National Laboratory of Membrane Protein Structure (MPS), Research Institute of Pharmaceutical Sciences, College of Pharmacy, Seoul National University, Seoul, Korea
National Laboratory of Membrane Protein Structure (MPS), Research Institute of Pharmaceutical Sciences, College of Pharmacy, Seoul National University, San 56-1, Shillim-Dong, Kwanak-Gu, Seoul 151-742, Korea
Helicobacter pylori is a spiral-shaped Gram-negative bacterium and a human bacterial pathogen that infects approximately half of the world's population.1 It has unique capacity to adapt in the extreme acidic environment in the stomach and to chronically colonize the epithelium of the stomach.2, 3 It is responsible for diverse gastric diseases such as peptic ulcers, chronic gastritis, mucosa-associated lymphoid tissue lymphoma, and gastric cancer.4–6
The genome of H. pylori has been fully sequenced for two prototype strains (strain 26695, strain j99). In the chromosome of strain 26695, 1,590 open reading frames (ORFs) were identified. Among the 1,590 ORFs, 499 ORFs have no homologs in other organisms and more have no putative function.7–9 Determining three-dimensional (3D) structure of these unknown proteins in the paradigm of structural genomics can lead to the inference of the biological function of those and identification of new drug targets.
As part of our structural genomics effort on H. pylori, we have determined the solution structure of HP0894. HP0894 (SwissProt/TrEMBL ID O25554) is an 88-residue, conserved hypothetical protein from H. pylori strain 26695 with calculated pI value of 8.5 and molecular weight of 10.38 kD.
Materials and Methods.
Protein Expression and Purification.
HP0894 gene was polymerase chain reaction-amplified from the H. pylori genomic DNA with specific primers. The amplified DNA fragment was cloned into the plasmid pET21a (Novagen). The recombinant plasmids harboring the target genes were transformed into Escherichia coli BL21(DE3) host cells for large-scale protein production. The resulting construct contains eight nonnative residues at the C-terminus (LEHHHHHH) that facilitate protein purification. Uniformly 15N- and 15N/13C-labeled proteins were prepared by growing bacteria in M9 medium using 15NH4Cl and 13C6-glucose as nitrogen and carbon sources. The protein was purified using Ni2+-affinity column (Chelating Sepharose Fast Flow resin, Pharmacia). All nuclear magnetic resonance (NMR) samples were dissolved in 90% H2O/10% D2O containing ∼1 mM 15N- and 15N/13C-labeled protein in 20 mM NaH2PO4/Na2HPO4 (pH 5.0), 500 mM NaCl, 0.1 mM DTT, and 0.1 mM EDTA.
NMR spectra were acquired on a Bruker AVANCE 500 and AVANCE 600 (equipped with a cryoprobe) spectrometer at 303 K. Spectra were processed using NMRPipe/NMRDraw10 and analyzed using NMRView. Backbone assignments were made from HNCO, HNCA, HN(CO)CA,11 HNCACB,12 HN(CO)CACB,13 and 3D 15N-separated NOESY-HSQC.14 Aliphatic side-chain assignments were made from 3D 15N-separated TOCSY-HSQC,15 HCCH-TOCSY, and 3D 13C-separated NOESY-HSQC. Aromatic ring resonances were assigned using 3D 13C-separated NOESY-HSQC. Chemical shifts were referenced to DSS. Slowly exchanging amide protons were monitored by dissolving the protein in D2O and acquiring a series of 15N-HSQC spectra.
Dihedral angle restraints were calculated from chemical shifts using the program TALOS.16 Upper distance limit restraints were obtained from 3D 15N- and 13C-separated NOESY-HSQC by manual and automatic assignment of NOESY spectra. CANDID module in CYANA 2.017 is used to make automatic assignment of NOESY peaks. Hydrogen-bond restrains were based on slowed hydrogen exchange and observation of regular secondary elements from CSI18 search and NOE patterns. Structure calculations were performed using the program CNS 1.119 and standard simulated annealing and torsion angle dynamics. Analyses of final structures were performed using the program PROCHECK-NMR20 and MOLMOL.21 The program MOLMOL was used to visualize the structures.
Results and Discussion.
The statistical parameters in Table I showed that a high-quality NMR structure was obtained. HP0894 structure (PDB ID: 1Z8M) has two α-helices, two 310-helices, and four β-strands (α - α -310- β -310 - β - β - β). β-Strands form four-stranded anti-parallel β-sheet [Fig. 1(A,B)].
Table I. Structural Statistics for the 30 Best Conformers of HP0894
The default parameters and force constants of protein-allhdg.param and anneal.inp in CNS 1.1 were used for calculation.
PROCHECK-NMR was used for calculation.
Short-range (|i − j| ≤ 1)
Medium-range (1 < |i − j| < 5)
Long-range (|i − j| ≥ 5)
Dihedral angle restraints
Root-mean-square deviation to the mean structure (Å) for residues 5–87
A PSI-BLAST22 analysis of the HP0894 sequence against the nonredundant proteins database identified a total of 64 homologs, most of which are annotated as hypothetical or uncharacterized protein. BLAST conserved domain search23 showed that HP0894 contains a conserved domain of DUF332 (Domain of Unknown Function), which is equivalent to COG 3041 in the National Center for Biotechnology Information Database of Clusters of Orthologous Groups. But, in the Pfam24 database, HP0894 belongs to plasmid stabilization system protein family (PF05016). Members of this family are involved in plasmid stabilization. But the exact molecular function of these proteins is not known.
A search for structural homologs with Z score higher than 3.0 using the programs DALI25 shows HP0894 is structurally similar to Archaeal RelE26 (hypothetical protein Psh013, PDB code 1WMI, Z score= 7.8, RMSD= 2.8) and Ganyloribonuclease(PDB code 1RGE, Z score = 3.3, RMSD = 3.4). These two proteins are both ribonucleases, have the similar number of residues to HP0894 (HP0894: 88 residues, Archaeal RelE: 90, Ganyloribonuclease: 96), share a similar β-sheet topology with HP0894, and have comparable location for two of their helices (Fig. 1C). But they have no detectable sequence homology with HP0894 in PSI-BLAST searches and Blast2 analyses. Archaeal RelE with highest structural similarity to HP0894, whose structure was released recently, is an Archaeal homolog of RelE of E. coli, and regarded as a ribosome-dependent ribonuclease.26 Interestingly, despite no sequential homology with HP0894, Archaeal RelE belongs to Plasmid stabilization system protein family as same as HP0894, according to Pfam database. Considering the result of structural homology search and Pfam database search, therefore, there is the possibility that HP0894 is a ribonuclease.