Synthesis of long-chain amide analogs of the cannabinoid CB1 receptor antagonist N-(piperidinyl)-5-(4-chlorophenyl)-1-(2,4-dichlorophenyl)-4-methyl- 1H-pyrazole-3-carboxamide (SR141716) with unique binding selectivities and pharmacological activities

Article abstract of DOI:10.1016/j.bmc.2005.06.005

An extended series of alkyl carboxamide analogs of N-(piperidinyl)-5-(4- chlorophenyl)-1-(2,4-dichlorophenyl)-4-methyl- 1H-pyrazole-3-carboxamide (SR141716; 5) was synthesized. Each compound was tested for its ability to displace the prototypical cannabin

Full text of DOI:10.1016/j.bmc.2005.06.005

Bioorganic & Medicinal Chemistry 13 (2005) 5463–5474
Synthesis of long-chain amide analogs of the cannabinoid
CB1 receptor antagonist N-(piperidinyl)-5-(4-chlorophenyl)-1-(2,4-
dichlorophenyl)-4-methyl-1H-pyrazole-3-carboxamide (SR141716)
with unique binding selectivities and pharmacological activities
Brian F. Thomas,a,* Ma. Elena Y. Francisco,a Herbert H. Seltzman,a
James B. Thomas,a Scott E. Fix,a Anne-Kathrin Schulz,a Anne F. Gilliam,a
Roger G. Pertweeb and Leslie A. Stevensonb
aScience and Engineering Group, Research Triangle Institute, Research Triangle Park, NC 27709, USA
bDepartment of Biomedical Sciences, Institute of Medical Sciences, University of Aberdeen, Foresterhill,
Aberdeen AB25 2ZD, Scotland, UK
Received 18 November 2004; revised 31 May 2005; accepted 1 June 2005
Available online 1 July 2005
Abstract—An extended series of alkyl carboxamide analogs of N-(piperidinyl)-5-(4-chlorophenyl)-1-(2,4-dichlorophenyl)-4-methyl-
1H-pyrazole-3-carboxamide (SR141716; 5) was synthesized. Each compound was tested for its ability to displace the prototypical
cannabinoid ligands ([3H]CP-55,940, [3H]2; [3H]SR141716, [3H]5; and [3H]WIN55212-2, [3H]3), and selected compounds were fur-
ther characterized by determining their ability to affect guanosine 50-triphosphate (GTP)-c-[35S] binding and their effects in the
mouse vas deferens assay. This systematic evaluation has resulted in the discovery of novel compounds with unique binding prop-
erties at the central cannabinoid receptor (CB1) and distinctive pharmacological activities in CB1 receptor tissue preparations. Spe-
cifically, compounds with nanomolar affinity which are able to fully displace [3H]5 and [3H]2, but unable to displace [3H]3 at similar
concentrations, have been synthesized. This selectivity in ligand displacement is unprecedented, in that previously, compounds in
every structural class of cannabinoid ligands had always been shown to displace each of these radioligands in a competitive fashion.
Furthermore, the selectivity of these compounds appears to impart unique pharmacological properties when tested in a mouse vas
deferens assay for CB1 receptor antagonism.
ꢀ 2005 Published by Elsevier Ltd.
1. Introduction
these receptors. For example, CB1 receptors are also
coupled to several types of potassium7,8 and calcium
channels,9 and thereby inhibit the release of several neu-
rotransmitters.10–13 Despite the euphoria production
and ÔhighÕ associated with agonists at the CB1 receptor,
these compounds have therapeutic utility as antiglauco-
ma agents, antiemetics or appetite stimulants, and ther-
apeutic potential as analgesics, or for the treatment of
symptoms of multiple sclerosis, and other indications.
Agonists at the CB2 receptor may hold promise as
immunomodulators.14 Similarly, CB1 antagonists are
being investigated for their therapeutic utility in treating
obesity and improving memory.15
There have been at least two types of cannabinoid recep-
tors characterized to date, CB11 and CB2.2 CB1 recep-
tors are found primarily in brain and neuronal tissue,
while CB2 receptors are found predominantly in im-
mune cells and tissues. Both are G-protein-coupled
receptors,3 sharing approximately 48% homology in
their amino acid sequences,4 and both responding to li-
gand binding through coupling with mitogen-activated
protein kinase5 and adenylyl cyclase.6 Other signal
transduction pathways have been demonstrated for
In addition to binding the naturally occurring cannabi-
noids such as D9-THC (1), the CB1 and CB2 receptors
can bind other structural classes of compounds, includ-
ing nonclassical cannabinoids (CP55940, 2), aminoalky-
Keywords: Cannabinoid; Ligand-receptor; Binding; Structure–activity
relationships; CB1; Antagonist.
*
Corresponding author. Tel.: +1 9195416552; fax: +1 9195416499;
e-mail: bft@rti.org
0968-0896/$ - see front matter ꢀ 2005 Published by Elsevier Ltd.
doi:10.1016/j.bmc.2005.06.005

5464
B. F. Thomas et al. / Bioorg. Med. Chem. 13 (2005) 5463–5474
lindoles (WIN55212-2, 3), arachidonoylethanolamides
(anandamide, 4), and diarylpyrazoles (SR141716, 5 or
SR144528, 6; Fig. 1). It has been suggested that all of
these compounds interact as competitive ligands at a
common or overlapping/interacting binding region on
the CB1 receptor, since heterologous displacement
curves have consistently been demonstrated between
these various ligand classes.16 Indeed, overlapping
superposition models have been hypothesized by several
investigators for nonclassical and classical cannabi-
noids,17,18 arachidonoylethanolamides and classical can-
sites that are being occupied by these compounds and
the radioligands are significantly different.
More recently, we published on a series of amide ana-
logs of the CB1 receptor antagonist 5.28 These com-
pounds included straight chain alkyl, hydroxy alkyl,
and hydrazide analogs up to six carbons in length. Ini-
tially, we elected to characterize these compoundÕs
binding affinities at the CB1 receptor using [3H]2 and
[3H]5. However, during subsequent experiments with
[3H]3 in rat brain CB1 (rCB1) preparations, subtle dif-
ferences were noted in the displacement curves ob-
tained with the three radioligands. This observation
led to the synthesis of alkyl side-chain analogs of even
greater length. Furthermore, we examined the binding
profiles of these compounds in cells transfected with
the human CB1 receptor and in human brain prepara-
tions. It was found that these compounds clearly pos-
sess unique binding selectivities, particularly in
human CB1 receptor (hCB1) preparations, representing
the first clear demonstration of nonhomologous dis-
placement curves between these three radioligands.
Specifically, some long-chain alkyl amide analogs can
displace [3H]5 and [3H]2 with reasonably high affinity,
but are unable to displace [3H]3. The demonstration
of nonheterologous displacement curves with these
radioligands agrees with previous hypotheses that 3
binds with a distinct recognition site, or in a unique
manner within an overlapping recognition site, com-
pared to 2 or 5 (as described later). By extension, the
binding of the ÔWIN-sparingÕ alkyl analogs of 5 is also
unique compared to the prototype antagonist 5. How-
ever, it remains to be determined whether this distinct
mode of binding imparts unique pharmacological
properties.
nabinoids,19–22
aminoalkylindoles
and
classical
cannabinoids,23–26 and diarylpyrazole antagonists and
classical cannabinoid agonists,16,27 and finally, a super-
position strategy for nonclassical, aminoalkylindole
and classical cannabinoids has been described.16
While these superpositions represent reasonable hypoth-
eses, in 1998, our laboratory had noted differences in the
ability of a variety of compounds to compete for the rat
brain CB1 receptor site when labeled with [3H]5 as com-
pared to when the site was labeled with [3H]2.27 For
example, the affinity of 2 differed by more than 38-fold
depending on what radioligand was used to determine
its Ki. WIN55212-2 (3) also had a greater ability to com-
pete for the receptor when labeled with [3H]2 as com-
pared to [3H]5. In contrast, 5 was more than 5-fold
more effective when competing with [3H]5 than when
competing with [3H]2. We hypothesized at that time that
the differences in Ki values of these compounds suggest
that their mode of binding or the population of binding
2. Results
2.1. CB1 receptor affinity in rat brain
All of the compounds were tested for their ability to dis-
place [3H]2, [3H]5, and [3H]3 in rat brain membrane
preparations (Table 1, Fig. 2). The CB1 receptor binding
data in rat brain for this series of analogs demonstrated
that as the size of the carbon chain is increased from C4
to C5, a slight increase in binding affinity is observed,
there is a modest decrease at C6, followed by a slightly
greater decrease at C7. Beyond this length, there appears
to be a further decrease in affinity. This pattern is not as
obvious with the branched alkyl amides; however, gen-
erally the pentyl and hexyl amides demonstrated the
highest affinity, with the heptyl and decyl analogs having
lower affinity. In addition, the data obtained in rat brain
suggested that as the carbon chain extended beyond C5
or C6, both the affinity and the ability to fully displace
[3H]3 decreased more rapidly than with [3H]2 or [3H]5.
In addition to these observations, characterization in
mouse vas deferens studies with 3 also indicated that
these compounds were pharmacologically unique (de-
scribed in greater detail below), and this encouraged
us to extend our observations into other CB1 receptor
preparations.
Figure 1. Structure of selected cannabinoid ligands.

B. F. Thomas et al. / Bioorg. Med. Chem. 13 (2005) 5463–5474
5465
Table 1. Amide analogs at the 3-position: displacement of various radioligands in whole rat brain (CB1) membrane preparations
Compound
Substituted group
[3H]2
[3H]5
[3H]3
Ki (nM) SEM
Maximum
displacement
(%)
Ki (nM) SEM Maximum
displacement
(%)
Ki (nM) SEM Maximum
displacement
(%)
a
5 (SR141716)
6 (SR144528)
6.18
74.1
2.46
13.4
1.2
11.4
86.6
84.7
1.18
81.7
1.07
12.8
6.83
10.7
0.10
19.3
96.5
68.6
1.97
2.23
85.7
8b
N-(1-Cyclohexyl)
N-(1-Butyl)
N-(1-Pentyl)
0.10 96.1
1.0 91.9
0.49 91.4
0.13 100
0.15 ꢀ92.6
9b
4.5
1.2
0.62
1.2
6.0
5.5
100
100
32.8
26.0
4.5
4.4
2.8
2.7
90.8
94.4
77.9
69.6
68.4
68.6
67.4
90.3
56.5
80.5
89.0
89.9
94.1
89.0
87.7
85.1
53.7
72.2
82.4
10b
11b
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
11.4
18.2
46.2
12.6
16.6
28.8
36.2
68.2
N-(1-Hexyl)
3.9
13.9
16.7
87.3
81.7
80.2
79.5
83.6
86.4
82.6
79.3
84.2
91.4
76.6
98.2
89.6
89.5
83.9
83.2
83.7
98.6
N-(1-Heptyl)
N-(1-Octyl)
19.0
37.4
88.2
45.0
25.5
53.7
97.9
75.2
85.8
92.5
83.8
95.6
97.6
95.7
96.3
96.2
96.5
95.1
95.7
91.3
76.4
97.3
135
333
92.8
N-(1-Nonyl)
N-(1-Decyl)
128
4.7
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
1393
a
a
a
a
a
a
a
a
a
a
N-(1-Undecyl)
N-(1-Dodecyl)
N-(1-Tridecyl)
Geranylamine
168
109
492
278
238
495
172
48.1
9.6
20.1
2.3
40.5
N-(1,3-Dimethylpentyl)
N-(1,4-Dimethylpentyl)
N-(1-Methylhexyl)
N-(1,5-Dimethylhexyl)
N-(1-Ethylhexyl)
N-(1-Methylheptyl)
13.7
28.6
3.37
2.81
10.7
8.97
19.6
26.2
2.60
3.57
8.6
1.84
5.76
25.8
10.8
5.8
10.4
22.3
N-[1-(1,1-Dimethylheptyl)] 181
N-(1-Methyldecyl) 27.1
5.8
21.6
9.7
3.3
3.3
a
a
a
a n = 1, compound obtained from parallel synthesis and assay concentration was grossly estimated.
b Some results for this compound have been published previously (see Ref. 28).
2.2. Human CB1 receptor affinity
similar results were obtained (Fig. 4). Thus, all evidence
obtained to date supports the conclusion that these
long-chain alkyl amides are unique from all other classes
of cannabinoid receptor ligands with regard to their
ability to displace the various radioligands tested. The
data also illustrate that despite a very high degree of se-
quence homology, the alkyl amides beyond C6 interact
at rat and hCB1 receptors quite differently.
In hCB1-transfected cells, affinity modestly increased
from C4 to C5 side chain. As the side chain was in-
creased beyond C5, affinity tended to decrease up to
C9, where a slight increase in affinity was noted up to
C11. Beyond this length, further extension led to de-
creased affinity. As was seen in rat brain, the branched
alkyl amides showed higher affinity within the pentyl
and hexyl analogs than that seen with the heptyl or decyl
branched side chains. Examination of the alkyl amide
analogs using transfected cell lines expressing the hCB1
receptor showed an even greater discrepancy in the abil-
ity of these compounds to displace [3H]3 as compared to
either [3H]2 or [3H]5 (Table 2, Fig. 3). It is interesting to
note that there is a structural trend by which systematic
increases in the alkyl side chain length past C6 leads to
an incremental decrease in affinity and percentage max-
imum displacement for [3H]5, a more pronounced loss
of affinity and percentage maximum displacement with
[3H]2, and a profound loss of affinity and displacing
ability with [3H]3. The difference between the rat brain
membrane preparation and the hCB1 receptor transfects
is quite remarkable when one considers that these two
CB1 receptors possess approximately 99% homology
in their sequences.
2.3. Human CB2 receptor affinity
Those compounds that were obtained by conventional
synthesis were further tested for their affinity at the
CB1 receptor (Table 3). In most instances, these com-
pounds were unable to completely displace [3H]2 from
the CB2 receptor at the highest concentration tested,
indicating that the relative selectivity of the analogs
for CB1 versus CB2 receptors was relatively unaffected
by structural modifications at this substituent position.
Indeed, the absence of CB2-selective ligands in our series
of analogs is consistent with the observation that the
CB2-selective antagonist 6, when compared to 5, has
structural modifications in addition to the change at
the aminopiperidine moiety that was explored in these
studies.
2.4. Cannabinoid receptor-mediated alteration in GTP-c-
It was possible that the augmentation in these com-
poundsÕ binding selectivity in hCB1 transfects could be
a result of the nature of transfected cell lines and not
due to the differences in receptor sequence and structure.
However, when these compounds were tested in human
brain membrane preparations (cerebellum and cortex),
[35S] binding
In addition to assessing CB1 and CB2 receptor affinities,
several compounds were screened in a GTP-c-[35S] assay
for characterization of the compounds as agonists, par-
tial agonists, antagonists, or inverse agonists. The results

5466
B. F. Thomas et al. / Bioorg. Med. Chem. 13 (2005) 5463–5474
Figure 2. Displacement curves for SR141716A analogs in rat brain membrane against [3H]CP55940 (top), [3H]SR141716A (middle) and
[3H]WIN55212 (bottom) for ( ) N-cyclohexyl, ( ) N-(1, 3-dimethylpentyl), ( ) N-(1, 4-dimethylpentyl), ( ) N-(1-methylhexyl), ( ) N-(1,5-
dimethylhexyl), ( ) N-(2-ethylhexyl), ( ) N-(1,1-dimethylheptyl), ( ) N-(1-methylheptyl), ( ) N-(1-methyldecyl) (left panel); and ( ) 1-geranyl, (h) N-
butyl, ( ) N-pentyl, ( ) N-hexyl, ( ) N-heptyl, ( ) N-octyl, ( ) N-nonyl, (+) N-decyl, ( ) N-undecyl, ( ) N-dodecyl, and ( ) N-tridecyl (right panel).
of GTP-c-[35S] studies (Table 4) demonstrate that, in all
instances, these compounds act like inverse agonists.
(46.2 nM vs. [3H]2 in rat brain), and producing a robust
inverse agonist effect in the absence of any application of
3. Thus, the data suggest that this compound can inter-
act at the receptor (displace [3H]2 and [3H]5) and pro-
duce inverse agonist effects, but not readily compete
for the WIN55212-2 (3) binding site, and thereby fail
to produce a dextral shift in the dose–response curve un-
til extremely high concentrations are used. This lends
pharmacological significance to the aforementioned
ÔWIN-sparingÕ displacement curves (Figs. 2 and 3) ob-
tained with the alkyl amide analogs. With extremely
high concentrations of 12 (>3 lM), the dose–response
curves for both 2 and 3 were shifted, albeit to a much
lesser extent than that achieved with the prototype
antagonist, 5.
When some of the alkyl amide compounds with the
highest affinity for the CB1 receptor were characterized
in an isolated mouse vas deferens, the results obtained
(Table 4) support the conclusion that some of the alkyl
amide analogs interact with the CB1 receptor in a un-
ique manner. Specifically, one can see from the data in
this table for the alkyl amides that as the chain length
was increased from the butyl analog to the heptyl ana-
log, the dextral shift in the dose–response curve of 3
was reduced. Indeed, in one experiment, the heptyl ana-
log (12) failed to produce an effect on the dose–response
curve, despite this compound having good affinity

B. F. Thomas et al. / Bioorg. Med. Chem. 13 (2005) 5463–5474
5467
Table 2. Amide analogs at the 3-position: displacement of various radioligands in human CB1 receptor-transfected cells
Compound
Substituted group
[3H]2
[3H]5
[3H]3
Ki (nM) SEM Maximum
displacement
(%)
Ki (nM) SEM Maximum
displacement
(%)
Ki (nM) SEM Maximum
displacement
(%)
a
a
a
5 (SR141716)
8c
3.92
7.06
79.5
0.76 92.3
2.43
1.02
4.79
6.35
85.7
99.9
97.4
89.9
75.4
67.9
67.8
86.9
95.4
90.4
89.8
87.3
95.7
4.67
5.36
77.9
96.8
93.2
89.9
52.7
53.4
35.2
28.3
12.4
74.6
31.4
68.1
78.3
78.7
91.5
100
N-(1-Cyclohexyl)
N-(1-Butyl)
N-(1-Pentyl)
0.22
0.89
2.4
0.24
15.6
5.9
9c
28.0
14.6
124
1.3
3.6
90.2
87.1
85.6
85.3
70.6
68.8
94.5
66.3
79.5
62.6
84.2
86.5
76.8
80.7
82.5
91.4
79.5
91.1
77.7
29.8
10c
11c
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
18.7
85.4
N-(1-Hexyl)
N-(1-Heptyl)
27.4
69.7
48.0
23.5
16.5
1.1
18.6
291
288
307
184
9.3
44.7
56.5
24.6
21.0
43.5
393
154
36.9
2826
N-(1-Octyl)
N-(1-Nonyl)
18.6
25.7
NDb
NDb
NDb
166
40.7
a
a
a
N-(1-Decyl)
N-(1-Undecyl)
N-(1-Dodecyl)
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
57.7
176
220
NDb
N-(1-Tridecyl)
Geranylamine
101
6.80
2.2
128
211
26.8
N-(1,3-Dimethylpentyl)
N-(1,4-Dimethylpentyl)
N-(1-Methylhexyl)
N-(1,5-Dimethylhexyl)
N-[1-(2-Ethyl)hexyl]
N-(1-Methylheptyl)
N-[1-(1,1-Dimethylheptyl)]
N-(1-Methyldecyl)
9.4
13.2
11.9
100
23.9
33.2
54.6
64.2
78.6
NDc
14.4
22.2
5.48
2.02
2.10
3.8
93.6
95.6
93.7
95.9
91.5
99.9
90.0
7.39
67.9
100
0.0
9.6
26.3
32.7
41.1
2.8
5.0
10.4
a
0.51
1.0
86.5
76.1
a
a
7.0
a n = 1, compound obtained from parallel synthesis and assay concentration was grossly estimated.
b The Ki value could not be determined since <50% displacement of the radiolabel occurred.
c Some results for this compound have been published previously (see Ref. 28).
Since compound 12 appeared to be unable to fully dis-
place [3H]3, it was further tested in GTP-c-[35S] experi-
ments to compare the ability of 5 and 12 to shift the
dose–response curve of CP55940 (2) or WIN55212-2
(3). The results of these assays are provided in Table
5. When the model is constrained to one where the
Schild slope is 1, the pA2 values are equal to the pKb
of the antagonist. In this instance, the pKb values are
consistent with the Ki values. For example, the pKi of
5 in rat brain is 8.21 against [3H]2 and 8.71 against
[3H]3, and in hCB1-transfected cells, it is 8.41 against
[3H]2 and 8.33 against [3H]3. These Ki values are in rea-
sonably close correspondence with the Kb values (Kb cal-
culated with a Schild slope constrained to 1) in rat brain
of 9.28 against [3H]2 and 9.49 against [3H]3, and 7.55
against [3H]2 and 7.79 against [3H]3 in hCB1-transfected
cells. Similarly, the pKi of 12 in rat brain is 7.34 against
[3H]2 and 7.54 against [3H]3, and in hCB1-transfected
cells, it is 6.54 against [3H]2 and 6.41 against [3H]3.
Comparing these Ki values to the Kb values in rat brain
of 9.00 against [3H]2 and 7.52 against [3H]3, and 7.24
against [3H]2 and 7.24 against [3H]3 in hCB1-transfected
cells. However, there does not appear to be a difference
in the ability of 12 to antagonize the effects of 2 as com-
pared to 3, as might have been expected based on 12Õs
selective binding displacement data.
branching of alkyl amide chains up to five carbons in
length results in modest increases in affinity at rat brain
CB1 receptor preparations, a trend that is also apparent
in hydroxy alkyl amide and hydrazide analogs. After the
pentyl analog, additional chain lengthening of alkyl
amides resulted in a further decrease in affinity. Howev-
er, as the carbon chain extended beyond C5 or C6, affin-
ity tends to decrease, and the ability to fully displace
[3H]3 decreases in a more pronounced fashion than ob-
served with [3H]2 and [3H]5. It is important to note that
the compounds with side chains extending beyond C9
(nonyl), and the majority of the branched alkyl amides,
were synthesized using parallel synthesis techniques.
While we used LC/UV/MS methods to ensure that the
yields were relatively consistent from one compound
to another, the concentrations of these compounds were
estimated based on approximately 50% recovery, as sug-
gested by the HPLC/UV response. Thus, substantive
conclusions regarding relative affinities within these
compounds should be avoided, but observations regard-
ing the percentage maximum displacement in binding
displacement studies are supportable, as long as the sig-
moidal curves show no further decrease with increasing
concentrations. Furthermore, characterization in mouse
vas deferens studies with 3 also indicated that these com-
pounds were pharmacologically unique, and this
encouraged us to extend our observations into other
CB1 receptor preparations. In CHO cells transfected
with hCB1 and in human postmortem tissue, this unique
selectivity was more pronounced, despite the high degree
of homology between the two receptors (hCB1 and
rCB1). This selectivity in displacement has not previous-
ly been reported for a cannabinoid compound in
3. Discussion
The nature of the substituent at the 3-position has a
marked effect on receptor binding affinity. Previous re-
search has shown that increasing the length and

5468
B. F. Thomas et al. / Bioorg. Med. Chem. 13 (2005) 5463–5474
Figure 3. Displacement curves for SR141716A analogs in hCB1-transfected cells against [3H]CP55940 (top), [3H]SR141716A (middle), and
[3H]WIN55212 (bottom) for ( ) N-cyclohexyl, ( ) N-(1, 3-dimethylpentyl), ( ) N-(1, 4-dimethylpentyl), ( ) N-(1-methylhexyl), ( ) N-(1,5-
dimethylhexyl), ( ) N-(2-ethylhexyl), ( ) N-(1,1-dimethylheptyl), ( ) N-(1-methylheptyl), ( ) N-(1-methyldecyl) (left panel); and ( ) 1-geranyl, (h) N-
butyl, ( ) N-pentyl, ( ) N-hexyl, ( ) N-heptyl, ( ) N-octyl, ( ) N-nonyl, (+) N-decyl, ( ) N-undecyl, ( ) N-dodecyl, and ( ) N-tridecyl (right panel).
wildtype CB1 receptors. That is, compounds within
structural classes that bind to the CB1 receptor (i.e.,
the aminoalkylindole agonists such as 3, the bicyclic
cannabinoids such as 2, and the pyrazole inverse ago-
nists such as 5) had always been shown to displace each
other in a competitive fashion.
receptors were transfected into human embryonic kid-
ney (HEK293) cells and used for ligand binding and
cAMP accumulation studies. The affinity of 3 for the
CB2F5.46V mutation was decreased by 14-fold, whereas
the CB(1)V5.46F mutation increased the affinity of 3 for
CB1 by 12-fold. However, these mutations did not
change the affinity of HU-210, 2, or 4. Similarly,
Lys192 of the third transmembrane domain of the
CB1 receptor has been mutated to either Arg
(K192R), Gln (K192Q), or Glu (K192E) receptors and
expressed in Chinese hamster ovary cells.32 In this in-
stance, only the Lys to Arg mutation allowed retention
of binding affinity to 2, whereas 3 bound to all the mu-
tant receptors in the same range as it bound to the wild-
type. More recently, Picone et al. reported at the 2002
Evidence of unique binding domains for structurally
divergent cannabinoid agonists has previously been pro-
vided only by receptor mutagenesis studies. For exam-
ple, Song and Bonner used site-directed mutagenesis of
the amino acids at position 5.46 of CB1 and CB2; a
Phe to Val mutation at the position 5.46 in CB2
(CB2F5.46V), and a corresponding Val to Phe mutation
at the position 5.46 in CB1 (CB1V5.46F).31 The mutant

B. F. Thomas et al. / Bioorg. Med. Chem. 13 (2005) 5463–5474
5469
Figure 4. Presented in the top panel are [3H]WIN55212-2 competition curves for SR141716A (solid symbols) and the N-(1-heptyl)-analog of
SR141716A (12, open symbols) as determined in rat brain (black symbols), human CB receptor-transfected cells (blue symbols), and human
cerebellar membrane preparations (red symbols). In the bottom panel, displacement curves in human cerebellum for the N-(1-heptyl)-analog are
shown as determined with [3H]CP55940 (h), [3H]SR141716A ( ), and [3H]WIN55212-2 ( ). In the inserts, the Ki and % maximum displacement are
provided as determined for each curve. In those instances where <50% displacement occurred with a particular ligand, the Ki was not determined
(ND).
meeting of the International Cannabinoid Research
Society that mutation of the CB1 receptor amino acid
residue C6.47(355) to S, A, I, and K impacts the recep-
tor binding affinity of [3H]CP55940, while leaving the
binding of [3H]WIN55212-2 unaffected.33 All of these
studies support the hypothesis that WIN55212-2 binds
differently to the CB1 receptor than CP55940 and possi-
bly other agonists and antagonists, which is consistent
with the displacement curves we have described for the
long-chain alkyl amides.
receptor interactions as compared to SR141716. Indeed,
it is possible that these compounds have other pharma-
cological properties that are unique from SR141716 or
other antagonists that displace all other cannabinoid li-
gands with equal capacity. The selective displacement of
the agonists [3H]2 and [3H]3 seen with 12 as compared to
5 did not appear to translate into observable differences
in antagonism of these agonists when tested in the GTP-
c-[35S] assay. However, in the mouse vas deferens exper-
iments, concentrations of 12 that possessed inverse
agonist activity when tested alone, appeared less able
to block the effects of either agonist when compared to
5. This discrepancy in pharmacological antagonism/in-
verse agonist activity may be due to several factors.
The heterologous displacement curves obtained with the
long-chain alkyl amides suggest that these ÔWIN-spar-
ingÕ compounds also appear to be unique in their

5470
B. F. Thomas et al. / Bioorg. Med. Chem. 13 (2005) 5463–5474
Table 3. Amide analogs at the 3-position: displacement of [3H]CP55940 in human CB2 receptor-transfected cells
Compound
Substituted group
[3H]2
Ki (nM)
SEM
Maximum displacement (%)
Ratio CB1/CB2
a
5 (SR141716)
6 (SR144528)
313
79.8
ND
32.3
57.1
76.0
55.4
13.8
342
4.92
0.39
1.5
88.4
84.5
75.6
75.9
66.9
47.8
45.3
53.8
52.8
8b
N-(1-Cyclohexyl)
N-(1-Butyl)
228
1598
9b
425
10b
11b
12
13
14
26
N-(1-Pentyl)
N-(1-Hexyl)
1110
6873
241
a
N-(1-Heptyl)
N-(1-Octyl)
4027
174
98,366
22,878
16,555
92,734
17,622
3445
N-(1-Nonyl)
N-[1-(1,1-Dimethylheptyl)]
74.5
506
a n = 1.
b Some results for this compound have been published previously (see Ref. 28).
Table 4. Amide analogs at the 3-position: inverse agonist/antagonist activity
Compound
Substituted group
GTP-c-[35S] in whole rat brain
Mouse vas deferens tissue assay
EC50 (nM)
SEM
Emax
Agonist used
[nM]
31.6
Kb (nM)
Dextral shift
81.4
% Inverse effect
65.2
5 (SR141716)
6 (SR144528)
56,305
8136
14,330
258
ꢀ37.8
ꢀ27.7
ꢀ22.1
ꢀ38.5
ꢀ7.4
3
0.4
8
N-(1-Cyclohexyl)
N-(1-Butyl)
26,030
8536
—
3
3
3
3
3
3
3
2
31.6
316.2
316.2
31.6
1.2
26.7
15.9
11.0
3.1
55.5
NSc (28.2)
9
3134
1656
16,135
3950
21.2
31.7
15.4
NS
83.9b
100.4
157.1
10
11
12
N-(1-Pentyl)
N-(1-Hexyl)
N-(1-Heptyl)
5270
29,375
212,950
86.6
NS (30.1)
50.6
NS (37.8b)
49.3
ꢀ13.0
ꢀ20.0
316.2
316.2b
NS
4.8b
32.5
21.1
3162
3162
NS (15.0)
13
14
N-(1-Octyl)
N-(1-Nonyl)
676,800a
293,500
187,400
60,400
ꢀ21.3
ꢀ20.1
a Value is above highest concentration on displacement curve.
b Due to unexpected result, compound was assayed again and both replicates are shown.
c NS = not significant.
Table 5. Schild analysis of the antagonism of CP55940 and WIN55212-2 by 5 and 12 in rat brain and hCB1-transfected CHO cells
Agonist SR141716A (5) 12
Rat brain
Slope/R2
hCB1 cells
Slope/R2
Rat brain
Slope/R2
hCB1 cells
Slope/R2
pA2/Kb
Schild slope constrained to 1
pA2/Kb
pA2/Kb
pA2/Kb
CP55940
WIN55212
9.28
9.49
1/0.01
1/0.61
7.55
7.79
1/ꢀ0.09
9.0
7.52
1/0.26
1/0.41
7.24
7.24
1/0.12
1/0.25
1/0.20
Agonist
pA2
Slope/R2
pA2
Slope/R2
pA2
Slope/R2
pA2
Slope/R2
Schild slope not constrained
CP55940
WIN55212
8.55
9.91
1.87/0.13
0.81/0.62
7.07
7.10
7.63/ꢀ0.08
6.81/ꢀ0.19
10.4
13.88
0.55/0.33
0.12/0.55
22.61
7.54
0.05/0.21
0.55/0.25
The data derived using the statistically better model (determined using global nonlinear regression and an F-test, see Methods) is shown in bold.
For example, it is possible that 2 and 3 and 5 and 12 dif-
fer in their influence on GPCR activation beyond their
effects on the relative proportion of the receptor active
state. It has been shown that chemically distinct classes
of cannabinoid agonists promote differential CB1 recep-
tor–Gi protein interactions.34 It is also likely that antag-
onists are able to induce differing populations of
microconformations of the CB1 receptor, which could
have an effect on their G-protein uncoupling and speci-
ficity. However, the significance in the binding affinity
and signaling differences between 5 and 12 needs to be
further characterized in in vitro and in vivo assays of
cannabinoid activity.
It is intriguing to speculate that there may be other can-
nabinoid ligands that are spared from displacement with
these unique SR141716 analogs. For example, it remains
to be determined whether endocannabinoids are dis-
placed by ÔWIN-sparingÕ SR141716 analogs. If they
are not, it is possible that the endocannabinoid system
could be left unaltered, while some exogenous cannabi-
noid compounds could be antagonized. However, it is

B. F. Thomas et al. / Bioorg. Med. Chem. 13 (2005) 5463–5474
5471
also possible that the unique binding of the alkyl com-
pounds imparts unique pharmacological properties in
the absence of a competitive agonist or agonist. These
possibilities will therefore be the subject of future explo-
ration in our laboratory.
combination of HPLC, GC–MS, high-resolution mass
spectra (HRMS), LC/MS, and NMR analytical tech-
niques described below. Compounds were shown to be
homogeneous by HPLC, employing two diverse solvent
mixtures on a Waters dual pump chromatography oper-
ating at 2.0 mL/min with a model 484 tunable absor-
bance detector, Waters Nova-Pak reversed phase C-18
(4 lm) RCM 8 mm · 100 mm column, and UV detec-
tion at 280 nm; eluants utilized were either CH3CN–
H2O or CH3OH–H2O mixtures as indicated in each
experimental procedure. GC–MS was measured on a
Hewlett–Packard 6890 GC System with a model 5973
Mass Selective Detector using EI ionization. LC/MS
spectra were obtained using an API3000 LC/MS-MS
system using atmospheric pressure chemical ionization
(APCI). HRMS were determined on a VG-70S Mass
Spectrometer (Micromass, Beverly, MA) and were per-
formed by the mass spectrometry laboratory at the Uni-
versity of South Carolina. 1H NMR spectra were
recorded on a Bruker Avance DPX-300 (300 MHz)
spectrometer and were determined in MeOH-d4 with
TMS (0.00 ppm) or MeOH (3.30 ppm) as the internal
reference unless otherwise noted.
4. Experimental
The synthesis of the target compounds 8–11 was de-
scribed previously,28 and compounds 12, 13, and 26
were synthesized in a manner similar to a previously
published synthesis of 528 by condensation of the respec-
tive amines with the pyrazole acid chloride 7, as shown
in Scheme 1.
Dimethylheptyl amine was prepared by the addition of
ethyl 2,2-dimethyloctanoate to sodium amide in tetrahy-
drofuran (THF),29 followed by Hofmann rearrangement
using the commercially available reagent [I,I-bis(triflu-
oroacetoxy)iodo]benzene.30 The synthesis of 26 was then
carried out as above (Scheme 1). The products were
characterized by 1H nuclear magnetic resonance
(NMR), high-resolution electron impact (EI) mass spec-
troscopy, and HPLC.
4.2. N-(1-Heptyl)-5-(4-chlorophenyl)-1-(2,4-dichlorophe-
nyl)-4-methyl-1H-pyrazole-3-carboxamide (12)
The synthesis of the target compounds 15–25 and 27 was
carried out by similar methods as previously pub-
lished,28 performed in parallel using an Argonaut Quest
210.
1-Heptylamine (172.5 mg, 222.0 lL, 1.50 mmol), trieth-
ylamine (420 lL, 3.00 mmol), and CH2Cl2 (7 mL) were
cooled to 0 ꢁC under argon and treated dropwise over
5 min with a solution of 728 (299.1 mg, 0.75 mmol) in
CH2Cl2 (8 mL). The solution was allowed to warm slow-
ly to 25 ꢁC under Ar and stirred for an additional 2 h.
The reaction was quenched by the addition of 10 mL
H2O and partitioned. The organic layer was washed
with 2 · 20 mL of 1 N HCl. The combined aqueous lay-
ers were backextracted with CH2Cl2, and the combined
CH2Cl2 layers were washed with saturated aqueous
NaHCO3. The aqueous layer was backextracted with
CH2Cl2, and the combined CH2Cl2 layers were dried
over Na2SO4. The CH2Cl2 was evaporated in vacuo,
yielding a yellow oil. Medium-pressure chromatography
(1:7, EtOAc–hexanes) yielded the product as a yellow oil
4.1. General methods
Reactions were conducted under N2 or Ar atmospheres
using oven-dried glassware. Parallel syntheses were per-
formed with an Argonaut Technologies Quest 210 Par-
allel Synthesizer. All solvents and chemicals used were
reagent grade. CH2Cl2 was passed through basic alumi-
˚
na and stored over 4 A molecular sieves under Ar. Et3N
was distilled from CaH2 and stored over NaOH pellets
under Ar. Unless otherwise mentioned, reagents were
obtained from commercial sources and used without
further purification. Medium-pressure column chroma-
tography was carried out on a Merck LoBar prepacked
silica gel column (240 mm, Si-60, 40–63 lm). Purity and
characterization of compounds were established by a
1
(286 mg, 80% yield). H NMR: d 7.59 (d, J = 2.2 Hz,
1H, Ar 3-H), 7.55 (d, J = 8.5 Hz, 1H, Ar 6-H), 7.46
(dd, J = 2.2, 8.5 Hz, 1H, Ar 5-H), 7.39 (d, J = 8.5 Hz,
Scheme 1.

5472
B. F. Thomas et al. / Bioorg. Med. Chem. 13 (2005) 5463–5474
2H, Ar0 3,5-H), 7.21 (d, J = 8.5 Hz, 2H, Ar0 2,6-H), 3.37
(t, J = 7.1 Hz, 2H, N–CH2–(CH2)5–CH3), 2.32 (s, 3H,
CH3), 1.62 (m, 2H, N–CH2–CH2–(CH2)4–CH3), 1.36
the solution was stirred in the dark at room temperature
for 12 h. The reaction mixture was diluted with water
(75 mL), concentrated HCl (8 mL) was added, and the
mixture was extracted with ether (2 · 75 mL). The aque-
ous layer was concentrated at reduced pressure to yield
the hydrochloride salt of 1,1-dimethylheptylamine as
white, needle-like crystals (662 mg, 96% yield). mp
112.2–112.8 ꢁC.
(m,
8H,
N–CH2–CH2–(CH2)4–CH3),
0.91
(t,
J = 6.8 Hz, 3H, N–(CH2)6–CH3). HPLC: 90%
CH3CN–H2O, tR 6.5 min (100%); 85% CH3OH–H2O,
Rt 12.2 min (100%). HREIMS m/z 477.1141 (calcd for
C25H28 35Cl3N3O, 477.1141).
4.3. N-(1-Octyl)-5-(4-chlorophenyl)-1-(2,4-dichlorophe-
nyl)-4-methyl-1H-pyrazole-3-carboxamide (13)
4.7. N-(1,1-Dimethylheptyl)-5-(4-chlorophenyl)-1-(2,4-
dichlorophenyl)-4- methyl-1H-pyrazole-3-carboxamide
(26)
Compound 7 was obtained from 4 and 1-octylamine
according to the procedure described for 6 and was iso-
lated as a yellow oil (175 mg, 90% yield). H NMR: d
The HCl salt of 1,1-dimethylheptyl amine (174.1 mg,
1.03 mmol), triethylamine (500 lL, 3.59 mmol) and
95% EtOH (5 mL) were cooled to 0 ꢁC under argon
and treated dropwise over 5 min with a solution of 4
(203.6 mg, 0.51 mmol) in CH2Cl2 (5 mL). The solution
was allowed to warm slowly to 25 ꢁC under Ar and stir-
red for an additional 2 h. The reaction was quenched by
addition of 10 mL H2O. The organic layer was washed
with 2 · 20 mL of 1 N HCl. The combined aqueous lay-
ers were backextracted with CH2Cl2, and the combined
CH2Cl2 layers were washed with saturated aqueous
NaHCO3. The aqueous layer was backextracted with
CH2Cl2, and the combined CH2Cl2 layers were dried
over Na2SO4. The CH2Cl2 was evaporated in vacuo,
yielding a dark yellow film. Flash chromatography
(1:11, EtOAc–hexanes) yielded the product as a yellow
oil (177.2 mg, 69% yield). 1H NMR: d 7.57 (d,
J = 2.1 Hz, 1H, Ar 3-H), 7.52 (d, J = 8.5 Hz, 1H, Ar 6-
H), 7.44 (dd, J = 2.2, 8.5 Hz, 1H, Ar 5-H), 7.36 (d,
J = 8.5 Hz, 2H, Ar0 3,5-H), 7.18 (d, J = 8.5 Hz, 2H,
Ar0 2,6-H), 2.28 (s, 3H, CH3), 1.80 (br t, J = 7.5 Hz,
2H, N–C(CH3)2–CH2–(CH2)4–CH3) 1.40 (s, 6H, N–
C(CH3)2–(CH2)5–CH3), 1.31 (m, 8H, N–C(CH3)2–
1
7.54 (d, J = 2.2 Hz, 1H, Ar 3-H), 7.50 (d, J = 8.5 Hz,
1H, Ar 6-H), 7.42 (dd, J = 2.2, 8.5 Hz, 1H, Ar 5-H),
7.35 (d, J = 8.5 Hz, 2H, Ar0 3,5-H), 7.18 (d,
J = 8.5 Hz, 2H, Ar0 2,6-H), 3.34 (t, J = 7.2 Hz, 2H, N–
CH2–(CH2)6–CH3), 2.30 (s, 3H, CH3), 1.59 (m, 2H,
N–CH2–CH2–(CH2)5–CH3), 1.28 (m, 10H, N–CH2–
CH2–(CH2)5–CH3), 0.88 (t, J = 7.0 Hz, 3H, N–(CH2)7–
CH3). HPLC: 90% CH3CN–H2O, tR 16.6 min (99%);
85% CH3OH–H2O, Rt 16.3 min (100%). HREIMS m/z
491.1277 (calcd for C25H28 35Cl3N3O, 491.1298).
4.4. N-(1-Nonyl)-5-(4-chlorophenyl)-1-(2,4-dichlorophe-
nyl)-4-methyl-1H-pyrazole-3-carboxamide (14)
Compound 8 was obtained from 4 and 1-nonylamine
according to the procedure described for 7 and was iso-
1
lated as a yellow oil (211 mg, 83% yield). H NMR: d
7.50 (d, J = 2.2 Hz, 1H, Ar 3-H), 7.47 (d, J = 8.5 Hz,
1H, Ar 6-H), 7.39 (dd, J = 2.2, 8.5 Hz, 1H, Ar 5-H),
7.31 (d, J = 8.5 Hz, 2H, Ar0 3,5-H), 7.14 (d,
J = 8.5 Hz, 2H, Ar0 2,6-H), 3.30 (t, J = 7.1 Hz, 2H, N–
CH2–(CH2)7–CH3), 2.26 (s, 3H, CH3), 1.54 (m, 2H,
N–CH2–CH2–(CH2)6–CH3), 1.24 (m, 12H, N–CH2–
CH2–(CH2)6–CH3), 0.84 (t, J = 6.8 Hz, 3H, N–(CH2)8–
CH3). HPLC: 90% CH3CN–H2O, tR 21.6 min (99%);
85% CH3OH–H2O, Rt 23.2 min (100%). HREIMS m/z
505.1463 (calcd for C26H30 35Cl3N3O, 505.1454).
CH2–(CH2)4–CH3),
0.88
(t,
J = 6.0 Hz,
3H,
N–C(CH3)2–(CH2)5–CH3). HPLC: 90% CH3CN–H2O,
Rt 24.7 min (99%); 80% CH3OH–H2O, tR 24.3 min
(100%). HREIMS m/z 505.1465 (calcd for C26H30
35Cl3N3O, 505.1454).
4.8. Preparation of long-chain unbranched- and
branched-alkyl amide analogs of (5): general procedure
4.5. Synthesis of 2,2-dimethyloctanamide29
NaNH2 (10.1 mmol) was treated with a solution of ethyl
2,2-dimethyloctanoate35,36 (10.1 mmol) in dry THF
(35 mL) under N2 and stirred at ambient temperature
for 2 h. The reaction was quenched with saturated aque-
ous NH4Cl. This was extracted with 2 · 20 mL CH2Cl2.
The organic layer was washed with 2 · 20 mL saturated
aqueous NH4Cl, and the combined organic layers were
dried over Na2SO4. The CH2Cl2 was evaporated in vac-
uo, yielding a beige powder (1.02 g, 59% yield). The
product was carried forth to the next step without fur-
ther purification.
To a mixture of the respective hydrazine or alkyl amine
(0.09 mmol) and Et3N (0.18 mmol) in anhydrous
CH2Cl2 (4 mL) under nitrogen, 7 (0.05 mmol) was add-
ed, and the reaction mixture was agitated for 2 h at
room temperature. The crude mixtures were washed
with water, the layers were separated, and the organics
were dried over Na2SO4. Filtration and solvent removal
yielded the desired compounds which were characterized
1
by H NMR and/or LC/MS.
4.9. N-(1-Decyl)-5-(4-chlorophenyl)-1-(2,4-dichlorophe-
nyl)-4- methyl-1H-pyrazole-3-carboxamide (15)
4.6. Synthesis of 1,1-dimethylheptyl amine30
1H NMR (CD3OD): d 7.56 (d, J = 2.2 Hz, 1H), 7.52 (d,
J = 8.5 Hz, 1H), 7.44 (dd, J = 2.2 Hz, 8.5 Hz, 1H), 7.36
(d, J = 8.5 Hz, 2H), 7.19 (d, J = 8.5 Hz, 2H), 3.35 (t,
J = 7.1 Hz), 2.30 (s, 3H), 1.60 (m, 2H), 1.28 (m, 14H),
0.89 (t, J = 6.6 Hz, 3H).
[I,I-bis(trifluoroacetoxy)iodo]benzene
(703 mg,
4.1
mmol) was dissolved in acetonitrile (6 mL) and glass-
distilled water (6 mL) was added. To this solution, 2,2-
dimethyloctanamide (1.94 g, 4.5 mmol) was added and

B. F. Thomas et al. / Bioorg. Med. Chem. 13 (2005) 5463–5474
5473
4.10. N-(1-Undecyl)-5-(4-chlorophenyl)-1-(2,4-dichlor-
ophenyl)-4- methyl-1H-pyrazole-3-carboxamide
(16)
4.16. N-(1-Methylhexyl)-5-(4-chlorophenyl)-1-(2,4-
dichlorophenyl)- 4-methyl-1H-pyrazole-3-carboxamide
(22)
1H NMR (CD3OD): d 7.56 (d, J = 2.2 Hz, 1H), 7.52 (d,
J = 8.5 Hz, 1H), 7.44 (dd, J = 2.2 Hz, 8.5 Hz, 1H), 7.36
(d, J = 8.5 Hz, 2H), 7.19 (d, J = 8.5 Hz, 2H), 3.35 (t,
J = 7.1 Hz, 2H), 2.30 (s, 3H), 1.59 (m, 2H), 1.28 (m,
16H), 0.89 (t, J = 6.6 Hz, 3H).
1H NMR (CDCl3): d 7.38 (s,1H), 7.24 (m, 4H), 7.00 (d,
J = 8.5 Hz, 2H), 6.70 (d, J = 8.2 Hz, 1H), 4.12 (m, 1H),
2.33 (s, 3H), 1.45 (m, 2H), 1.33 (m, 2H), 1.21 (m, 4H),
1.18 (d, J = 6.5 Hz, 3H), 0.83 (t, J = 6.6 Hz, 3H).
4.17. N-(1,5-Dimethylhexyl)-5-(4-chlorophenyl)-1-(2,4-
dichlorophenyl)-4-methyl- 1H-pyrazole-3-carboxamide
(23)
4.11. N-(1-Dodecyl)-5-(4-chlorophenyl)-1-(2,4-dichlor-
ophenyl)-4- methyl-1H-pyrazole-3-carboxamide (17)
1H NMR (CD3OD): d 7.56 (d, J = 2.1 Hz, 1H), 7.52 (d,
J = 8.5 Hz, 1H), 7.44 (dd, J = 2.1 Hz, 8.5 Hz, 1H), 7.36
(d, J = 8.4 Hz, 2H), 7.19 (d, J = 8.4 Hz, 2H), 3.52 (t,
J = 7.1 Hz, 2H), 2.30 (s, 3H), 1.60 (m, 2H), 1.28 (m,
18H), 0.89 (t, J = 6.6 Hz, 3H).
1H NMR (CDCl3): d 7.42 (s, 1H), 7.29 (m, 4H), 7.05 (d,
J = 8.5 Hz, 2H), 6.73 (d, J = 8.0 Hz, 1H), 4.16 (m, 1H),
2.38 (s, 3H), 1.50 (m, 3H), 1.35 (m, 2H), 1.20 (m, 5H),
0.85 (d, J = 6.6 Hz, 6H). LC/UV-MS analysis revealed
a single UV peak at 11.59 min (ꢁ94% integrated area)
with an intense ion (base peak) at m/z 514 (predicted
mass of sodium adduct).
4.12. N-(1-Tridecyl)-5-(4-chlorophenyl)-1-(2,4-dichlor-
ophenyl)-4- methyl-1H-pyrazole-3-carboxamide (18)
4.18. N-(2-Ethylhexyl)-5-(4-chlorophenyl)-1-(2,4-dichlor-
ophenyl)-4- methyl-1H-pyrazole-3-carboxamide (24)
1H NMR (CD3OD): d 7.56 (d, J = 2.2 Hz, 1H), 7.52 (d,
J = 8.5 Hz, 1H), 7.44 (dd, J = 2.2 Hz, 8.5 Hz, 1H), 7.36
(d, J = 8.5 Hz, 2H), 7.19 (d, J = 8.5 Hz, 2H), 3.35 (t,
J = 7.1 Hz, 2H), 2.30 (s, 3H), 1.58 (m, 2H), 1.28 (m,
20H), 0.89 (t, J = 6.6 Hz, 3H).
1H NMR (CDCl3): d 7.43 (s, 1H), 7.28 (m, 4H), 7.05 (d,
J = 8.5 Hz, 2H), 6.94 (t, J = 8.0 Hz, 1H), 3.36 (m, 1H),
2.38 (s, 3H), 1.54 (m, 2H), 1.37 (m, 8H), 0.89 (m, 6H).
LC/UV-MS analysis revealed a single UV peak at
11.67 min (ꢁ94% integrated area) with an intense ion
(base peak) at m/z 514 (predicted mass of sodium
adduct).
4.13. N-(1-Geranyl)-5-(4-chlorophenyl)-1-(2,4-dichlor-
ophenyl)-4- methyl-1H-pyrazole-3-carboxamide (19)
1H NMR (CDCl3): d 7.42 (s, 1H), 7.28 (m, 4H), 7.01 (d,
J = 8.5 Hz, 2H), 6.86 (t, J = 9.0 Hz, 1H), 5.28 (t,
J = 8.3 Hz, 1H), 5.08 (t, J = 8.3 Hz, 1H), 4.03 (t,
J = 8.2 Hz, 2H), 2.38 (s, 3H), 2.03 (m, 4H), 1.70 (s,
3H), 1.66 (s, 3H), 1.56 (s, 3H). LC/UV-MS analysis re-
vealed a single UV peak at 11.68 min (ꢁ93% integrated
area) with an intense ion (base peak) at m/z 538 (predict-
ed mass of sodium adduct).
4.19. N-(1-Methylheptyl)-5-(4-chlorophenyl)-1-(2,4-
dichlorophenyl)- 4-methyl-1H-pyrazole-3-carboxamide
(25)
1H NMR (CDCl3): d 7.43 (s, 1H), 7.28 (m, 4H), 7.05 (d,
J = 8.5 Hz, 2H), 6.73 (d, J = 8.2 Hz, 1H), 4.13 (m, 1H),
2.38 (s, 3H), 1.54 (m, 2H), 1.39 (m, 8H), 0.86 (t,
J = 6.6 Hz, 3H). LC/UV-MS analysis revealed a single
UV peak at 111.66 min (ꢁ93% integrated area) with
an intense ion (base peak) at m/z 514 (predicted mass
of sodium adduct).
4.14. N-(1,3-Dimethylpentyl)-5-(4-chlorophenyl)-1-(2,4-
dichlorophenyl)-4- methyl-1H-pyrazole-3-carboxamide
(20)
1H NMR (CDCl3): d 7.43 (s, 1H), 7.30 (m, 4H), 7.04 (d,
J = 8.5 Hz, 2H), 6.73 (d, J = 8.0 Hz, 5H), 6.67 (d,
J = 8.0 Hz, 5H), 4.27 (m, 1H), 2.38 (s, 3H), 1.47 (m,
3H), 1.20 (m, 5H), 0.89 (m, 6H). LC/UV-MS analysis re-
vealed a single UV peak at 11.16 min (ꢁ97% integrated
area) with an intense ion (base peak) at m/z 500 (predict-
ed mass of sodium adduct).
4.20. N-(1-Methyldecyl)-5-(4-chlorophenyl)-1-(2,4-
dichlorophenyl)- 4-methyl-1H-pyrazole-3-carboxamide
(27)
1H NMR (CDCl3): d 7.42 (s, 1H), 7.28 (m, 4H), 7.05 (d,
J = 8.5 Hz, 2H), 6.73 (d, J = 8.7 Hz, 1H), 4.14 (m, 1H),
2.38 (s, 3H), 1.48 (m, 2H), 1.23 (m, 17H), 0.87 (t,
J = 7.7 Hz, 3H). LC/UV-MS analysis revealed a single
UV peak at 12.64 min (ꢁ91% integrated area) with an
intense ion (base peak) at m/z 558 (predicted mass of
sodium adduct).
4.15. N-(1,4-Dimethylpentyl)-5-(4-chlorophenyl)-1-(2,4-
dichlorophenyl)-4- methyl-1H-pyrazole-3-carboxamide
(21)
1H NMR (CDCl3): d 7.43 (s, 1H), 7.29 (m, 4H), 7.06 (d,
J = 8.5 Hz, 2H), 6.73 (d, J = 8.0 Hz, 1H), 4.14 (m, 1H),
2.37 (s, 3H), 1.52 (m, 3H), 1.28 (m, 5H), 0.87 (d, J =
6.6 Hz, 6H). LC/UV-MS analysis revealed a single UV
peak at 11.19 min (ꢁ78% integrated area) with an in-
tense ion (base peak) at m/z 500 (predicted mass of sodi-
um adduct).
4.21. CB1 receptor affinity and efficacy determination
All animal procedures were carried out in accordance
with the Institutional Animal Care and Use Committee
at the Research Triangle Institute and with the 1996
Guide for the Care and Use of Laboratory Animals as
adapted and promulgated by the National Institute on

5474
B. F. Thomas et al. / Bioorg. Med. Chem. 13 (2005) 5463–5474
10. Schlicker, E.; Timm, J.; Gothert, M. Naunyn Schmiede-
bergs Arch. Pharmacol. 1996, 354, 791.
11. Szabo, B.; Dorner, L.; Pfreundtner, C.; Norenberg, W.;
Starke, K. Neuroscience 1998, 85, 395.
12. Levenes, C.; Daniel, H.; Soubrie, P.; Crepel, F. J. Physiol.
1998, 510, 867.
13. Shen, M.; Piser, T.; Seybold, V. S.; Thayer, S. A. J.
Neurosci. 1996, 16, 4322.
14. Pertwee, R. G. Pharm. Sci. 1997, 3, 539.
15. Le Fur, G.; Arnone, M.; Rinaldi-Carmona, M.; Barth, F.;
Heshmati, H. Abstracts of Papers, 11th Annual Sympo-
sium on the Cannabinoids, Madrid, Spain, Jun 28–30,
2001; International Cannabinoid Research Society: Bur-
lington, VT, 2001.
16. Shim, J. Y.; Welsh, W. J.; Cartier, E.; Edwards, J. L.;
Howlett, A. C. J. Med. Chem. 2002, 45, 1447.
17. Thomas, B. F.; Compton, D. R.; Martin, B. R.; Semus, S.
F. Mol. Pharmacol. 1991, 40, 656.
18. Schmetzer, S.; Greenidge, P.; Kovar, K. A.; Schulze-
Alexandru, M.; Folkers, G. J. Comput. Aided Mol. Des.
1997, 11, 278.
19. Thomas, B. F.; Adams, I. B.; Mascarella, S. W.; Martin,
B. R.; Razdan, R. K. J. Med. Chem. 1996, 39, 471.
20. Fichera, M.; Cruciani, G.; Bianchi, A.; Musumarra, G. J.
Med. Chem. 2000, 43, 2300.
Health. Mouse vas deferens methods were performed as
previously described in Ref. 28. For the competition as-
says utilizing rat brain membrane preparations, male
CDꢂ rats (Charles River Laboratories, Rayleigh, NC)
weighing 220–225 g were killed. The whole brains were
quickly removed and placed into a 55-mL Potter–Elveh-
jem glass homogenizer tube maintained on ice. The
methods for preparation of rat brain membranes were
essentially those described by Devane et al.1 as modified
later by our laboratory.27
GTP-c-[35S] assays were also performed to determine
the ability of 5 and 12 to shift the binding curves of
the agonists 2 or 3. Reaction mixtures consisted of
either 2 (2.5 pM to 25 lM) to 3 (10 pM to 100 lM),
20 lM GDP, and 100 pM GTP-c-[35S] in 50 mM
Tris–HCl, pH 7.4, 1 mM EDTA, 5 mM MgCl2,
100 mM NaCl, and 1 mg/mL BSA. The effects of 5
and 12 on agonist binding were compared at concen-
trations of 1, 10, and 100 nM vs. reactions with no
antagonist in a final reaction mixture volume of
0.5 mL. Binding was determined using membrane prep-
arations as previously described.28 Data analysis was
performed using global nonlinear regression analysis
of the dose–response curves (Prism, GraphPad), and
pA2 values were calculated. The calculations were per-
formed with the slope of the Schild line constrained to
1, as well as unconstrained, and an F-test (P < 0.05)
was used to determine the best model.
21. Tong, W.; Collantes, E. R.; Welsh, W. J.; Berglund, B. A.;
Howlett, A. C. J. Med. Chem. 1998, 41, 4207.
22. Howlett, A. C.; Mukhopadhyay, S.; Shim, J. Y.; Welsh,
W. J. Life Sci. 1999, 65, 617.
23. Eissenstat, M. A.; Bell, M. R.; DÕAmbra, T. E.; Alexan-
der, E. J.; Daum, S. J.; Ackerman, J. H.; Gruett, M. D.;
Kumar, V.; Estep, K. G.; Olefirowicz, E. M. J. Med.
Chem. 1995, 38, 3094.
24. Xie, X. Q.; Eissenstat, M.; Makriyannis, A. Life Sci. 1995,
56, 1963.
Acknowledgments
25. Shim, J. Y.; Collantes, E. R.; Welsh, W. J.; Subramaniam,
B.; Howlett, A. C.; Eissenstat, M. A.; Ward, S. J. J. Med.
Chem. 1998, 41, 4521.
The authors thank Dr. Michael D. Walla, University of
South Carolina, for the high-resolution mass spectra,
and Vicki McCall for her assistance in the preparation
of the manuscript. This work was supported by the
National Institute on Drug Abuse (NIDA) Grant DA-
19217 (B.F.T.) and NIDA Grant DA-09789 (R.G.P).
26. Kumar, V.; Alexander, M. D.; Bell, M. R.; Eissenstat, M.
A.; Casiano, F. M.; Chippari, S. M.; Haycock, D. A.;
Luttinger, D. A.; Kuster, J. E.; Miller, M. S.; Stevenson, J.
I.; Ward, S. J. Bioorg. Med. Chem. Lett. 1995, 5, 381.
27. Thomas, B. F.; Gilliam, A. F.; Burch, D. F.; Roche, M. J.;
Seltzman, H. H. J. Pharmacol. Exp. Ther. 1998, 285, 285.
28. Francisco, M. E.; Seltzman, H. H.; Gilliam, A. F.;
Mitchell, R. A.; Rider, S. L.; Pertwee, R. G.; Stevenson,
L. A.; Thomas, B. F. J. Med. Chem. 2002, 45, 2708.
29. Pang, Y.-P.; Kozikowski, A. P. J. Org. Chem. 1991, 56,
4499.
30. Loudon, G. M.; Radhakrishna, A. S.; Almond, M. R.;
Blodgett, J. K.; Boutin, R. H. J. Org. Chem. 1984, 49, 4272.
31. Song, Z. H.; Bonner, T. I. Mol. Pharmacol. 1996, 49, 891.
32. Chin, C. N.; Lucas-Lenard, J.; Abadji, V.; Kendall, D. A.
J. Neurochem. 1998, 70, 366.
33. Picone, R. P.; Ayotte, L. A.; Hurst, D.; Reggio, P. R.;
Khanolkar, A. D.; Fournier, D. J.; Makriyannis, A.
Abstract of Papers, 12th Annual Symposium on the
Cannabinoids, Pacific Grove, CA, July 10–14, 2002;
International Cannabinoid Research Society: Burlington,
VT, 2002.
34. Mukhopadhyay, S.; Howlett, A. C. Mol. Pharmacol. 2005,
67, 2016.
35. Williams, T. R.; Sirvio, L. M. J. Org. Chem. 1980, 45,
5082–5088.
References and notes
1. Devane, W. A.; Dysarz, F. A., 3rd; Johnson, M. R.;
Melvin, L. S.; Howlett, A. C. Mol. Pharmacol. 1988, 34,
605.
2. Munro, S.; Thomas, K. L.; Abu-Shaar, M. Nature 1993,
365, 61.
3. Matsuda, L. A.; Lolait, S. J.; Brownstein, M. J.; Young,
A. C.; Bonner, T. I. Nature 1990, 346, 561.
4. Abood, M. E.; Martin, B. R. Int. Rev. Neurobiol. 1996, 39,
197.
5. Bouaboula, M.; Poinot-Chazel, C.; Bourrie, B.; Canat, X.;
Calandra, B.; Rinaldi-Carmona, M.; Le Fur, G.; Casellas,
P. Biochem. J. 1995, 312, 637.
6. Howlett, A. C.; Qualy, J. M.; Khachatrian, L. L. Mol.
Pharmacol. 1986, 29, 307.
7. Deadwyler, S. A.; Hampson, R. E.; Mu, J.; Whyte, A.;
Childers, S. J. Pharmacol. Exp. Ther. 1995, 273, 734.
8. Poling, J. S.; Rogawski, M. A.; Salem, N., Jr.; Vicini, S.
Neuropharmacology 1996, 35, 983.
36. Seltzman, H. H.; Fleming, D. N.; Thomas, B. F.; Gilliam,
A. F.; McCallion, D. S.; Pertwee, R. G.; Compton, D. R.;
Martin, B. R. J. Med. Chem. 1997, 40, 3626.
9. Mackie, K.; Lai, Y.; Westenbroek, R.; Mitchell, R. J.
Neurosci. 1995, 15, 6552.