Arachidonylsulfonyl derivatives as cannabinoid CB1 receptor and fatty acid amide hydrolase inhibitors

Article abstract of DOI:10.1016/S0960-894X(03)00721-2

Arachidonylsulfonyl fluoride (3), reported here for the first time, is similar in potency to its known methyl arachidonylfluorophosphonate (2) analogue as an inhibitor of mouse brain fatty acid amide hydrolase activity (IC₅₀ 0.1 nM) and cannabinoid CB1 agonist [³H]CP 55,940 binding (IC₅₀ 304-530 nM). Interestingly, 3 is much more selective than 2 as an inhibitor for fatty acid amide hydrolase relative to acetylcholinesterase, butyrylcholinesterase and neuropathy target esterase. N-(2-Hydroxyethyl)arachidonylsulfonamide (4) is at least 2500-fold less potent than N-(2-hydroxyethyl)arachidonamide (anandamide) (1) at the CB1 agonist site.

Full text of DOI:10.1016/S0960-894X(03)00721-2

Bioorganic& Medicinal Chemistry Letters 13 (2003) 3301–3303
Arachidonylsulfonyl Derivatives as Cannabinoid CB1 Receptor
and Fatty Acid Amide Hydrolase Inhibitors
Yoffi Segall,y Gary B. Quistad, Daniel K. Nomura and John E. Casida*
Environmental Chemistry and Toxicology Laboratory, Department of Environmental Science, Policy and Management,
University of California, Berkeley, CA 94720-3112, USA
Received 30 August 2002; accepted 30 May 2003
Abstract—Arachidonylsulfonyl fluoride (3), reported here for the first time, is similar in potency to its known methyl arachidonyl-
fluorophosphonate (2) analogue as an inhibitor of mouse brain fatty acid amide hydrolase activity (IC50 0.1 nM) and cannabinoid
CB1 agonist [3H]CP 55,940 binding (IC50 304–530 nM). Interestingly, 3 is much more selective than 2 as an inhibitor for fatty acid
amide hydrolase relative to acetylcholinesterase, butyrylcholinesterase and neuropathy target esterase. N-(2-Hydroxy-
ethyl)arachidonylsulfonamide (4) is at least 2500-fold less potent than N-(2-hydroxyethyl)arachidonamide (anandamide) (1) at the
CB1 agonist site.
# 2003 Elsevier Ltd. All rights reserved.
Arachidonic acid derivatives play key roles in the func-
tion and understanding of the brain cannabinoid sys-
tem. N-(2-Hydroxyethyl)arachidonamide (anandamide)
(1) is an important endogenous agonist for the canna-
binoid CB1 receptor,1 binding to the same site as Á9-
tetrahydrocannabinol and [3H]CP 55,940, the most
widely used radioligand for the agonist target.2 Anand-
amide is hydrolyzed by fatty acid amide hydrolase
(FAAH) which is very sensitive to irreversible inhibition
by methyl arachidonylfluorophosphonate (MAFP) (2).3
and specifically reports the preparation and unusual
potency and selectivity properties of 3 and 4 (Fig. 1).
The synthesis of 3 and 4 involved a three-step conver-
sion of arachidonic acid, via the corresponding alcohol7
and arachidonyl bromide,8 to arachidonylsulfonyl chlo-
ride (5).9 Compound 5 was converted to 310 by treat-
ment with Bu4N+Fꢀ and to 411 with ethanolamine
(Scheme 1).
The inhibitory potencies of arachidonyl derivatives 1–4
and two dodecyl analogues were compared for CB1 and
FAAH of mouse brain (Table 1). The findings for 1 and
the dodecyl analogues were generally similar to earlier
studies with rat, thereby validating the assay methods
for evaluation of the new arachidonylsulfonyl deriva-
tives 3 and 4 (Table 1).
Agonist binding is also irreversibly inhibited by 2,3,4
involving several possible mechanisms: the arachidonyl
moiety of 2 either occluding a site within the receptor
which would otherwise be occupied by anandamide3 or
causing a conformational change; phosphonylation of
an amino acid moiety in the vicinity of the agonist site.
Alkanesulfonyl derivatives are also of interest since
n-dodecanesulfonyl fluoride5 and methyl n-dodecyl-
fluorophosphonate6 are similarly potent inhibitors of
both FAAH and CB1. The exceptional activity of 2
suggests that arachidonylsulfonyl fluoride (3) might also
be potent at FAAH or CB1 or both. This study incor-
porates the novel arachidonylsulfonyl substituent in
both the fluoride and N-(2-hydroxyethyl)amide series
The most remarkable property of 3 is its outstanding
potency as an inhibitor of mouse brain FAAH, with an
IC50 of 0.1 nM as with the exceptionally-active 2.14
*Corresponding author. Tel.: +1-510-642-5424; fax: +1-510-642-
6497; e-mail: ectl@nature.berkeley.edu
yCurrent address: Israel Institute for Biological Research, PO Box 19,
Ness-Ziona 74100, Israel.
Figure 1. Structural relationship between anandamide (1), methyl
arachidonylfluorophosphonate (2) and their sulfonyl analogues [ara-
chidonylsulfonyl fluoride (3) and N-(2-hydroxyethyl)arachidonyl-
sulfonamide (4)].
0960-894X/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00721-2

3302
Y. Segall et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3301–3303
Lower potency is observed for the CB1 receptor with
IC50 values for agonist binding of 530 nM for 2 and 304
nM for 3.15 Sulfonyl fluoride 3 is also much less potent
and more selective than fluorophosphonate 2 as an
inhibitor of other mouse brain esterases with IC50 values
(nM) (15-min pre-incubation) as follows: 3 780ꢁ120
and 2 0.6ꢁ0.116 for neuropathy target esterase; 3
>100,000 and 2 124ꢁ1716 for acetylcholinesterase; 3
>100,000 and 2 81 for butyrylcholinesterase.17
stituent very effectively positions the –C(O)NHCH2–
CH2OH moiety of 1 for hydrolysis and the –CH2SO2F
and –CH2P(O)(OCH3)F substituents of 3 and 2 for sul-
fonylation and phosphonylation, respectively. These
relationships hold for CB1 although with lower inhibi-
tory potency. However, replacement of the –C(O)–
moiety of 1 with –CH2SO2– of 4 greatly reduces the
activity at CB1. These findings are consistent with low
potency of 4 for the agonist site but moderate potency
of 2 and 3 for a coupled CB1 subsite.
Inhibitory potencies were also compared for carbox-
amide 1 and sulfonamide 4 determined with [3H]CP
55,940 in the absence and presence of phenylmethane-
sulfonyl fluoride (PMSF) to minimize hydrolysis by
FAAH. Carboxamide 1 has an IC50 of 39 nM with
PMSF (vs 40 nM for rat brain CB1 with PMSF)3 or 12
mM without PMSF. Sulfonamide 4 (IC50>100 mM) is at
least 2500-fold less potent than 1 with PMSF. The dif-
ference could be due to either the sulfonyl replacement
for the carbonyl group or the additional methylene of 4
compared with 1.18 Arachidonylsulfonamide 4 (IC50
ꢂ10 mM) is presumably a competitive inhibitor of
FAAH.
Acknowledgements
This study was supported by Grant R01 ES08762 from
the National Institute of Environmental Health Sciences
(NIEHS), NIH, and its contents are solely the respon-
sibility of the authors and do not necessarily represent
the official views of the NIEHS, NIH.
References and Notes
1. 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.
2. Ross, R. A.; Brockie, H. C.; Fernando, S. R.; Saha, B.;
Razdan, R. K.; Pertwee, R. G. Br. J. Pharmacol. 1998, 125,
1345.
In conclusion, 3 has outstanding potency and selectivity
for FAAH relative to other esterases and lower activity
at the CB1 receptor. Sulfonamide 4 is also a moderately
potent FAAH inhibitor with low potency for the ago-
nist site. Thus, with FAAH the arachidonyl-type sub-
3. Deutsch, D. G.; Omeir, R.; Arreaza, G.; Salehani, D.; Pre-
stwich, G. D.; Huang, Z.; Howlett, A. Biochem. Pharmacol.
1997, 53, 255.
4. Fernando, S. R.; Pertwee, R. G. Br. J. Pharmacol. 1997,
121, 1716.
5. Deutsch, D. G.; Lin, S.; Hill, W. A. G.; Morse, K. L.;
Salehani, D.; Arreaza, G.; Omeir, R. L.; Makriyannis, A.
Biochem. Biophys. Res. Commun. 1997, 231, 217.
6. Martin, B. R.; Beletskaya, I.; Patrick, G.; Jefferson, R.;
Winckler, R.; Deutsch, D. G.; Di Marzo, V.; Dasse, O.;
Mahadevan, A.; Razdan, R. K. J. Pharmacol. Exp. Ther.
2000, 294, 1209.
7. Suhara, Y.; Takayama, H.; Nakane, S.; Miyashita, T.;
Waku, K.; Sugiura, T. Chem. Pharm. Bull. 2000, 48, 903.
8. Easton, C. J.; Xia, L.; Pitt, M. J.; Ferrante, A.; Poulos, A.;
Rathjen, D. A. Synthesis 2001, 451.
Scheme 1. (a) LiAlH4, Et2O, 0 ꢃC, N2; (b) CBr4, (C6H5)3P, CH2Cl2,
N2, 0 ꢃC; (c) (1) Mg, THF (BrCH2CH2Br), rt, 16 h; (2) SO2Cl2, THF/
hexane, ꢀ60 ꢃC, 30 min; (d) (1) Bu4N+Fꢀ, THF, rt, 2 h; (2) 5% HCl;
(e) NH2CH2CH2OH, Et3N, CHCl3, 0 ꢃC.
9. (all-Z)- Eicosa-5,8,11,14-tetraenesulfonyl chloride (arachi-
donylsulfonyl chloride) (5). Mg turnings were cut into small
pieces and purified by three washes with each of deionized
water, MeOH, acetone and hexane in sequence, then stored
under hexane. Freshly prepared Mg turnings (93. 6 mg, 3. 9
mmol) were covered with freshly dried THF (2 mL) under N2
atmosphere. Dibromoethane (168 mg, 0. 9 mmol) was added
and the slurry was stirred for 30 min (exothermic) until the Mg
was activated. To that solution, arachidonyl bromide (1056
mg, 3 mmol) in THF (3 mL) was added and the slurry was
stirred for 16 h at rt in a sealed flask. Addition of more THF
(7 mL) dissolved all the salt formed during the preparation.
The resulting Grignard reagent was filtered through oven
dried glass wool and added dropwise to freshly distilled
SO2Cl2 (402 mg, 3 mmol) in hexane (25 mL) over 30 min at
ꢀ60 ꢃC. A yellow precipitate was formed and the solution was
stirred an additional 2 h at rt. The reaction was quenched with
5% HCl (30 mL) and the solution was extracted with CHCl3
(3ꢄ20 mL). After removal of the solvent, the resulting color-
less oil was chromatographed on a silica column eluted with
CHCl3/hexane (1:1). The major fraction (500 mg) had Rf 0. 95
Table 1. Inhibitory potencies of arachidonyl derivatives 1–4 and two
dodecyl analogues for [3H]CP 55,940 agonist binding at the CB1
receptor and FAAH activity
IC50ꢁSE (nM, n=3)
CB1
FAAH
Compd
Mouse
Rat
Mouse
Rat
2.53
1
1+PMSF
2
12,000ꢁ300
39ꢁ14
403
203
530ꢁ150
304ꢁ23
0.10ꢁ0.0212
0.11ꢁ0.05a
ꢂ10,000
213
3
4+PMSF
n-C12H25SO2F
n-C12H25P(O)(OR)Fb 1.8ꢁ0.813 2.5ꢁ0.36
>100,000
6.9ꢁ3.613
185
35
213
3.0ꢁ0.26
aThe inhibition curve for 3, but not 2, is apparently biphasic, suggest-
ing the involvement of two enzyme components of differential sensi-
tivity to 3.
bR=i-C3H7 and CH3 for assays with mouse and rat brain membranes,
respectively.

Y. Segall et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3301–3303
3303
on TLC/silica with the same solvent system and was identified
1
3.25 (2H, q), 3.05 (2H, m), 2.81 (6H, m), 2.66 (1H, br s), 2.71
(4H, m), 1.83 (2H, m), 1.53 (2H, m), 1.30 (6H, m), 0.90 (3H, br
s); 13C NMR: 130.73, 129.14, 128.99, 128.84, 128.51, 128.30,
128.03, 127.73, 62.10, 52.94, 45.54, 31.70, 30.56, 29.52, 28.44,
27.42, 26.85, 25.86, 25.77, 23.47, 22.75, 14.27. EI/MS: m/z 397,
M+ (20%), 79 (100%). EI-HRMS: calcd for C22H39NO3S
(M+): 397.2651 found: 397.2658.
12. Quistad, G. B.; Sparks, S. E.; Casida, J. E. Toxicol. Appl.
Pharmacol. 2001, 173, 48.
13. Segall, Y.; Quistad, G. B.; Casida, J. E. Synth. Commun.
2003, 33, 2151.
by H NMR [5.38 (8H, m), 2. 83 (6H, m), 2.06 (6H, m), 1.70
(2H, m), 1.33 (6H, m), 0. 90 (6H, br s)] and EI/MS (m/z 274,
M+) as eicosa-5,8,11,14-tetraene. The fraction at Rf 0.3–0.5
was collected and rechromatographed using similar conditions
1
to give 120 mg of pure 5 as a colorless oil (11%). H NMR:
5.38 (8H, m), 3.75 (2H, m), 2.83 (4H, m), 2.19 (2H, br q), 2.05
(4H, m), 1.60 (2H, m), 1.28 (8H, m), 0.89 (3H, br s). EI/MS:
m/z 337, MꢀCl+ (20%), 273, MꢀCl–SO+2 (20%), 79 (100%).
EI-HRMS calculated for C20H33SO2 (MꢀCl+): 337. 2201,
found: 337. 2199.
10. (all-Z)- Eicosa-5,8,11,14-tetraenesulfonyl fluoride (arachi-
donylsulfonyl fluoride) (3). Using a syringe, Bu4N+Fꢀ (80 mL
of 1 M solution in THF, 0.081 mmol) was introduced into a
solution of 5 (20 mg, 0.054 mmol) in THF (1 mL). The solu-
tion became brown and was stirred for 2 h. It was then quen-
ched into 5% HCl (20 mL) at 0 ꢃC and extracted with CHCl3
(2ꢄ25 mL), dried (Na2SO4), concentrated to 2 mL and passed
with CHCl3 through a short silica column to give 18 mg pure 3
(94%). 1H NMR: 5.39 (8H, m), 3.37 (2H, m), 2.82 (6H, br
quintet), 2.14 (2H, br q), 2.05 (2H, br q), 1.97 (2H, br quintet),
1.57 (4H, m), 1.31 (4H, m), 0.90 (3H, br s); 13C NMR: 130. 78,
129. 62, 128. 93, 128. 72, 128. 51, 128. 12, 128. 00, 127. 76, 51.
10 (d, JF-P=15.8 Hz), 38.50, 31.76, 30.59, 29.55, 27.99, 27.48,
26.58, 25.92, 23.23, 22.78, 14.24. EI/MS: m/z 356, M+ (40%),
258, MꢀF–SO2–CH2–H+ (15%), 79 (100%). EI-HRMS:
calcd for C20H33SO2F (M+):356. 2185 found: 356. 2195.
11. (all-Z)- Eicosa-5,8,11,14-tetraenesulfonic acid N-(2-hydroxy-
ethyl)amide [N-(2-hydroxyethyl)arachidonylsulfonamide] (4).
Arachidonylsulfonyl chloride (5, 50 mg, 0. 13 mmol) in CHCl3
(2 mL, ethanol free) was added to a stirred and cooled (0 ꢃC)
solution of NH2CH2CH2OH (41 mg, 0.67 mmol) and Et3N
(14 mg, 0.13 mmol) in CHCl3 (3 mL). A precipitate of
14. FAAH inhibition was determined for mouse brain mem-
branes (700 g supernatant, 10,000 g pellet; 8 mg protein) in Tris
buffer (100 mM, pH 7.0, with 1 mM EDTA) incubated with
test inhibitors (2–4) for 15 min, then with [14C]oleamide (1
mM) for 15 min at 25 ꢃC. An extraction procedure was used to
analyze [14C]oleic acid liberated by hydrolysis according to ref
12.
15. CB1 binding assays used the same mouse brain mem-
branes (150–200 mg protein) as those for FAAH in Tris buffer
(50 mM, pH 7.4, with 1 mM EDTA and 1 mM MgCl2) and 3
mg/mL bovine serum albumin. Test inhibitors (1–4) or ligand
for non-specific binding (10 mM 2) were preincubated 15 min
at rt prior to addition of the cannabinoid radioligand [3H]CP
55,940 (10 nM). After incubation for 90 min at 30 ꢃC, CB1-
bound radiolabel was recovered by vacuum filtration for
quantification. Potential hydrolysis of 1 and 4 by FAAH was
examined and minimized by assaying with PMSF (50 mM, 15-
min pre-incubation) or alone. Assay conditions of Quistad, G.
B.; Nomura, D. K.; Sparks, S. E.; Segall, Y.; Casida, J. E.
Toxicol. Lett. 2002, 135, 89.
16. Quistad, G. B.; Sparks, S. E.; Segall, Y.; Nomura, D. K.;
Casida, J. E. Toxicol. Appl. Pharmacol. 2002, 179, 57.
17. Assays according to refs 12 and 16.
.
HOCH2CH2NH2 HCl was formed. Monitored by TLC (silica,
EtOAc, Rf 0.85), the solution was stirred for 1 h at rt. The
CHCl3 was removed and the residue was dissolved in EtOAc
(3 mL) and passed through a short (8 cm) silica column with
EtOAc(50 mL). The solvent was distilled to leave 45 mg of
pure 4 as a colorless oil (87%). The product gave one spot on
TLC/silica developed with EtOAc/CHCl3 1:1 (Rf 0.5). 1H
NMR: 5.38 (8H, m), 5.11 (1H, t, J=6.2 Hz), 3.75 (2H, br s),
18. Similar potency as an FAAH inhibitor is observed for 2
(Fig. 1) and its analogue lacking one methylene on the ara-
chidonyl moiety [i.e., RP(O)(OCH3)F] according to Seltzman,
H. H.; Fleming, D. N.; Deutsch, D. G.; Glaser, S. T.; Ste-
venson, L.; Ross, R.; Pertwee, R. G. Abstracts of Papers, 2001
Symposium on the Cannabinoids, International Cannabinoid
Research Society: Burlington, VT, 2001.