Fluorinated Adenosine A2A Receptor Antagonists Inspired by Preladenant as Potential Cancer Immunotherapeutics

Antagonism of the adenosine A2A receptor on T cells blocks the hypoxia-adenosinergic pathway to promote tumor rejection. Using an in vivo immunoassay based on the Concanavalin A mouse model, a series of A2A antagonists were studied and identified preladenant as a potent lead compound for development. Molecular modeling was employed to assist drug design and subsequent synthesis of analogs and those of tozadenant, including fluorinated polyethylene glycol PEGylated derivatives. The efficacy of the analogs was evaluated using two in vitro functional bioassays, and compound 29, a fluorinated triethylene glycol derivative of preladenant, was confirmed as a potential immunotherapeutic agent.


Synthesis of phenyl (4-hydroxy-7-morpholinobenzo[d]thiazol-2-yl)carbamate (41).[16]
To a solution of compound 38 (0.6 g, 1.48 mmol) in dichloromethane (31.2 mL) was added BBr 3 (10.6 mL, 1 M solution in dichloromethane) dropwise at -78 °C. The mixture was allowed to warm to room temperature and stirred overnight. Water (20 mL) was added to the mixture followed by NH 4 OH until pH = 7.0 was achieved. The mixture was extracted with dichloromethane (3 x 20 mL) and the extracts condensed in vacuo, then purified using flash

Induction of liver injury
Mice were injected intravenously with Con A (20 mg kg -1 ) in sterile PBS, and serum samples were taken or mice euthanized at indicated time points. Some mice were co-injected intraperitoneally with CGS21680 (2 mg kg -1 ), 2 (2 mg kg -1 ), 3 (2 mg kg -1 ) and 4 (2 mg kg -1 ) separately just before treatment with Con A. The magnitude or liver damage was evaluated by serum alanine aminotransferase (ALT) levels and liver tissue histology.

Measuring functionality of A 2A R antagonism by cAMP assay
Stimulation of intracellular cAMP production and measurement of cAMP levels were performed as described previously.
[18b] Lymphocytes were isolated from the spleen of C57/BL6 mice and suspended at a concentration of 400,000 cells/well, and then treated with 1 µM CGS 21680 6 (A 2A R-specific agonist; from Tocris, Ellisville, MO) with or without KW-6002 2, KW-PEG 3, preladenant 4, tozadenant 5a and their PEGylated analogs 27-31 and 40 (at a concentration of 1 mM/mL). The cells were incubated for 15 min at 37 °C, and the reaction was stopped by addition of 1 N hydrochloric acid. Levels of cAMP were determined by ELISA (Amersham Biosciences, Buckinghamshire, UK). All treatment groups were performed in triplicate.

Cytokine release assay
Splenocytes (2 x 10 6 /mL) were isolated from the spleen of C57/BL6 mice and activated with 0.1 µg/mL CD3 mAb to induce production of IFN-gamma. Immediately following the addition of mAb-CD3, the cells were treated with or without 1 µM CGS 21680 6 agonist and 1 uM KW-6002 2, KW-PEG 3, Preladenant 4 and compound 29. Supernatants were collected after 24 h and levels of IFN-gamma were assayed by ELISA (Amersham Biosciences, Buckinghamshire, UK) using paired mAb and standard purchased from BD Pharmingen. All treatment groups were performed in triplicate.

Homology modeling
The homology model from a previous report was employed for the docking study.
[6] It was constructed via YASARA (Yet Another Scientific Artificial Reality Application) based on the crystal structure of A2AR in complex with ZM241385 1 (PDB 3EML)[10] to fix the second extracellular loop (ECL2) residues Gln148 to Ser156. Residues were missing due to weak experimental electron density in that region. The quality of the homology model was examined by PROCHECK and verified as appropriate.

Computational studies
All calculations performed in this work were carried out on two Cooler Master Centurion 5 (Intel Core-i7 Quad CPU Q6600 @ 3.33GHz) operating systems running Maestro 10.4 (Schrödinger, LLC, New York, NY, 2015), ChemBioDraw® Ultra 14.0 (PerkinElmer) and YASARA. All pictures presented in this study were generated by Maestro and YASARA.

Molecule preparation
The

Aqueous pH 7.4 Solubility
Compounds were dried down from 10 mM DMSO solutions using centrifugal evaporation techniques. Phosphate buffer (0.1 M pH 7.4) was added and StirStix inserted in the glass vials.
Agitation was then performed at a constant temperature of 25°C for 20-24 hours. The vials were then subjected to double centrifugation (with a tip wash in between) to eliminate any residues of the dried compound. The solutions were then diluted and quantitated using LC/MS/MS.

Log D 7.4
A shake-flask method was employed to determine octanol-water distribution coefficients at pH 7.4 (Log D 7.4 ). The aqueous solution used was 10 mM sodium phosphate pH 7.4 buffer. The method was validated for Log D 7.4 ranging from -2 to 5.0.

Human Plasma Protein Binding (PPB)
PPB was determined using equilibrium dialysis (RED device) to separate free from bound compound. The amount of compound in plasma (10 µM initial concentration) and in dialysis buffer (pH 7.4 phosphate buffer) was measured by LC-MS/MS after equilibration at 37°C in a dialysis chamber. The fraction unbound (fu) is reported.

Intrinsic clearance
In vitro intrinsic clearance was determined from human liver microsomes or rat hepatocytes using standard methods. Following incubation and preparation, the samples were analyzed using LC/MS/MS. Refined data were uploaded to IBIS and displayed as CL int (intrinsic clearance) in µl/min/mg or µl/min/1 million cells.