PII S0024-3205(97)00690-5
PHARMACOLOGY LETTERS
Accelerated Communication
1Medical Department, Brookhaven National Laboratory, Upton, NY 11973 and 2School of Pharmacy and Department of Molecular and Cell Biology, U-92, University of Connecticut 372 Fairfield Road, Storrs, CT 06269-2092 (Submitted March 24, 1997; accepted April 17, 1997; received in final form June 23, 1997)
Introduction
We recently synthesized and labeled with iodine-123 the novel compound N-(piperidin-1-yl)-5-(4-iodophenyl)-1-(2,4-dichlorophenyl)-4-methyl-1H-pyrazole-3-carboxamide (AM251), which is an analog of the recently described cannabinoid CB1 receptor antagonist, SR141716A, containing a 4-iodophenyl group in place of 4-chlorophenyl (1). [ 123I]AM251 bound with nanomolar affinity to binding sites in mouse hippocampal membranes, in a manner which was inhibited by Delta9-tetrahydrocannabinol (THC) and by SR141716A (2). Furthermore, [ 123I]AM251 accumulated in mouse brain following intravenous administration, with peak uptake of about 1% of the administered activity at 2 h and a subsequent half clearance time of about 8 h. Coinjection of SR141716A inhibited the in vivo brain binding of [ 123I]AM251 but did not alter the concentration of radioactivity in other major organs. These observations (2) indicated that [ 123I]AM251 binds to the murine cannabinoid CB1 receptor both in vitro and in vivo. The present studies were undertaken to further characterize the in vitro binding of [ 123I]AM251. In particular, we wished to determine whether the labeled antagonist [123I]AM251 recognized the same binding site as the commercially available labeled cannabinoid receptor agonist [3H]CP 55,940 (3).
Methods
Animals. Mice (male Swiss-Webster, 25-35g) and rats (male Sprague-Dawley, 250-300g) were obtained from Taconic Farms, NY.
Preparation of [ 123I]AM251. Labeling in good yield (62%) and radiochemical purity (> 95%), and high specific activity (> 2500 Ci/mmol) was achieved by an iododestannylation reaction using the tributyltin precursor, no carrier added I-123 iodide, and chloramine-T, as previously described (1).
Equilibrium Binding Experiments. Brain membranes (P2 fraction) for brain regional binding experiments were prepared as previously described (2). Bmax (pmol/g tissue wet wt) and Kd values were determined by Scatchard analysis of experiments where the amount of non-radioactive ligand (AM251 or CP 55,940) was varied in the presence of a fixed activity of [ 123I]AM251 or [ 3H]CP 55,940.
For determination of IC50 values and dissociation kinetics, mouse cerebellum was homogenized in 20 mL ice-cold medium containing 1% bovine serum albumin and 50 mM tris.HCl, 5 mM MgCl2 and 1 mM EDTA, pH 7.4. A hand-held Tissue-Tearor homogenizer was used on setting 5 for 1 min. Incubation tubes, which contained 0.25 mL of homogenate in a total volume of 0.51 mL, were held at 30 ° for 2 h, in a shaking water bath. The radioligand concentrations of [ 123I]AM251 and [ 3H]CP 55,940 were routinely 1 nM and 0.5 nM, respectively. Incubations were terminated by addition of 5 mL of 50 mM tris HCl, pH 7.4 containing 0.25% bovine serum albumin and filtration through GF/B filters using a Brandel cell harvester. The filters were washed a further 3 times with the same buffer, and counted for 123I or 3H. Ki values were calculated using the Cheng-Prusoff equation.
Dissociation Experiments. Cerebellar homogenates were incubated for 2h at 30° in a final volume of 0.51 mL as described above. To each tube was then added 5 mL of incubation medium, containing 20 µM CP 55,940 or SR141716A where indicated. The final incubation mixtures were filtered as described above at times from 0 to 75 minutes.
Autoradiographic Experiments. After pre-soaking in buffer (1% bovine serum albumin and 50 mM tris-HCl, 5 mM MgCl2 and 1 mM EDTA, pH 7.4) for 5 min, slide mounted sections of rat brains (20 µm) were incubated at room temperature for 1 h in the same buffer containing 1 nM [ 123I]AM251 (approximately 1 µCi/mL). Slides were then washed briefly with buffer to remove adherent radioligand, and then incubated for 2 x 30 min in fresh ice-cold buffer. Following a brief water rinse they were air dried and apposed to Amersham MP autoradiography film overnight. Films were developed using a clinical X-ray film processor.
Results
Brain Regional Binding of Radioligands. We conducted homogenate binding experiments using washed membranes from three brain regions with [3H]CP 55,940 and [123I]AM251 under identical conditions (Table 1). Each tissue exhibited a lower Kd value for [123I]AM251 (range 0.23-0.62 nM) than for [3H]CP 55,940 (range 1.3-4.0 nM), while a similar Kd value for each radioligand was found across the 3 tissues examined. Similar Bmax values (range 15-41 pmol/g wet wt) were obtained in each tissue regardless of the radioligand used.
Dissociation Experiments. Subsequently to conducting experiments with washed brain membranes, we found greater specific/non-specific binding ratios for [ 123I]AM251 when using crude homogenates. Subsequent binding experiments therefore employed mouse cerebellar homogenates. Following dilution of incubation mixtures with a 10-fold excess of medium, the apparent dissociation half-time for [ 123I]AM251 was >75 min. However, the presence of a 20 µM concentration of either SR141716A or CP 55,940 decreased the dissociation half-time to about 10 min (Fig 1).
Fig 1 Semilogarithmic plot of dissociation of [123I]AM251 from binding sites in cerebellar homogenate. (filled square) control binding; (filled triangle) in the presence of 20 µM CP 55,940; (inverted filled triangle) in the presence of 20 µM SR141716A.
Fig 2 Representative binding experiments using [123I]AM251 and [3H]CP 55,940 as radioligand and crude cerebellar homogenate as the source of cannabinoid receptor.
Fig 3 Examples of in vitro [123I]AM251 binding to slide mounted sagittal sections rat brain in three separate experiments (Panels A, B and C-E, respectively). Panel A shows binding in the hippocampus, striatum and molecular layer of the cerebellum. Note also heterogeneous binding in the cerebral cortex. Panel B is a magnified image of a section showing (left to right) substantia nigra, hippocampus, entopeduncular nucleus and globus pallidus. Panels C, D and E show binding in the cerebellum in a control section (C) and in the presence of 30 µM CP55,940 (D) or 30 µM SR141716A (E).
Determination of IC50 Values. Using mouse cerebellar homogenates, we examined the abilities of (-)-Delta9-THC, methanandamide, SR 141716A (4), WIN 55,212-2 and CP 55,940 to inhibit the binding of [ 123I]AM251 (Fig 2). These compounds were chosen to represent the classes of compounds known to bind to cannabinoid receptors: the prototypic ligand (5), an anandamide analog with reduced rate of metabolism (6), a diarylpyrazole (4), an aminoalkylindole (7), and a high affinity non-classical cannabinoid (8,9). The increasing rank order of affinity (IC50 values given in Table 2) was methanandamide < THC < WIN 55,212-2 < CP 55,940 < SR141716A. Using identical incubation conditions, we examined the same compounds with the agonist tritiated CP 55,940 as radioligand. The increasing rank order of affinity was THC < methanandamide < WIN 55,212-2 < SR 141716A < CP 55,940. Thus although all the compounds are fully competitive with apparent Hill slopes of about unity, SR141716A appears to have higher affinity when competing against its congener AM251 than when competing against the classical cannabinoid CP 55,940.
Autoradiographic Studies in Rat Brain. High binding densities were observed in globus pallidus, substantia nigra and parts of the cerebellum (Figure 3). Intermediate levels were seen in striatum and hippocampus and parts of the cerebral cortex. Binding in the brain stem was slight. Specific binding was almost abolished by either 30 µM CP 55,940 or 30 µM SR141716A. The results were essentially identical with those in the literature obtained using [ 3H]CP 55,940 (10,11), and other high affinity agonists (12,13).
Discussion
Four kinds of observations indicate that the diarylpyrazole radioligand [123I]AM251 (a congener of the CB1 receptor antagonist SR141716A) binds to the same brain cannabinoid receptor as the non-classical cannabinoid radioligand [3H]CP 55,940. Firstly, Bmax values for [3H]CP 55,940 and [123I]AM251 are identical within experimental error in washed membranes from cerebellum, striatum or hippocampus. Secondly, CP 55,940 or SR141716A identically accelerated the dissociation of [123I]AM251 from a cerebellar homogenate. Thirdly, in vitro autoradiographs of slide-mounted sections of rat brain prepared using [123I]AM251 were very similar to published autoradiographs obtained using [3H]CP 55,940. Fourthly, members of different classes of cannabinoid receptor ligands were each able to fully compete with the binding of [123I]AM251 as well as [3H]CP 55,940 in cerebellar homogenates.
Our data provide strong support for the idea that [123I]AM251 and [3H]CP 55,940 both bind to the cannabinoid CB1 receptor in rodent brains. However, the two radioligands may each interact with a different set of aminoacid residues at or close to the active site. This is suggested by the dependence of IC50 values for cannabinoid receptor ligands on the radioligand used (Fig 2). This dependence even led to a change in rank order of binding affinity, since SR141716A had higher affinity than CP 55,940 when [123I]AM251 was used as radioligand, but a lower affinity than CP 55,940 when [3H]CP 55,940 was used as radioligand. The lipophilic nature of cannabinoid receptor ligands makes IC50 values more than usually dependent on assay conditions. However, neither this factor nor differences in tissue from individual animals can be responsible for the binding discrepancies because we conducted parallel incubations using the same homogenates with the two radioligands, as well as using identical incubation conditions. The notion that members of different classes of CB1 receptor ligand may bind at different areas of the THC binding site is also supported by the recent demonstration that lys192 of the receptor is less important for binding of WIN 55212-2 than for binding of other agonists (14).
Dependence of binding on assay conditions, however, may explain why in the present experiments with cerebellar homogenates there was a smaller difference of relative inhibitory potencies for THC and SR141716A against [ 123I]AM251 binding (3 orders of magnitude) than our previously published work with washed hippocampal membranes (>4 orders of magnitude). However, there was still a larger difference between the IC50 values of THC and SR141716A than when [3H]CP 55,940 was used as radioligand (Fig 2). Clearly, more work is required to characterize the in vitro binding of [ 123I]AM251 and other cannabinoid receptor radioligands in terms of brain regions, tissue concentrations and other factors. A recently published preliminary binding study using tritiated SR 141716A in rat whole brain minus cerebellum reported a 3 order of magnitude difference between SR141716A and THC, and a higher affinity for SR141716A than CP 55,940, in agreement with our observations using [ 123I]AM251 (15).
Notwithstanding the differences in apparent affinities obtained when using [ 123I]AM251 rather than [3H]CP 55,940 in binding experiments, radioiodinated AM251 appears to have significant advantages. These include its high ratio of specific to non-specific binding and its availability at higher specific activity (2,200 Ci/mmol for iodine-125). For autoradiographic studies [ 123I]AM251 or [125I]AM251 offer reduced wash times compared with [3H]CP 55,940 as well as the shorter film exposure times characteristic of these iodine isotopes relative to tritium. One aim of our work has been to develop a radiotracer for the visualization and quantitation of cannabinoid receptors in the living human brain (16). Unfortunately, we were unable to visualize baboon brain cannabinoid receptors in SPECT (single photon emission computed tomography) imaging experiments, for reasons that are still under investigation (Gatley et al., unpublished).
Acknowledgements
This work was carried out at Brookhaven National Laboratory under contract DE-AC02-76CH00016 with the U.S. Department of Energy and supported by its Office of Health and Environmental Research. The research was also supported by the National Institute on Drug Abuse (RO1 DA03801). The helpful comments of Dr YS Ding are gratefully acknowledged.
References
1. R. Lan, S.J. Gatley and A. Makriyannis, J. Label Cmpds. 38 875-881 (1996).
2. S.J. Gatley, A.N. Gifford, N.D. Volkow, R. Lan and A. Makriyannis, Eur. J. Pharmacol. 307 301-308 (1996).
3. W.A. Devane, F.A. Dysarz, M.R. Johnson, L.S. Melvin and A.C. Howlett, Mol. Pharmacol. 34 605-613 (1988).
4. M. Rinaldi-carmona, F. Barth, M. Heaulme, D. Shire, B. Calandra, C. Congy, S. Martinez, J. Maruani, G. Neliat, D. Caput, P. Ferrara, P. Soubrie, J.C. Breliere and G. Lefur, FEBS Lett. 350 240-244 (1994).
5. Y. Gaoni and R. Mechoulam, J. Am. Chem. Soc. 86 1646-1647 (1964).
6. J. Romero, E. Garc'a-Palomero, S.Y. Lin, J.A. Ramos, A. Makriyannis and J.J. Fernndez-Ruiz, Life Sci 58 1249-1257 (1996).
7. T.E. D'Ambra, K.G. Estep, M.R. Bell, M.A. Eissenstat, K.A. Josef, S.J. Ward, D.A. Haycock, E.R. Baizman, F.M. Casiano, N.C. Beglin, S.M. Chippari, J.D. Grego, R.K. Kullnig and G.T. Daley, J Med Chem 35 124-135 (1992).
8. P.J. Little, D.R. Compton, M.R. Johnson, L.S. Melvin and B.R. Martin, J Pharmacol Exp Ther 247 1046-1051 (1988).
9. L.S. Melvin, M.R. Johnson, C.A. Harbert, G.M. Milne and A. Weissman, J Med Chem 27 67-71 (1984).
10. M. Herkenham, A.B. Lynn, M.D. Little, M.R. Johnson, L.S. Melvin, B.R. DeCosta and K.C. Rice, Proc. Natl. Acad. Sci. 87 1932-1936 (1990).
11. M. Herkenham, A.B. Lynn, M.R. Johnson, L.S. Melvin, B.R. deCosta and K.C. Rice, J. Neurosci. 11 563-583 (1991).
12. B.F. Thomas, X. Wei and B.R. Martin, J Pharmacol Exp Ther 263 133-1390 (1992).
13. E.M. Jansen, D.A. Haycock, S.J. Ward and V.S. Seybold, Brain Res 575 93-102 (1992).
14. Z.H. Song and T.I. Bonner, Mol. Pharmacol. 49 891-896 (1996).
15. M. Rinaldi-Carmona, F. Pialot, C. Congy, E. Redon, F. Barth, A. Bachy, J.C. Breliere, P. Soubrie and G. Lefur, Life Sci. 58 1239-1247 (1996).
16. A. Charalambous, G. Marciniak, C.Y. Shiue, S.L. Dewey, D.J. Schlyer, A.P. Wolf and A. Makriyannis, Pharmacol. Biochem. Behav. 40 503-507 (1991).
Correspondence and requests for reprints to S. John Gatley PhD, Medical Department, Brookhaven National Laboratory, Upton, NY 11973. 516-344-4394 (FAX 5311) gatley@brain.med.bnl.gov.
