Synthesis and pharmacology of a very potent cannabinoid lacking a phenolic hydroxyl with high affinity for the CB2 receptor

Full text of DOI:10.1021/jm960394y

J . Med. Chem. 1996, 39, 3875-3877
3875
Communications to the Editor
hydroxyl of 2, while Val 196 creates steric hindrance
behind the carbocyclic ring of 11-hydroxy-∆8-THC-DMH.
The existence of such a sterically occluded region has
been postulated in cannabinoid SAR for some time.7,8
Syn th esis a n d P h a r m a cology of a Ver y
P oten t Ca n n a bin oid La ck in g a P h en olic
Hyd r oxyl w ith High Affin ity for th e CB2
Recep tor
Since traditional cannabinoid SAR, mutation studies,
and the modeling studies described above indicate that
the phenolic hydroxyl is a key interaction site at CB1,
it was assumed that a 1-deoxy cannabinoid would show
little affinity for the CB1 receptor and little potency in
vivo.3,4 The SAR and modeling data suggest that a
cannabinoid lacking the phenolic hydroxyl should ex-
hibit poor affinity for the wild-type CB1 receptor similar
to the poor affinity observed for the traditional cannab-
inoids, CP 55,940 (1) and 11-hydroxy-∆8-THC-DMH (2),
for binding to the mutant receptor in which a lysine has
been replaced by an alanine. A suitable compound to
test this hypothesis appeared to be 3, the 1-deoxy
analogue of the very potent cannabinoid 11-hydroxy-
∆8-THC-DMH.
The synthesis of deoxy cannabinoid 3 was carried out
using methodology developed for the synthesis of 1-nor-
9-carboxy-∆9-tetrahydrocannabinol,9 and the enantiodi-
vergent synthesis of both enantiomers of nabilone.10 As
shown in Scheme 1, the aromatic component for the
synthesis, 2-bromo-5-(1,1-dimethylheptyl)methoxyben-
zene (4),11 was prepared in four steps from 1,3-dimethoxy-
5-(dimethylheptyl)benzene (5).10,12 The aryllithium de-
rived from 4 was added to apoverbenone to give, after
oxidative rearrangement, enone 6.11 Dissolving metal-
reduction of 6 proceeded stereoselectively to provide
saturated ketone 7.11 Ether cleavage, followed by
sequential treatment with SnCl4 and AlCl3, gave deoxy-
nabilone (8).11 Although ketone 6 could be converted
to the corresponding enolate, this enolate was unreac-
tive to a variety of reagents designed to provide triflate
9.13,14 The triflate was ultimately prepared by the
procedure of Stang, employing triflic anhydride in the
presence of 2,6-di-tert-butylpyridine.15 Palladium-medi-
ated carbonylation of 9 in the presence of methanol gave
ester 10,11,16 which provided 311 after reduction.
J ohn W. Huffman,*,† Shu Yu,† Vincent Showalter,‡
Mary E. Abood,‡ J enny L. Wiley,‡ David R. Compton,‡
Billy R. Martin,‡ R. Daniel Bramblett,§ and
Patricia H. Reggio§
Department of Chemistry, Clemson University, Clemson,
South Carolina 29634-1905, Department of Pharmacology
and Toxicology, Medical College of Virginia, Virginia
Commonwealth University, Richmond, Virginia 23298, and
Department of Chemistry, Kennesaw State University,
1000 Chastain Road, Kennesaw, Georgia 30144-5591
Received J une 3, 1996
Song and Bonner have very recently described a
mutant (Lys 192 Ala) CB1 receptor in which the binding
of CP 55,940 (1), and the very potent cannabinoid HU-
210 (the dimethylheptyl analogue of 11-hydroxy-∆8-
tetrahydrocannabinol, 11-hydroxy-∆8-THC-DMH, 2) is
greatly attenuated.1 These authors conclude that it is
likely that in the wild-type receptor Lys 192 is hydrogen
bonded to the phenolic hydroxyl of cannabinoids 1 and
2 and that the loss of a terminal amino group when Lys
192 is replaced by alanine is responsible for the decrease
in affinity.1
The affinity of 3 for the CB1 receptor was determined
by measuring its ability to displace [3H]CP 55,940 (1)
from its binding site in a membrane preparation from
rat brain as described by Compton et al.17 The affinity
for the peripheral (CB2) receptor was determined by
measuring the ability of 3 to displace [3H]CP 55,940
from a cloned human receptor preparation using the
procedure described by Showalter et al.18 In contrast
to expectations, 3 has high affinity for the CB1 receptor
(Ki ) 1.2 ( 0.1 nM) and even greater affinity for the
CB2 receptor (Ki ) 0.032 ( 0.19 nM). These data are
included in Table 1 along with those for 11-hydroxy-
∆8-THC-DMH (2), ∆9-tetrahydrocannabinol (∆9-THC,
14), the active constituent of marijuana, and the 1′,1′-
dimethylheptyl analogue of ∆8-tetrahydrocannabinol
(∆8-THC DMH, 11). The in vivo pharmacology of 3 was
evaluated in the mouse model of cannabimimetic activ-
ity which consists of measuring spontaneous activity
(SA), antinociception (as tail flick, TF), and hypothermia
(as rectal temperature, RT).19 These data are also
A computer model of the cannabinoid brain (CB1)
receptor2 was examined for possible ligand interactions
(see the Experimental Section included in the Support-
ing Information for model building details). On the
basis of Song and Bonner’s mutation studies of Lys 192,
11-hydroxy-∆8-THC-DMH (2) was docked so that Lys
192 could interact with the phenolic hydroxyl group at
C-1. An extensive amount of cannabinoid structure-
activity relationship (SAR) data indicate that the can-
nabinoid C-3 side chain is a key element in CB1 receptor
recognition.3-6 It was found that with the phenolic
hydroxyl of 2 interacting with Lys 192, the C-3 side
chain can interact with a hydrophobic pocket formed by
Val 351 and Ile 354. In this docking position, Tyr 275
at the top of Helix 5 can hydrogen bond to the 11-
* Address correspondence to this author at: Department of Chem-
istry, Clemson University, Clemson, South Carolina 29634-1905.
† Clemson University.
‡ Medical College of Virginia, Virginia COmmonwealth University.
§ Kennesaw State University.
S0022-2623(96)00394-9 CCC: $12.00 © 1996 American Chemical Society

3876 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 20
Communications to the Editor
Ta ble 1. Pharmacological Responses (Mean ( SEM) of 1-Deoxy-11-hydroxy-∆8-THC-DMH (3) and Related Compounds
Ki (nM)
ED50 (µmol/kg)
compound
CB1
CB2
SA
TF
RT
∆9-THC (14)
41 ( 2a
36 ( 10b
0.52 ( 0.05d
ND
0.032 ( 0.019
2.9 ( 1.6
2.9c
4.8c
4.5c
11-hydroxy-∆8-THC-DMH (2)
∆8-THC-DMH (11)
0.73 ( 0.11a
0.77 ( 0.11c
1.2 ( 0.1
0.01e
0.27c
0.12
6.4
0.02e
0.14c
0.15
1.6
0.05e
0.15c
0.39
3.1
1-deoxy-11-hydroxy-∆8-THC-DMH (3)
deoxy-∆8-THC-DMH (12)
23 ( 7
a
b
Reference 14. Reference 18. c Martin, B. R.; Compton, D. R.; Semus, S.; Lin, S.; Marciniak, C.; Grzybowska, J .; Charalambous, A.
d
Makriyannis, A. Pharmacol. Biochem. Behav. 1993, 46, 295-301. Felder, C. C.; J oyce, K. E.; Briley, E. M.; Mansouri, J .; Mackie, K.;
Olivier, B.; Lai, Y.; Ma, A. L.; Mitchell, R. L. Mol. Pharmacol. 1995, 48, 443-450. e Little, P. J .; Compton, D. R.; Mechoulam, R.; Martin,
B. R. Pharmacol. Biochem. Behav. 1989, 32, 661-666.
Sch em e 1a
that the side chain of Lys 192 has the length and
flexibility to form a hydrogen bond with the C-11
hydroxyl, while the C-3 side chain interacts with the
hydrophobic binding pocket.22 There is a net loss of one
hydrogen-bonding interaction upon removal of the phe-
nolic hydroxyl group of 11-hydroxy-∆8-THC-DMH, which
may account for the decrease in potency of 3 relative to
2.
On the basis of this model, it appeared probable a
cannabinoid lacking both the C-11 and phenolic hy-
droxyls should show much attenuated potency. Accord-
ingly, 1-deoxy-∆8-THC DMH (12) was prepared from ∆8-
THC DMH (11) by reaction with diethyl chlorophosphate,
followed by reduction of the phosphate ester using Li
in liquid ammonia. In contrast to expectations, deoxy-
cannabinoid 12 showed high affinity for both the CB1
and CB2 receptors (Ki ) 23 ( 7 and 2.9 ( 1.6 nM,
respectively). The in vivo pharmacology (Table 1)
indicated that 12 was approximately equal in potency
to ∆9-THC. Docking studies indicated that the orienta-
tion of 12 would have to be inverted relative to that of
∆9-THC, 11-hydroxy-∆8-THC-DMH (2), or 1-deoxy-11-
hydroxy-∆8-THC-DMH (3) in order to account for the
a
(a) NaSPr/DMF, 120 °C; (b) (EtO)2P(O)H, Et3N/CCl4, 0-25 °C;
(c) Li/NH3, THF, -78 °C; (d) Br2/HOAc, 0-25 °C; (e) BuLi/THF, 0
°C then apoverbenone; (f) PDC/CH2Cl2, 25 °C; (g) SnCl4/CH2Cl2,
25 °C; (h) AlCl3/CH2Cl2, 25 °C; (i) Tf2O, 2,6-di-tert-butyl-4-
methylpyridine/CH2Cl2, 40 °C; (j) Et3N, Pd(OAc)2, Ph3P, CO,
CH3OH/DMF, 45 °C; (k) LiAlH4/Et2O, 0-25 °C.
receptor affinity. In this inverted orientation, the pyran
oxygen hydrogen bonds to Lys 192, which increases the
distance between the hydrophobic binding pocket and
C-3, the atom to which the side chain is attached.
Although there is now a longer distance to traverse for
interaction with the hydrophobic binding pocket, the
dimethylheptyl side chain of 12 can still reach this
pocket, albeit with fewer carbon atoms participating in
the hydrophobic interaction.
Following the completion of this work, the synthesis
of 12 and its binding affinity for the human CB1 and
CB2 receptors was reported by Gareau et al.23 The
method employed for the preparation of 12 was the
same as that outlined above; however, the affinity data
differed considerably from those cited above. The
reported affinities for the human CB1 and CB2 recep-
tors are Ki ) 249.7 ( 31.0 and 20.8 ( 11.2 nM,
respectively, both of which are 1 order of magnitude less
than that determined in this work. The difference in
included in Table 1, with those of 11-hydroxy-∆8-THC-
DMH (2) and ∆9-THC (14). 1-Deoxy-11-hydroxy-∆8-
THC-DMH (3) also produced dose dependent cannabimi-
metic effects in the rat discrimination procedure (ED50
) 0.42 mg/kg, 95% CL: 0.12-1.5), an animal model of
the subjective effects of cannabis intoxication in hu-
mans.20 The drug discrimination studies were carried
out using the procedure described by Wiley et al.21
As can be seen from the data summarized in Table 1,
1-deoxy-11-hydroxy-∆8-THC-DMH (3) is a very potent
cannabinoid, both in vitro and in vivo in contrast to
expectations based on traditional SAR.3,4 In order to
provide a rationalization for the enhanced potency of
3, modeling studies have been carried out which indicate

Communications to the Editor
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 20 3877
(5) Melvin, L. S.; Milne, G. M.; J ohnson, M. R.; Subramian, B.;
Wilken, G. H.; Howlett, A. C. Structure-activity relationships
for cannabinoid receptor-binding and analgesic activity: studies
of bicyclic cannabinoid analogs. Mol. Pharmacol. 1993, 44,
1008-1015.
(6) Ryan, W.; Singer, M.; Razdan, R. K.; Compton, D. R.; Martin,
B. R. A novel class of potent tetrahydrocannabinols (THCs): 2′-
yne-∆8-and ∆9-THCs. Life Sci. 1995, 56, 2013-2020.
(7) Reggio, P. H.; Panu, A. M.; Miles, S. Characterization of a region
of steric interference at the cannabinoid receptor using the active
analog approach. J . Med. Chem. 1993, 36, 1761-1771.
(8) Reggio, P. H.; Greer, K. V.; Cox, S. M. The importance of the
orientation of the C9 substituent to cannabinoid activity. J .
Med. Chem. 1989, 32, 1630-1635.
(9) Huffman, J . W.; Zhang, X.; Wu, M.-J .; J oyner, H. H.; Pennington,
W. T. Synthesis of (()-11-nor-9-carboxy-∆9-tetrahydrocan-
nabinol: new synthetic approaches to cannabinoids. J . Org.
Chem. 1991, 56, 1481-1489.
(10) Huffman, J . W.; J oyner, H. H.; Lee, M. D.; J ordan, R. D.;
Pennington, W. T. Synthesis of both enantiomers of nabilone
from a common intermediate. Enantiodivergent synthesis of
cannabinoids. J . Org. Chem. 1991, 56, 2081-2086.
(11) All new compounds were characterized by 1H and 13C NMR and
gave either acceptable analytical or HRMS data. Purity was
established by TLC and 13C NMR.
(12) Dominianni, S. J .; Ryan, C. W.; De Armitt, C. W. Synthesis of
5-(tert-alkyl)resorcinols. J . Org. Chem. 1977, 42, 344-346.
(13) McMurry, J . E.; Scott, W. J . A method for the regioselective
synthesis of enol triflates by enolate trapping. Tetrahedron Lett.
1983, 24, 979-982.
(14) Comins, D. L.; Dehghani, A. Pyridine derived triflating re-
agents: an improved preparation of vinyl triflates from metallo
enolates. Tetrahedron Lett. 1992, 33, 6299-6302.
(15) Stang, P. J .; Treptow, W. Single step improved synthesis of
primary and other vinyl trifluoromethanesulfonates. Synthesis
1980, 283-284.
(16) Cacchi, S.; Morera, E.; Ortar, G. Palladium catalyzed carbonyl-
ation of enol triflates. A novel method for one-carbon homolo-
gation of ketones to R,â-unsaturated carboxylic acid derivatives.
Tetrahedron Lett. 1985, 26, 1109-1112.
the affinity data for CB1 may be due to the fact that
Gareau et al. employed the human receptor, while our
data were obtained using a rat membrane preparation;
however, the human and rat receptors are virtually
identical (97% homology).24 Both sets of binding data
for CB2 were obtained using the human receptor, and
the reported differences may be due to differences in
the experimental procedures employed in obtaining the
affinity data.
In contrast to the classical generalizations of cannab-
inoid SAR, which state that for those compounds based
on the dibenzopyran skeleton, a phenolic hydroxyl at
C-1 is necessary, both 1-deoxy-11-hydroxy-∆8-THC-
DMH (3) and 1-deoxy-∆8-THC-DMH (12) have signifi-
cant affinity for both the CB1 and CB2 receptors.
Further, both compounds exhibit pharmacological prop-
erties in the mouse similar to those of ∆9-THC. 1-Deoxy-
11-hydroxy-∆8-THC-DMH (3) also shows typical can-
nabinoid behavior in the rat discrimination procedure.
To the best of our knowledge, the only other 1-deoxy
cannabinoid which has been evaluated using contem-
porary methodology is 1-deoxy-CP 55,940 (13), which
has approximately the same affinity as ∆9-THC for the
CB1 receptor (Ki ) 40.2 ( 13.5 nM), but is ap-
proximately 200 times less potent than CP 55,940 in
vivo.5 On the basis of these data, it is apparent that a
phenolic hydroxyl group at C-1 is not essential for
cannabinoid activity. The modeling studies described
above suggest that the structural features necessary for
typical cannabinoid activity are the presence of an
oxygen atom to which Lys 192 can hydrogen bond, and
a lipophilic structural unit which can simultaneously
interact with the lipophilic pocket on Helix 6 of the CB1
receptor.
(17) Compton, D. R.; Rice, K. C.; De Costa, B. R.; Razdan, R. K.;
Melvin, L. S.; J ohnson, M. R.; Martin, B. R. Cannabinoid
structure-activity relationships: correlation of receptor binding
and in vivo activities. J . Pharmacol. Exp. Ther. 1993, 265, 218-
226.
(18) Showalter, V. M.; Compton, D. R.; Martin, B. R.; Abood, M. E.
Evaluation of binding in a transfected cell line expressing a
peripheral cannabinoid receptor (CB2): identification of can-
nabinoid receptor subtype selective ligands. J . Pharmacol. Exp.
Ther., in press.
Ack n ow led gm en t. The work at Clemson was sup-
ported by Grant DA03590, that at Kennesaw State
University by Grant DA03934, and that at Virginia
Commonwealth University by Grants DA03672 (Billy
R. Martin) and DA05274 (Mary E. Abood) and Training
Grant DA07027 (Vincent M. Showalter), all from the
National Institute on Drug Abuse.
(19) Little, P. J .; Compton, D. R.; J ohnson, M. R.; Melvin, L. S.;
Martin, B. R. Pharmacology and stereoselectivity of structurally
novel cannabinoids in mice. J . Pharmacol. Exp. Ther. 1988, 247,
1046-1051.
(20) Balster, R. L.; Prescott, W. R. ∆9-Tetrahydrocannabinol dis-
crimination in rats as a model for cannabis intoxication. Neu-
rosci. Biobehav. Rev. 1992, 16, 55-62.
Su p p or tin g In for m a tion Ava ila ble: Details of the syn-
thesis of 1-deoxy-HU-210 (3) and 1-deoxy-∆8-THC-DMH (12)
and details of the modeling studies (13 pages). Ordering
information is given on any current masthead page.
(21) Wiley, J . L.; Lowe, J . A.; Balster, R. L.; Martin, B. R. Antagonism
of the discriminative stimulus effects of ∆9-tetrahydrocannabinol
in rats and rhesus Monkeys. J . Pharmacol. Exp. Ther. 1995,
275, 1-6.
(22) An alternative, but energetically less favorable, docking position
maintains the hydrogen bond of the C-11 hydroxyl with Tyr 5.39,
and the C-3 side chain interacts with the hydrophobic binding
pocket.
(23) Gareau, Y.; Dufresne, C.; Gallant, M.; Rochette, C.; Sawyer, N.;
Slipetz, D. M.; Tremblay, N.; Weech, P. K.; Metters, K. M.;
Labelle, M. Structure activity relationships of tetrahydrocan-
nabinol analogues on human cannabinoid receptors. BioMed.
Chem. Lett. 1996, 6, 189-194.
Refer en ces
(1) Song, Z. H.; Bonner, T. I. A lysine residue of the cannabinoid
receptor is critical for receptor recognition by several agonists
but not WIN-55,212. Mol. Pharmacol. 1996, 49, 891-896.
(2) Bramblett, R. D.; Panu, A. M.; Ballesteros, J . A.; Reggio, P. H.
Construction of a 3D model of the cannabinoid CB1 receptor:
determination of helix ends and helix orientation. Life Sci. 1995,
56, 1971-1982.
(3) Razdan, R. K. Structure-activity relationships in cannabinoids.
Pharmacol. Rev. 1986, 38, 75-149.
(4) Rapaka, R. S.; Makriyannis, A. Structure-Activity Relationships
of the Cannabinoids; NIDA Research Monograph 79; National
Institute on Drug Abuse: Rockville, MD, 1987.
(24) Ge´rard, C. M.; Mollereau, C.; Vassart, G.; Parmentier, M.
Molecular cloning of a human cannabinoid receptor which is also
expressed in testis. Biochem. J . 1991, 279, 129-134.
J M960394Y