N-(4-Substituted-benzoyl)-N'-(beta-d-glucopyranosyl)ureas as inhibitors of glycogen phosphorylase: Synthesis and evaluation by kinetic, crystallographic, and molecular modelling methods

Our reference: BMC 9770 P-authorquery-v10 AUTHOR QUERY FORM Journal: BMC Article Number: 9770 Please e-mail or fax your responses and any corrections to: E-mail: [email protected] Fax: +31 2048 52799 Dear Author, Please check your proof carefully and mark all corrections at the appropriate place in the proof (e.g., by using on-screen annotation in the PDF file) or compile them in a separate list. Note: if you opt to annotate the file with software other than Adobe Reader then please also highlight the appropriate place in the PDF file. To ensure fast publication of your paper please return your corrections within 48 hours. For correction or revision of any artwork, please consult http://www.elsevier.com/artworkinstructions. Any queries or remarks that have arisen during the processing of your manuscript are listed below and highlighted by flags in the proof. Click on the ‘Q’ link to go to the location in the proof. Location in article Query / Remark: click on the Q link to go Please insert your reply or correction at the corresponding line in the proof Q1 Please confirm that given names and surnames have been identified correctly. Q2 Please check all affiliations, and correct if necessary. Q3 Please check the telecommunication data for corresponding authors. Q4 Please check the edit made in legend of Figure 3. Thank you for your assistance. Graphical abstract pp xxx–xxxN-(4-Substituted-benzoyl)-N0-(bDD -glucopyranosyl)ureas as inhibitors of glycogen phosphorylase: Synthesis and evaluation by kinetic, crystallographic, and molecular modelling methods Veronika Nagy, Nóra Felföldi, Bálint Kónya, Jean-Pierre Praly, Tibor Docsa, Pál Gergely, Evangelia D. Chrysina * , Costas Tiraidis, Magda N. Kosmopoulou, Kyra-Melinda Alexacou, Maria Konstantakaki, Demetres D. Leonidas, Spyros E. Zographos, Nikos G. Oikonomakos, Stanislav Kozmon, Igor Tvaroška, László Somsák * OH N OH HO HO HO H N O O R R = Me, Ph, Cl, OH, OMe, NO2, NH2, COOH, COOMe Best inhibitor: R = Me K i = 2.3 μM BMC 9770 No. of Pages 1, Model 5G 23 January 2012 1 N-(4-Substituted-benzoyl)-N0-(b-D-glucopyranosyl)ureas as inhibitors of glycogen phosphorylase: Synthesis and evaluation by kinetic, crystallographic, and molecular modelling methods q Veronika Nagy a,b ,Nóra Felföldi a ,Bálint Kónya a ,Jean-Pierre Praly b ,Tibor Docsa c ,Pál Gergely d , Evangelia D. Chrysina e, ⇑ ,Costas Tiraidis e ,Magda N. Kosmopoulou e ,Kyra-Melinda Alexacou e , Maria Konstantakaki e ,Demetres D. Leonidas e, ,Spyros E. Zographos e ,Nikos G. Oikonomakos e,z , Stanislav Kozmon f ,Igor Tvaroška f ,László Somsák a, ⇑ a Department of Organic Chemistry, University of Debrecen, POB 20, H-4010 Debrecen, Hungary 10 b Université Claude-Bernard Lyon 1, ICBMS, UMR UCBL-CNRS-INSA-CPE 5246, CPE-Lyon, 43 Boulevard du 11 Novembre 1918, F-69622 Villeurbanne, France c Cell Biology and Signaling Research Group of The Hungarian Academy of Sciences, University of Debrecen, Egyetem tér 1, H-4032 Debrecen, Hungary d Department of Medical Chemistry, Medical and Health Science Centre, University of Debrecen, Egyetem tér 1, H-4032 Debrecen, Hungary e Institute of Organic and Pharmaceutical Chemistry, The National Hellenic Research Foundation, 48, Vassileos Constantinou Avenue, GR-116 35 Athens, Greece f Institute of Chemistry, Slovak Academy of Sciences, Dubravska cesta 9, SK-845 38 Bratislava, Slovakia article info Article history: 20 Received 5 October 2011 Revised 28 December 2011 Accepted 29 December 2011 Available online xxxx Keywords: N-Acyl-N 0 -bD -glucopyranosyl ureas Glycogen phosphorylase Inhibitor X-ray crystallography Molecular docking 30 abstract N-(4-Substituted-benzoyl)-N 0 -b-D-glucopyranosyl ureas (substituents: Me, Ph, Cl, OH, OMe, NO 2 ,NH 2 , COOH, and COOMe) were synthesised by ZnCl 2 catalysed acylation of O-peracetylated b-D-glucopyranosyl urea as well as in reactions of O-peracetylated or O-unprotected glucopyranosylamines and acyl-isocyanates. O-deprotections were carried out by base or acid catalysed transesterifications where necessary. Kinetic studies revealed that most of these compounds were low micromolar inhibitors of rabbit muscle glycogen phosphorylase b(RMGPb). The best inhibitor was the 4-methylbenzoyl compound (K i = 2.3 l M). Crystallographic analyses of complexes of several of the compounds with RMGPbshowed that the analogues exploited, together with water molecules, the available space at the b-pocket subsite and induced a more extended shift of the 280s loop compared to RMGPbin complex with the unsubstituted benzoyl urea. The results suggest the key role of the water molecules in ligand binding and structure-based ligand design. Molecular docking study of selected inhibitors was done to show the ability of the binding affinity prediction. The binding affinity of the highest scored docked poses was calculated and correlated with experimentally measured K i values. Results show that correlation is high with the R-squared (R 2 ) coefficient over 0.9. Ó2012 Elsevier Ltd. All rights reserved. 1. Introduction 50 Type 2 diabetes mellitus is currently estimated to affect more than 5% of the adult population in Western societies, and its incidence is expected to increase considerably in the future, in particular owing to the dramatic increase in obesity. The global incidence of type 2 diabetes is projected to afflict more than 300 million people worldwide over the next 20 years, and many of those affected will be young adults. 1 The disease is characterised by hyperglycaemia associated with defective insulin production and hepatic and peripheral insulin resistance. Due to long term complications, it is a major cause of 60 blindness and renal disease, and is known to significantly increase the risk of cardiovascular disorders. 2 Current preventive and therapeutic strategies do not achieve adequate control of blood glucose to prevent chronic morbidity, and are completely ineffective in 40% of all diagnosed cases. 3 As a consequence of this, there is a pressure on the research community (both academic and industrial) to develop novel healthcare interventions to address this substantial biomedical challenge. 4,5 Hepatic glucose output is elevated in type 2 diabetic patients and current evidence indicates that glycogenolysis (release of 70 monomeric glucose from the glycogen polymer storage form) is 0968-0896/$ - see front matter Ó2012 Elsevier Ltd. All rights reserved. doi:10.1016/j.bmc.2011.12.059 q PDB ID Codes: 2QNB, 2QLM, 2QLN, 2QN3, 2QN7, 2QN8, 2QN9. ⇑ Corresponding authors. Tel.: +30 2107273851; fax: +30 2107273831 (E.D.C.); tel.: +36 52512900/22453; fax: +36 52453836 (L.S.). E-mail addresses: [email protected],[email protected] (E.D. Chrysina), somsak@ tigris.unideb.hu (L. Somsák).  Present address: Department of Biochemistry and Biotechnology, University of Thessaly, 26 Ploutonos St., 41221 Larissa, Greece. z Dr. Nikos G. Oikonomakos sadly passed away on the 31st of August 2008 while this manuscript was in preparation. This paper is dedicated to his memory. Q3 Q1 Q2 Bioorganic & Medicinal Chemistry xxx (2012) xxx–xxx Contents lists available at SciVerse ScienceDirect Bioorganic & Medicinal Chemistry journal homepage: www.elsevier.com/locate/bmc BMC 9770 No. of Pages 17, Model 5G 25 January 2012 Please cite this article in press as: Nagy, V.; et al. Bioorg. Med. Chem. (2012), doi:10.1016/j.bmc.2011.12.059 an important contributor to the abnormally high production of glucose by the liver. 2,6,7 Glycogen phosphorylase (GP) is the enzyme responsible for glycogen breakdown to produce glucose and related metabolites for energy supply. 4,8,9 Due to its key role in modulation of glycogen metabolism, pharmacological inhibition of GP has been regarded as an effective therapeutic approach to treating type 2 diabetes. 10,11 Besides the liver tissue GP isoenzymes are also located in muscle and brain tissue with increased overall homology and high levels of identity at the catalytic site. Hence, 80 series of studies have been underway, employing both the liver and the muscle isoenzyme for the design of new antidiabetic agents. Several classes of compounds have been reported and extensively surveyed 3,10,12,13 for the inhibition of and in complex with liver and rabbit muscle GP (RMGP). Distinct binding sites identified in GP 14 include the catalytic site, the purine inhibitory site (also known as I-site), the allosteric site, the glycogen storage site, a novel allosteric inhibitor site and the newly discovered benzimidazole-binding site. The quest for compounds with higher potency than glucose to inhibit GP activity 90 led to the design of various glucose-based analogues. For this purpose the properties of the catalytic site of GP in the Tstate (inactive) conformation were investigated. Specifically, the crystal structure of rabbit muscle enzyme revealed that the catalytic site is a long channel secluded from the bulk solvent. The entrance to this channel is constrained by a flexible loop (280s) comprising residues 282–286. During allosteric transition the 280s loop changes conformation and acts as a tollgate allowing access of the substrate to the core of the molecule. Dissection of the catalytic site in the presence of a collection of b-substituents of D -glucose unveiled two sub-sites 100 of the catalytic channel, previously filled with water molecules in the native enzyme. These sites are lined by residues of mixed ionic/hydrophobic character. 10 Glucose analogues, binding primarily to the catalytic site, represent the most populated family of GP inhibitors 3,15 comprising among others various derivatives of Nacyl-bD -glucopyranosylamines, glucopyranosylidene-spiro-heterocycles, Nand C-bD -glucopyranosyl heterocycles as well as bD -glucopyranosyl thiosemicarbazone derivatives. A selection of N-acyl-bD -glucopyranosylamine type compounds collected in Chart 1 indicate that increasing hydrophobicity 110 of the acyl moiety makes the inhibition stronger (among 1to 8,4is the best inhibitor) while the a - D -anomer 9is ineffective. Phosphorylated bD -glucopyranosylamine 10 has a very weak effect. Spirocyclization with the proper ‘anomeric’ configuration (11 and 12 vs 13) makes very good inhibitors. Attachment of a further amide moiety to the first one (14,15) with a hydrophobic part results in inhibitor 15 whose efficiency is comparable to that of the spiro-hydantoins 11 and 12. Crystallographic analyses of complexes of the above compounds with rabbit muscle glycogen phosphorylase b(RMGPb) demonstrated that in most cases a H-bridge 120 exists between the bD -glucopyranosylamine NH and main-chain carbonyl of His377 next to the active site (for illustration and references see Chart 1). This hydrogen bond in 1–4,6,7, and 10– 12 makes an important contribution to the binding. Consequently, a - D -configurated derivatives 9and 13 are much weaker inhibitors. This hydrogen bridge is absent in the complexes of compounds 5, 8, and 14 which again show poor binding. In the case of benzoyl urea 15 this H-bridge has not been observed in the crystal 16 but the binding is stronger than in any other N-acyl-bD -glucopyranosylamine derivative. This points to the role of interactions of the 130 inhibitor molecule in the b-channel of the enzyme. The aim of the present study is to investigate interactions of Nbenzoyl-N 0 -bD -glucopyranosyl urea derivatives in the b-channel. To this end 15 was modified by placing neutral apolar and polar, as well as acidic and basic substituents in position 4 of the phenyl ring. Experimentally determined and computed inhibition constants were compared to evaluate the predictivepower of molecular docking methods applied to the catalytic site of GP, while crystallographic studies of the enzyme-inhibitor complexes allowed to assess binding peculiarities of the molecules. 140 2. Results & discussion 2.1. Synthesis Although there are many examples of N-substituted-N 0 -glycosyl urea derivatives in the literature, at the outset of this work we could find only two structures for N-acyl-N 0 -glycosyl ureas: perO-acetylated N-acetyland N-benzoyl-N 0 -(bD -glucopyranosyl)urea 18 (per-O-acetyl protected 14 and 15, respectively). These derivatives were made by exhaustive acetylation of bD -glucopyranosyl urea by Ac 2 O/ZnCl 2 , and N-benzoylation of 2,3,4,6-tetra-Oacetyl-bD -glucopyranosyl urea by BzCl/Py, respectively. For the 150 synthesis of the planned new derivatives disconnections Aand B (Scheme 1) were envisaged requiring investigation of reactions between bD -glucopyranosylamines and acyl-isocyanates (A) and acylations of bD -glucopyranosyl urea (B). Intermediates for the preparation of the target compounds (Scheme 2) were obtained from 2,3,4,6-tetra-O-acetyl-bD -glucopyranosyl azide (17). Per-O-acetylated bD -glucopyranosylamine 18 was prepared by Raney-nickel reduction of 17. The protected bD -glucopyranosyl urea 16 was synthesized by a slight modification of a published protocol. N-Acylation of 16 with acid chlorides 160 catalysed by ZnCl 2 in CHCl 3 gave compounds 20–24 in acceptable yields in most cases. For details of these transformations and referencing, please, consult Supplementary data. The reaction of glucosylamine 18 with in situ prepared acyl-isocyanates 19 was probed next. In these reactions the suggested solvent CH 3 CN had to be dried with extreme care because even traces of moisture prevented the formation of the acyl-isocyanate. The yields of these transformations were also good except that for 23. It can be concluded that for the preparation of the per-O-acetylated target compounds 19–24 N-acylation of urea 16 appears 170 superior as compared to the other route taking into account the number and simplicity of the necessary manipulations. Removal of the protecting groups was performed by the Zemplén method to get 15,27, and 28. These reactions had to be closely controlled by TLC in order to avoid loss of the N-acyl group after longer reaction times (cf. Scheme 2). This could not be achieved during Zemplén deprotection of 20,21, and 24, therefore, the respective 25,26, and 29 were obtained by mild acid catalyzed transesterification. These problems associated with the deprotection were circumvented by elaborating a procedure to convert 180 unprotected glucopyranosylammonium carbamate 20 (35) and acyl-isocyanates to N-acyl-N 0 -bD -glucopyranosyl ureas, and for the preparation of 31–33 these conditions were used. Catalytic reduction of 29 furnished amino derivative 30, while that of 33 gave carboxylic acid 34. Structure elucidation of the new compounds was straightforward by NMR methods and needs no detailed comments (see Supplementary data for data). Carbonyl resonances of the acyl urea moieties appeared between 149–157 ppm (–NHCONH–) and 162–171 ppm (–NHCOAr). 190 2.2. Enzyme kinetics and X-ray crystallographic results The inhibitory effect of ten new N-bD -glucopyranosyl urea analogues 25–34 (Table 1) on the enzyme activity was evaluated by kinetic assays performed in the direction of glycogen synthesis at pH 6.8 and 30 °C. The results showed that all compounds are competitive inhibitors of the enzyme activity with K i values in the low l M range (Table 1). Among the substituents introduced in the paraposition of the phenyl ring only the –CH 3 (compound 25) appeared 2V. Nagy et al. / Bioorg. Med. Chem. xxx (2012) xxx–xxx BMC 9770 No. of Pages 17, Model 5G 25 January 2012 Please cite this article in press as: Nagy, V.; et al. Bioorg. Med. Chem. (2012), doi:10.1016/j.bmc.2011.12.059 to bring about slightly improved affinity (K i value of 2.3 l M) compared to the lead compound 15. 200 With the aim to interpret the results obtained from the kinetic experiments and provide rationalizations of the binding affinities, structural studies of RMGPbin complex with six new compounds 25–30 were performed at high resolution (Table 1). All derivatives of 15 bound at the catalytic site, as it was clearly indicated by the 2F o F c and F o F c electron density maps (Fig. 1a–f). Additional electron density was observed at the new allosteric site for compound 29 (Fig. 1g), suggesting weak binding. All six inhibitors induced extended conformational changes mainly in the 280s loop upon binding, similar, not identical though, to those observed in 210 the RMGPb:15 complex. 16 More detailed analysis of the complex structures was performed by mapping the key hydrogen bond and van der Waals interactions formed with residues lining the catalytic site (Tables S2b and S2c). The interactions formed by the peripheral hydroxyl groups of the glucopyranose moiety at the catalytic site are maintained in almost all complexes. Therefore, the analyses focused on the alterations in the network of contacts that occur when a new neutral, apolar and polar or slightly acidic and basic functional group was introduced to the parent molecule (at position 4 of the phenyl ring). 220 2.2.1. Compound 15 The crystal structure of RMGPb–15 complex (Bzurea) was previously determined to 1.8 Å resolution 16 ; the structural results showed that upon binding of 15 at the catalytic site, there is a significant rearrangement of the 280s loop. In specific, Asn284 is sandwiched between the side chains of residues Phe285 and Tyr613, resulting in the disruption of the inhibitor site located at the entrance of the catalytic channel. The dramatic shift observed in the 280s loop (shifts 1.3 to 3.7 Å of C a atoms compared to N–H···O=C H-bridge Ki [μM] + 1 R = CH3 32 + 2 R = CF3 75 + 3 R = C6H5 81 + 4 R = C10H7 (2-naphthyl) 10 – 5 R = NH2 140 + 6 R = CH2N3 49 + 7 R = CH2CH2CO2H 20 OH N OH HO HO HO R O – 8 R = COOCH3 210 ON OH HO HO O R O N H N HN O H H His-377 H-bridges between His-377 and N-acyl-β-D-glucopyranosylamine type inhibitors in the catalytic site of GP O HO HO HO HO HN O CF3 – 9 non inhibitory OH N OH HO HO HO P OCH3 O OCH3 + 10 5900 N–H···O=C H-bridge Ki [μM] + 11 X = O 3.1 O HO HO HO HO N H H N X O+ 12 X = S 5.1 OH N OH HO HO HO NH O CH3 O – 14 370 O HO HO HO HO HN NH O O – 29 13 320 105 OH N OH HO HO HO H N O O – 15 4.6 Chart 1. Selected N-acylD -glucopyranosylamine type inhibitors of GP. 15,17 OH N OR RO RO RO H N O Ar O ONH2 OR CN O Ar O OH N OR NH2 O Ar O X ++ AB AB Scheme 1. Retrosynthetic disconnections of N-acyl-N 0 -(bD -glucopyranosyl) urea. V. Nagy et al. /Bioorg. Med. Chem. xxx (2012) xxx–xxx 3 BMC 9770 No. of Pages 17, Model 5G 25 January 2012 Please cite this article in press as: Nagy, V.; et al. Bioorg. Med. Chem. (2012), doi:10.1016/j.bmc.2011.12.059 the native structure) was attributed to the bulky benzoyl moiety; 230 however, these shifts were not in the direction of Tto Rallosteric conversion. The conformational changes resulted in increased contacts between the inhibitor and the protein, providing a rationalization for the K i with a value of 4.6 l M. However, the ‘characteristic’ hydrogen bond between amide nitrogen N1 and the backbone O of His377 was disrupted in RMGPb–15 complex. 16 This hydrogen bond was present in most RMGPbstructures determined in complex with bD -glucopyranosylamine analogues illustrated in Chart 1. 2.2.2. Compound 25 240 Introduction of a methyl (–CH 3 ) group in the para-position of the phenyl ring resulted in an improved inhibitor with a K i value of 2.3 l M compared to the lead compound 15. Structural studies of RMGPb–25 complex showed that 25 bound at the catalytic site and formed a total of 18 hydrogen bonds and 94 van der Waals interactions (slightly reduced compared to 15, 20 and 103, respectively) (Tables S2b and S2c and Fig. 2a and b). The aforementioned ‘characteristic’ hydrogen bond between N1 and the main chain O of His377 was disrupted in accordance with the results observed forBzurea. Thesubstituted phenylring pointedat asimilar direction 250 tothat of15 with anegligible incline(atomsshifted by0.4 to0.7 Å) towards Ala383 (Fig. 3a). The solvent structure was comparable in both complex structures except for two water molecules, Wat161 O and Wat235 O (numbering from RMGPb:15 complex structure) that were displaced to avoid clashes with C15 of the methyl group. A few minor shifts were also recorded in the solvent lying in the vicinity of the hydrophobic –CH 3 group, that is Wat84 O shifts by 0.7 Å, Wat215 O by 0.6 Å and Wat209 O by 0.5 Å (Wat93 O, Wat251 O, Wat183 O numbering from RMGPb–15 complex, respectively). The rearrangement of the solvent structure induces more 260 profound changes in the 280s loop compared to those observed in the Bzurea complex. In particular, the side chain of Asn282 became less stable mainly due to the displacement of Wat235 O in the presence of the –CH 3 group and changed conformation by rotation of its dihedrals ( v 1 , v 2 )by(120°,33°), respectively. As a result the side chain of Asn284 is subjected to a minor shift in all atoms by 0.5 Å. Similar move (by 0.5 Å in all atoms) was recorded for Phe285 since the hydrogen bond formed between the backbone oxygen and Asn282 ND2 was disrupted. Analogous changes were induced to the side-chain atoms of Phe286 (atoms shifted by 0.5 to 0.7 Å and 0w rotated by 10°). However, the most profound modifications, possibly interrelated to the new position of Asn282, were observed in Glu287 that adopted a different conformation (rotation of dihedral angles ( v 1 , v 2 , v 3 )by(55°,18°,170°), respectively) compared to RMGPb:15 complex structure. The peptide bond between Glu287 and Gly288 also flipped to restore stereochemistry. ON 3 OAc AcO AcO Ac O OH N OAc AcO A cO AcO OH N OH HO HO HO H N O O 16 19-24 15, 25-34 b d H N O ONH 2 OAc AcO AcO Ac O a 17 c R OH N OAc AcO AcO Ac O 18 NH 2 O O R ONH 2 .HO 2 CNH 2 OH HO HO HO e or f or g i 35 Conditions Yield (%) R R Reaction time Yield (%) c 50 H 19 e 15 H 2 h 90 c 34 CH320 f 25 CH3 12 d 54 c 43 C6H521 f 26 C 6H5 6 d 86 d 67 Cl 22 e 27 Cl 1 h 86 c d 72 20 OAc 23 e 28 OH 0.5 h 87 c d 60 65 NO224 g 29 NO2 5 d 99 29 h 30 NH2 1 h 99 35 i 31 OCH3 2 d 50 35 i 32 COOMe 5 d 39 35 i 33 COOBn 4 d 43 33 j 34 COOH 3 h 73 Scheme 2. Reagents and conditions: (a) PPh 3 , EtOAc, NH 3 ,CO 2 , rt, 88%; (b) H 2 , Raney-Ni, EtOAc, rt, 71%; (c) 4-R-C 6 H 4 –COCl, ZnCl 2 , CHCl 3 , reflux; (d) 4-R-C 6 H 4 –CONCO, CH 3 CN, Ar, rt; (e) NaOMe, MeOH, rt; (f) AcCl, MeOH, rt; (g) KHSO 4 , MeOH, rt; (h) H 2 , Raney-Ni, MeOH, rt; (i) 4-R-C 6 H 4 –CONH 2 converted to 4-R-C 6 H 4 –CONCO (by (COCl) 2 in ClCH 2 CH 2 Cl at reflux temp., then reaction with 35 in Py, rt; (j) H 2 , Pd/C, MeOH, reflux. 4V. Nagy et al. / Bioorg. Med. Chem. xxx (2012) xxx–xxx BMC 9770 No. of Pages 17, Model 5G 25 January 2012 Please cite this article in press as: Nagy, V.; et al. Bioorg. Med. Chem. (2012), doi:10.1016/j.bmc.2011.12.059 Although binding of 25 seemed to provoke more profound changes in the 280s loop than those of 15, and some perturbation in the solvent structure, introduction of –CH 3 appeared to be overall energetically favourable since it outweighed those changes and resulted 280 in a better inhibitor than 15. 2.2.3. Compound 26 Replacement of the methyl group by another hydrophobic but much bulkier substituent (–C 6 H 5 ) in the para-position of the phenyl ring was also investigated. The new compound exhibited similar affinity to 25 (K i value of 3.4 l M) and bound at the catalytic site with the phenyl ring been subjected to minor incline (atoms shifted by 0.3 to 0.5 Å) compared to the corresponding one in 15 complex (Fig. 3b) towards Asn133 and Glu88 (unlike RMGPb:25 that moves slightly towards Ala383). The new aromatic ring intro290 duced was rotated by 38°to optimize its interactions with residues in the vicinity Asn282, Phe286, Arg292, and His341. The (–C 6 H 5 ) group occupied the subsite of the b-pocket at the catalytic channel 21 displacing three water molecules Wat84 O, Wat161 O and Wat215 O upon binding to avoid clashes with C17 atom of the second phenyl ring. These two waters were hydrogen bonded to Wat235 O, also absent in the new complex structure that was in turn interacting with Asn282 ND2. Disruption of this water-mediated network destabilized the side chain of Asn282 that adopted a different conformation by rotation of its dihedrals 300 ( v 1 , v 2 )by(130°,35°) similar to that in RMGPb:25. The alteration of Asn282 side chain resulted in shifts in the range of 0.3 to 0.7 Å in both Asn284 and Phe285 side chain atoms and more pronounced modifications in Glu287. The peptide bond formed between Glu287 and Gly288 flipped as in RMGPb:26 complex and the side chain of the former also changed (rotation of dihedral angle v 3 by 61°). Introduction of such a bulky substituent affects also the side chain of Arg292 the atoms of which are slightly shifted away from the ligand (by 0.5 Å). Overall, 26 upon binding maintains the number of hydrogen bond interactions 310 formed compared with the methyl compound 25; however, the increased number of van der Waals interactions (113) did not seem to promote ligand binding (Fig. 2c, Tables S2b and S2c). The preference though of the b-pocket subsite for more hydrophobic and the space availability for even bulkier substituents became evident. 2.2.4. Compound 27 To further investigate the type of interactions formed at the catalytic site a chlorine atom (–Cl) was placed in position 4 of the phenyl ring. Kinetic results indicated that affinity of the new 320 analogue was equivalent to that of the parent molecule (K i value of 4.4 l M). Analysis of the RMGPb:27 crystal structure showed that it bound at the catalytic site in a similar fashion to the previous analogues (Fig. 3c) making a total of 19 hydrogen bond and 91 Van der Waals interactions (Fig. 2d, Tables S2b and S2c). The phenyl ring was Table 1 Kinetic data obtained with rabbit muscle GPband crystallographic numbering of the compounds O HO HO OH OH O5 O2 O6 O3 O4 C5 C3 C2 C6 C4 H N O H N O R C8 O8O7 N2 C7 N1 Compound R K i ( l M) Compound R K i ( l M) 15 C10 C11 C12 C13 C14 C9 4.6 ± 0.90 16 30 NH 2 C13 C12 C11 C10 C14 C9 N3 6.0 ± 0.60 25 CH3 C11 C12 C13 C14 C10 C9 C15 2.3 ± 0.03 a 31 OCH3 3.2 ± 0.16 26 C17 C18 C19 C20 C16 C15 C11 C10 C13 C14 C9 C12 3.7 ± 0.06 32 COOCH 3 4.0 ± 0.16 27 Cl C11 C12 C13 C14 C9 C10 4.4 ± 0.09 a 33 COOCH 2 Ph 150 ± 12 a 28 OH C13 C12 C11 C10 C9 C14 O12 6.3 ± 0.30 34 COOH 85 ± 6.5 a 29 N C11 C10 C13 C14 C9 O O O9 O10 C12 N3 3.3 ± 0.30 a K i values for these compounds were calculated for comparison purposes by the Cheng–Prusoff equation: K i =IC 50 /(1 + [S]/K m ). V. Nagy et al. /Bioorg. Med. Chem. xxx (2012) xxx–xxx 5 BMC 9770 No. of Pages 17, Model 5G 25 January 2012 Please cite this article in press as: Nagy, V.; et al. Bioorg. Med. Chem. (2012), doi:10.1016/j.bmc.2011.12.059 almost coplanar with the one in 15, and only slight shift (by 0.3 to 0.5 Å) of the carbon atoms towards Ala383 was observed compared to 15 when bound to RMGPb(Fig. 3c). Two water molecules are displaced upon binding of 27 at the catalytic site; Wat161 O to Figure 1. Schematic representation of the 2F o F c electron density maps contoured at 1.0 r level of the refined Bzurea analogues bound at the catalytic site of RMGPb(a, b, c, d, e, f for compounds 25,26,27,28,29,30) and the new allosteric site (g, compound 29). 6V. Nagy et al. / Bioorg. Med. Chem. xxx (2012) xxx–xxx BMC 9770 No. of Pages 17, Model 5G 25 January 2012 Please cite this article in press as: Nagy, V.; et al. Bioorg. Med. Chem. (2012), doi:10.1016/j.bmc.2011.12.059 330 avoid clashes with the chlorine atom and its neighboring Wat235 O that was not tightly bound at the catalytic site any more also due to the conformational change of Asn282 side chain (dihedral angles v 1 , v 2 rotate by 120°,40°, respectively) that resulted in the disruption of the second hydrogen bond interaction it formed with ND2 atom of the same residue. Additional alterations observed in the solvent structure involved shifts of Wat215 O (Wat257 O numbering from GPb:25 complex), Wat84 O (Wat87 O), Wat236 O (Wat186 O), Wat148 O (Wat270 O) by 0.8, 0.7, 0.3, 0.9 Å, respectively. Overall, residues of the 280s loop follow 340 the same rearrangement observed for in the RMGPb:15 crystal structure except for Asn282. Upon ligand binding, residues Asn284, Phe285, Phe286 are subjected to minor disturbance (in the range of 0.3 to 0.7 Å). More extensive changes are recorded for Glu287 the peptide bond of which with Gly288 flipped in this complex structure as well. Despite the increased perturbation of Figure 2. Stereo representation of the molecular interactions of compounds 15 (a), 25 (b) 26 (c), 27 (d), 28 (e), 29 (f), 30 (g) when bound at the catalytic site of RMGPband protein atoms in the vicinity. Water molecules are labeled as w. Emphasis is given in the new substituent introduced in the para-position of the phenyl ring of 15. V. Nagy et al. /Bioorg. Med. Chem. xxx (2012) xxx–xxx 7 BMC 9770 No. of Pages 17, Model 5G 25 January 2012 Please cite this article in press as: Nagy, V.; et al. Bioorg. Med. Chem. (2012), doi:10.1016/j.bmc.2011.12.059 we have no rational for such a large error. However, we cannot ex590 clude a possibility of a different binding mode for these inhibitors, caused by the so called 280s loop (280–289 amino acid residues) flexibility. Also it is known that lipophilic and stacking interactions of the methylphenyl group are usually poorly described by scoring functions. We also note that our case is not unique, e. g. swainsonine that is most active inhibitor of mannosidase II was predicted to be one of the weakest inhibitors with several scoring functions, though its binding mode was predicted correctly. This shows that there is a room for improvement of scoring functions. 3. Conclusions 600 N-(4-Substituted-benzoyl)-N 0 -bD -glucopyranosyl urea derivatives were prepared by ZnCl 2 catalysed acylation of O-peracetylated bD -glucopyranosyl urea or reacting O-peracetylated bD -glucopyranosylamine with acyl-isocyanates and subsequent deprotection. Some compounds were prepared, to avoid protecting group manipulations, in reactions of bD -glucopyranosylammonium carbamate with acyl-isocyanates. Substituents of neutral polar and apolar as well as acidic and basic character were introduced in the 4-position of the phenyl ring with the aim to exploit the available space at the subsite of the b-pocket of the active site of glycogen phosphorylase 610 taking into account the solvent structure. Most of the new compounds were low micromolar inhibitors of rabbit muscle glycogen phosphorylase b, however, no significant improvement of the inhibitory efficiency as compared to that of N-benzoyl-N 0 -bD -glucopyranosyl urea could be observed. These findings might indicate the lack of a specific and crucial interaction from this position within the catalytic site. Crystallographic analyses of the 3D structures of the new compounds with RMGPbshowed clearly that all analogues induced a more extended shift of the backbone atoms of the 280s loop com620 pared to RMGPb:15. This shift was facilitated by the flipping of the peptide bond formed between Glu287 and Gly288 to create more space at the catalytic site of the enzyme. The most profound differences were observed in the side chain of Asn282 that adopted a modified conformation upon ligand binding in all complexes except for –OH and –NH 2 groups. Overall, the structural results were in accordance with those obtained by the kinetic evaluation of the compounds. The best inhibitor identified was compound 25 (4-CH 3 group) that formed increased van der Waals interactions with the residues lining the catalytic site. The second best inhibi630 tor, out of those studied by X-ray crystallography, was compound 29 (4-NO 2 group) that stabilized the closed conformation of the 280s loop and as well as residues at the far end of the b-pocket such Arg292 through an extended network of water-mediated interactions. Compound 29 formed the largest number of hydrogen bonds at the catalytic site compared to the rest of the compounds. The results obtained suggest the key role of the solvent structure in structure-based ligand design. This could be exploited for the design of more potent inhibitors of enzyme activity coupling the knowledge derived from each complex structure with geometric 640 algorithms. 25 The compounds 15,25–32 were docked into the Bzurea and 2-Nap-urea crystal structure of the glycogen phosphorylase, respectively. The binding affinity of the highest ranked docked poses was predicted by the LIAISON program (LiaScore) and correlated with experimentally measured K i values of the studied inhibitors. The quality of the correlation was assessed using the Rsquared (R 2 ) coefficient. The value of the R 2 ranges from 0.1080 up to 0.9180. Generally, a better correlation was found for the structures docked into the 2-Nap-urea structure. The best correla650 tion was achieved for the correlation where two outliners, 25 and 32, were excluded from the linear fit, with linear correlation coefficients A and B (A + B⁄x) having values of 23.89 and 7.12. Interestingly, the best measured inhibitor appears as an outliner. This situation might be caused by the different binding mode predicted by the docking and absence of the water molecules present in the active site as have been shown in presented crystal structure. Despite all this, presented linear fit can be used to predict the binding affinity for compounds having the same scaffold structure by the LIAISON program. 660 4. Experimental 4.1. General synthetic methods Melting points were measured in open capillary tubes or on a Kofler hot-stage and are uncorrected. Optical rotations were determined with a Perkin-Elmer 241 polarimeter at room temperature. NMR spectra were recorded with Bruker WP 200 SY (200/50 MHz for 1 H/ 13 C), Bruker AM 360 (360/90 MHz for 1 H/ 13 C) or Bruker AM 400 (400/100 MHz for 1 H/ 13 C) spectrometers. Chemical shifts are referenced to Me 4 Si ( 1 H), or to the residual solvent signals ( 13 C). TLC was performed on DC-Alurolle Kieselgel 60 F 254 (Merck) 670 (eluent EtOAc–hexane 1:2, unless stated otherwise), and the spots were visualized under UV light and by gentle heating. For column chromatography Kieselgel 60 (Merck, particle size 0.063– 0.200 mm) was used. Organic solutions were dried over anhydrous MgSO 4 and concentrated under diminished pressure at 40–50 °C (bath temperature). Acetonitrile was distilled from P 4 O 10 and stored over molecular sieves (3 Å). Dry methanol was distilled from magnesium methylate. Other solvents of commercial analytical grade quality were used without further purification. 4.2. General methods for the preparation of N-acyl-N 0 -(2,3,4,6680 tetra-O-acetyl-bD -glucopyranosyl)ureas 4.2.1. Method A To a solution of an acyl chloride (7.7 mmol) in 20 ml of dry chloroform anhydrous zinc chloride (80 mg, 0.59 mmol) and 2,3,4,6-tetra-O-acetyl-bD -glucopyranosyl urea (16, 1 g, 2.56 mmol) were added with stirring. The reaction mixture was refluxed until TLC showed the complete transformation of 16. Then the reaction mixture was poured into ice-water and was extracted with chloroform (2). The organic phases were collected and washed with satd aq Na 2 CO 3 solution and water. After drying the solvent was evaporated 690 and the residue was purified by column chromatography (eluent: EtOAc–hexane = 1:2). 4.2.2. Method B To a suspension of NaOCN (1.21 g, 18.7 mmol) in dry acetonitrile an acyl chloride (14.4 mmol) in acetonitrile (30 ml) and SnCl 4 (84 l l, 0.72 mmol) were added under argon with stirring. The mixture was stirred under reflux for 8 h and after cooling to rt 2,3,4,6tetra-O-acetyl-bD -glucopyranosyl amine (18, 1 g, 2.88 mmol) was added under argon. After stirring for 30 min some drops of water were added and the mixture was filtered. The solvent was evapo700 rated and the residue purified by column chromatography (eluent: EtOAc–hexane = 1:1). 4.3. General methods for the preparation of N-acyl-N 0 - (bD -glucopyranosyl)ureas 4.3.1. Method C A solution of an N-acyl-N 0 -(2,3,4,6-tetra-O-acetyl-bD -glucopyranosyl)urea in dry methanol was treated with a catalytic amount 14 V. Nagy et al. / Bioorg. Med. Chem. xxx (2012) xxx–xxx BMC 9770 No. of Pages 17, Model 5G 25 January 2012 Please cite this article in press as: Nagy, V.; et al. Bioorg. Med. Chem. (2012), doi:10.1016/j.bmc.2011.12.059 of a methanolic solution of NaOMe at rt. After TLC had shown disappearance of the starting material the reaction mixture was neutralized with a cation exchange resin (Amberlyst 15, H + form). 710 After filtration the solvent was removed and the residue was purified by crystallisation. 4.3.2. Method D A solution of an N-acyl-N 0 -(2,3,4,6-tetra-O-acetyl-bD -glucopyranosyl)urea in dry methanol was treated with a catalytic amount of acetyl chloride at rt. After TLC had shown disappearance of the starting material the reaction mixture was neutralized with solid NaHCO 3 . After filtration the solvent was removed and the residue was purified by column chromatography (eluent: CHCl 3 – MeOH =9:1). 720 4.3.3. Method E Oxalylchloride (1.1 equiv) was added to a suspension of a carboxamide (0.15 mmol) in anhydrous 1,2-dichloroethane (2 mL) and the mixture was heated at reflux temp. for 1 d. The volatiles were distilled off under diminished pressure and toluene (2 5 mL) was evaporated from the residue to remove the rest of oxalylchloride. The acyl-isocyanate obtained in this way was mixed with a solution of bD -glucopyranosylammonium-carbamate 20 (1 equiv) in anhydrous pyridine (100 mg/14 mL) and the mixture was stirred until TLC (CHCl 3 –MeOH =7:1) showed no 730 more change (transformations were incomplete). Pyridine was distilled off under diminished pressure and evaporation of toluene (2 30 mL) removed traces of pyridine. The crude was purified by column chromatography (CHCl 3 –MeOH =7:1). Particular syntheses and compound characterization data are presented in the Supplementary data. 4.4. Enzyme preparation RMGPbwas isolated from rabbit skeletal muscle and purified as described previously. 26 Kinetic studies were performed in the direction of glycogen synthesis in the presence of various concen740 trations of inhibitors as indicated in Table 1. Enzyme activity was measured at pH 6.8 by the release of inorganic phosphate as described previously by Oikonomakos et al. 27 (compounds 27,29 and 30) and Saheki et al. 28 (compounds 21,23 and 25). 4.5. Crystal complex formation and X-ray crystallographic data collection and processing Native T-state GPbcrystals were grown in the tetragonal lattice, space group P4 3 2 1 2 29 and prior to data collection were soaked in a buffered solution (10 mM Bes, pH 6.7) with 6.2 mM of 21 (for 2 h), 0.76 mM of 23 (in 20% DMSO for 84 h), 1mMof25 (in 2% DMSO for 750 10 h), 10 mM of 29 (for 1 h). Co-crystals of RMGPbcomplexed with 27 and 30 were obtained in a medium consisting of 20 mg/ml enzyme, 1 mM spermine, 3 mM DTT, 10 mM BES, 0.1 mM EDTA, 0.02% sodium azide, pH 6.7 (16 °C) with either 5 mM of 27 or 7mMof30. Diffraction data for all complexes were collected from single crystals at room temperature, using synchrotron radiation source at EMBL-Hamburg outstation, Germany, beamlines X31 (k= 0.8123 Å), X11 (k= 0.813 or 0.8468 Å), BW7A (k= 0.9076 Å) and BW7B (k= 0.8441 Å), and SRS-Daresbury Laboratory, beamline PX9.6 (k= 0.92 Å). Integration of the reflections and data reduction 760 was performed using the programs DENZO and SCALEPACK from HKL-package. 30 4.6. Crystal structure determination The structure of RMGPbin complex with 15, previously determined at 1.8 Å resolution 16 with XPLOR (pdb code 1K06) was refined using REFMAC. 31 The results showed that residues lining the 280s loop exhibited differences in their side chain orientations with the most profound ones localized in residues Asp283 ( u , w , v 1 , v 2 change by 14°,51°, 166°,20°, respectively) and Glu287 ( u , w , v 3 ) change by 22°,35°, 156°, respectively). The new model 770 deposited with the protein data bank (pdb code 2QNB), as replacement of 1K06, was used as a starting model for the structure determination of all six complexes. Crystallographic refinement was carried out using a standard protocol as implemented by REFMAC. 31 2F o F c and F o F c electron density maps calculated were visualized using the program for molecular graphics ‘O’. 32 Ligand models were fitted to the electron density maps after adjustment of their torsion angles. Alternate cycles of manual rebuilding with ‘O’ and refinement with REFMAC improved the quality of the models. The data collection and refinement statistics along with the 780 model quality are summarized in Supplementary data Table S2a. The stereochemistry of the protein residues was validated by PROCHECK. 33 The analyses of the complex structures comprised mapping of the hydrogen bond and van der Waals interactions of the analogues with the residues lining the catalytic site using CONTACT 33 applying a distance cut off 3.3 and 4.0 Å between the electronegative atoms, respectively. The program calculates the angle OHN(where the hydrogen position is unambiguous) and the angle sourceoxygen-bonded carbon atom. Suitable values are 120°and 90°. The network of interactions is described in 790 Tables S2b and S2c. Structural comparisons were performed with the program ‘O’ 32 by superposition of the atomic coordinates of the new complexes with those of the RMGPb:Bzurea structure. The root mean square deviation in C a positions were determined for residues (24–249), (261–281), (289–313), (326–549) and (558–830) using LSQKAB. 33 All figures were prepared with the programs MolScript 34 and BobScript 35 and rendered with Raster3D. 36 The coordinates of the new structures have been deposited with the RCSB Protein Data Bank (http://www.rcsb.org/pdb) with codes: 800 2QNB-15, 2QLM-25, 2QLN-26, 2QN3-27, 2QN7-28, 2QN8-29 and 2QN9-30. 4.7. Preparation of protein and ligand structures for docking The initial structures of the glycogen phosphorylase in a complex with pyridoxal-5 0 -phosphate and N-benzoyl-N 0 -bD -glucopyranosyl urea 15 and N-(2-naphthoyl)-N 0 -bD -glucopyranosyl urea (2-Nap-urea) were obtained from the PDB database under the codes Bzurea and 2-Nap-urea, respectively, and prepared using Schrodinger’s Maestro and Protein Preparation Wizard 37 as follows. Water molecules were removed, hydrogen atoms were added 810 and protonation states were assigned based on a residue pK a ’s at their normal pH (7.0). Atom types and partial charges were assigned according to the OPLS_2001 force field, also known as OPLS-AA. 38 Docking grids were prepared for each of the prepared protein structures, Bzurea and 2-Nap-urea, using the same procedure. The centers of the cubic grid boxes were placed on the centroid of the bound ligands and the box sizes were set to 14 Å in all three dimensions. Nine inhibitors with available experimental K i data were se820 lected for this study. The structures of all inhibitors were obtained through the energy minimization using Jaguar v7.0 program included in the Schrodinger software 37 at the DFT B3LYP 39,40 level with the 6-31+G⁄basis set 41–46 prior to the docking. The calculated ESP charges were used as input partial charges for ligand atoms in the docking calculations. Docking was carried out using the Glide v4.5 package of Schrodinger suite 2007. 37 Recent review of docking programs showed 47 that Glide belongs amongst the most accurate docking programs. V. Nagy et al. /Bioorg. Med. Chem. xxx (2012) xxx–xxx 15 BMC 9770 No. of Pages 17, Model 5G 25 January 2012 Please cite this article in press as: Nagy, V.; et al. Bioorg. Med. Chem. (2012), doi:10.1016/j.bmc.2011.12.059 We have used flexible ligand docking with the standard precision 830 (SP) algorithm. 48 The parameters for van der Waals radii were scaled by 0.80 for ligand atoms with partial atomic charge less then 0.15. Ligand poses were clustered with RMSDs less than 0.5 Å and within maximum atomic displacement less than 1.3 Å. After the docking procedure 20 poses of each docked ligand with the best GlideScore 49 were saved and used for the analysis. Binding affinities were recalculated for all saved docking poses employing the program Liaison included in the Schrodinger software. 37 The structures of the complexes using the OPLS2005 force field 50 were optimized and the LiaScore’s (‘GlideScore in Liaison’) 840 analyzed for the relaxed complexes. Enzyme ligand complexes were optimized with a restrained mobility of receptor residues at distance larger than 12 Å and with frozen residues at distance larger than 16 Å from ligand, respectively. The same DFT calculated partial atomic charges were used for the ligand as for the docking. Implicit solvent continuum model have been used in computation to model solvent effects. The calculated LiaScore for the best docked pose were correlated with the experimentally measured lnK i values. Acknowledgments 850 This work was supported by the Hungarian Scientific Research Fund (OTKA T37210, T46081, CK77712) as well as by the TÁMOP 4.2.1/B-09/1/KONV-2010007 project co-financed by the European Union and the European Social Fund. Financial aid to the collaboration between Athens and Debrecen (GR-4/03) as well as Lyon and Debrecen (F-11/05) was provided by GSRT (Greece) as well as EGIDE (France), respectively, and the Agency for Research Fund Management and Research Exploitation (KPI, Hungary), and also by CNRS (France) and the Hungarian Academy of Sciences (PICS 4576). VN thanks the French Embassy in Budapest for supporting 860 her co-tutored PhD Thesis prepared in Lyon and Debrecen. Additional support for this work was provided by Greek GSRT through ENTER-EP6/2001, PENED-204/2001, the Joint Research and Technology Projects between Greece and Hungary (2005–2007), Marie Curie Host Fellowship for the Transfer of Knowledge (ToK) contact no MTKD-CT-2006-042776 under FP6; the FP7 Capacities coordination and support actions REGPOT-2008-1-No 230146 ‘EUROSTRUCT’ and REGPOT-2009-1-No 245866 ‘ARCADE’; The research leading to these results has received funding from the European Community’s Sixth (RII3/CT/2004/5060008, IHPP HPRI-CT-1999870 00012) and Seventh Framework Programme (FP7/2007-2013) under grant agreement No. 226716 awarded to both SRS-Daresbury Laboratory and EMBL-Hamburg outstation. We would also like to thank Agence Nationale de la Recherche for funding ANR-08BLAN-0305 ‘GPdia’ project coordinated by JPP for supporting our collaborative research network and acknowledge the assistance of the staff at EMBL beamlines X31, X13, BW7A, BW7B, at the DORIS storage ring in Hamburg and at SRS-Daresbury beamline PX9.6 for providing excellent facilities for X-ray data collection. Supplementary data 880 Supplementary data (details of syntheses and crystallographic analyses) associated with this article can be found, in the online version, at doi:10.1016/j.bmc.2011.12.059. References and notes 1. Alberti, G.; Zimmet, P.; Shaw, J.; Bloomgarden, Z.; Kaufman, F.; Silink, M. Diabetes Care 2004,27, 1798. 2. Moller, D. E. Nature 2001,414, 821. 3. Praly, J. P.; Vidal, S. Mini-Rev. Med. Chem. 2010,10, 1102. 4. Ross, S. A.; Gulve, E. A.; Wang, M. H. Chem. Rev. 2004,104, 1255. 5. Gershell, L. Nat. Rev. Drug Disc. 2005,4, 367. 8906. Radziuk, J.; Pye, S. Diabetes Metab. Res. Rev. 2001,17, 250. 7. Staehr,P.;Hother-Nielsen,O.;Beck-Nielsen,H.Diabetes Obes. Metab. 2002,4,215. 8. Kurukulasuriya, R.; Link, J. T.; Madar, D. J.; Pei, Z.; Richards, S. J.; Rohde, J. J.; Souers, A. J.; Szczepankiewicz, B. G. Curr. Med. Chem. 2003,10, 123. 9. Rotella, D. P. J. Med. Chem. 2004,47, 4111. 10. Oikonomakos, N. G. Curr. Protein Pept. Sci. 2002,3, 561. 11. Baker, D. J.; Greenhaff, P. L.; Timmons, J. A. Expert Opin. Ther. Patents 2006,16, 459. 12. Gimisis, T. Mini-Rev. Med. Chem. 2010,10, 1127. 13. Loughlin, W. A. Mini-Rev. Med. Chem. 2010,10, 1139. 90014. Chrysina, E. D. Mini-Rev. Med. Chem. 2010,10, 1093. 15. Somsák, L. Compt. Rend. Chim. 2011,14, 211. 16. Oikonomakos, N. G.; Kosmopolou, M.; Zographos, S. E.; Leonidas, D. D.; Somsák, L.; Nagy, V.; Praly, J.-P.; Docsa, T.; Tóth, B.; Gergely, P. Eur. J. Biochem. 2002,269, 1684. 17. Somsák, L.; Czifrák, K.; Tóth, M.; Bokor, É.; Chrysina, E. D.; Alexacou, K. M.; Hayes, J. M.; Tiraidis, C.; Lazoura, E.; Leonidas, D. D.; Zographos, S. E.; Oikonomakos, N. G. Curr. Med. Chem. 2008,15, 2933. 18. Helferich, B.; Kosche, W. Chem. Ber. 1926,59, 69. 19. Deng, M. Z.; Caubere, P.; Senet, J. P.; Lecolier, S. Tetrahedron 1988,44, 6079. 91020. Likhosherstov, L. M.; Novikova, O. S.; Shibaev, V. N. Dokl. Chem. 2002,383, 89. 21. Gregoriou, M.; Noble, M. E. M.; Watson, K. A.; Garman, E. F.; Krülle, T. M.; Fuente, C.; Fleet, G. W. J.; Oikonomakos, N. G.; Johnson, L. N. Protein Sci. 1998,7, 915. 22. Hayes, J. M.; Leonidas, D. D. Mini-Rev. Med. Chem. 2010,10, 1156. 23. Benltifa, M.; Hayes, J. M.; Vidal, S.; Gueyrard, D.; Goekjian, P. G.; Praly, J.-P.; Kizilis, G.; Tiraidis, C.; Alexacou, K.-M.; Chrysina, E. D.; Zographos, S. E.; Leonidas, D. D.; Archontis, G.; Oikonomakos, N. G. Bioorg. Med. Chem. 2009,17, 7368. 24. Alexacou, K. M.; Hayes, J. M.; Tiraidis, C.; Zographos, S. E.; Leonidas, D. D.; Chrysina, E. D.; Archontis, G.; Oikonomakos, N. G.; Paul, J. V.; Varghese, B.; Loganathan, D. Proteins: Struct. Funct. Bioinform. 2008,71, 1307. 92025. Chrysina, E. D.; Chajistamatiou, A.; Chegkazi, M. Curr. Med. Chem. 2011,18, 2620. 26. Melpidou, A. E.; Oikonomakos, N. G. FEBS Lett. 1983,154, 105. 27. Oikonomakos, N. G.; Kontou, M.; Zographos, S. E.; Watson, K. A.; Johnson, L. N.; Bichard, C. J. F.; Fleet, G. W. J.; Acharya, K. R. Protein Sci. 1995,4, 2469. 28. Saheki, S.; Takeda, A.; Shimazu, T. Anal. Biochem. 1985,148, 277. 29. Oikonomakos, N. G.; Melpidou, A. E.; Johnson, L. N. Biochim. Biophys. Acta 1985, 832, 248. 30. Otwinowski, Z.; Minor, W. Methods Enzymol. 1997,276, 307. 31. Murshudov, G. N.; Vagin, A. A.; Dodson, E. J. Acta Crystallogr., Sect. D 1997,53, 930240. 32. Jones, T. A.; Zou, J. Y.; Cowan, S. W.; Kjeldgaard, M. Acta Crystallogr., Sect. A 1991,47, 110. 33. Acta Crystallogr., Sect. D 1994,50, 760. 34. Kraulis, P. J. Appl. Crystallogr. 1991,24, 946. 35. Esnouf, R. M. J. Mol. Graphics Modell. 1997,15, 132. 36. Merritt, E. A.; Bacon, D. J. Macromol. Crystallogr., Pt B 1997,277, 505. 37. Schrödinger Suite, 2007. Schrödinger, Inc., Portland, OR. 38. Jorgensen, W. L.; Maxwell, D. S.; Tirado-Rives, J. J. Am. Chem. Soc. 1996,118, 11225. 94039. Becke, A. D. J. Chem. Phys. 1993,98, 1372. 40. Lee, C. T.; Yang, W. T.; Parr, R. G. Phys. Rev. B 1988,37, 785. 41. Ditchfield, R.; Hehre, W. J.; Pople, J. A. J. Chem. Phys. 1971,54, 724. 42. Hehre, W. J.; Pople, J. A. J. Chem. Phys. 1972,56, 4233. 43. Binkley, J. S.; Pople, J. A. J. Chem. Phys. 1977,66, 879. 44. Hariharan, P. C.; Pople, J. A. Theor. Chim. Acta 1973,28, 213. 45. Hehre, W. J.; Ditchfield, R.; Pople, J. A. J. Chem. Phys. 1972,56, 2257. 46. Francl, M. M.; Pietro, W. J.; Hehre, W. J.; Binkley, J. S.; Gordon, M. S.; Defrees, D. J.; Pople, J. A. J. Chem. Phys. 1982,77, 3654. 47. Englebienne, P.; Fiaux, H.; Kuntz, D. A.; Corbeil, C. R.; Gerber-Lemaire, S.; Rose, 950D. R.; Moitessier, N. Proteins: Struct. Funct. Bioinform. 2007,69, 160. 48. Friesner, R. A.; Banks, J. L.; Murphy, R. B.; Halgren, T. A.; Klicic, J. J.; Mainz, D. T.; Repasky, M. P.; Knoll, E. H.; Shelley, M.; Perry, J. K.; Shaw, D. E.; Francis, P.; Shenkin, P. S. J. Med. Chem. 2004,47, 1739. 49. Eldridge, M. D.; Murray, C. W.; Auton, T. R.; Paolini, G. V.; Mee, R. P. J. Comput. Aided Mol. Des. 1997,11, 425. 50. Kaminski, G. A.; Friesner, R. A.; Tirado-Rives, J.; Jorgensen, W. L. J. Phys. Chem. B 2001,105, 6474. 16 V. Nagy et al. / Bioorg. Med. Chem. xxx (2012) xxx–xxx BMC 9770 No. of Pages 17, Model 5G 25 January 2012 Please cite this article in press as: Nagy, V.; et al. Bioorg. Med. Chem. (2012), doi:10.1016/j.bmc.2011.12.059