Naunyn-Schmiedeberg's Arch Pharmacol (2016) 389:349–352 DOI 10.1007/s00210-015-1197-z

BRIEF COMMUNICATION

Inhibitors of membranous adenylyl cyclases with affinity for adenosine receptors Karl-Norbert Klotz 1 & Sonja Kachler 1

Received: 1 December 2015 / Accepted: 2 December 2015 / Published online: 14 December 2015 # Springer-Verlag Berlin Heidelberg 2015

Abstract Membrane-bound adenylyl cyclases constitute an interesting therapeutic target for various diseases that affect a large number of patients including asthma or congestive heart failure. Many inhibitors of adenylyl cyclases are competitive inhibitors at the ATP binding site and may, therefore, also interact with one or several of numerous ATP-binding proteins other than adenylyl cyclases. Several such inhibitors also show structural similarity to adenosine receptor ligands, providing a risk for side effects mediated by an unwanted interaction with these receptors. We have investigated a potential specific binding of four representative adenylyl cyclase inhibitors and found binding with pharmacologically relevant affinity to A 1 and A 2A receptors for NKY80 (2-amino-7-(2-furanyl)-7,8-dihydro-5(6H)-quinazolinone) and SQ22,536 (9-(tetrahydro-2-furanyl)-9H-purin-6-amine). These results underscore the importance to consider potential side effects mediated via adenosine receptors in the development of potent and specific inhibitors of adenylyl cyclases. Keywords Adenylyl cyclase . Inhibitor . Selectivity . Adenosine receptor

Abbreviations AR Adenosine receptor CCPA 2-Chloro-N6-cyclopentyladenosine HEMADO 2-Hexyn-1-yl-N6-methyladenosine

* Karl-Norbert Klotz [email protected] 1

Institut für Pharmakologie und Toxikologie, Universität Würzburg, Versbacher Str. 9, D-97078 Würzburg, Germany

NECA NB001 NKY80 SQ22,536 CHO cells GPCR mAC

Adenosine-5′-N-ethyluronamide 5-[[2-(6-amino-9H-purin-9-yl)ethyl] amino]-1-pentanol 2-Amino-7-(2-furanyl)-7,8-dihydro-5 (6H)-quinazolinone 9-(tetrahydro-2-furanyl)-9H-purin-6-amine Chinese hamster ovary cells G protein-coupled receptor Membrane-bound adenylyl cyclase

1. Introduction Membrane-bound adenylyl cyclases (mACs) play a central role in the signaling network of every cell. Their activity is regulated by G protein-coupled receptors (GPCRs) which interact with Gs or Gi in order to transmit a stimulatory or inhibitory signal to mACs, respectively. The nine know isoforms of mACs show cell-specific expression patterns in distinct cell types and play, therefore, specific roles in different tissues. Numerous studies including results from isoform-specific knockout mice suggested that mACs might be promising therapeutic targets in, e.g., analgesia, neurodegenerative diseases, asthma, and congestive heart failure (for review, see Pierre et al. 2009). Therefore, specific inhibition of individual mACs is of paramount interest for the development of pharmacological treatment options for such pathophysiological conditions. However, the development of potent and selective inhibitors poses several challenges and complications as most endeavors to develop such inhibitors were aimed at the catalytic site of the enzyme. For such compounds, a high risk for off-target effects must be taken into account due to the large number of ATP-binding proteins. In addition to inhibitors with close structural similarity to ATP, an assortment of purine-based mAC inhibitors with

350

some analogy to various adenosine receptor ligands are known (Seifert et al. 2012). This structural relationship of inhibitors like vidarabine (9-β-D-arabinofuranosyladenin), first introduced as an antiviral drug (Seifert 2014), suggests a potential for undesirable effects mediated by adenosine receptors. For fludarabine, which represents the cycotoxic 2-fluoro derivative of vidarabine, specific binding to the A1 adenosine receptor was previously shown (Jensen et al. 2012). The structurally related cytotoxic nucleosides cladribine and clofarabine bound to adenosine receptor subtypes as well and presented as A1 and A2A agonists (Jensen et al. 2012). The plentiful physiological effects mediated by adenosine receptors (Fredholm et al. 2001, 2011) constitute a significant risk for off-target effects for such compounds. The aim of this study was, therefore, to investigate a potential cross-reaction of a subclass of mAC inhibitors with adenosine receptors. From a comprehensive list of mAC inhibitors (Seifert et al. 2012), four compounds were selected based on their structural resemblance to known adenosine receptor ligands.

Material and methods Material The radioligands [3H]CCPA and [3H]NECA were from GE Healthcare, München, Germany, [3H]HEMADO was from Tocris Bioscience, Bristol, UK. [α- 32 P]ATP was from Hartmann Analytik, Braunschweig, Germany. NKY80 and SQ22,536 were from Merck, Darmstadt, Germany; all other chemical compounds including vidarabine were from Sigma-Aldrich, München, Germany. Media and fetal calf serum for cell culture were from PanSystems, Aidenbach, Germany; penicillin (100 U/ml), streptomycin (100 μg/ml), L-glutamine, and G-418 were purchased from Gibco-Life Technologies, Eggenstein, Germany. All other materials were from sources as described earlier (Klotz et al. 1998, 2007). Methods All methods used for the pharmacological characterization of compounds were described in detail in previous studies. Human adenosine receptors were characterized in membranes from CHO cells stably transfected with the individual subtypes (Klotz et al. 1998). The radioligands used in competition experiments were 1 nM [3H]CCPA for A1, 10 nM [3H]NECA for A2A, and 1 nM [3H]HEMADO for A3 receptors (Klotz et al. 1998, 2007). CHO cells were grown at 37 °C in 5 % CO2/95 % air in Dulbecco’s modified Eagles medium, supplemented with nutrient mixture F12 (DMEM/F12) without nucleosides, including 10 % fetal calf serum, penicillin (100 U/ml), streptomycin (100 μg/ml), L-glutamine (2 mM), and Geneticin (G-418, 0.2 mg/ml) (Klotz et al. 1998).

Naunyn-Schmiedeberg's Arch Pharmacol (2016) 389:349–352

Membranes for radioligand binding were prepared using a two-step centrifugation protocol while a crude membrane fraction prepared with just one high-speed centrifugation step was used for membranes prepared for measurement of adenylyl cyclase activity (for details, see Klotz et al. 1998). Membranes for radioligand binding were frozen in liquid nitrogen and stored at −80 °C while membranes for adenylyl cyclase assays were used immediately according to previously published procedures (Klotz et al. 1998). Adenylyl cyclase activity was used to detect affinity of the tested compounds for A2B adenosine receptors. None of the compounds showed detectable interaction with the A2B subtype (not shown), thus, only binding data for A1, A2A, and A3 receptors are reported in Table 1.

Results and discussion In Fig. 1, the structures of four mAC inhibitors are compared with the structure of adenosine. Vidarabine is a stereoisomer of adenosine and seems to have the closest structural relationship to adenosine of all compounds investigated in this study. From animal studies, it was concluded that vidarabine might be an AC5 inhibitor and useful for the treatment of heart failure (Vatner et al. 2013) and cancer (De Lorenzo et al. 2013). However, in a recent commentary, it was discussed that vidarabine is not a potent mAC inhibitor, and its selectivity was questioned, in particular as it does not discriminate between AC5 and AC6 (Seifert 2014). Despite the apparent similarity to adenosine, vidarabine showed no measureable affinity for any of the receptor subtypes characterized in radioligand binding experiments (Table 1). The 2′- and 3′-hydroxy groups of the ribose in the adenosine molecule play an important role in ligand-receptor interaction and receptor activation (Lebon et al. 2011; Dal Ben et al. 2014) and changing the stereochemistry of the 2′-hydroxy group is not well tolerated (Jensen et al. 2012). Removal of both the 2′- and 3′-hydroxy groups turns the resulting compound into an antagonist (Lohse et al. 1988). This is in line with the current study where SQ22,536, a compound lacking these hydroxyl groups, presents as an antagonist at the A1 and the A2A receptor (Table 1). The compound shows compelling affinity for the A2A receptor in the low micromolar range similar to the Table 1

NKY80 SQ22,536 Vidarabine NB001

Binding data at adenosine receptor subtypes A1

A2A

A3

23,900 (13,600–42,100) 13,800 (8530–22,200) >100,000 >100,000

17,300 (9530–31,400) 3020 (2600–3520) >100,000 >100,000

>60,000 >100,000 >100,000 >100,000

Ki values are given in nM as geometric means of at least three independent experiments with 95 % confidence limits in parentheses

Naunyn-Schmiedeberg's Arch Pharmacol (2016) 389:349–352

351

Depending on the selectivity profile for mACs and adenosine receptor subtypes, the combined response might be beneficial or it might be characterized by more pronounced side effects. For the interpretation of in vivo results of mAC inhibitors, it seems crucial to realize their distinct pharmacological interaction with adenosine receptors. adenosine

SQ22,536

vidarabine

Conclusion

NKY80

NB001

Fig. 1 Structures of adenosine and characterized compounds

clinically used theophylline (Klotz et al. 1998) and, most importantly, similar to the IC50 for the inhibition of AC5 by SQ22,536 (Seifert et al. 2012). Therefore, clinically useful effects of AC5 inhibition would inevitably be associated with effects mediated through A2A adenosine receptors. NB001 represents a potential mAC inhibitor with poorly characterized mechanism of action leading to a reduction of cellular cAMP accumulation. It seems likely that the reduction of cAMP levels via AC1 inhibition is the consequence of indirect effects and is possibly achieved through several mechanisms (Seifert et al. 2012; Brand et al. 2013). Blockade of an A2 receptor would be one such mechanism compatible with the effect on cellular cAMP levels. However, our data show that NB001 does not bind to any subtype of adenosine receptor in concentrations up to 100 μM (Table 1). Many different heterocyclic cores apart from the classical xanthine scaffold (Klotz 2000) were successfully used for the development of adenosine receptor antagonists over the last decades (e.g., Soudijn et al. 2003; Gillespie et al. 2009; Inamdar et al. 2013; Matos et al. 2015). This led us to examine whether the quinazolinone NKY80 shows specific binding to adenosine receptors. It turned out that this compound interacts with micromolar affinity with A1 and A2A receptors, whereas no measurable binding to the A3 subtype was detected (Table 1). Although the affinity for A1 and A2A receptors is quite moderate, it seems to be of significance as the lowest reported IC50 values for adenylyl cyclase inhibition are in the same order of magnitude (AC5: 8300 nM; Seifert et al. 2012). Cross reaction of potential adenylyl cyclase inhibitors with adenosine receptors will directly affect the magnitude of the response caused by mAC inhibition. The blockade of an A2 receptor will amplify the reduction in cAMP levels of the cell while antagonism at A1 or A3 receptors will counteract the effect of mAC inhibition. Both mACs and adenosine receptor subtypes display a tissue-specific distribution making predictions about the outcome of simultaneous cyclase inhibition and blockade of a given adenosine receptor difficult.

In addition to the potential interaction of mAC inhibitors with ATP-binding proteins other than specific adenylyl cyclases, compounds with a purine scaffold or even vaguely related structures might cross-react with adenosine receptors resulting in a hitherto unnoticed cause for undesirable effects of such potential drugs. Our results document the importance of careful screening of novel mAC inhibitors not only for mAC selectivity but emphasize the need to consider effects mediated by additional targets like adenosine receptors.

References Brand CS, Hocker HJ, Gorfe AA, Cavasetto CN, Dessauer CW (2013) Isoform selectivity of adenylyl cyclase inhibitors: characterization of known and novel compounds. J Pharmacol Exp Ther 347:265–275 Dal Ben D, Buccioni M, Lambertucci C, Kachler S, Falgner N, Marucci G, Thomas A, Cristalli G, Volpini R, Klotz K-N (2014) Different efficacy of adenosine and NECA derivatives at the human A3 adenosine receptor: Insight into the receptor activation switch. 87:321–331 De Lorenzo MS, Chen W, Baljinnyam E, Carlini MJ, La Perle K, Bishop SP, Wagner TE, Rabson AB, Vatner DE, Puricelli LI, Vatner SF (2013) Reduced malignancy as a mechanism for longevity in mice with adenylyl cyclase type 5 disruption. Aging Cell 13:102–110 Fredholm BB, IJzerman AP, Jacobson KA, Klotz K-N, Linden J (2001) International union of pharmacology. XXV. Nomenclature and classification of adenosine receptors. Pharmacol Rev 53:527–552 Fredholm BB, IJzerman AP, Jacobson KA, Linden J, Muller CE (2011) International union of basic and clinical pharmacology. LXXXI. Nomenclature and classification of adenosine receptors–an update. Pharmacol Rev 63:1–34 Gillespie RJ, Bamford SJ, Gaur S, Jordan AM, Lerpiniere J, Mansell HL, Stratton GC (2009) Antagonists of the human A2A receptor. Part 5: highly bio-available pyrimidine-4-carboxamides. Bioorg Med Chem Lett 19:2664–2667 Inamdar GS, Pandya AN, Thakar HM, Sudarsanam V, Kachler S, Sabbadin D, Moro S, Klotz K-N, Vasu KK (2013) New insight into adenosine receptors selectivity derived from a novel series of [5substituted-4-phenyl-1,3-thiazol-2-yl] benzamides and furamides. Eur J Med Chem 63:924–934 Jensen K, Johnson LA, Jacobson P, Kirstein M, Lamba J, Klotz K-N (2012) Cytotoxic purine nucleoside analogues bind to A1, A2A and A3 receptors. Naunyn-Schmiedeberg’s Arch Pharmacol 385:519–525 Klotz K-N (2000) Adenosine receptors and their ligands. NaunynSchmiedeberg’s Arch Pharmacol 362:382–391 Klotz K-N, Hessling J, Hegler J, Owman C, Kull B, Fredholm BB, Lohse MJ (1998) Comparative pharmacology of human adenosine

352 receptor subtypes - characterization of stably transfected receptors in CHO cells. Naunyn Schmiedeberg’s Arch Pharmacol 357:1–9 Klotz K-N, Falgner N, Kachler S, Lambertucci C, Vittori S, Volpini R, Cristalli G (2007) [3H]HEMADO—a novel tritiated agonist selective for the human adenosine A3 receptor. Eur J Pharmacol 556:14–18 Lebon G, Warne T, Edwards PC, Bennett K, Langmead CJ, Leslie AGW, Tate CG (2011) Agonist-bound adenosine A2A receptor structures reveal common features of GPCR activation. Nature 474:524–525 Lohse MJ, Klotz K-N, Diekmann E, Friedrich K, Schwabe U (1988) 2′,3′Dideoxy-N6-cyclohexyladenosine: an adenosine derivative with antagonist properties at adenosine receptors. Eur J Pharmacol 156:157–160 Matos MJ, Vilar S, Kachler S, Celeiro M, Vazquez-Rodriguez S, Santana L, Uriarte E, Hripcsak G, Borges F, Klotz K-N (2015) Development of novel adenosine receptor ligands based on the 3-amidocoumarin scaffold. Bioorg Chem 61:1–6

Naunyn-Schmiedeberg's Arch Pharmacol (2016) 389:349–352 Pierre S, Eschenhagen T, Geisslinger G, Scholich K (2009) Capturing adenylyl cyclases as potential drug targets. Nat Rev Drug Discov 8:321–335 Seifert R (2014) Vidarabine is neither a potent nor a selective AC5 inhibitor. Biochem Pharmacol 87:543–546 Seifert R, Lushington GH, Mou T-C, Gille A, Sprang SR (2012) Inhibitors of membranous adenylyl cyclases. Trends Pharmacol Sci 33:64–78 Soudijn W, van Wijngaarden I, IJzerman AP (2003) Medicinal chemistry of adenosine A1 receptor ligands. Curr Top Med Chem 3:355–367 Vatner SF, Park M, Yan L, Lee GJ, Lai L, Iwatsubo K, Ishikawa Y, Pessin J, Vatner DE (2013) Adenylyl cyclase type 5 in cardiac disease, metabolism, and aging. Am J Physiol Heart Circ Physiol 305:H1–H8