* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Determination and Characterization of a Cannabinoid Receptor in
Drug interaction wikipedia , lookup
Discovery and development of direct thrombin inhibitors wikipedia , lookup
Discovery and development of neuraminidase inhibitors wikipedia , lookup
5-HT2C receptor agonist wikipedia , lookup
DNA-encoded chemical library wikipedia , lookup
Discovery and development of non-nucleoside reverse-transcriptase inhibitors wikipedia , lookup
NMDA receptor wikipedia , lookup
Discovery and development of cephalosporins wikipedia , lookup
Discovery and development of beta-blockers wikipedia , lookup
CCR5 receptor antagonist wikipedia , lookup
Discovery and development of tubulin inhibitors wikipedia , lookup
Discovery and development of direct Xa inhibitors wikipedia , lookup
Discovery and development of integrase inhibitors wikipedia , lookup
Toxicodynamics wikipedia , lookup
Psychopharmacology wikipedia , lookup
5-HT3 antagonist wikipedia , lookup
Discovery and development of angiotensin receptor blockers wikipedia , lookup
Drug design wikipedia , lookup
Nicotinic agonist wikipedia , lookup
Neuropharmacology wikipedia , lookup
Discovery and development of antiandrogens wikipedia , lookup
Neuropsychopharmacology wikipedia , lookup
0026-895X/88/050605-09$02.OO/O Copyright © by The American Society for Pharmacology All rights of reproduction in any form reserved. MOLECULAR PHARMACOLOGY, 34:605-613 ACCELERA TED Therapeutics and Characterization WILLIAM A. DEVANE, ALLYN C. HOWLETT FRANCIS of Pharmacology, Groton, Connecticut A. DYSARZ III, M. St. Louis University Medical 06340 (M.R.J. and L.S.M.) August 4, 1988; Accepted Received Experimental COMMUNICATION Determination Rat Brain Department Research, and August ROSS School, of a Cannabinoid JOHNSON,1 St. Louis, LAWRENCE Missouri 63104 Receptor S. MELVIN, (WAD., FAD. in and and A.C.H.) and Pfizer Central 30, 1988 , Various preparations traditionally been ical manifestations (2)]. \9-THC is the have effects \9-THC on the include to and memory, humans (1, cannabinoid a characteristic work FAD. was University 1 Present was supported supported Medical Address: 27709. in extracts alterations in of cannabis include ataxia in part by a Medical School. Glaxo, Inc., altered by National Student 5 Moore Institutes Drive, with in monkeys, of Health Fellowship Research binoid drugs present time, ical neurons NC in models the (3). have the poorly been actions effects Ref. active model 4 for for the may that cannabinoid in of canna- At the neuroanatomdrugs, that this obstacle class of drugs evaluation). interact by has remained Our recent demonstrating cannabinoid drugs inhibit adenylate neuronal system (5, 6). The ability Gpp(NH)p, studied the clas- with can- drugs have CNS. a thorough this change Although understood. concerning to cannabinoid pathways or the a biphasic rodents (3), remain is known responsive cells, in One reason for our lack of insight Concerning cannabinoid drugs in the CNS is that a clearly overcome Park, of cells nabinoceptive from Triangle brain little neurotransmitter mechanism(s) Louis animal in the very and activity relationships in these location sical immobility, locomotor and ROlto A.C.H. T32-NS07254. St. cannabinoid structure-activity humans on analGrant Award K04-NS00868 Training Grant Summer a in hypothermia, a typical spontaneous extensive and associated behavior in responses well as performance and gesia, to in cognition patterns in dogs, Career Development by a Neuropharmacology CNS as psychomotor behavioral have psychologand Dewey THC, tetrahydrocannabinol; DALN, desacetyllevonantradol; HEPES, 4-(2-hydroxyethyl)-1 -piperazineethanesulfonic acid. ABBREVIATIONS: system; drowsiness, actions static DA03690 and Research W.A D. was supported by CNS (3). The predominant analgesia and antiemesis, Animal drug (marihuana) and for their Hollister (1) compound a decrement 2). sativa Cannabis major high,” “psychological This of used therapeutically [see reviews the kinetic constants approximately 55 M and exceeded 200 pM, respectively. The binding of the agonist ligand [3H]CP-55,940 was decreased by the nonhydrolyzable GTP analog guanylylimidodiphosphate. The guanine nucleotide induced a more rapid dissociation of the ligand from the binding site, consistent with an allostenc regulation of the putative receptor by a G protein. The binding was also sensitive to MgCI2 and CaCI2. Binding of [3H]CP-55,940 was displaced by cannabinoid drugs in the following order of potency: CP-55,940 desacetyllevonantradol > 11OH-9-THC = 9-THC > cannabinol. Cannabidiol and cannabigerol displaced [3H]CP-55,940 by less than 50% at 1 M concentrations. The (-)-isomer of CP-55,940 displaced with 50-fold greater potency than the (+)-isomer. This pharmacology is comparable to both the inhibition of adenylate cyclase in vitro and the analgetic activity of these compounds in vivo. The criteria for a high affinity, stereoselective, pharmacologically distinct cannabinoid receptor in brain tissue have been fulfilled. guanyl ($,-y)-imidodiphosphate; that the actions of defined cellular elusive studies the [see have centrally cyclase activity in a of cannabinoid drugs CNS, central nervous 605 Downloaded from molpharm.aspetjournals.org at Univ of Washington on April 14, 2009 SUMMARY The determination and characterization of a cannabinoid receptor from brain are reported. A biologically active bicyclic cannabinoid analgetic CP-55,940 was tntium-Iabeled to high specific activity. Conditions for binding to rat brain P2 membranes and synaptosomes were established. The pH optimum was between 7 and 8, and specific binding could be eliminated by heating the membranes to 60#{176}. Binding to the P2 membranes was linear within the range of 1 0 to 50 g of protein/mi. Specific binding (defined as total binding displaced by 1 M 9-tetrahydrocannabinol (9 THC) or 1 00 nM desacetyllevonantradol) was saturable. The Kd determined from Scatchard analysis was 1 33 pM, and the Bmax for rat cortical P2 membranes was 1 .85 pmol/mg of protein. The Hill coefficient for [3H]CP-55,940 approximated 1 indicating that, under the conditions of assay, a single class of binding sites was determined that did not exhibit cooperativity. The binding was rapid (k00 2.6 x 1 0 pM min) and reversible (K 0.016 min) and (k06’ > 0.06 min). The two Kd values estimated from 606 to Devane regulate et al. adenylate cyclase determined was to be to National effects in were could be related Institute on Drug Abuse. DALN and the isomers of CP-55,940 was demonstrated for the inhibition of adenylate cyclase that paralleled the isomeric selectivity exhibited in analgetic tests in animals (8). The inhibition of adenylate cyclase occurred at Pfizer Central Research. Cannabinoid drugs were stored as 10 mM stock solutions in absolute ethanol at -20. Drugs were initially diluted to 20 M in 9.4 mg/ml fatty acid-deficient bovine serum albumin using Regisil-treated glassware. All subsequent dilutions were made into a vehicle containing 5 mg/ml bovine serum albumin. [3H)CP-55,940 was radiolabeled at DuPont NEN by tritiuin reduction, in the presence of a Pd catalyst, of a double bond between carbons 2 and 3 of the A-ring alkyl side chain (Fig. 1). Labile tritium was removed by several washes with methanol. Product was purified by high performance liquid chromatography on a 25-cm Zorbax ODS only on column as (65:35). Purified material was judged by high performance liquid chromatography to be greater than 97% chemically pure. The specific activity was determined to be 93.4 Ci/mmol, using the UV absorbance to quantitate the yield of product. Tritium exchange with labile hydro- the ability of humans these and compounds animal models to produce 8). The (7, CNS response produced at submicromolar concentrations (7, 8) and thus would be consistent with drug levels that might be expected to be present in the brain during peak activity (9-11). Using a series of cannabinoid Research for in certain adenylate cell G1 (6, of studies that effect such membrane demonstrated requirement protein receptors suggested fluidity the regulatory hormone activity. strongly the to The evidence the presence that ultimately accrued of from pharma- cologically unique cannabinoid receptors on the cultured neuronal cells. Logically, neurons in the CNS should also possess cannabinoid receptors. The tools to search for a cannabinoid receptor in the brain were not available until recently. The relatively low potency and tendency to partition into biological membranes suggest that 9-THC is a poor candidate for a radiolabeled ligand for the detection 9-THC nM (7). would molar characterization of cannabinoid adenylate cyclase with a K10h of 430 It might be expected that radioactively labeled 9-THC have an affinity for cannabinoid receptors in the nanorange and would bind to receptors estimated to be present brain Reports in of the the range of fmoles membrane/buffer per milligram partition of coefficient tissue. for i- THC have ranged from 400 (15) to 12,500 (16). It can be calculated that the amount of labeled z9-THC binding to receptors could potentially be 5 or 6 orders ofmagnitude smaller than the amount that would be expected to partition into A collaborative interaction between lowed the investigation of cannabinoid our laboratories has alreceptors in the brain using cannabinoid a highly potent analgetic bicyclic compound, CNS receptor functional that groups mediates for analgesia agonist-receptor hydroxyl, functional groups of to adenylate cyclase were found in vitro cyclase by CP-55,940 was 2) side found important were the phenolic chain. These to be required (8). The interaction be 1) the C-ring hydroxyl, and 3) the A-ring alkyl posed (12). for the The regulation to exhibit a K10h accounts 25 nM, for binding studies to characterize receptor. The studies reported for [3H]CP-55,940 and provide binding site is the elusive the cannabinoid and than the enantiouseful cannabinoid here describe the binding convincing evidence that activity 55,940 was Materials. The natural cannabinoid Procedures were provided by the labeling was in NaH2PO4, excess stored at of pH the 1 mCi/mi 4.3 theoretical in ethanol at to the radioligand inhibit the was adenylate also monitored. cyclase [3H]CP- activity of N18TG2 membranes in a dose-dependent manner, using the protocol previously described (6) (data not shown). Membrane preparations. Male Sprague-Dawley rats weighing 250 to 370 g were decapitated, and the brains were rapidly removed and with on ice. a washed of two Unless or three rats were 5 mM MgCl2. mm. The indicated, P2 preparation in 45 ml of a solution and all consisting with of 320 homogenate supernatant was reported as follows. homogenized The results prepared mM were The and the cortices a Dounce glass homogenizer sucrose, 2 mM Tris.EDTA, was centrifuged saved, obtained entire at 1600 pellets were x g for 10 washed twice as above. The combined supernatant fractions were then centrifuged at 39,000 x g for 15 mm. The pellet was resuspended in 90 ml of buffer A (50 mM Tris.HC1, pH 7.0 at 30*, 2 mM Tris.EDTA, 5 mM MgCl2), incubated at 37* The membranes mm, and site in for 10 mm, were centrifuged and at 11,000 to be important equilibrium centrifuged resuspended at in buffer x g for 15 mm. for observing studies (see 23,000 These a single Results). x g for A, incubated 10 mm. at 30* for 40 two washing steps homogeneous The final binding pellet was resus- pended in buffer B (50 mM Tris. HC1, pH 7.4 at 30, 1 mM Tris . EDTA, 3 mM MgCl2) at a protein concentration of 4 to 5 mg/ml and stored at Storage Protein bovine for up to using modifications. The Dounce glass tubes Ti50 at rotor no noticeable effect on Tris . Ligand initial by for from and x g for the 10 mm. was performed homogenate was The were performed The final binding. (17) using were also the hippo- preparation with several using a 50- centrifuged in differential sedimentations at 50,000 rpm in a Beckman synaptosomal pellet was resuspended protein in a buffer containing 25 mM Tris . HC1, pH 7.4, 1 EDTA, and 16.6 mM sucrose and was stored at -80. binding assays. Ligand binding assays were performed in test and exception derived homogenization homogenizer 12 mm. Regisil-treated started preparation cortex of the rat. The synaptosomal the protocol of Dodd et al. (18) 1600 for to 5 mg/ml mM had a synaptosomal campus plus prefrontal was made following ml 6 weeks values were determined by the method of Bradford ‘y-globulin as the standard. Some of the studies performed 30’ receptor. drugs of able radioligand, site this the mM for long Biological ml Experimental for CH3CN/25 [3H]CP-55,940 over 1.2M and 0.8M sucrose of adenylate of system term storage and at 20* for routine usage. Purity of the stored material was monitored by thin layer chromatography on silica gel GHLF plates using the solvent system ether/isopropanol (98:2). -80’ 50-ml inhibition the (-)-isomer was found to be 200-fold more potent poorly analgetic (+)-isomer (8). The high affinity and selectivity exhibited by CP-55,940 made it a potentially radioligand pro- A-ring same solvent activity. -80. CP-55,940 (8, 12). This structure is one ofa series of compounds that conform to a postulated three-point agonist-receptor interaction model proposed for the cannabinoid association with the probably specific were found membranes. the gens dissected receptors. was able to inhibit the in and using the tubes in addition 50 mm. After mm at 13,000 and counted The microfuge drugs experiments, incubation, microfuge x g. After to 1 ml containing specified. The samples and centrifugation, were the drained concentration on were were With carried the out at to centrifuged supernatant of drying pins B, was transferred immediately the buffer incubation protein. all reactions the tubes determine tubes of as of 20 to 50 .sg of membrane of the kinetic polypropylene a volume cannabinoid was free 1.5for aspirated [3H]CP-55,940. for 30 mm, after 9 Downloaded from molpharm.aspetjournals.org at Univ of Washington on April 14, 2009 these cyclase Central phenomenon nucleotide of Pfizer enantioselectivity arguing cannabinoid-induced responses adenylate 14), clearly at (12), a universal a guanine the decrease (13, not studies 13), mediates activity types was be expected Further changes. developed analgetic cyclase would for compounds, their synthesized Cannabinoid Receptors 607 in Rat Brain )H Ta Fig. 1. The synthesis of [3H] CP-55,940 by the tritium reduction of compound 1. T 1 3H, Results Conditions studies for addressed radiolabeled and cannabinoid the separation ligand. Williams In agreement (16) and receptor binding. of unbound Harris ligand with the et al. (19) Initial using THC as the glass cellulose nitrate excessive added. and detergents, not provide the free with of the the filters unsuccessful using by adsorption the acetate concentration with polyethyleneimine, or acceptable conditions [:3H]Cp55940 coal was also were achieved above. tubes varied Treatment or cellulose onto 19). was radioligand organic bovine serum for separation. (see also Ref. sedimentation filters of various solvents, albumin did dextran-coated Optimal procedure ware 55,940 Separating charconditions described The incubations were performed in Regisil-treated glass in an effort to minimize the adsorption of cannabinoid surface in the the (16). the optimal different buffers bound to the for incubation tried were effectively to the during the exact the amount and adhering each assay. conditions at glass- [3H]CP- of glassware by determining for of bovine also concentration ligand in the supernatant after the sedimentation To determine presence would drug free adherence accounted of radiolabeled microfuge tube the mixture in the from were Similarly, of cannabinoid alterations resulting several (20). incubation amount Any incubation to the for binding, various concentrations pH values. Of the buffers tested, including K HEPES, KW-tris[Hydroxyrnethyl]methyl-2-aminoethanesulfonate, K phosphate, and imidazole Ct, none performed any better than 50 mM Tris C1. Optimal binding was observed between pH 7 and 8, with specific binding of only 60% of optimal at pH 6 or 9 (data not shown). The phenolic moiety of [3H]CP-55,940 might be expected to have a pK in the vicinity ofpH 10. The 40% loss of specific binding observed between pH 8 and 9 cannot be entirely accounted for by a modification of the ligand. The pH requirement can be postulated to aLso reflect the optimal pH for amino acids that would interact with the ligand or be required to maintain the optimaiprotein conformation. Under the experimental conditions used, linear binding extended from 10 to 50 j.tg/ml of protein for the P2 membrane preparation (data not shown). Experiments were routinely conducted using 20 to 40 g/ml P2 protein, and in this range specific binding typically represented 85% of the total binding. Specific binding was linear through 120 to 150 g/ml of protein for the synaptosomalpreparation (data not shown). As the concentration of membranes was increased, there was a decline in the percentage of the total binding that could be described as specific. Thermolability of the specific binding was demonstrated. Membrane preparations that were incubated at 600 for 12 mm and before of free [3H]CP-55,940 using a filtralittle success. The binding of ligand to and decrease shown). separation met with to the albumin of Roth labeled ligand, tion technique fiber serum bound from experience compounds assay failed These [3H]CP-55,940 is subject to show significant specific binding ing to thermal nonspecific denaturation. of therefrom was obtained from four pmol/mg the by Scatchard sites ofprotein Hill for (four transformation (21), and the Scatchard error). The plot of 133 ± 11 pM (21) of the data Kd preparation yielding An ex- inset). ± standard experiments). increas- 2A, transformation P2 cortical with (Fig. in Fig. 2. A (mean the binding site. was saturable, to increase isotherm are depicted experiments of binding continued [3H]CP-55,940 binding of a saturation obtained by binding concentrations ample not findings would be consistent with the binding of to a protein component of the membranes that Characterization of the [3HJCP-55,940 [3H]CP-55,940 binding to cortical membranes whereas (data Downloaded from molpharm.aspetjournals.org at Univ of Washington on April 14, 2009 which the tips of the tubes were sliced, using a heated spatula and a cutting block designed to ensure that the cut tips were of identical size. The tips were then placed in scintillation vials and submersed in 2 ml of a solubilizing solution (5% ethanol, 5% Triton X100, 0.2 N NaOH). The vials were then shaken for at least 4 hr in order to dissolve the pelleted membranes. Scintillation cocktail (10 ml) was added to each vial and the radioactivity was determined using a Beckman LS1800 with an efficiency for tritium of 30%. Nonspecific binding to the microfuge tip was assessed in control tubes having radioligand but no protein (typically 2% to 3% of the total radioactivity available in the incubation medium). Subtracting this value from total binding gave the total binding in the pelleted membranes. Specific binding was defined as the difference between total binding to the membranes in the absence and presence of either 100 nM DALN or 1 M 9-THC. Nonspecific binding to the membranes was typically 15% to 30% of the total binding to the membranes, dependent upon the membrane and ligand concentrations (see Fig. 2). Assays were carried out in triplicate with an average coefficient of variation for the samples of 2.5%, and experiments were repeated at least three times. Metabolism of [3HJCP-55,940. To determine whether [3H]CP55,940 was metabolized during the binding assay, 40 g of membrane protein were incubated with 70 pM [3H]CP-55,940 for 90 mm at 30* in the standard assay buffer described above. After centrifugation, the supernatant and the pellet were separated and the radioactivity was extracted from each using ether. After drying with N2 gas, the samples and an unincubated [3H]CP-55,940 control were resuspended in absolute ethanol, and thin layer chromatography was performed using the procedure described above. The plate was sprayed with EN3HANCE (NEN, Boston, MA), placed against Kodak X-Omat film and stored at _80* for 6 days. Upon developing, a single band was observed for the samples which comigrated with the control. These studies demonstrated that neither the free nor bound [3H]CP-55,940 was metabolized or chemically modified during the assay procedure. -CP-55,940 The was data were a straight This density 1.85 ± 0.26 analyzed line (Fig. One IDI IIIIIIIIIIIIIIIIIII10 IIIIIIIII K9SR-X5R.- EQAQ 608 Devane et al. 60 A 0 . -, E 9- / 40 D C 3 0 I. . I .0 0 20 4- 0 Fig. 2. Equilibrium binding of [3H]CP55,940. Membranes (43 /Lg of protein) were incubated with various concentrations of [3H]CP-55,940. A, The saturation isotherm of specific binding. Inset; binding a) 0. U) Fr.. (3OcP-664O M 0 0 2000 1000 [3H]CP-55,940 ‘+L/tItJ transformation of [3H]CP-55,940 binding data from A with the bound ligand being expressed in terms of concentration (pM). This experiment exhibited Kd and Bmax values of 1 39 M and 1 .3 pmol/mg of protein, respectively. Inset; The Hill transformation of data from A. F; free drug concentration; B; specifically bound drug. The Hill coefficient (nH) was calculated to be 0.90 for this experiment. The lines drawn represent the best fit as determined by least squares linear regression analysis. (pM) 0.60 a) a) 0.40 9- -o C 3 0 0.20 0.00 0 10 20 30 40 [3H]CP-55,940 2B, inset). and the The was flH that the n binding sites conditions 0.88 Kinetic than at 30. hr. specific time of the through The 3A). binding (pM) ± 12 pM that class of the assay cooperativity is specific binding plateau consistent incubation binding was of maximal with being of exists the component reached and with rapidly, binding within for at least determination that steady state no the already earliest ment [3H]CP-55,940-receptor dissociation more rapid earliest One change the by the addition These rectly studies in the were performed microfuge tubes before of 100 nM DALN is depicted by establishing rather than by sedimentation. Semilog in Fig. equilibrium transferring plots (K_1 mechanism or interaction with free affinities dithe ies. In preliminary was observed equilibrium the this were not that binding for k_1 may the studies the be slower the component having that these of transient affinity states not of tightly for that equilibrium unwashed kinetic receptor. interaction Evidence in the two bound such shown). binding be discerned An an binding P2 membranes, could A at t,. two the possessing using 45 mm). = forms It is of interest (data displaceAssuming (t,. nucleotides. multiple of was that the be discerned, either below. studies the for complete expect experiments) also G proteins discernible time represent It is possible of guanine 3B. suggest (21), (three is described the of manipulation. interchangeable for receptor When addition at equilibrium would may period min). 3B). upon would have sedimented within Thus, the data obtained for could 0.06 that one depicted min1 represent binding be noted the dissociation may specific However, points during ± 0.001 11 mm in- time (Fig. immediately of membranes of centrifugation. first-order states complex of the was GDP of the 20% It should occurring 0.016 monophasic centrifuged is 10 mm. major fraction the first 3 mm the not were displaced. sedimentation non- at the was tubes nM DALN, The 3 hr. dissociation mixture 100 dur- further 70 dissociation altered Procedures). underwent the with attained stable or chemically Experimental measurable ligand reached remained metabolized (see to P2 mem- the 60 microfuge observation of [3H}CP-55,940 association itiated reaction The a single under no significant Equilibrium The is not point experiments). that by [3H]CP-55,940 that 116 was suggests labeled a rapid 90% This radioligand analysis bound sites. (Fig. 50 mm such (four one and analysis greater the ± 0.08 is being indicates receptor from approaches binding branes ing derived described among 2 Kd 50 explanation studit in Downloaded from molpharm.aspetjournals.org at Univ of Washington on April 14, 2009 Free of [3H]CP-55,940 the absence (0) or presence (U) of 1 ;LMin DALN. B, Scatchard A E’sE\I 3000 Cannabinoid 0 50 100 Time for these 100 50 results might tightly be that the GDP could bound 200 (mm) Time possessing (mm) population be 150 of G proteins greater in to for Kd binding dissociation mated pM’ the for rapidly rate experiments). the slowly method (21, 22). Using (and k+1 ological to thus been of G proteins reviewed by on agonist-receptor Birnbaumer and and Gilman (24). To summarize believed to exist in a heterotrimer form bound GDP in the presence of the values calculated binding methodology has Allosteric been Thus, that system via anisms similar to those homologous x the The protein intermediate for agonist. methods briefly, the receptor a k0h, as the receptor theoretical and of the linear transformation an internal the k+, may and the Kd of consistency One would its signal would that be have receptors. regulated been Our to the by the current hypothesize cyclase allosteric demonstrated unit. Dissociation mechfor understanding other GTP the the decrease in The of the G protein into a and the a t/. from calculated in the 0.059 pM. for ± 0.009 This dissociating dissociation min value component the described for from The effector a sub- receptor and hy- the system to allow of hormone. were 45 mm Kd to not in 12 mm. and An the for in-depth Kd the the addition The of Gpp(NH)p estimated above. (data analyzed (Fig. 4). The presence experiments) to ligand GTP-bound the nucleotide (three is similar affinity dissociates of [3H]CP-55,940 dissociation presence or absence ofguanine of Gpp(NH)p reduced the In equilibrium binding studies iM Gpp(NH)p resulted in a binding of high agonist the subunit of 100 specific dissociates. 3y subunits. from receptor, a suf- a nonhydrolyzable concentrations presence kinetics or the to altered cannabinoid that shown). on GDP affinity hormone with confers a. relatively with are tightly receptor-hormone-G and interacts of the respond indicated 40% separates the the the GTP and interaction change that of the cyclase) of For for the adenylate and perpetually similar binding (23) G proteins possessing Upon exhibits complex complex, adenylate drolysis such is decreased, (e.g., method- are Kd to (afl’y) nucleotides, Upon this analog for estimating estimates the system interactions colleagues a conformational of guanine Kd for pM. to the absence An Kd 180 rate, min’ complex, energy the a pseudo-first-order dissociation 10 pM’ of binding. transmits a G protein 3.4 ± 0.76 value, demonstrated. regulation a receptor functionally using data. was this 47 pM. receptor-agonist ficient esti- k+1 be to estimate using have kinetic appreciably to exceed of these values) Kd the equilibrium that the limitations, both was have would the slower 2.6 ± 0.2 x was estirequired 3 The (22) site would data association contribute Using to equilibrium be calculated to be would be 62 pM. Although the not displacing of the proceeds the of incubation. of binding site treatment reaction 5 mm (three from rate of association of the membranes would after dissociating alternative calculated initial min’ calculated the dissociation until from 10 that the reaction be The initial pelleting rates. that and the the may by assuming mm influence has Casey The mated the of unwashed membranes. and Fig. 3. Time course of association (A) and dissociation (B) of [3H]CP-55,940. [3H]CP-55,940 (81 pM) was incubated with 28 ;hg of P2 membranes at 30#{176}. A, The times indicated are those that elapsed between the addition of protein (start of the reaction) and the start of centrifugation of the microfuge tubes. Specific and nonspecific binding were determined with 100 flM DALN as described in Expenmental Procedures. B, After equilibrium binding of [3H]CP55,940 had been reached (70 mm), 100 n DALN was added (t = 0) and dissociation was monitored. Data presented are a first-order representation with B/B0 denoting the specific binding at the time indicated/specific binding at t = 0. The results are means of triplicate determinations from a single representative experiment, which was performed three times. k was was 176 rapidly analysis Downloaded from molpharm.aspetjournals.org at Univ of Washington on April 14, 2009 0 150 609 in Rat Brain Receptors 610 Devane et al. I Fig. 4. Effect of Gpp(NH)p on the dissociation rate of [3HJCP-55,940. [3H]CP-55,940 (81 pM) was incubated with 28 g of P2 preparation membranes for 70 mm at 30#{176} and dissociation was monitored after addition of 1 00 nM DALN (t = 0). Control, addition of 1 00 nM DALN; +Gpp(NH)p, simultaneous addition of 1 00 nM DALN plus 1 00 M Gpp(NH)p. The x axis indicates the time that elapsed between the addition of the above compounds and 2 mm after the start of centrifugation. The y axis is a log-scale presentation of the percentage of specific binding at the indicated time/specific binding at t = 0. The data are the means of triplicate determinations from a single representative experiment, which was repeated three times. I -j 0 15 30 Time interaction proteins most facile the of the in various nucleotides with Divalent for possesses at divalent presence affinity one has for been the whether G The at than 50% with CaCl2. similar Studies of the the (see 26 and receptors opioid affinity associated (data not which manner, (26, 27). indicated Na has In as 1 mM binding response to does for the an effort - opioid Na act Ca2 Na of K NaCl 80%. The appear to be great, specific had and by about not intra- reduced minimal 100 mM selectivity KC1 of this suggesting that salt concentrations as is altering the ability of the ligand to bind receptor. the of of [3H]CP-55,940 ability the (Fig. 6). determined using 1987) (28). A one-site for an pmol/mg computer data using CP-55,940 ± 0.20 ment the both CP-55,940 criterion had analysis model fit the and on natural a K mem- of 68 ± 6.2 (three better cannabinoid residual variances for cortical pM experiments) (version data other the several of homologous program speci- [3H]CP-55,940 in the protein LIGAND F test sites The by determining and with compete binding Unlabeled receptor. established compounds to specific by was synthetic compounds of cannabinoid binding of related Bmax of 1.75 model receptor mM, mM tested inhibition may not be the result of a specific interaction a Na site. It is possible that a chaotropic effect of higher branes to determine 120 were this with occupancy as to cations monochloride to be present concentrations of specific cannabinoid to adenylate Low 20 by of various be expected At 40%. effects 5). Monovalent might inhibited the effects were binding in that Na may been demonstrated KCI Concentrations and is coupled In by about of [3H]CP-55,940 the extracellularly. effects. ficity [3H]CP-55,940 similar specific (Fig. Pharmacology the binding observed, that or to the the be determined binding aden- (25). as low ligand ligands therein). to shown). receptor, of agonist in higher with Maguire agonist this to increase of MgCl2 of the references a much of Mgi by for of GDP in addition having (Fig. 5). Qualitatively MnCl2 also stimulated cyclase in an inhibitory an allosteric regulator decrease for discussed to MgCl2 A role complex, effects affinity G protein dissociation a site The binding 24). the were of the could concentrations cellularly of a the cations salts is that interaction regulation valent by guanine that (23, the concentrations specific a manner for with ligands have Mgi shown 24). investigation, Ref. here to influence receptor-hormone (23, of agonist by greater observed with with study. presented It is believed associated Mgi stimulated site of this be influenced reported site been of the cyclase present been receptors. least functions for results can consistent have their cation affinity scope a G protein. cations of agonists ylate site binding the of the binding in a manner receptor other is beyond interpretation [4H]CP-55,940 the [3H]CP-55,940 states (mm) displace2.3.11; than May, a two-site drugs tested at the level MgC2 200 0 150 Fig. 5. The influence of cations on specific binding of [3H] CP-55,940. Control samples contained 50 m Tns . HCI, 1 C 0 0 0 mM C a) 0 a) Tns.EDTA, and 0.1 mM MgCl2. Experimental samples contained the same buffer plus the indicated concentrations of salts. The data are the means ± standard error of triplicate determinations from a single representative experiment. All values were different from control at p < 0.05 except 20 mM NaCI and 5 mM KCl. Similar results were observed in two other experiments. 100 50 0 20 120 5 100 1 Millimolar 3 added 10 1 3 10 Downloaded from molpharm.aspetjournals.org at Univ of Washington on April 14, 2009 the of 60 45 Cannabinoid Receptors 611 in Rat Brain 0) c -o c -O Fig. 6. Competitive inhibition of [3H]CP55,940 binding by various synthetic and natural cannabinoid drugs. [3H]CP-55,940 (5070 pM) was incubated with P2 membranes (1 6-30 g) for 50 mm at 30#{176} with either the indicated concentrations of drug or vehicle alone. The results were normalized to 100% of specific binding, which was determined with 1 00 nM DALN as described in Experimental Procedures. Data points represent the averages of triplicate determinations from single representative experiments. The K1 values listed in the inset table were determined using the LIGAND program and represent the mean ± standard error of three 0 4- 0 a) 0. U) 9- 0 a) 0 a) independent 12 11 10 9 -Log of p = values analysis 8 [drug] 7 CP-55,940 was 50-fold less potent than the (-)-isomer, having a K, of 3.4 nM. DALN was nearly equipotent with CP-55,940, having a K, of 123 pM. z9-THC and 11-OH-9-THC both exhibited high affinity binding with K1 values of 1.6 nM. Cannabinol was 8-fold less potent than 9-THC, having a K1 of 13 This with The cyclase selective tor. nabinoid fail compounds in solution. These concentrations latter two of cancompounds to exhibit cannabinoid activity in humans or animal models (1-3). One of several explanations for the binding results could apply. 1) The two compounds could bind to the receptor with low affinity, but the concentrations required to observe a biological response may not be possible to achieve in vivo. 2) The cannabinoid by drug response to these two compounds may be masked concentrations, such as membrane perturbation in in vitro studies and CNS depression in in vivo studies. 3) The observed binding displacement may be the result of a contaminant in the drug preparation. The latter explanation is a possible artifact that must be dealt with in future investigations. These drugs were isolated from organic extracts of cannabis that originally contained a variety of cannabinoid compounds, including \9-THC. effects at high Discussion Studies the ability suggestive of a cannabinoid receptor were based on of centrally active cannabinoid drugs to interact guanine for this interaction The findings argument would with presented The rat from what would CNS (30). It may for be expected investigations to drugs have system. To strengthen binoid have receptor evidence duction with drugs rat (31, brain 32). in the site the its cortical was more binding site than 2 orders its Kh regulation of adenylate the model of the canna- CNS, Cyclic we AMP now pro- in response agonist in the absence of magnitude cyclase by cannacell is decreased affinity [H]CPPrevious of the preparations. regions for brain. cyclase in The receptor association 55,940 for nucleotides for studies. with putative The of [3H]CP- is consistent binding in in CNS remarkably binding ligand cyclase a and for agrees neuroblastoma slice rapid reported of adenylate adenylate brain The Kd for binding the via consistent Bmax determined a cloned the using site. the cyclase inhibition utilized in several cannabinoid that adenylate of the binoid system a neuromodulatory be hypothesized is linked The a receptor are constants agonist for strongest of a neuromodulator equilibrium this the with values The kinetic from determined and with (30). the Kd obtained the affinity binding toxin receptor. messenger expected consistent receptors derived this saturability are provide thermolability properties binding cortex pertussis consistent a second the require- (6, 13). a cannabinoid and for are neuromodulator 55,940 with the are cation receptors is entirely structure binding receptor. with here be associated a protein divalent study for recep- cyclase demonstrated this The pH sensitivity reversible 55,940 in available described G protein. the enantiothat these mediating adenylate of G1-linked currently site with and and and of a receptor nucleotide characteristic potent, hypothesis system. adenylate membrane-bound in favor drugs messenger inhibit to the a biological of cannabinoid sensitivity that with arguments characteristic ments cellular second compounds cell type-specific, would support interact interaction the higher in a reversible, manner (5-8) Additional would to maintain characterized, of cannabinoid compounds binding for be necessary 6 a well ability order of potency generally parallels the order of both CNS activity in vivo (1, 3, 12, 29) and inhibition of adenylate cyclase in vitro (7, 8). Cannabidiol and cannabigerol were much less potent, binding the [3H]CP-55,940 site with K1 values estimated to be greater than 500 nM. Cannabidiol and cannabigerol were unable to fully displace the specifically bound [3H]CP-55,940 at the highest concentrations tested (1 1uM). Concentrations greater than this were not tested due to the limited solubility of cannabinoid drugs above 1 M and the confounding aspects of increasing the solvent or bovine serum albumin concentration, which nM. for each drug. (M) 0.05. Using 68 pM as the Kd for [3H]CP-55,940, the K1 of the various compounds were determined by LIGAND of heterologous displacement data. The (+)-isomer of potency experiments in the to [3H]CP- of guanine greater than neuronal cell Downloaded from molpharm.aspetjournals.org at Univ of Washington on April 14, 2009 cL 612 Devane model of (8). eta!. However, Gpp(NH)p adenylate for ligand this site are inhibition One responses designed to possess activity plate, for CP-55,940 phenylbenzylquinone tests to the (-)agrees reported here binoid been displace rat latter was not dependent scribed by Nye thylammonium to this site behavioral the analog does not models CNS depression with our (35). taming the hydrophobic from the A-ring (8). The to discern 34) ligand ity of the stereoisomers and in vivo effects of z9-THC may was (33). z8-THC was binding of [3H]trimethylammonium was 3 orders of magnitude lower 8-THC than the that, throughout adsorption Evidence analysis finding of the may these to suggests glass equilibrium in part be that this yielded a K determined binding data the aqueous not site the (34). linked a G that the is definitely described here suggested reported or that [3H] different using [3H]CP cannabinoid neuromodulators (36) that the drugs to their receptors. concentrations 11-OH-z9-THC increased additional that drugs in specific studies similarly altered from high fluidity in excess the presence binding that of a of the f3-adre- for opioid receptors cannabinoid We have previously demonstrated in neuroblastoma agonists duction indicate in the cells corthat that is which of plasma cannabi- membranes (37). Vaysse with certain by addition drugs laboratory, concentrations of synaptic by fluorescence polarization (38) reported a similar interference of several of high as and colbinding concentrations presence of 100 mM that site of cannabinoid is not the related to the to the #{244} opioid receptor or to subsequent (14). The studies reported in the present the presence of a pharmacologically ethanol. binding signal work selective, of transclearly high affinity binding site for cannabinoid drugs. However, membrane perturbation may be the mechanism by which high concentrations of cannabinoid drugs may interfere with binding determinations made for a variety of other receptor types. The development noid class of drugs that brain have that of a ligand will allow previously binding assay investigations for the cannabiof cannabinoid been Pathways in the action can can be be more not may be involved in The cellular regulation examined. cannabinoid and can drugs its interaction be assessed. can possible. cannabinoid of the receptor with Perhaps be developed alternative second an antagonist for now that a binding site has been found that correlates with a cellular function (inhibition of adenylate cyclase). Furthermore, efforts to search for putative endogenous of the make a major ligand can characterization impact now proceed. of a cannabinoid on research in this Thus, the receptor field. Acknowledgments The excellent clerical assistance of Maggie Klevorn is gratefully appreciated. References 1. Hollister, 2. Dewey, (1986). 3. Razdan, L. E. Health aspects W. L. Cannabinoid R. K. Structure-activity of cannabis. pharmacology. relationships Pharmacol. PharmacoL Rev. in cannabinoids. 38:1-20 (1986). Rev. 38:151-178 Pharmacol. Rev. 5. 6. 7. and (33). to by detected leagues by This solubility have of other vehicle supported the that site nergic antagonist ligand [3H]dihydroalprenolol in mouse tical homogenates. The interpretation of this finding was the cannabinoid drugs altered membrane properties such the binding behavior was modified (36). This mechanism observation considered is not detergent 4. and for binding Hillard and Bloom of 3 tM 9-THC for 2-fold, Kd were binding than derived by ligands site z8- by nonhydrolyzable It is clear binding receptor ligand. laboratories the will selectiv- levonantradol explained experiments, of the less for constants Scatchard The (34). z8-THC cannabinoid importance that several neither [3H]trimethylammonium decreased as the Other a of main- binding kinetic unusual of is consistent be questioned. observed the alter the animal exception models or in humans of this binding site The dextronantradol the 8-THC and no stereoselectivity for binding 8-THC indicated greatest affinity in and cannabidiol) are [3Hjtrimethylammonium from of than nucleotides fully characterized messenger systems side chain extending profile for displace- agonists nor antagonists in in vivo animal (33). Thus, the pharmacological relevance to cannabinoid used importance alkyl de- [3H]5’-trime- activity of the (16). was binding rather of guanine actions which in typical with biological nature ofthe pharmacological affinity 9-THC the activity action, ment of [3H]5’-trimethylammonium cannabinoid compounds having brain regions (e.g., cannabigerol to however, a high using The poor able adsorption, offree biological This were membranes demonstration and synaptosomal concentration of 8-THC. not and as trimethylammonium action have Roth 8-THC; 1 jM membrane of cannabinoid previous canna- [3H]8-THC group in brain analogs assays pharmacological displacewas not performed (19). (33, exhibit and receptor to analgesia. site colleagues and a cannabinoid of unable was enhanced noid including typical or purified with than binding and of the former inasmuch THC demonstrated Other this in addition and ligands of CP-55,940 Thus, characterize binding upon affinity order hypothermia, certain The other the order receptor. for binding were of binding dem- of drugs, (29). to crude brains. of the This The or in vivo functions colleagues (19) and and investigators A high class and in vitro of the the activity, demonstrated up to 10% component to in animals the binding was not saturable ment by other cannabinoid The binding find with Harris from analgetic site. by demonstrated respectively, membranes The enantioselectivity is mimicked with to receptor associated met with success. [3HJ9-THC, (12). in the tail flick, hot tail clamp, and flinch cannabinoid observed (16) receptor site specifically was binding locomotor attempts Williams data protein 55,940 activity to be enantioselective responses of potencies described for reported [3H]CP-55,940 This ligand 50-fold the to be associated Previous with 8. 38:75-149 (1986). Martin, B. R. Cellular effects of cannabinoids. PharmacoL Rev. 38:45-74 (1986). Howlett, A. C., and R. M. Fleming. Cannabinoid inhibition of adenylate cyclase: pharmacology of the response in neuroblastoma cell membranes. Mol Pharmacol. 26:532-538 (1984). Howlett, A. C. Cannabinoid inhibition of adenylate cyclase: biochemistry the response in neuroblastoma cell membranes. Mol. Pharmacol. 27:429-436 (1985). Howlett, A. C. Cannabinoid inhibition of adenylate cyclase: relative activity of constituents and metabolites of marihuana. Neuropharmacology 26:507512 (1987). Howlett, A. C., M. R. Johnson, L. S. Melvin, and G. M. Milne. Nonclassical cannabinoid analgetics inhibit adenylate cyclase: development of a cannabinoid receptor model. Mol. Pharmacol. 33:297-302 (1988). of Downloaded from molpharm.aspetjournals.org at Univ of Washington on April 14, 2009 appears the for also shown (7, 8). activity of the have been previously [3H]CP-55,940 in spontaneous have with cyclase demonstrated writhing, with the analgetic immobility, site for typical changes order addition (8, 12). The ratio of the activities in the analgetic tests was 200-fold. for functions concurrent analgetic was well here potency by the state that the is analgesia. in rodents (+)-isomer of potency promoted (23, 24). The enantioselectivity the potent reasonably onstrated state prevalent consistent in vivo regulate jump the of adenylate the of may affinity be cyclase regulation interaction and binding for the the would Cannabinoid 9. 10. 11. 12. 13. 14. 15. 16. 18. 19. 20. 21. 22. 23. 24. Harris, L. S., R. A. Carchman, and B. R. Martin. Evidence for the existence of specific cannabinoid binding sites. Life Sci. 22:1131-1138 (1978). Garrett, E. R., and C. A. Hunt. Physicochemical properties, solubility, and protein binding of 9-tetrahythocannabinol. J. Pharm. Sci. 63:1056-1064 (1974). Limbird, L. E. Cell Surface Receptors: A Short Course on Theory and Methods. Martinus Nijhoff Publishing, Boston, 51-96 (1986). Bennett, J. P., and H. I. Yamamura. Neurotransmitter, hormone, or drug receptor binding methods, in Neurotransmitter Receptor Binding (H. I. Yamamura, S. J. Enna, and M. J. Kuhar, eds.). Raven Press, New York 61-89 (1985). Birnbaumer, L., J. Codina, R. Mattera, R. A. Cerione, Sunyer, F. J. Rojas, M. G. Caron, R. J. Lefkowitz, and of hormone receptors and adenylyl cyclases by guanine proteins. Recent Prog. Horm. Res. 41:41-99 (1985). Casey, P. J., and A. G. Gilman. G protein involvement coupling. J. Biol. Chem. 263:2577-2580 (1988). J. D. Hildebrandt, R. Iyenga.r. Regulation nucleotide binding 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. Maguire, M. E. Hormone-sensitive magnesium transport and magnesium regulation of adenylate cyclase. Trends PharmacoL Sci. 5:73-77 (1984). Blume, A. J., D. Lichtshtein, and G. Boone. Coupling of opiate receptors to adenylate cyclase: requirement for Na and GTP. Proc. NatL Acad. Sci. USA 76:5626-5630 (1979). Koski, G., R. A. Streaty, and W. A. Klee. Modulation of sodium-sensitive GTPase by partial opiate agonist. An explanation for the dual requirement for Na and GTP in inhibitory regulation ofadenylate cyclase. J. BiOL Chem. 257:14035-14040 (1982). Munson, P. J., and D. Rodbard. LIGAND: a versatile computerized approach for characterization of ligand-binding systems. AnaL Biochem. 107:220-239 (1980). Little, P. J., D. R. Compton, M. R. Johnson, L. S. Melvin, and B. R. Martin. Pharmacology and stereoselectivity of structurally novel cannabinoids in mice. J. PharmacoL Exp. Ther., in press. Burt, D. R. Criteria for receptor identification, in Neurotransmitter Receptor Binding (H. I. Yamamura, S. J. Enna, and M. J. Kuhar, ode.). Raven Press, New York, 41-60 (1985). Bidaut-Russell, M., and A. C. Howlett. Cannabinoid drugs regulate cyclic AMP metabolism in rat brain. Trans. Am. Soc. Neurochem. 19:163 (1988). Bidaut-Russell, M., and A. Howlett. Opioid and cannabinoid analgetics both inhibit cyclic AMP production in the rat striatum. Adv. Biosci. in press. Nye, J. S., H. H. Seltzman, C. G. Pitt, and S. H. Snyder. High-affinity cannabinoid binding sites in brain membranes labeled with (3H]-5’ -trimethylammonium 18-tetrahydrocannabinol. J. PharmacoL Exp. Ther. 234:784791 (1985). Nye, J. S., and S. H. Snyder. The high-affinity cannabinoid binding site in brain: regulation by guanine nucleotides and isolation of an endogenous inhibitor. NatL Inst. Drug Abuse Res. Monogr. 79:134-147 (1987). Compton, D. R., P. J. Little, and B. R. Martin. Cannabinoid behavioral versus non-specific actions, in Marijuana ‘87: Proceedings of Symposium on Cannabis (G. B. Cheshire, P. Consroe, and R. Musty, ode.). NCADA Monograph, Australian Government Publishing Service, Canberra, 213-218 (1988). Hillard, C. J., and A. S. Bloom. 19-Tetrahythacannabinol-induced changes in 9-adrenergic receptor binding in mouse cerebral cortex. Brain Res. 235:370-377 (1982). Hillard, C. J., R. A. Harris, and A. S. Bloom. Effects of the cannabinoids on physical properties of brain membranes and phospholipid vesicles: fluorescence studies. J. PharmacoL Exp. Ther. 232:579-588 (1985). Vaysse, P. J.-J., E. L. Gardner, and R. S. Zukin. Modulation of rat brain opioid receptors by cannabinoids. J. PharmacoL Exp. Ther. 241:534-539 (1987). Send St. in receptor-effector 613 effects: specific the Melbourne T. N in Rat Brain reprint requests to: Dr. Allyn Louis University School of Medicine, MO 63104. C. Howlett, Department 1402 South Grand of Pharmacology, Boulevard, St. Louis, Downloaded from molpharm.aspetjournals.org at Univ of Washington on April 14, 2009 17. Gill, E. W., and G. Jones. Brain levels of 1’-tetrahydrocannabinol and its metabolites in miceCorrelation with behaviour, and the effect of the metabolic inhibitors SKF 525A and piperonyl butoxide. Biochem. PharmacoL 21:2237-2248 (1972). Ho, B. T., V. S. Estevez, and L. F. Englert. The uptake and metabolic fate of cannabinoids in rat brains. J. Pharm. PharmacoL 25:488-490 (1973). Ohlsson, A., M. Widman, S. Carlsson, T. Ryman, and C. Strid. Plasma and brain levels of -THC and seven monooxygenated metabolites correlated to the cataleptic effect in the mouse. Acta PharmacoL ToxicoL 47:308-317 (1980). Johnson, M. R., and L. S. Melvin. The discovery of nonclassical cannabinoid analgetics, in Cannabinoids as Therapeutic Agents (R. Mechoulam, ed). CRC Press, Boca Raton, FL, 121-145 (1986). Howlett, A. C., J. M. Qualy, and L. K. Khachatrian. Involvement of G in the inhibition of adenylate cyclase by cannabimimetic drugs. MoL PharmacoL 29:307-313 (1986). Devane, W. A., J. W. Spain, C. J. Coscia, and A. C. Howlett. An assessment of the role of opioid receptors in the response to cannabimimetic drugs. J. Neurochem. 46:1929-1935 (1986). Seeman, P., M. Chau-Wong, and S. Moyyen. The membrane binding of morphine, diphenylhydantoin and tetrahydrocannabinol. Can. J. PhysioL PharmacoL 50:1193-1200 (1972). Roth, S. H., and P. J. Williams. The non-specific membrane binding properties of 9-tetrahydrocannabino1 and the effects of various solubilizers. J. Pharm. PharmacoL 31:224-230 (1979). Bradford, M. M. A rapid and sensitive method for the quantitation of microgram quantities ofprotein utilizing the principle of protein-dye binding. Anal. Biochem. 72:248-254 (1976). Dodd, P. R., J. A. Hardy, A. E. Oakley, J. A. Edwardson, E. K. Perry, and J.-P. Delaunoy. A rapid method for preparing synaptosomes: Comparison with alternative procedures. Brain Res. 226:107-118 (1981). Receptors