Portal de Periódicos da CAPES

Structure of Human Phytanoyl-CoA 2-Hydroxylase Identifies Molecular Mechanisms of Refsum Disease*

2005; Elsevier BV; Volume: 280; Issue: 49 Linguagem: Inglês

10.1074/jbc.m507528200

1083-351X

M.A. McDonough, K.L. Kavanagh, Danica Butler, Timothy Searls, Udo Oppermann, Christopher J. Schofield,

Cancer, Hypoxia, and Metabolism

Refsum disease (RD), a neurological syndrome characterized by adult onset retinitis pigmentosa, anosmia, sensory neuropathy, and phytanic acidaemia, is caused by elevated levels of phytanic acid. Many cases of RD are associated with mutations in phytanoyl-CoA 2-hydroxylase (PAHX), an Fe(II) and 2-oxoglutarate (2OG)-dependent oxygenase that catalyzes the initial α-oxidation step in the degradation of phytenic acid in peroxisomes. We describe the x-ray crystallographic structure of PAHX to 2.5 Å resolution complexed with Fe(II) and 2OG and predict the molecular consequences of mutations causing RD. Like other 2OG oxygenases, PAHX possesses a double-stranded β-helix core, which supports three iron binding ligands (His175, Asp177, and His264); the 2-oxoacid group of 2OG binds to the Fe(II) in a bidentate manner. The manner in which PAHX binds to Fe(II) and 2OG together with the presence of a cysteine residue (Cys191) 6.7 Å from the Fe(II) and two further histidine residues (His155 and His281) at its active site distinguishes it from that of the other human 2OG oxygenase for which structures are available, factor inhibiting hypoxia-inducible factor. Of the 15 PAHX residues observed to be mutated in RD patients, 11 cluster in two distinct groups around the Fe(II) (Pro173, His175, Gln176, Asp177, and His220) and 2OG binding sites (Trp193, Glu197, Ile199, Gly204, Asn269, and Arg275). PAHX may be the first of a new subfamily of coenzyme A-binding 2OG oxygenases. Refsum disease (RD), a neurological syndrome characterized by adult onset retinitis pigmentosa, anosmia, sensory neuropathy, and phytanic acidaemia, is caused by elevated levels of phytanic acid. Many cases of RD are associated with mutations in phytanoyl-CoA 2-hydroxylase (PAHX), an Fe(II) and 2-oxoglutarate (2OG)-dependent oxygenase that catalyzes the initial α-oxidation step in the degradation of phytenic acid in peroxisomes. We describe the x-ray crystallographic structure of PAHX to 2.5 Å resolution complexed with Fe(II) and 2OG and predict the molecular consequences of mutations causing RD. Like other 2OG oxygenases, PAHX possesses a double-stranded β-helix core, which supports three iron binding ligands (His175, Asp177, and His264); the 2-oxoacid group of 2OG binds to the Fe(II) in a bidentate manner. The manner in which PAHX binds to Fe(II) and 2OG together with the presence of a cysteine residue (Cys191) 6.7 Å from the Fe(II) and two further histidine residues (His155 and His281) at its active site distinguishes it from that of the other human 2OG oxygenase for which structures are available, factor inhibiting hypoxia-inducible factor. Of the 15 PAHX residues observed to be mutated in RD patients, 11 cluster in two distinct groups around the Fe(II) (Pro173, His175, Gln176, Asp177, and His220) and 2OG binding sites (Trp193, Glu197, Ile199, Gly204, Asn269, and Arg275). PAHX may be the first of a new subfamily of coenzyme A-binding 2OG oxygenases. In humans, the plasma level of the diet-derived isoprenoid, phytanic acid, is normally low ( 95% purity by SDS-PAGE analysis. Electrospray ionization mass spectrometry revealed that the mass of the purified PAHX was consistent with loss of the N-terminal methionine from the predicted amino acid sequence (observed, 35,436 Da; calculated without N-terminal methionine, 35,435 Da). Amino acid sequence analysis by Edman degradation confirmed the identity of the protein and loss of the N-terminal methionine (observed, STGISS). The activity of purified PAHX was confirmed for 2OG turn-over using the reported assay monitoring release of 14CO2 (32Lee H.-J. Lloyd M.D. Harlos K. Schofield C.J. Biochem. Biophys. Res. Commun. 2000; 267: 445-448Crossref PubMed Scopus (28) Google Scholar).Selenomethionine-substituted Protein Expression—Selenomethionine (SeMet)-substituted PAHX was produced using a metabolic inhibition protocol and LeMaster medium supplemented with 50 mg/liter l-selenomethionine. Selenomethionine incorporation was >95% by electrospray ionization mass spectrometry (observed, 35,858 Da; calculated, 35,859 Da).Crystallization—Crystallization conditions were initially sought using high throughput robotic screening methods at the Oxford Protein Production Facility (33Walter T.S. Diprose J. Brown J. Pickford M. Owens R.J. Stuart D.I. Harlos K. J. Appl. Crystallogr. 2003; 36: 308-314Crossref Scopus (92) Google Scholar). Optimization of the initial crystallization condition was performed using the hanging drop vapor diffusion method in VDX™ plates (Hampton Research, Aliso Viejo, CA). Hanging drops containing 2 μl of 5.2 mg/ml PAHX and 2 μl of well solution were suspended over 500 μl of well solution containing 21% polyethylene glycol 3350, 0.3 m triammonium citrate, pH 7.1, at 18 °C. Crystallization with 1 mm iron(II) sulfate and 2 mm 2OG in an anaerobic environment (Bell Technologies glove box under an argon atmosphere) produced square, plate-shaped crystals over a period of 3 weeks to a maximum size of 200 μm × 100 μm × 50 μm. SeMet PAHX crystals were grown aerobically using the same conditions, except that iron(II) sulfate and 2OG were substituted with zinc(II) chloride and N-oxalylglycine.Crystallographic Data Collection and Structure Solution—A single crystal, anaerobically grown (200 × 100 × 50 μm) was transferred to cryoprotectant (1:7 glycerol/well solution) and immediately cryocooled in liquid nitrogen. Native data were collected at 100 K using beamline 10.1 (34Cianci M. Antonyuk S. Bliss N. Bailey M.W. Buffey S.G. Cheung K.C. Clarke J.A. Derbyshire G.E. Ellis M.J. Enderby M.J. Grant A.F. Holbourn M.P. Laundy D. Nave C. Ryder R. Stephenson P. Helliwell J.R. Hasnain S.S. J. Synchrotron Radiat. 2005; 12: 455-466Crossref PubMed Scopus (42) Google Scholar) of the Synchrotron Radiation Source (SRS, Daresbury, UK) equipped with a MAR 225CCD detector. The native data were processed with MOSFLM and SCALA of the CCP4 suite version 5.0.2 (35Project Collaborative Computational Acta Crystallogr. Sect. D Biol. Crystallogr. 1994; 50: 760-763Crossref PubMed Scopus (19707) Google Scholar, 36Leslie A.G.W. Joint CCP4 and ESF-EAMCB Newsletter on Protein Crystallography. 1992; 26Google Scholar) with an I222 lattice (TABLE ONE). Calculation of a Matthews coefficient of 2.1 Å3/dalton implied a single monomer in the asymmetric unit. Attempts at molecular replacement and isomorphous replacement were unsuccessful, so crystals were produced from SeMet-substituted protein. A single SeMet crystal (120 μm × 70 μm × 5 μm) was transferred to cryoprotectant (1:9 glycerol/well solution) before cryocooling in liquid nitrogen. Single wavelength anomalous dispersion (SAD) data were collected at 100 K and at the selenium peak wavelength at beamline X10A (Swiss Light Source, Villigen, Switzerland). Data were integrated with HKL2000 version 1.98.0 (37Otwinowski Z. Minor W. Methods Enzymol. 1997; 276: 307-326Crossref Scopus (38368) Google Scholar) and merged, and anomalous differences were analyzed with XPREP (38Bruker A. XPREP: Data Preparation and Reciprocal Space Exploration, Version 6.14/UNIX. Bruker Nonius, Madison, WI2003Google Scholar). The substructure was solved with SHELXD of the SHELX-97 suite (39Sheldrick G.M. Schneider T.R. Methods Enzymol. 1997; 277: 319-343Crossref PubMed Scopus (1874) Google Scholar) using anomalous differences to 3.5 Å in space group I222, and the sites and chirality were confirmed with SHELXE. The substructure (top seven selenium sites) was refined with SHARP version 2.0.4 (40delaFortelle E. Bricogne G. Methods Enzymol. 1997; 276: 472-494Crossref PubMed Scopus (1796) Google Scholar) against the SAD data, and two additional selenium sites were identified. Phases were calculated to 3 Å using SHARP, and the combined anaerobic native and SAD data sets result in a figure of merit of 0.33. Phases were improved by density modification using RESOLVE version 2.08 (41Terwilliger T. Acta Crystallogr. Sect. D Biol. Crystallogr. 2000; 56: 965-972Crossref PubMed Scopus (1631) Google Scholar) to 2.7 Å with a final figure of merit of 0.85, resulting in core secondary structure elements and a 7σ peak in the core region attributable to the iron atom clearly visible in the electron density.TABLE ONECrystallographic data and structure statisticsSeMet (SLS beamline X10A)Native anaerobic (SRS Daresbury beamline 10.1)Data collectionWavelength (Å)0.97900.9800Resolution (Å)50-2.6 (2.69-2.6)30-2.5 (2.64-2.50)Space groupI222I222Unit cell (Å)a = 67.6, b = 85.3, c = 97.2a = 67.9, b = 86.7, c = 97.5Reflections observed/unique31,391/700351,212/9583Mean I/σ (I)22.9 (7.9)13.5 (2.0)Rmerge0.078 (0.428)0.119 (0.841)Completeness (%)78.5 (17.7)94.2 (97.2)Redundancy4.5 (1.7)5.3 (5.2)RefinementResolution (Å)40-2.5Completeness (%)93.05Rcryst/Rfree0.20/0.27Root mean square deviation bond length (Å)0.012Root mean square deviation bond angle (degrees)1.360No. of atomsProtein (average Bfactor Å2)1961 (37.7)Iron (average Bfactor Å2)1 (40.1)2OG (average Bfactor Å2)10 (54.9)Waters (average Bfactor Å2)15 (44.4)Ramachandran statisticsCore 85.0%Allowed 11.6%Generous 0.5%Disallowed 0.0% Open table in a new tab The initial model was built using the program COOT version 0.0.31 (42Emsley P. Cowtan K. Acta Crystallogr. Sect. D Biol. Crystallogr. 2004; 60: 2126-2132Crossref PubMed Scopus (22815) Google Scholar) into composite omit maps calculated with RESOLVE. Before refinement commenced, 5% of the data were flagged for calculation of a free R-factor (Rfree). Initially, simulated annealing was performed in CNS version 1.1 using combined phases calculated from the model and phase probabilities from SHARP. Iterative refinement using CNS version 1.1 and model building using COOT continued until Rfree was below 30%. At this stage, restrained TLS refinement was performed using REFMAC5 (43Murshudov G.N. Vagin A.A. Dodson E.J. Acta Crystallogr. D Biol. Crystallogr. 1997; 53: 240-255Crossref PubMed Scopus (13779) Google Scholar) with only the model phases. Iterative rounds of manual refitting and crystallographic refinement using the programs COOT and REFMAC5 continued until Rfree was no longer improved. Statistics are given in TABLE ONE.RESULTS AND DISCUSSIONThe PAHX Structure—PAHX crystallizes as a monomer in space group I222 with one molecule per asymmetric unit. The solvent content of the crystals is 40%, and the crystals contain large solvent channels that coincide with access to the active site. Three disordered loops that border the active site face this solvent channel and include residues 165-172, 223-233, and 303-318. Other disordered regions are near the N terminus and comprise residues 31-42 and 51 and 52. The most extensive contact between monomers in the lattice, 841 Å2 per monomer, is a ∼18-Å-long extended pair of antiparallel β-strands (β-3 and β-4) in which β-4 is antiparallel to β-4 of a symmetry-related molecule along a crystallographic 2-fold axis. Another significant lattice contact involves loop 240-244, which packs against Lys321 of the C-terminal helix, and residues 288-291 of one symmetry-related molecule and residue 94 of another.The PAHX fold is a mixed α-β structure composed of a major and a minor β-sheet surrounded by five α-helices and four 310 helices (Fig. 3). As in other 2OG oxygenases for which structures are available (30Valegard K. Terwisscha van Scheltinga A.C. Lloyd M.D. Hara T. Ramaswamy S. Perrakis A. Thompson A. Lee H.-J. Baldwin J.E. Schofield C.J. Haidu J. Andersson I. Nature. 1998; 394: 805-809Crossref PubMed Scopus (312) Google Scholar, 31Elkins J.M. Hewitson K.S. McNeil L.A. Seibel J.F. Schlemminger I. Pugh C.W. Ratcliffe P.J. Schofield C.J. J. Biol. Chem. 2003; 278: 1802-1806Abstract Full Text Full Text PDF PubMed Scopus (327) Google Scholar, 44Elkins J.M. Ryle M.J. Clifton I.J. Hotopp J.C.D. Lloyd M.D. Burzlaff N.I. Baldwin J.E. Hausinger R.P. Roach P.L. Biochemistry. 2002; 41: 5185-5192Crossref PubMed Scopus (190) Google Scholar, 45Muralidharan F.N. Muralidharan V.B. Biochim. Biophys. Acta. 1985; 835: 36-40Crossref PubMed Scopus (16) Google Scholar, 46Zhang Z.-H. Ren J.S. Stammers D.K. Baldwin J.E. Harlos K. Schofield C.J. Nat. Struct. Biol. 2000; 7: 127-133Crossref PubMed Scopus (250) Google Scholar, 47Wilmouth R.C. Turnbull J.J. Welford R.W.D. Clifton I.J. Prescott A.G. Schofield C.J. Structure. 2002; 10: 93-103Abstract Full Text Full Text PDF PubMed Scopus (262) Google Scholar, 48Clifton I.J. Hsueh L.-C. Baldwin J.E. Harlos K. Schofield C.J. Eur. J. Biochem. 2001; 268: 6625-6636Crossref PubMed Scopus (98) Google Scholar, 49Chance M.R. Bresnick A.R. Burley S.K. Jiang J.S. Lima C.D. Sali A. Almo S.C. Bonanno J.B. Buglino J.A. Boulton S. Chen H. Eswar N. He G. Huang R. Ilyin V. McMahan L. Pieper U. Ray S. Vidal M. Wang L.K. Protein Sci. 2002; 11: 723-738Crossref PubMed Scopus (148) Google Scholar), the core of the protein consists of a DSBH fold composed of eight β-strands (Fig. 4 and supplemental Fig. 2S). However, in PAHX, only seven β-strands of the DSBH are apparent: β-6, β-8, β-9, β-10, β-11, β-12, and β-13, here defined as β-strands I, III, IV, V, VI, VII, and VIII, respectively. In PAHX, the residues that correspond to DSBH β-strand II (β-7) (residues 165-172) in other 2OG oxygenases are disordered and left out of the model. This strand is probably involved in substrate binding (see below). Four additional β-strands (β-1, β-2, β-5, and β-15) hydrogen-bond antiparallel to the core DSBH, resulting in the eight-stranded major β-sheet (Fig. 3). It is interesting to note that both β-strands β-5 and β-15 are paired antiparallel to DSBH β-strand I (β-6), a unique feature among the 2OG oxygenase family.FIGURE 3Stereoview from the crystal structure of the PAHX-Fe(II)-2OG complex as a ribbon representation. The conserved eight stranded DSBH core found in all Fe(II) and 2OG-dependent oxygenases is colored yellow. Only seven strands (β-6, β-13, β-8, and β-11 of the major sheet and β-9, β-10, and β-12 of the minor sheet) are observed in PAHX, since the residues making up the DSBH β-strand II are disordered. Additional β-strands attached to the major β-sheet are colored slate blue. The Fe(II)-binding residues and 2OG C5′-carboxylate interacting arginine of PAHX along with the 2OG co-substrate and Fe(II) cofactor are shown as a stick representation. α- and 310-helices are shown in red.View Large Image Figure ViewerDownload Hi-res image Download (PPT)FIGURE 4Comparison of the PAHX structure with other 2OG oxygenases. A, PAHX (Protein Data Bank code 2A1X); B, FIH (Protein Data Bank code 1H2N) (31Elkins J.M. Hewitson K.S. McNeil L.A. Seibel J.F. Schlemminger I. Pugh C.W. Ratcliffe P.J. Schofield C.J. J. Biol. Chem. 2003; 278: 1802-1806Abstract Full Text Full Text PDF PubMed Scopus (327) Google Scholar); C, DAOCS (Protein Data Bank code 1DCS) (30Valegard K. Terwisscha van Scheltinga A.C. Lloyd M.D. Hara T. Ramaswamy S. Perrakis A. Thompson A. Lee H.-J. Baldwin J.E. Schofield C.J. Haidu J. Andersson I. Nature. 1998; 394: 805-809Crossref PubMed Scopus (312) Google Scholar); D, CAS (Protein Data Bank code 1DS1) (46Zhang Z.-H. Ren J.S. Stammers D.K. Baldwin J.E. Harlos K. Schofield C.J. Nat. Struct. Biol. 2000; 7: 127-133Crossref PubMed Scopus (250) Google Scholar); E, proline-3-hydroxylase (Protein Data Bank code 1E5S) (48Clifton I.J. Hsueh L.-C. Baldwin J.E. Harlos K. Schofield C.J. Eur. J. Biochem. 2001; 268: 6625-6636Crossref PubMed Scopus (98) Google Scholar). The Fe(II)-binding residues and Lys/Arg 2OG C5′-carboxylate-binding residues are shown as sticks.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Fe(II) Binding Site—The Fe(II)-containing active site is located between one end of the β-sheets that form the DSBH core in a manner similar to other 2OG oxygenases (30Valegard K. Terwisscha van Scheltinga A.C. Lloyd M.D. Hara T. Ramaswamy S. Perrakis A. Thompson A. Lee H.-J. Baldwin J.E. Schofield C.J. Haidu J. Andersson I. Nature. 1998; 394: 805-809Crossref PubMed Scopus (312) Google Scholar, 31Elkins J.M. Hewitson K.S. McNeil L.A. Seibel J.F. Schlemminger I. Pugh C.W. Ratcliffe P.J. Schofield C.J. J. Biol. Chem. 20