Crystal structure of rabbit muscle glycogen phosphorylase a in complex with a potential hypoglycaemic drug at 2.0 Å resolution

Biochimica et Biophysica Acta 1647 (2003) 325 – 332 www.bba-direct.com

Crystal structure of rabbit muscle glycogen phosphorylase a in complex ˚ resolution with a potential hypoglycaemic drug at 2.0 A Nikos G. Oikonomakos *, Evangelia D. Chrysina, Magda N. Kosmopoulou, Demetres D. Leonidas Institute of Biological Research and Biotechnology, The National Hellenic Research Foundation, 48 Vas. Constantinou Avenue, Athens 11635, Greece Received 17 June 2002; received in revised form 12 November 2002; accepted 22 January 2003

Abstract CP320626 has been identified as a potent inhibitor, synergistic with glucose, of human liver glycogen phosphorylase a (LGPa), a possible target for type 2 diabetes therapy. CP320626 is also a potent inhibitor of human muscle GPa. In order to elucidate the structural basis of the mechanism of CP320626 inhibition, the structures of T state rabbit muscle GPa (MGPa) in complex with glucose and in ˚ resolution, and refined to crystallographic R values of 0.179 complex with both glucose and CP320626 were determined at 2.0 A ˚ from the catalytic site, where (Rfree = 0.218) and 0.207 (Rfree = 0.235), respectively. CP320626 binds at the new allosteric site, some 33 A glucose binds. The binding of CP320626 to MGPa does not promote extensive conformational changes except for small shifts of the side chain atoms of residues R60, V64, and K191. Both CP320626 and glucose promote the less active T state, while structural comparisons of MGPa – glucose – CP320626 complex with LGPa complexed with a related compound (CP403700) and a glucose analogue inhibitor indicate that the residues of the new allosteric site, conserved in the two isozymes, show no significant differences in their positions. D 2003 Elsevier Science B.V. All rights reserved. Keywords: Type 2 diabetes; Glycogen metabolism; Glycogen phosphorylase; Indole-2-carboxamide; Inhibition; X-ray structure

1. Introduction Glycogenolysis is catalysed in liver, muscle and brain by tissue-specific isoforms of glycogen phosphorylase (GP), a key enzyme of special importance for the mobilisation of glycogen deposits. GP catalyses the phosphorolytic cleavage of a-1,4-linked glucosyl units in glycogen with the formation of a-D-glucose-1-phosphate. In the muscle, glucose-1-phosphate is utilised via glycolysis to generate metabolic energy, while in the liver, it is mostly

Abbreviations: GP, glycogen phosphorylase; GPb, glycogen phosphorylase b; GPa, glycogen phosphorylase a; LGPa, human liver GPa; MGPb, rabbit muscle GPb; MGPa, rabbit muscle GPa; Ser14-P, Ser14-phosphate; a-D-glucose, glucose-1-P, a-D-glucose 1-phosphate; 1-GlcNAc, N-acetyl-hD-glucopyranosylamine; CP320626, 5-chloro-1H-indole-2-carboxylic acid [1-(4-fluorobenzyl)-2-(4-hydroxypiperidin-1-yl)-2-oxoethyl]amide; rms deviation, root mean square deviation; W1807, ()(S)-3-isopropyl 4-(2chlorophenyl)-1,4-dihydro-1-ethyl-2-methyl-pyridine-3,5,6-tricarboxylate * Corresponding author. Tel.: +30-10-7273761; fax: +30-10-7273758. E-mail address: [email protected] (N.G. Oikonomakos).

converted to glucose, which is released for the benefit of other tissues [1,2]. Because of the biological importance in glycogen metabolism, GP has been exploited as a possible molecular target for potent inhibitors, which may prevent unwanted glycogen breakdown and that may be relevant to the control of elevated blood glucose concentrations in type 2 diabetes [3 –5]. GP is a regulatory enzyme that can exist in two interconvertible forms: the dephosphorylated low-activity glycogen phosphorylase b (GPb) and the phosphorylated high-activity glycogen phosphorylase a (GPa). In both forms, allosteric effectors can promote equilibrium between a less active T state and an active R state according to the Monod – Wyman – Changeux model for allosteric proteins [6]. Through structural studies on rabbit muscle GPb (MGPb) and GPa (MGPa) [7 –10] and recently on the human liver GPa (LGPa) [11], the conformational changes, which take place following conversion of GPb to GPa by phosphorylation and allosteric transition (T ! R), have been characterised. In the T state structure of MGPb, Nterminal residues 10– 22 (residues 1– 10 are not located)

1570-9639/03/$ - see front matter D 2003 Elsevier Science B.V. All rights reserved. doi:10.1016/S1570-9639(03)00085-2

326

N.G. Oikonomakos et al. / Biochimica et Biophysica Acta 1647 (2003) 325–332

are in an extended h-stranded conformation, poorly ordered and make intrasubunit contacts. On phosphorylation, the N-terminal residues 5 – 22 which contain the phosphorylation site (Ser14) fold into a distorted 310 helix and change their contacts from intrasubunit to intersubunit. The Ser14-phosphate (Ser14-P) forms ion pairs with Arg69 of the a2 helix (residues 47– 78) and Arg43V in the capV (residues 36V to 47V) of the symmetry-related subunit (Fig. 1). On the T ! R allosteric transition, one subunit is rotated relative to the other by 9.6j so as to bring the two subunits closer together at the a2V-cap interface and causing them to move further apart at the tower/tower’ interface of the dimer, on the opposite side of the structure to the capV/a2 helix interactions. As a result, the capV/a2 interface is tightened and the other major subunit – subunit contact involving the tower helices (residues 262 –276) is weak-

ened, allowing communication of changes at the allosteric site to those at the catalytic site and vice versa. In the T state enzyme, there is no access from the buried catalytic site to the surface: access to this site is partly blocked by the 280s loop (residues 282 to 286). The T ! R allosteric transition involves a movement of the 280s loop away from the entrance to the catalytic site that allows a crucial arginine, Arg569, to enter the catalytic site in place of Asp283. Access for the glycogen substrate to the catalytic site is open, and the 380s loop (residues 376 – 386) becomes disordered [12,13]. CP320626 (Scheme 1), a member of the indole2-carboxamide series, is a potent inhibitor of LGPa (IC50 = 205 nM) that acts synergistically with glucose. Additionally, CP320626 is a potent inhibitor of human MGPa (IC50 = 83 nM) [14]. Related compounds have been studied clinically, and are currently in phase II

Fig. 1. A schematic diagram of the MGPa – glucose – CP320626 dimeric complex molecule viewed down the 2-fold. The positions are shown for the Ser14-P site, the catalytic, and the new allosteric inhibitor sites. The catalytic site is buried at the center of the subunit, and is accessible to the bulk solvent through a 15˚ -long channel. Glucose (shown in red), a competitive inhibitor of the enzyme that promotes the less active T state through stabilization of the closed position A ˚ from the catalytic site. The new allosteric of the 280s loop (cream), binds at this site. The Ser14-P (magenta) of the N-terminal tail (cream) is over 40 A ˚ from the inhibitor site, located inside the central cavity formed on association of the two subunits, binds the CP320626 molecule (orange) and is some 33 A catalytic site. Inset: Details of the interactions of Ser14-PVwith Arg69Vfrom the a2Vhelix and Arg43 from the cap region of the other monomer. The figure was prepared using the programs Molscript [32] and Raster3D [33].

N.G. Oikonomakos et al. / Biochimica et Biophysica Acta 1647 (2003) 325–332

Scheme 1. 5-chloro-1H-indole-2-carboxylic acid [1-(4-fluorobenzyl)-2-(4hydroxypiperidin-1-yl)-2-oxoethyl]amide (CP320626), showing the numbering system used.

trials [4]. To understand the structural basis of its potency and mechanism of inhibition, the structures of MGPb – CP320626 and MGPb – CP320626 – glucose complexes ˚ [15] and 1.76 A ˚ resoluhave been determined at 2.3 A tion [16], respectively. Moreover, LGPa was independently reported to bind two related compounds, CP403700 (IC50 = 45 nM) and CP526423 (IC50 = 6 nM) potent inhibitors of the liver isozyme [17]. The structures revealed a surprising, new allosteric binding site, located in the region of the central cavity, at the subunit interface. These studies have given further insights into structural aspects of the enzyme enhancing our understanding of how the enzyme recognizes and specifically binds allosteric ligands, which could be of potential therapeutic value in the treatment of type 2 diabetes. The observed synergism between CP320626 and glucose could be an additional physiological benefit of a LGPa inhibitor, expected to modulate its effects only when glucose concentrations are increased in vivo, and therefore should minimize the risk of hypoglycaemia. The importance of the search for allosteric modulators of cell-surface receptors such as ligand-gated ion channels and G protein-coupled receptors has been recently reviewed [18]. To elucidate the molecular basis of CP320626 potent inhibitory action on MGPa, we have now determined the crystal structures of the co-crystallised MGPa –glucose and ˚ resolution. MGPa– glucose– CP320626 complexes at 2.0 A The structure of the MGPa –glucose– CP320626 complex reveals that CP320626 binds at the new allosteric inhibitor site and provides rationalisations for the high affinity of this compound.

2. Materials and methods 2.1. Crystallisation and data collection Tetragonal ( P4 3 2 1 2), T state MGPa crystals were grown as described previously [19]. Similarly, the MGPa – glucose – CP320626 complex was co-crystallised with

327

CP320626 at a molar ratio of CP320626/enzyme c 4 with 50 mM glucose, after the mixture was filtered out to remove precipitates. Data for MGPa – glucose crystals were collected at room temperature, on an image plate on the ˚ ) in Hamburg, at a maximum beamline X11 (k = 0.9058 A ˚ resolution of 2.0 A. Data for MGPa– glucose– CP320626 ˚ on the complex were also collected at a resolution of 2.0 A ˚ beamline station X31 (k = 1.05 A) in Hamburg Outstation. Crystal orientation, integration of reflections, inter-frame scaling, partial reflection summation, data reduction and post-refinement were all completed using the HKL program suite [20] (Table 1). 2.2. Refinement Crystallographic refinement of the MGPa –glucose complex was performed with X-PLOR version 3.8 [21] using bulk solvent correction and including all reflections ˚ . The starting protein structure between 15.0 and 2.0 A was a room temperature model of the MGPa – glucose (with water molecules removed) kindly provided by Prof. R.J. Fletterick. Interpretation of the difference, SIGMAA [22] weighted, density maps (mFo-DFc and 2mFo-DFc) using the program O [23] showed strong density for glucose bound at the catalytic site. The model was then refined using the conventional positional and restrained individual B-factor refinement protocol in X-PLOR (Table 1). A Luzatti plot [24] suggests an average positional error ˚ . The stereochemistry of the of approximately 0.21 A model was examined with the program PROCHECK [25] and the Ramachandran plot showed 91.0% of residues in most favoured regions. In the case of the MGPb– glucose– CP320626 complex, ˚ resolution the starting protein structure was the 2.0 A refined model of the rmGPa – glucose complex. The difference Fourier maps indicated binding of CP320626 at the new allosteric site and glucose at the catalytic site. Crystallographic refinement was performed as above with all ˚ resolution and no sigma cut-off data between 30.0 and 2.0 A (Table 1). The Luzatti plot suggests an average positional ˚ , and the model displays good error of approximately 0.25 A stereochemistry with 90.3% of residues in most favoured regions. The structures were analysed with the graphics program O [23]. Hydrogen-bonds were assigned if the distance ˚ between the electronegative atoms was less than 3.3 A and if both angles between these atoms and the preceding atoms were greater than 90j. van der Waals’ interactions were assigned for non-hydrogen atoms separated by less ˚ rms deviations in Ca atoms positions were than 4 A determined using the program LSQKAB [26]. Coordinate sets used for comparison were: LGPa – 1-GlcNAc – CP403700 complex (PDB code 1EXV), and MGPb –glucose –CP320626 complex (PDB code 1H5U). Coordinates ˚ resolution T state MGPa –glucose and MGPa– for the 2.0 A glucose –CP320626 complexes have been deposited with

328

N.G. Oikonomakos et al. / Biochimica et Biophysica Acta 1647 (2003) 325–332

Table 1 Diffraction data and refinement statistics for T state MGPa – Glc and MGPa – Glc – CP320626 complexes

Space group Number of images (degrees) Unit cell ˚) dimensions (A Resolution ˚) range (A Number of observations Number of unique reflections < I/r (I)b > (outermost shell) ˚ 2) B values (A (Wilson plot) Completeness (outermost shell) (%) Rmerge (outermost shell)c ˚) Outermost shell (A Multiplicity (outermost shell) Refinement ˚) (resolution) (A Number of reflections used (free) Residues included

MGPa – glucose

MGPa – glucose – CP320626

P43212 79 (0.9j)a

P43212 60 (0.9j)a

Table 1 (continued )

Average B (A˚2) for ligands CP320626 Water

MGPa – glucose

MGPa – glucose – CP320626

– 39.1

68.8 40.5

a

0.9j is the rotation range per image. r(I) is the standard deviation of I. c Rmerge = Ai Ah | < Ih>  Iih|/Ai Ah Iih, where < Ih> and Iih are the mean and ith measurement of intensity for reflection h, respectively. d Crystallographic R = A | | Fo |  | Fc| |/A | Fo|, where | Fo| and | Fc| are the observed and calculated structure factor amplitudes, respectively. Rfree is the corresponding R value for a randomly chosen 5% of the reflections that were not included in the refinement. b

a = b = 128.3, c = 116.9 15.0 – 2.0

a = b = 128.4, c = 116.4 30.0 – 2.0

453,933

343,075

65,781 (3229)

64,780 (3121)

16.7 (4.1)

13.2 (3.3)

26.4

27.4

99.3 (99.2)

97.5 (95.7)

the RCSB Protein Data Bank (http://www.rcsb.org/) (codes 1LWN and 1LWO, respectively).

3. Results and discussion 0.067 (0.490)

0.078 (0.658)

2.03 – 2.00 5.9 (5.8)

2.03 – 2.00 4.8 (4.2)

14.8 – 2.0

29.4 – 2.0

62,405 (3336)

61,460 (3287)

(5 – 251), (261 – 314), (325 – 838) 6630

(5 – 251), (261 – 314), (325 – 838) 6630

327

309

15 (PLP), 5 (Ser14-P), 12 (Glc) 17.9 (21.8) 25.4 (29.7)

15 (PLP), 5 (Ser14-P), 12 (Glc), 31 (CP320626) 20.7 (23.5) 30.7 (31.5)

0.007

0.007

1.31

1.26

3.1. Interactions between glucose and catalytic site residues

24.3

24.1

0.68

0.69

Average B (A˚ 2) for residues Overall CA, C, N, O Side chain

31.4 28.8 33.4

33.8 31.5 35.1

Average B (A˚ 2) for ligands PLP Glucose Ser14-P

18.6 20.5 38.0

20.0 19.8 59.7

The mode of binding and the interactions that glucose makes with MGPa in the absence or presence of CP320626, are almost identical with those previously reported for MGPa [19,27] and for MGPb [16,28], respectively. Binding of glucose at the catalytic site promotes localisation of the 280s loop in its inactive T state conformation by making a hydrogen bond to Asp283 via a water molecule (through its 1-OH), a hydrogen bond to Asp284 (through its 2-OH) and van der Waals’ contacts to Asn284 CG1, OD1 and ND2 atoms (through its 1-OH and 2-OH). The structural results indicate that, in the presence of CP320626, glucose can be accommodated at the catalytic site of MGPa with essentially

Number of protein atoms Number of water molecules Number of ligand atoms

Final R (Rfree) (%)d R (Rfree) (outermost shell) (%) Rmsd in bond ˚) lengths (A Rmsd in bond angles (j) Rmsd in dihedral angles (j) Rmsd in improper angles (j)

Crystallographic data collection, processing and refinement statistics for the structure determinations of co-crystallised MGPa – glucose and MGPa – glucose – CP320626 complexes are summarised in Table 1. The overall architecture of the T state MGPa with the location of the Ser14-P site, catalytic, and the new allosteric inhibitor site, with the two subunits of the dimer related by a crystallographic 2fold symmetry axis is presented in Fig. 1. For both complexes, electron density maps indicated that glucose bound tightly at the catalytic site of MGPa. Additional density at the new allosteric site, in the MGPa –glucose –CP320626, indicated binding of CP320626. After refinement, the (2Fo – Fc) Fourier map suggested partial occupancy of the site by CP320626. There was good density for the 4-fluorobenzyl and 4-hydroxy-piperidyl moieties, and the chlorine of the 1H-indole ring and less satisfactory density for the 1H-indole ring. Despite the low occupancy of CP320626, the density was well resolved to indicate the position of CP320626. We describe briefly the glucose interactions at the catalytic site and in more detail the CP320626 interactions at the new allosteric site.

N.G. Oikonomakos et al. / Biochimica et Biophysica Acta 1647 (2003) 325–332

no disturbance of the structure (there are no significant differences in amino acid side chain positions) and the interactions it makes are similar to those observed in the MGPa– glucose complex. 3.2. Interactions between CP320626 and new allosteric site residues CP320626 binds in the central binding cavity, a solvent filled central cavity, surrounded on each side by residues from the N-terminal domains of both subunits. The cavity is closed at one end by residues from the cap and a2 helices (Arg33, His34, Arg60 and Asp61 and their symmetryrelated residues) and at the other end by the tower helices (residues Asn270, Glu273, Ser276 and their symmetryrelated residues). The CP320626 binding site is some 33

329

˚ from the catalytic site. Fig. 2 shows the contacts made at A the new allosteric site in the MGPa– glucose– CP320626 complex. The contacts are very similar to those observed for the MGPb – CP320626 complex [15,16]. In brief, they comprise aromatic/aromatic interactions (4-fluoro-benzyl group/Phe53V side chain), amino – aromatic interactions (His57V/4-fluoro-benzyl group), CH/k electron interactions (Arg60 side chain/indole ring, CE3 and CZ3 of Trp67/ indole ring, CG and CD of Pro229/indole ring, Val40Vside chain/indole ring, CA, CB, CD of Pro188V/4-fluoro-benzyl group), and non-polar/non-polar interactions (CB of Ala192/C15 of piperidyl group, CG2 of Thr38V/C17, and CD1 of Leu39V/C17). These contacts, together with the specific hydrogen bonding interactions appear to be the major source of the binding energy that characterises the nM affinity of the ligand for human muscle GPa.

˚ 2Fo  Fc electron density map (contoured at 1.0r level) of the bound CP320626 to MGPa calculated before incorporating the coordinates of Fig. 2. (a) The 2 A the ligand. (b) Interactions between CP320626 and MGPa and water structure in the vicinity of the new allosteric site, shown in stereo. Residues from subunit 1 are coloured green, the CP320626 molecule in orange and their symmetry-related equivalents (from subunit 2) blue and cyan for the protein and ligand, respectively. The figure was produced using Xobjects (M.E.M. Noble, unpublished results).

330

N.G. Oikonomakos et al. / Biochimica et Biophysica Acta 1647 (2003) 325–332

3.3. Structural comparisons and the mechanism of inhibition Although the quaternary structure of MGPa –glucose is T state, the conformation of residues at the Ser14-P site and the allosteric site are almost identical to those observed in the R state structure of MGPa [9]; both the N-terminal tail with Ser14-P is localised and the AMP site is in the highaffinity state. However, the presence of glucose at the catalytic site of MGPa prevents the propagation of conformational changes from the subunit interface to the catalytic site and the MGPa – glucose structure is considered to represent a conformation intermediate in the T ! R allosteric transition [19]. In the T state LGPa [11], the N-terminal region is completely disordered, as expected for the binding

of a T state inhibitor that promotes the transition from the R to T state. The quaternary structure of the MGPa –glucose– CP320626 complex is almost identical to that of the MGPa –glucose complex; both the N-terminal tail and the AMP allosteric site are localised in the intermediate T ! R state. The LSQKAB comparison of the structure of the MGPa –glucose complex with the structure of the MGPa– glucose –CP320626 complex showed that the positions of the Ca atoms for well-defined residues (24 – 208, 212– 249, 261– 313, 325 –547, and 558 –836) deviate from their mean ˚ , indicating that the complexes are positions by 0.118 A similar in their overall conformation to within the limits of the resolution of the study. The most significant shifts that accompany CP320626 binding are at the new allosteric site (Fig. 3a). These include small shifts of the atoms surround-

Fig. 3. (a) Stereodiagram showing a comparison between MGPa – glucose complex and MGPa – glucose – CP320626 complex in the vicinity of the new allosteric site. Green: MGPa – glucose – CP320626 complex (subunit 1); orange: CP320626 (subunit 1); blue: MGPa – glucose – CP320626 complex (subunit 2); cyan: CP320626 (subunit 2); white: MGPa – glucose complex. (b) Stereodiagram showing a comparison of CP320626 bound to MGPa – glucose complex with bound CP403700 to LGPa – 1-GlcNAc complex (code 1exv) in the vicinity of the new allosteric inhibitor site. Green: MGPa – glucose – CP320626 complex (subunit 1); orange: CP320626 (subunit 1); blue: MGPa – glucose – CP320626 complex (subunit 2); cyan: CP320626 (subunit 2); white: LGPa – 1-GlcNAc – CP403700 complex. The figures were produced using Xobjects (M.E.M. Noble, unpublished results).

N.G. Oikonomakos et al. / Biochimica et Biophysica Acta 1647 (2003) 325–332

ing the inhibitor, i.e. of residues Arg60, Val64, and Lys191 that undergo conformational changes to accommodate the ligand. There are no changes at the catalytic site, while the tower helices and the capV-a2 helix contacts in the Nterminal region of the molecule have the same positions as in the MGPa –glucose complex. On the contrary, the binding of W1807 to the allosteric site of the T state MGPa promotes extensive conformational changes both in the vicinity of the allosteric site and the subunit interface, such as a disordering of the N-terminal tail, and shifts of residues 192 –196 and residues 43V–49Vto accommodate the ligand [19,29]. These changes transform the MGPa– glucose intermediate T ! R state to a TVstate like, already observed with MGPb– W1807 complex [30]. The structure of the T state LGPa in complex with 1GlcNAc, a glucose analogue inhibitor of MGPb and MGPa [31], and CP403700 (IC50 = 45 nM) has been recently ˚ resolution [17]. Structural comparison described at 2.4 A between the CP403700 complex and the CP320626 complex over the conserved 27 residues of the new allosteric inhibitor site (37V– 40V, 53V– 57V, 60 –67, 188 – 192, 185V– ˚ for Ca atoms. 188V, and 229) gave rms deviations of 0.595 A The superposition of the LGPa – 1-GlcNAc – CP403700 complex on the MGPa – glucose – CP320626 complex is shown in Fig. 3b. In order to assess the validity of using MGPb as a model for the LGPa in structure-assisted drug design, we had previously superimposed the LGPa – 1GlcNAc – CP403700 complex with the MGPb – glucose – CP320626 complex over residues of the new allosteric inhibitor site [16]. The superposition showed only minor differences in the positions of the conserved residues, with ˚ for Ca atoms, which indicates rms deviations of 0.307A little conformational change between the two complexes. These observations suggest that the crystal structure of T state MGPb, compared to that of T state MGPa, is a better model in the design and optimisation of inhibitors. In conclusion, CP320626 binds at a site distant from the catalytic site and exerts its effects by stabilising a T statelike conformation, and hence inhibits MGPa by locking the enzyme in an inactive state. Glucose, bound at the catalytic site, interacts with the side chains of Asp283 and Asn284 holding the 280s loop (residues 282 – 286) in its close conformation that blocks access for the substrate to the catalytic site. By promoting the T-like state, CP320626 will also promote synergistic binding of glucose.

Acknowledgements This work was supported in part by Greek GSRT through PENED-204/2001, ENTER-EP6/2001and a Joint Research and Technology project between Greece and Cyprus (2001 – 2003), a Royal Society Joint Project (to Prof. L.N. Johnson and N.G. Oikonomakos) and EMBL, Hamburg Outstation (HPRI-CT-1999-00017). We thank V.T. Skamnaki, K.E. Tsitsanou and S.E. Zographos for help in crystallisations

331

and X-ray data collection. We also wish to acknowledge the assistance of the staff at EMBL, Hamburg, for providing excellent facilities for X-ray data collection.

References [1] L.N. Johnson, J. Hajdu, K.R. Acharya, D.I. Stuart, P.J. McLaughlin, N.G. Oikonomakos, D. Barford, Glycogen phosphorylase b, in: G. Herve (Ed.), Allosteric Enzymes, CRC Press, Boca Raton, FL, 1989, pp. 81 – 127. [2] C.B. Newgard, P.K. Hwang, R.J. Fletterick, The family of glycogen phosphorylases: structure and function, CRC Crit. Rev. Biochem. Mol. Biol. 24 (1989) 69 – 99. [3] L. Somsa´k, L. Kova´cs, M. To´th, E. OUsz, L. Szila´gyi, Z. Gyo¨rgydea´k, Z. Dinya, T. Docsa, B. To´th, P. Gergely, Synthesis of and a comparative study on the inhibition of muscle and liver glycogen phosphorylases by epimeric pairs of D-gluco- and D-xylopyranosylidene-spiro(thio)hydantoins and N-(D-glucopyaranosyl) amides, J. Med. Chem. 44 (2001) 2843 – 2848. [4] J.L. Treadway, P. Mendys, D.J. Hoover, Glycogen phosphorylase inhibitors for treatment of type 2 diabetes mellitus, Expert Opin. Investig. Drugs 10 (2001) 439 – 454. [5] N.G. Oikonomakos, M. Kosmopoulou, S.E. Zographos, D.D. Leonidas, E.D. Chrysina, L. Somsa´k, V. Nagy, Z. Hadady, T. Docsa, B. To´th, P. Gergely, Binding of N-acetyl-NV-h-D-glucopyranosyl urea Nbenzoyl-NV-h-D-glucopyranosyl urea to glycogen phosphorylase b: kinetic and crystallographic studies, Eur. J. Biochem. 269 (2002) 1684 – 1696. [6] J. Monod, J. Wyman, J.-P. Changeux, On the nature of allosteric transitions: a plausible model, J. Mol. Biol. 12 (1965) 88 – 118. [7] S.R. Sprang, K.R. Acharya, E.J. Goldsmith, D.I. Stuart, K. Varvill, R.J. Fletterick, N.B. Madsen, L.N. Johnson, Structural changes in glycogen phosphorylase induced by phosphorylation, Nature 336 (1988) 215 – 221. [8] D. Barford, L.N. Johnson, The allosteric transition of glycogen phosphorylase, Nature 340 (1989) 609 – 616. [9] D. Barford, S.H. Hu, L.N. Johnson, Structural mechanism for glycogen phosphorylase control by phosphorylation and AMP, J. Mol. Biol. 218 (1991) 233 – 260. [10] S.R. Sprang, S.G. Withers, E.J. Goldsmith, R.J. Fletterick, N.B. Madsen, Structural basis for activation of glycogen phosphorylase b by adenosine monophosphate, Science 254 (1991) 1367 – 1371. [11] V.L. Rath, M. Ammirati, P.K. LeMotte, K.F. Fennell, M.N. Mansour, D.E. Danley, T.R. Hynes, G.K. Schulte, D.J. Wasliko, J. Pandit, Activation of human liver glycogen phosphorylase by alteration of the secondary structure and packing of the catalytic core, Mol. Cell 6 (2000) 139 – 148. [12] L.N. Johnson, Glycogen phosphorylase: control by phosphorylation and allosteric effectors, FASEB J. 6 (1992) 2274 – 2282. [13] L.N. Johnson, R.J. Lewis, Structural basis for control by phosphorylation, Chem. Rev. 101 (2001) 2209 – 2242. [14] D.J. Hoover, S. Lefkowitz-Snow, J.L. Burgess-Henry, W.H. Martin, S.J. Armento, I.A. Stock, R.K. McPherson, P.E. Genereux, E.M. Gibbs, J.L. Treadway, Indole-2-carboxamide inhibitors of human liver glycogen phosphorylase, J. Med. Chem. 41 (1998) 2934 – 2938. [15] N.G. Oikonomakos, V.T. Skamnaki, K.E. Tsitsanou, N.G. Gavalas, L.N. Johnson, A new allosteric site in glycogen phosphorylase b as a target for drug interactions, Structure 8 (2000) 575 – 584. [16] N.G. Oikonomakos, S.E. Zographos, V.T. Skamnaki, G. Archontis, ˚ resolution crystal structure of glycogen phosphorylase b The 1.76 A complexed with glucose, and CP320626, a potential antidiabetic drug, Bioorg. Med. Chem. 10 (2002) 1313 – 1319. [17] V.L. Rath, M. Ammirati, D.E. Danley, J.L. Ekstrom, E.M. Gibbs, T.R. Hynes, A.M. Mathiowetz, R.K. McPherson, T.V. Olson, J.L.

332

[18]

[19]

[20] [21] [22] [23]

[24] [25]

[26]

N.G. Oikonomakos et al. / Biochimica et Biophysica Acta 1647 (2003) 325–332 Treadway, D.J. Hoover, Human liver glycogen phosphorylase inhibitors bind at a new allosteric site, Chem. Biol. 7 (2000) 677 – 682. Christopoulos, Allosteric binding sites on cell-surface receptors: novel targets for drug discovery, Nat. Rev., Drug Discov. 1 (2002) 198 – 210. N.G. Oikonomakos, K.E. Tsitsanou, S.E. Zographos, V.T. Skamnaki, S. Goldmann, H. Bischoff, Allosteric inhibition of glycogen phosphorylase a by the potential antidiabetic drug 3-isopropyl 4-(2-chlorophenyl)1,4-dihydro-1-ethyl-2-methyl-pyridine-3,5,6-tricarboxylate, Protein Sci. 8 (1999) 1930 – 1945. Z. Otwinowski, W. Minor, Processing of X-ray diffraction data collected in oscillation mode, Methods Enzymol. 276 (1997) 307 – 326. A.T. Bru¨nger, X-PLOR Version 3.1. A System for X-ray Crystallography and NMR, Yale Univ. Press, New Haven, 1992. R.J. Read, Improved coefficients for map calculation using partial structures with errors, Acta Crystallogr. A42 (1986) 140 – 149. T.A. Jones, J.Y. Zou, S.W. Cowan, M. Kjeldgaard, Improved methods for the building of protein models in electron density maps and the location of errors in these models, Acta Crystallogr. A47 (1991) 110 – 119. V. Luzatti, Traitement statistique des erreurs dans la determination des structure crystallines, Acta Crystallogr. 5 (1952) 802 – 810. R.A. Laskowski, M.W. MacArthur, D.S. Moss, J.M. Thornton, PROCHECK: a programme to check the stereochemical quality of protein structures, J. Appl. Crystallogr. 26 (1993) 283 – 291. Collaborative computational project, number 4, the CCP4 suite: programs for protein crystallography, Acta Crystallogr. D50 (1994) 760 – 763.

[27] N.G. Oikonomakos, S.E. Zographos, V.T. Skamnaki, K.E. Tsitsanou, L.N. Johnson, Flavopiridol inhibits glycogen phosphorylase by binding at the inhibitor site, J. Biol. Chem. 275 (2000) 34566 – 34573. [28] J.L. Martin, K. Veluraja, L.N. Johnson, G.W.J. Fleet, N.G. Ramsden, I. Bruce, N.G. Oikonomakos, A.C. Papageorgiou, D.D. Leonidas, H.S. Tsitoura, Glucose analogue inhibitors of glycogen phosphorylase: the design of potential drugs for diabetes, Biochemistry 30 (1991) 10101 – 10116. [29] K.E. Tsitsanou, V.T. Skamnaki, N.G. Oikonomakos, Structural basis of the synergistic inhibition of glycogen phosphorylase a by caffeine and a potential antidiabetic drug, Arch. Biochem. Biophys. 384 (2000) 245 – 254. [30] S.E. Zographos, N.G. Oikonomakos, K.E. Tsitsanou, D.D. Leonidas, E.D. Chrysina, V.T. Skamnaki, H. Bischoff, S. Goldman, M. Schram, K.A. Watson, L.N. Johnson, The structure of glycogen phosphorylase b with an alkyl-dihydropyridine-dicarboxylic acid compound, a novel and potent inhibitor, Structure 5 (1997) 1413 – 1425. [31] N.G. Oikonomakos, M. Kontou, S.E. Zographos, K.A. Watson, L.N. Johnson, C.J.F. Bichard, G.W.J. Fleet, K.R. Acharya, N-acetyl-h-Dglucopyranosylamine: a potent T state inhibitor of glycogen phosphorylase. A comparison with a-D-glucose, Protein Sci. 4 (1995) 2469 – 2477. [32] P. Kraulis, MOLSCRIPT: a program to produce both detailed and schematic plots of protein structures, J. Appl. Crystallogr. 24 (1991) 946 – 950. [33] E.A. Merritt, D.J. Bacon, Raster3D: photorealistic molecular graphics, Methods Enzymol. 277 (1997) 505 – 524.