Crystal structure of human PNP complexed with guanine

Arts & Architecture

6 pages

Please download to get full document.

View again

of 6
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Crystal structure of human PNP complexed with guanine
  Crystal structure of human PNP complexed with guanine Walter Filgueira de Azevedo Jr., a,b,* Fernanda Canduri, a,b Denis Marangoni dos Santos, a,b Jos  ee Henrique Pereira, a,b M  aarcio Vinicius Bertacine Dias, a Rafael Guimar ~ aaes Silva, c Maria Anita Mendes, b,d Luiz Augusto Basso, c M  aario S  eergio Palma, b,d and Di  oogenes Santiago Santos c,e,* a Departamento de F   ıısica, UNESP, S  ~ aa o Jos  ee  do Rio Preto, SP 15054-000, Brazil  b Center for Applied Toxinology, Instituto Butantan, S  ~ aa o Paulo, SP 05503-900, Brazil  c Rede Brasileira de Pesquisas em Tuberculose, Departamento de Biologia Molecular e Biotecnologia, UFRGS, Porto Alegre, RS 91501-970, Brazil  d Laboratory of Structural Biology and Zoochemistry-CEIS/Department of Biology, Institute of Biosciences, UNESP, Rio Claro, SP 13506-900, Brazil  e Faculdade de Farm  aa cia/Instituto de Pesquisas Biom  ee dicas, Pontif   ııcia Universidade Cat  oo lica do Rio Grande do Sul, Porto Alegre, RS, Brazil  Received 23 October 2003 Abstract Purine nucleoside phosphorylase (PNP) catalyzes the phosphorolysis of the  N  -ribosidic bonds of purine nucleosides anddeoxynucleosides. PNP is a target for inhibitor development aiming at T-cell immune response modulation and has been submittedto extensive structure-based drug design. More recently, the 3-D structure of human PNP has been refined to 2.3  AA resolution, whichallowed a redefinition of the residues involved in the substrate-binding sites and provided a more reliable model for structure-baseddesign of inhibitors. This work reports crystallographic study of the complex of Human PNP:guanine (HsPNP:Gua) solved at 2.7  AAresolution using synchrotron radiation. Analysis of the structural differences among the HsPNP:Gua complex, PNP apoenzyme,and HsPNP:immucillin-H provides explanation for inhibitor binding, refines the purine-binding site, and can be used for futureinhibitor design.   2003 Elsevier Inc. All rights reserved. Keywords:  PNP; Synchrotron radiation; Structure; Drug design PNP catalyzes the reversible phosphorolysis of theribonucleosides and 2 0 -deoxyribonucleosides of guanine,hypoxanthine, and a number of related nucleosidecompounds [1], except adenosine (Fig. 1). Human PNPis an attractive target for drug design and it has beensubmitted to extensive structure-based design. PNP in-hibitors could be used in the following applications: (1)treatment of T-cell leukemia; (2) suppression of thehost-vs.-graft response in organ transplantation recipi-ents; (3) treatment of secondary or xanthine gout byrestricting purine catabolites to the more soluble nucle-osides; and (4) in combination with nucleosides to pre-vent their degradation by PNP metabolism [2]. Morerecently, the 3-D structure of human PNP has beenrefined to 2.3  AA resolution [3], which allowed a redefi-nition of the residues involved in the substrate-bindingsites and provided a more reliable model for structure-based design of inhibitors. The crystallographic struc-ture is a trimer and analysis of human PNP in solution,using SAXS, confirmed that the crystallographic trimeris conserved in solution [4].We have obtained the crystallographic structure of the complex between HsPNP and guanine (HsPNP:Gua). Previously reported structure for the same com-plex showed poor stereochemistry quality [2,5] and therefined model does not show water molecules. Ouranalyses of the HsPNP:Gua structural data and struc-tural differences between the PNP apoenzyme andHsPNP:Gua complex provide explanation for substratebinding, refine the purine-binding site, identify watermolecules, a new phosphate-binding site, and can beused for future inhibitor design. * Corresponding authors. Fax: +55-17-221-2247. E-mail addresses: (W.F. de AzevedoJr.), (D.S. Santos).0006-291X/$ - see front matter    2003 Elsevier Inc. All rights reserved.doi:10.1016/j.bbrc.2003.10.190Biochemical and Biophysical Research Communications 312 (2003) 767–772 BBRC  Materials and methods Crystallization and data collection.  Recombinant human PNP wasexpressed and purified as previously described [6]. HsPNP:Gua wascrystallized using the experimental conditions described elsewhere[7,8]. In brief, a PNP solution was concentrated to 13mgmL  1 against10mM potassium phosphate buffer (pH 7.1) and incubated in thepresence of 0.6mM of guanine (Sigma). Hanging drops were equili-brated by vapor diffusion at 25  C against reservoir containing 19%saturated ammonium sulfate solution in 0.05M citrate buffer (pH 5.3).In order to increase the resolution of the HsPNP:Gua crystal, wecollected data from a flash-cooled crystal at 104 K. Prior to flashcooling, glycerol was added, up to 50% by volume, to the crystalliza-tion drop. X-ray diffraction data were collected at a wavelength of 1.4310  AA using the Synchrotron Radiation Source (Station PCr, Lab-orat  oorio Nacional de Luz S  ııncrotron, LNLS, Campinas, Brazil) and aCCD detector (MARCCD) with an exposure time of 30s per image ata crystal to detector distance of 120mm. X-ray diffraction data wereprocessed to 2.7  AA resolution using the program MOSFLM and scaledwith the program SCALA [9].Upon cooling the cell parameters shrank from  a  ¼  b  ¼  142 : 90  AA, c  ¼  165 : 20  AA to  a  ¼  b  ¼  141 : 07  AA, and  c  ¼  162 : 37  AA. For HsPNP:Guacomplex the volume of the unit cell is 2.847  10 6   AA 3 compatible withone monomer in the asymmetric unit with  V   m  value of 4.92  AA 3 /Da.Assuming a value of 0.25cm 3 g  1 for the protein partial specificvolume, the calculated solvent content in the crystal is 75% and thecalculated crystal density 1.09gcm  3 . Crystal structure.  The crystal structure of the HsPNP:Gua wasdetermined by standard molecular replacement methods using theprogram AMoRe [10], using as search model the structure of HsPNP (PDB Access Code: 1M73) [3]. Structure refinement wasperformed using X-PLOR [11]. The atomic positions obtained frommolecular replacement were used to initiate the crystallographic re-finement. The overall stereochemical quality of the final model forHsPNP:Gua complex was assessed by the program PROCHECK[12]. Atomic models were superposed using the program LSQKABfrom CCP4 [9]. Results and discussion Molecular replacement and crystallographic refinement The standard procedure of molecular replacementusing AMoRe [10] was used to solve the structure. Aftertranslation function computation the correlation was of 74% and the  R factor  of 31%. The highest magnitude of the correlation coefficient function was obtained for theEuler angles  a  ¼  113 : 7  ,  b  ¼  57 : 5  , and  c  ¼  158 : 0  . Thefractional coordinates are  T   x  ¼  0 : 164,  T   y   ¼  0 : 625, and T   z   ¼  0 : 032. At this stage 2  F  obs    F  calc  omit maps werecalculated. These maps showed clear electron density forthe guanine in the complex. Further refinement inX-PLOR continued with simulated annealing using the Fig. 1. The enzymatic reaction catalyzed by PNP.Table 1Data collection and refinement statisticsCell parameters  a  ¼  b  ¼  141 : 07  AA, c  ¼  162 : 37  AA a  ¼  b  ¼  90 : 00  , c  ¼  120 : 00  Space group R32No. of measurements with  I   >  2 r  ( I  ) 46,457No. of independent reflections 18,226Completeness in the range from 56.80 to2.60  AA (%)91.0  R syma (%) 7.0Highest resolution shell (  AA) 2.85–2.70Completeness in the highest resolution shell (%) 96.0  R syma in the highest resolution shell (%) 37.1Resolution range used in the refinement (  AA) 8.0–2.7  R factorb (%) 21.8  R freec (%) 29.3  B  values d (  AA 2 )Main chain 34.45Side chains 38.07Guanine 28.72Waters 32.14Sulfate groups 37.70No. of water molecules 38No. of sulfate groups 4 a  R sym  ¼  100 P j  I  ð h Þ  h  I  ð h Þij = P  I  ð h Þ  with  I  ð h Þ , observed intensityand  h  I  ð h Þi , mean intensity of reflection  h  over all measurement of   I  ð h Þ . b  R factor  ¼  100   P j  F  obs    F  calc j = P ð  F  obs Þ , the sums being takenover all reflections with  F  = r  ð  F  Þ  >  2 cutoff. c  R free  ¼  R factor  for 10% of the data, which were not included duringcrystallographic refinement. d  B  values ¼ average  B  values for all non-hydrogen atoms.Table 2Structural quality of the present structure and 1ULBStructure 1ULB Present workResidues in most favored regionsof the Ramachandran plot (%)73.5 82.0Residues in addition allowedregions of the Ramachandran plot (%)21.2 13.5Residues in generously allowedregions of the Ramachandran plot (%)2.4 3.7Residues in disallowed regionsof the Ramachandran plot (%)2.9 0.8Observed r.m.s.d. from ideal geometryBond lengths (  AA) 0.038 0.013Bond angles (  ) 29.64 24.90Dihedrals (  ) 1.50 1.79Highest resolution (  AA) 2.75 2.70768  W.F. de Azevedo Jr. et al. / Biochemical and Biophysical Research Communications 312 (2003) 767–772  slow-cooling protocol, followed by alternate cycles of positional refinement and manual rebuilding usingXtalView [13]. Finally, the positions of guanine, water,and sulfate molecules were checked and corrected in  F   obs   F   calc  maps. The final model has an  R factor  of 21.8%and an  R free  of 29.3%, with 38 water molecules, 4 sulfateions, and the guanine.Ignoring low-resolution data, a Luzzati plot [14] givesthe best correlation between the observed and calculateddata for a predicted mean coordinate error of 0.36  AA.The average  B  factor for main chain atoms is 34.45  AA 2 ,whereas that for side chain atoms is 38.07  AA 2 (Table 1).Comparison of the present structure with previouslydeposited atomic coordinates for the same complex in-dicates that the present structure shows better overallstereochemistry (Table 2). Furthermore, analysis of theelectron-density maps of the present structure allowedthe determination of water molecules, not identified inthe previous structure. In addition, three new humanPNP structures [3,7,8] made possible structural com-parison presented here. Overall description Analysis of the crystallographic structure of HsPNP:Gua complex indicates a trimeric structure.Each PNP monomer displays an  a = b  fold consisting of amixed  b -sheet surrounded by  a  helices. The structurecontains an eight-stranded mixed  b -sheet and a five-stranded mixed  b -sheet, which join to form a distorted b -barrel. Fig. 2 shows schematic drawings of theHsPNP:Gua complex. Ligand-binding conformational changes There is a conformational change in the PNP struc-ture when guanine binds in the active site. The overallchange is relatively small, with an r.m.s.d. in the coor-dinates of all C a  of 1.29  AA after superimposition. Thelargest movement was observed for His257 side chain,which partially occupies the purine subsite in the nativeenzyme. The residues 241–260 act as a gate that opensduring substrate binding. This gate is anchored near thecentral  b -sheet at one end and near the C-terminal helixat the other end and it is responsible for controllingaccess to the active site. The gate movement involves ahelical transformation of residues 257–265 in the tran-sition apoenzyme-complex. Fig. 3 shows the gate Fig. 2. Ribbon diagram of HsPNP:Gua generated by Molscript [23]and Raster3d [24].Fig. 3. Gate movement after binding of guanine to human PNP. W.F. de Azevedo Jr. et al. / Biochemical and Biophysical Research Communications 312 (2003) 767–772  769  movement observed in the transition from the apoen-zyme to the complex HsPNP:Gua. Phosphate-binding sites The present structure of HsPNP shows clear electron-density peaks for four sulfate groups, which are presentin high concentration in the crystallization experimentalcondition. Three of these sulfate groups have beenpreviously identified in the high-resolution structure of human PNP [3] and one new site was identified in thepresent structure. The first sulfate site, which is thecatalytic phosphate-binding site, is positioned to formhydrogen bonds to Ser33, Arg84, His86, and S220. Thesecond sulfate-binding site lies near Leu35 and Gly36and is exposed to the solvent and whether it is mecha-nistically significant or an artifact resulting from thehigh-sulfate concentration used to grow the crystals isnot known. The third identified sulfate group makeshydrogen bonds, involving residues Gln144 and Arg148from adjacent subunit. The fourth sulfate-binding sitemakes hydrogen bonds with residues Ser33 (3.5  AA) andTyr88 (3.3  AA) and it is close to guanine, making onehydrogen bond to the nitrogen (2.9  AA), and occupies the Fig. 4. Multiple modes of binding to human PNP. (A) HsPNP:guanine, (B) HsPNP:acyclovir, and (C) HsPNP:immucillin-H.770  W.F. de Azevedo Jr. et al. / Biochemical and Biophysical Research Communications 312 (2003) 767–772  ribose-binding site (Fig. 4A). The ribose-binding site ismostly hydrophobic, which is composed of several ar-omatic amino acids, including Tyr88, Phe159 (of theadjacent subunit), Phe200, His86, His257, and Met219.It is tempting to speculate that the presence of a sulfate(phosphate) group at the ribose-binding site may offerfurther hindrance to the binding of substrates, whichmay also contribute to the larger IC 50  observed forseveral inhibitors in the presence of higher phosphateconcentrations [7,8]. Interactions with guanine The specificity and affinity between enzyme and itsligand depend on directional hydrogen bonds and io-nic interactions, as well as on shape complementarityof the contact surfaces of both partners [15 – 21]. The electrostatic potential surface of the guanine com-plexed with HsPNP was calculated with GRASP [22](figure not shown). The analysis of the charge distri-bution of the binding pocket indicates the presence of some charge complementarity between inhibitor andenzyme, though most of the binding pocket is hy-drophobic.Comparison of the present structure with humanPNP complexed with acyclovir (HsPNP:Acy) [7] andimmucillin-H (HsPNP:ImmH) [8] indicates that humanPNP presents multiple modes of binding to the activesite. Figs. 4A–C show the interaction between ligandsand PNP. The main residues involved in binding in allcomplexes are Glu201, Thr242, and Asn242. Analysisof the hydrogen bonds between immucillin-H and PNPreveals eight hydrogen bonds, involving the residuesHis86, Tyr88, Glu201, Asn243, and His257. For thecomplex HsPNP:Acy five hydrogen bonds were ob-served. These hydrogen bonds involve Glu201 andAsn243. Five hydrogen bonds between guanineand human PNP, involving residues Glu201, Thr242,and Asn243 were observed. The previously describedparticipation of Lys244 [5] in ligand binding was notidentified in the present study and in the structures of human PNP complexed with inhibitors. Analysis of thecomplexes indicates that Glu201 and Thr242 occupyapproximately the same position in all the complexes.The side-chain of Asn243 shows some flexibility, whichcauses differences in the hydrogen bond pattern of thisresidue, the complexes HsPNP:ImmH and HsPNP:Guashow intermolecular hydrogen bonds involving thefollowing atom pairs: Asn243 ND2-O6 and Asn243OD1-N7. The participation of Asn243 OD1 is notobserved in the HsPNP:Acy complex. The precisedefinition of the modes of binding to human PNP mayhelp in future structure-based design of inhibitors.The atomic coordinates and the structure factors forthe complex HsPNP:Gua have been deposited in thePDB with the Accession Code: 1V2H. Acknowledgments We acknowledge the expertise of Denise Cantarelli Machado forthe expansion of the cDNA library and Deise Potrich for the DNAsequencing. This work was supported by grants from FAPESP(SMOLBNet, Proc.01/07532-0 and 02/04383-7), CNPq, CAPES andInstituto do Mil ^ eenio (CNPq-MCT). WFA (CNPq, 300851/98-7), MSP(CNPq, 300337/2003-5), and LAB (CNPq, 520182/99-5) are re-searchers for the Brazilian Council for Scientific and TechnologicalDevelopment. References [1] J.A. Montgomery, Purine nucleoside phosphorylase: a target fordrug design, Med. Res. Rev. 13 (3) (1993) 209–228.[2] S.E. Ealick, Y.S. Babu, C.E. Bugg, M.D. Erion, W.C. Guida, J.A.Montgomery, J.A. Secrist III, Application of crystallographic andmodelingmethodsinthedesignofpurinenucleosidephosphorylaseinhibitors, Proc. Natl. Acad. Sci. USA 91 (1991) 11540–11544.[3] W.F. De Azevedo, F. Canduri, D.M. dos Santos, R.G. Silva, J.S.Oliveira, L.P.S. Carvalho, L.A. Basso, M.A. Mendes, M.S. Palma,D.S. Santos, Crystal structure of human purine nucleosidephosphorylase at 2.3  AA resolution, Biochem. Biophys. Res. Com-mun. 308 (3) (2003) 545–552.[4] W.F. De Azevedo, G.C. Santos, D.M. dos Santos, J.R. Olivieri, F.Canduri, R.G. Silva, L.A. Basso, M.A. Mendes, M.S. Palma, D.S.Santos, Docking and small angle X-ray scattering studies of purine nucleoside phosphorylase, Biochem. Biophys. Res. Com-mun. 309 (2003) 928–933.[5] A. Bzowska, E. Kulikowska, D. Shugar, Purine nucleosidephosphorylases: properties, functions, and clinical aspects, Phar-macol. Ther. 88 (2000) 349–425.[6] R.G. Silva, L.P. Carvalho, J.S. Oliveira, C.A. Pinto, M.A.Mendes, M.S. Palma, L.A. Basso, D.S. Santos, Cloning, overex-pression, and purification of functional human purine nucleosidephosphorylase, Protein Expr. Purif. 27 (1) (2003) 158–164.[7] D.M. dos Santos, F. Canduri, J.H. Pereira, M.V.B. Dias, R.G.Silva, M.A. Mendes, M.S. Palma, L.A. Basso, W.F. de Azevedo,D.S. Santos, Crystal structure of human purine nucleosidephosphorylase complexed with acyclovir, Biochem. Biophys.Res. Commun. 308 (3) (2003) 553–559.[8] W.F. De Azevedo, F. Canduri, D.M. dos Santos, J.H. Pereira,M.V.B. Dias, R.G. Silva, M.A. Mendes, L.A. Basso, M.S. Palma,D.S. Santos, Structural basis for inhibition of human PNP byimmucillin-H, Biochem. Biophys. Res. Commun. 309 (2003) 922– 927.[9] Collaborative Computational Project, Number 4. Acta Crystal-logr. D 50 (1994) 760–763.[10] J. Navaza, AMoRe: an automated package for molecularreplacement, Acta Crystallogr. A 50 (1994) 157–163.[11] A.T. Br € uunger, X-PLOR Version 3.1: a System for Crystallographyand NMR, Yale University Press, New Haven, 1992.[12] R.A. Laskowski, M.W. MacArthur, D.K. Smith, D.T. Jones,E.G. Hutchinson, A.L. Morris, D. Naylor, D.S. Moss, J.M.Thorton, PROCHECK v.3.0  —  Program to Check the Stereochem-istry Quality of Protein Structures  —  Operating Instructions, 1994.[13] D.E. McRee, XtalView/Xfit  —  a versatile program for manipulat-ing atomic coordinates and electron density, J. Struct. Biol. 125(1999) 156–165.[14] P.V. Luzzati, Traitement statistique des erreurs dans la determi-nation des structures cristallines, Acta Crystallogr. 5 (1952) 802– 810.[15] W.F. De Azevedo, H.J. MuellerDieckmann, U. SchulzeGahmen,P.J. Worland, E. Sausville, S.H. Kim, Structural basis forspecificity and potency of a flavonoid inhibitor of human W.F. de Azevedo Jr. et al. / Biochemical and Biophysical Research Communications 312 (2003) 767–772  771
Related Documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks