Homologues and isomers of noladin ether, a putative novel endocannabinoid: Interaction with rat cannabinoid CB1 receptors

Article abstract of DOI:10.1016/S0960-894X(02)00839-9

Two regioisomers and 13 analogues of the putative endocannabinoid noladin ether (2-arachidonyl glyceryl ether, 2-AGE, 1) were synthesized and tested for their interaction with CB₁ receptors in rat brain membranes. The results showed that a C-20 tetra-unsaturated moiety is necessary for high affinity, and that a series of alkyl glyceryl ethers of potential occurrence in brain tissues have less affinity than 2-AGE for CB₁ receptors.

Full text of DOI:10.1016/S0960-894X(02)00839-9

Bioorganic & Medicinal Chemistry Letters 13 (2003) 43–46
Homologues and Isomers of Noladin Ether, a Putative Novel
Endocannabinoid: Interaction with Rat Cannabinoid
CB1 Receptors
Giovanni Appendino,a,* Alessia Ligresti,b Alberto Minassi,a Nives Daddario,a
Tiziana Bisognob and Vincenzo Di Marzob,*
aDiSCAFF, Viale Ferrucci 33, 28100 Novara, Italy
bEndocannabinoid ResearchGroup, Istituto di Chimica Biomolecolare, CNR, Via Campi Flegrei 34, Comprensorio A. Olivetti,
Building 70, 80078 Pozzuoli (NA), Italy
Received 19 July 2002; revised 23 September 2002; accepted 4 October 2002
Abstract—Two regioisomers and 13 analogues of the putative endocannabinoid noladin ether (2-arachidonyl glyceryl ether,
2-AGE, 1) were synthesized and tested for their interaction with CB1 receptors in rat brain membranes. The results showed that a
C-20 tetra-unsaturated moiety is necessary for high affinity, and that a series of alkyl glyceryl ethers of potential occurrence in brain
tissues have less affinity than 2-AGE for CB1 receptors.
# 2002 Elsevier Science Ltd. All rights reserved.
Noladin ether (1),1 the 2-glyceryl ether of arachidonyl
alcohol, is the latest polyunsaturated fatty acid deriva-
tive to be proposed as an endogenous ligand for the
cannabinoid CB1 receptor,2 one of the two recognition
sites for marijuana’s psychoactive principle (À)-Á9-tetra-
hydrocannabinol (THC) and for the two previously
described endocannabinoids,3 N-arachidonoyl-ethano-
lamine (anandamide, AEA, 2) and 2-arachidonoyl gly-
cerol (2-AG, 3). While the chemical bases for the
interaction of AEA and 2-AG with cannabinoid CB1
receptors have been investigated in detail,4 nothing is
known on the structure–activity relationships under-
lying the cannabimimetic activity of alkyl glyceryl
ethers. Hence, two enantiomeric 2-AGE regioisomers
(sn-1 and sn-3 arachidonyl glyceryl ether) and a series of
2-AGE analogues (Table 1) were prepared and tested
for CB1 affinity in specific binding assays (Fig. 1).
Results and Discussion
The terminal glyceryl ethers 5a–d were obtained by
reacting l- or d-solketal (glycerol acetonide, 4a,b) with
arachidonyl mesylate, affording, after deprotection, the
sn-1- and sn-3 ethers 5a,b, respectively (Scheme 1). The
2-alkyl glyceryl ethers 7a–m were obtained by treatment
of 1,3-benzilydenglycerol (6)1 with fatty alcohol mesy-
lates (final compounds 7a–i) or isoprenoid bromides
(final compounds 7l,m), followed by acidic deprotection
(HCl in MeOH) (Scheme 1). The base employed in the
etherification was dimsyl anion (Scheme 1, Conditions
B), except when skipped double bonds were present. In
this case, KOH in toluene had to be employed (Scheme
1, Conditions A).1
Figure 1. Chemical structures of endocannabinoids.
*Corresponding authors. Tel.: +39-081-867-5093; fax: +39-081-804-
1770; e-mail: vdimarzo@icmib.na.cnr.it
Scheme 1. Synthesis of the glyceryl ethers 5a–d and 7a–m. Conditions A:
RCH2–OMs, KOH, benzene, 40 ꢀC; Conditions B: RCH2–OMs (RCH2–
Br for the isoprenoid residues), NaH, DMSO, tetrabutylammonium
iodide, rt. For the nature of R and the yields, see Table 1.
(V. Di Marzo); Tel.:
+39-011-670-7684; fax: +39-011-670-7687; e-mail: appendin@
pharm.unito.it (G. Appendino).
0960-894X/03/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00839-9

44
G. Appendino et al. / Bioorg. Med. Chem. Lett. 13 (2003) 43–46
Table 1. Affinity for rat cannabinoid CB1 receptors of noladin ether analoguesa,b
Entry
R
Etherification site
(conditions, yield)
Ki (mM)
Krel
5a
5b
5c
5d
7a
7b
7c
7d
7e
sn-1 (A, 34%)
sn-3 (A, 33%)
sn-3 (B, 48%)
sn-1 (B, 52%)
2 (A, 21%)
5.0Æ0.8
2.5Æ0.4
>50
1250
625
>12,500
>12,500
3475
>50
13.9Æ1.2
22.5Æ1.8
25.6Æ2.1
>50*
2 (A, 26%)
5625
2 (B, 38%)
6400
2 (B, 39%)
>12,500
>12,500
2 (B, 47%)
>50
7f
2 (B, 26%)
2 (B, 36%)
>50*
>50
>12,500
>12,500
7g
7h
7i
2 (B, 41%)
2 (B, 64%)
>50
>50
>12,500
>12,500
7l
2 (B, 50%)
2 (B, 77%)
>50
>50
>12,500
>12,500
7m
aThe site of etherification, stereochemistry and type of chain are shown (see Scheme 1).
bThe affinity is expressed in terms of absolute Ki (mM) for the displacement of [3H]SR141716A from cannabinoid CB1 receptors in rat brain mem-
brane preparations, as well as in terms of relative Ki, that is Krel=Ki/Ki of HU-210. The Ki for the displacement by HU-210 of [3H]SR141716A from
rat brain membranes was 4.0Æ0.3 nM. For 2-AGE, Ki=2.5Æ0.3 mM; Krel=625. The symbol * indicates that it was possible to measure some dis-
placement at a 50 mM concentration.
This change gave overall lower yields, but avoided double
bond conjugation and preserved homoconjugation.5
higher than that reported in synaptosomal membranes
from rat brain (without brainstem) using [3H]HU-243 as
radioligand, and centrifugation instead of filtration to
separate bound from unbound ligand.1 Variations of
this type are not uncommon.2,7 On the other hand, the
comparison of the ratios between the Ki constants of 2-
AGE and HU-210 measured in rat brain membranes
with [3H]SR141716A and in rat brain synaptosomes
with [3H]HU-243 showed very similar values, with 2-
AGE being about 2.5 orders of magnitude less potent
than HU-210 as a CB1 receptor ligand in both cases.8
Therefore, the relative CB1 affinity constants within a
series of homologues should always be the same, inde-
pendently of the binding assay used. For this reason,
Table 1 shows not only the Ki values of each com-
pounds but also their ratios with the Ki value calculated
here for HU-210 (Krel).
The choice of the alkyl moiety, though far from
exhaustive, was done making reference to fatty and iso-
prenoid alcohols of potential natural occurrence as gly-
ceryl ethers. Since arachidonic acid is not easily
amenable to point mutations, modification of the olefin
moiety (cyclopropanation, isomerization and inclusion
in a conjugate system) were carried out for the C-18
monounsaturated alkyl residue. The fifteen alkyl gly-
ceryl ethers prepared were then tested as cannabinoid
CB1 receptor ligands using a widely accepted proce-
dure,2,6 namely the equilibrium competition of increas-
ing concentrations of the test compounds with a fixed
concentration of the high affinity CB1 ligand,
[3H]SR141716A, in rat brain membranes, which contain
high levels of CB1 receptors.2,7
The first finding of this study is that the two enantio-
meric 2-AGE regioisomers sn-1- and sn-3-arachidonyl
glyceryl ethers (5a,b) exhibit affinity for CB1 receptors
similar to that of 2-AGE (Fig. 2). However, the enan-
The Ki value for the compounds tested are reported in
Table 1. The binding affinity we measured for 2-AGE
(Ki=2.5 mM) in our assay system was about 100-fold

G. Appendino et al. / Bioorg. Med. Chem. Lett. 13 (2003) 43–46
45
methylene derivative of 2-erucyl glyceryl ether (7h) was
devoid of significant activity. Finally, replacement of the
arachidonyl moiety with three widespread alkyl groups
(palmityl (7i), farnesyl (7l) and geranyl (7m)) also abol-
ished activity. No activity was also observed for the sn-3
palmityl and sn-1 geranyl ethers (5c and 5d, respectively)
(Table 1).
Taken together, our findings: (1) extend to 2-AGE pre-
vious observations on the relevance of a tetra-unsatu-
rated C-20 chain for high affinity at CB1 receptors;4,10
and (2) indicate that a range of alkyl glyceryl ethers of
potential natural occurrence would not be more potent
endocannabinoids than 2-AGE. These results support
the view that 2-AGE, rather than its homologues and
analogues, is an endocannabinoid, and provide infor-
mation on the structural basis of its interactions with
the CB1 receptor binding site. The hydrophilic head of
2-AGE and its regioisomers provide ample opportunity
for chemical modification. Given the increased potency
attained by modifications of the polar moiety of other
endocannabinoids and the metabolic stability of gly-
ceryl ethers, 2-AGE analogues that retain the arachido-
nyl moiety and bear suitable modifications in the
glyceryl head might represent an interesting opportunity
to overcome the pharmacokinetics shortcomings of
other types of endocannabinoids.
Figure 2. Displacement by increasing doses of noladin ether (2-AGE)
or its positional isomers, sn-1 and sn-3-arachidonyl glyceryl ethers, of
[3H]SR141716A from cannabinoid CB1 receptors in rat brain mem-
brane preparations. The displacement assay was carried out as descri-
bed in the Methods. Data points are means of three separate
experiments. Error bars are not shown for the sake of clarity and were
never higher than 5%.
tiomer more likely to be naturally occurring (sn-1, 5a) is
2-fold less potent than 2-AGE (1) and its sn-3 enantio-
mer 5b (Table 1). This is not surprising, since a high
degree of enantioselectivity for CB1 binding was repor-
ted for several chiral cannabinoid ligands, including
THC itself.2,7 Our finding is important in view of the
controversy generated by the discovery of 2-AGE in
animal tissues.1 Before the discovery of 2-AGE, the only
natural glyceryl ethers known were those with the alkyl
(or alkenyl) chain in the sn-1 position, as exemplified by
PAF and the plasmalogens.9 Our findings show that,
even if the sn-1 regioisomer (5a) of 2-AGE was to be
recognized as the most abundant naturally occurring
arachidonyl glyceryl ether in mammalian brain, this
compound would have lower affinity for CB1 receptors
than 2-AGE.
Acknowledgements
We are grateful to Chemi (Cinisello Balsamo, MI, Italy)
for a generous gift of d- and l-solketal, and to Giu-
seppe Guilla for his help in the synthesis of several gly-
ceryl ethers. Support from MURST (grant 3933 to
V.D.M.) is acknowledged.
References and Notes
The other original finding of this study is that, as with
AEA and 2-AG,10 also for alkyl gliceryl ethers a poly-
unsaturated C-20 chain is important for cannabinoid
activity. Thus, shortening of the side chain without a
substantial decrease of the overall degree of unsatura-
tion in terms of double bonds per carbon atoms, as
featured in the C-18 o-6 polyunsaturated 2-AGE
analogue 2-g-linolenoyl-glyceryl ether (7a), caused a
5.5-fold loss of potency (Table 1). Furthermore, within
the C-18 series of analogues, affinity for CB1 receptors
gradually decreased with saturation, the 2-linoleyl (7b)
and 2-oleyl (7c) homologues being 9- and 10-fold less
potent than 2-AGE, and 1.6- and 1.85-fold less potent
than 2-g-linolenoyl-glyceryl ether (7a), respectively
(Table 1). In compounds from the C-18 series, one cis
double bond was sufficient to impart a certain activity at
CB1 receptors, but this did not hold true for higher
homologues, since the C-22 homologue, 2-erucyl gly-
ceryl ether (7d), was almost inactive. Cyclopropanation
of 2-oleyl-glyceryl ether as in 7e, or replacement of the
double bond with a trans-enine construct as in 7f were
both detrimental for activity, and the residual affinity of
2-ximenyl-glyceryl ether (7f) was totally abolished by
cyclopropanation of the double bond (7g). Likewise, the
1. Hanus, L.; Abu-Lafi, S.; Fride, E.; Breuer, A.; Vogel, Z.;
Shalev, D. E.; Kustanovich, I.; Mechoulam, R. Proc. Natl.
Acad. Sci. U.S.A. 2001, 98, 3662.
2. Pertwee, R. G. Pharmacol. Ther. 1997, 74, 129.
3. (a) Devane, W. A.; Hanus, L.; Breuer, A.; Pertwee, R. G.;
Stevenson, L. A.; Griffin, G.; Gibson, D.; Mandelbaum, A.;
Etinger, A.; Mechoulam, R. Science, 1992, 258, 1946. (b)
Mechoulam, R.; Ben-Shabat, S.; Hanus, L.; Ligumsky, M.;
Kaminski, N. E.; Schatz, A. R.; Gopher, A.; Almog, S.; Mar-
tin, B. R.; Compton, D. R.; Pertwee, R. G.; Griffin, G.;
Bayewitch, M.; Barg, J.; Vogel, Z. Biochem. Pharmacol. 1995,
50, 83. (c) Sugiura, T.; Kondo, S.; Sukagawa, A.; Nakane, S.;
Shinoda, A.; Itoh, K.; Yamashita, A.; Waku, K. Biochem.
Biophys. Res. Commun. 1995, 215, 89.
4. (a) Di Marzo, V.; Bisogno, T.; De Petrocellis, L.; Melck,
D.; Martin, B. R. Curr. Med. Chem. 1999, 6, 721. (b) Palmer,
S. L.; Khanolkar, A. D.; Makriyannis, A. Curr. Pharm. Des.
2000, 6, 1381. (c) Reggio, P. H. Prostaglandins Leukot. Essent.
Fatty Acids 2002, 66, 143.
5. All compounds were characterized by IR, 1H and 13C
NMR, as well as MS data. Reaction conditions A were adop-
ted without modification from ref 1. The synthesis of 2-farne-
syl gyceryl ether (7l) is presented as an example of procedure
B: To a suspension of NaH (60% dispersion in mineral oil, 70
mg, 1.63 mMol, 1.5 mol equiv) in dry THF (5 mL), 1,3-

46
G. Appendino et al. / Bioorg. Med. Chem. Lett. 13 (2003) 43–46
benzylidenglycerol (196 mg, 1.09 mMol, 1 mol. equiv) was
added. When gas evolution ceased, tetrabutylammonium
iodide (20 mg, cat.) and E,E-farnesylbromide (310 mg, 295 mL,
1.09 mMol) were added. After stirring at rt for 3 h, the reac-
tion was worked up by dilution with petroleum ether–EtOAc
3:1 and washed with NH4Cl. After washing with brine, drying
(Na2SO4), and evaporation, the residue was dissolved in MeOH
(5 mL) and treated with 500 mL of a soln obtained by reacting
AcCl (560 mL) in MeOH (10 mL). After 2 h, the reaction was
worked up by neutralization with NaHCO3 and extraction with
EtOAc. After removal of the solvent, the residue was purified
by CC (13 g silica gel, petroleum ether–EtOAc 9:1 as eluant) to
give 161 mg (50%) 7l as a colourless oil.
Noel, M. A.; Showalter, V. M.; Tao, Q. Biochem. Pharmacol.
1997, 53, 207.
7. Pertwee, R. G. Curr. Med. Chem. 1999, 6, 635.
8. In separate experiments, we observed that the Ki for 2-AGE
using [3H]HU-243 was 0.70Æ0.16 mM (n=3) with rat brain
membranes (Di Marzo, V.; Bisogno, T.; Ligresti, A. unpub-
lished observations).
9. Nagan, N.; Zoeller, R. A. Prog. Lipid. Res. 2001, 40, 199.
10. (a) Felder, C. C.; Joyce, K. E.; Briley, E. M.; Man-
souri, J.; Mackie, K.; Blond, O.; Lai, Y.; Ma, A. L.;
Mitchell, R. L. Mol. Pharmacol. 1995, 48, 443. (b)
Sugiura, T.; Kodaka, T.; Natane, S.; Miyashita, T.; Kondo,
S.; Suhara, Y.; Takayama, H.; Waku, K.; Seki, C.; Baba, N.;
Ishima, Y. J. Biol. Chem. 1999, 274, 2794. (c) Sheskin, T.;
Hanus, L.; Slager, J.; Vogel, Z.; Mechoulam, R. J. Med.
Chem. 1997, 40, 659.
6. (a) Melck, D.; Bisogno, T.; De Petrocellis, L.; Chuang, H.;
Julius, D.; Bifulco, M.; Di Marzo, V. Biochem. Biophys. Res.
Commun. 1999, 262, 275. (b) Abood, M. E.; Ditto, K. E.;