Crystallization and preliminary crystallographic analysis of the nickel-responsive regulator NikR from Pyrococcus horikoshii

The nickel-responsive repressor from Pyrococcus horikoshii OT3 (PhNikR) has been crystallized in the apo form (PhNikR-apo) and two nickel-bound forms (PhNikR-Ni-1 and PhNikR-Ni-2). The PhNikR-apo crystals belong to space group P2₁, with unit-cell parameters a = 75.78, b = 54.32, c = 77.28 Å, β = 116.07°, and diffract to 2.2 Å. The PhNikR-Ni-1 crystals belong to space group P4₁2₁2, with unit-cell parameters a = b = 99.89, c = 97.98 Å, and diffract to 3.0 Å and the PhNikR-Ni-2 crystals belong to space group P3₁21 or P3₂21, with unit-cell parameters a = b = 109.95, c = 79.0 Å, and diffract to 2.1 Å. The crystals obtained were suitable for detailed structural studies.

1. Introduction

Nickel is an essential nutrient for microorganisms and is critical for survival under anaerobic conditions (Maroney, 1999). Two types of nickel permeases, single-component and multi-component, have been discovered. Single-component permeases related to HoxN from Ralstonia eutropha have been identified in Gram-negative and Gram-positive bacteria, while high-affinity multi-component permeases related to the ATP-dependent NikABCDE permease from Escherichia coli (Navarro et al., 1993) have been identified in both bacteria and archaea (Eitinger & Mandrand-Berthelot, 2000). The transcription of the nikABCDE operon is repressed by the nickel-regulatory protein NikR at high intracellular nickel concentrations (De Pina et al., 1999).

NikR belongs to the ribbon–helix–helix family of transcription factors and functions as a homotetramer (Chivers & Sauer, 1999). The crystal structure of apo-form E. coli NikR (EcNikR) shows that the molecule consists of a central tetrameric domain (TD) flanked by dimeric DNA-binding domains (DBD; Schreiter et al., 2003). The TD binds Ni²⁺ ions avidly with a K_d of 7 pM (Chivers & Sauer, 2002). Four similar metal-binding sites are located at the tetramerization interface and the coordination of nickel at these sites is square planar (Schreiter et al., 2003). Full occupation of these sites results in binding to the operator site of the nikABCDE promoter with 30 nM affinity; however, with a 20–50 µM excess of nickel the DNA-binding affinity increases dramatically to a K_d of 15 pM, pointing to the presence of a second low-affinity nickel-binding site (Chivers & Sauer, 2000, 2002).

The most intriguing questions of how NikR senses nickel at low concentrations, how the nickel coordination responds to DNA binding and what the role of nickel is in DNA recognition remain unanswered. In order to resolve some of the remaining questions further structural studies are required. Here, we report the crystallization and preliminary crystallographic studies of a product of the Pyrococcus horikoshii OT3 gene PH0601 (PhNikR) in the apo form (PhNikR-apo) and two nickel-bound forms (PhNikR-Ni-1 and PhNikR-Ni-2) prepared with a protein:nickel ratio of 1:1. Crystals soaked in solution with high nickel content (PhNikR-Ni-2h) and with high nickel content and phosphate ions (PhNikR-Ni-2h-PO₄) were also prepared and characterized.

2. Materials and methods

2.1. Expression and purification

The protocols used for the expression and purification of PhNikR were similar to those described for phosphopantetheine adenylyltransferase by Takahashi et al. (2004).

Dynamic light-scattering measurements of the purified protein were performed at 291 K using DynaPro MS/X (Protein Solution) and the data were analyzed using DYNAMICS software (Protein Solutions).

2.2. Crystallization

Initial crystallization conditions were obtained by screening at a temperature of 298 K in 96-well plates using the sitting-drop vapour-diffusion method with Crystal Screens I and II (Hampton Research; Jancarik & Kim, 1991; Cudney et al., 1994). 1 µl drops of protein solution (10 mg ml⁻¹ PhNikR, 49 mM NaCl, 20 mM Tris–HCl pH 8.0, with or without 0.65 mM NiCl₂) mixed with 1 µl reservoir solution were equilibrated against 100 µl reservoir solution. Some conditions produced small crystals. Based on the conditions producing these crystals, grid screens with variations in precipitant concentrations, salt types and concentrations and pH were prepared and tested. Finally, we obtained one type of diffraction-quality crystals for apo protein and two types of crystals for nickel-bound protein. The PhNikR-apo crystals of thin elongated tetragonal prism shape (Fig. 1a) grew to dimensions of 0.01 × 0.02 × 0.5 mm in 2–3 d using reservoir solution containing 21%(v/v) PEG 400, 200 mM magnesium chloride hexahydrate and 100 mM HEPES buffer pH 7.5. PhNikR-Ni-1 crystals of tetragonal bipyramid shape (Fig. 1b) appeared in 2–3 d and grew to dimensions of 0.4 × 0.4 × 0.4 mm in two weeks using a reservoir containing 30 mM lithium sulfate monohydrate and 0.6%(w/v) PEG 8000. Finally, PhNikR-Ni-2 crystals of hexagonal prism shape (Fig. 1c) appeared after 1–2 weeks and grew to dimensions of 0.15 × 0.15 × 0.4 mm in one month using reservoir solution containing 4%(w/v) PEG 8000, 4%(v/v) ethylene glycol and 50 mM HEPES buffer pH 7.5.

Figure 1 Photomicrographs of (a) PhNikR-apo, (b) PhNikR-Ni-1 and (c) PhNikR-Ni-2 crystals.

2.3. Crystal soaking

Previous studies have indicated that the EcNikR molecule contains two types of nickel-binding sites: low-affinity and high-affinity sites (Chivers & Sauer, 2000, 2002). The crystal structure of the nickel-bound C-­terminal tetramerization domain of EcNikR has confirmed the presence of four high-affinity sites, one per subunit (Schreiter et al., 2003). The position of the low-affinity site remains unknown. As an increase of nickel concentration resulted in heavy precipitation of PhNikR, we were only able to obtain PhNikR-Ni-1 and PhNikR-Ni-2 crystals at a protein:nickel ratio of 1:1. In order to locate the low-affinity nickel-binding sites, a PhNikR-Ni-2 crystal was soaked for 1.5 h in reservoir solution with the addition of 20 mM NiCl₂. The X-­ray diffraction pattern of the soaked crystal (PhNikR-Ni-2h) was comparable to that of the PhNikR-Ni-2 crystal (Table 1).

Table 1
Crystal parameters and data-collection statistics

Values in parentheses are for the last shell.

Crystal	PhNikR-apo	PhNikR-Ni-1	PhNikR-Ni-2	PhNikR-Ni-2h		PhNikR-Ni-2h-PO4
Crystal parameters
Unit-cell parameters
a (Å)	75.78	99.89	109.95	110.21		108.54
b (Å)	54.32	99.89	109.95	110.21		108.54
c (Å)	77.28	97.98	79.00	78.60		78.82
β (°)	116.07
Space group	P21	P41212	P3121 or P3221	P3121 or P3221		P3121 or P3221
Z	8	16	12	12		12
Solvent content (%)	62.4	68.2	71.8	71.8		70.9
Data collection
Temperature (K)	100	100	100	100	100	100
X-ray source	BL26B1	BL26B1	BL26B1	BL26B1	BL26B1	Rigaku FR-D
Detector	R-AXIS V	R-AXIS V	R-AXIS V	R-AXIS V	R-AXIS V	R-AXIS V
Wavelength (Å)	1.00	1.00	1.00	1.00	1.4851	1.54178
Resolution (Å)	30–2.2 (2.28–2.2)	30–3.0 (3.11–3.0)	30–2.1 (2.18–2.1)	30–2.1 (2.18–2.1)	30–2.4 (2.49–2.4)	30–3.0 (3.11–3.0)
Unique reflections	28283	18601	32158	32178	21834	10353
Redundancy	3.1 (2.5)	9.4 (5.6)	8.3 (5.9)	5.2 (3.5)	6.6 (6.1)	2.6 (2.0)
Completeness (%)	97.8 (94.7)	99.0 (98.8)	99.3 (99.2)	98.6 (96.1)	99.9 (99.9)	93.3 (84.3)
Rsym (%)	5.3 (28.2)	6.5 (29.2)	6.3 (28.2)	4.2 (29.7)	5.6 (33.7)	6.5 (36.3)
〈I/σ(I)〉	21.9 (4.5)	20.8 (4.8)	24.8 (5.5)	23.1 (3.5)	18.5 (5.8)	26.1 (3.8)

K-edge X-ray absorption spectroscopy (XAS) of Ni-bound EcNikR in the presence and absence of bound DNA shows that the nickel coordination is DNA-dependent (Carrington et al., 2003). It is most likely that the nickel coordination is influenced by the phosphate groups of bound DNA. In an attempt to simulate the possible effect of DNA phosphates, the PhNikR-Ni-2 crystals were soaked in a reservoir solution with the addition of 20 mM nickel chloride and 5 mM sodium phosphate. Unexpectedly, after 30 min soaking the crystals had completely dissolved. This phenomenon also occurred at increased concentrations of precipitant. In order to reduce the crystal damage, the concentrations of nickel chloride and sodium phosphate were reduced to 5 mM and 1 mM, respectively. The crystal was soaked for only 10 min and was immediately cryoprotected and flash-cooled. The resulting crystal (PhNikR-Ni-2h-PO₄) diffracted to 3 Å.

2.4. Data collection

For data collection, the crystals were soaked in cryoprotectant for a few seconds, mounted in nylon-fibre loops and flash-cooled in a dry nitrogen stream at 100 K. 36%(v/v) PEG 400 was used as a cryoprotectant for all crystals, although in the case of the nickel-containing crystals 0.65, 5 or 20 mM NiCl₂ was added to the solution. Complete data sets were collected at 100 K using synchrotron radiation at SPring-8 beamline BL26B1 or using Cu Kα radiation from a Rigaku FR-D rotating-anode generator (Table 1). All intensity data were indexed, integrated and scaled with DENZO and SCALEPACK from the HKL2000 program package (Otwinowski, 1993; Otwinowski & Minor, 1997). The crystal parameters and data-processing statistics are summarized in Table 1.

3. Results and discussion

Dynamic light-scattering measurements show that both apo and nickel-bound molecules of PhNikR are monodisperse with a molecular weight of 62 kDa. This suggests that PhNikR subunits may form tetramers both in solution and in crystals, similar to the EcNikR molecule (Schreiter et al., 2003), which shares 33.8% amino-acid sequence identity with PhNikR. Solvent-content calculations (Matthews, 1968) show that if four subunits (a possible homotetramer) of PhNikR-apo are assigned to the asymmetric unit of the crystal, the solvent content of the crystal will be 62.4%. In the case of the PhNikR-Ni-1 and PhNikR-Ni-2 crystals, assuming the presence of two protein subunits in the asymmetric unit (half a possible homotetramer) gives solvent contents of 68.2 and 71.8%, respectively. The crystals obtained are suitable for detailed structural studies of PhNikR.