Crystal structure of AICAR transformylase IMP cyclohydrolase (TM1249) from Thermotoga maritima at 1.88 Å resolution
Year: 2008
Type: article
Abstract: The TM1249 gene (purH) of Thermotoga maritima encodes a bifunctional enzyme with 5-aminoimidazole-4-carboxamide ribonucleotide (AICAR) transformylase (AICAR Tfase; EC 2.1.2.3) and inosine 5′-monophosphate (IMP) cyclohydrolase (IMPCH; EC 3.5.4.10) activities. These activities represent the final two steps of the de novo purine biosynthesis pathway. PurH contains two functional regions, with the IMPCH activity performed by the N-terminal domain and the AICAR Tfase activity by the C-terminal region, which is composed of two structural domains. The AICAR Tfase region catalyzes the transfer of the formyl group from the cofactor 10-formyl-tetrahydrofolate (10-f-THF) to the substrate AICAR, while the IMPCH domain catalyzes the intramolecular cyclization of 5-formyl-AICAR (FAICAR) to IMP [Fig. 1(A)]. Crystal structure of PurH from Thermotoga maritima. A: The reactions carried out by PurH. B: Stereo ribbon diagram of the TM1249 monomer, color-coded from N-terminus (blue) to C-terminus (red), showing the domain organization. Helices (H1–H18) and β-strands (β1–β21) are labeled. C: Diagram showing the secondary structural elements of TM1249 superimposed on its primary sequence. The α-helices, 310-helices, β-strands, β-turns, and γ-turns are indicated. The β-hairpins are depicted as red loops. The four β-sheets are labeled as A, B, C, and D, respectively. The TM1249 monomer has a molecular weight of 49.7 kDa (residues 1–452) and a calculated isoelectric point of 5.49. Crystal structures for the avian1 and human2 forms of this enzyme have been previously reported. In bacteria and eukaryotes, the IMPCH and AICAR Tfase activities reside on a single, bifunctional enzyme. A similar bifunctional protein coded by a purH gene has not been found in archaea. Instead, these activities have been identified in a nonhomologous IMPCH3 and a separate AICAR Tfase.4 Previously, we reported the crystal structures of PurS (TM1244)5 and smPurL (TM1246)6 from Thermotoga maritima. These two proteins, along with TM1245, form the formylglycinamide ribonucleotide amidotransferase complex (EC 6.3.5.3), which catalyzes an earlier step in the de novo purine biosynthesis pathway. Here, we report the crystal structure of PurH from Thermotoga maritima (TM1249), a bacterial form of this bifunctional enzyme, determined using the semiautomated, high-throughput pipeline of the Joint Center for Structural Genomics (JCSG).7 Because this enzyme serves an important role in nucleotide biosynthesis in bacteria and eukaryotes, it is a potential target for the development of antibacterial and anticancer agents. The TM1249 structure offers new insights about the mechanistic features of this essential biochemical reaction. PurH from Thermotoga maritima (TIGR: TM1249; Swiss-Prot: Q9X0X6) was amplified by polymerase chain reaction (PCR) from genomic DNA using PfuTurbo DNA polymerase (Stratagene) and primers corresponding to the predicted 5′ and 3′ ends. The PCR product was cloned into plasmid pMH4, which encodes an expression and purification tag (MGSDKIHHHHHH) at the amino terminus of the full-length protein. The cloning junctions were confirmed by DNA sequencing. Protein expression was performed in a modified Terrific Broth using the Escherichia coli strain GeneHogs (Invitrogen). At the end of fermentation, lysozyme was added to the culture to a final concentration of 250 μg/mL, and the cells were harvested. After one freeze/thaw cycle, the cells were sonicated in lysis buffer [50 mM Tris pH 7.9, 50 mM NaCl, 10 mM imidazole, 1 mM Tris(2-carboxyethyl)phosphine hydrochloride (TCEP)], and the lysate was clarified by centrifugation at 32,500 g for 30 min. The soluble fraction was passed over nickel-chelating resin (GE Healthcare) pre-equilibrated with lysis buffer, the resin was washed with wash buffer [50 mM Tris pH 7.9, 300 mM NaCl, 40 mM imidazole, 10% (v/v) glycerol, 1 mM TCEP], and the protein was eluted with elution buffer [20 mM Tris pH 7.9, 300 mM imidazole, 10% (v/v) glycerol, 1 mM TCEP]. The eluate was diluted 10-fold with buffer Q [20 mM Tris pH 7.9, 5% (v/v) glycerol, 0.25 mM TCEP] containing 50 mM NaCl and loaded onto a RESOURCE Q column (GE Healthcare) pre-equilibrated with the same buffer. The protein was eluted with a linear gradient of 50–500 mM NaCl in buffer Q, buffer exchanged with crystallization buffer [20 mM Tris pH 7.9, 150 mM NaCl, 0.25 mM TCEP], and concentrated for crystallization assays to 20 mg/mL by centrifugal ultrafiltration (Millipore). TM1249 was crystallized using the nanodroplet vapor diffusion method8 with standard JCSG crystallization protocols.7 The crystallization reagent that produced the crystal used for the structure solution contained 20% (w/v) polyethylene glycol (PEG) 6000 and 0.1 M citrate pH 5.0. PEG 200 was added as a cryoprotectant to a final concentration of 10% (v/v). Initial screening for diffraction was carried out using the Stanford Automated Mounting system (SAM)9 at the Stanford Synchrotron Radiation Laboratory (SSRL, Menlo Park, CA). The crystal was indexed in triclinic space group P1 (Table I).10, 11 Molecular weight and oligomeric state of TM1249 were determined using a 1 cm × 30 cm Superdex 200 column (GE Healthcare) in combination with static light scattering (Wyatt Technology). The mobile phase consisted of 20 mM Tris pH 8.0, 150 mM NaCl, and 0.02% (w/v) sodium azide. Diffraction data were collected at the Advanced Light Source (ALS, Berkeley, CA) on beamline 5.0.1 at 100K using a Quantum 210 CCD detector (ADSC). Data were integrated and reduced using Denzo12 and then scaled with the program SCALEPACK.12 Primary phasing was accomplished using the JCSG molecular replacement (MR) protocol.13 A conserved functional domain analysis14 of the amino acid sequence of TM1249 showed that it contains an N-terminal IMPCH domain and a C-terminal AICAR Tfase region. MR search models were constructed from the avian homolog (PDB accession code: 1g8m) using WHATIF,15 based on FFAS16 alignments. To accommodate a possible difference in the relative positioning of the domains in TM1249 compared with the avian homolog, the N-terminal domain (residues 1–156) and the C-terminal region (residues 187–452) were treated as independent entities in the search model. The hinge region (residues 157–186), which separates the N-terminal domain and the C-terminal region, was omitted from the search. After initial phasing with MOLREP17 from the CCP4 suite,11 a preliminary model was traced from the MR-phased electron density maps using ARP/wARP.18 Further refinement was carried out using REFMAC519 and XtalView.20 Data collection, model, and refinement statistics are summarized in Table I. Analysis of the stereochemical quality of the model was accomplished using AutoDepInputTool,21 MolProbity,22 SFcheck 4.0,23 and WHATIF 5.0.15 Figure 1(C) was adapted from an analysis using PDBsum,24 and all other figures were prepared with PyMOL (DeLano Scientific). Atomic coordinates and experimental structure factors of TM1249 have been deposited in the PDB and are accessible under the code 1zcz. The three-dimensional structure of TM1249 [Fig. 1(B)] was determined to 1.88 Å resolution by MR. The refined structure includes a dimer (residues 1–452 for chains A and B), two K+ ions, two PEG 200 molecules, and 636 water molecules in the asymmetric unit. The two PEG molecules are symmetrically positioned near the noncrystallographic twofold axis relating the two monomers of the dimer, and each is within hydrogen bonding distance of Ala 161 in the hinge region that links the N-terminal domain and the C-terminal region. Electron density was not observed for residues from the expression and purification tag. The Matthews' coefficient (Vm)25 is 2.3 Å3/Da, and the estimated solvent content is 46.1%. A Ramachandran plot produced by MolProbity22 shows that 98.3% and 100% of the residues are in favored and allowed regions, respectively. The peptide bond between Ser 355 and Asn 356 (in chain A) is in the cis conformation, with ϕ = −54.8° and ψ = −78.0° for Ser 355, and ϕ = −64.3° and ψ = 135.5° for Asn 356. These residues surround a cavity that forms the proposed AICAR Tfase active site. The K+ ion bound to each monomer is located ∼4 Å from the side-chain hydroxyl of Ser 355 [Fig. 2(A)]. A bound K+ ion and the cis peptide bond are also observed in the avian1 and human structures,2 suggesting that these conserved features are important determinants of AICAR Tfase activity. The biologically relevant dimer of PurH from Thermotoga maritima, and a shift in the relative orientation of the two functional regions of TM1249 compared to the avian enzyme. A: Stereo ribbon diagram of the dimer of TM1249 showing the locations of the IMPCH and AICAR Tfase regions. The active sites of IMPCH and AICAR Tfase are labeled with cyan and orange asterisks, respectively. Bound K+ atoms are shown as spheres (grey) and PEG molecules as sticks (pink) (B) (Left) The relative shift in orientation between the N-terminal IMPCH domain of TM1249 (blue) and that of the avian enzyme (grey) (PDB accession code: 1m9n). (Right) FATCAT-optimized alignment of the N- and C-terminal regions of the avian enzyme and TM1249 after implementing a ∼90° twist in the hinge region of TM1249. A monomer of the avian protein is shown in grey, the N-terminal domain of TM1249 is in green, and the C-terminal domains of TM1249 are in red. The TM1249 monomer contains 21 β-strands (β1–β21) in four β-sheets, 14 α-helices (H2, H3, H6–H15, H17, and H18), and four 310-helices (H1, H4, H5, and H16) [Fig. 1(B)]. The total β-strand, α-helical, and 310-helical content is 22.1%, 35.2%, and 2.7%, respectively. TM1249 comprises three structural domains: an N-terminal domain, which functions as an IMPCH, and two C-terminal domains, which function as an AICAR Tfase [Fig. 1(B)]. The N-terminal IMPCH domain (residues 1–156), which adopts a Rossmann fold topology, has been classified in the SCOP25, 26 database as a methylglyoxal synthase-like fold. Compared with the canonical nucleotide-binding Rossmann fold, which contains a six-stranded β-sheet, this α/β fold lacks the last β-strand and, therefore, consists of a β-sheet formed by five parallel β-strands. The second (residues 187–301) and third (residues 328–452) structural domains are related by tandem duplication and, along with residues 302–327, compose the AICAR Tfase region (residues 187–452) [Fig. 1(B)]. When aligned with each other, these two domains show an 18% sequence identity and an RMSD of 1.4 Å over 112 aligned Cα atoms. These domains adopt a three-layered α+β fold comprised of a mixed six-stranded β-sheet and four α-helices arranged as βαββαβαβαβ, which is similar to the cytidine deaminase-like fold in SCOP.25, 26 In addition, both domains are preceded by β-hairpins (1st: residues 164–182; 2nd: residues 307–318) that connect the three domains. These β-hairpins are involved in extensive inter-subunit interactions and contribute significantly to the dimeric interface, which buries a surface area of 5270 Å2 from each monomer [Fig. 2(A)]. The dimerization interface is very extensive, running along a noncrystallographic twofold axis that covers the full length of the long axis of the molecule. As in the avian1 and human structures,2 the AICAR Tfase active site of TM1249 is formed by the dimerization of the AICAR Tfase regions. Moreover, analytical size exclusion chromatography coupled with static light scattering also indicates that the dimer is the biologically relevant form. Unexpectedly, β-strands β6 from the two monomers form a two-stranded antiparallel β-sheet that packs against α-helix H9 from the IMPCH domain in the TM1249 dimer [Fig 2(A)]. This feature is not present in the avian1 or human structures.2 The top of this β-sheet forms the bottom of a hydrophobic circular cleft delineated by β7, β14, and β15. This hydrophobic patch is located at the hinge region that separates the two functional halves of TM1249. In the crystal structure of TM1249, a PEG 200 molecule partially shields this region from the solvent [Fig. 2(A)]. As anticipated, a structural comparison of TM1249 using the DALI server26 revealed several highly similar structures with statistically significant Z-scores. The avian enzyme (PDB: 1g8m, 1m9n) was found as the top hit (Z = 33.6). Individually, the IMPCH domain and AICAR Tfase region in TM1249 show topological similarity with the corresponding regions in the avian enzyme. The RMSD for the structural alignment of the N-terminal IMPCH domains of TM1249 (residues 1–156) and the avian enzyme (residues 1–197) is 2.2 Å over 154 aligned residues with 42% sequence identity. For the C-terminal AICAR Tfase regions, the RMSD is 1.9 Å over 281 aligned residues with 36% sequence identity. Although these proteins share a similar architecture for the individual regions, the arrangement of the N-terminal IMPCH domain with respect to the C-terminal AICAR Tfase region is significantly different between TM1249 and the avian or human enzyme. Consequently, the RMSD for a structural alignment of TM1249 and the avian enzyme reported by DALI is 12.3 Å over 389 aligned Cα atoms with 38% sequence identity [Fig. 2(B), left]. A similar alignment of these two proteins can be performed using the FATCAT server27 with an RMSD of 13.4 Å (opt-RMSD of 6.3 Å) over 383 Cα atoms when treating the protein chains as rigid bodies. However, the FATCAT server can further optimize this structural alignment by incorporating twists around the hinge regions of the protein chains. When conformational flexibility (one twist) is assumed, these structures can be superimposed with an opt-RMSD of 2.1 Å over 442 Cα atoms [Fig. 2(B), right], revealing the striking difference in the arrangement of the N-terminal IMPCH domain with respect to the C-terminal AICAR Tfase region between the avian and T. maritima structures. The relative arrangement of the two functional regions of this bifunctional enzyme in different species is determined by the structure of the hinge region. In the TM1249 dimer, the two-stranded anti-parallel β-sheet, comprised of strands β6 [Fig. 2(A)] from the symmetry-related monomers, forms a rigid linker that dictates that both functional regions remain on the same side of the dimerization axis, as opposed to the 90° twist that is observed in the avian and human structures. In the avian protein, the linker region is more flexible and is not constrained by the β-sheet formation. Although TM1249 and numerous homologs share a significant degree of overall sequence identity (i.e., >30%), their structures suggest an important role for the hinge region (157–186 in TM1249) in the relative positioning of the two functional regions. Therefore, an NCBI-BLAST29 search was performed using residues from the hinge region of TM1249, which, surprisingly, revealed only a few examples of similar sequences. For example, the hinge region of TM1249 shows 73% sequence identity with the corresponding region in AICAR Tfase IMPCH from Erythrobacter sp. NAP1. This suggests that the interactions between the IMPCH and AICAR Tfase regions in Thermotoga maritima and Erythrobacter sp. NAP1 are likely to be similar and that the two bacterial proteins adopt a similar arrangement of the two regions. The IMPCH and AICAR Tfase active sites of TM1249 were inferred from sequence and structural comparison to the avian enzyme. The IMPCH active sites are located in clefts on the two monomers [Fig. 3(A)]. One boundary of this cleft, which is most likely adjacent to the ribose-phosphate moiety of the substrate and product, is at one end of a β-sheet comprised of β1, β2, β4, and β5. The N-terminal ends of H6 and H7 form a second boundary of this active site cleft for interaction with the 5-aminoimidazole-4-carboxamide moiety of the substrate. The active site region contains the highly conserved residues Asp 94, Tyr 88, Lys 63, Thr 64, Thr 34, Ser 31, Lys 11, and Ser 7, which adopt side-chain conformations similar to those in the avian homolog [Fig. 3(A)]. The active sites of PurH from Thermotoga maritima. A: Stereo diagram of residues in the putative IMPCH active site of TM1249 superimposed on the avian structure (PDB ID: 1m9n). The carbon atoms on the side chains of TM1249 are shown in green, while the carbon atoms on the active site ligand (XMP) and on the side chains of the avian enzyme are in cyan. Residues from the avian enzyme are shown in parentheses. The side chain of Lys 11 in TM1249 is disordered and, consequently, only the β carbon atom is shown. The location of XMP, an inhibitor of IMPCH, from the avian crystal structure is shown. B: Stereo diagram of residues in the putative AICAR Tfase active site of TM1249 superimposed on the avian structure, with the same color and labeling scheme used in (A). The AICAR substrate is from the avian structure. The two AICAR Tfase active sites are situated in large clefts at the dimer interface [Figs. 2(A), 3(B)]. A structural comparison of TM1249 to the avian enzyme reveals similarities in their AICAR Tfase active sites [Fig. 3(B)]. As in the avian enzyme, the AICAR Tfase active site of TM1249 is situated between the two cytidine deaminase-like domains of one monomer and the C-terminal cytidine deaminase-like domain of the second monomer.28 The substrate-bound avian structure reveals several conserved residues that are involved in AICAR binding and in the activation of the 5-amino group of AICAR for formyl group transfer. These residues are conserved in TM1249 and include Phe 401, Arg 376, and His 452 from one monomer, and Glu 275, His 224, Lys 223, Tyr 171, and Arg 170 from the other. In TM1249, difference electron density was observed near one of the putative AICAR-binding sites adjacent to Lys 223 and His 224. This difference electron density was not modeled in the deposited coordinates but can likely be attributed to citrate from the crystallization buffer. The PurH family contains hundreds of homologs in bacteria and eukaryotes, and the TM1249 structure provides a more suitable template for the bacterial homologs. Models for these homologs can be accessed at http://www1.jcsg.org/cgi-bin/models/get_mor.pl?key=1ZCZA. The two independent active sites in this bacterial bifunctional enzyme are very similar to those in its eukaryotic counterparts. The most striking structural difference between TM1249 and the avian homolog is the ∼90° rotation in the relative orientation of the N-terminal IMPCH domain and C-terminal AICAR region. TM1249 shares a >30% sequence identity with both the avian and human homologs, and, not surprisingly, the architecture of the individual IMPCH and AICAR Tfase regions shows a substantial degree of structural similarity across this broad phylogenetic spectrum. However, a prediction of the relative positioning of the two conserved functional regions is more challenging, and the structures can reveal significant differences in the interactions between the regions. These are likely related to differences in coupling between IMPCH and AICAR Tfase activities in TM1249 versus the eukaryotic enzymes. The JCSG has developed The Open Protein Structure Annotation Network (TOPSAN), a wiki-based community project to collect, share, and distribute information about protein structures determined at PSI centers. TOPSAN offers a combination of automatically generated, as well as comprehensive, expert-curated annotations, provided by JCSG personnel and members of the research community. Additional information about TM1249 is available at https://www.topsan.org/1ZCZ. Portions of this research were carried out at the Stanford Synchrotron Radiation Laboratory (SSRL) and the Advanced Light Source (ALS). The SSRL is a national user facility operated by Stanford University on behalf of the U.S. Department of Energy, Office of Basic Energy Sciences. The SSRL Structural Molecular Biology Program is supported by the Department of Energy, Office of Biological and Environmental Research, and by the National Institutes of Health (National Center for Research Resources, Biomedical Technology Program, and the National Institute of General Medical Sciences). The ALS is supported by the Director, Office of Science, Office of Basic Energy Sciences, Materials Sciences Division, of the U.S. Department of Energy under Contract No. DE-AC03-76SF00098 at Lawrence Berkeley National Laboratory. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute of General Medical Sciences or the National Institutes of Health.
DOI: 10.1002/prot.21967
Authors: Herbert L. Axelrod, Daniel McMullan, S. Sri Krishna, Mitchell D. Miller, Marc‐Andre Elsliger, Polat Abdubek, Eileen Ambing, Tamara Astakhova, Dennis Carlton, Hsiu‐Ju Chiu, Thomas Clayton, Lian Duan, Julie Feuerhelm, Slawomir K. Grzechnik, Joanna Hale, Gye Won Han, Justin Haugen, Lukasz Jaroszewski, Kevin K. Jin, Heath E. Klock, Mark W. Knuth, Eric Koesema, Andrew T. Morse, Edward Nigoghossian, Linda Okach, Silvya Oommachen, Jessica Paulsen, Kevin Quijano, Ron Reyes, Christopher L. Rife, Henry van den Bedem, Dana Weekes, Aprilfawn White, Guenter Wolf, Qingping Xu, Keith O. Hodgson, John Wooley, Ashley M. Deacon, Adam Godzik, Scott A. Lesley, Ian A. Wilson
Language: English
Subfield: Molecular Biology
Domain: Life Sciences
Open Access status: bronze
OpenAlex ID: W2018969721