Structure of the hypothetical DUF1811-family protein GK0453 from Geobacillus kaustophilus HTA426

The crystal structure of a conserved hypothetical protein, GK0453, from Geobacillus kaustophilus has been determined to 2.2 Å resolution. The crystal belonged to space group P4₃2₁2, with unit-cell parameters a = b = 75.69, c = 64.18 Å. The structure was determined by the molecular-replacement method and was refined to a final R factor of 22.6% (R_free = 26.3%). Based on structural homology, the GK0453 protein possesses two independent binding sites and hence it may simultaneously interact with two proteins or with a protein and a nucleic acid.

1. Introduction

As part of the RIKEN Structural Genomics Initiative (RSGI) project, in collaboration with UK Structural Genomics, we selected the hypothetical protein GK0453 (13 kDa, 113 residues) from Geo­bacillus kaustophilus HTA426 to predict its function from analysis of its crystal structure. The GK0453 protein is a member of the DUF1811 family in the Pfam database (Bateman et al., 2002). G. kaustophilus, from the Bacillaceae family, was isolated from deep-sea sediment from the Mariana Trench (Takami et al., 1997). It is an aerobic, endospore-forming, Gram-positive bacterium that grows optimally at 333 K, with an upper temperature limit of 347 K (Takami et al., 2004). There are 174 uncharacterized proteins in the DUF1811 family, and many are from Bacillus and Staphylococcus species that are known to cause a wide variety of diseases such as nosocomial infections. Thus, the proteins in this family may represent potential drug targets for highly selective bactericides or novel chemotherapies for these pathogens. The crystal structure of YfhH from B. subtilis, which belongs to this family, has been determined (PDB entry 1sf9 ; Midwest Center for Structural Genomics, unpublished work); however, the function of this protein is still unclear. Here, we describe the crystal structure of the hypothetical DUF1811-family protein GK0453 from G. kaustophilus and discuss its function based on structural homology.

2. Methods and materials

2.1. Cloning, expression and purification

The gene encoding the GK0453 protein (gi:56418988) was amplified via PCR using G. kaustophilus HTA426 genomic DNA and was cloned into the pET-15b expression vector (Merck Novagen, Darmstadt, Germany). The tobacco etch virus (TEV) protease recognition sequence was inserted in the N-terminal tag region of the expression vector, which was then introduced into the Escherichia coli Rosetta (DE3) strain (Merck Novagen, Darmstadt, Germany). The recombinant strain was cultured in 5 l LB medium containing 30 µg ml⁻¹ chloramphenicol and 50 µg ml⁻¹ ampicillin. The harvested cells (23.3 g) were lysed by sonication on ice in 35 ml of 20 mM Tris–HCl buffer pH 8.0 containing 500 mM NaCl, 5 mM β-mercaptoethanol and 1 mM phenylmethylsulfonyl fluoride. The cell lysate was heat-treated at 343 K for 13 min and then centrifuged at 15 000g for 30 min at 277 K. The supernatant was applied onto a HisTrap HP column (GE Healthcare Biosciences) equilibrated with 20 mM Tris–HCl buffer pH 8.0 containing 500 mM NaCl and 20 mM imidazole and eluted with a linear (20–500 mM) gradient of imidazole. The target sample, which eluted in the 500 mM imidazole fraction, was collected and applied onto a HiLoad 16/60 Superdex 200 pg column (GE Healthcare Biosciences) equilibrated with 20 mM Tris–HCl buffer pH 8.0 containing 200 mM NaCl and 20 mM imidazole. The eluted fractions containing the target sample were collected and treated with TEV protease at 303 K for 60 min. The sample was then applied onto a HisTrap HP column (GE Healthcare Biosciences) equilibrated with 20 mM Tris–HCl buffer pH 8.0 containing 500 mM NaCl and 20 mM imidazole. The flowthrough fraction was collected and desalted by fractionation on a HiPrep 26/10 column (GE Healthcare Biosciences) with 20 mM Tris–HCl buffer pH 8.0 containing 200 mM NaCl. The protein sample was analyzed by SDS–PAGE and its identity was confirmed by N-terminal amino-acid sequencing. After concentration to 28.5 mg ml⁻¹ by ultrafiltration, the protein yield was 42.8 mg from 23.3 g of cells.

2.2. Protein crystallization, data collection and processing

Crystallization was performed by the microbatch-under-oil method at 291 K. A 0.5 µl aliquot of crystallization reagent was mixed with 0.5 µl of the 28.5 mg ml⁻¹ protein solution and was covered with 15 µl of silicone and paraffin oil. In the initial screening, small crystals appeared in a drop composed of 0.1 M Tris–HCl buffer pH 8.5 containing 20%(w/v) PEG MME 2000 and 0.01 M nickel(II) chloride hexahydrate (Crystal Screen 2 condition No. 45; Hampton Research). After optimization, large crystals were obtained from a crystallization reagent consisting of 0.1 M Tris–HCl buffer pH 8.1 containing 13.3%(w/v) PEG MME 2000 and 0.01 M NiCl₂. Crystals suitable for X-ray data collection appeared within 1 d and reached dimensions of 0.42 × 0.15 × 0.12 mm (Fig. 1a). The crystals were flash-cooled in a nitrogen-gas stream at 100 K using 10%(v/v) glycerol as a cryoprotectant. An X-ray diffraction data set was collected using a MAR Mosaic 225 CCD detector on beamline PX10.1 at the Daresbury Synchrotron Radiation Source (SRS), England. The data were integrated and scaled using the HKL-2000 software package. The data-reduction statistics are summarized in Table 1.

Table 1
Summary of data-collection and refinement statistics

Values in parentheses are for the highest resolution shell.

Data collection
Source	SRS PX10.1
Wavelength (Å)	1.117
Space group	P43212
Unit-cell parameters (Å)	a = b = 75.7, c = 64.2
Resolution (Å)	20.0–2.2
Completeness (%)	99.8 (99.6)
Multiplicity	10.3 (10.6)
Rmerge† (%)	7.8 (27.9)
Refinement statistics
No. of molecules in asymmetric unit	1
Resolution limits (Å)	20.0–2.2
σ cutoff	0
No. of reflections	9710
R factor‡/Rfree§ (%)	22.6/26.3
No. of protein residues	104
No. of water molecules	170
R.m.s. deviations
Bond lengths (Å)	0.011
Bond angles (°)	1.4

†R_merge = .
‡R = , where F_obs and F_calc are the observed and calculated structure factors, respectively.
§R_free was calculated with 5% of data that were omitted from refinement.

Figure 1 The structure of GK0453 from G. kaustophilus. (a) Crystals of the GK0453 protein. (b) Cartoon representation of the tertiary structure of GK0453 coloured in a rainbow ramp from blue at the N-terminus to red at the C-terminus. All figures were produced with PyMOL (Schrödinger) unless mentioned otherwise.

2.3. Structure determination and refinement

The crystal structure of GK0453 was determined by the molecular-replacement method, using the YfhH protein structure as a search model (PDB entry 1sf9 ; Midwest Center for Structural Genomics, unpublished work). The program MOLREP from the CCP4 suite (Winn et al., 2011) was used for structure determination. It generated a distinct peak with an R factor of 48.9% and a correlation coefficient of 45.1% for data in the resolution range 20–4 Å. The structure unambiguously revealed that the crystal belonged to space group P4₃2₁2 and contained one molecule in the asymmetric unit. The model was refined with CNS (Brünger et al., 1998) and several rounds of manual fitting and re-fitting were performed using the program O (Jones et al., 1991), with careful inspection of the 2F_o − F_c, F_o − F_c and OMIT electron-density maps. The final R factor and R_free were 22.6 and 26.3%, respectively, at 2.2 Å resolution. In the final structure, four residues (residues 1–4) in the N-terminal region and five residues in the C-terminal region (residues 109–113) were absent owing to poor electron density in these regions. The stereochemistry of the GK0453 structure was good as assessed by MolProbity (Chen et al., 2010). The structure was deposited in the PDB under accession code 2yxy . The refinement statistics are summarized in Table 1.

3. Results and discussion

3.1. Overall structure

The overall tertiary structure of G. kaustophilus GK0453 consists of two small domains at the N- and C-terminal regions (Fig. 1b). The N-terminal region contains a helix–turn–helix motif (α1, Lys14–Met34; α2, Val37–Tyr53). The C-terminal domain possesses a β-­barrel-like structure with four β-strands (β1, Glu64–Ile68; β2, Ala71–Lys82; β3, Phe85–Arg90; β4, Glu98–Pro101). A long loop containing a 3₁₀-helix (Pro57–Asp59) connects the N- and C-­terminal domains (Figs. 1b and 2a).

Figure 2 Structural comparisons of GK0453. (a) Sequence alignment of GK0453 (Q5L2U2_GEOKA) with the hypothetical protein YfhH (YFHH_BACSU) and representative structurally similar proteins UvrB (UvrB_Ecoli; amino acids 628–673) and the MRG15 chromodomain protein (MO4L1_HUMAN; amino acids 6–65) corresponding to the N- and C-terminal regions of GK0453, respectively. The secondary-structure elements of GK0453 are indicated above the alignment and residues that are similar between GK0453 and YfhH are coloured red. The figure was generated by ESPript (Gouet et al., 1999). Superimpositions are shown of (b) GK0453 (pink) on the YfhH protein (green), (c) the N-terminal region of GK0453 on the C-terminal domain of UvrB (light blue) and on the Rab-binding domain of rabenosyn-5 (yellow) and (d) the C-terminal region of GK0453 on the MRG15 chromodomain (cyan) and on the type II dihydrofolate reductase DHFR (orange). The methylated histone-tail recognizing residues Tyr26, Tyr46 and Trp49 in the MRG15 chromodomain are depicted by sticks.

3.2. Structure comparison and functional prediction

A DALI (Holm & Rosenström, 2010) search was performed for the GK0453 structure to identify structural homologues within the RCSB PDB. The search revealed that the GK0453 structure is very similar to that of the hypothetical protein YfhH (60.8% sequence identity; PDB entry 1sf9 ). Superimposition of the GK0453 structure on the YfhH structure yielded a Z-score of 15.4 and an r.m.s.d. of 1.8 Å for 101 C^α atoms (Figs. 2a and 2b). All other results from the search showed that the structural similarity occurred within the distinct domains of either the 44-amino-acid N-­terminal region or the 54-amino-acid C-terminal region. The N-­terminal region of GK0453 is structurally homologous to the C-­terminal domain of UvrB (PDB entry 1qoj ; 19% identity; Z-score of 7.3 and r.m.s.d. of 0.6 Å for 43 C^α atoms; Sohi et al., 2000) and to the minimal Rab-binding domain of rabenosyn-5 (PDB entry 1z0j ; 23% identity; Z-score of 6.9 and r.m.s.d. of 1.7 Å for 48 C^α atoms; Eathiraj et al., 2005) (Fig. 2c). The UvrB C-terminal domain interacts with the UvrC C-terminal domain during excision repair in E. coli, and the UvrBC complex is part of the UvrABC endonuclease system, which catalyzes DNA damage repair (Sohi et al., 2000). The other protein family, including rabenosyn-5, selectively recognizes distinct subunits of Rab GTPases exclusively through interactions with the switch and inter-switch regions in the helix–turn–helix motif (Eathiraj et al., 2005). A similar structural feature is also observed in several PDB structures of proteins that interact with 23S ribosomal RNA. For example, the ribosomal protein L29 (PDB entry 2gya , chain W; Mitra et al., 2006) yielded 20% identity and a Z-score of 7.2 and an r.m.s.d. of 1.7 Å for 50 C^α atoms. This ribosomal subunit protein, which contains a helix–turn–helix motif, extensively interacts with the 23S ribosomal RNA. Hence, we speculated from this analysis that the N-­terminal region of GK0453 may be involved in a protein–protein or a protein–nucleic acid interaction.

In contrast, the C-terminal domain of GK0453 is structurally homologous to those of the MRG15 chromodomain (PDB entry 2f5k ; 22% sequence identity; Z-score of 6.7 and r.m.s.d. of 2.3 Å for 54 C^α atoms; Zhang et al., 2006), the nuclear protein KIN17 (PDB entry 2ckk , chain A; 16% sequence identity; Z-score of 6.4 and r.m.s.d. of 2.2 Å for 52 C^α atoms; le Maire et al., 2006) and type II R-plasmid-encoded R67 dihydrofolate reductase (R67 DHFR; PDB entry 2rh2 , chain A; 12% identity; Z-­score of 6.5 and r.m.s.d. of 2.7 Å for 51 C^α atoms; Krahn et al., 2007) (Fig. 2d).

The R67 DHFR protein is an NADPH-dependent enzyme that catalyzes the reduction of dihydrofolate (DHF) to tetrahydrofolate (THF). THF is essential for the synthesis of thymidylate, purine nucleosides, methionine and other metabolic intermediates (Krahn et al., 2007). Since the functional tetramerization and the critical residues for enzymatic function are absent in GK0453, it is unlikely that GK0453 possesses an activity similar to that of DHFR. The human KIN17 protein is an essential nuclear component that plays a critical role in maintaining the integrity of the human global genome-repair machinery. The SH3-like β-barrel domain of KIN17 has been shown to interact with RNA (le Maire et al., 2006). As GK0453 possesses a hydrophobic environment in the corresponding region, it is unlikely that GK0453 interacts with RNA or DNA through this region (Fig. 3a).

Figure 3 The electrostatic surface potentials of (a) the GK0453 protein and (b) the MRG15 chromodomain protein. The arrows indicate the similar hydrophobic environments present in both GK0453 and MRG15. The surface is coloured red and blue for potential values below −5kBT and above +5kBT, respectively, where kB is the Boltzmann constant and T is room temperature.

The MRG15 chromodomain participates in chromatin remodelling and transcription regulation by interacting with the methylated histone tail of the nucleosome (Zhang et al., 2006; Steiner et al., 2002). The β-barrel core forms a hydrophobic pocket containing three conserved residues, Tyr26, Tyr46 and Trp49, as a potential binding site for interaction with the methylated histone tail (Fig. 2d). The domain bearing these residues, which are responsible for recognizing the methylated histone tail, is absent in GK0453; however, the GK0453 and MRG15 chromodomain structures both possess similar hydrophobic environments (Fig. 3). However, the structural analysis suggested that the β-barrel domain of GK0453 may also be involved in protein–protein interactions with an unknown function.

In conclusion, the crystal structure of the hypothetical protein GK0453 revealed two small domains: a helix–turn–helix motif at the N-terminal region and an SH3-like β-barrel structure at the C-­terminal region. Based on structural comparisons, we speculate that the GK0453 protein may simultaneously interact with two proteins or with a protein and a nucleic acid to exert its unknown function.