PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of actafjournal home pagethis articleInternational Union of Crystallographysearchsubscribearticle submission
 
Acta Crystallogr Sect F Struct Biol Cryst Commun. May 1, 2009; 65(Pt 5): 450–454.
Published online Apr 24, 2009. doi:  10.1107/S1744309109012767
PMCID: PMC2675582
Geobacillus stearothermophilus 6-phosphogluconate dehydrogenase complexed with 6-phosphogluconate
Scott Cameron,a Viviane P. Martini,b Jorge Iulek,b and William N. Huntera*
aDivision of Biological Chemistry and Drug Discovery, College of Life Sciences, University of Dundee, Dundee DD1 5EH, Scotland
bUniversidade Estadual de Ponta Grossa, Departamento de Química, Avenida Carlos Cavalcanti, 4748 Uvaranas, 84030-000 Ponta Grossa, Paraná, Brazil
Correspondence e-mail: w.n.hunter/at/dundee.ac.uk
Received March 13, 2009; Accepted April 3, 2009.
Two crystal structures of recombinant Geobacillus stearothermophilus 6-phosphogluconate dehydrogenase (Gs6PDH) in complex with the substrate 6-­phosphogluconate have been determined at medium resolution. Gs6PDH shares significant sequence identity and structural similarity with the enzymes from Lactococcus lactis, sheep liver and the protozoan parasite Trypanosoma brucei, for which a range of structures have previously been reported. Comparisons indicate that amino-acid sequence conservation is more pronounced in the two domains that contribute to the architecture of the active site, namely the N-terminal and C-terminal domains, compared with the central domain, which is primarily involved in the subunit–subunit associations required to form a stable dimer. The active-site residues are highly conserved, as are the interactions with the 6-phosphogluconate. There is interest in 6PDH as a potential drug target for the protozoan parasite T. brucei, the pathogen responsible for African sleeping sickness. The recombinant T. brucei enzyme has proven to be recalcitrant to enzyme–ligand studies and a surrogate protein might offer new opportunities to investigate and characterize 6PDH inhibitors. The high degree of structural similarity, efficient level of expression and straightforward crystallization conditions mean that Gs6PDH may prove to be an appropriate model system for structure-based inhibitor design targeting the enzyme from Trypanosoma species.
Keywords: pentose phosphate pathway, 6-phosphogluconate dehydrogenase, Geobacillus stearothermophilus
The enzyme 6-phosphogluconate dehydrogenase (6PDH; EC 1.1.1.44) catalyzes the conversion of 6-phosphogluconate (6PG) to ribulose 5-phosphate (R5P), a reaction that occurs in the pentose phosphate pathway that generates NADPH and pentose sugars (Barrett, 1997 [triangle]). This is an essential activity in eukaryotic cells (Lobo & Maitra, 1982 [triangle]; Hughes & Lucchesi, 1977 [triangle]) and the deleterious effects generated by compromising 6PDH activity have been ascribed to the disruption of glycolysis by 6PG inhibition of phosphoglucose isomerase (Marchand et al., 1989 [triangle]). The essential nature of 6PDH suggests that it is a potential drug target in Trypanosoma brucei, a protozoan parasite that is responsible for African sleeping sickness (Barrett, 1997 [triangle]). Further support for this conclusion comes from the observation that selective inhibition of the parasite over mammalian enzymes appears to be possible (Hanau et al., 2004 [triangle]). The structures of 6PDH from sheep (Oa6PDH; Adams et al., 1994 [triangle]), T. brucei (Tb6PDH; Phillips et al., 1998 [triangle]), Lactococcus lactis (Ll6PDH; Sundaramoorthy et al., 2007 [triangle]) and Saccharomyces cerevisiae (He et al., 2007 [triangle]) have been reported.
Our studies with 6PDH from a variety of species, in particular the T. brucei enzyme, have proven to be problematic, with limitations in protein stability and solubility compounding difficulties in structural studies. The enzyme from L. lactis enabled us to generate some data on protein–ligand complexes (Tetaud et al., 1998 [triangle]; Sundaramoorthy et al., 2007 [triangle]). However, this enzyme does not produce good-quality crystals in a reproducible fashion and the presence of one long unit-cell length (>240 Å) renders it difficult to routinely measure diffraction data to high resolution. An alternative source of suitable crystals was sought to provide a surrogate model to assist structure-based ligand studies that guide the development of novel inhibitors (Hunter, 2009 [triangle]). The crystallization of 6PDH from the moderate thermophile Geobacillus stearothermophilus has been reported (Gs6PDH; Pearse & Harris, 1973 [triangle]) and we sought to establish whether this particular enzyme would provide a useful source of crystals. We now report the preparation of an efficient recombinant protein-production system, crystallization, structure determination and comparisons with previously determined structures.
2.1. Cloning, expression and purification
The gene encoding 6PDH was identified using a BLAST search (Altschul et al., 1990 [triangle]), with sequence information based on 6PDH from L. lactis (accession No. NP_266778), against the unfinished G. stearothermophilus genome (http://www.genome.ou.edu/bstearo.html). Genomic DNA from G. stearothermophilus (American Type Culture Collection 12980) was used as template for PCR with the following primers designed to amplify the 6PDH open reading frame and incorporating NdeI and XhoI restriction sites (bold), respectively: 5′-­CAT-ATG-GCT-AAA-CAG-CAA-ATC-GGC-G-3′ and 5′-CTC-GAG-TTA-TTT-CAA-CCA-TTC-CGT-ATG-3′. The PCR product was ligated into pCR-BluntII-TOPO vector using the Zero Blunt TOPO PCR cloning kit (Invitrogen). Once identified, the gene was excised with NdeI/XhoI and then ligated into a modified pET15b vector (Novagen) in which the sequence that encodes a thrombin protease cleavage site had been replaced with a tobacco etch virus (TEV) protease recognition site. This plasmid therefore encodes a product with an N-terminal hexahistidine tag, which is removed by TEV protease. The plasmid was amplified in XL-1 Blue Escherichia coli and the integrity of the PCR product was verified by sequencing and then used to transform E. coli strain BL21 (DE3) pLysS for protein production.
Bacteria were cultured with shaking at 200 rev min−1 in autoinduction media (Studier, 2005 [triangle]) with 50 mg l−1 carbenicillin for approximately 2 h at 310 K followed by 22 h at 295 K. Cells were harvested by centrifugation (3500g, 20 min, 277 K) and resuspended in 50 mM Tris–HCl pH 7.4, 250 mM NaCl. DNAse I (100 µg) and an EDTA-free protease-inhibitor cocktail tablet (Roche) were added to the mixture before the cells were lysed in a TS-75 cell disruptor (Constant Systems) and the soluble fraction was isolated by centrifugation (50 000g, 30 min, 277 K). The cell extract was filtered and applied onto a HisTrap HP 5 ml column (GE Healthcare). The His-tagged protein was eluted with a 0–1 M imidazole gradient in the same buffer. Histidine-tagged TEV protease (1 mg in the same buffer per 20 mg 6PDH) was added to the product and the mixture was dialysed against the same buffer as used for cell lysis. Passage over a His-Trap column removed the TEV protease, the cleaved His-tag peptide and any 6PDH which retained the tag and the processed 6PDH was collected. A final purification step involved gel filtration (Superdex 200, 26/60 column, GE Healthcare) before buffer exchange into 50 mM Tris–HCl pH 7.4, 20 mM NaCl and concentration to 12 mg ml−1 for crystallization trials. The high level of protein purity was confirmed by SDS–PAGE and matrix-assisted laser desorption ionization–time-of-flight mass spectrometry. The yield of purified protein was approximately 20 mg per litre of bacterial culture.
2.2. Crystallization, data collection and processing
Crystals grew over 2 d in hanging drops consisting of 1 µl protein solution and 1 µl reservoir solution (0.2 M lithium sulfate, 2.2 M ammonium sulfate or 0.2 M ammonium acetate, 2.2 M ammonium sulfate) equilibrated against 500 µl reservoir solution at 293 K. Crystals were also obtained in the presence of 2 mM 6PG and 2 mM NADP+ or 2 mM R5P and 2 mM NADP+. Single crystals (blocks with approximate dimensions 0.2 × 0.2 × 0.2 mm) were placed in a nylon loop (Hampton Research), mounted on a goniostat, cooled to 100 K with gaseous nitrogen and then characterized using a Micromax-007 rotating-anode generator and an R-AXIS IV++ dual image-plate detector (Rigaku). Suitable crystals were stored in liquid nitrogen and X-ray diffraction data were then measured with a wavelength of 0.979 Å and a Q315r charge-coupled device detector (Area Detector Systems Corporation) on beamline ID14-4 of the European Synchrotron Radiation Facility. Although all crystals were isomorphous, those grown in the presence of lithium sulfate diffracted to higher resolution than those obtained with sodium acetate and so they were used for the analyses. Two data sets were obtained: data set I from a crystal (hereafter called crystal I) grown in the presence of 6PG which diffracted to 2.3 Å resolution and data set II acquired from two crystals grown in the presence of R5P which diffracted to 2.2 Å resolution. The subsequent analysis of data set II revealed only the presence of 6PG in the corresponding crystals.
Diffraction data from crystal I were processed using MOSFLM/SCALA (Leslie, 2006 [triangle]; Evans, 2006 [triangle]), whilst data set II was generated using XDS/XSCALE (Kabsch, 2001 [triangle]). Relevant results are shown in Table 1 [triangle].
Table 1
Table 1
Crystallographic statistics
2.3. Molecular replacement, model building and refinement
The data sets were processed and the structures solved and refined independently, but the same R free set was maintained. Both a polyalanine model and mixed model, with conserved residues left as they are, based on an L. lactis 6PDH subunit (PDB code 1iyo) were constructed using CHAINSAW from the CCP4 suite of programs (Collaborative Computational Project, Number 4, 1994 [triangle]). These two proteins share 66% sequence identity. These models were used for molecular-replacement calculations in Phaser (McCoy et al., 2005 [triangle]) to position the two subunits that constitute the asymmetric unit. Refinement was carried out using REFMAC5 (Murshudov et al., 1997 [triangle]) interspersed with electron-density and difference density map inspection, model manipulation and the incorporation of solvent and ligands using Coot (Emsley & Cowtan, 2004 [triangle]). The asymmetric unit consists of two monomers and they were treated independently during refinement. Crystallographic statistics are given in Table 1 [triangle]. Figures were prepared with PyMOL (http://www.pymol.org) and ALINE (Bond & Schüttelkopf, 2009 [triangle]).
3.1. Structure determination
Two isomorphous crystal structures were determined (Table 1 [triangle]). Data set I was obtained from a sample grown in the presence of 6PG and NADP+ and data set II was obtained from crystals grown in the presence of R5P and NADP+. In neither structure was there electron density that could be attributed to NADP+. In structure I, 6PG is present in both active sites of the dimer that constitutes the asymmetric unit. However, structure II contained only 6PG, not R5P, and then only in one active site with a reduced occupancy of 0.7. This value was assigned following several rounds of refinement testing different levels of occupancy to elucidate that which gave clean difference density maps and thermal parameters consistent with the amino-acid side chains in the vicinity of the ligand. We have no definitive explanation of why we only see partially ordered 6PG in one active site of structure II, although we note a difference in the time that elapsed between crystal growth and cryo-freezing. Crystal I was frozen and stored in liquid nitrogen within one week of setting up the crystallization conditions. The crystals used to generate data set II were frozen after a period of two months.
A pronounced similarity of the two subunits in the asymmetric units of both crystals is observed despite not employing noncrystallographic symmetry restraints during refinement. The overlay of 467 Cα positions of structure I subunit A on subunit B results in an r.m.s.d. of 0.53 Å; for structure II an r.m.s.d. of 0.70 Å is observed from the overlay of 470 Cα atoms. The models derived from the independent analyses are also highly conserved. The overlay of the structure I and structure II dimers, 935 Cα atoms, gives an r.m.s.d. of 0.29 Å. In both structures Lys3, Gly137, Asp175 and Thr453 of both subunits are in disallowed regions of the Ramachandran plot.
To all intents, the protein structures are identical. The only significant difference is that structure II only has 6PG in a single active site and at a reduced occupancy. We therefore detail structure I.
3.2. Overall structure
Gs6PDH shows the same overall fold and three-domain architecture as the other 6PDH structures known (Fig. 1 [triangle]; Adams et al., 1994 [triangle]; Sundaramoorthy et al., 2007 [triangle]). The N-terminal domain, which is formed from the residues up to 174, harbours the site in which NADP+ binds. This domain adopts the Rossmann nucleotide-binding domain fold. The central domain is a largely helical domain (ten α-­helices, residues 175–433) that is primarily involved in dimer formation. The C-­terminal domain or ‘tail domain’ consists of a single α-helix followed by two β-strands and extrudes from the central domain deep into the partner subunit. The active site of 6PDH is formed largely by residues contributed from the N-terminal domain but with one side of the cleft lined by residues from the tail domain of the partner subunit; this will be discussed later.
Figure 1
Figure 1
Secondary and domain structure of Gs6PDH. (a) A ribbon diagram of one subunit. The N-terminal domain is blue, the central domain is yellow and the C-terminal domain is red. (b) Ribbon diagram of the dimer, coloured as in (a). The ligand 6PG is shown as (more ...)
3.3. Structural and sequence comparisons
The G. stearothermophilus 6PDH models display high similarity to the L. lactis, O. aries and T. brucei 6PDH structures (Fig. 2 [triangle] a). The least-squares superpositions of Cα positions of subunit A of structure I with a subunit of each of these orthologues results in r.m.s.d. values of 1.51 Å (461 residues), 0.90 Å (464 residues) and 1.60 Å (454 residues), respectively.
Figure 2
Figure 2
Comparison of the primary and secondary structures of Gs6PDH, Ll6PDH, Tb6PDH and Oa6PDH. (a) Stereoview Cα trace of Gs6PDH (grey) superimposed on Ll6PDH (black) and Tb6PDH (red). (b) Sequence alignment of four 6PDH enzymes, with the assigned secondary (more ...)
Gs6PDH shares 66% sequence identity with Ll6PDH and 57% and 36% identity with Oa6PDH and Tb6PDH, respectively. A sequence alignment of these four enzymes is presented in Fig. 2 [triangle](b). About 26% of the residues are strictly conserved across all four sequences; however, the level of conservation varies according to the domain. In the N-terminal, central and tail domains there are 32, 21 and 36% of residues, respectively, that are strictly conserved. If this comparison is restricted to the two bacterial enzymes the values are 66, 63 and 75%, respectively. Comparison of the Gs6PDH and Tb6PDH sequences reveals values of 49, 26 and 44% for the three domains. The N-terminal and tail domains contribute to the creation of the active site and display a higher level of sequence conservation compared with the central domain, which contributes to dimer formation.
Cofactor binding to 6PDH occurs in a groove on the periphery of the N-terminal domain, with the adenine ribose placed at the β1–α1 turn that carries a nucleotide-binding motif. In Gs6PDH the sequence is Gly10-Leu11-Ala12-Val13-Met14-Gly15 and this matches closely with that observed in most organisms: a Gly-X-Ala-X-X-Gly motif. In Tb6PDH the motif is altered to Gly-X-Gly-X-X-Gly (Phillips et al., 1998 [triangle]; Fig. 2 [triangle] b). At present it is unclear whether this glycine–alanine difference will influence the catalytic properties of the respective enzymes or lead to structural differences when the cofactor is bound. Further structures of both Tb6PDH and Gs6PDH in complex with the cofactor would be required in order to address this point.
3.4. Ligand binding
The interactions formed by 6PG with 6PDH have previously been detailed for the L. lactis enzyme and compared with other structures (Sundaramoorthy et al., 2007 [triangle]). The 6PG substrate-binding site of Gs6PDH and the interactions formed with this ligand (Fig. 3 [triangle]) are highly conserved in other 6PDH structures. The phosphate of 6PG is positioned by accepting hydrogen bonds from Tyr190, Lys260, Arg287 of one subunit and Arg446 of the partner subunit. Ordered water-mediated hydrogen bonds also link the phosphate group to Asn186, Glu189 and Thr262. Additional hydrogen bonds, including several water-mediated interactions, involving His552 from one subunit and Asn102, Ser128, Gly129, Gly130 and Asn186 from the partner hold the rest of the ligand in place. These residues include two that have been shown to be critical for the enzyme mechanism, Lys182 (Zhang et al., 1999 [triangle]) and Glu189 (Karsten et al., 1998 [triangle]), and others (for example, Ser128, Asn187 and His186) that are responsible for ligand binding and catalysis (Li, Dworkowski et al., 2006 [triangle]; Li, Zhang et al., 2006 [triangle]). All of these 13 residues are strictly conserved in the 6PDH sequences from diverse species ranging from a higher eukaryote to a protozoan eukaryote and two different Gram-positive bacteria. Such a high level of sequence and structural conservation suggests that Gs6PDH could serve as an appropriate surrogate for structure-based ligand design targeting the pathogenic protozoan T. brucei.
Figure 3
Figure 3
The 6PG-binding site of Gs6PDH and comparison with Ll6PDH. (a) The Gs6PDH active site. Atoms are coloured according to type: N, blue; O, red; P, orange; the C atoms of residues associated with subunit A are shown in grey, those associated with subunit (more ...)
Supplementary Material
PDB reference: 6-phosphogluconate dehydrogenase, 2w8z, r2w8zsf
PDB reference: 2w90, r2w90sf
Acknowledgments
JI and VPM thank CAPES for a postdoctoral fellowship BEX 2000/04-0 and an MSc fellowship, respectively. WNH thanks the BBSRC (Structural Proteomics of Rational Targets, BBS/B/14434) and The Wellcome Trust (grant Nos. 082596 and 083481). We gratefully acknowledge the provision of beam time at the European Synchrotron Radiation Facility, Grenoble. The Bacillus (Geobacillus) stearo­thermophilus Genome Sequencing Project, from which we obtained sequence information, was funded by the NSF EPSCoR Program (Experimental Program to Stimulate Competitive Research) Grant #EPS-9550478.
  • Adams, M. J., Ellis, G. H., Gover, S., Naylor, C. E. & Phillips, C. (1994). Structure, 2, 651–668. [PubMed]
  • Altschul, S. F., Gish, W., Miller, W., Myers, E. W. & Lipman, D. J. (1990). J. Mol. Biol.215, 403–410. [PubMed]
  • Barrett, M. P. (1997). Parasitol. Today, 13, 11–16. [PubMed]
  • Bond, C. S. & Schüttelkopf, A. W. (2009). Acta Cryst. D65, 510–512. [PubMed]
  • Collaborative Computational Project, Number 4 (1994). Acta Cryst. D50, 760–763. [PubMed]
  • Emsley, P. & Cowtan, K. (2004). Acta Cryst. D60, 2126–2132. [PubMed]
  • Evans, P. (2006). Acta Cryst. D62, 72–82. [PubMed]
  • Hanau, S., Rinaldi, E., Dallocchio, F., Gilbert, I. H., Dardonville, C., Adams, M. J., Gover, S. & Barrett, M. P. (2004). Curr. Med. Chem.11, 2639–2650. [PubMed]
  • He, W., Wang, Y., Liu, W. & Zhou, C.-Z. (2007). BMC Struct. Biol.7, 38. [PMC free article] [PubMed]
  • Hughes, M. B. & Lucchesi, J. C. (1977). Science, 196, 1114–1115. [PubMed]
  • Hunter, W. N. (2009). J. Biol. Chem., doi:10.1074/jbc.R800072200.
  • Kabsch, W. (2001). International Tables for Crystallography, Vol. F, edited by M. G. Rossmann & E. Arnold, pp. 730–734. Dordrecht: Kluwer Academic Publishers.
  • Karsten, W. E., Chooback, L. & Cook, P. F. (1998). Biochemistry, 37, 15691–15697. [PubMed]
  • Leslie, A. G. W. (2006). Acta Cryst. D62, 48–57. [PubMed]
  • Li, L., Dworkowski, F. S. & Cook, P. F. (2006). J. Biol. Chem.281, 25568–25576. [PubMed]
  • Li, L., Zhang, L. & Cook, P. F. (2006). Biochemistry, 45, 12680–12686. [PubMed]
  • Lobo, Z. & Maitra, P. K. (1982). Mol. Gen. Genet.185, 367–368. [PubMed]
  • Marchand, M., Kooystra, U., Wierenga, R. K., Lambeir, A. M., Van Beeumen, J., Opperdoes, F. R. & Michels, P. A. (1989). Eur. J. Biochem.184, 455–464. [PubMed]
  • McCoy, A. J., Grosse-Kunstleve, R. W., Storoni, L. C. & Read, R. J. (2005). Acta Cryst. D61, 458–464. [PubMed]
  • Murshudov, G. N., Vagin, A. A. & Dodson, E. J. (1997). Acta Cryst. D53, 240–255. [PubMed]
  • Pearse, B. M. & Harris, J. J. (1973). FEBS Lett.38, 49–52. [PubMed]
  • Phillips, C., Dohnalek, J., Gover, S., Barrett, M. P. & Adams, M. J. (1998). J. Mol. Biol.282, 667–681. [PubMed]
  • Tetaud, E., Hall, D. R., Gourley, D. G., Leonard, G. A., Arkison, S., Barrett, M. P. & Hunter, W. N. (1998). Acta Cryst. D54, 1422–1424. [PubMed]
  • Studier, F. W. (2005). Protein Expr. Purif.41, 207–234. [PubMed]
  • Sundaramoorthy, R., Iulek, J., Barrett, M. P., Bidet, O., Ruda, G. F., Gilbert, I. H. & Hunter, W. N. (2007). FEBS J.274, 275–286. [PubMed]
  • Zhang, L., Chooback, L. & Cook, P. F. (1999). Biochemistry, 38, 11231–11238. [PubMed]
Articles from Acta Crystallographica Section F: Structural Biology and Crystallization Communications are provided here courtesy of
International Union of Crystallography