SO4
Buffer	100 mM MES pH 6.5	100 mM sodium citrate pH 5.6
Precipitant	2.0 M NaCl	1 M Li2SO4
Crystal dimensions (µm)	400 × 50 × 50	150 × 150 × 250

Figure 1.

(a) Crystals obtained from condition I. (b) Crystals obtained from condition II.

2.4. X-ray diffraction collection and processing

X-ray diffraction data were collected from a single crystal at 100 K using a MAR CCD detector (MAR Research, Norderstadt, Germany) at beamline BW7A at the DORIS storage ring, DESY, Hamburg. 206 images were collected at a wavelength of 0.9116 Å and with a rotation angle of 1°. The images were processed using the HKL programs DENZO and SCALEPACK (Otwinowski & Minor, 1997 ▶). The molecular-replacement trials were performed using PHASER (from CCP4; Read, 2001 ▶).

3. Results and discussion

Ss-LrpBC-term was overexpressed and purified from pET24aSs-lrpBC-term transformants of strain BL21 (DE3). A combination of heat treatment and affinity chromatography resulted in electrophoretically pure protein (3.2 mg per litre of culture) which has a theoretical molecular weight of 10.22 kDa and migrated accordingly on SDS–PAGE (Fig. 2 ▶). Crystallization of the protein fragment proved to be easy. Crystals were found in two conditions of the screening set (conditions I and II; Table 1 ▶). This probably indicates that the C-terminal part has a stable tightly packed structure, as for the LrpA from P. furiosus (Leonard et al., 2001 ▶). Indeed, the same secondary structure (RAM domain: βαββαβ fold) was predicted (3DPSSM; Kelley et al., 2000 ▶). The exact formulation of conditions I and II and the estimated dimensions of the crystals are given in Table 1 ▶. An attempt was made to optimize crystal growth under these conditions, but changing the protein, salt or precipitant concentration frequently resulted in too much nucleation.

Figure 2.

SDS–PAGE of a sample of purified Ss-LrpBC-term after heat treatment and Ni²⁺ ion affinity purification.

A crystal grown from condition I (Fig. 1 ▶ a) diffracted to 1.8 Å. PEG 400 was used as a cryoprotectant at a concentration of 30%(v/v). A data set to a resolution of 2 Å was collected from a single crystal. The crystals belong to the orthorhombic space group P2₁2₁2, with unit-cell parameters a = 59.35, b = 74.86, c = 38.08 Å. Data-collection information and statistics are given in Table 2 ▶. The values of the Matthews coefficient V _M (Matthews, 1968 ▶), determined assuming the presence of a dimer in the asymmetric unit, is 2.1 ų Da⁻¹. The corresponding value for the solvent content is 41%. Crystals grown from condition II (Fig. 1 ▶ b) were not stable when transferred into cryoprotectant solution, either PEG 400 or glycerol, at concentrations ranging from 10–40%(v/v). Other cryoprotectants were not tested owing to the limited availability of crystals.

Table 2. Data-collection statistics.

Values in parentheses correspond to the highest resolution shell.

Unit-cell parameters (Å, °)	a = 59.35, b = 74.86, c = 38.08, α = β = γ = 90
Space group	P21212
Wavelength (Å)	0.9116
Resolution (Å)	30–2.0 (2.07–2.00)
Total No. of reflections	96635
No. of unique reflections	11970
Completeness (%)	98.6 (98.7)
Redundancy	8.1 (8.1)
Percentage of reflections with I/σ(I) < 3 (%)	12.4 (36.6)
Rsym (%)	4.3 (23.5)
Mosaicity (°)	0.65

Ss-LrpBC-term shares 27.8% amino-acid identity (55.7% similarity) with the corresponding C-terminal part of LrpA and only 16.5% amino-acid identity (46.8% similarity) with the corresponding C-terminal part of FL11. Preliminary attempts at molecular replace­ment using the C-terminal part of the LrpA and FL11 as a model, using the dimer as well as the monomer, have not been successful. No clear solution was found. In order to solve the structure a SeMet derivative is being prepared.