The atomic coordinates for MTH1020 (code 1KUU) have been deposited in the Protein Data Bank, Research Collaboratory for Structural Bioinformatics, Rutgers University, New Brunswick, NJ (http://www.rcsb.org).
Use of the Advanced Photon Source was supported by the Basic Energy Sciences, Office of Science, United States Department of Energy, under Contract W-31-109-Eng-38. Use of the BioCARS Sector 14 was supported by the National Center for Research Resources, National Institutes of Health, under Grant RR07707.
The structure of MTH1020 was determined by the MAD method using selenium as the anomalous scatterer. The resulting electron density was of high quality and allowed the placement of all 202 residues. Refinement at 2.2 Å resolution resulted in an Rcryst of 0.229 and an Rfree of 0.248. According to PROCHECK, 99.4 % of the residues are in the allowed regions, and one residue (Asp102) is in the disallowed region of the Ramachandran plot. This aspartate, however, is represented by excellent electron density including a clearly visible carbonyl oxygen establishing its unusual Φ, Ψ-angles.
The overall fold of this single domain protein consists of a four-layered α-β-β-α core structure that is formed by two antiparallel β-sheets packed against each other, and these β-sheets are covered by α-helices on one face of the molecule [Fig. 1(A)]. The protein was determined to be tetrameric from gel filtration studies, which is consistent with the crystal structure analysis. The β-sheets are composed of seven and six strands, respectively, and the topology of strands in the first β-sheet is 11-10-1-2-12-13-3 and in the second β-sheet is 9-8-7-6-5-4.
A number of structural homologues of MTH1020 were identified (with DALI Z-scores ranging from 9.9 to 6). They belong to the N-terminal nucleophile-(NTN-) hydrolase superfamily,1 which contains a four-layered α-β-β-α core structure. This family of hydrolases includes penicillin acylase, 20S proteasome, and heat shock locus V.2–4 The mechanism of activation of these proteins is conserved, although they differ in their substrate specificities. All known members catalyze the hydrolysis of amide bonds in either proteins or small molecules, and each one of them is synthesized as a preprotein. For each, an autocatalytic endoproteolytic process generates a new N-terminal residue. This mature N-terminal residue is central to catalysis and acts as both a polarizing base and a nucleophile during the reaction. The N-terminal amino group acts as the proton acceptor and activates either the nucleophilic hydroxyl in a Ser or Thr residue or the nucleophilic thiol in a Cys residue. The position of the N-terminal nucleophile in the active site and the mechanism of catalysis are conserved in this family, despite considerable variation in the protein sequences.
In MTH1020, a putative active site was identified by superposition with homologous NTN-hydrolase superfamily members and searching for a pocket that contained a structurally conserved N-terminal nucleophile. We identified a deep pocket on the surface of MTH1020 in a position equivalent to that of the active sites of the NTN-hydrolase superfamily members; however, we were unable to locate an N-terminal nucleophile. In MTH1020, this site contains the following conserved polar residues: Tyr2, Arg5, Tyr20, Arg30, Tyr56, Tyr59, Asn60, Asn73, His76, Asp78, Glu104 Arg112, and Tyr148. All of these residues are absolutely conserved between the MTH1020 family members, thus reinforcing the correct identification of the active site location.
The structural analysis of MTH1020 reveals an NTN-hydrolase fold but fails to assign an unequivocal function as the protein neither seems to be processed, nor does it contain an appropriate amino acid in the position of the conserved N-terminal nucleophile. In previously identified NTN-hydrolase family members, a threonine, serine or cysteine residue occupies this position; however, in MTH1020, as well as its sequence homologues, an Arg residue (Arg5) is found at this site [Fig. 1(B)]. This amino acid cannot act as a nucleophile.
Full-length MTH1020 was found in the crystal structure; therefore, no processing had occurred. However, the protein was purified at temperatures well below the usual growth conditions for M. thermoautotrophicum. To investigate whether MTH1020 exhibited autohydrolase activity at higher temperature, MTH1020 was incubated at 65°C (the ambient temperature for M. thermoautotrophicum) for varying times, and the protein sample was analyzed on SDS-PAGE (data not shown). We found that the protein remained intact even after 2 h at 65°C, indicating that MTH1020 does not undergo autohydrolysis.
In conclusion, we found that MTH1020 is structurally but not functionally similar to members of the NTN-hydrolase family. Primary sequence analysis was unable to predict that MTH1020 would fold into a four-layered α-β-β-α core structure or that it would be structurally similar to the NTN-hydrolase family.
Cloning, Purification, and Crystallographic Studies.
Cloning, purification, and crystallization experiments have been described elsewhere for other MT proteins.5 The morphology of single crystals of MTH1020 is trigonal bipyramidal, and they appear after approximately 24 h in crystallization setups containing methyl-pentanediol (MPD) as precipitant. Crystals selected for diffraction experiments were grown in 14% MPD, 0.2 M Mg acetate, and 100 mM HEPES at pH 7.5 at 20°C. The crystals belonged to the tetragonal space group, I4122, with the following unit cell parameters: a = b = 107.0, and c = 87.0 Å. The Matthew's coefficient, VM, was determined as 2.8 Å3 Da−1 resulting in a solvent content of 57% with a single molecule in each asymmetric unit.
X-Ray Diffraction and Structure Determination.
A three-wavelength MAD experiment was carried out at 100 K on beamline BM14D, APS, and data from a native crystal of MTH1020 were collected on beamline BM14C, APS. MAD and native data were processed and scaled with the DENZO/SCALEPACK suite of programs. Data collection statistics are presented in Table I. SOLVE was used to locate the selenium sites and to calculate the phases, and RESOLVE was used to modify the density. Electron density visualization and model building were done with O. Rigid body and simulated annealing torsion angle refinement were normally followed by individual B-factor refinement and performed by using CNS 1.0. Several rounds of refinement were combined with model rebuilding in O after inspection of both 2Fo-Fc and Fo-Fc maps. Refinement statistics are found in Table I.
Table I. X-Ray Data Collection and Refinement Statistics
Numbers in parentheses represent values in the highest resolution shell (native 2.33–2.25 Å and SeMet 1.93–1.86 Å.
Rsym = Σ|I−<I>|/ΣI, where I is the observed integrated intensity, <I> is the average integrated intensity obtained from multiple measurements, and the summation is over all observed reflections.
FOMMAD = figure of merit after MAD phasing.
FOMDM = figure of merit after density modification.
Rcryst = |Fobs Fcalc|/|Fobs|.
Rfree was calculated by using randomly selected reflections (10%).
We thank Alexey G. Murzin for helpful discussions and the staff of BioCARS for help during data collection at Sector 14 of the Advanced Photon Source. AME and CHA are Scientists of the Canadian Institutes of Health Research; DC was supported by a Best Fellowship.