Nucleoside diphosphate kinase from B. anthracis has been crystallized. Preliminary crystallographic analysis shows that there is one monomer in the asymmetric unit of the crystal.
Keywords: nucleoside diphosphate kinase, Bacillus anthracis, sporulation
Abstract
Bacillus anthracis nucleoside diphosphate kinase (BaNdk) is an enzyme whose primary function is to maintain deoxynucleotide triphosphate (dNTP) pools by converting deoxynucleotide diphosphates to triphosphates using ATP as the major phosphate donor. Although the structures of Ndks from a variety of organisms have been elucidated, the enzyme from sporulating bacteria has not been structurally characterized to date. Crystals of the B. anthracis enzyme were grown using the vapour-diffusion method from a hanging drop consisting of 2 µl 10 mg ml−1 protein in 50 mM Tris–HCl pH 8.0, 50 mM NaCl, 5 mM EDTA equilibrated against 500 µl reservoir solution consisting of 2.25 M ammonium formate and 0.1 M HEPES buffer pH 7.25. Diffraction data extending to 2.0 Å were collected at room temperature from a single crystal with unit-cell parameters a = b = 107.53, c = 52.3 Å. The crystals are hexagonal in shape and belong to space group P6322. The crystals contain a monomer in the asymmetric unit, which corresponds to a Matthews coefficient (V M) of 2.1 Å3 Da−1 and a solvent content of about 36.9%.
1. Introduction
Nucleoside diphosphate kinases (Ndks) are a highly conserved group of enzymes. The primary function of the enzyme is to maintain deoxynucleotide triphosphate (dNTP) pools by converting deoxynucleotide diphosphates to triphosphates using ATP as the major phosphate donor. The reaction proceeds by the formation of a covalent intermediate in which the enzyme is phosphorylated at a catalytic histidine residue. The dNTPs thus formed serve as precursors for DNA and RNA synthesis (Lascu & Gonin, 2000 ▶). Ndk is postulated to be involved in numerous other physiological processes that are independent of its metabolic functions (Chakrabarty, 1998 ▶; Hartsough & Steeg, 2000 ▶; Bernard et al., 2000 ▶; Postel, 2003 ▶; Shen et al., 2004 ▶; Wallet et al., 1990 ▶; Biggs et al., 1990 ▶; Miller et al., 2002 ▶; Munoz-Dorado et al., 1990 ▶). Human Ndk/NM23 homologues have been identified as tumour suppressors with implications in tumour metastasis and pathogenesis (Hartsough & Steeg, 2000 ▶; Leone et al., 1991 ▶). Another isoform of the same enzyme appeared to act as a transcription factor, with the c-myc gene as one of its targets (Postel, 2003 ▶; Postel et al., 1993 ▶). DNA-binding and DNA-cleavage activities have also been reported for Mycobacterium tuberculosis Ndk, with implications in dissemination of the intracellular pathogen (Saini et al., 2004 ▶). Ndks from various organisms share primary, secondary and tertiary structural similarity, but differ in their quaternary structure. Ndks from eukaryotes, halophilic archaea, Bacillus species and M. tuberculosis form hexamers, although the axial symmetry and stability of the hexamer is species-dependent. Some bacterial enzymes have been reported to also have tetrameric structures (Mock & Fouet, 2001 ▶; Singh et al., 2004 ▶). Ndk oligomerization has been correlated with its ability to interact with other molecules and carry out its function. However, the diverse cellular functions of Ndk from a sporulating bacterium have not been studied.
B. anthracis, the causative agent of anthrax, forms metabolically dormant spores in response to nutrient deprivation and high cell density similar to other members of the genus Bacillus (Setlow, 2003 ▶; Setlow et al., 2001 ▶). The spores of B. anthracis are resistant towards extreme temperature, fluctuations in pH, hazardous radiation and noxious chemicals. These properties allow B. anthracis spores to be used as a potent biological weapon for mass destruction. These spores can persist in their habitat for years and can germinate in response to favourable conditions. The genome sequence of B. anthracis revealed that it has ‘a large metabolic tool kit’, similar to those of other soil bacteria, that gives the bacterium the versatility to import nutrients and respond to signals from its environment (Setlow et al., 2002 ▶).
We have purified, crystallized and collected high-resolution data from B. anthracis Ndk and have solved the structure using molecular replacement with human nucleoside diphosphate kinase (PDB code 1jxv; 56% identity) as a template. The structure, when refined, will form the basis for a structure-based mutational analysis of catalytically important residues to probe the structure–function relationship in the enzyme.
2. Experimental methods
2.1. Protein expression and purification
B. anthracis genomic DNA was used as a template for the amplification of the gene encoding Ndk (BAS1425). The nucleotide sequences of the two primers were 5′-CCATGGAAAAAACATTTCTAATGG-3′ to introduce a BamHI site at the 5′ end and 5′-CTCGAGTGGAATTCTTAGTAAATCCATTCA-3′ with an EcoRI site at the 3′ end. The amplified product was digested with BamHI and EcoRI and was ligated into BamHI–EcoRI-digested pPROEx-HTc expression vector (Invitrogen). The resultant plasmid carrying ndk was designated pndk.
Recombinant protein was expressed and purified from Escherichia coli BL21 cells. In brief, the culture was harvested and suspended in 50 mM Tris pH 8.5, 5 mM 2-mercaptoethanol and 1 mM PMSF at 277 K, which was followed by sonication at 2.5 W with 30 s pulses for 10 min with a microtip probe (Misonix 2000). The sonicated sample was centrifuged, the supernatant was incubated with Ni–NTA (trinitriloacetate) resin (Qiagen) and the protein was purified according to the manufacturer’s instructions. Fractions containing the desired protein were pooled after analysis on 15% SDS–PAGE and were dialyzed against PBS containing 20% glycerol. The concentration of protein was determined using Bradford’s reagent (Bradford, 1976 ▶) with BSA as a standard and stored in aliquots until further use. The purified protein gave a single band corresponding to a molecular weight slightly above the expected ∼21 kDa on SDS–PAGE. Size-exclusion chromatography indicates that BaNdk is hexameric in nature.
2.2. in vitro activity
The ability of purified Ndk to autophosphorylate was determined in an independent assay. The protein was incubated in a kinase buffer containing 50 mM Tris–HCl pH 7.4, 5 mM MgCl2 in the presence of 0.5 µCi [γ32P]-ATP as a phosphate donor for 15 min at room temperature to assay autophosphorylation. The reactions were terminated by addition of SDS sample buffer, resolved on 15% SDS–PAGE and autoradiographed to determine the phosphorylation of Ndk.
Ndk activity was determined using a coupled-enzyme assay in which ATP was used as a donor and thymidine diphosphate (TDP) was used as a substrate. The decrease in absorbance owing to oxidation of NADH was measured at 340 nm at room temperature in a spectrophotometer (Perkin–Elmer, lambda25). Negative-control assays were performed in the absence of TDP and Ndk.
2.3. Crystallization
Crystallization experiments were set up at 295 K using the hanging-drop vapour-diffusion method and by exploring crystallization ‘space’ using sparse-matrix approaches (Mazeed et al., 2003 ▶; Jancarik & Kim, 1991 ▶; Cudney et al., 1994 ▶). Crystals were obtained in various screening solutions. Large single crystals (1.5 × 0.8 × 0.5 mm) were grown in 2–3 d from a hanging drop consisting of 2 µl 10 mg ml−1 protein and 2 µl reservoir solution in 50 mM Tris–HCl pH 8.0, 50 mM NaCl, 5 mM EDTA equilibrated against 500 µl of a reservoir solution consisting of 2.25 M ammonium formate and 0.1 M HEPES buffer pH 7.25 (Fig. 1 ▶).
Figure 1.
Crystal of B. anthracis nucleoside diphosphate kinase.
2.4. Data collection
A single crystal was mounted in a capillary. Data were collected to 2.0 Å at room temperature using a MAR 345 area-detector system and a Rigaku RU-300 rotating-anode generator operating at 50 kV and 100 mA. The data were processed using the MOSFLM/CCP4 suite of programs (Collaborative Computational Project, Number 4, 1994 ▶). The crystals were hexagonal, with unit-cell parameters a = 107.53, b = 107.53, c = 52.3 Å and space group P6322. The data-collection statistics are summarized in Table 1 ▶. The crystal mosaicity refined to around 0.6° and an overall data completeness of 100% was obtained. Calculation of the Matthews coefficient (Matthews, 1968 ▶) suggests that the asymmetric unit contains a single subunit. Assuming that the asymmetric unit contains a monomer, the calculated Matthews coefficient is 2.1 Å3 Da−1, corresponding to a solvent content of about 36.9%.
Table 1. Data-collection statistics.
Values in parentheses are for the highest resolution shell.
| Wavelength (Å) | 1.5418 |
| Space group | P6322 |
| Unit-cell parameters | a = 107.53, b = 107.53, c = 52.3 |
| Resolution range (Å) | 26.7–2.0 (2.11–2.0) |
| No. of measured reflections | 79277 (11305) |
| No. of unique reflections | 12490 (1784) |
| Multiplicity | 6.3 (6.3) |
| I/σ(I) | 14.9 (3.4) |
| Completeness (%) | 99.9 (100.0) |
| Rmerge | 0.087 (0.55) |
| Rr.i.m.† | 0.095 (0.59) |
R
r.i.m. =
, where N is the multiplicity.
2.5. Data processing and structure solution
A total of 79 277 measured reflections were merged into 12 490 unique reflections with an R merge of 8.7%. A sequence comparison of BaNdk against the structures in the Protein Data Bank (http://ww.rcsb.org) showed that BaNdk exhibits 56% identity to the human enzyme. Molecular-replacement calculations were carried out using AMoRe (Navaza, 1994 ▶). A clear solution with good packing was obtained with a correlation coefficient of 66.3 and an R factor of 42.4% using data between 20 and 2 Å resolution. The packing of the molecules in the crystal was good (Fig. 2 ▶) and suggested that the hexamer observed in size-exclusion chromatography apparently arises from association as a dimer of trimers. Model building and refinement are presently in progress.
Figure 2.
Packing of molecules in the crystal structure of B. anthracis Ndk. The association of the hexamer as a dimer of trimers is clearly visible.
Acknowledgments
GM and AA acknowledge fellowships from the Council of Scientific and Industrial Research (CSIR), India. This paper bears CDRI communication No. 7390 and is supported by CSIR network grant NWP0038 (formerly SMM003). Support for the computational work came from CSIR grant CMM0017.
References
- Bernard, M. A., Ray, N. B., Olcott, M. C., Hendricks, S. P. & Mathews, C. K. (2000). J. Bioenerg. Biomembr.32, 259–267. [DOI] [PubMed]
- Biggs, J., Hersperger, E., Steeg, P. S., Liotta, L. A. & Shearn, A. (1990). Cell, 63, 933–940. [DOI] [PubMed]
- Bradford, M. M. (1976). Anal. Biochem.72, 248–254. [DOI] [PubMed]
- Chakrabarty, A. M. (1998). Mol. Microbiol.28, 875–882. [DOI] [PubMed]
- Collaborative Computational Project, Number 4 (1994). Acta Cryst. D50, 760–763.
- Cudney, R., Patel, S., Weisgraber, K., Newhouse, Y. & McPherson, A. (1994). Acta Cryst. D50, 414–423. [DOI] [PubMed]
- Hartsough, M. T. & Steeg, P. S. (2000). J. Bioenerg. Biomembr.32, 301–308. [DOI] [PubMed]
- Jancarik, J. & Kim, S.-H. (1991). J. Appl. Cryst.24, 409–411.
- Lascu, I. & Gonin, P. (2000). J. Bioenerg. Biomembr.32, 237–246. [DOI] [PubMed]
- Leone, A., Flatow, U., King, C. R., Sandeen, M. A., Margulies, I. M., Liotta, L. A. & Steeg, P. S. (1991). Cell, 65, 25–35. [DOI] [PubMed]
- Matthews, B. W. (1968). J. Mol. Biol.33, 491–497. [DOI] [PubMed]
- Mazeed, S., Ofek, G., Belachew, A., Huang, C. C., Zhou, T. & Kwong, P. D. (2003). Structure, 11, 1061–1070. [DOI] [PubMed]
- Miller, J. H., Funchain, P., Clendenin, W., Huang, T., Nguyen, A., Wolff, E., Yeung, A., Chiang, J. H., Garibyan, L., Slupska, M. M. & Yang, H. (2002). Genetics, 162, 5–13. [DOI] [PMC free article] [PubMed]
- Mock, M. & Fouet, A. (2001). Annu. Rev. Microbiol.55, 647–671. [DOI] [PubMed]
- Munoz-Dorado, J., Inouye, M. & Inouye, S. (1990). J. Biol. Chem.265, 2702–2706. [PubMed]
- Navaza, J. (1994). Acta Cryst. A50, 157–163.
- Postel, E. H. (2003). J. Bioenerg. Biomembr.35, 31–40. [DOI] [PubMed]
- Postel, E. H., Berberich, S. J., Flint, S. J. & Ferrone, C. A. (1993). Science, 261, 478–480. [DOI] [PubMed]
- Saini, A. K., Maithal, K., Chand, P., Chowdhury, S., Vohra, R. & Goyal, A. (2004). J. Biol. Chem.279, 50142–50149. [DOI] [PubMed]
- Setlow, B., Loshon, C. A., Genest, P. C., Cowan, A. E., Setlow, C. & Setlow, P. (2002). J. Appl. Microbiol.92, 362–375. [DOI] [PubMed]
- Setlow, B., Melly, E. & Setlow, P. (2001). J. Bacteriol.183, 4894–4899. [DOI] [PMC free article] [PubMed]
- Setlow, P. (2003). Curr. Opin. Microbiol.6, 550–556. [DOI] [PubMed]
- Shen, R., Olcott, M. C., Kim, J., Rajagopal, I. & Mathews, C. K. (2004). J. Biol. Chem.279, 32225–32232. [DOI] [PubMed]
- Singh, Y., Liang, X. & Duesbery, N. S. (2004). Molecular and Cellular Biology, edited by T. Proft, pp. 285–312. Norwich: Horizon.
- Wallet, V., Mutzel, R., Troll, H., Barzu, O., Wurster, B., Veron, M. & Lacombe, M. L. (1990). J. Natl Cancer Inst.82, 1199–1202. [DOI] [PubMed]


