Synthesis of a potential photoactivatable anandamide analog

Article abstract of DOI:10.1016/j.bmcl.2006.04.059

A potential photoreactive analog of anandamide was synthesized via selective hydrogenation of a skipped tetrayne intermediate. This compound might be a useful tool to search for new cannabinoid receptors.

Full text of DOI:10.1016/j.bmcl.2006.04.059

Bioorganic & Medicinal Chemistry Letters 16 (2006) 3765–3768
Synthesis of a potential photoactivatable anandamide analog
Laurence Balas,a,* Maria Grazia Cascio,b,c Vincenzo Di Marzob and Thierry Duranda
aUMR CNRS 5074, Faculty of Pharmacy, Universite´ Montpellier 1, 15 av. Charles Flahault B.P. 14491,
34093 Montpellier Cedex 5, France
bEndocannabinoid Research Group, Institute of Biomolecular Chemistry, Via Campi Flegrei 34, 80078 Pozzuoli (NA), Italy
c
`
Dipartimento di Scienze Farmaceutiche, Universita degli Studi di Salerno, Fisciano, Italy
Received 20 March 2006; revised 18 April 2006; accepted 18 April 2006
Available online 6 May 2006
Abstract—A potential photoreactive analog of anandamide was synthesized via selective hydrogenation of a skipped tetrayne
intermediate. This compound might be a useful tool to search for new cannabinoid receptors.
Ó 2006 Elsevier Ltd. All rights reserved.
Several unsaturated fatty acid derivatives have been
identified as the endogenous ligands for the two cannab-
inoid receptor subtypes cloned thus far: the CB1 and
CB2 receptors. These lipid mediators are referred to as
endocannabinoids (eCBs) in order to distinguish them
from the natural cannabinoid, D9-tetrahydrocannabinol
(THC), derived from the Cannabis sativa plant.1 The
eCBs identified to date include2,3 the eicosanoids; N-ara-
chidonoylethanolamide (anandamide), 2-arachidonoyl-
glycerol (2-AG), 2-arachidonylglyceryl ether (noladin
ether), O-arachidonoylethanolamide (virohamine), and
N-arachidonoyl-dopamine, together with the ethanol-
amides of docosatetraenoic and homo-c-linolenic acids.
They have been suggested to participate in several phys-
iological and pathological conditions in mammals.4,5
Several excellent reviews on endocannabinoids and their
synthesis, metabolism, and functions have appeared
during the last three years.6 However, despite the recent
considerable advances and the extensive studies under-
taken over the past 14 years, the eCB system is still
not completely understood.7
Recently, it was reported that GPR55, an orphan G-
protein-coupled receptor, might represent a new CB
receptor.10 In addition, thanks to the development of
selective CB1 or CB2 antagonists and to the availability
of gene knock-out and double knock-out mice (CB1À/À,
CB2À/À, and CB1À/À/CB2À/À), there is emerging evi-
dence suggesting that novel cannabinoid receptors may
exist for anandamide in various tissues.11,12 Interesting-
ly, some of these molecular targets are not sensitive to
plant-derived cannabinoids.13,14 To date, none of these
putative receptors have been cloned.
Within this context, we reasoned that an anandamide-
derived and potentially photoactivatable tool might be
useful for the identification of non-CB1, non-CB2 eCB
receptors. We therefore focussed our attention on the
synthesis of an aryl azide probe of anandamide. Our
design featured a flexible synthetic strategy allowing us
to readily change the head and/or tail group at a later
stage 1.
O
While the anandamide cellular uptake mechanism is cur-
rently the source of much controversial debate,8,9
numerous contemporary experiments suggest that eCBs
cause some of their effects independently of the known
CB1 and CB2 receptors.
OH
N
H
Anandamide
O
head group
photoreactive
1
*
tail
Corresponding author. Tel.: +33 467 548623; fax: +33 467 548625;
e-mail: balas@univ-montp1.fr
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.04.059

3766
L. Balas et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3765–3768
Since little is known about the targeted putative recep-
tors, structural requirements are rather unpredictable
at this time. However, it was clearly established that
methylation at the alpha position to the nitrogen atom
significantly enhances the metabolic stability toward
the FAAH enzyme (and also its CB1 binding affinity).15
In addition, SAR with both CB receptors highlights the
importance of the number of double bonds and their
location on the carbon skeleton.16 Thus, we have chosen
to introduce the potentially photoactivatable group in
the tail of (R)-methanandamide. This will allow us also
to infer the influence of the terminal bulky group on the
affinity of anandamide analogs for CB1 and CB2 recep-
tors, a type of information that is limited to date.
the previously21,22 described linear strategy which built
consecutively the diyne, triyne, and then tetrayne back-
bone, we have developed a convergent route to the
skipped tetrayne 4, by coupling the two skipped diynes
5 and 6.
The synthesis of 2 is depicted in Scheme 2.
The precursor 5 was prepared according to Razdan.23
Thus, starting from the commercially available 5-hexy-
noic acid 7, esterification followed by a cross-coupling
reaction with 4-chlorobut-2-yn-1-ol 9 and subsequent
bromination with CBr4, in the presence of PPh3, gave
the expected bromide 5 in 73% yield. The synthesis of
precursor 6 was achieved in two steps in 73% yield via a
coupling reaction between the commercially available 3-
bromo-1-(trimethylsilyl)-1-propyne 12, and 3-butyn-1-ol
11 and subsequent deprotection of the silyl group with
fluoride. The key coupling reaction between the precur-
sors 5 and 6 was carried out under mild conditions24 using
Cs2CO3 and afforded tetrayne 4 in 45% yield. In our
hands, this skipped tetrayne 4 was found to be unstable
when stored in a freezer under argon. However, if used
immediately its selective reduction using the P2-nickel
catalyst subsequently yielded the expected all cis key tetr-
aene 325 (47% yield).
We report herein the total synthesis of compound 2. The
2-azido-5-iodobenzoate ester group has already been suc-
cessfully used in photolabeling experiments.17 Its radiola-
beled analog (125I) is readily obtainable.26 Moreover,
besides their greater stability, aryl azides have the advan-
tage over alkyl azides that their irradiation can be per-
formed with an activation wavelength higher (>300 nm)
than the ultraviolet absorption for most proteins18,19
(thus limiting protein damage). It is hypothesized20 that
the irradiation of the aryl azide by UV light will generate
a nitrene intermediate which would bind to putative non-
CB1, non-CB2 eCB receptors, establish an irreversible
cross-link to these proteins, and permit the isolation and
mapping of the receptor site.
The x-hydroxyl function of tetraene 3 was protected as a
dimethoxytrityl group. Upon saponification of the result-
ing tetraene 14, the amide coupling reaction with the O-
silylated (R)-(À)-2-amino-1-propanol 16, using the mixed
anhydride procedure, provided the expected amide 17
along with its corresponding x-free hydroxyl analog 18.
This partial deprotection of the dimethoxytrityl group oc-
curred during the purification step by column chromato-
graphy using Florisil. The remaining protected material
17 was converted into the desired alcohol 18 by treatment
with hexafluoro-2-propanol. Steglich esterification of the
key tetraene 18 with 2-azido-5-iodobenzoic acid 19 (pre-
pared as described by Perrier et al.26) followed by removal
Retrosynthetically, the strategy involves the saponifica-
tion of the ester function of the key intermediate 3 to
introduce the anandamide head group, and esterification
of the x-hydroxy function with 2-azido-5-iodobenzoic
acid (Scheme 1).
The tetraene backbone comes from selective hydrogena-
tion of the tetrayne 4. The three skipped diynes of tetra-
yne 4 were assembled using cross-coupling between
copper(I) alkynides and propargylic halides. Instead of
O
CH3
H
OH
N
O
H
OMe
3
O
O
2
N3
OH
I
O
HO
OMe
4
O
Br
HO
6
OMe
5
Scheme 1. Key advanced intermediates to probe 2.

L. Balas et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3765–3768
3767
OH
HO
b
OH
9
Cl
O
O
O
a
O
d
HO
Br
OH
c
OMe
OMe
OMe
5
7
10
8
SiMe
3
f
e
SiMe
3
+
HO
Br
g
13
HO
HO
11
12
6
O
O
O
h
i
OMe
OMe
OMe
14
3
ODMTr
4
OH
OH
j
O
OH
15
ODMTr
H C
3
OTBDPS
H N 16
k
2
O
CH
O
H
CH
3
3
H
O
CH
H
H
3
OTBDPS
n
OH
OTBDPS
OTBDPS
N
H
N
N
H
H
17
O
O
O
O
ODMTr
l
N
3
N
3
m
2
O
CH
3
COOH
N
20
I
I
N
H
19
3
I
18
OH
Scheme 2. Reagents and conditions: (a) APTS, MeOH, CH2Cl2, reflux, 24 h, 96%; (b) SOCl2, PhH, 55%; (c) CuI, NaI, K2CO3, DMF, 30 °C, 20 h,
85%; (d) CBr4, PPh3, CH2Cl2, 1 h, 90%; (e) CuI, NaI, Cs2CO3, DMF, 30 °C, 18 h, 77%; (f) TBAF, THF, 0 °C, 1 h, 95%; (g) CuI, NaI, Cs2CO3, DMF,
35 °C, 18 h, 45%; (h); P-2 Ni, H2, EtOH, 8 h, 13 °C, 47%; (i) DMTrCl, DMAP, NEt3, CH2Cl2, rt, 15 h, 81%; (j) LiOH, MeOH, 55 °C, 22 h, 96%; (k) i-
BuOCOCl, NMN, MeCN, À5 °C, 3 h; (l) HFIP, 12 h, rt, 65%; (m) DCC, DMAP, CH2Cl2, rt, 15 h, 82%; (n) TBAF, THF, rt, 1 h30 min, 75%.
of the silyl group with TBAF, afforded the expected com-
pound 227 in 61% yield.
allow its use to identify other non-CB1, non-CB2
cannabinoid receptors for anandamide. Its synthesis
was achieved in 11 steps at a 1.6 mmol scale. The
skipped tetrayne backbone was obtained using a con-
vergent strategy starting from four commercially
available reagents; that is, 5-hexynoic acid 7, 1,4-
butynol, 3-bromo-1-(trimethylsilyl)-1-propyne 12 and
3-butyn-1-ol 11. We think that the synthesis of differ-
ent endocannabinoid probes will be attainable in a
similar fashion replacing 2-(R)-aminopropan-1-ol with
a suitable alternative head group (for example, dopa-
mine, aminoacids, etc., or alcohols such as glycerol,
ethanolamine, etc.). The synthesis of these novel
probes together with their binding affinities at CB1
and CB2, and photoaffinity labeling experiments with
probe 2 are in progress and will be reported in due
course.
Compound 2 was tested for its affinity for human re-
combinant CB1 and CB2 receptors using a previously
described procedure and employing membranes from
COS cells overexpressing either receptor and
[3H]CP55-940 as the high affinity ligand.28 The com-
pound exhibited Ki (lM) = 0.57 0.05 and 0.22 0.03
for CB1 and CB2 receptors, respectively (means SE,
N = 3). In the same conditions, anandamide shows Ki
(lM) = 0.07 0.03 and 0.18 0.02, respectively. Thus,
these assays showed, for the first time, that a cumber-
some ‘tail’ group reduces the binding of putative ligands
to CB1 but not CB2 receptors. This residual affinity for
the CB2 receptor might suggest that compound 2 might
be used as a tool for yet-to-be characterized new eCB
receptors. On the other hand, the reduced affinity for
CB1 receptors as compared to anandamide also suggests
that compound 2 might turn out to be a rather selective
tool for the putative new receptor(s).
Acknowledgment
In summary, probe 2 was found to bind to CB1 and,
particularly, CB2 receptors with affinities that should
The authors deeply thank Prof. Paul Evans for proof-
reading the manuscript.

3768
L. Balas et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3765–3768
20. Brase, S.; Gil, C.; Knepper, K.; Zimmermann, V. Angew
Chem., Int. Ed. 2005, 44, 5188.
References and notes
21. Altundas, R.; Mahadevan, A.; Razdan, R. K. Tetrahedron
Lett. 2004, 45, 5449.
22. Li, C.; Xu, W.; Vadivel, S. K.; Fan, P.; Makriyannis, A.
J. Med. Chem. 2005, 48, 6423.
1. Howlett, A. C.; Barth, F.; Bonner, T. I.; Cabral, G.;
Casellas, P.; Devane, W. A.; Felder, C. C.; Herkenham,
M.; Mackie, K.; Martin, B. R.; Mechoulam, R.; Pertwee,
R. G. Pharmacol. Rev. 2002, 54, 161.
23. Dasse, O.; Mahadevan, A.; Han, L.; Martin, B. R.;
Di Marzo, V.; Razdan, R. K. Tetrahedron 2000, 56,
9195.
2. Rodriguez de Fonseca, F.; Del Arco, I.; Bermudez-Silva,
F. J.; Bilbao, A.; Cippitelli, A.; Navarro, M. Alcohol.
Alcoholism 2005, 40, 2.
24. Caruso, T.; Spinella, A. Tetrahedron 2003, 59, 7787.
25. Rf = 0.27 (cyclohexane/AcOEt 7:3). IR (neat, e mÀ1):
3640–3129; 3012; 2927; 2855; 1737; 1436; 1367; 1317; 1241;
1200; 1170; 1048. MS (electro-spray, positive mode): 315.2
(M+Na). 1H NMR (d ppm, 300 MHz, CDCl3 at
7.24 ppm): 5.6–5.15 (m, 8H, H(olefin)); 3.61–3.55 (m,
5H, OMe and H17); 2.83–2.70 (m, 6H, H7, H10 and H13);
2.40–2.20 (m, 4H, H16 and H2); 2.18–1.92 (m, 2H, H4);
1.80–1.58 (m 2H, H3); 1.58–1.45 (br s, OH). 13C NMR (d
ppm, 75 MHz, CDCl3 at 77.0 ppm): 174.07 (COOMe);
130.98 (CH); 128.93 (CH); 128.84 (CH); 128.28 (CH);
128.01 (CH); 125.68 (CH); 62.19 (CH2); 51.46 (OCH3);
33.43 (CH2); 30.85 (CH2); 26.54 (CH2); 25.75 (CH2); 25.62
(CH2); 24.75 (CH2). Anal. Calcd for C18H28O3: C, 73.93;
H, 9.65. Found: C, 74.11; H, 9.47.
3. Hanus, L.; Gopher, A.; Almog, S.; Mechoulam, R. J.
Med. Chem. 1993, 36, 3032.
4. Di Marzo, V.; Bifulco, M.; De Petrocellis, L. Nat. Rev.
Drug. Discov. 2004, 3, 771.
5. Howlett, A. C.; Breivogel, C. S.; Childers, S. R.; Dead-
wyler, S. A.; Hampson, R. E.; Porrino, L. J. Neurophar-
macology 2004, 47(Suppl. 1), 345.
6. (a) Pertwee, R. G. Br. J. Pharmacol. 2006, 147(Suppl. 1),
S163; (b) Jonsson, K. O.; Holt, S.; Fowler, C. J. Basic Clin.
Pharmacol. Toxicol. 2006, 98, 124; (c) Lambert, D. M.;
Fowler, C. J. J. Med. Chem. 2005, 48, 5059.
7. De Petrocellis, L.; Cascio, M. G.; Di Marzo, V. Br. J.
Pharmacol. 2004, 141, 765.
8. Hillard, C. J.; Jarrahian, A. Br. J. Pharmacol. 2003, 140,
802.
26. Perrier, H.; Prasit, P.; Wang, Z. Tetrahedron Lett. 1994,
35, 1501.
9. Glaser, S. T.; Kaczocha, M.; Deutsch, D. G. Life Sci.
2005, 77, 1584.
27. Rf = 0.27 (CH2Cl2/MeOH 94:6). IR (neat, e mÀ1): 3518–
3149; 3012; 2929; 2854; 2124; 1722; 1639; 1550; 1475;
1293; 1239; 1135; 1083. MS (electro-spray, positive
mode): 629.1 (M+Na). 1H NMR (d ppm, 300 MHz,
CDCl3 at 7.24 ppm): 8.10 (d, 1H, J = 2.1 Hz, H(arom));
7.79 (dd, 1H, J = 2.1 Hz and J = 8.5 Hz, H(arom)); 6.96
(d, 1H, J = 8.5 Hz, H(arom)); 5.75–5.25 (m, 9H, NH,
H(olefin)); 4.40–4.20 (m, 2H, H17); 4.15–3.95 (m, 1H,
CHN); 3.70–3.57 (m, 1H, CH2OH); 3.55–3.40 (m, 1H,
CH2OH); 3.00–2.65 (m, 6H, H7, H10 and H13); 2.60–
2.40 (m, 2H, H16); 2.17 (t, 2H, J = 7.6 Hz, H2); 2.16–
19.5 (m, 2H, H4); 1.75–1.59 (m, 2H, H3); 1.14 (d, 3H,
J = 6.7 Hz, CH3). 13C NMR (d ppm, 75 MHz, CDCl3 at
77.0 ppm): 173.63 (CON); 163.75 (COOR); 141.78 (CH);
140.31 (CH); 140.30 (C(quat)); 131.00 (CH); 129.10(CH);
128.75 (CH); 128.41 (CH); 128.28 (CH); 128.04 (CH);
127.82 (CH); 124.70 (CH); 124.43 (C(quat)); 121.74
(CH); 87.33 (C(quat)); 67.33 (CH2OH); 64.94 (CH2O);
47.82 (CHN); 36.06 (CH2); 26.84 (CH2); 26.63 (CH2);
25.64 (CH2); 25.46 (CH2); 17.04 (CH3). Anal. Calcd for
C27H35IN4O4: C, 53.47; H, 5.82; N, 9.24. Found: C,
53.19; H, 5.65; N, 9.40.
10. Baker, D.; Pryce, G.; Davies, W. L.; Hiley, C. R. Trends
Pharmacol. Sci. 2006, 27, 1.
11. Pertwee, R. G. Curr. Neuropharmacol. 2004, 2, 9.
12. Begg, M.; Pacher, P.; Batkai, S.; Osei-Hyiaman, D.;
Offertaler, L.; Mo, F. M.; Liu, J.; Kunos, G. Pharmacol.
Ther. 2005, 106, 133.
13. Baskfield, C. Y.; Martin, B. R.; Wiley, J. L. J. Pharmacol.
Exp. Ther. 2004, 309, 86.
14. Di Marzo, V.; Breivogel, C. S.; Tao, Q.; Bridgen, D. T.;
Razdan, R. K.; Zimmer, A. M.; Zimmer, A.; Martin, B. R.
J. Neurochem. 2000, 75, 2434.
15. Abadji, V.; Lin, S.; Taha, G.; Griffin, G.; Stevenson, L. A.;
Pertwee, R. G.; Makriyannis, A. J. Med. Chem. 1994, 37,
1889.
16. Sheskin, T.; Hanus, L.; Slager, J.; Vogel, Z.; Mechoulam,
R. J. Med. Chem. 1997, 40, 659.
17. Mancini, J. A.; Abramovitz, M.; Cox, M. E.; Wong, E.;
Charleson, S.; Perrier, H.; Wang, Z.; Prasit, P.; Vickers, P.
J. FEBS Lett. 1993, 318, 277.
18. Dorman, G.; Prestwich, G. D. Trends Biotechnol. 2000, 18,
64.
19. Durand, D.; Hullot, P.; Vidal, J. P.; Girard, J. P.; Baneres,
J. L.; Parello, J.; Muller, A.; Bonne, C.; Rossi, J. C.
Bioorg. Med. Chem. Lett. 2000, 10, 811.
28. Brizzi, A.; Brizzi, V.; Cascio, M. G.; Bisogno, T.; Sirianni,
R.; Di Marzo, V. J. Med. Chem. 2005, 48, 7343.