Purification, crystallization and crystallographic analysis of Dictyostelium discoideum phenylalanine hydroxylase in complex with dihydrobiopterin and Fe^III

Phenylalanine hydroxylase from D. discoideum has been crystallized. Diffraction data were collected to 2.07 Å resolution.

Abstract

Dictyostelium discoideum phenylalanine hydroxylase (DicPAH; residues 1–415) was expressed in Escherichia coli and purified for structural analysis. Apo DicPAH and DicPAH complexed with dihydrobiopterin (BH₂) and Fe^III were crystallized using 0.06 M PIPES pH 7.0, 26%(w/v) PEG 2000 by the hanging-drop vapour-diffusion method. Crystals of apo DicPAH and the DicPAH–BH₂–Fe^III complex diffracted to 2.6 and 2.07 Å resolution, respectively, and belonged to space group P2₁, with unit-cell parameters a = 70.02, b = 85.43, c = 74.86 Å, β = 110.12° and a = 70.97, b = 85.33, c = 74.89 Å, β = 110.23°, respectively. There were two molecules in the asymmetric unit. The structure of DicPAH has been solved by molecular replacement.

1. Introduction

The nonhaem iron-dependent enzyme phenylalanine hydroxylase (PAH; EC 1.14.16.1) catalyzes the hydroxylation of the aromatic amino acid l-phenylalanine (l-Phe) to l-tyrosine (l-Tyr) in the presence of the specific cofactor tetrahydrobiopterin (BH₄) and dioxygen (O₂). As l-Tyr is the precursor of the neurotransmitter dopamine, the catabolism of l-Phe plays an essential role in the development of the neural system (Kaufman, 1993 ▶). Disorder of the PAH system leads to the diseases phenylketonuria and hyperphenyl­alaninaemia, which result in brain-development problems unless treated immediately after birth (Scriver & Kaufman, 1995 ▶). Full-length mammalian PAHs consist of three domains: an N-terminal regulatory domain, a central catalytic domain and a short C-­terminal oligomerization domain for tetramer formation. The catalytic site includes a nonhaem Fe atom coordinated by a His-His-Glu catalytic triad that is highly conserved in aromatic amino-acid hydroxylases. Dysfunction of human PAH is mostly caused by mutations that are found in the catalytic domain. These mutations demonstrate different clinical, metabolic and enzymatic phenotypes (Erlandsen et al., 2004 ▶). Crystallographic studies of human and rat PAHs revealed the binding sites of the pterin cofactor and the substrate, and provided insights into the substrate-specificity and catalytic mechanism of the enzyme (Erlandsen et al., 1997 ▶; Fusetti et al., 1998 ▶; Kobe et al., 1999 ▶). Mammalian PAH responds with positive cooperativity to increased concentration of the substrate (Kaufman, 1993 ▶; Knappskog et al., 1996 ▶).

There are four possible BH₄ stereoisomers: l-erythro-BH₄, l-threo-BH₄, d-erythro-BH₄ and d-threo-BH₄. l-erythro-BH₄ is ubiquitous in animals as the natural cofactor for aromatic amino-acid hydroxylases, whereas the social amoeba Dictyostelium discoideum produces both l-erythro-BH₄ and d-threo-BH₄ (DH₄; Klein et al., 1990 ▶; Cho et al., 1999 ▶). Interestingly, Dictyostelium PAH (DicPAH) uses both as co­factors, although it exhibits higher activity with DH₄ than with l-­erythro-BH₄ (Siltberg-Liberles et al., 2008 ▶). This indicates that DicPAH possesses dual cofactor specificity (Choi et al., 2005 ▶). However, no structural information is available on the binding of DH₄ to PAH. In order to understand the enzyme-catalysis mechanism and dual cofactor specificity, DicPAH (residues 1–415) was purified and crystallized for structural analysis. Three-dimensional structures of DicPAH will provide a deeper understanding of the cofactor specificity of the hydroxylase reaction and its regulatory properties.

2. Experimental

2.1. Protein expression and purification

DicPAH (residues 1–415; DicPAHΔ415) was cloned into the bacterial expression vector pProEx HTa (Life Technologies, Carlsbad, California, USA) to produce a recombinant protein with an N-­terminal hexahistidine tag and a TEV protease cleavage site (MSYYHHHHHHDYDIPTTENLYFQG). Escherichia coli strain BL21 (DE3) was used for protein expression. 100 ml aliquots of an overnight culture were seeded into 1000 ml fresh Luria–Bertani (LB) medium (10 g Bacto tryptone, 5 g yeast extract, 10 g NaCl per litre of solution) containing ampicillin (100 µg ml⁻¹) at 310 K with vigorous shaking until the OD₆₀₀ reached 0.6. Protein expression was induced for 5 h with 0.4 mM isopropyl β-d-1-thiogalactoside (IPTG) at 303 K and cells were harvested by centrifugation (6000 rev min⁻¹, 6 min, 277 K). The harvested cells were washed with phosphate-buffered saline (PBS; 8 g NaCl, 0.2 g KCl, 1.44 g NaH₂PO₄ per litre of solution; pH 7.4), resuspended in lysis buffer (50 mM sodium phosphate pH 7.4, 500 mM NaCl, 5 mM imidazole) and disrupted by sonication. After centrifugation (1 h at 12 000 rev min⁻¹) at 277 K, the clear supernatant was filtered (qualitative filter paper, Advantec, Japan) and loaded onto an open column of nickel–NTA beads (Qiagen, Hilden, Germany) pre-equilibrated with binding buffer. The column was washed first with ten column volumes of binding buffer and then with ten column volumes of washing buffer (50 mM sodium phosphate pH 7.4, 500 mM NaCl, 20 mM imidazole). The recombinant DicPAHΔ415 was eluted with 50 mM Tris–HCl pH 8.0, 100 mM NaCl, 300 mM imidazole. Fractions containing DicPAHΔ415 were pooled, concentrated and exchanged into 50 mM Tris–HCl pH 8.0, 1 mM EDTA by ultrafiltration (Centriprep YM-50, Millipore, Bedford, Massachusetts, USA). The N-terminal hexahistidine tag was then removed using 0.02 mg TEV protease (Invitrogen, Carlsbad, California, USA) per milligram of protein at 277 K overnight, leaving artifactual residues GAMDPEF at the N-terminus. After removal of the His tag, the DicPAHΔ415 was further purified by anion-exchange chromatography on a Resource 15Q column (GE Healthcare, Piscataway, New Jersey, USA). The protein was eluted using a salt gradient and eluted at ∼200 mM NaCl pH 7.5. The fractions containing the DicPAHΔ415 protein were finally purified by gel-filtration chromatography on a Superdex 200 column (GE Healthcare, Piscataway, New Jersey, USA) in 20 mM Tris–HCl pH 8.0, 150 mM NaCl. The DicPAHΔ415 protein was pooled and concentrated to 10 mg ml⁻¹ in 20 mM Tris–HCl pH 8.0 by ultrafiltration (Centricon YM-30, Millipore Corporation, Bedford, Massachusetts, USA) for crystallization. The protein purity was examined by SDS–PAGE and native PAGE. The protein concentration was determined using the Bradford assay (Bradford, 1976 ▶).

2.2. Crystallization and data collection

Crystallization of apo DicPAHΔ415 was initially carried out using Crystal Screens I and II and Index Screen from Hampton Research (California, USA), Wizard Screens I and II and Cryo Screens I and II from Emerald BioStructures (Bainbridge Island, Washington, USA) and laboratory-made solutions using a microbatch crystallization method at 291 K. Drops containing equal volumes (1 µl) of protein solution (6 mg ml⁻¹ in 20 mM Tris–HCl pH 8.0) and screening solution were equilibrated under Al’s oil in a 72-well microbatch plate.

After crystals of apo DicPAHΔ415 were produced from the initial screenings, further screenings to find optimal crystallization conditions were performed using hanging-drop vapour-diffusion trials, varying the pH range, type of precipitant, precipitant concentration and volume of the drop. To obtain crystals of the DicPAHΔ415–BH₂ complex, BH₂ (the oxidized form of BH₄) was used as BH₄ is not stable under aerobic conditions. Apo DicPAHΔ415 was incubated with 2 mM BH₂ (Fluka, Steinheim, Switzerland) for 1 h on ice prior to setting up crystallization. The DicPAHΔ415–BH₂ complex crystals grew in the same condition as the apo DicPAHΔ415 crystals. To incorporate Fe^III into the protein, the crystals of DicPAHΔ415–BH₂ were soaked in well solution containing 5 mM Fe(NH₄)₂(SO₄)₂ (Sigma, Steinheim, Switzerland) for 1 h before flash-freezing. Crystals were flash-frozen in liquid nitrogen for data collection after soaking for 1 min in reservoir solution containing 12%(v/v) ethylene glycol. X-ray diffraction data were collected from apo DicPAHΔ415 and DicPAHΔ415–BH₂–Fe^III crystals using an ADSC Quantum 210 CCD detector with X-rays of wavelength 1.23985 Å on beamlines 4A and 6C1 of Pohang Accelerator Laboratory (PAL), Pohang, Republic of Korea, with exposure times of 30 s and 10 s for 1° oscillations at crystal-to-detector distances of 200 and 150 nm, respectively. All diffraction data were indexed, integrated and scaled using HKL-2000 (Otwinowski & Minor, 1997 ▶). Molecular replacement was performed using the program AMoRe (Collaborative Computational Project, Number 4, 1994 ▶; Navaza, 2001 ▶).

3. Results

Residues 1–415 of D. discoideum PAH (DicPAHΔ415), corresponding to the N-terminal regulatory and central catalytic domains, were expressed in E. coli and purified for crystallization by nickel-affinity, ion-exchange and gel-filtration chromatography. The molecular weight of the protein was estimated to be about 48 kDa from SDS–PAGE, which is similar to the theoretical molecular weight of 47.5 kDa (Fig. 1 ▶ a). The protein was separated by gel-filtration chromatography with an effective mass of 80 kDa, suggesting that the protein exists as a dimer in solution (Fig. 1 ▶ b). From the initial crystallization trials, dendritic apo DicPAHΔ415 crystals were found using the screening solution 0.1 M PIPES pH 6.5, 30%(w/v) PEG 4000 (Fig. 2 ▶ a). Apo DicPAHΔ415 crystals that were suitable for diffraction experiments were obtained by the hanging-drop vapour-diffusion method at 291 K in a drop containing 5 µl 6 mg ml⁻¹ protein solution and 5 µl of a mixture of 9 µl reservoir solution [0.06 M PIPES pH 7.0, 26%(w/v) PEG 2000] and 1 µl 0.1 M l-proline as an additive. The crystals grew to maximum dimensions of 0.3 × 0.4 × 0.2 mm in 3–4 d (Figs. 2 ▶ b and 2 ▶ c). Data were collected to 2.6 Å resolution from a single crystal of apo DicPAHΔ415 and were pro­cessed in the monoclinic space group P2₁, with unit-cell parameters a = 70.02, b = 85.43, c = 74.86 Å, β = 110.12° (Table 1 ▶). DicPAHΔ415–BH₂ complex crystals were obtained using the same condition as the apo DicPAHΔ415 crystals and grew to maximum dimensions of 0.3 × 0.3 × 0.3 mm in 3–4 d (Fig. 2 ▶ d). The crystals of the DicPAHΔ415–BH₂–Fe^III complex were produced by soaking the DicPAHΔ415–BH₂ complex crystals. The DicPAH–BH₂–Fe^III complex crystal diffracted to 2.07 Å resolution and belonged to space group P2₁, with unit-cell parameters a = 70.97, b = 85.33, c = 74.89 Å, β = 110.23° (Table 1 ▶). For both crystal forms it was clear that higher resolution data were available and we plan to collect these data as soon as possible. The asymmetric unit contained two DicPAHΔ415 molecules, with a Matthews coefficient V _M of 2.24 ų Da⁻¹ and an estimated solvent content of 45.23% (Matthews, 1968 ▶). The crystal of the DicPAHΔ415–BH2–Fe^III complex had essentially the same Matthews coefficient and therefore the same number of molecules in the asymmetric unit and the same percentage solvent content. The structure of DicPAHΔ415 could be determined by molecular replace­ment using the program AMoRe (Collaborative Computational Project, Number 4, 1994 ▶; Navaza, 2001 ▶). The human PAH catalytic domain (PDB code 1j8u; Andersen et al., 2001 ▶) was used as a search model for molecular replacement, as it shares 64% amino-acid sequence identity in the catalytic domain. After a rotation and translation search, a solution consisting of two catalytic domains with a correlation coefficient of 33.3 and an R factor of 52.2% could be found and was refined to a correlation coefficient of 53.8 and an R factor of 45.7% using the fitting function in AMoRe. The missing N-­terminal regulatory domain could be built into the electron densities calculated from the molecular-replacement solution (data not shown). Detailed discussion of the refined structures of both crystals will be published elsewhere.

Figure 1.

Purification of DicPAHΔ415. (a) The purity of the DicPAHΔ415 was confirmed by SDS–PAGE. (b) Analytical gel-filtration chromatography of DicPAHΔ415. A single peak was observed that was estimated using protein standards to correspond to the size of a dimer. The inset shows logK _d against log(molecular weight) for protein standards (circles) and DicPAHΔ415 (filled circle). K _d was calculated from the equation K _d = (V _e − V ₀)/(V _t − V ₀), where V _e, V _t and V ₀ represent the eluted volume, total volume and void volume, respectively. The standard proteins were ferritin (440 kDa), aldolase (158 kDa), BSA (67 kDa), ovalbumin (43 kDa), chymotrypsinogen A (25 kDa) and cytochrome c (12.4 kDa).

Figure 2.

Crystals of DicPAHΔ415. (a) Dendritic crystals obtained from initial screening. (b) Overlapping rod-shaped crystals obtained after optimization. (c) The best large single crystal of apo DicPAHΔ415 obtained after additive screening and controlling the drop size. (d) The best large single crystal complexed with BH₂ used for soaking with Fe^III and diffraction.

Table 1. Data statistics for DicPAHΔ415 crystals.

Values in parentheses are for the highest resolution shell.

	Apo DicPAHΔ415	DicPAHΔ415–BH2–FeIII
Resolution (Å)	50–2.60 (2.69–2.60)	50–2.07 (2.14–2.07)
Wavelength (Å)	1.23985	1.23985
Space group	P21	P21
Unit-cell parameters (Å, °)	a = 70.02, b = 85.43, c = 74.86, β = 110.12	a = 70.97, b = 85.33, c = 74.89, β = 110.23
No. of molecules per ASU	2	2
VM (Å3 Da−1)	2.24	2.24
Solvent content (%)	45.23	45.11
Unique reflections	25829 (2516)	50608 (4958)
Completeness (%)	99.2 (98.4)	99.1 (97.8)
Multiplicity	3.8 (3.8)	3.8 (3.7)
〈I/σ(I)〉	26.3 (27.3)	9.3 (9.9)
Rmerge† (%)	5.5 (9.8)	6.7 (59.8)

^† R _merge = , where I _i(hkl) is the ith intensity measurement of reflection hkl, including symmetry-related reflections, and 〈I(hkl)〉 is its average.