Polyoxometalates--potent and selective ecto-nucleotidase inhibitors.

G Model

BCP-12133; No. of Pages 11 Biochemical Pharmacology xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Biochemical Pharmacology journal homepage: www.elsevier.com/locate/biochempharm

Polyoxometalates—Potent and selective ecto-nucleotidase inhibitors Sang-Yong Lee a,1, Amelie Fiene a,1, Wenjin Li a, Theodor Hanck a, Konstantin A. Brylev b,c, Vladimir E. Fedorov b,c, Joanna Lecka d,e, Ali Haider f, Hans-Ju¨rgen Pietzsch g, Herbert Zimmermann h, Jean Se´vigny d,e, Ulrich Kortz f, Holger Stephan g, Christa E. Mu¨ller a,* a

PharmaCenter Bonn, Pharmaceutical Institute, Pharmaceutical Chemistry I, University of Bonn, An der Immenburg 4, D-53121 Bonn, Germany Nikolaev Institute of Inorganic Chemistry, Siberian Branch of the Russia Academy of Sciences, 3 Acad. Lavrentiev prospect, 630090 Novosibirsk, Russia Novosibirsk State University, 2 Pirogova Str., 630090 Novosibirsk, Russia d De´partement de Microbiologie-Infectiologie et d’Immunologie, Faculte´ de Me´decine, Universite´ Laval, Que´bec City, QC, Canada G1V 0A6 e Centre de Recherche du CHU de Que´bec, Que´bec City, QC, Canada G1V 4G2 f School of Engineering and Science, Campus Ring 8, Jacobs University, 28759 Bremen, Germany g Institute of Radiopharmaceutical Cancer Research, Helmholtz Zentrum Dresden-Rossendorf, Bautzner Landstrasse 400, 01328 Dresden, Germany h Institute of Cell Biology and Neuroscience, Molecular and Cellular Neurobiology, Goethe University, 60438 Frankfurt am Main, Germany b c

A R T I C L E I N F O

A B S T R A C T

Article history: Received 30 September 2014 Accepted 4 November 2014 Available online xxx

Polyoxometalates (POMs) are inorganic cluster metal complexes that possess versatile biological activities, including antibacterial, anticancer, antidiabetic, and antiviral effects. Their mechanisms of action at the molecular level are largely unknown. However, it has been suggested that the inhibition of several enzyme families (e.g., phosphatases, protein kinases or ecto-nucleotidases) by POMs may contribute to their pharmacological properties. Ecto-nucleotidases are cell membrane-bound or secreted glycoproteins involved in the hydrolysis of extracellular nucleotides thereby regulating purinergic (and pyrimidinergic) signaling. They comprise four distinct families: ecto-nucleoside triphosphate diphosphohydrolases (NTPDases), ecto-nucleotide pyrophosphatases/phosphodiesterases (NPPs), alkaline phosphatases (APs) and ecto-50 -nucleotidase (eN). In the present study, we evaluated the inhibitory potency of a series of polyoxometalates as well as chalcogenide hexarhenium cluster complexes at a broad range of ecto-nucleotidases. [Co4(H2O)2(PW9O34)2]10 (5, PSB-POM142) was discovered to be the most potent inhibitor of human NTPDase1 described so far (Ki: 3.88 nM). Other investigated POMs selectively inhibited human NPP1, [TiW11CoO40]8 (4, PSB-POM141, Ki: 1.46 nM) and [NaSb9W21O86]18 (6, PSB-POM143, Ki: 4.98 nM) representing the most potent and selective human NPP1 inhibitors described to date. [NaP5W30O110]14 (8, PSB-POM144) strongly inhibited NTPDase1–3 and NPP1 and may therefore be used as a pan-inhibitor to block ATP hydrolysis. The polyoxoanionic compounds displayed a non-competitive mechanism of inhibition of NPPs and eN, but appeared to be competitive inhibitors of TNAP. Future in vivo studies with selected inhibitors identified in the current study are warranted. ß 2014 Elsevier Inc. All rights reserved.

Keywords: Alkaline phosphatase Ecto-50 -nucleotidase Ecto-nucleotidase inhibitor NPP1 NTPDase1 Polyoxometalate

Abbreviations: AP, alkaline phosphatase; CDP-Star, disodium 2-chloro-5-(4-methoxyspiro[1,2-dioxetane-3,20 -(5-chlorotricyclo[3.3.1.13.7]decan])-4-yl]-1-phenyl phosphate; CE, capillary electrophoresis; CHES, 2-(N-cyclohexylamino)ethanesulfonic acid; HEPES, 4-(2-hydroxyethyl)piperazine-1-ethanesulfonic acid; eN, ecto-50 nucleotidase; LDC (18:1), 1-oleoyl-sn-glycero-3-phosphocholine; NPP, ecto-nucleotide pyrophosphatase/phosphodiesterase; NTPDase, ecto-nucleoside triphosphate diphosphohydrolase; POM, polyoxometalate; POM1, Na6[H2W12O40]21H2O; POM4, K6H2[TiW11CoO40]13H2O; POM6, (NH4)18[NaSb9W21O86]14H2O; PSB-POM141, K6H2[TiW11CoO40]13H2O; PSB-POM142, K10[Co4(H2O)2(PW9O34)2]22H2O; PSB-POM143, (NH4)18[NaSb9W21O86]14H2O; PSB-POM144, Na14[NaP5W30O110]30H2O; PV4, K6H2[TiW11CoO40]13H2O; TNAP, tissue-nonspecific alkaline phosphatase; TOOS, 3-(N-ethyl-3-methylanilino)-2-hydroxypropanesulfonic acid; Tris, tris(hydroxymethyl)aminomethane. * Corresponding author. E-mail addresses: [email protected], [email protected] (C.E. Mu¨ller). 1 Both authors contributed equally to this study. http://dx.doi.org/10.1016/j.bcp.2014.11.002 0006-2952/ß 2014 Elsevier Inc. All rights reserved.

Please cite this article in press as: Lee S-Y, et al. Polyoxometalates—Potent and selective ecto-nucleotidase inhibitors. Biochem Pharmacol (2014), http://dx.doi.org/10.1016/j.bcp.2014.11.002

G Model

BCP-12133; No. of Pages 11 2

S.-Y. Lee et al. / Biochemical Pharmacology xxx (2014) xxx–xxx

1. Introduction Polyoxometalates (POMs) are negatively charged inorganic compounds which contain early transition metal ions such as tungsten (W), molybdenum (Mo), niobium (Nb), antimony (Sb) or vanadium (V), surrounded by oxygen atoms [1–3]. In POMs transition-metal ions are typically in their highest oxidation state [4]. Those anionic complexes are relatively stable, some of them are even highly stable in aqueous solutions at physiological pH values [4]. The diversity in structure and composition of POMs allows for a wide versatility in terms of shape, polarity, redox potential, surface charge distribution, and acidity, resulting in many possible applications in the fields of catalysis, electronics, magnetic materials and nanotechnology [5,6]. Furthermore, POMs have been shown to exhibit biological activities in vitro as well as in vivo, including anticancer [7,8], antibacterial [9,10], antiprotozoal [11], antiviral[12,13], and antidiabetic activities [1,14]. However, their biological mechanisms of action at the molecular level are not well understood. Due to their anionic character and their high negative charge at physiologic pH values, POMs will hardly be able to penetrate cells [15]. Therefore, it has been speculated that POMs are likely to act extracellularly inhibiting several different enzyme families such as phosphatases [11], kinases [15], sulfotransferases [9], sialyltransferases [9], and ecto-nucleotidases [16–20], which are mostly located on the plasma membrane and display extracellular binding sites [21]. Ecto-nucleotidases regulate extracellular levels of nucleotides by catalyzing their hydrolysis, eventually leading to the formation of the respective nucleosides and inorganic phosphates [22– 25]. Since extracellular nucleosides and nucleotides act as signaling molecules in almost all tissues and organs by activating P1 and P2 purinergic receptors, respectively, ecto-nucleotidases have recently gained considerable interest due to their important roles in modulating purinergic signal transduction [26–28]. At least four protein families display ecto-nucleoside activity: ectonucleoside triphosphate diphosphohydrolases (NTPDases, EC 3.6.1.5), ecto-50 -nucleotidase (eN, EC 3.1.3.5), ecto-nucleotide pyrophosphatases/phosphodiesterases (NPPs, EC 3.1.4.1, and EC 3.6.1.9), and alkaline phosphatases (APs, EC. 3.1.3.1) [26,27,29]. Nucleoside 50 -triphosphates and 50 -diphosphates are hydrolyzed by members of the E-NTPDase family, the E-NPP family, and by APs (e.g. ATP or ADP to AMP), whereas nucleoside 50 -monophosphates are hydrolyzed by APs and eN (e.g. AMP to adenosine) [26,30,31]. Many cancer cells show an overexpression of NTPDases and eN leading to increased extracellular adenosine levels. Since adenosine promotes angiogenesis, tumor growth, and immunosuppression, inhibitors of eN and NTPDases have considerable potential for the treatment of various diseases which are associated with elevated adenosine concentrations, e.g. cancer or immunodeficiency disorders [17,31,32]. Beside mammalian NTPDases, microbial NTPDases have been found to be expressed by some important pathogens (e.g. Legionella pneumophila) and reported to contribute to their virulence. Therefore NTPDase inhibitors have been further proposed as novel anti-bacterial therapeutics [16,26,33–35]. NPPs have been involved in various biological processes including bone mineralization, blood coagulation and regulation of insulin receptor signaling. Moreover, expression of NPPs in cancer cells has been demonstrated to promote angiogenesis, lymphocyte trafficking and tumor growth [29,36–39]. Thus inhibition of NPPs has been proposed as a new potential therapeutic strategy, e.g., for the treatment of diabetes and cancer. Tissue-nonspecific alkaline phosphatase (TNAP) ensures normal bone mineralization by hydrolyzing extracellular diphosphate (PPi), a potent inhibitor of hydroxyapatite formation [40–43]. TNAP levels rise when bones are growing due to bone fractures or tumors [44]. TNAP is further involved in pathophysiological abnormalities,

which can lead to ankylosis, vascular calcification, and osteoarthritis [45,46]. Therefore TNAP inhibitors may be therapeutically useful for the treatment of diseases such as arterial calcification, ankylosis and cancer. Furthermore potent and selective inhibitors are required as pharmacological tools to further investigate the (patho)physiological roles of ecto-nucleotidases in various tissues and pathological contexts. We previously identified several polyoxotungstates as potent inhibitors of rat NTPDases with submicromolar potency [16]. Subsequently, POMs have been applied in a number of biological studies to inhibit NTPDases [17–20]. The compound [H2W12O40]6, designated POM1, has been used in physiological studies, where it was found to increase infarct sizes and abolish the protective effects of cardiac and renal ischemia preconditioning by NTPDase1 inhibition [17]. This compound has been subsequently tested in rat cerebellar and hippocampal slice preparations and was found to be more effective in blocking ATP breakdown than the standard NTPDase1 inhibitor (ARL 67156) [18]. Additionally, the inhibition of NTPDase1 activity by POM1 significantly inhibited hepatic metastatic colonic tumour growth in mice [19]. Recently, PV4 (or POM4, [TiW11CoO40]8), which was found to be a particularly potent inhibitor of rat NTPDase2 [16], was shown to inhibit the hydrolysis of the extracellular ATP under physiological and ischemic conditions in vivo in the rat striatum [20]. Also, crystallographic studies of two different POMs (POM1 and (NH4)6[Mo7O24]) demonstrated that the negatively charged POMs bind electrostatically to the positively charged domain of rat NTPDase1 [47]. Inhibition of enzyme function by POMs was thereby proposed to be induced by changes in the Y409 substrate-binding loop structure restricting enzyme flexibility [47]. Very recently, POM1, a potent inhibitor of rat NTPDase1, and POM6 ([NaSb9W21O86]18), a selective inhibitor of rat NTPDase2 and -3 [16], have been investigated in plasma and blood cell samples from human volunteers, and the ADPase activity in hematocytes was shown to be primarily blocked by POM1, while its activity in plasma was potently inhibited by POM6. This suggests the presence of NTPDase1 on cell membranes of hematocytes and the presence of NTPDase2 or -3 in blood plasma [48]. Although a number of reports have documented the inhibitory effect of POMs on NTPDases, their effect on the other ectonucleotidases have scarcely been investigated so far. Moreover, the previous enzyme inhibition studies were performed only on rat, but not on human NTPDases. In the present study, we investigated the inhibitory potency of an extended series of 16 POMs and structurally related chalcogenide hexarhenium cluster complexes on human NTPDases. In addition, their inhibitory activity on further ecto-nucleotidases, including human NPP1, -2, and -3, TNAP, and rat ecto-50 -nucleotidase (eN) was investigated, and selectivity profiles of the compounds were established. This led to the discovery of the first POM inhibitor of NPP1 and the most potent and selective NPP1 inhibitors known to date. Finally, we analyzed their mechanism of inhibition of the various ecto-nucleotidases. 2. Materials and methods 2.1. Materials 2-(N-cyclohexylamino)ethanesulfonic acid (CHES), 4-(2-hydroxyethyl)piperazine-1-ethanesulfonic acid (HEPES), levamisole hydrochloride and tris(hydroxymethyl)aminomethane (Tris) were purchased from Applichem (Darmstadt, Germany). pAcGP67-A and pAcG2T baculovirus expression vectors were purchased from BD BaculoGold (Heidelberg, Germany). The mouse monoclonal ADP antibody (3 mg/mL), the monoclonal AMP/GMP antibody (1.2 mg/ mL), the ADP AlexaFluor1 633 tracer (400 nM), the AMP/GMP AlexaFluor1 633 tracer (800 nM) and ADP (5 mM) were obtained from BellBrook Laboratories (Madison, WI, USA). Agarose, disodium


G Model

BCP-12133; No. of Pages 11 S.-Y. Lee et al. / Biochemical Pharmacology xxx (2014) xxx–xxx

hydrogenphosphate and ethylenediaminetetraacetic acid (EDTA) were purchased from Carl Roth (Karlsruhe, Germany). Cellfectin, COS-7 cells, Lipofectamine, pcDNA3 expression vector, pUC19 expression vector and Spodoptera frugiperda sf9 cells was obtained from Invitrogen (Darmstadt, Germany). Insect Xpress media was obtained from Lonza Group (Basel, Switzerland). Amicon1Ultra 15 mL filters were purchased from Merck Millipore (Schwalbach, Germany). Deoxynucleotides (dNTPs), disodium 2-chloro-5-(4methoxyspiro[1,2-dioxetane-3,20 -(5-chlorotricyclo[3.3.1.13.7]decan])-4-yl]-1-phenyl phosphate (CDP-Star), restriction enzymes (BamHI, SmaI and XbaI), and T4 DNA polymerase was purchased from New England BioLabs (Frankfurt am Main, Germany). ADP, 4aminoantipyrine, AMP, aprotinin, ATP, bovine serum albumin, Brij1 L23, calcium chloride, choline oxidase, diethanolamine (DEA), dimethyl sulfoxide (DMSO), Dowex 50WX8, glycerol, magnesium chloride, 1-oleoyl-sn-glycero-3-phosphocholine (LPC (18:1)), peroxidase from horseradish, phenylmethylsulfonyl fluoride (PMSF), phosphotungstic acid, sodium chloride, sodium hydroxide, 3-(Nethyl-3-methylanilino)-2-hydroxypropanesulfonic acid (TOOS), uridine and zinc chloride were obtained from Sigma (Steinheim, Germany). Zymo Gel DNA Recovery Kit and Zymo DNA Clean & Concentrator were obtained from Zymo Research (Irvine, USA). Na6[H2W12O40]21H2O (1), H3[PW12O40]H2O (2), K7[Ti2W10PO40]8H2O (3), K6H2[TiW11CoO40]13H2O (4), K10[Co4(H2O)2(PW9O34)2]22H2O (5), (NH4)18[NaSb9W21O86]14H2O (6) have previously been described [16]. K4[(Re6S8)(OH)6]8H2O (13), K4[(Re6Se8)(OH)6]8H2O (14), K4[(Re6S8)(CH3COO)6]8H2O (15) and K4[(Re6S8)(HCOO)6]3H2O (16) were prepared as described [49–51]. Na14[NaP5W30O110]30H2O (8) [52], Na20[P6W18O79]37H2O (9) [53], Na33[H7P8W48O184]92H2O (10) [54], Na16[(O3POPO3)4W12O36]38H2O (11) [55], and Na16[(O3PCH2PO3)4W12O36]16H2O (12) were prepared according to published procedures [55]. Na3[PW12O40]7H2O (7) was obtained by converting phosphotungstic acid to its sodium salt by ion exchange chromatography on a Dowex 50WX8 column. 2.2. Enzyme preparations of TNAP, eN and NTPDases Human recombinant soluble alkaline phosphatase (TNAP), expressed in NS0 cells from murine myeloma, was obtained from R&D Systems GmbH (Wiesbaden, Germany). Rat recombinant soluble ecto-50 -nucleotidase (eN), expressed in Sf9 insect cells was prepared as previously described [56]. Membrane preparations of COS-7 cells recombinantly expressing human NTPDase1, -2, -3, or 8, respectively, were obtained as described previously [57,58]. Briefly, COS-7 cells (in 15 cm dishes) were transfected with an expression vector (pcDNA3) incorporating the cDNA that encodes each ecto-nucleotidase using Lipofectamine. Cells were harvested 40–72 h later. To obtain the protein extracts, transfected cells were washed three times with 95 mM NaCl and 45 mM Tris, pH 7.5, at 4 8C, collected by scraping using the same buffer supplemented with 0.1 mM phenylmethylsulfonyl fluoride (PMSF) and washed twice by centrifugation (300 g, 10 min, 4 8C). The cells were then resuspended in harvesting buffer supplemented with 10 mg/mL aprotinin and sonicated. Nucleus and cellular debris were discarded after another centrifugation (300 g, 10 min, 4 8C). The resulting supernatant (hereinafter called protein extract) was aliquoted and stored at -80 8C after estimation the protein concentration by a Bradford microplate assay using bovine serum albumin as a standard [59]. 2.3. Expression and enzyme preparation of human NPP1, NPP2 and NPP3 Full length human cDNAs of NPP1, NPP2 and NPP3 (Genbank accession no. NM_006258, NM_006209, and NM_005021,

3

respectively) were obtained from Origene (Rockville, USA). To amplify the cDNAs polymerase chain reactions (PCR) were performed by using human NNP1-forward primer (50 GATCGGATCC GCCACGAAAGAAGTTAAAAGTTGCAAAGG-30 ) and human NPP1-reverse primer (50 -GATCTCTAGATCAGTCTTCTTGGCTAAAGGTTG-30 ) for human NPP1, human NPP2-forward (50 GATCGGATCCGCCACCATGGCAAGGAGGAGCTCGTTCC-30 ) and human NPP2-reverse primer (50 -GATCCCCGGGTTAAATCTCGCTCTCATATGTATG-30 ) for human NPP2, human NPP3-forward primer (50 GATCGGATCCGCCACCATGTCACTTGGATTAGGCCTGG-30 ), and human NPP3-reverse primer (50 -GATCCTCGAGTTAAATAGTGGT TTCAAATGTTGG-30 ) for human NPP3. Amplification was started by an incubation of 2 min at 95 8C which was followed by 35 cycles of 20 s denaturation at 95 8C, 20 s annealing at 56 8C (65 8C for NPP1) and 1 min primer extension at 70 8C, and ending with 10 min incubation at 70 8C. The PCR products of approximately 2.5–2.7 kb were purified on agarose gel using the Zymo Gel DNA Recovery Kit. The DNA fragment of NPP1 was then digested with BamHI and XbaI and ligated into the pAcGP67-A vector previously digested with BamHI and XbaI. Plasmid DNAs were then purified with Zymo DNA Clean & Concentrator, digested with NotI and the sticky end of NPP1 plasmid was filled at 12 8C for 15 min with desoxynucleotides (dNTPs) by using a T4 DNA polymerase. Afterwards, they were digested with BamHI and ligated into the baculovirus expression vector pAcG2T previously digested with BamHI and SmaI and finally purified with Zymo DNA Clean & Concentrator. The purified PCR products of NPP2 and NPP3 were digested with BamHI and subcloned into the pUC19 vector previously digested with BamHI and SmaI. Both plasmid DNAs were then purified with Zymo DNA Clean & Concentrator, then digested with SmaI and purified again with Zymo DNA Clean & Concentrator. Additionally, the sticky end of NPP3 plasmid was filled at 12 8C for 15 min with dNTPs by using T4 DNA polymerase. Afterwards, both plasmid DNAs were digested with BamHI and subcloned into the baculovirus expression vector pAcG2 T previously digested with BamHI and SmaI and finally purified with Zymo DNA Clean & Concentrator. S. frugiperda Sf9 insect cells were grown at 27 8C in Insect Xpress media containing 0.1 mg/mL of gentamicin. After initial transfection, recombinant baculoviruses were produced, which were subsequently used for infecting Sf9 cells. For transfection, 8 105 Sf9 cells per well were seeded into six well plates containing 2 mL of medium without antibiotics. Prepared NPP1–3 plasmids (each 2 mg) were mixed with medium without antibiotics and incubated Cellfectin for 20 min. The transfection mixture was added drop-wise to the cells, which were left at 25 8C for 30 min, and then incubated overnight at 27 8C. Afterward, the transfection mixture was replaced by Insect Xpress medium containing gentamicin, and the cells were cultured at 27 8C until signs of viral infection were observed, usually after 5– 6 days. The supernatant, containing the P1 virus generation, was used to produce a high virus titer stock, which was then used to infect cells for protein production. After multiple infections of insect cells (at least 5 amplification rounds), the culture supernatant of Sf9 cells containing the respective recombinant enzymes was collected and further concentrated by centrifugation through a filter (Amicon1Ultra 15 mL, 50 kDa cutoff, 48,000 g, 40 min). Subsequently, the residual Insect Xpress medium of enzyme concentrates was exchanged with 50 mM Tris (pH 7.5), 10 mM NaCl and 5% glycerol by centrifugation through the filter (Amicon1Ultra 15 mL, 50 kDa cutoff, 24,000 g, 20 min). Finally, samples were kept at 80 8C. 2.4. NTPDase assays Inhibitory activity on NTPDases was initially assayed in a final inhibitor concentration of 20 mM using fluorescence polarization (FP) assays established in our group [63]. Compounds dissolved in


G Model



10% aq. DMSO (2 mL) were pipetted into wells of a 384-well microplate together with substrate solution (either 4 mL of ATP, final concentration of 20 mM for human NTPDase2, human NTPDase3 and human NTPDase8, or ADP, final concentration of 10 mM for human NTPDase1) in assay buffer (5 mM CaCl2 and 80 mM Tris with the pH value adjusted to 7.4). The reaction was initiated by adding 4 mL of the appropriate enzyme suspension in assay buffer to each test well (final concentrations of 1.0 ng/mL human NTPDase1, 0.25 ng/mL human NTPDase2, 0.80 ng/mL human NTPDase3 and 1.5 ng/mL human NTPDase8). The plate was incubated for 10 min at 37 8C. To stop the reaction 10 mL of the appropriate detection reagent was added to each well containing 23 mg/mL ADP antibody and 4 nM ADP Alexa 633 tracer in detection reagent buffer (final concentration of 20 mM HEPES, 40 mM EDTA and 0.02% Brij1 L23 with the pH value adjusted to 7.5) for 20 mM ATP as substrate or 8 mg/mL AMP antibody and 8 nM AMP Alexa 633 tracer in detection reagent buffer for 10 mM ADP as substrate. After equilibration for 1 h (ADP detection assay), or 2 h (AMP detection assay), respectively, at room temperature (with orbital shaking) the FP signals were measured on a BMG PheraStar FS plate reader (BMG Labtech GmbH, Ortenberg, Germany) using an excitation wavelength of 590 nm and an emission wavelength of 675 nm (50 nm bandwidth). Product formation leads to a decrease in FP values. For the determination of IC50 values, 6–8 different concentrations of the respective inhibitor spanning 3 orders of magnitude were used and IC50 values were determined by nonlinear curve fitting using the program PRISM 4.0 (GraphPad, San Diego, USA). 2.5. NPP1 and NPP3 assays The test compounds were initially screened in a concentration of 10 mM. Solutions were prepared in the reaction buffer (1 mM MgCl2, 2 mM CaCl2, 10 mM CHES (2-(N-cyclohexylamino)ethanesulfonic acid), pH 9.0) together with the substrate ATP (400 mM). The reaction was initiated by adding 20 mL of human NPP1 (1.7 mg) or human NPP3 (43 mg) and was incubated at 37 8C for 30 min (NPP1) or 60 min (NPP3). After the incubation, the reaction was stopped by heating at 90 8C for 3 min. After cooling in ice, the reaction samples were transferred into CE (capillary electrophoresis) vials and injected into the CE instrument. The operation conditions in CE were as follows. All experiments were carried out using a P/ACE MDQ capillary electrophoresis system (Beckman Instruments, Fullerton, CA, USA) equipped with a DAD detection system. Data collection and peak area analysis were performed by the P/ACE MDQ software 32 KARAT obtained from Beckman Coulter (Fullerton, CA, USA). The electrophoretic separations were carried out using a polyacrylamide-coated capillary (60 cm [50 cm effective length], 50 mm (id) obtained from CS-chromatography (Langerwehe, Germany)). Electrokinetic injections were performed using a voltage of 6 kV for 60 s and separations were carried out by a voltage of 20 kV. Analytes were detected using direct UV absorbance at 260 nm. The capillary temperature was kept constant at 15 8C and the temperature of the storing unit was adjusted to 15 8C. The running buffer consisted of 100 mM phosphate buffer (pH 6.5). Between separations, the capillary was washed with water for 2 min (20 psi) and subsequently with running buffer for 2 min (20 psi) before each injection. The determination of IC50 values was performed by the same procedures as described in Section 2.4 (with ATP as a substrate). Mechanisms of inhibition were determined using five different concentrations of ATP (from 20 to 500 mM), and three different concentrations (0, 0.5-fold and 2-fold IC50) of inhibitor. The type of inhibition was evaluated graphically for each inhibitor from the Lineweaver–Burk plots using PRISM 5.0.

2.6. NPP2 assays Initial screening assays were carried out in an inhibitor concentration of 10 mM. The enzyme inhibition assays were performed in a final volume of 50 mL. The reaction mixture contained 5 mM MgCl2, 5 mM CaCl2, 100 mM Tris, pH 9.0, and 400 mM LPC (18:1). The reaction was started by the addition of 10 mL of human NPP2 (44 mg). The mixture was incubated at 37 8C for 60 min, and subsequently the released choline was quantified colorimetrically at 555 nm after incubation at 37 8C for 10 min with 50 mL of each, the peroxidase reagent (50 mM Tris at pH 9.0, 2 mM TOOS, 5 U/mL peroxidase), and the choline oxidase reagent (50 mM Tris at pH 9.0, 2 mM 4-aminoantipyrine, 5 U/mL choline oxidase). Determination of IC50 values and of the mechanism of inhibition was performed by the procedures described in Sections 2.4 and 2.5 (but with LPC (18:1) as a substrate) and data were calculated using Prism 5.0. 2.7. Ecto-50 -nucleotidase assays The investigation of compounds for inhibition of rat eN was performed as previously described using a CE method [60]. The test compounds were initially screened in a concentration of 100 mM. They were dissolved in the assay buffer (in mM: 1 CaCl2, 1 MgCl2, 140 NaCl, 20 HEPES, pH 7.4) together with the substrate AMP (500 mM). The enzyme reaction was started by adding 0.13 mg of recombinant rat eN, then incubated at 37 8C for 15 min, and subsequently the reaction was terminated by heating at 99 8C for 5 min. Finally, 50 mL of the reaction mixture were transferred into mini-CE vials containing 50 mL of the internal standard uridine (final concentration 6.25 mM). The operation conditions in CE were as follows: P/ACE MDQ capillary electrophoresis system, fusedsilica capillary (40 cm [30 cm effective length], 75.5 mm (id) obtained from Polymicro Technologies (Kehl, Germany)), hydrodynamic injection (0.5 psi, 5 s), separation voltage of 20 kV, running buffer (40 mM borax at pH 9.1), and detection at 260 nm. Between separations, the capillary was washed with 0.1 N aq. NaOH solution for 2 min (20 psi) and subsequently with running buffer for 2 min (20 psi) before each injection. Determination of IC50 values and the mechanism of inhibition were performed by the same procedures as described in Sections 2.4 and 2.5 (but with AMP as a substrate), and analysed using the program PRISM 5.0. 2.8. Tissue-nonspecific alkaline phosphatase (TNAP) assays For the testing of polyoxometalates and chalcogenide hexarhenium cluster complexes at TNAP we utilized a luminescent assay based on the method described by Sergienko et al. [43] using CDP-Star, a 1,2-dioxetane-based synthetic substrate of TNAP. Stock solutions (10 mM) of the compounds were prepared in water and dilutions were made in 10% aq. DMSO. The inhibition experiments were carried out in 96-well plates (final volume of 50 mL) and a final compound concentration of 20 mM was employed. After addition of 20 mL CDP-Star dissolved in water (final concentration 105 mM) to 10 mL of compound solution the enzymatic reaction was started by adding 20 mL of recombinant human TNAP (final concentration 0.02 ng/mL) dissolved in assay buffer (125 mM DEA, 2.5 mM MgCl2, 0.05 mM ZnCl2, pH 9.8). Samples containing the standard inhibitor levamisole (final concentration 1 mM) instead of test compound were used as positive controls. After 30 min of incubation at RT, the enzyme activity was monitored luminometrically on a microtiter plate reader (Berthold Technologies). The enzymatic dephosphorylation of CDP-Star leads to the formation of a meta-stable dioxetane phenolate anion, which decomposes to a stable product with light emission in the lower region of the visible spectrum. The measured light intensity increases proportionally to


G Model


5

Table 1 Investigated polyoxometalates (POMs) and chalcogenide hexarhenium cluster complexes. Compd

Formula

Charge at pH 7.4a

Molecular weight (g/mol)

Code names

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Na6[H2W12O40]21H2O H3[PW12O40]H2O K7[Ti2W10PO40]8H2O K6H2[TiW11CoO40]13H2O K10[Co4(H2O)2(PW9O34)2]22H2O (NH4)18[NaSb9W21O86]14H2O Na3[PW12O40]7H2O Na14[NaP5W30O110]30H2O Na20[P6W18O79]37H2O Na33[H7P8W48O184]92H2O Na16[(O3POPO3)4W12O36]38H2O Na16[(O3PCH2PO3)4W12O36]16H2O K4[(Re6S8)(OH)6]8H2O K4[(Re6Se8)(OH)6]8H2O K4[(Re6S8)(CH3COO)6]8H2O K4[(Re6S8)(HCOO)6]3H2O

6 3 7 8 10 18 3 14 20 40 16 16 4 4 4 4

3366 2880 3023 3240 5518 6932 3074 8320 5888 14446 4532 4128 1776 2151 2028 1854

POM1

a

POM4, PV4, PSB-POM141 PSB-POM142 POM6, PSB-POM143 PSB-POM144

Compounds are fully deprotonated at pH 7.4 and at higher pH values.

product formation. The determination of IC50 values was performed by the same procedures as described in Section 2.4 (but with CDP-Star as a substrate). Mechanisms of inhibition were determined using different concentrations of CDP-Star (from 0 to 600 mM), and three different concentrations of inhibitor. The type of inhibition was evaluated graphically from the Lineweaver–Burk plots using PRISM 4.0.

2.9. Statistical analyses for selectivity determination In order to evaluate the selectivity of each compound for a particular enzyme all numerical data were expressed as means SEM of pKi values, in which pKi is equal to log Ki. The obtained IC50 values were transformed into Ki values by using the Cheng– Prusoff equation [61].

Fig. 1. Structures of polyoxometalates (POMs) and chalcogenide hexarhenium cluster complexes used in the present study: Keggin-type structures of compounds 1–4 and 7; trivacant Keggin-derived sandwich structure of compound 5; crypt-like structure of compound 6; Preyssler-structure of compound 8; banana-shaped structure of compound 9; wheel-shaped structure of compound 10; saddle-like structure of compounds 11 and 12; structure of anionic cluster units [(Re6Q8)L6]4 in 13 (Q = S, L = OH), 14 (Q = Se, L = OH), 15 (Q = S, L = CH3COO) and 16 (Q = S, L = HCOO).


G Model


6

Table 2 Potency of polyoxometalates and chalcogenide hexarhenium cluster complexes at human NTPDases. Ki SEM (mM)

Compd

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

NTPDase1

NTPDase2

NTPDase3

NTPDase8

0.800 0.210 6.87 1.81 1.28 0.50 0.173 0.036 0.00388 0.00140 >20 (38%)a 3.70 0.88 0.0280 0.0046 7.85 0.66 0.626 0.200 8.81 2.65 15.6 2.4 >20 (26%)a >20 (17%)a >20 (12%)a 2.44 0.43

3.29 0.16 6.64 0.70 3.03 0.40 0.0882 0.0362 0.0184 0.0035 3.80 0.34 3.65 0.36 0.0218 0.0058 4.21 0.54 0.194 0.051 10.4 2.2 9.59 1.59 >20 (11%)a >20 (7%)a >20 (1%)a 0.680 0.151

0.663 0.056 1.17 0.30 0.596 0.179 0.110 0.031 0.0596 0.0092 0.782 0.116 0.820 0.074 0.0777 0.0126 1.99 0.86 0.310 0.103 3.29 0.54 18.5 5.3 > 20 (28%)a > 20 (20%)a > 20 (2%)a 0.582 0.050

>20 >20 >20 >20 >20 >20 >20 >20 >20 >20 >20 >20 >20 >20 >20 >20

(12%)a (11%)a (9%)a (4%)a (2%)a (5%)a (12%)a (15%)a (15%)a (4%)a (19%)a (24%)a (20%)a (3%)a (1%)a (2%)a

Note. Inhibition measured at membrane preparations of human NTPDase1, -2, -3 and -8. For the experimental procedure see Section 2.4. The Ki values are expressed as means SEM of three to five independent experiments. a Percent inhibition at 20 mM (n = 2).

where [S] = substrate concentration (mM) and Km = Michaelis– Menten constant (mM). For non-competitive inhibition (with the assumption a = 1) [62]:

of ecto-nucleotidases. Table 1 lists the investigated POMs including twelve polyoxotungstates (1–12) and four rhenium clusters (13– 16). Compounds 1–4 and 7 are Keggin structure complexes, 5 is a trivacant Keggin-derived sandwich complex, 6 features a cryptlike structure, 8 a Preyssler-structure, 9 is a banana-shaped structure, 10 is a wheel-shaped structure, and 11 and 12 are saddle-like structures; compounds 13–16 are chalcogenide hexarhenium cluster complexes (see Fig. 1) [13,16,49–51,55].

K i ¼ IC50

3.1. Effects of compounds on human NTPDases

For competitive inhibition: Ki ¼

IC50 ð1 þ ½S=K m Þ

In the case of weakly potent compounds no concentration– inhibition curves were recorded. The IC50 values were set equal to the maximally tested inhibitor concentrations and maximal Ki values were deduced using the equations above. Statistical data analyses with pKi values were performed using Prism 5.0 software. Data were tested for statistical significance by analysis of variance (one-way ANOVA) as appropriate. When significant differences were observed, the Newman–Keuls multiple comparison test (one-way ANOVA) was done. A value of P < 0.05 was considered significant.

Table 3 Potency of polyoxometalates and chalcogenide hexarhenium cluster complexes at human NPPs.

3. Results and discussion In the present study we investigated the inhibitory potency of a series of 12 different polyoxometalates (POMs) and of four chalcogenide hexarhenium cluster compounds on a broad range

120

residual activity of enzymes (%)

100 80

human NTPDase1 human NTPDase2 human NTPDase3 human NTPDase8

60 40 20 0 10 -11

10 -9

10 -7

The POMs and chalcogenide hexarhenium cluster complexes were screened for their inhibitory activities on human NTPDases that are expressed on the plasma membrane with an extracellular active site, namely NTPDase1, -2, -3 and -8. The Ki values are presented in Table 2. In general, none of the tested compounds exhibited any significant inhibitory activity on NTPDase8 in a high compound concentration of 20 mM, whereas potent inhibitors

10 -5

10 -3

[compound 5], M Fig. 2. Concentration–inhibition curve of compound 5 at human NTPDase1 (Ki = 3.88 1.4 nM, n = 3), human NTPDase2 (Ki = 18.4 3.5 nM, n = 3), human NTPDase3 (Ki = 59.6 9.2 nM, n = 4) and human NTPDase8 (Ki > 20 000 nM, n = 2).

Compd

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Ki SEM (mM) NPP1

NPP2

NPP3

0.0118 0.0002 0.0365 0.0011 0.0343 0.0030 0.00146 0.00001 0.0690 0.0038 0.00498 0.00042 0.0311 0.0035 0.0245 0.0009 0.0722 0.0004 0.0130 0.0009 0.0200 0.0013 0.0410 0.0066 0.537 0.009 0.161 0.008 0.419 0.059 0.100 0.007

>10 (9%)a >10 (15%)a >10 (40%) a 12.0 0.1 >10 (41%)a >10 (2%)a >10 (37%)a >10 (1%)a >10 (14%)a 16.8 1.0 >10 (35%)a >10 (45%)a >10 (14%)a >10 (12%)a >10 (16%)a >10 (17%)a

>10 (34%)a >10 (57%)a >10 (31%)a 1.37 0.02 >10 (14%)a >10 (39%)a >10 (59%)a >10 (46%)a 3.73 0.21 0.232 0.008 >10 (53%)a >10 (55%)a >10 (1%)a >10 (1%)a >10 (0%)a >10 (10%)a

Note. Inhibition measured at human soluble NPP1, -2, and -3. For the experimental procedure, see Sections 2.5 and 2.6. The Ki values are expressed as means SEM of three independent experiments. a Percent inhibition at 10 mM (n = 3).


G Model

BCP-12133; No. of Pages 11

B

120 100 80 60

human NPP1 human NPP2 human NPP3

40 20 0 10 -12 10 -10 10 -8 10 -6 10 -4 10 -2


A



7

120 100 80 60

human NPP1 human NPP2 human NPP3

40 20

[compound 4], M

0 10 -12 10 -10 10 -8 10 -6 10 -4 10 -2

[compound 10], M

Fig. 3. Concentration–inhibition curves for the most potent polyoxometalates of human NPP1, 2, and 3. (A) Concentration–inhibition curve of compound 4 at human NPP1 (Ki = 0.00146 0.00001 mM, n = 3), human NPP2 (Ki = 12.0 0.1 mM, n = 3) and human NPP3 (Ki = 1.37 0.02 mM, n = 3). (B) Concentration–inhibition curve of compound 10 at human NPP1 (Ki = 0.0130 0.0009 mM, n = 3), human NPP2 (Ki = 16.8 1.0 mM, n = 3) and human NPP3 (Ki = 0.232 0.008 mM, n = 3).

6 showed weak potency on both, human and rat, NTPDase1 (Ki value >20 mM) [16]. The type of inhibition of NTPDases was not determined in this study. But Zebisch et al. [47] already showed for rat NTPDase1 that POMs can bind electrostatically to a loop which is involved in binding of the nucleobase. After binding they might inhibit NTPDase activity by changing the loop structure or limiting its flexibility [47]. Moreover, Wall et al. [18] have investigated the inhibitory potency and physiological effects of POM-1 (1) on cerebellar and hippocampal slices. They confirmed that POM1 is a potent NTPDase inhibitor and provided evidence for its noncompetitive inhibition type [18]. Therefore we assume a noncompetitive inhibition mechanism for the investigated POMs.

could be identified for the other NTPDase subtypes. Compounds 1–12 were found to be more potent at human NTPDase1, -2 and -3 than the rhenium cluster compounds 13–16, which showed negligible inhibitory activity except for compound 16 (Ki values ranging from 0.582 to 2.44 mM for NTPDase1, -2 and -3). At NTPDase1 only compound 6, a tungsten-based metal complex with a crypt-like structure, exhibited low inhibition (38%), while all other polyoxotungstates displayed Ki values in the low nanomolar to micromolar range (0.00388–15.6 mM). Contrary to NTPDase1, the NTPDase subtypes 2 and 3 were also effectively inhibited by compound 6 (Ki values of 3.80 mM and 0.782 mM, respectively; see Table 2 and Fig. 7). We identified POM 5 as the most potent inhibitor of NTPDase1, -2 and -3 with Ki values of 3.88 nM, 18.4 nM and 59.6 nM, respectively (Fig. 2). Subsequently, our results with human NTPDases were compared with already published inhibition data of compounds 1–6 tested at rat NTPDase1–3 [16]. The investigated compounds were generally more potent at the human enzymes: Ki values were found to range from 0.00388 to 20 mM, while Ki values of these compounds when tested at rat NTPDase1, -2 and -3 were previously reported to range from 0.140 to 37.4 mM [16]. For compound 5, which was identified as the most potent inhibitor of human NTPDase1 (Ki = 0.00388 mM), NTPDase2 (Ki = 0.0184 mM) and NTPDase3 (Ki = 0.0596 mM), Ki values of 0.480 mM for rat NTPDase1, 1.53 mM for rat NTPDase2 and 2.61 mM for rat NTPDase3 have been published [16]. Thus, compound 5 shows at least 123-, 83- and 44-fold higher inhibitory potency on the human NTPDase1, -2 and -3, respectively. Interestingly, compound

A

In subsequent experiments, the effects of polyoxometalates and rhenium complexes on NPPs including NPP1 (PC-1), NPP2 (autotaxin) and NPP3 were investigated. All tested compounds were found to exhibit high inhibitory activity on NPP1 with Ki values in the range of 1.46 to 537 nM (Table 3). The tungsten-based metal complexes (1–12) were considerably more potent in blocking NPP1 than the rhenium clusters (13–16). While chalcogenide hexarhenium complexes exhibited Ki values between 100 and 537 nM, the complexes with tungsten showed Ki values of 1.46 to 72.2 nM. Among the investigated compounds, 4 and 6 were the most potent NPP1 inhibitors with Ki values of 1.46 nM and 4.98 nM, respectively (Fig. 3). Among the rhenium cluster

B

C 0.4

6

0.08

0.3

0.06 0.04

1/v

4

1/v

1/v

3.2. Effects on human NPPs

0.2

2 0.02

0.1

0.02 0.04

-0.04 -0.02

1/[S]

-0.02

0.02

1/[S]

0.04

-0.02

0.02

0.04

0.06

1/[S]

Fig. 4. (A) Lineweaver–Burk plot of NPP1 inhibition by compound 4. S, substrate concentration of ATP (mM), v, velocity of enzyme (nmol/min/mg protein); concentration of compound 4: green circle, 0 mM; blue triangle, 0.001 mM; and red triangle, 0.004 mM. (B) Lineweaver–Burk plot of NPP2 inhibition by compound 4. S, substrate concentration of LPC (18:1) (mM), v, velocity of enzyme (nmol/min/mg protein); concentration of compound 4: green circle, 0 mM; blue triangle, 5 mM; and red triangle, 20 mM. (C) Lineweaver–Burk plot of NPP3 inhibition by compound 4. S, substrate concentration of ATP (mM), v, velocity of enzyme (nmol/min/mg protein); concentration of compound 4: green circle, 0 mM; blue triangle, 1 mM; and red triangle, 5 mM. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)


G Model



Table 4 Potency of polyoxometalates and chalcogenide hexarhenium cluster complexes at human tissue-nonspecific alkaline phosphatase and ecto-50 -nucleotidase.

and 16.8 mM (10)). In case of NPP3, three compounds, 4, 9 and 10, were found to inhibit enzymatic activity with Ki values ranging from 0.232 to 3.73 mM. The Lineweaver–Burk plot for NPP1, NPP2 and NPP3 with compound 4 revealed a non-competitive mechanism of inhibition by showing the same x-intercept for uninhibited and inhibited enzyme (Fig. 4).

Ki SEM (mM)

Compd

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Rat eN

Human TNAP

>100 (26%)a 689 55 >100 (1%)a 14.1 1.1 >100 (24%)a >100 (0%)a >100 (-2%)a >100 (11%)a >100 (-18%)a >100 (48%)a >100 (-23%)a (11.4 1.3)b 11.4 1.8 52.3 2.7 >100 (20%)a 4.57 1.28

>10 (23%)c 3.45 0.17 >10 (34%)c 5.35 0.35 >10 (36%)c >10 (56%)c 3.85 0.12 >10 (18%)c 2.27 0.12 >10 (51%)c 3.06 0.04 3.33 0.38 >10 (17%)c >10 (21%)c >10 (12%)c >10 (19%)c

3.3. Effects on human tissue-nonspecific alkaline phosphatase The polyoxometalates and rhenium cluster complexes (1–16) were investigated for their potential to inhibit human TNAP. All rhenium-based metal cluster complexes (13–16) were nearly inactive at TNAP in a high concentration of 20 mM (12–21% inhibition), and only six compounds exhibited moderate TNAP inhibition with Ki values in the micromolar concentration range (2.27 to 5.35 mM, Table 4). Compound 9 was identified as the most potent TNAP inhibitor of the present series (Ki = 2.27 mM, Fig. 5). Therefore we decided to determine its mechanism of enzyme inhibition (Fig. 5). The three lines (different inhibitor concentrations) in the Lineweaver–Burk plot intersect at the y-axis (same Vmax value) indicating a competitive mechanism of inhibition versus the employed substrate CDP-Star. This is the first study that has investigated the inhibitory potential of POMs on human TNAP. Five compounds (1, 9, 10, 11 and 12) were previously published to inhibit porcine kidney alkaline phosphatase using the standard spectrophotometric assay with p-nitrophenyl phosphate as an artificial substrate [44]. When compared to the inhibitory potency on the human enzyme determined in the present study and using the artificial substrate CDP-Star, the previously published compounds revealed higher inhibitory potency on porcine kidney alkaline phosphatase with

Note. Inhibition measured at human soluble TNAP and membrane preparations of rat eN. For the experimental procedure, see Sections 2.7 and 2.8. The Ki values are expressed as means SEM of three independent experiments. a Percent inhibition at 100 mM (n = 3). b EC50 value in means SEM of three separate experiments. c Percent inhibition at 20 mM (n = 3),

compounds, 14 and 16 showed the best NPP1 inhibition with Ki values of 161 nM and 100 nM, respectively. Generally, NPP2 and NPP3 were only moderately inhibited by POMs in a test concentration of 10 mM. NPP3 was more effectively inhibited by POMs than NPP2. Only two compounds, 4 and 10; had a (moderately) inhibitory effect on NPP2 (Ki values: 12.0 mM (4)

B

120 100

0.0003

80

1/RLU

residual activity of human TNAP (%)

A

60 40

0.0002

0.0001

20 0 10 -8

10 -6

10 -4

-0.050 -0.025 0.000 0.025 0.050

10 -2

1/[S]

[compound 9], M

B

C

4

100 3

80 60

2

40 1

20 0 10 -8 10 -7 10 -6 10 -5 10 -4 10 -3

[compound 16], M

-0.10 -0.05

0.05

1/[S]

0.10

residual activity of rat eN (%)

500

120

1/v

A

residual activity of rat eN (%)

Fig. 5. (A) Concentration–inhibition curve for the most potent polyoxometalate 9 at human TNAP (Ki = 2.27 0.12 mM, n = 3). (B) Lineweaver–Burk plot of human TNAP inhibition by compound 9. S, substrate concentration of CDP-Star (mM), RLU, relative luminescent unit; concentration of compound 9: green circle, 0 mM; blue triangle, 3 mM; and red triangle, 6 mM. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

400 300 200 100 0 10 -8 10 -7 10 -6 10 -5 10 -4 10 -3

[compound 12], M

Fig. 6. (A) Concentration–inhibition curve of compound 16 at rat eN (Ki = 4.57 1.28 mM, n = 3). (B) Lineweaver–Burk plot of rat eN inhibition by compound 16. S, substrate concentration of AMP (mM), v, velocity of enzyme (nmol/min/mg protein); concentration of compound 16: green circle, 0 mM; blue triangle, 1 mM; and red triangle, 4 mM. (C) Concentration–activation curve of compound 12 at rat eN (EC50 = 11.4 1.3 mM, n = 3). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)


G Model


Compound 2

Compound 1

Compound 4

Compound 3

10

10

9

10

10

*

*

8

6

4

4

N TP e1 D N ase TP 2 D N ase TP 3 D as e8 N PP 1 N PP 2 N PP 3 TN A P N

8

8

8

TP

N

eN eN

N

as D TP N

TP

N

N

e2 N

N

TP

D

as

as D

eN

4

eN

4

TP D N ase TP 1 D N ase TP 2 D N ase TP 3 D as e8 N PP 1 N PP 2 N PP 3 TN A P

4

D a TP se3 D as e8 N PP 1 N PP 2 N PP 3 TN A P

6

e1 D a TP s e 2 D N ase TP 3 D as e8 N PP 1 N PP 2 N PP 3 TN A P

pKi

pKi

*

6

e1

eN

*

6

TP N

TP e1 D a TP se2 D N ase TP 3 D as e8 N PP 1 N PP 2 N PP 3 TN A P N

N

*

pKi

pKi

e1 D a TP se2 D N ase TP 3 D as e8 N PP 1 N PP 2 N PP 3 TN A P

Compound 16

8

as

TP

N

D TP N

Compound 15 10

D

N

as

eN

N ase TP 1 D N ase TP 2 D N ase TP 3 D as e8 N PP 1 N PP 2 N PP 3 TN A P

D

eN

TP N

TP e 1 D a TP s e 2 D a N TP se3 D as e8 N PP 1 N PP 2 N PP 3 TN A P

N

N

Compound 14 10

TP

eN

as D

4 #

as TP D N

6

10

4

*

pKi

6

Compound 13

6

TP e 1 D a TP se2 D a N TP s e 3 D as e8 N PP 1 N PP 2 N PP 3 TN A P

pKi eN

TP N

8

*

10

*

eN

pKi D TP N

N

TP

N

N

10

pKi

8

pKi eN

Compound 12

10

*

N

e1 D a TP se2 D N as e TP 3 D as e8 N PP 1 N PP 2 N PP 3 TN A P

as D N

TP

N

TP

D N

N

Compound 11

Compound 10

4

eN

N TP e1 D a TP s e 2 D N ase TP 3 D as e8 N PP 1 N PP 2 N PP 3 TN A P

D TP N

pKi

4

eN

4

e1 D a TP se 2 D N ase TP 3 D as e8 N PP 1 N PP 2 N PP 3 TN A P

4

TP

TP

N

N

6

as

eN

e1 D a TP s e 2 D N ase TP 3 D as e8 N PP 1 N PP 2 N PP 3 TN A P

as D TP

6

10

8

8

*

6

Compound 9 10

Compound 8 10

8

pKi

pKi 4

pKi

4

Compound 7

*

8

TP D N as e TP 1 D N as e TP 2 D N as e TP 3 D as e8 N PP 1 N PP 2 N PP 3 TN A P

4

10

* 8

6

6

Compound 6 10

6

6

N

Compound 5

N

eN

e1 as TP e 2 D N ase TP 3 D as e8 N PP 1 N PP 2 N PP 3 TN A P D

N

TP

as N

N

N

TP

TP

D

as D TP N

N

4

eN

4

e1 D a TP se2 D N ase TP 3 D as e8 N PP 1 N PP 2 N PP 3 TN A P

6

8

*

pKi

pKi

pKi 6

10

N

8

*

as

8

as

*

8

Fig. 7. Selectivity of inhibition by POMs and rhenium sulfur/selenium cluster complexes of different ecto-nucleotidases. Data are means SEMs of pKi values. The bars in green represent the pKi values transformed from obtained IC50 values and those in grey the maximal pKi values derived from the effects at the highest tested inhibitor concentrations using the Cheng–Prusoff equation for competitive and non-competitive inhibitions (e.g. maximal pKi values for human NTPDases, 4.70 mM; human NPPs, 5.00 mM; human TNAP, 5 mM; rat eN, 4.00 mM). # No data are available for the compound 12 at rat eN, because it acts as an enzyme activator. * P < 0.05. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)


G Model


10

Ki values in the nanomolar range (0.135–0.315 mM) [44]. These differences might be due to species differences, or also due to the different substrates employed [44,46]. 3.4. Effects on ecto-50 -nucleotidase Only few of the investigated compounds showed inhibition of eN. The rhenium-based metal complexes (13–16) were considerably more potent at blocking eN than the polyoxotungstates (1–12) (Table 4). Among the investigated compounds, 4, 13 and 16 were the strongest eN inhibitors with Ki values ranging from 4.57 to 14.1 mM (Fig. 6A). The Lineweaver–Burk plot of eN inhibition for compound 16 indicated a non-competitive mechanism of inhibition by showing the same x-intercept for uninhibited and inhibited enzyme (Fig. 6B). Interestingly, compound 12 did not inhibit eN activity but, on the contrary, activated its enzymatic activity with an EC50 value of 11.4 mM (Fig. 6C), and can thus be characterized as an allosteric activator of eN. 3.5. Selectivity Fig. 7 highlights the inhibitory potencies of POMs on different ecto-nucleotidases and analyzes their inhibition profile and potential selectivity. Table 5 displays the selectivity factor for each compound which describes the margin between the Ki value for the enzyme at which it shows the strongest inhibition and the Ki value for the enzyme at which it displays the second best inhibition. Except for compounds 5 and 8, all investigated compounds blocked the activity of NPP1 more potently than that of the other ecto-nucleotidases (P < 0.05). Most of them were at least 5-fold selective for NPP1 versus other ecto-nucleotidases. The most potent NPP1 inhibitors, compounds 4 and 6, were found to inhibit most selectively NPP1 activity with selectivity factors greater than 60. On the contrary, compound 5 blocked NTPDase1 selectively (P < 0.05), showing 5-fold selectivity. However, in the present study, no selective inhibitors could be identified for NTPDase2, NTPDase3, NTPDase8, NPP2 and NPP3, TNAP or eN. Compound 8 exhibited strong inhibition of NTPDase1–3 and NPP1 (see Fig. 7 and Tables 2–4), and may therefore prove useful as a non-specific pan-inhibitor of ATP hydrolysis.

In conclusion, an extended series of 12 POMs and 4 chalcogenide hexarhenium clusters was investigated in the present study for their potential to inhibit a broad range of ecto-nucleotidases, including human NTPDase1–3 and -8, human NPP1–3, human TNAP and rat eN. The anionic cluster complexes were found to inhibit the degradation of negatively charged nucleotides with distinct preferences for certain ecto-nucleotidase subfamilies and enzyme subtypes depending on the compound’s composition and structure. In general, they showed little or no activity on NTPDase8, NPP2 and NPP3, TNAP and eN, but on the other hand, they exhibited high inhibitory potential on NTPDase1–3 and NPP1 with nano- or even subnanomolar Ki values. Except for 5 and 8, all investigated compounds were found to block NPP1 activity selectively versus other ecto-nucleotidases, and compounds 4 and 6 in particular, were shown to be highly potent and selective NPP1 inhibitors with low nanomolar Ki values. They belong to the most potent NPP1 inhibitors described so far. Compound 5 was found to be a very potent and somewhat selective NTPDase1 inhibitor with a low nanomolar Ki value, which therefore belongs to the most potent human NTPDase1 inhibitors described to date. Finally, we performed studies to determine the mechanisms of inhibition of ecto-nucleotidases, in which the compounds behaved as non-competitive inhibitors of human NPPs and rat eN, but as competitive inhibitors of human TNAP. Our findings provide novel insights into the interactions of polyoxometalates and related rhenium complexes with ecto-nucleotidases and can therefore provide a solid basis for future research. Moreover the newly identified potent inhibitors of NTPDase1 and NPP1 can serve as novel pharmacological tools for investigating the pathophysiological and physiological roles of these ubiquitous and (patho) physiologically highly relevant ecto-enzymes. Acknowledgments J.S. was supported by grants from the Canadian Institutes of Health Research (CIHR) (MOP-102472, MOP-93683) and he was also a recipient of a ‘‘Chercheur National’’ award from the Fonds de recherche du Que´bec–Sante´ (FRQS).

References Table 5 Selectivity of investigated polyoxometalates (POMs) and chalcogenide hexarhenium cluster complexes. Compd

Most potent at

Second most potent at

Selectivity factor

1 2 3 4 5 6 7 8 9 10 11 12 13

Human Human Human Human Human Human Human Human Human Human Human Human Human

54 32 17 60 5 157 26 1 32 24 146 80 19

14

Human NPP1

15

Human NPP1

16

Human NPP1

Human NTPDase3 Human NTPDase3 Human NTPDase3 Human NTPDase2 Human NTPDase2 Human NTPDase3 Human NTPDase3 Human NPP1 Human NTPDase3 Human NTPDase3 Human TNAP Human TNAP Human TNAP, human NTPDase2 and human NTPDase3 Human TNAP, human NTPDase2 and human NTPDase3 Human TNAP, human NTPDase2 and human NTPDase3 Human NTPDase3

NPP1 NPP1 NPP1 NPP1 NTPDase1 NPP1 NPP1 NTPDase2 NPP1 NPP1 NPP1 NPP1 NPP1

49 34 6

Note. The Ki values of each compound were used to calculate the selectivity factor between the best and second best enzyme inhibitions. For the calculation of Ki values and selectivity determination see Section 2.9 and Tables 2–4.

[1] Hasenknopf B. Polyoxometalates: introduction to a class of inorganic compounds and their biomedical applications. Front Biosci 2005;10:275–87. [2] Pope MT, Kortz U. Polyoxometalates, in encyclopedia of inorganic and bioinorganic chemistry. Chichester: Wiley; 2012. [3] Ogata A, Mitsui S, Yanagie H, Kasano H, Hisa T, Yamase T, et al. A novel antitumor agent, polyoxomolybdate induces apoptotic cell death in AsPC-1 human pancreatic cancer cells. Biomed Pharmacother 2005;59:240–4. [4] Geisberger G, Paulus S, Carraro M, Bonchio M, Patzke GR. Synthesis, characterisation and cytotoxicity of polyoxometalate/carboxymethyl chitosan nanocomposites. Chem Eur J 2011;17:4619–25. [5] Cartuyvels E, Van Hecke K, Van Meervelt L, Go¨rller-Walrand C, Parac-Vogt TN. Structural characterization and reactivity of gamma-octamolybdate functionalized by proline. J Inorg Biochem 2008;102:1589–98. [6] Long D-L, Burkholder E, Cronin L. Polyoxometalate clusters, nanostructures and materials: from self assembly to designer materials and devices. Chem Soc Rev 2007;36:105–21. [7] Wang X, Liu J, Li J, Yang Y, Liu J, Li B, et al. Synthesis and antitumor activity of cyclopentadienyltitanium substituted polyoxotungstate [CoW11O39(CpTi)]7 (Cp = 5-C5H5). J Inorg Biochem 2003;94:279–84. [8] Mitsui S, Ogata A, Yanagie H, Kasano H, Hisa T, Yamase T, et al. Antitumor activity of polyoxomolybdate, [NH3Pri]6[Mo7O24]3H2O, against, human gastric cancer model. Biomed Pharmacother 2006;60:353–8. [9] Seko A, Yamase T, Yamashita K. Polyoxometalates as effective inhibitors for sialyl- and sulfotransferases. J Inorg Biochem 2009;103:1061–6. [10] Inoue M, Segawa K, Matsunaga S, Matsumoto N, Oda M, Yamase T. Antibacterial activity of highly negative charged polyoxotungstates, K27[KAs4W40O140] and K18[KSb9W21O86], and Keggin-structural polyoxotungstates against Helicobacter pylori. J Inorg Biochem 2005;99:1023–31. [11] Turner TL, Nguyen VH, McLauchlan CC, Dymon Z, Dorsey BM, Hooker JD, et al. Inhibitory effects of decavanadate on several enzymes and Leishmania tarentolae in vitro. J Inorg Biochem 2012;108:96–104.


G Model

BCP-12133; No. of Pages 11 S.-Y. Lee et al. / Biochemical Pharmacology xxx (2014) xxx–xxx [12] Qi Y, Xiang Y, Wang J, Qi Y, Li J, Niu J, et al. Inhibition of hepatitis C virus infection by polyoxometalates. Antiviral Res 2013;100:392–8. [13] Rhule JT, Hill CL, Judd DA, Schinazi RF. Polyoxometalates in medicine. Chem Rev 1998;98:327–58. [14] Nomiya K, Torii H, Hasegawa T, Nemoto Y, Nomura K, Hashino K, et al. Insulin mimetic effect of a tungstate cluster. Effect of oral administration of homopolyoxotungstates and vanadium-substituted polyoxotungstates on blood glucose level of STZ mice. J Inorg Biochem 2001;86:657–67. [15] Prudent R, Moucadel V, Laudet B, Barette C, Lafaneche`re L, Hasenknopf B, et al. Identification of polyoxometalates as nanomolar noncompetitive inhibitors of protein kinase CK2. Chem Biol 2008;15:683–92. [16] Mu¨ller CE, Iqbal J, Baqi Y, Zimmermann H, Ro¨llich A, Stephan H. Polyoxometalates—a new class of potent ecto-nucleoside triphosphate diphosphohydrolase (NTPDase) inhibitors. Bioorg Med Chem Lett 2006;16:5943–7. [17] Ko¨hler D, Eckle T, Faigle M, Grenz A, Mittelbronn M, Laucher S, et al. CD39/ ectonucleoside triphosphate diphosphohydrolase 1 provides myocardial protection during cardiac ischemia/reperfusion injury. Circulation 2007;116: 1784–94. [18] Wall MJ, Wigmore G, Lopata´r J, Frenguelli BG, Dale N. The novel NTPDase inhibitor sodium polyoxotungstate (POM-1) inhibits ATP breakdown but also blocks central synaptic transmission, an action independent of NTPDase inhibition. Neuropharmacology 2008;55:1251–8. [19] Sun X, Wu Y, Gao W, Enjyoji K, Csizmadia E, Mu¨ller CE, et al. CD39/ENTPD1 expression by CD4 + Foxp3+ regulatory T cells promotes hepatic metastatic tumor growth in mice. Gastroenterology 2010;139:1030–40. [20] Melani A, Corti F, Stephan H, Mu¨ller CE, Donati C, Bruni P, et al. Ecto-ATPase inhibition: ATP and adenosine release under physiological and ischemic in vivo conditions in the rat striatum. Exp Neurol 2012;233:193–204. [21] Stephan H, Kubeil M, Emmerling F, Mu¨ller CE. Polyoxometalates as versatile enzyme inhibitors. Eur J Inorg Chem 2013;2013:1585–94. [22] Kukulski F, Le´vesque SA, Lavoie EG, Lecka J, Bigonnesse F, Knowles AF, et al. Comparative hydrolysis of P2 receptor agonists by NTPDases 1, 2, 3 and 8. Purinergic Signal 2005;1:193–204. [23] Stefan C, Jansen S, Bollen M. Modulation of purinergic signaling by NPP-type ectophosphodiesterases. Purinergic Signal 2006;2:361–70. [24] Zalatan JG, Fenn TD, Herschlag D. Comparative enzymology in the alkaline phosphatase superfamily to determine the catalytic role of an active site metal ion. J Mol Biol 2009;384:1174–89. [25] Stra¨ter N. Ecto-50 -nucleotidase. Structure function relationships. Purinergic Signal 2006;2:343–50. [26] Zimmermann H, Zebisch M, Stra¨ter N. Cellular function and molecular structure of ecto-nucleotidases. Purinergic Signal 2012;5:437–502. [27] Yegutkin GG. Nucleotide- and nucleoside-converting ectoenzymes: important modulators of purinergic signalling cascade. Biochim Biophys Acta 2008;1783: 673–94. [28] Al-Rashida M, Iqbal J. Therapeutic potentials of ecto-nucleoside triphosphate diphosphohydrolase, pyrophosphatase/phosphodiesterase, ecto-5-nucleotidase, and alkaline. Med Res Rev 2013;34:703–43. [29] Goding JW, Grobben B, Slegers H. Physiological and pathophysiological functions of the ecto-nucleotide pyrophosphatase/phosphodiesterase family. Biochim Biophys Acta 2003;1638:1–19. [30] Baqi Y, Weyler S, Iqbal J, Zimmermann H, Mu¨ller CE. Structure–activity relationships of anthraquinone derivatives derived from bromaminic acid as inhibitors of ectonucleoside triphosphate diphosphohydrolases (E-NTPDases). Purinergic Signal 2009;5:91–106. [31] Freundlieb M, Zimmermann H, Mu¨ller CE. A new, sensitive ecto-50 -nucleotidase assay for compound screening. Anal Biochem 2014;446:53–8. [32] Spychala J. Tumor-promoting functions of adenosine. Pharmacol Ther 2000;87:161–73. [33] Sansom FM, Robson SC, Hartland EL. Possible effects of microbial ecto-nucleoside triphosphate diphosphohydrolases on host–pathogen interactions. Microbiol Mol Biol Rev 2008;72:765–81. [34] Sansom FM, Riedmaier P, Newton HJ, Dunstone MA, Mu¨ller CE, Stephan H, et al. Enzymatic properties of an ecto-nucleoside triphosphate diphosphohydrolase from Legionella pneumophila: substrate specificity and requirement for virulence. J Biol Chem 2008;283:12909–18. [35] Firon A, Dinis M, Raynal B, Poyart C, Trieu-Cuot P, Kaminski PA. Extracellular nucleotide catabolism by the group B Streptococcus ectonucleotidase NudP increases bacterial survival in blood. J Biol Chem 2014;289:5479–89. [36] Terkeltaub R. Physiologic and pathologic functions of the NPP nucleotide pyrophosphatase/phosphodiesterase family focusing on NPP1 in calcification. Purinergic Signal 2006;2:371–7. [37] Abate N, Chandalia M, Paola R Di, Foster DW, Grundy SM, Trischitta V. Mechanisms of disease: ectonucleotide pyrophosphatase phosphodiesterase 1 as a gatekeeper of insulin receptors. Nat Clin Pract Endocrinol Metab 2006;2:694–701. [38] Bageritz F, Puccio RM, Hovestadt V, Phillips E, Pankert T, Lohr J, et al. Stem cell characteristic in glioblastoma are maintained by the ecto-nucleotidase E-NPP1. Cell Death Differ 2014;21:929–40.

11

[39] Aerts I, Martin J-J, De Deyn PP, Van Ginniken C, Van Ostade X, Kockx M, et al. The expression of ecto-nucleotide pyrophosphatase/phosphodiesterase 1 (E-NPP1) is correlated with astrocytic tumor grade. Clin Neurol Neurosurg 2011;113:224–9. [40] Millań JL. Alkaline phosphatases: structure, substrate specificity and functional relatedness to other members of a large superfamily of enzymes. Purinerg Signal 2006;2:335–41. [41] Orimo H. The mechanism of mineralization and the role of alkaline phosphatase in health and disease. J Nippon Med Sch 2010;77:4–12. [42] Sergienko E, Su Y, Chan X, Brown B, Hurder A, Narisawa S, et al. Identification and characterization of novel tissue-nonspecific alkaline phosphatase inhibitors with diverse modes of action. J Biomol Screen 2009;14:824–37. [43] Sergienko E, Millań JL. High-throughput screening of tissue-nonspecific alkaline phosphatase for identification of effectors with diverse modes of action. Nat Protoc 2010;5:1431–9. [44] Raza R, Matin A, Sarwar S, Barsukova-Stuckart M, Ibrahim M, Kortz U, et al. Polyoxometalates as potent and selective inhibitors of alkaline phosphatases with profound anticancer and amoebicidal activities. Dalton Trans 2012;41: 14329–36. [45] Chung TDY, Sergienko E, Millań JL. Assay format as a critical success factor for identification of novel inhibitor chemotypes of tissue-nonspecific alkaline phosphatase from high-throughput screening. Molecules 2010;15: 3010–37. [46] Kozlenkov A, Le Du MH, Cuniasse P, Ny T, Hoylaerts MF, Millań JL. Residues determining the binding specificity of uncompetitive inhibitors to tissuenonspecific alkaline phosphatase. J Bone Miner Res 2004;19:1862–72. [47] Zebisch M, Krauss M, Scha¨fer P, Stra¨ter N. Crystallographic evidence for a domain motion in rat nucleoside triphosphate diphosphohydrolase (NTPDase) 1. J Mol Biol 2012;415:288–306. [48] Jiang ZG, Wu Y, Csizmadia E, Feldbru¨gge L, Enjyoji K, Tigges J, et al. Characterization of circulating microparticle-associated CD39 family ecto-nucleotidases in human plasma. Purinergic Signal Aug 2014 [Epub ahead of print]. [49] Yarovoi S, Mironov Y, Naumov D, Gatilov Y, Kozlova S, Kim S-J, et al. Octahedral hexahydroxo rhenium cluster complexes [Re6Q8(OH)6]4 (Q = S, Se): synthesis, structure, and properties. Eur J Inorg Chem 2005;8:3945–9. [50] Brylev KA, Mironov YV, Fedorov VE, Kim S-J, Pietzsch H-J, Stephan H, et al. A new hexanuclear rhenium cluster complex with six terminal acetate ligands: synthesis, structure, and properties of K4[Re6S8(CH3COO)6]8H2O. Inorg Chim Acta 2010;363:2686–91. [51] Brylev KA, Mironov YV, Kozlova SG, Fedorov VE, Kim S-J, Pietzsch H-J, et al. The first octahedral cluster complexes with terminal formate ligands: synthesis, structure, and properties of K4[Re6S8(HCOO)6] and Cs4[Re6S8(HCOO)6]. Inorg Chem 2009;48:2309–15. [52] Alizadeh MH, Harmalker SP, Jeannin Y, Martin-Frere J, Pope MT. A heteropolyanion with fivefold molecular symmetry that contains a nonlabile encapsulated sodium ion. The structure and chemistry of [NaP5W30O110]14. J Am Chem Soc 1985;107:2662–9. [53] Acerete R, Server-Carrio J, Vegas A, Martinez-Ripoll M. Synthesis and x-ray crystal structure determination of a novel chiral heteropolyanion: the 3:1 octadecatungstohexaphosphate. J Am Chem Soc 1990;112:9386–7. [54] Contant R, Klemperer WG, Yaghi O. Potassium octadecatungstodiphosphates(V) and related lacunary compounds. Inorganic syntheses, vol. 27. Hoboken, NJ: Wiley; 1990. [55] Kortz U, Jameson GB, Pope MT. Polyoxometalate diphosphate complexes. Folded macrocyclic dodecatungstates, [(O3PXPO3)4W12O36]16 (X = O, CH2). J Am Chem Soc 1994;116:2659–60. [56] Servos J, Reila¨nder H, Zimmermann H. Catalytically active soluble ecto-50 nucleotidase purified after heterologous expression as a tool for drug screening. Drug Dev Res 1998;276:269–76. [57] Lecka J, Gillerman I, Fausther M, Salem M, Munkonda MN, Brosseau J-P, et al. 8BuS-ATP derivatives as specific NTPDase1 inhibitors. Br J Pharmacol 2013;169: 179–96. [58] Gillerman I, Lecka J, Simhaev L, Munkonda MN, Fausther M, Martıń-Satue´ M, et al. 2-hexylthio-b,g-CH2-ATP is an effective and selective NTPDase2. J Med Chem 2014;57:5919–34. [59] Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein–dye binding. Anal Biochem 1976;72:248–54. [60] Baqi Y, Lee S-Y, Iqbal J, Ripphausen P, Lehr A, Scheiff AB, et al. Development of potent and selective inhibitors of ecto-50 -nucleotidase based on an anthraquinone scaffold. J Med Chem 2010;53:2076–86. [61] Cheng Y, Prusoff WH. Relationship between the inhibition constant (Ki) and the concentration of inhibitor which causes 50 percent inhibition (I50) of an enzymatic reaction. Biochem Pharmacol 1973;22:3099–108. [62] Copeland RA. Enzymes: a practical introduction to structure, mechanism, and data analysis. 2nd ed. New York, NY: Wiley; 2000. 286. [63] Fiene A, Baqi Y, Lecka J, Se´vigny J, Mu¨ller CE. Fluorescence polarization immunoassays for monitoring nucleoside triphosphate diphosphohydrolase (NTPDase) activity. Analyst Nov 2014 [Epub ahead of print].


Selective JAK inhibitors.

Selective serotonin reuptake inhibitors.

Selective serotonin reuptake inhibitors.

Selective serotonin reuptake inhibitors.

Identification of potent and selective MTH1 inhibitors.

Developing selective inhibitors of calpain.

SAH derived potent and selective EZH2 inhibitors.

Discovery of novel and selective SIRT6 inhibitors.

Progress toward JAK1-selective inhibitors.

Selective serotonin reuptake inhibitors for fibromyalgia syndrome.

Selective Phosphodiesterase 4B Inhibitors: A Review.

Histone H3 peptide based LSD1-selective inhibitors.

Sulfonate-containing thiiranes as selective gelatinase inhibitors.

Selective inhibitors of biosynthesis of aminergic neurotransmitters.

Selective serotonin reuptake inhibitors: depressing perioperative outcomes?

The selective effects of serotonin reuptake inhibitors.

Selective serotonin reuptake inhibitors in OCD.

Castanospermine-glucosides as selective disaccharidase inhibitors.

Fluorinated hydroxypiperidines as selective β-glucosidase inhibitors.

Synthesis and SAR study of potent and selective PI3Kδ inhibitors.

Resveratrol-salicylate derivatives as selective DNMT3 inhibitors and anticancer agents.

Selective serotonin reuptake inhibitors and perioperative bleeding in endoscopic dacryocystorhinostomy.

Selective and potent small-molecule inhibitors of PI3Ks.

Functionalized N,N-Diphenylamines as Potent and Selective EPAC2 Inhibitors.