Crystal structure of HP0242, a hypothetical protein from <i>Helicobacter pylori</i> with a novel fold

Abstract

Helicobacter pylori is a spiral-shaped, gram-negative microorganism that was found in 1979 and isolated in 1982.1 In 1994, the International Agency for Cancer Research declared H. pylori to be a carcinogen of human.2 Two complete genome sequences of H. pylori, strain 26695 and strain J99, have been determined by the whole-genome random sequencing method.3, 4 About 33% protein sequences in the whole genome are annotated as “hypothetical proteins” whose functions and three-dimensional structures have never been identified. The better understanding of these proteins' cellular processes could provide the basis for discoveries of potential antibacterial drug targets. Therefore, the determination of the structural and functional relationship of these proteins has thus drawn much research attention. HP0242 is a hypothetical protein encoded from H. pylori strain 26695. The genomic microarray analysis reveals that HP0242 is an acid-adaptive protein.5 It indicates that HP0242 may have an important function in bacteriological physiology when H. pylori colonize in the highly acidic environment of the human stomach. Although HP0242 has no significant sequence similarity with other functional proteins, a PSI-BLAST search with HP0242 shows four homology proteins (Fig. 1). The HP0242 gene is located next to the napA gene (HP0243/HP-NAP),3 which upstream contains a ferric-uptake regulator binding site. This operon governs coordinated expression of seven proteins totally. There are five functional proteins (TIGR: http://www.tigr.org/): HP0243 (neutrophil activating protein), HP0240 (octaprenyl-diphosphate synthase), HP0239 (glutamyl-tRNA reductase), HP0238 (prolyl-tRNA synthetase), and HP0237 (porphobilinogen deaminase), and two hypothetical proteins, HP0242 and HP0241. The biological functions of HP0243,6, 7 HP0239,3 and HP02373 have been determined to be related to the iron storage and heme biosynthesis. Multiple sequence alignment of HP0242 and its homologues. Four sequences are from Helicobacter pylori 26695 (HP0242), H. pylori J99 (JHP0227), Wolinella succinogenes DSM 1740 (WS1584), and Helicobacter hepaticus ATCC 51449 (HH1722). Sequence alignment was generated by CLUSTALW.28 Completely conserved residues are colored in purple, highly conserved residues are colored in blue, and similar residues are colored in red. Secondary structure elements of HP0242 are shown. We have determined the three-dimensional structure of HP0242 by multiwavelength anomalous dispersion (MAD) phasing from a selenomethinoine (Se-HP0242) protein. The novel folding of HP0242 will be discussed and the possible functional regions will be proposed from the detailed structure analysis. Primers for polymerase chain reaction (PCR) amplification contained a BamHI restriction site in the forward primer (5′-CGGGATCCATTGAGGGTCGCATGAGAGATTACA- GCGAGCTTG-3′) and an XhoI site in the reverse primer (5′- CCGCTCGAGTCATTCATTCCCGCTGAGAATATTC-3′).3 The genomic DNA is purified from H. pylori strain 26695 to amplify the HP0242 gene. PCR was performed by using Taq DNA Polymerase. The amplified DNA was digested with BamHI and XhoI and then cloned into the BamHI/XhoI-digested expression vector pET-30a, prior to transform into Escherichia coli BL21 (DE3). The transformed BL21 cells were cultured overnight in 10 mL LB medium with 100 μg mL−1 kanamycin at 37°C with shaking at 200 rpm. Five milliliters of overnight culture were then transferred to 500 mL LB medium and incubated under the same conditions until OD595 reached 0.7. Protein expression was induced by the addition of IPTG (isopropyl-β-D-thiogalactopyranoside) at a final concentration of 0.5 mM. Cells were harvested by centrifugation at 4000 × g at 4°C for 15 min. The pellet was resuspended in 10 mL lysis buffer (20 mM Tris-HCl pH 7.9, and 300 mM NaCl). The mixture was then lysed by Homogenizer (EmulsiFlex-C5), followed by centrifugation at 18,000 × g at 4°C for 20 min. The supernatant was loaded onto a Ni-NTA column equilibrated with binding buffer (50 mM Tris pH 8.0, 100 mM NaCl), and subsequently washed with wash buffer (50 mM Tris pH 8.0, 100 mM NaCl, 4 mM imidazole) at a flow rate of 2 mL min−1. Finally, the HP0242 protein was eluted with elution buffer (50 mM Tris pH 8.0, 100 mM NaCl, 100 mM imidazole). Crystallization trails of HP0242 were carried out by the hanging-drop vapor-diffusion method.8 One microliter of protein solution was mixed with 1 μL reservoir solution and equilibrated against a 500-μL reservoir solution in Linbro plates. Initial crystallization conditions were obtained using crystal screen kits (Hampton Research) and then further optimized to obtain diffraction quality crystals. The Se-HP0242 derivative crystals grow in a buffer of 20 mM Tris-HCl (pH 8.6) and 200 mM NaCl using a precipitant of 3.2 M sodium formate. The crystals grew to maximum dimensions of 0.1 × 0.1 × 0.2 mm within 3 days at 25°C. The X-ray diffraction data Se-HP0242 crystals was collected at BL12B2 Taiwan beamline at SPring-8, Japan. Three wavelengths (peak: 0.9797 Å, inflection: 0.9799 Å, and high remote: 0.9618 Å) were chosen for the data collection of the Se-HP0242 crystal. Crystals were flash frozen in a liquid–nitrogen stream at 100 K without any cryoprotectant. The data were processed and scaled using the program HKL2000.9 The Se-HP0242 crystal diffracted to 2.27 Å belongs the tetragonal P4322 space group with unit-cell parameters, a = 44.3 Å, b = 44.3 Å, and c = 119 Å, and an Rmerge of 4.4% (Table I). The VM was calculated to be 2.62 Å3Da−1, corresponding to a solvent content of 53%, containing 1 mol per asymmetric unit. The structural phase of Se-HP0242 was determined by the selenium MAD method. The selenium sites were calculated by SOLVE.10 The initial phase was generated to 2.5 Å. The density modification was calculated with RESOLVE11 using solvent flattening. The electron density map showed distinguishable protein and solvent regions. The model was autobuilt with RESOLVE and subsequent iterations of model building was carried out by XtalView.12 The inflection selenium data set was used for the structural refinement using CNS13 at 2.27 Å. The final model of HP0242 consists of 90 residues having four residues missing (6–7 and 93–94). The detailed refinement statistics are shown in Table I. The residues showing higher B factors (>60 Å2) are located at the N-terminal loop and the loop H3-H4. The quality of the model was evaluated continuously during the refinement process with PROCHECK.14 Analysis of the Ramachandran plot showed no violations of accepted backbone torsion angles. Atomic coordinates have been deposited in the Protein Data Bank (PDB ID code 2bo3). The overall structure of HP0242 folds like a musical instrument triangle containing four helices (H1 through H4) with no cysteine [Fig. 2(A)]. It has the dimensions of approximately 38 × 25 × 27 Å. Four helices continually fold approximately in the same plane and H2, H3, and H4 act as the three edges of a triangle, and then H1 and H4 interact to close up the triangle folding. Interestingly, Helix2 is an amphipatic α-helix with a distinctive sequence [Fig. 2(B)]. One side of H2 comprises hydrophobic residues with a total of eight leucine residues, four of which (Leu29, Leu34, Leu37, and Leu38) are highly conserved (Fig. 1). The other side of H2 consists of hydrophilic residues with one lysine, one arginine, and five glutamate residues, three of which (Glu33, Glu39, and Glu45) are also highly conserved (Fig. 1). A structure-based homology analysis of HP0242 structure was carried out using DALI,15 CATH,16 and CE.17 The results of DALI search revealed low Z-scores with <4.0. The top three candidates are 3.8 (for 57 pairs of aligned Cα atoms for cytochrome c oxidase; 2occ-B from Bos taurus; RMSD: 3.9 Å), 3.2 (for 59 pairs of aligned Cα atoms for transcription regulator from Bacillus subtilis; 1vi0-A; RMSD: 3.6 Å), 3.2 (for 51 pairs of aligned Cα atoms for transmembrane glycoprotein gp41 fragment mutant; 1k34-A; RMSD: 6.3 Å). However, there is no functional relationship among these candidate proteins based on the structural classification search by CATH. In addition, a structural alignment between HP0242 and each of these candidates using the program CE shows low Z-scores (<5.0) and high RMSD values (>3.0 Å). Taken together, these results indicate that HP0242 has no obvious similarity to those proteins with known three-dimensional structures. HP0242. (A) A ribbon representation of HP0242 monomer. Four helices were labeled with H1, H2, H3, and H4. The figure was generated by MOLSCRIPT.29 (B) The helical wheel projection of Helix2 (residues 34–51) represents an amphipatic helix. Nonpolar, polar but uncharged, acidic, and basic residues are colored in yellow, green, pink, and blue, respectively. The results from the size-exclusion chromatography and the analytical ultracentrifugation experiments demonstrate that HP0242 exist a strong dimer in solution (data not shown). HP0242 also forms a compact homodimer in crystal, and the central helix “Helix2” plays the key role in the dimer formation [Fig. 3(A) ]. Helix2 interacts with the H2′, H3′, and H4′ of its partner molecule to intertwine two monomers tightly through a significant hydrophobic core from H2 and H2′ with four leucine residues, Leu37, Leu38, Leu40, and Leu41. Several additional forces contribute to this firm and stable dimer conformation, for example, the hydrophobic interactions between H2 and H3′ and hydrophilic interactions of three hydrogen bonds between H2 and H4′. Therefore, Helix2 is essential in the protein folding of the HP0242 dimer, and it may carry out an important role in the protein function. Two dimers assemble into a tetramer using H1 and H4 of one monomer of each dimer by a four-helix bundle interaction [Fig. 3(B)]. There are numerous hydrophobic interactions made by Pro14, Lys17, and Ile21 of H1 as well as Thr79, Ile83, Met86, Ala87, and Leu90 of H4. Because H1 is shorter than H4 with only 10 residues versus 20 residues, fewer hydrophobic interactions are created in the four-helix bundle formation of the HP0242 tetramer. The interaction of the HP0242 tetramer is weaker than those of other similar four-helix bundle structures. HP0242 dimer and tetramer. (A) A stereoview of ribbon diagram showing the structural relationship of two monomers (color in green and red) in the HP0242. (B) A ribbon diagram representation of HP0242 four monomers A, B, C, and D are colored in green, red, blue, and yellow, respectively. (C) The conserved residues of HP0242 and its homologues are colored in yellow. Two putative functional sites (PS1 and PS2) are labeled. (D) Electrostatic surface potential of the HP0242 dimer, the most positive charged regions are colored in blue and the most negative charged regions are colored in red. The figure was generated by MOLSCRIPT29 and GRASP.30 A resembling knot folding18-24 was observed in the dimer structure of HP0242. The HP0242 dimer mimics a deep figure-of-eight knot folding, which is formed by Helix2 of monomer threading through the other chain. Because the knot folding is crucial for the stability of dimer,19, 22 the dimerization of HP0242 might be facilitated by the knot folding. Normally, there is a cavity in the knot structure for the substrate binding, and the knot structure seems to be important for the protein function.18, 19 However, the knot folding of HP0242 is tighter than that of other knot structures because it is made by only six helices from two monomers.23, 24 There is no potential cavity in HP0242 to construct an active site like other knot structures. Therefore, we suggest that the functional site of HP0242 may be located on its surface. From the sequence alignment (Fig. 1), several highly conserved residues found between HP0242 and its homology proteins are located in two significant regions on the dimer surface [Fig. 3(C)]. These regions are assigned as two putative functional sites, PS1 and PS2. PS1 is located around the interface of H1–H2′ revealing a hydrophobic electrostatic surface potential by 14 residues with only three charged residues, Lys17, Asp20, and Glu33 [Fig. 3(C) and (D)]. PS1 might involve in the protein–protein interactions. PS2 is located around the interface of H2–H3 showing a hydrophilic electrostatic surface potential of a highly negative charged distribution with five glutamine acid residues, Glu45, Glu49, Glu54, Glu55, and Glu58 [Fig. 3(C)]. PS2 might contain the probable binding to positively charged molecules [Fig. 3(D)]. We propose that PS1 and PS2 may be the potential function sites for HP0242. The HP0242 gene is next to the napA gene, and both genes possibly belong to the same operon. The biological function of neutrophil activating protein has been determined to be related to iron storage.25 Generally, the iron metal is coordinated by carboxylated residues such as aspartic and glutamic acids. Additional residues, histidine and tryptophan, provide another coordinated sites to make an iron center.26, 27 Histidine plays a role in the redox process in the iron binding.27 Based on this concept, the corresponding residues and the possible binding environment were examined in HP0242. Because HP0242 contains only one histidine, His24, at the C-terminal of H1, the surrounding environment of His24 was evaluated for the possibility of iron binding. However, there is no additional coordinated residue, such as asparatic and glutamic acids around His24 to construct the preferred iron-binding center. Meanwhile, we found no indication of bound metal ion around His24 as well as the whole protein molecule in the electron density map of HP0242. Additionally, the data of inductively coupled plasma-atomic emission spectrometer (ICP-AES) show no identification of metal ion. To date, there is no three-dimensional structure of HP0242 homologues been determined. However, HP0242 and its homology proteins share the similar secondary structure prediction, with high helix content and an amphipatic Helix2. According to the detailed structure analysis, numerous important residues involved in the protein folding of HP0242 are most all conserved among its homologous. Therefore, we strongly suggest that these homology proteins might have the similar protein folding like HP0242.