Discovery of benzothiazolylquinoline conjugates as novel human A3 receptor antagonists: biological evaluations and molecular docking studies

Adenosine is known as an endogenous purine nucleoside and it modulates a wide variety of physiological responses by interacting with adenosine receptors. Among the four adenosine receptor subtypes, the A3 receptor is of major interest in this study as it is overexpressed in some cancer cell lines. Herein, we have highlighted the strategy of designing the hA3 receptor targeted novel benzothiazolylquinoline scaffolds. The radioligand binding data of the reported compounds are rationalized with the molecular docking results. Compound 6a showed best potency and selectivity at hA3 among other adenosine receptors.


Introduction
Adenosine is an endogenous purine nucleoside which regulates many physiological functions through the activation of four specific receptor subtypes, classified as A 1 , A 2A , A 2B and A 3 adenosine receptors (AR), belonging to the family of G-proteincoupled receptors [1]. These receptors are widely distributed in mammalian tissues. The A 3 AR subtype is the most recently characterized member of the family which was first cloned from a rat testis cDNA library [2] and is still undergoing intensive pharmacological characterization. The A 3 AR subtype is implicated in various pathological conditions such as cardiac and cerebral ischaemia, neurodegenerative diseases as well as inflammatory pathologies including rheumatoid arthritis and asthma [3]. Furthermore, A 3  neoplastic cells including HL-60 leukaemia, human Jurkat T-cell lymphoma, astrocytoma, A378 melanoma, B16-F10 and solid tumour (e.g. a two-threefold increase in colon carcinomas), while low or almost no receptor expression was found in normal cells [4,5]. Similar results were found in studies of the receptor expression levels in tumour tissues derived from patients with colon, breast, small cell lung, pancreatic and hepatocellular carcinomas and melanoma in direct comparison with adjacent body normal tissues [6][7][8][9][10][11][12][13][14][15]. Higher A 3 AR expression in the tumour versus adjacent non-neoplastic tissue was further confirmed by reverse transcription-PCR analysis of colon and breast carcinoma. Protein analysis of A 3 AR expression in fresh tumours derived from colon (n = 40) or breast (n = 17) revealed 61% and 78% higher expression in the tumour than adjacent normal tissue, respectively [16]. Thus the A 3 receptor could be a prospective therapeutic target and biological predictive marker in cancer therapy. The high A 3 AR expression level in the tumour tissues was associated with elevated nuclear factor κB and cyclin D1 levels [16]. High A 3 AR mRNA expression was also exhibited in other solid tumour types. Mechanistic studies demonstrated that A 3 AR activation by synthetic agonists or antagonists induces down-regulation of key cell growth-regulatory proteins including cyclin D1 and nuclear factor κB [10,17,18]. Hence, discovery of selective A 3 AR targeting ligands has been a great challenge in last two decades. Moreover, A 3 AR antagonists research not only aids in the development of therapeutic agents but also in the development of diagnostic agents [2,19]. Nowadays, the diagnostic approaches have been significantly developed by using fluorescently labelled pharmacophores.
Benzothiazole possesses several biological activities such as anti-inflammatory, antimicrobial, anti-HIV, anticancer and amyloid marker [30][31][32][33][34][35][36]. These scaffolds are also able to arrest metal promoted amyloid fibril build-up [37]. Likewise, 8-hydroxy quinoline has also been developed as potent bioactive scaffold [38]. Herein, we have highlighted the strategy of designing the novel 2-(2hydroxyphenyl)benzothiazole (HBT) scaffolds having dual properties (pharmacophore and fluorophore) via molecular simplification followed by a molecular docking approach (figure 1). The designed molecules have already shown potency and selectivity in hA 3 AR overexpressed cancer cell lines than normal cell lines [38]. In our earlier report, the cellular localization was also observed using those scaffolds. To justify the molecular pathway of these drugs, we have initiated the molecular docking approach as well as radioligand binding study at hA 3 AR.   we have simplified the structure of the tricyclic A 3 antagonist (triazoloquinoxaline) into bicyclic motif (2-benzothiazolyl phenol). In the course of structure design, we have attached the bicyclic fluorescent pharmacophore (2-benzothiazolyl phenol) with another bioactive 8-hydroxy quinolone unit through a linker to enhance the A 3 binding ability (figure 2).

Synthesis
In our earlier report, a series of 2-(2 -hydroxyphenyl)benzothiazolylquinoline scaffolds were prepared in conventional way and also under microwave condition in one pot sequence (scheme 1) [38]. We followed the earlier reported green method to synthesize the following scaffolds and then characterized

Molecular docking studies 2.3.1. Homology modelling
The crystallographic structure of hA 2A AR complexed with ZM-241385 as a high affinity antagonist (PDB code: 3EML) [39]. It was already being used to build up a homology model of the hA 3 AR by our group [40]. Considering the high resolution (2.6 Å) and accuracy of the structure of hA 2A AR, 3EML was selected as a template by various research groups [41,42]. MODELLER 9.11 was used to perform the homology modelling [43][44][45][46] and the quality of the model was evaluated using the Ramachandran plot. Subsequently, the prediction ability of the constructed hA 3 AR homology model was evaluated in the molecular docking experiments using the GLIDE tool from Schrödinger maestro.

Molecular docking
Our approach was to select a new class of hA 3 AR targeting compounds from a group of hypothetically designed 2-phenylbenzothiazole (HBT)-based scaffolds. We performed molecular docking of 12 different HBT-based ligands using the GLIDE tool from Schrödinger maestro to identify the hypothetical binding mode at the hA 3 AR. Finally, we have identified the best inhibitor for targeted hA 3 AR from GLIDE scores (  All the newly synthesized benzothiazolylquinoline scaffolds were docked into the orthosteric transmembrane-binding cavities of hA 3 AR. From the ligand docking, we have inferred that out of 12 synthesized scaffolds, compound 6a displayed the best GLIDE score with the lowest binding energy. In figure 4, the hypothetical binding pose of compound 6a is clearly observed at the hA 3 AR. In particular, the most prominent aromatic π-π stacking interactions are established between Phe 168 and ligands and it was anchored properly within the binding cleft. Moreover, a strong hydrogen bond with Phe 168 was also appeared within the binding pocket. Likewise, a strong π-π stacking interaction was observed between Tyr 265 and compound 6f. Compound 6g also formed a hydrogen bond with Val 169 (electronic supplementary material, figure S1)    There is no suitable radioligand for hA 2B AR found and hence the antagonists activity was determined in adenylyl cyclase experiments in CHO cells expressing the hA 2B AR [47,48]. K i (dissociation constant) value of the data was calculated using the Cheng and Prusoff equation [49], with geometric means of at least three experiments including 95% confidence intervals. From table 2, it was observed that most of the compounds exhibited a K i value at hA 3 in the range of 2-6 µM. Compound 6c and 6i showed least binding efficacy with hA 3 which is clearly rationalized with their binding energy profile (GLIDE score). Compound 6a exhibited the most binding potency at hA 3 in the 2.6 µM range. This compound has also shown 38-fold and 11-fold more selective potency at hA 3 than hA 1 and hA 2A , respectively. While increasing the length of the -CH 2linker from 2 to 3, binding potency of the compound 6e at hA 3 is reduced to 3.8 µM. Selectivity of this compound at hA 3 with respect to hA1 and hA 2 has also been reduced to some extent. Consequently, compound 6i showed much less potency and selectivity in hA 3 . Compounds (6f-h and 6j-l) having electronegative groups and lengthy linkers (n = 3, 4) showed good potency and selectivity at hA 3 . It was also observed that compound 6b having two electronegative bromine groups with a small linker (n = 2) exhibited good potency (3.2 µM) and selectivity at hA 3 .

Adenylyl cyclase activity
The potency of antagonists at the A 2B AR was determined in adenylyl cyclase experiments [47][48][49][50]. For the measurement of adenylyl cyclase activity, only one high speed centrifugation of the homogenate was used. The resulting crude membrane pellet was resuspended in 50 mM Tris-HCl, pH 7.4 and immediately used for the cyclase assay.

Data analysis
Inhibitory binding constants, K i , were calculated from the IC 50 values according to the Cheng and Prusoff equation K i = IC 50 /(1 + [C*]/KD*), where [C*] is the concentration of the radioligand and KD* its dissociation constant [49,50]. A weighted nonlinear least-squares curve fitting program LIGAND was also used for computer analysis of inhibition experiments. Potency values (IC 50 ) obtained in cyclic AMP assays were calculated by nonlinear regression analysis using the equation for a sigmoid concentrationresponse curve (Graph Pad Prism, San Diego, CA, USA). All experimental data are expressed as geometric mean with 95% confidence limits in parentheses of three or four independent experiments performed in duplicate.

Conclusion
In summary, we have designed a class of benzothiazolylquinoline scaffolds for the hA 3 target. Molecular simplification and molecular docking approach using the GLIDE tool from Schrödinger maestro has been employed for the design of these drugs. The effective binding modes of the scaffolds with the receptor binding sites were clearly explained. In addition, we have also performed radioligand binding assay of these scaffolds at hA 1 AR, hA 2A AR, hA 2B AR and hA 3 AR. We observed that compound 6a exhibited maximum potency and selectivity in hA 3 AR with respect to hA 1 AR, hA 2A AR and hA 2B AR which is rationalized with a docking study. Finally, it was concluded that these cytotoxic molecules are selectively targeting to the hA 3 AR.
Ethics. The department was ethically approved by UGC. Data accessibility. Molecular docking images of selected compounds can be found in the electronic supplementary material.