Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Compounds that bind to brain cannabinoid (CB₁) receptors show a large degree of diversity in chemical structure and include classical tricyclic and bicyclic cannabinoids, aminoalkylindoles, indoles, pyrroles, and anandamides. Each of these classes of compounds shares a similar profile of pharmacological activity in vivo with the prototypic tricyclic cannabinoid, ⁹-tetrahydrocannabinol (⁹-THC), albeit they differ in potency and there are some differences in efficacy in individual assays (Compton et al., 1992, 1993; Adams et al., 1995; Wiley et al., 1998). These in vivo cannabimimetic effects include hypoactivity, hypothermia, antinociception, and catalepsy in mice (Martin et al., 1991; Smith et al., 1994), ⁹-THC-like discriminative stimulus effects in rats and monkeys (see Wiley, 1999 for review), and static ataxia in dogs (Lichtman et al., 1998).

With the synthesis of a CB₁ cannabinoid antagonist, SR141716A, a new class of cannabinoids was revealed (Rinaldi-Carmona et al., 1994). SR141716A selectively binds to cannabinoid CB₁ receptors without producing cannabimimetic activity in vivo (Compton et al., 1996), suggesting that binding and activation of cannabinoid receptors are separable events. Consequently, structure-activity relationship (SAR) studies with analogs of this antagonist provide a unique opportunity to compare the structural requirements for binding and antagonist activity to those required for binding and agonist efficacy. To date, only a couple of studies have been published, which systematically examined the SAR of cannabinoid CB₁ antagonists (Thomas et al., 1998; Lan et al., 1999). Although both of these studies reported CB₁ binding values for SR141716A analogs, neither involved measurement of in vivo activity of the compound alone and in combination with an active cannabinoid.

The purpose of the present study was synthesis of a series of analogs of SR141716A and subsequent in vitro and in vivo testing. These analogs retained a central pyrazole structure with manipulation of one of four other areas of the molecule: 1) substitution for carboxyamide and/or piperidine substituent (3-substituent substitution); 2) substitution for the 2,4-dichlorophenyl group (1-substituent substitution); 3) substitution for chlorophenyl group (5-substituent substitution); or 4) substitution for the methyl (4-substituent substitution) (Fig. 1). Cannabinoid receptor binding affinities were determined then followed by in vivo testing in mice. Selected compounds with binding affinity (K_i) less than 100 nM were further tested in combination with active dose(s) of ⁹-THC to evaluate potential antagonist effects.

View larger version (11K): [in this window] [in a new window]   Fig. 1.   Chemical structure of SR141716A with points of substituent attachment marked by numbers surrounding the pyrazole core: 1 (dichlorophenyl group), 2 (hydrogen), 3 (carboxyamide and piperidine groups), 4 (methyl group), and 5 (chlorophenyl group). For comparison purposes, the chemical structure of 9-THC is also presented.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Subjects. Male ICR mice (25-32 g), purchased from Harlan (Dublin, VA), were housed in groups of five. All animals were kept in a temperature-controlled (20-22°C) environment with a 12-h light/dark cycle (lights on at 7:00 AM). Separate mice were used for testing each drug dose in the in vivo behavioral procedures. The mice were maintained on a 14:10-h light:dark cycle and received food and water ad libitum. Brain tissue for binding studies was obtained from male Sprague-Dawley rats (150-200 g) purchased from Harlan Laboratories (Dublin, VA).

Apparatus. Measurement of spontaneous activity in mice occurred in standard activity chambers interfaced with a Digiscan Animal Activity Monitor (Omnitech Electronics, Inc., Columbus, OH). A standard tail-flick apparatus [described by Dewey et al. (1970)] and a digital thermometer (Fisher Scientific, Pittsburgh, PA) were used to measure antinociception and rectal temperature, respectively.

Drugs. ⁹-THC (National Institute on Drug Abuse, Rockville, MD) and CP 55,940 (Pfizer, Groton, CT) were suspended in a vehicle of absolute ethanol, Emulphor-620 (Rhone-Poulenc, Inc., Princeton, NJ), and saline in a ratio of 1:1:18. SR141716A (National Institute on Drug Abuse) and novel pyrazole cannabinoids (synthesized in our laboratories) were also mixed in 1:1:18 vehicle. All drugs were administered to the mice intravenously in the tail vein at a volume of 0.1 ml/10 g.

Membrane Preparation and Binding. The methods used for tissue preparation and binding have been described previously (Compton et al., 1993) and are similar to those described by Devane et al. (1988). All assays, as described briefly below, were performed in triplicate, and the results represent the combined data from three to six individual experiments.

Procedure. Before testing in the behavioral procedures, mice were acclimated to the experimental setting (ambient temperature 22-24°C) overnight. Preinjection control values were determined for rectal temperature and tail-flick latency (in seconds). For agonism tests, mice were injected intravenously with drug or vehicle and, 5 min later, were placed in individual activity chambers where spontaneous activity was measured for 10 min. Activity was measured as total number of interruptions of 16 photocell beams per chamber during the 10-min test and was expressed as percentage inhibition of activity of the vehicle group. Tail-flick latency was measured at 20 min postinjection. A maximum latency of 10 s was used. Antinociception was calculated as percentage of maximum possible effect {%MPE = [(test  control latency)/(10  control)] × 100}. Control latencies typically ranged from 1.5 to 4.0 s. At 30 min postinjection, rectal temperature was measured. This value was expressed as the difference between control temperature (before injection) and temperatures following drug administration (°C). Different mice (n = 5-6) were tested for each dose of each compound. Each mouse was tested in each of the three procedures. Antagonism tests were conducted using an identical procedure with the exception that the antagonist analog was injected 10 min before the injection of 3 mg/kg ⁹-THC.

Data Analysis. Based on data obtained from numerous previous studies with cannabinoids, maximal cannabinoid effects in each procedure were estimated as follows: 90% inhibition of spontaneous activity, 100% MPE in the tail-flick procedure, and 6°C change in rectal temperature. ED₅₀ values were defined as the dose at which half-maximal effect occurred. For drugs that produced one or more cannabinoid effect, ED₅₀ values were calculated separately using least-squares linear regression on the linear part of the dose-effect curve for each in vivo measure, plotted against log₁₀ transformation of the dose. For the purposes of potency comparison, potencies were expressed as millimoles per kilogram. Data collected during combination tests (analog dose + 3 mg/kg ⁹-THC) were converted to percentage antagonism [(mean score of group that received vehicle and 3 mg/kg ⁹-THC  score obtained with analog dose and 3 mg/kg ⁹-THC)/(mean score of group that received vehicle and 3 mg/kg ⁹-THC) × 100]. When the resulting values showed dose-responsiveness, AD₅₀ values were calculated separately using least-squares linear regression on the linear part of the percentage antagonism curve for each in vivo measure, plotted against log₁₀ transformation of the dose. For the purposes of potency comparison, antagonist potencies were expressed as millimoles per kilogram.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Binding Affinities. Table 1 shows binding affinities for pyrazole analogs in which the carboxyamide group of the 3-substituent of SR141716A was replaced with an alkylether group. Substitution of an alkylether group for the carboxyamide group with retention of the terminal piperidine group, as in O-848, greatly decreased binding affinity for CB₁ receptors. Although affinity was improved (compared with O-848) by substitution of various cyclic, bicyclic, or tricyclic structures for the piperidine ring of O-848, most compounds listed in Table 1 still had relatively little affinity for the CB₁ receptor (K_i > 100 nM). Notable exceptions were O-852, O-889, and O-1043, each of which had CB₁ affinity <100 nM. In addition to substitution of an alkylether for the carboxyamide at position 3 on the pyrazole core (as with all compounds in this series), these compounds had substitutions of naphthalene (O-852), 4-fluorophenyl (O-889), and 2,4-difluorophenyl (O-1043) groups for the piperidine of the parent compound, SR141716A. Nevertheless, the CB₁ affinities of these three compounds were substantially less than that of SR141716A.

View this table: [in this window] [in a new window]   TABLE 1 Pharmacological effects of 1-(2,4-dichlorophenyl)-4-methyl-5-(4-chlorophenyl)-1H-pyrazoles with cyclic 3-substituents For Tables 1-4: # indicates the maximal effect that was produced by the analog and the dose (mg/kg) at which it occurred in parentheses. ">dose" indicates that 50% activity was not achieved at this dose (mg/kg), which was the highest dose of the compound that was tested. All ED50 values are expressed as µmol/kg (with 95% confidence limits in parentheses).

View this table: [in this window] [in a new window]   TABLE 2 Pharmacological effects of 1-(2,4-dichlorophenyl)-4-methyl-5-(4-chlorophenyl)-1H-pyrazoles with carbon chain 3-substituents

View this table: [in this window] [in a new window]   TABLE 3 Pharmacological effects of N-(piperidin-1-yl)-5-(4-chloro-phenyl)-3-carboxamide-4-methyl-1H-pyrazole with various 1-substituents

View this table: [in this window] [in a new window]   TABLE 4 Pharmacological effects of N-(piperidin-1-yl)-1-(2,4-dichloro-phenyl)-3-carboxamide-4-methyl-1H-pyrazole with various 5- and 4-substituents

Structure-Activity Relationship for Agonist Activity in Mice. 3-Substituent substitution of an alkylether group for the amide and various cyclic structures for the piperidine of SR141716A resulted in analogs that engendered slight in vivo cannabimimetic effects. Minor activity (30-70% of maximum effect) was observed with several compounds (Table 1). The most potent cannabimimetic activity in this series was produced by a compound (O-889) with a 3-substituent substitution of a p-fluorophenyl methoxy group. O-889 had full or partial activity in all three assays and also had one of the highest CB₁ receptor affinities in the series (Table 1). In addition, O-889 stimulated locomotor activity by about 30% at a dose lower than those that produced suppression of locomotor activity (Table 5). O-852 also stimulated locomotion by 52%, but unlike O-889, this compound did not inhibit locomotor activity at higher doses, nor was it active in the antinociceptive or hypothermia assays.

View this table: [in this window] [in a new window]   TABLE 5 Maximum stimulation and inhibition of spontaneous locomotor activity by pyrazole analogs Maximum (max) % stimulation (stim) and % inhibition (inhibit) produced by pyrazole analogs tested alone (left) and tested in combination with 3 mg/kg 9-THC (right). Dose(s) (mg/kg) at which the effect occurred are given in parentheses.

Structure-Activity Relationship for Antagonist Activity in Mice. To assess antagonist activity, SR141716A and its analogs from each series with good binding affinities (<100 nM) were tested in combination with an active dose of ⁹-THC (3 mg/kg i.v.). The results of these tests are presented in Table 6. As expected, SR141716A fully antagonized the suppression of locomotor activity, antinociceptive, and hypothermic effects induced by 3 mg/kg ⁹-THC. Analogs with 3-substituent substitutions produced partial antagonism at best and often were ineffective. Maximum antagonist activity was obtained with O-1271, which produced an average of 59% antagonism across the three measures and did not have agonist properties at doses up to 30 mg/kg. With the exception of O-1253, 1-substituent substitution also did not result in marked antagonist activity. Interestingly, O-1253 also had the highest CB₁ binding affinity in this series of compounds. When tested in combination with 3 mg/kg ⁹-THC, O-1253 produced full, dose-dependent antagonism of the antinociceptive and hypothermic effects of this dose of ⁹-THC, but antagonized its locomotor suppressant effects only at a single dose (1 mg/kg) with stimulation at higher doses and no antagonism at lower doses.

View larger version (25K): [in this window] [in a new window]   Fig. 2.   Effects of 5-substituted pyrazole analogs of SR141716A on percentage antagonism of the antinociceptive (left panels) and hypothermic (right panels) effects of a 3 mg/kg (i.v.) dose of 9-THC in mice. Chemical structures of each analog are presented in Table 4, and AD50 values are provided in Table 6.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

SR141716A binds to CB₁ receptors and competitively antagonizes many of the CB₁ receptor-mediated effects of cannabinoids; hence, its structure would be expected to contain regions of overlap with those of cannabinoid agonists. An area of receptor recognition that is crucial for all known CB₁ agonists is a lipophilic side chain (e.g., THC and anandamides) or comparable moiety (e.g., nitrogen substituent of indole-derived cannabinoids) (Martin et al., 1991; Huffman et al., 1994; Thomas et al., 1996; Wiley et al., 1998). Changes in the length, branching, and flexibility of this side chain affects CB₁ receptor binding affinity and in vivo potency of cannabinoid agonists (Compton et al., 1993; Huffman et al., 1997; Martin et al., 1999). A goal of this study was to determine whether any of the pyrazole substituents of SR141716A might correspond to the C3 side chain of ⁹-THC.

Molecular modeling suggests a possible superpositioning of the para-position of the 5-substituent in SR141716A with the pentyl side chain in ⁹-THC (Thomas et al., 1998). Structure-activity relationships (SAR) of SR141716A analogs presented here and elsewhere (Thomas et al., 1998; Lan et al., 1999) are consistent with this proposed alignment. Retention of the phenyl group is critical for receptor affinity and antagonism, as illustrated with O-1559, which had an alkyl group at position 5 rather than a phenyl. Substitution of the para-portion of the phenyl substituent is also important. Deletion of the p-chloro group (Lan et al., 1999) greatly decreased affinity, whereas substitution of an alkyl group or an iodo/bromo (Thomas et al., 1998) enhanced affinity. Interestingly, lengthening of the pentyl side chain of ⁸-THC (Martin et al., 1999), methylation at the first or second carbon of the chain (Huffman et al., 1997), and halogenation at the terminal end of the chain (Charalambous et al., 1991) resulted in analogs that were agonists in vivo and that had enhanced CB₁ affinity compared with the parent compound.

Although all of the p-pentylphenyl analogs of SR141716A (Table 4) have good affinity for CB₁ receptors, none of these analogs show cannabinoid activity in vivo. Indeed, with the exception of O-1710 (the phenyl analog with the least affinity), all are potent antagonists of the antinociceptive and hypothermic effects of ⁹-THC. Presumably, they also will block activation of CB₁ receptors, although this hypothesis has yet to be tested in functional assays. Hence, the 5-substituent of pyrazole cannabinoids appears to be involved both in receptor recognition and in antagonism of receptor activation. Consistent with the hypothesis that this position is important for receptor recognition, Howlett et al. (2000) have shown that covalent binding of an azido or isothiocyano group to the p-position of the 5-phenyl ring of SR141716A irreversibly displaces [H³]CP 55,940 from its binding site.

Consistent with the proposed overlap of the C3 side chain of ⁹-THC and the 5-substituent of SR141716A, position 4 of the pyrazole core would correspond with either C2 or C4 of ⁹-THC (see Fig. 1). Addition of an iodine or bromine at this position of the p-pentylphenyl analog of SR141716A did not substantially alter affinity, whereas hydrogen substitution (O-1710) decreased it. By comparison, halogenation of C2 of ⁸-THC resulted in agonist analogs with decreased CB₁ affinity, and halogenation of C4 produced inactive analogs with little affinity (Martin et al., 1993). Based on these results, we suggest alignment of position 4 with C2 of ⁹-THC; however, given the paucity of substitutions that have been made at these positions, this suggestion is tentative, pending the results of further SAR studies.

Another area likely to be involved in the antagonist actions of SR141716A is the 1-substituent. Thomas et al. (1998) have suggested that the 2,4-dichlorophenyl of SR141716A is its most unique area compared with ⁹-THC, and that it may represent the "antagonist-conferring" region. To date, results of SAR studies support this hypothesis. Manipulation of this area by removal of one or both of the chlorine atoms (present study; Lan et al., 1999), addition of a 3-chloro or 3- or 6-iodo group (Thomas et al., 1998), or substitution of an n-alkyl chain for the p-chloro group (present study) resulted in substantial decreases in CB₁ affinity and decreased potency or loss of antagonism. Of the analogs presented here, only the branched p-1'-methylpropylphenyl analog (O-1253) had reasonable binding affinity and antagonist activity, although this analog still had less affinity than SR141716A and it was not as potent an antagonist. Although the positioning of the two chlorine atoms is important in determining the CB₁ affinity of these 1-substituent analogs, the presence of the 1-phenyl group is crucial for their antagonist activity. Replacement of the phenyl with an alkyl chain resulted in analogs that were partial agonists in a [³⁵S]GTPS assay of G-protein activation (Houston et al., 1997). In contrast, analogs in which an alkyl chain was added to the phenyl group at the p-position showed decreased affinity and were inactive in vivo (present study). Together, these findings demonstrate that small changes in the structure in the 1-substituent result in loss of antagonism, lending support to the hypothesis that this area is important in conferring receptor recognition and antagonist activity to pyrazole cannabinoids.

The 3-substituent of the pyrazole core, the fourth area of SR141716A that was manipulated in the present study, appears to be involved in receptor recognition, as analogs that were ethers, alkyl amides, ketones, alcohol, or alkane showed greatly decreased CB₁ binding affinity. These results are in agreement with those of Lan et al. (1999). Only three of these analogs showed CB₁ binding affinity of less than 100 nM: naphthalene, 4-fluorophenyl, and 2,4-difluorophenyl substitutions. The other 3-substituent analogs that showed reasonable binding affinity were some of the alkyl amides and ketones, with the best binding affinity exhibited by the n-pentyl and n-heptyl amides and the n-pentyl ketone. It is noteworthy that, in each of the pairs of alkylamides and ketones, the analog with n-pentyl substitution had the best affinity, suggesting that substituent length affected CB₁ receptor binding in both series. Although none of the 3-substituent analogs that were tested completely blocked ⁹-THC's effects in all assays, several were agonists or partial agonists in vivo, although most were not as efficacious as ⁹-THC in producing the full profile of cannabimimetic effects. In addition, all of the active 3-substituent analogs were less potent than ⁹-THC, even though some of them had approximately the same affinity for the CB₁ receptor. Together, these results suggest that the 3-substituent region is involved in receptor recognition and agonist activity.

A final issue examined in this study was the degree to which pyrazoles produce their effects through inverse agonism. SR141716A produces effects that have been considered possible indications of inverse agonism, including stimulation of locomotor activity in mice (Compton et al., 1996), inhibition of G-protein-gated inwardly rectifying potassium channels in Xenopus oocytes (McAllister et al., 1999), reduction in [³⁵S]GTPS binding (Landsman et al., 1997), and increased twitch response in guinea pig ileum (Coutts et al., 2000). In the present study, substantial locomotor stimulation was observed with some analogs, particularly those with 1- and 5-substituent substitutions. Because these analogs also showed the most antagonist activity, it is tempting to speculate that this antagonism may have resulted from inverse agonism; however, several observations argue against this hypothesis. First, ⁹-THC produces a biphasic effect on locomotor activity with stimulation at lower doses and suppression at higher doses (Evans et al., 1976). It is possible that any locomotor stimulation may be related to the agonist or partial agonist activity of some of these analogs. Second, the locomotor stimulation does not appear to be correlated with the CB₁ affinity of these analogs nor with their potency for antagonizing the in vivo effects of ⁹-THC. For example, the greatest degree of stimulation was produced by O-1559; yet, this analog did not have good CB₁ affinity nor was it an antagonist in vivo. These results suggest that the stimulatory effect that we observed with some of these analogs is not strongly related to action at the CB₁ receptor.

In summary, this study was undertaken to examine the pharmacological profile of various SR141716A analogs in which the 1-, 3-, 4-, and 5-positions of the pyrazole core were replaced by substituents known to impart potent agonist activity in tetrahydrocannabinols. Our results suggest that, although all three positions are important for receptor recognition, the effects of the positions differ with respect to receptor activation. The 3-position appears to be involved in agonism and receptor activation. In contrast, the 1-, 4-, and 5-positions seem to be involved in antagonism. Furthermore, the present evaluation of locomotor stimulatory effects does not support the hypothesis that the antagonist activity of pyrazole cannabinoids is related to inverse agonism. In conclusion, the present results suggest that binding and activation of the cannabinoid CB₁ receptor are separable events and that the structural properties of 1- and 5-substituents are primarily responsible for the antagonist activity of SR141716A.