Crystal structure of γ‐glutamyl phosphate reductase (TM0293) from <i>Thermotoga maritima</i> at 2.0 Å resolution

摘要

The TM0293 gene of Thermotoga maritima encodes γ-glutamyl phosphate reductase (GPR; EC 1.2.1.41), an enzyme that catalyzes the second step of proline biosynthesis, the reversible nicotinamide adenine dinucleotide phosphate (NADPH)-dependent reduction of L-γ-glutamyl phosphate to phosphate and L-glutamate 5-semialdehyde. The product then undergoes spontaneous cyclization to form 1-pyrroline-5-carboxylate.1 In higher organisms, such as plants and humans, GPR is the second domain of the bifunctional enzyme Δ-1-pyrroline-5-carboxylate synthetase (P5CS),2 which catalyzes the first two steps in proline biosynthesis. In plants, P5CS is the first enzyme in the metabolic pathway used for proline production in response to environmental factors, such as drought and salinity, that lead to osmotic stress.2 In humans, mutations in this enzyme result in developmental disorders characterized by joint hyperlaxity, skin hyperelasticity, cataracts, and mental retardation.3 Here, we report the crystal structure of T. maritima GPR at 2.0 Å resolution that represents the first structure of any GPR family member, a subfamily of the aldehyde dehydrogenase extended family.4 The GPR structure was determined using the automated high-throughput pipeline at the Joint Center for Structural Genomics (http://www.jcsg.org).5 The TM0293 structure (Fig. 1) was determined using the multiple-wavelength anomalous dispersion (MAD) method. Data collection, model, and refinement statistics are summarized in Table I. The final model includes TM0293 residues 2–415 and 299 water molecules. The Matthews coefficient (Vm) for TM0293 is 2.1 Å3/Da with an estimated solvent content of 52.3%.6 The Ramachandran plot produced by PROCHECK 3.47 shows that 92.7% of the residues lie in the most favored regions, 7.0% in additionally allowed regions, 0.0% in generously allowed regions, and 0.3% (His 390) in the disallowed regions. His 390 is the central residue in a classical γ-turn,8 with dihedral angles (ϕ = 75.8°, ψ = −61.4°) consistent with those expected for this type of secondary structural element. (A) Ribbon diagram of the TM0293 monomer with its three domains labeled; α-helices are in red and β-strands are in cyan. The proposed catalytic cysteine, Cys255, is represented in ball-and-stick, with the sulfur in yellow; a prominent difference density around the Cys sulfhydryl group suggests it may be in an oxidized form (density not shown). The NADPH and hinge regions are indicated by arrows, and the N and C termini are labeled. (B) The crystallographic TM0293 tetramer; each monomer is colored separately. (C) Diagram showing the secondary structure elements of TM0293 superimposed on its primary sequence. Helices are labeled H1 to H19. Strands are labeled according to their respective β-sheets: A (the oligomerization domain III), B (the cofactor binding domain I), C (the catalytic binding domain II) and D (the hinge). The locations of the hairpin loops, as well as the location of β turns and γ turns, are also depicted in the diagram. TM0293 was originally identified as a member of the aldehyde dehydrogenase family based on sequence homology. A structural similarity search, performed with the program DALI9 using the coordinates of TM0293, indicated that the closest known structure is the class 3 aldehyde dehydrogenase from rat (PDB ID: 1AD3),10 with a 4.4 Å root-mean-square deviation (RMSD) over 361 residues (87%), and a sequence identity of 17%. The relatively high RMSD is mostly caused by domain rotation, and can be reduced to less than 2.0 Å RMSD when each domain (the cofactor binding domain I and the catalytic domain II; see below) is superimposed separately. TM0293 consists of a single 415–amino acid polypeptide chain composed of 16 α-helices and 16 β-strands [Fig. 1(A, C)]. Like other aldehyde dehydrogenases, GPR consists of three distinct domains. The first two domains, the cofactor binding domain I (residues 1–223) and the catalytic domain II (residues 224–371), both have an