Structural modifications of the cannabinoid side chain towards C3-aryl and 1′,1′-cycloalkyl-1′-cyano cannabinoids

Article abstract of DOI:10.1016/j.bmcl.2005.12.026

The compounds reported in this study are Δ⁸-THC analogues in which the C3 five-carbon linear side chain of Δ⁸-THC was replaced with aryl and 1′,1′-cycloalkyl substituents. Of the compounds described here analogues 2d (CB₁,

Full text of DOI:10.1016/j.bmcl.2005.12.026

Bioorganic & Medicinal Chemistry Letters 16 (2006) 1616–1620
Structural modifications of the cannabinoid side chain towards
C3-aryl and 10,10-cycloalkyl-10-cyano cannabinoids
Demetris P. Papahatjis,a,* Victoria R. Nahmias,a Thanos Andreou,a Pusheng Fanb
and Alexandros Makriyannisb
aInstitute of Organic and Pharmaceutical Chemistry, National Hellenic Research Foundation,
48 Vass. Constantinou, Athens 116-35, Greece
bCenter for Drug Discovery, Northeastern University, 360 Huntington Avenue, 116 Mugar Life Sciences Building,
Boston, MA 02115, USA
Received 9 September 2005; revised 29 November 2005; accepted 7 December 2005
Available online 4 January 2006
Abstract—The compounds reported in this study are D8-THC analogues in which the C3 five-carbon linear side chain of D8-THC
was replaced with aryl and 10,10-cycloalkyl substituents. Of the compounds described here analogues 2d (CB1, Ki = 11.7 nM. CB2,
Ki = 9.39 nM) and 2f (CB1, Ki = 8.26 nM. CB2, Ki = 3.86 nM) exhibited enhanced binding affinities for CB1 and CB2, exceeding that
of D8-THC. Efficient procedures for the synthesis of these novel cannabinoid analogues are described.
Ó 2006 Elsevier Ltd. All rights reserved.
In the quest for cannabinoid receptor binding affinity
and selectivity, pharmacophoric requirements of classi-
cal cannabinoids have been extensively examined,
though recurrently revised. During the last decade,
numerous selective ligands for each of the cannabinergic
proteins were designed and synthesized.1 Many of these
agents serve as important molecular probes, providing
structural information about receptor binding sites.
Structure–activity relationship (SAR) studies have indi-
cated four pharmacophores within the cannabinoid
structure:2 the phenolic hydroxyl, the lipophilic alkyl
side chain, the northern aliphatic hydroxyl and the
southern aliphatic hydroxyl. Among these, the cannabi-
noid side chain has been the object of extensive research
for more than half a century. Structural variations with-
in this pharmacophore can result in analogues varying
by up to three orders of magnitude in receptor affinity
and pharmacological potency. The spatial orientation
of the side chain seems to play a pivotal role in canna-
binergic activity. Introduction of a triple bond in the
benzylic position of the n-heptyl-D8-THC analogue led
to the classical cannabinoid 1a with high CB1 affinity,3
originally synthesized in our laboratory. Thus, several
efforts towards chain orientation with regard to the tri-
cyclic cannabinoid moiety were instigated.4 Tetracyclic
analogues synthesized a year later,5 also aimed at
restricting the cannabinoid side chain, exhibited de-
creased affinity for the CB1 receptor, probably due to
unfavourable interaction between the fourth ring and
the receptor. On the other hand, the remarkable binding
affinity and CB1 selectivity exhibited by the bulky ada-
mantyl-D8-THC6 analogue 1b, indicated greater toler-
ance for steric bulk in the receptor subsite. Quite
recently, it has been shown that variations in the ada-
mantyl substituents can lead to cannabinoid analogues
possessing higher affinities and selectivities for each of
the two receptors depending on the relative orientation
of the adamantyl group with respect to the tricyclic can-
nabinoid structure7 (Fig. 1).
Taking these results into consideration, we were
prompted to seek novel substituents for the cannabi-
noid side chain, which would introduce bulk close to
the phenolic ring of D8-THC and, at the same time,
could constrain the orientation of the alkyl side chain,
since the chain is flexible and can achieve any one of a
number of conformations. Here we describe the syn-
thesis of a carefully designed series of analogues, in
which a variety of aryl substituents were introduced
Keywords: Cannabinoid analogues; Suzuki coupling; Biaryls.
*
at the C3 position of the phenolic
D8-THC, as depicted in Figure 2.
A ring of
Corresponding author. Tel.: +30 10 7273885; fax: +30 10 7273831;
e-mail: dpapah@eie.gr
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.12.026

D. P. Papahatjis et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1616–1620
1617
OH
OH
OH
1´
6´
CN
O
CN
O
R
O
R =
Figure 3. Structural modification of the side chain towards bulky,
cyano functionalized cannabinoids.
(-)- 8-THC
1b : R =
1a : R =
OH
2d : R =
2e : R =
2f : R =
2a : R =
2b : R =
R
HO
CN
CN
R =
cannabidiol
CN
7e : R =
7a : R =
7d : R =
2c : R =
Figure 1.
CN
7f : R =
Figure 4.
OH
1
OH
2
3
cycloalkyl-10-cyano-substituted cannabinoids, as depict-
ed in Figure 3. Conformational restriction within the
chain was achieved through the incorporation of the
side chain carbons into five- and six-membered rings.
The C10 cyano functionalization served to ensure en-
hanced receptor binding affinities.10
O
O
1'
4
4'
Figure 2. Structural modification of the side chain towards C3-aryl
cannabinoids.
The presence of the biaryl unit in a great number of
pharmaceutical compounds has also been a significant
motivation to us, in designing biaryl cannabinoids.
The biphenyl framework is without doubt a privileged
substructure. This motif has shown activity across a
wide range of therapeutic classes, which include anti-
fungal, anti-inflammatory, antirheumatic, antitumor,
antihypertensive and antiatherosclerotic agents.8 The
protein binding of drugs that contain aromatic substitu-
ents is dominated by interactions with aromatic and
hydrophobic residues. In addition, aromatics have been
shown to form favourable interactions with polar sub-
stituents, such as amides or hydroxyl groups, and even
positively charged moieties.9 This diversity in receptor
selectivity is not overly surprising when one considers
that aromatic moieties have long been considered as ma-
jor players in molecular recognition.8 Thus, we have
sought to enrich our library of cannabinoid analogues
with the synthesis of biaryl-containing compounds, in
an effort to obtain high-affinity ligands for the two
cannabinoid receptors (Figs. 1 and 2).
We have also included in our SAR studies the synthetic
bicyclic intermediates 7a, 7d, 7e and 7f which represent
modifications of cannabindiol (Fig. 4), a significant can-
nabis constituent, which has been shown to exhibit weak
binding affinity for the cannabinoid receptors.
The key step in the C3-aryl cannabinoid synthetic path-
way was the formation of 5-aryl resorcinols, through the
application of the Suzuki–Mijaura reaction.11 The com-
mercially available nickel catalyst NiCl2(PPh3)2, reduced
in situ to the active Ni(0) species, was used over
3,5-dimethoxychloride with various arylboronic sub-
strates (Scheme 1).
4-Alkyl-phenylboronic acids were employed to achieve
varying chain lengths, whereas the naphthyl substrate
provided further bulk at the C10 position. The resulting
4-alkyl biaryls were obtained in increasing yields as the
chain length decreased. However, the Suzuki–Mijaura
procedure did not prove to be as efficient on the naph-
thyl-boronic substrate, which exhibited the lowest reac-
tion yield (Table 1).
Additionally, in order to further explore the pharmaco-
phoric requirements of the cannabinoid side chain with-
in the classical tetrahydrocannabinol template, we have
extended our studies to include the synthesis of 10,10-
Compounds 5a–5d were efficiently deprotected using
highly selective B-I-9BBN reagent to afford the respective

1618
D. P. Papahatjis et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1616–1620
OH
OMe
NiCl2(PPh3)2, PPh3,
K3PO4.H2O
OMe
OMe
OMe
a
b
ArB(OH)2
+
toluene, 90oC
CN
( )n
CN
( )n
CN
HO
MeO
MeO
MeO
Cl
MeO
Ar
5e, n=3
5f, n=4
6e, n=3
6f, n=4
3
4a, 4b, 4c, 4d
5a, 5b, 5c, 5d
OH
OH
a : Ar =
c : Ar =
d : Ar =
c
d
CN
( )n
CN
( )n
O
HO
7e, n=3
7f, n=4
2e, n=3
2f, n=4
b : Ar =
Scheme 3. Reagents and conditions: (a) Br(CH2)3CH2Br for 5e and
Br(CH2)4CH2Br for 5f, KHMDS, dry THF, 0 °C; (b) BBr3, CH2Cl2,
ꢁ78 °C; (c) (+)-cis/trans-p-mentha-2,8-dien-1-ol, p-TSA, dry benzene,
5 °C, (d) BF3ÆEt2O, CH2Cl2, 0 °C.
Scheme 1. Application of the Suzuki–Mijaura reaction on various
aryl-boronic acids.
Table 1. Reaction yields of the Suzuki–Mijaura coupling
Table 2. Affinities (Ki) of tetrahydrocannabinol and cannabidiol
analogues for cannabinoid receptors CB1 and CB2
Compound
Yield (%)
Reaction time (h)
5a
5b
5c
5d
73
50
44
35
12
14
14
12
Compound
Kia (nM)
Ratio CB1/CB2
CB1
CB2
D8-THC
1a
1b
47.6
0.65
6.80
39.3
3.1
52
1.2
0.2
0.1
resorcinols.12 (+)-cis/trans-p-Mentha-2,8-dien-1-ol13 was
then introduced, in the presence of catalytic amount of
p-TSA, followed by ring closure through treatment with
2a
2b
95.49 ( 16.67) 71.81 ( 12.17) 1.3
119 ( 30.04)
57.77 ( 7.17)
11.73 ( 1.62)
27.9 ( 5.06)
8.26 ( 0.96)
51.7 ( 12.79)
107.8 ( 18.45) 0.5
2.3
2c
2d
3
9.39 ( 1.54)
25.2 ( 4.79)
3.86 ( 0.47)
1.2
1.1
2.1
BF3ÆEt2O in dry CH2Cl2 , leading to biaryl cannabi-
2e
noids 2a, 2b, 2c and 2d14 (Scheme 2).
2f
7a
638.1 ( 106.4) 374.4 ( 60.55) 1.7
753.5 ( 178.1) 221.6 ( 36.75) 3.4
The synthesis of 10,10-cycloalkyl-10-cyano cannabinoid
analogues involves cyclobisalkylation of 3,5-dimethoxy-
phenylacetonitrile with the appropriate a,x-dibromoal-
kane,15 followed by demethylation through the use of
boron tribromide.4b Terpene coupling and ring closure,
as mentioned above, afforded cannabinoid analogues 2e
and 2f14 (Scheme 3).
7d
7e
255.4 ( 45.2)
319 ( 67.55)
105 ( 26.1) 2.4
110.7 ( 22.68) 2.9
7f
a Affinities for CB1 and CB2 were determined using rat brain (CB1) or
mouse spleen (CB2) membranes and [3H]CP-55,940 as the radioli-
gand following previously described procedures.4,16 Ki values were
obtained from three independent experiments run in duplicate and
are expressed as means of three values.
The abilities of 2a–2f and cannabidiol (7a, 7d, 7e and 7f)
analogues to displace radiolabelled CP-55,940 from
purified rat forebrain synaptosomes and mouse spleen
synaptosomes were determined as described else-
where.4,16 Ki values calculated from the respective dis-
placement curves are listed in Table 2 and serve as
indicators for the affinities of these D8-THC analogues
for the CB1 and CB2 receptors.
OH
OMe
OH
a
b
HO
Ar
MeO
Ar
Ar
7a, 7b, 7c, 7d
The compounds reported in this study are D8-THC ana-
logues in which the C3 five-carbon linear side chain of
D8-THC was replaced with aryl and 10,10-cycloalkyl sub-
stituents. Interestingly, replacing the side chain of D8-
THC with bulky groups did not abolish binding affinity.
Both 3-(1-naphthyl)-D8-tetrahydrocannabinol and 10,10-
cyclohexyl-10-cyano analogues, 2d and 2f, exhibited en-
hanced affinities for CB1 and CB2 exceeding that of
D8-THC. Conversely, the 4-alkyl-phenyl cannabinoids
2a–2c exhibited modest binding affinity for both can-
nabinoid receptors. More specifically, the methyl-phenyl
analogue 2a exhibited comparable binding affinity for
the two cannabinoid receptors, whereas the ethyl-phenyl
HO
5a, 5b, 5c, 5d
6a, 6b, 6c, 6d
OH
c
O
Ar
2a, 2b, 2c, 2d
Scheme 2. Reagents and conditions: (a) B-I-9-BBN, hexanes, rt; (b)
(+)-cis/trans-p-mentha-2,8-dien-1-ol, p-TSA, dry benzene, 5 °C; (c)
BF3ꢀEt2O, CH2Cl2, 0 °C.

D. P. Papahatjis et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1616–1620
1619
10. Singer, M.; Ryan, W. J.; Saha, B.; Martin, B. R.; Razdan,
R. K. J. Med. Chem. 1998, 41, 4400.
analogue 2b and the propyl-phenyl analogue 2c showed
moderate receptor selectivity. Analogue 2b exhibited 2-
fold CB2 selectivity, while analogue 2c showed prefer-
ence for the CB1 cannabinoid receptor. 10,10-Cyclopen-
tyl-10-cyano analogue 2e exhibited adequate Ki values
for both receptors without any selectivity whatsoever.
11. (a) Miyaura, N.; Suzuki, A. Synth. Commun. 1981, 11, 513;
(b) Inada, K.; Miyaura, N. Tetrahedron 2000, 56, 8657.
12. Furstner, A.; Seidel, G. J. Org. Chem. 1997, 62, 2332.
13. (+)-cis/trans-p-Mentha-2,8-dien-1-ol was supplied by
Firmenich, Princeton, NJ.
14. All new compounds 2a, 2b, 2c, 2d, 2e, 2f, 7a, 7d, 7e and
7f were fully characterized by NMR, MS and HRMS
spectra. Selected data of final cannabinoids.
Cannabidiol analogues 7a, 7d, 7e and 7f exhibited en-
hanced CB2 receptor selectivity although their binding
affinities are approximately one order of magnitude low-
er than the binding values of their respective D8-tetrahy-
drocannabinol analogues.
Compound 2a: 1H NMR (300 MHz, CDCl3) d 7.43 (d,
J = 7.9 Hz, 2H), 7.19 (d, J = 7.9 Hz, 2H), 6.68 (d,
J = 1.8 Hz, 1H, 4-H), 6.50 (d, J = 1.8 Hz, 1H, 2-H),
5.45 (br s, 1H, 8-H), 4.84 (s, 1H, OH), 3.23 (dd,
J = 17.1 Hz, J = 4.3 Hz, 1H, 10a-H), 2.76 (ddd as td,
J = 11.0 Hz, J = 4.3 Hz, 1H, 10a-H), 2.36 (s, 3H, CH3),
2.17 (m, 1H, 7a-H), 1.95–1.82 (m, 3H, 10b-H, 7b-H, 6a-
H), 1.71 (s, 3H, 9-CH3), 1.40 (s, 3H, 6b-CH3), 1.13 (s,
3H, 6a-CH3); mass spectrum m/z (relative intensity) 334
(M+, 100), 291 (30), 266 (22), 251 (95), 205 (25), 84 (93).
Exact mass calculated for C23H26O2, 334.1933. Found:
334.1931.
In conclusion, the present study shows that bulky substit-
uents on the C3 position of the cannabinoid moiety are
well tolerated by both receptors, whereas some preference
for CB1 or CB2 is observed based on different chain
lengths of the 4-alkyl-phenyl analogues or upon ring size
variation at the C10-position in the case of C10-cyano-
substituted cannabinoid analogues. The above findings
provide interesting additions to the currently available
SAR for the cannabinoid side chain, while opening up a
new field of research on cannabinoid analogues character-
ized by the privileged biaryl substructure.
Compound 2b: 1H NMR (300 MHz, CDCl3) d 7.45 (d,
J = 7.9 Hz, 2H), 7.22 (d, J = 7.9 Hz, 2H), 6.69 (d,
J = 1.8 Hz, 1H, 4-H), 6.50 (d, J = 1.8 Hz, 1H, 2-H), 5.45
(br s, 1H, 8-H), 4.86 (s, 1H, OH), 3.24 (dd, J = 17.7 Hz,
J = 4.4 Hz, 1H, 10a-H), 2.79–2.63 (m, 3H, 10a-H, –CH2–),
2.17 (m, 1H, 7a-H), 1.95–1.82 (m, 3H, 10b-H, 7b-H, 6a-H),
1.72 (s, 3H, 9-CH3), 1.41 (s, 3H, 6b-CH3), 1.26 (t,
J = 7.3 Hz, 3H, –CH3), 1.14 (s, 3H, 6a-CH3). Exact mass
calculated for C24H28O2, 348.2089. Found: 348.2091.
Compound 2c: 1H NMR (300 MHz, CDCl3) d 7.45 (d,
J = 7.9 Hz, 2H), 7.19 (d, J = 7.9 Hz, 2H), 6.68 (d,
J = 1.7 Hz, 1H, 4-H), 6.50 (d, J = 1.7 Hz, 1H, 2-H), 5.45
(br s, 1H, 8-H), 4.87 (s, 1H, OH), 3.23 (dd, J = 16.5 Hz,
J = 4.3 Hz, 1H, 10a-H), 2.77 (m, 1H, 10a-H), 2.60 (t,
J = 7.3 Hz, 2H, –CH2–), 2.16 (m, 1H, 7a-H), 1.95–1.82 (m,
3H, 10b-H, 7b-H, 6a-H), 1.72 (s, 3H, 9-CH3), 1.65 (m, 2H,
–CH2–), 1.40 (s, 3H, 6b –CH3), 1.14 (s, 3H, 6a-CH3), 0.95 (t,
J = 7.3 Hz, 3H, –CH3); mass spectrum m/z (relative
intensity) 362 (M+, 100), 347 (13), 319 (26), 294 (18), 279
(54), 241 (15). Exact mass calculated for C25H30O2, 362.246.
Found: 362.2246.
Acknowledgments
This work was supported by the National Hellenic
Research Foundation and by the G.S.R.T. Programs
‘Location and use of research results by the creation
of new enterprises (spin-off)’ and YB/60.
References and notes
1. (a) Pertwee, R. G. Curr. Med. Chem. 1999, 6, 635; (b)
Khanolkar, A. D.; Palmer, S. L.; Makriyannis, A. Chem.
Phys. Lipids 2000, 108, 37; (c) Palmer, S. L.; Thakur, G.
A.; Makriyannis, A. Chem. Phys. Lipids 2002, 121, 3; (d)
Thakur, G. A.; Nikas, S. P.; Makriyannis, A. Mini-Rev.
Med. Chem. 2005, 5, 631.
Compound 2d: 1H NMR (300 MHz, CDCl3) d 8.04 (d,
J = 7.9 Hz, 1H), 7.88 (d, J = 7.3 Hz, 1H), 7.80 (d,
J = 7.9 Hz, 1H), 7.50–7.40 (m, 4H), 6.59 (d, J = 1.8 Hz,
1H, 4-H), 6.40 (d, J = 1.8 Hz, 1H, 2-H), 5.48 (br s, 1H, 8-H),
4.87 (s, 1H, OH), 3.29 (dd, J = 15.9 Hz, J = 4.3 Hz, 1H, 10a-
H), 2.82 (m, 1H, 10a-H), 2.19 (m, 1H, 7a-H), 1.98–1.87 (m,
3H, 10b-H, 7b-H, 6a-H), 1.74 (s, 3H, 9-CH3), 1.41 (s, 3H,
6b-CH3), 1.18 (s, 3H, 6a-CH3); mass spectrum m/z (relative
intensity) 370 (M+, 100), 327 (36), 287 (80), 205 (60), 149
(67). Exact mass calculated for C26H26O2, 370.1933. Found:
370.1930.
2. Makriyannis, A.; Rapaka, R. S. Life Sci. 1990, 47,
2173.
3. Papahatjis, D. P.; Kourouli, T.; Abadji, V.; Goutopoulos,
A.; Makriyannis, A. J. Med. Chem. 1998, 41, 1195.
4. (a) Papahatjis, D. P.; Nikas, S. P.; Andreou, T.; Makri-
yannis, A. Bioorg. Med. Chem. Lett. 2002, 12, 3583; (b)
Papahatjis, D. P.; Nikas, S. P.; Kourouli, T.; Chari, R.;
Xu, W.; Pertwee, R. G.; Makriyannis, A. J. Med. Chem.
2003, 46, 3221.
Compound 2e: 1H NMR (300 MHz, CDCl3) d 6.51 (d,
J = 1.8 Hz, 1H), 6.39 (d, J = 1.8 Hz, 1H), 5.70 (s, 1H, OH),
5.42 (d, J = 3.7 Hz, 1H, 8-H), 3.21 (dd, J = 15.5 Hz,
J = 4.3 Hz 1H, 10a-H), 2.70 (td, J = 11.0 Hz, J = 4.9 Hz,
1H, 10a-H), 2.37 (m, 2H, –CH2–), 2.11–1.86 (m, 10H, 7a-H,
10b-H, 7b-H, 6a-H, 6H–CH2–), 1.70 (s, 3H, 9-CH3), 1.37 (s,
3H, 6b-CH3), 1.09 (s, 3H, 6a-CH3); mass spectrum m/z
(relative intensity) 337 (60), 294 (47), 254 (100), 205 (65).
Exact mass calculated for C22H27NO2, 337.2042. Found:
337.2041.
5. Khanolkar, A. D.; Lu, D.; Fan, P.; Tian, X.; Makriyannis,
A. Bioorg. Med. Chem. Lett. 1999, 9, 2119.
6. (a) Luk, T.; Jin, W.; Zvonok, A.; Lu, D.; Lin, X.-Z.;
Chavkin, C.; Makriyannis, A.; Mackie, K. Br. J. Pharma-
col. 2004, 142, 495; (b) Jarbe, T. U. C.; DiPatrizio, N. V.; Lu,
D.; Makriyannis, A. Behav. Pharmacol. 2004, 15, 517.
7. Lu, D.; Meng, Z.; Thakur, G. A.; Fan, P.; Steed, J.;
Tartal, C. L.; Hurst, D. P.; Reggio, P. H.; Deschamps, J.
R.; Parrish, D. A.; George, C.; Jarbe, T. U. C.; Lamb, R.
J.; Makriyannis, A. J. Med. Chem. 2005, 48, 4576.
8. Horton, D. A.; Bourne, G. T.; Mark, L.; Smythe, M. L.
Chem. Rev. 2003, 103, 893.
Compound 2f: 1H NMR (300 MHz, CDCl3) d 6.56 (d,
J = 1.8 Hz, 1H), 6.41 (d, J = 1.8 Hz, 1H), 5.87 (s, 1H, OH),
5.42 (d, J = 3.6 Hz, 1H, 8-H), 3.23 (dd, J = 16.0 Hz,
J = 4.3 Hz, 1H, 10a-H), 2.71 (ddd as td, J = 11.0 Hz,
J = 4.3 Hz, 1H, 10a-H), 2.11 (m, 3H, 7a-H, 2H–CH2–),
9. Hajduk, P. J.; Bures, M.; Praestgaard, J.; Fesik, S. W.
J. Med. Chem. 2000, 43, 3443.

1620
D. P. Papahatjis et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1616–1620
1.81–1.64 (m, 14H, 10b-H, 7b-H, 6a-H, 9-CH3, 8H –CH2–),
15. Papahatjis, D. P.; Nikas, S.; Tsotinis, A.; Vlachou, M.;
Makriyannis, A. Chem. Lett. 2001, 192.
16. Dodd, P. R.; Hardy, J. A.; Oakley, A. E.; Edwardson,
J. A.; Perry, E. K.; Delaunoy, J. P. Brain Res. 1981,
226, 107.
1.37 (s, 3H, 6b-CH3), 1.09 (s, 3H, 6a-CH3); mass spectrum
m/z (relative intensity) 351 (85), 308 (52), 268 (93), 205 (100).
Exact mass calculated for C23H29NO2, 351.2198. Found:
351.2198.