Development of the first ultra-potent "capsaicinoid" agonist at transient receptor potential vanilloid type 1 (TRPV1) channels and its therapeutic potential

Article abstract of DOI:10.1124/jpet.104.074864

Olvanil (N-9-Z-octadecenoyl-vanillamide) is an agonist of transient receptor potential vanilloid type 1 (TRPV1) channels that lack the pungency of capsaicin and was developed as an oral analgesic. Vanillamides are unmatched in terms of structural simplici

Full text of DOI:10.1124/jpet.104.074864

0022-3565/05/3122-561–570$20.00
THE JOURNAL OF PHARMACOLOGY AND EXPERIMENTAL THERAPEUTICS
Copyright © 2005 by The American Society for Pharmacology and Experimental Therapeutics
JPET 312:561–570, 2005
Vol. 312, No. 2
74864/1184193
Printed in U.S.A.
Development of the First Ultra-Potent “Capsaicinoid” Agonist
at Transient Receptor Potential Vanilloid Type 1 (TRPV1)
Channels and Its Therapeutic Potential
Giovanni Appendino, Luciano De Petrocellis, Marcello Trevisani, Alberto Minassi,
Nives Daddario, Aniello Schiano Moriello, David Gazzieri, Alessia Ligresti, Barbara Campi,
Gabriele Fontana, Christian Pinna, Pierangelo Geppetti, and Vincenzo Di Marzo
Dipartimento di Scienze Chimiche Alimentari, Farmaceutiche e Farmacologiche (G.A., A.M., N.D.), Novara, Italy;
Endocannabinoid Research Group, Istitute of Cibernetica “Eduardo Caianiello” (L.D.P., A.S.M.) and Institute of Biomolecular
Chemistry (A.S.M., A.L., V.D.M.), Consiglio Nazionale delle Ricerche, Pozzuoli (NA), Italy; Department of Clinical and
Experimental Medicine (M.T., D.G., B.C.), Pharmacology Unit, University of Ferrara, Ferrara, Italy; Indena Ricerche, S.p.A.
(G.F.), Milan, Italy; Dipartimento di Scienze Farmacologiche (C.P.), University of Milan, Milan, Italy; and Department of Critical
Care Medicine and Surgery (P.G.), Clinical Pharmacology Unit, University of Florence, Florence, Italy
Received July 23, 2004; accepted September 2, 2004
ABSTRACT
Olvanil (N-9-Z-octadecenoyl-vanillamide) is an agonist of tran- (90 pM), phenylacetylrinvanil (PhAR, IDN5890) is the most po-
sient receptor potential vanilloid type 1 (TRPV1) channels that tent vanillamide ever described with potency comparable with
lack the pungency of capsaicin and was developed as an oral that of resiniferatoxin (EC50, 11 pM). Benzoyl- and phenylpro-
analgesic. Vanillamides are unmatched in terms of structural pionylrinvanil were as potent and less potent than PhAR, re-
simplicity, straightforward synthesis, and safety compared with spectively, whereas configurational inversion to ent-PhAR and
the more powerful TRPV1 agonists, like the structurally com- cyclopropanation (but not hydrogenation or epoxidation) of the
plex phorboid compound resiniferatoxin. We have modified the double bond were tolerated. Finally, iodination of the aromatic
fatty acyl chain of olvanil to obtain ultra-potent analogs. The hydroxyl caused a dramatic switch in functional activity, gen-
insertion of a hydroxyl group at C-12 yielded a compound erating compounds that behaved as TRPV1 antagonists rather
named rinvanil, after ricinoleic acid, significantly less potent than agonists. Since the potency of PhAR was maintained in rat
than olvanil (EC50 ϭ 6 versus 0.7 nM), but more versatile in dorsal root ganglion neurons and, particularly, in the rat urinary
terms of structural modifications because of the presence of an bladder, this compound was investigated in an in vivo rat model
additional functional group. Acetylation and phenylacetylation of urinary incontinence and proved as effective as resinifera-
of rinvanil re-established and dramatically enhanced, respec- toxin at reducing bladder detrusor overactivity.
tively, its potency at hTRPV1. With a two-digit picomolar EC50
The “transient receptor potential” (TRP) channels are stretching. Of the vanilloid-type (TRPV) subfamily of TRP
characterized by six trans-membrane domains and by perme-
ability to several cations, including Ca2ϩ. Several members
of this large family of plasma membrane channels function as
sensors for physical stimuli such as temperature higher or
lower than physiological, changes in osmotic pressure, and
receptors (Gunthorpe et al., 2002), TRPV1 and TRPV4 re-
spond to stimulation with natural products, with capsaicin
and resiniferatoxin being the best known and most thor-
oughly studied natural TRPV1 agonists (Sterner and Szal-
lasi, 1999) and 4␣-phorbols being capable of activating
TRPV4 (Nilius et al., 2004). Another common feature of
TRPV1 and TRPV4 is their capability of being gated by
endogenous ligands, which are distinct arachidonate deriva-
tives in both cases (Di Marzo et al., 2002a; Nilius et al., 2004).
It is now recognized that TRPV1 functions as a molecular
This work was partly funded by a research grant from Indena, S.p.A. (to
G.A. and V.D.M.).
Article, publication date, and citation information can be found at
http://jpet.aspetjournals.org.
doi:10.1124/jpet.104.074864.
ABBREVIATIONS: TRP, transient receptor potential; TRPV, vanilloid-type subfamily of TRP receptors; PhAR, phenylacetylrinvanil; DCC, dicy-
clohexylcarbodiimide; DMAP, 4-dimethylaminopyridine; ESI-MS, electrospray ionization-mass spectrometry; PPAA, propylphosphonic acid
anhydride; MCPBA, meta-chloroperoxybenzoic acid; DIAD, diisopropylaxodicarboxylate; HEK, human embryonic kidney; AMT, anandamide
membrane transporter; DRG, dorsal root ganglia; PBS, phosphate-buffered solution; DMEM, Dulbecco’s modified Eagle’s medium; FBS, fetal
bovine serum; FAAH, fatty acid amide hydrolase.
561

562
Appendino et al.
integrator of nociceptive stimuli, including heat, protons, and
plant toxins, and is most abundant in peripheral sensory
fibers of the C type. Studies carried out with transgenic mice
lacking a functional TRPV1 receptor have implicated this
protein in the perception of thermal and inflammatory pain
(Caterina et al., 2000; Davis et al., 2000). Other investiga-
tions showed that TRPV1 is also involved in inflammatory
bowel disorders (Yiangou et al., 2001), neuropathic pain
(Walker et al., 2003), and pathological cough (see for review,
Chung and Chang, 2002). TRPV1 might also play an impor-
tant role in physiological conditions, for example, by allowing
Materials and Methods
Synthesis of Compounds
The synthetic procedure for the preparation of acylrinvanils and
their iodinated analogs (Fig. 1) is schematized in Scheme 1. The
experimental details and the characterization data for the key inter-
mediates and final products were as follows.
Synthesis of O-Acylrinvanils. The exemplificative synthesis of
phenylacetylrinvanil (PhAR, 1e) is as follows:
2Ј,2Ј,2Ј-Trichloroethylricinoleate: To a solution of ricinoleic acid (3
g, 10.07 mmol) in toluene (30 ml), 2,2,2-trichloroethanol (1.9 ml,
3.0 g, 20.14 mmol, 2 mol Eq), dicyclohexylcarbodiimide (DCC) (2.0 g,
the bladder to function correctly (Birder et al., 2002). The 10.07 mmol, 1 mol Eq), and 4-dimethylaminopyridine (DMAP) (1.23
g, 10.07 mmol, 1 mol Eq) were added. After stirring at room temper-
ature for 18 h, the reaction was filtered and the filtrate evaporated.
The residue was purified by gravity column chromatography (15 g
silica gel, petroleum ether/ethyl acetate 9:1 as eluant) to afford 4.3 g
(quantitative) of a colorless viscous oil. IR (KBr) cmϪ1: 3049, 1758,
1651, 1377, 1098, 808, 721, 569. 1H NMR (300 MHz): ␦ 5.53 (m, 1H),
5.41 (m, 1H), 4.73 (s, 2H), 3.60 (br t, J ϭ 6.0 Hz, 1H), 2.44 (t, J ϭ 7.4
Hz, 2H), 2.20 (t, J ϭ 6.3 Hz, 2H), 2.03 (m, 2H), ca. 1.68 (m, 2 H), ca.
1.20 (br m, 20H), 0.87 (br t, J ϭ 7.1 Hz, 3H).
12-Phenylacetyl-2Ј,2Ј,2Ј-trichloroethylricinoleate: To a solution of
2Ј,2Ј,2Ј-trichloroethylricinoleate (4.3 g, 10.7 mmol) in toluene (30
ml), phenylacetic acid (3.4 g, 25.2 mmol, 2.5 mol Eq), DCC (5.0 g, 25.2
mmol, 2.5 mol Eq), and DMAP (1.8 g, 15.0 mmol, 1.5 mol Eq) were
added. After stirring at room temperature for 30 min, the reaction
was worked up by filtration and evaporation. The residue was puri-
fied by gravity column chromatography on silica gel (35 g, petroleum
ether/ethyl acetate 95:5 as eluant) to afford 5.4 g (quantitative) of a
colorless syrup. IR (KBr) cmϪ1: 3051, 1760, 1736, 1454, 1372, 1260,
1134, 1027. 1H NMR (300 MHz, CDCl3): ␦ 7.25 (m, 5H), 5.41 (m, 1H),
5.29 (m, 1H), 4.86 (quint, J ϭ 6.0 Hz, 1H), 4.73 (s, 2H), 3.58 (s, 2H),
2.45 (br t, J ϭ 6.0 Hz, 2H), 2.26 (m, 2H), 1.96 (m, 2H), 1.68 (m, 2 H),
1.51 (m, 2H), ca. 1.29 (br m), ca. 1.21 (br m), 0.86 (br t, J ϭ 7.1 Hz,
3H).
12-Phenylacetylricinoleic acid: To a stirred solution of 12-pheny-
lacetyl-2Ј,2Ј,2Ј-trichloroethylricinoleate (4.6 g, 8.4 mmol) in acetic
acid/methanol 1:1 (40 ml), activated zinc powder (4.6 g) was added.
After stirring overnight at room temperature, the reaction was
worked up by filtration over Celite and washing with sat. NaHCO3.
After evaporation of the solvent, the residue was purified by gravity
column chromatography on silica gel (60 g, petroleum ether/ethyl
acetate 95:5 as eluant) to afford 1.85 g (53%) of a colorless syrup. IR
(KBr) cmϪ1: 3300–2800 (broad), 1731, 1603, 1586, 1497, 1255, 1106,
964, 724. 1H NMR (300 MHz): ␦ 7.25 (m, 5H), 5.41 (m, 1H), 5.26 (m,
1H), 4.86 (quint, J ϭ 6.0 Hz, 1H), 3.58 (s, 2H), 2.34 (t, J ϭ 6.0 Hz, 2H),
2.28 (br t, J ϭ 6.7 Hz, 2H), 1.97 (m, 2H), 1.62 (m, 2 H), 1.51 (m, 2H),
ca. 1.29 (br m), ca. 1.21 (br m), 0.86 (br t, J ϭ 7.1 Hz, 3H). Electro-
spray ionization-mass spectrometry (ESI-MS): 439 [M ϩ Na]ϩ
[C26H40O4 ϩ Na]ϩ.
Phenylacetylrinvanil (1e): To a solution of 12-phenylacetylricino-
leic acid (1.85 g, 4.4 mmol) in dry dichloromethane (15 ml), vanil-
lamine hydrochloride (835 mg, 4.4 mmol, 2 mol Eq), triethylamine
(2.45 ml, 1.78 g, 17.6 mmol, 4 mol Eq), and propylphosphonic acid
anhydride (PPAA) (50% in ethanol, 3.4 ml, 1.68 g, 5.28 mmol, 1.2 mol
Eq) were added. The reaction was stirred at room temperature for 3 h
and then worked up by evaporation. The residue was purified by
finding of TRPV1 in several brain nuclei (Mezey et al., 2000),
as well as in keratinocytes (Inoue et al., 2002), epithelial cells
(Birder et al., 2002), endothelial cells (Yamaji et al., 2003),
mast cells (Stander et al., 2004), and astrocytes (Doly et al.,
2004), widens considerably its biological importance, sug-
gesting its involvement in the control of several physiological
and pathological functions.
The advantages of targeting TRPV1 for therapeutic pur-
poses are highlighted by the recent finding of a significant
up-regulation of this protein in several pathological condi-
tions, ranging from pruritus and inflammatory or diabetic
neuropathic pain (Rashid et al., 2003; Luo et al., 2004;
Stander et al., 2004) to fecal hyperactivity (Chan et al., 2003),
vulvodynia (Tympanidis et al., 2004), and cancer of the cervix
(Contassot et al., 2004). Indeed, nonpungent synthetic ago-
nists capable of immediately desensitizing TRPV1, and even
selectively deleting TRPV1-expressing nociceptors (Karai et
al., 2004), can be used against inflammatory hyperalgesia,
bladder hyperactivity, emesis, cancer growth, and neuronal
excitotoxicity (Szallasi, 2002; Veldhuis et al., 2003). The re-
spiratory side effects of most agonists, however, argue in
favor of the use of selective TRPV1 antagonists. For many
years, both basic and preclinical studies have relied on olva-
nil (Brand et al., 1987) as the prototypical nonpungent cap-
saicinoid with promising analgesic activity and on capsaz-
epine, which remained the only known TRPV1 antagonist for
over a decade (Bevan et al., 1992). On the other hand, the
most potent TRPV1 agonist discovered so far, resiniferatoxin,
is undergoing clinical trials for the treatment of urinary
incontinence (Giannantoni et al., 2002; Szallasi and Fowler,
2002) and following the iodination of its vanillyl moiety, led
to the most potent TRPV1 antagonist available to date, 5Ј-
iodoresiniferatoxin (Wahl et al., 2001). We have previously
reported that iodination of another TRPV1 agonist, nordihy-
drocapsaicin (Appendino et al., 2003), also yields a potent
TRPV1 antagonist and have found that applying this chem-
ical modification to other “capsaicinoids” transforms them
into antagonists (G. Appendino and V. Di Marzo, unpub-
lished data). Their potential of being rendered potent antag-
onists upon iodination of the aromatic ring represents one
further reason to develop new potent TRPV1 agonists from
long-chain vanillylamides.
gravity column chromatography (50 g of silica gel, petroleum ether/
25
ethyl acetate 7:3) to afford 848 mg (35%) of a colorless syrup. [␣]D
:
In this study, we identified ultra-potent TRPV1 agonists
from the progressive derivatization of the fatty acid chain of
olvanil. We report the finding of the most potent capsaicinoid
TRPV1 agonist ever discovered exhibiting high efficacy in a
rat model of urinary incontinence and describe its structure
activity relationships and its interactions with proteins of the
endocannabinoid system.
ϩ3.2 (methanol, c 1.0), IR (KBr) cmϪ1: 3300–2800 (broad), 3278,
1731, 1515, 1454, 1362, 1261, 1034, 798. 1H NMR (300 MHz, CDCl3):
␦ 7.26 (m, 5 H), 6.85 (d, J ϭ 7.9 Hz, 1H), 6.80 (d, J ϭ 2.0 Hz, 1H), 6.74
(dd, J ϭ 7.9, 2.0 Hz, 1H), 5.69 (br s, 1H), 5.65 (br s, 1H), 5.41 (m, 1H),
5.27 (m, 1H), 4.85 (quint, J ϭ 6.0 Hz, 1H), 4.34 (d, J ϭ 5.8 Hz, 2H),
3.86 (s, 3H), 3.57 (s, 2H), 2.25 (m, 2H), 2.17 (t, J ϭ 7.4 Hz), 1.95 (m,
2H), 1.63 (m, 2 H), 1.50 (m, 2H), 1.27 (br m), 1.20 (br m), 0.85 (br t,

New Ultra-Potent TRPV1 Agonists
563
Fig. 1. Structure of capsaicin (CPS, 1a), olvanil
(1b), rinvanil (1), resiniferatoxin (2), PhAR (1e),
and the other compounds synthesized in this
study.
J ϭ 7.1 Hz, 3H). 13C NMR (75 MHz, CDCl3): ␦ 177.0 (s), 173.5 (s), brine. After removal of the solvent, the residue was purified by
171.7 (s), 156.6 (s), 147.2 (s), 145.4 (s), 141.6 (d), 136.8 (s), 132.7 (s), gravity column chromatography (15 ml silica gel, petroleum ether/
130.2 (d), 129. 5 (d), 129.3 (d), 128.5 (d), 128.3 (d), 127.5 (d), 127.0 (d), ethyl acetate 6:4 as eluant) to afford 149 mg (40%) of 1h as a colorless
126.2 (d), 124.1 (d), 120.6 (d), 114.9 (d), 110.9 (d), 106.1 (d), 74.6 (d),
56.3 (q), 43.8 (t), 43.4 (t), 41.9 (t), 36.6 (t), 31.8 (t), 29.1 (t), 27.2 (t),
syrup. IR (KBr) cmϪ1: 3150, 1731, 1651, 1515, 1274, 1157, 1033, 818,
721. 1H NMR (300 MHz): ␦ 7.26 (m, 5H), 6.86 (d, J ϭ 7.9 Hz, 1H), 6.80
25.9 (t), 25.3 (t), 22.6 (t), 13.9 (q). ESI-MS: 574 [M ϩ Na]ϩ (d, J ϭ 2.0 Hz, 1H), 6.74 (dd, J ϭ 7.9, 2.0 Hz, 1H), 5.70 (br s, 1H), 5.65
[C34H49NO5 ϩ Na]ϩ.
Chemical Modification of Phenylacetylrinvanil
(br s, 1H), 4.95 (quint, J ϭ 6.0 Hz, 1H), 4.33 (d, J ϭ 5.8 Hz, 2H), 3.86
(s, 3H), 3.60 (s, 2H), 2.17 (t, J ϭ 7.4 Hz), 1.63–1.52 (m, 6H), 1.27 (br
Epoxyphenylacetylrinvanil (1g): To a solution of phenylacetylrin- m), 0.84 (br t, J ϭ 7.1 Hz, 3H), 0.58 (m, 3H), Ϫ0.30 (m, 1H). ESI-MS:
vanil (400 mg, 0.72 mmol) in dichloromethane (10 ml), meta-chlo- 586 [M ϩ Na]ϩ [C35H51NO5].
roperoxybenzoic acid (MCPBA) (75%, 318 mg, 1.8 mmol, 2.5 mol Eq)
Dihydrophenylacetylrinvanil (1i): To a solution of phenylacetylr-
was added. After stirring at room temperature overnight, the reac- invanil (300 mg, 0.54 mmol) in methanol (2 ml), Pd(C) was added (ca.
tion was worked up by dilution and washed with 5% Na2S2O3 and 5% 5 mg), and the stirred solution was hydrogenated (balloon pressure).
NaOH. After removal of the solvent, the residue was purified by After 12 h, the reaction was worked up by filtration over Celite and
gravity column chromatography (10 g silica gel, petroleum ether/ evaporation to afford 196 mg (quantitative) of 1i as a colorless syrup.
ethyl acetate 8:2 as eluant) to afford 151 mg (35%) of 1g as a colorless IR (KBr) cmϪ1: 3273, 1731, 1651, 1372, 1273, 1074, 1034, 721. 1H
syrup. IR (KBr) cmϪ1: 3150, 1732, 1654, 1513, 1454, 1271, 1125, NMR (300 MHz): ␦ 7.27 (m, 5H), 6.86 (d, J ϭ 7.9 Hz, 1H), 6.84 (d, J ϭ
1033, 723. 1H NMR (300 MHz, CDCl3): ␦ 7.26 (m, 5H), 6.87 (d, J ϭ 7.9 2.0 Hz, 1H), 6.80 (dd, J ϭ 7.9, 2.0 Hz, 1H), 5.67 (br s, 1H), 4.85 (q, J ϭ
Hz, 1H), 6.82 (br s, 1H), 6.78 (br d, J ϭ 7.9 Hz, 1H), 5.69 (br s, 1H), 6 Hz), 4.35 (d, J ϭ 5.8 Hz, 2H), 3.86 (s, 3H), 3.57 (s, 2H), 2.18 (t, J ϭ
5.64 (br s, 1H), 5.05 (quint, J ϭ 6.0 Hz, 1H), 4.37 (d, J ϭ 5.8 Hz, 2H), 7.4 Hz), 1.63–1.52 (m, 6H), 1.27 (br m), 0.80 (br t, J ϭ 7.1 Hz, 3H).
3.89 (s, 3H), 3.66 (m, 2H), 2.90 (m, 1H), 2.84 (m, 1H), 2.20 (t, J ϭ 7.4 ESI-MS: 576 [M ϩ Na]ϩ [C34H51NO5].
Hz, 2H), ca. 1.76 (m), 1.44 (m), 1.27 (br m), 1.20 (br m), 0.88 (br t, J ϭ
ent-Phenylacetylrinvanil (1j): To a solution of rinvanil (300 mg,
7.1 Hz, 3H). 13C NMR (75 MHz, CDCl3): ␦ 173.0 (s), 171.4 (s), 146.8 0.69 mmol) in tetrahydrofuran (10 ml), phenylacetic acid (90 mg,
(s), 145.2 (s), 134.2 (s), 130.4 (s), 129.3 (d), 128.6 (d), 128.3 (d), 127.3 0.69 mmol), 180 mg of triphenylphosphine (0.69 mmol, 1 mol Eq),
(d), 127.2 (d), 120.9 (d), 114.5 (d), 110.8 (d), 73.0 (d), 57.0, 56.4 (d), and DIAD (0.136 ml, 140 mg, 0.69 mmol, 1 mol Eq) were sequentially
56.0 (t), 53.6, 53.0 (d), 43.6 (t), 41.8 (t), 36.8 (t), 31.7 (t), 29.4 (t), 29.2
(t), 27.5 (t), 25.8 (t), 25.3 (t), 22.6 (t), 14.2 (q). ESI-MS: 590 [M ϩ Na]ϩ was worked up by evaporation. Toluene (ca. 5 ml) was then added,
[C33H47NO6 ϩ Na]ϩ.
and the solution was stored overnight at Ϫ18°C. After filtration, the
added. After stirring at room temperature for 2 days, the reaction
Methylenphenylacetylrinvanil (1h): To a solution of phenylacetyl- solution was evaporated and purified by gravity column chromatog-
rinvanil (380 mg, 0.69 mmol) in dry toluene (29 ml), diethylzinc (1.0 raphy (10 g of silica gel, petroleum ether/ethyl acetate 7:3 as eluant)
M in hexanes, 10.35 mmol, 10.35 ml, 15 mol Eq) and diiodomethane to afford 124 mg (32%) 1i. [␣]D25: Ϫ3.0 (methanol, c 1.2).
(0.837 ml, 2.78 g, 10.35 mmol, 15 mol Eq) were added. The solution
was stirred at 65°C for 7 h and then worked up by cooling to 0°C and
the addition of 2 N H2SO4. The mixture was extracted with ethyl
Synthesis of 5؅- and 6؅-Iodophenylacetylrinvanyl (3a and
3b). The exemplicative synthesis of 3b is as follows:
O-Mem-6Ј-iodorinvanil: To a solution of ricinoleic acid (666 g, 1.46
acetate and the organic phase was washed with sat. NaHCO3 and mmol) in dry dichloromethane (5 ml), 6-iodovanillamine (1.07 g, 2.9

564
Appendino et al.
Scheme 1. General synthetic scheme for the
synthesis of the acylrinvanils 1c–f and 1g–i
(A), 1j (B), and 3a,b (C). Reagents and condi-
tions: i, trichloroethanol, DCC, DMAP, quan-
titative. ii, RCOOH, DCC, DMAP, quantita-
tive. iii, Zn, acetic acid, methanol, 50–60%. iv,
vanillamine, PPAA, triethylamine, 30–40%.
v, MCPBA, dichloromethane, 35%. vi, Et2Zn,
CH2I2, 40%. vii, H2, Pd(C), quantitative. viii,
DIAD, triphenylphosphine, PhCH2COOH,
32%. ix, PPAA, triethylamine, 30–40%. x,
PhCH2COCl, pyridine, 60–70%. xi, SnCl4, tet-
rahydrofuran, 75–80%.
mmol, 2 mol Eq), triethylamine (0.816 ml, 0.59 g, 5.9 mmol, 4 mol 48 h, the reaction was worked up by dilution with ethyl acetate and
Eq), and PPAA (50% in ethanol, 1.1 ml, 0.56 g, 1.76 mmol, 1.2 mol washed with sat. NaHCO3. After removal of the solvent, the residue
Eq) were added. The reaction was stirred at room temperature for 3 h
and then worked up by evaporation. The residue was purified by
was purified by gravity column chromatography (10 g silica gel,
petroleum ether/ethyl acetate 7:3 as eluant) to afford 190 mg (80%)
gravity column chromatography (15 g of silica gel, petroleum ether/ of 3b as a colorless oil. IR (KBr) cmϪ1: 3240, 1728, 1669, 1598, 1497,
ethyl acetate 7:3 as eluant) to afford 331 mg (35%) of a colorless oil. 1440, 1376, 1256. 1H NMR (300 MHz, CDCl3): ␦ 7.34 (s, 1H), 7.28 (m,
IR (KBr) cmϪ1: 3305, 1641, 1543, 1260, 999, 723, 694. 1H NMR (300 5H), 6.94 (s, 1H), 6.05 (br t, J ϭ 5.9 Hz, 1H), 5.40 (m, 1H), 5.30 (m,
MHz, CDCl3): ␦ 7.63 (s, 1H), 6.97 (s, 1H), 5.99 (br t, J ϭ 5.6 Hz, 1H), 1H), 4.87 (quint, J ϭ 6.0 Hz, 1H), 4.39 (d, J ϭ 5.8 Hz, 2H), 3.86 (s,
5.54 (m, 1H), 5.39 (m, 1H), 5.27 (s, 2H), 4.40 (d, J ϭ 5.1 Hz, 2H), 3.83 3H), 3.59 (br s, 2H), 2.28 (m, 2H), 2.19 (t, J ϭ 7.4 H, 2H), 1.975 (m,
(s, 3H), 3.77 (m, 2H), 3.61 (quint, J ϭ 6.0 Hz, 1H), 3.57 (m, 2H), 3.39 2H), 1.60 (m, 2 H), 1.53 (m, 2H), 1.27 (br m), 1.20 (br m), 0.87 (br t,
(s, 3H), 2.33 (m, 2H), 2.19 (t, J ϭ 7.6 Hz), 1.99 (m, 2H), 1.67 (m, 2 H), J ϭ 7.1 Hz, 3H). 13C NMR (75 MHz, CDCl3): ␦ 172.9 (s), 171.3 (s),
1.46 (m, 2H), 1.27 (br m), 1.20 (br m), 0.87 (br t, J ϭ 7.1 Hz, 3H). 147.0 (s), 145.7 (s), 134.2 (s), 133.08 (d), 132.5 (s), 129.6 (s), 129.5 (d),
Chemical ionization-mass spectrometry: 648 [M ϩ H]ϩ [C30H50INO6 129.1 (d), 129.4 (d), 124.9 (d), 124.6 (d), 120.1 (d), 112.6 (d), 87.6 (s),
ϩ H]ϩ.
74.4 (d), 55.9 (d), 47.7 (t), 47.6 (t), 41.6 (t), 36.6 (t), 33.4 (t), 31.8 (t),
6Ј-Iodophenylacetylrinvanil (3b): To a solution of O-mem-6Ј-iodor- 31.5 (t), 29.6 (t), 29.1 (t), 29.0 (t), 28.9 (t), 28.9 (t), 27.1 (t), 25.5 (t),
invanil (383 mg, 0.5 mmol) in pyridine (5 ml), phenylacetylchloride 25.1 (t), 22.45 (t), 14.0 (t). Chemical ionization-mass spectrometry:
(0.33 ml, 156 mg, 1 mmol, 2 mol Eq) was added. After stirring 678 [M ϩ H]ϩ [C34H48INO5 ϩ H]ϩ.
overnight at room temperature, the reaction was worked up by the
addition of water/methanol 9:1 (5 ml) and extraction with ethyl
acetate. The organic phase was sequentially washed with 2N H2SO4,
TRPV1 Assays
sat. NaHCO3 and brine and evaporated. The residue was purified by
gravity column chromatography (10 silica gel, petroleum ether/ethyl man TRPV1 were kindly donated by GlaxoSmithKline (Uxbridge,
acetate 7:3 as eluant) to afford 239 mg of mem-protected 3b (70%) as Middlesex, UK). The pharmacological responses of this cell line to
Human embryonic kidney (HEK)-293 cells overexpressing the hu-
a gum. The latter was dissolved in dry tetrahydrofuran (5 ml) and vanilloid agonists and antagonists are very similar to those previ-
treated with SnCl4 (5 drops). After stirring at room temperature for ously described by Witte et al. (2002) for the same cell line overex-

New Ultra-Potent TRPV1 Agonists
565
pressing the human TRPV1. Cells were grown as monolayers in
PBS and then placed in 2 ml of cold Dulbecco’s modified Eagle’s
minimum essential medium supplemented with nonessential amino medium (DMEM) supplemented with 10% fetal bovine serum (FBS,
acids, 10% fetal calf serum, and 0.2 mM glutamine and maintained heat inactivated), 2 mM L-glutamine, 100 U mlϪ1 penicillin, and 100
under 95%:5% O2/CO2 at 37°C. The effect of the substances on
␮g mlϪ1 streptomycin. The ganglia were then dissociated into single
[Ca2ϩ
] was determined by using Fluo-3, a selective intracellular cells by several passages through a series of syringe needles (23G
i
fluorescent probe for Ca2ϩ (Di Marzo et al., 2002b). One day prior to down to 25G). Finally, the complex of medium and ganglia cells were
experiments, cells were transferred into six-well dishes coated with sieved through a 40-␮m filter to remove debris and topped off with 8
poly-L-lysine (Sigma-Aldrich, St. Louis, MO) and grown in the cul-
ml of DMEM and centrifuged (200g for 5 min). The final cell pellet
ture medium mentioned above. On the day of the experiment, the was resuspended in DMEM [supplemented with 100 ng mlϪ1 mouse
cells (50–60,000 per well) were loaded for 2 h at 25°C with 4 ␮M nerve growth factor (mouse-NGF-7S) and cytosine-b-D-arabino-
Fluo-3 methylester (Molecular Probes, Eugene, OR) in dimethyl sul- furanoside free base (ARA-C) 2.5 ␮M]. Cells were plated on poly-L-
foxide containing 0.04% pluoronic. After the loading, cells were lysine (8.3 ␮M)- and laminin (5 ␮M)-coated 25-mm glass coverslips
washed with Tyrode’s solution pH ϭ 7.4, trypsinized, resuspended in and kept for 2 to 5 days at 37°C in a humidified incubator gassed
Tyrode, and transferred to the cuvette of the fluorescence detector
with 5% CO2 and air. Cells were fed on the second day (and subse-
(PerkinElmer LS50B; PerkinElmer Life and Analytical Sciences, quent alternate days) with DMEM (with 1% instead of 10% FBS).
Boston, MA) under continuous stirring. Experiments were carried Experiments were performed as previously reported (Tognetto et al.,
out by measuring cell fluorescence at 25°C (␭EX ϭ 488 nm, ␭EM ϭ 540 2001). Briefly, plated neurons (2 to 5 days) were loaded with Fura-
nm) before and after the addition of the test compounds at various 2/acetoxymethyl ester (3 ␮M) in Ca2ϩ buffer solution with the fol-
concentrations. In antagonism experiments, varying doses of PhAR lowing composition: 1.4 mM CaCl2, 5.4 mM KCl, 0.4 mM MgSO4, 135
were added 10 min prior to capsaicin (100 nM). Data were expressed
mM NaCl, 5 mM D-glucose, and 10 mM HEPES with 0.1% bovine
for the agonists as the concentration exerting a half-maximal effect serum albumin at pH 7.4 for 40 min at 37°C, washed twice with the
(EC50) and for the antagonists as the concentration exerting a half- Ca2ϩ buffer solution, and transferred to a chamber on the stage of
maximal inhibition (IC50), both calculated by GraphPad. The efficacy Nikon eclipse TE300 microscope. The dye was excited at 340 and 380
of the agonists was determined by comparing it to the analogous nm to indicate relative [Ca2ϩ
] changes by the F340/F380 ratio re-
i
effect observed with 4 ␮M ionomycin.
corded with a dynamic image analysis system (Laboratory Automa-
tion 2.0; RCS, Florence, Italy). The effect of PhAR (1 pM–100 ␮M)
was studied in the presence or absence of the TRPV1 antagonist,
5Ј-iodoresiniferatoxin (10 nM). In antagonism studies, capsaicin (0.1
␮M) was used as the agonist, added 10 min after increasing concen-
trations of the antagonists.
Assays for the Interactions with Proteins of the
Endocannabinoid System
Binding Assays. For CB1 and CB2 receptor binding, [3H]-(Ϫ)-cis-
3-[2-hydroxy-4-(1,1-dimethylheptyl)-phenyl]-trans-4-(3-hydroxy-pro-
pyl)-cyclohexanol (CP-55,940) (Kd ϭ 690 pM) was incubated with P2
membranes from COS cells transfected with either the human CB1
or CB2 receptor as described by the manufacturer (PerkinElmer Life
and Analytical Sciences). Displacement curves were generated by
incubating drugs with 0.5 nM [3H]CP-55,940. In all cases, Ki values
were calculated by applying the Cheng-Prusoff equation to the IC50
values (obtained by GraphPad Software Inc., San Diego, CA) for the
displacement of the bound radioligand by increasing concentrations
of the test compounds.
Fatty Acid Amide Hydrolase Assays. The effect of compounds
on the enzymatic hydrolysis of [14C]anandamide (6 ␮M) was studied
by using membranes prepared from rat brain incubated with in-
creasing concentrations of compounds in 50 mM Tris-HCl, pH 9, for
30 min at 37°C (Di Marzo et al., 2002b). [14C]Ethanolamine produced
from [14C]anandamide hydrolysis was measured by scintillation
counting of the aqueous phase after extraction of the incubation
mixture with 2 volumes of CHCl3/CH3OH (2:1 by volume).
Rat Urinary Bladder Assay
Rats were sacrificed by cervical dislocation, and the urinary blad-
der was removed. Strips of urinary bladders (approximately 1 cm in
length) were suspended in an organ bath with a resting tension of
1 g, bathed (37°C), and aerated (95% O2 and 5% CO2) with Krebs’
solution with the following composition: 119 mM NaCl, 25 mM
NaHCO3, 1.2 mM KH2PO4, 1.5 mM MgSO4, 2.5 mM CaCl2, 4.7 mM
KCl, 11 mM D-glucose, 0.001 mM phosphoramidon, and 0.001 mM
captopril. Tissues were allowed to equilibrate for 60 min prior to the
beginning of each set of experiments (washed every 5 min). In all
experiments, tissues were first contracted with carbachol (1 ␮M),
then after another 60 min, cumulative concentration-response
curves with PhAR (0.1 pM–1 ␮M) were performed. The effects of
PhAR were also tested in the presence of 5Ј-iodoresiniferatoxin (10
nM) or its respective vehicles.
Anandamide Membrane Transporter (AMT) Assays. The ef-
fect of compounds on the uptake of anandamide by RBL-2H3 cells
was studied as described previously (Di Marzo et al., 2002b). Cells
were incubated with [14C]anandamide (4 ␮M) for 5 min at 37°C in
the presence or absence of varying concentrations of the inhibitors.
Residual [14C]anandamide in the incubation media after extraction
with CHCl3/CH3OH (2:1 by volume) was used as a measure of the
anandamide that was taken up by cells. Data are expressed as the
Rat Urinary Incontinence Assay
Female Sprague-Dawley rats (200–225 g) were used in this inves-
tigation. Procedures involving animals and their care were con-
ducted in compliance with local institutional guidelines from the
University of Milan. Animals were anesthetized with ketamine (40
mg/kg) and xylazine (20 mg/kg). The urethra was exposed by a low
midline abdominal incision. According to Steers and De Groat (1988),
a double-silk (4.0) ligature was placed around the urethra and tied in
the presence of an extraluminally placed (1-mm o.d.) polyethylene
tubing which was then removed. Sham-operated rats underwent
urethral manipulation with no urethral ligature. Six to eight weeks
from partial urethral ligature, rats were anesthetized with urethane
concentration exerting 50% inhibition of anandamide uptake (IC50
calculated with GraphPad Software Inc.
)
Assays on Dorsal Root Ganglion Neurons
Newborn rats (2–3 days old) were terminally anesthetized and (1.1 g/kg i.p.), and they underwent anesthetized cystometry. The
decapitated. The dorsal root ganglia (DRG) were removed and rap- bladder was exposed by a low midline abdominal incision, the intra-
idly placed in cold phosphate-buffered solution (PBS) before being vesical urine was removed, and its volume was measured. Then, a
transferred to collagenase/dispase (1 mg mlϪ1 dissolved in Ca2ϩ
-
double-lumen catheter was inserted into the bladder. The outside
Mg2ϩ-free PBS) for 35 min at 37°C. Enrichment of the fraction of catheter (0.8-mm o.d.) was connected to a pressure transducer
nociceptive neurons was obtained following the methods reported (Gould P23 ID), and the intravesical pressure was recorded contin-
previously (Gilabert and McNaughton, 1997). After the enzymatic uously on an ink-write two-channel Gemini7070 (Basile, Comerio,
treatment, ganglia were rinsed three times with Ca2ϩ-Mg2ϩ-free Italy). The inner catheter (0.5-mm o.d.) was connected to a peristaltic

566
Appendino et al.
pump Peri-Star (World Precision Instruments, New Haven, CT), and Epoxydation, cyclopropanation, and hydrogenation of the cis-
saline at 37°C was infused at an infusion rate set to 0.11 ml/min for
control and operated animals. Distal urethra was catheterized with
a polyethylene tube to collect and measure the volume of fluid during
each voiding phase. Bladder outflow obstruction was successfully
produced 6 to 8 weeks from the surgical operation in about 70% of the
animals, and about 40% of these animals also showed detrusor
overactivity. Acute retention was probably the cause of death in 10%
of animals during the first week after operation. The residual 20% of
animals showed abnormal cystometry and were not included in the
experiments. The cystometric parameters evaluated in anesthetized
rats were: resting pressure, pressure threshold, maximal amplitude
of contraction, micturition volume, residual volume, bladder capac-
ity, and frequency of micturition. The effect of PhAR (10–100 nM)
and that of the reference compound resiniferatoxin (10–100 nM) on
cystometric parameters was studied after bladder instillation for 30
min. Both drugs were solubilized in 100% ethanol, and dilutions (10,
50, and 100 nM) were obtained in 1% ethanol. At this concentration,
ethanol did not alter cystometric parameters.
9,10 double bond did not ameliorate the potency and efficacy
of PhAR, and neither did the inversion of its configuration at
the only chiral carbon atom to ent-PhAR (Table 1). None of
the compounds exerted any notable activity in wild-type,
nontransfected HEK-293 cells (data not shown).
Functional Activity of PhAR at Native Rat TRPV1
Channels. PhAR was also extremely potent at the native rat
TRPV1 by elevating Ca2ϩ in neurons from rat dorsal root
ganglia in a way sensitive to the highly selective TRPV1
antagonist 5Ј-iodoresiniferatoxin and with an EC50 ϭ 11.3 Ϯ
0.15 nM (Fig. 3a) similar to that of resiniferatoxin (6.20 Ϯ
0.05 nM) when using a Ca2ϩ imaging assay in intact cultured
neurons. The potency of PhAR and resiniferatoxin in this
assay was lower than that using recombinant human TRPV1
overexpressed in HEK-293 cells, but nevertheless both com-
pounds were extremely powerful in contracting the rat uri-
nary bladder with EC50 ϭ 0.55 Ϯ 0.19 and 0.17 Ϯ 0.02 nM,
respectively (Fig. 3b and data not shown).
Results
Activity of PhAR and Its Analogs on Proteins of the
Endocannabinoid System. N-Acyl-vanyllamides were re-
ported to interact with proteins of the endocannabinoid sys-
tem with low to moderate affinity, for cannabinoid CB1 re-
ceptors, or as inhibitors of the activity of the proteins
mediating endocannabinoid degradation, the fatty acid
amide hydrolase (FAAH), and the putative AMT (Di Marzo et
al., 2002b). Therefore, we decided to test the activity of PhAR
on recombinant human CB1 and CB2 receptors as well as on
Functional Activity of the Novel Compounds at Hu-
man TRPV1 Receptors in HEK-293 Cells. We first as-
sessed the activity at the human TRPV1 of the vanyllamide
of ricinoleic acid by using a typical functional assay of TRPV1
activity, the enhancement of intracellular Ca2ϩ in HEK-293
cells overexpressing the human receptor. Rinvanil, as we
named this new vanyllamide (Fig. 1), exhibited less potency,
albeit higher efficacy, than olvanil (Table 1). The activity of
rinvanil could be modulated by acylation of the hydroxyl
group on the fatty acid moiety. Thus, acetylation afforded a
compound with vanilloid potency comparable with that of
olvanil (Table 1). Acylation of rinvanil with benzoic-, pheny-
lacetic-, and ␤-phenylpropionic acid provided compounds
that exhibited significantly higher potency and efficacy than
olvanil and rinvanil (Figs. 1–2, Table 1). In particular, PhAR
is the first compound structurally unrelated to resinifera-
toxin that shows efficacy comparable with the natural prod-
uct in assays of vanilloid activity with potency, although
[
14C]anandamide cellular uptake in RBL-2H3 cells and hy-
drolysis by rat brain membranes. As shown in Table 1, PhAR
was inactive on FAAH up to 50 ␮M, although it significantly
interacted with human CB2 receptors in a typical binding
assay (Ki ϭ 300 nM) and was weakly active at human CB1
receptors (Ki ϭ 2.2 ␮M) and on the AMT (Ki ϭ 12.2 ␮M). The
other compounds did not exert any strong activity in any of
the assays.
Epoxydation, cyclopropanation, and hydrogenation of the
8-fold lower, in the same picomolar EC50 range (Table 1). cis-9,10 double bond of PhAR significantly decreased its af-
TABLE 1
Summary of the pharmacological effects of the novel compounds developed in this study
Data on the maximal effect at TRPV1 are expressed as percentage of the maximal observable effect induced by 4 ␮M ionomycin. For a comparison, the EC50 of resiniferatoxin
tested side-by-side and under the same conditions was 0.011 Ϯ 0.002 nM, whereas the IC50 (nM) of other TRPV1 antagonists tested side-by-side and under the same
conditions are: 5Ј-I-resiniferatoxin, 0.6 Ϯ 0.2 nM; SB-366791, 32.0 Ϯ 5.1 nM; capsazepine, 50.0 Ϯ 0.6 nM.
Ki
TRPV1
Rat
Human
Percentage of
Maximum
Effect
EC50
(IC50 against capsaicin, 100 nM)
AMT
FAAH
CB1
CB2
␮M
nM
Capsaicin (1a)
Ͼ25
7.3 Ϯ 1.6
Ͼ25
Ͼ50
Ͼ50
Ͼ10
Ͼ10
Ͼ10
Ͼ10
Ͼ10
Ͼ10
40 Ϯ 0.3
0.7 Ϯ 0.05
6.0 Ϯ 0.3
0.50 Ϯ 0.07
0.09 Ϯ 0.02
0.10 Ϯ 0.02
0.13 Ϯ 0.02
0.20 Ϯ 0.02
5.0 Ϯ 0.2
66.2 Ϯ 4.1
67.2 Ϯ 6.2
85.0 Ϯ 4.9
93.5 Ϯ 5.9
98.2 Ϯ 2.1
95.2 Ϯ 2.3
86.1 Ϯ 2.2
98.4 Ϯ 2.7
71.3 Ϯ 3.2
80.1 Ϯ 2.3
72.2 Ϯ 3.5
ND
Olvanil (1b)
Rinvanil (1)
Ͼ50
Ͼ10
Acetylrinvanil (1c)
PhAR (1e)
Ͼ25
Ͼ50
Ͼ10
12.2 Ϯ 2.3
7.4 Ϯ 0.4
10.2 Ϯ 1.4
NT
Ͼ50
2.2 Ϯ 0.3
2.4 Ϯ 0.4
1.3 Ϯ 0.2
NT
0.3 Ϯ 0.02
0.4 Ϯ 0.01
1.4 Ϯ 0.3
NT
6.0 Ϯ 0.8
1.4 Ϯ 0.2
1.1 Ϯ 0.1
2.2 Ϯ 0.2
ent-PhAR (1j)
Ͼ50
Benzoylrinvanil (1d)
Phenylpropionylrinvanil (1f)
Epoxy-PhAR (1g)
Cycloproyl-PhAR (1 h)
Dihydro-PhAR (1i)
6-I-PhAR (3b)
24.3 Ϯ 2.2
Ͼ50
Ͼ25
Ͼ50
9.5 Ϯ 1.1
5.6 Ϯ 0.8
Ͼ5
15.9 Ϯ 3.6
9.4 Ϯ 1.3
17.6 Ϯ 2.6
Ͼ50
0.13 Ϯ 0.02
0.59 Ϯ 0.03
Ͼ10 ␮M
Ͼ50
Ͼ50
2.5 Ϯ 0.4
(6.0 Ϯ 0.5)
Ͼ10 ␮M
(900 Ϯ 97)
5-I-PhAR (3a)
9.1 Ϯ 1.3
Ͼ50
Ͼ5
1.5 Ϯ 0.2
ND
NT, not tested; ND, not determinable; SB-366791, N-(3-methoxyphenyl)-4-chlorocinnamide.

New Ultra-Potent TRPV1 Agonists
567
Fig. 2. Activity of some of the compounds synthe-
sized in this study on human recombinant TRPV1
channels. Activity was measured as the capability
of increasing the intracellular Ca2ϩ concentration
in HEK-293 cells overexpressing the human
TRPV1 and is expressed as percentage of the max-
imal observable effect induced by 4 ␮M ionomycin.
Data are the means of n ϭ 3 separate experiments,
and the standard error bars (never higher than
10% of the means) are not shown for the sake of
clarity. No effect was observed in wild-type HEK-
293 cells. Phpropionylrinvanil, phenylpropionylr-
invanil.
finity for both cannabinoid receptor subtypes (Table 1). In- min blocked micturition reflexes in three of six obstructed
version of configuration to ent-PhAR, however, improved only rats.
the inhibitory activity on AMT and did not modify the affinity
for either cannabinoid receptor subtype (Table 1).
Iodination of PhAR Leads to TRPV1 Antagonists.
Iodination of the 5Ј or 6Ј carbon atoms of the vanillyl moiety
Activity of PhAR in a Rat Model of Urinary Inconti- of PhAR totally abolished its agonist activity at the human
nence. We used anesthetized rats with ligature-intact par- TRPV1 in HEK-293 cells and, in the case of 6Ј-iodo-PhAR, led
tial urethral obstruction as a well established animal model to a potent TRPV1 antagonist with a IC50 ϭ 6 nM, which is
of this disorder. Figure 4a shows a typical cystometrogram 10-fold higher than that of the most potent TRPV1 antago-
obtained in an anesthetized control rat; the infusion rate of nist developed to date, 5Ј-iodoresiniferatoxin (Table 1, Fig.
0.11 ml ⅐ minϪ1 induced a slowly developing increase in 5a). The compound behaved as a noncompetitive antagonist
bladder pressure. Cystometrograms obtained in anesthetized (Kd ϭ 0.54 nM, Schild plot slope ϭ 0.62 Ϯ 0.06) (Fig. 5b). The
rats with partial urethral obstruction (6–8 weeks) showed a 6Ј-iodo-PhAR was also tested on the effect of capsaicin (100
more pronounced spontaneous activity than those obtained nM) on cytoplasmatic Ca2ϩ concentration in cultured rat
in control rats during the filling phase (Fig. 4b). Bladder DRG neurons (Fig. 5c). Similar to the shift observed for the
capacity increased from 0.91 Ϯ 0.05 in controls to 4.45 Ϯ 0.39 agonist (i.e., PhAR), the IC50 for 6Ј-iodo-PhAR against cap-
ml in obstructed rats. Residual volume increased from 0.23 Ϯ saicin (747 nM) was 125-fold higher in rat DRG neurons
0.04 to 3.95 Ϯ 0.48 ml in obstructed rats. Micturition volume compared with transfected HEK-293 cells. 5Ј-Iodoresinifera-
slightly decreased from 0.67 Ϯ 0.05 ml in controls to 0.49 Ϯ toxin retained its potency between the two assays (IC50
,
0.07 ml in obstructed rats, and frequency of micturition in- 0.87 Ϯ 0.01 nM in DRG neurons), whereas capsazepine was
creased from 15.3 Ϯ 0.68 to 30.7 Ϯ 3.23 hϪ1 in obstructed 3-fold less potent than 6Ј-iodo-PhAR (Appendino et al., 2003).
rats. Bladder instillation with either PhAR (10–50 and 100
nM) or resiniferatoxin (10–50 nM) inhibited detrusor over-
activity, restored the resting pressure, slightly increased the
Discussion
micturition volume compared with obstructed rats, and sig-
We have used a medicinal chemistry approach to generate
nificantly decreased the frequency of micturition (Fig. 4, novel and ultra-potent TRPV1 agonists suitable for use in
c–e). Resiniferatoxin (100 nM) instilled in the bladder for 30 vitro and in vivo and with a high potential for therapeutic use
Fig. 3. Effects of PhAR (IDN5890, E) on cytoplasmatic Ca2ϩ concentration ([Ca2ϩ]i) in cultured rat DRG neurons (a) and isolated strips of rat urinary
bladder (b). In both models, the TRPV1 antagonist, 5Ј-iodoresiniferatoxin (I-RTX) (10 nM, F), significantly reduced the effect of PhAR demonstrating
its selective agonistic activity at the capsaicin receptor. PhAR-induced contractions and [Ca2ϩ]i mobilization were expressed as percentage of the
maximal effect elicited by 100 nM resiniferatoxin. Data are means Ϯ S.E.M. of at least six experiments.

568
Appendino et al.
Fig. 5. Antagonistic activity of iodo-phenylacetylrinvanil (I-PhAR) iso-
mers on human recombinant and rat native TRPV1 channels. a, effect of
increasing concentrations of 5Ј- and 6Ј-I-PhAR on the action of capsaicin
(100 nM) on the intracellular Ca2ϩ concentration in HEK-293 cells over-
expressing the human TRPV1. b, effect of different doses of 6Ј- I-PhAR on
the dose-response curve of capsaicin. Data are expressed as percentage of
the maximal observable effect induced by 4 ␮M ionomycin and are means
of n ϭ 3 separate experiments. Standard error bars (never higher than
10% of the means) are not shown for the sake of clarity. c, effects of
increasing doses 6Ј-iodo-phenylacetylrinvanil on the effect of capsaicin
(100 nM) on cytoplasmatic Ca2ϩ concentration ([Ca2ϩ]i) in cultured rat
DRG neurons. Capsaicin-induced [Ca2ϩ]i mobilization was measured as
detailed under Materials and Methods. Data are expressed as the net
increase in the F340/F380 ratio and are means of n ϭ 5 separate experi-
ments The IC50 of 6Ј-I-PhAR is shown.
Fig. 4. Original tracing showing a cystometry from control rat (a), and
from rat with partial urethral obstruction (8 weeks) (b). Six to eight
weeks after surgical operation, rat and bladder mean weights were 295 Ϯ
13 versus 293 Ϯ 8 g and 102 Ϯ 9 versus 462 Ϯ 93 mg in sham-operated
(control) and in partial obstructed rats, respectively. In rats with outflow
obstruction, spontaneous bladder contractions (detrusor overactivity)
were often observed during the filling phase (infusion rate of 0.11 ml
minϪ1). c, d, and e, collected data obtained in anesthetized controls (n ϭ
10), obstructed (n ϭ 10), obstructed after bladder instillation of PhAR
(IDN5890, 50 nM) (n ϭ 8), or resiniferatoxin (RTX) (50 nM) (n ϭ 8). The
two drugs were also tested at 10 and 100 nM and found not to be
significantly different from “obstructed” and as active as the 50 nM
concentration, respectively (not shown). Statistical significance (one-way
ANOVA with Tukey post hoc test, p Ͻ 0.05): ء, versus control; #, versus
obstructed. RP, resting pressure; PT, pressure threshold; MAC, maximal
amplitude of contraction; MV, micturition volume; RV, residual volume;
BC, bladder capacity.
against urinary incontinence. Furthermore, we have used
one of these agonists to generate a very potent antagonist by
simply iodinating its vanillyl moiety. Olvanil (N-oleoylvanil-
lamine), a validated TRPV1 agonist devoid of pungency,
served as the starting structural scaffold, and an intracellu-

New Ultra-Potent TRPV1 Agonists
569
lar Ca2ϩ assay widely employed to determine TRPV1 activity icant affinity for the human CB2 receptor. Given the analge-
was used as the initial endpoint. sic and anti-inflammatory activity of both CB2 (Malan et al.,
By introducing in the acyl chain of olvanil a hydroxy group 2003) and TRPV1 agonists (see Introduction), this finding
highly suitable to further derivatization and starting from provides a strong rationale to conduct future studies that
observation that resiniferatoxin shows an aromatic moiety on assess the functional activity of PhAR at CB2 receptors and
its lipophilic diterpenoid core (Fig. 1), we first synthesized its effects on inflammatory pain in vivo.
rinvanil and then derivatized it with acids containing aryl
Since iodination of the vanillyl moiety of capsaicin and
moieties. Using this strategy, we developed PhAR, the most resiniferatoxin causes a dramatic switch in vanilloid func-
potent capsaicinoid TRPV1 agonist ever reported, and the tional activity (Wahl et al., 2001; Appendino et al., 2003), we
first compound structurally unrelated to resiniferatoxin that also investigated the effect of iodination on PhAR. As ex-
shows efficacy comparable with the natural product in assays pected, this modification totally abolished PhAR agonist ac-
of vanilloid activity with potency, although 8-fold lower, in tivity and, in the case of the 6Ј-iodo-PhAR, led to a TRPV1
the same two-digit picomolar EC50 range. The ultra-potency antagonist exhibiting a potency in vitro comparable with that
of PhAR was only decreased by further chemical modifica- of some other recently developed TRPV1 antagonists (Appen-
tions of its fatty acid chain but was conserved in assays of dino et al., 2003; Rami et al., 2004; Szallasi and Appendino,
TRPV1 activity carried out with the native rat receptor. In 2004, for review). The activity in vivo of this compound will
fact, the very low EC50 observed for PhAR in the TRPV1- need to be assessed before suggesting its possible therapeutic
mediated contraction of the rat urinary bladder suggested its value.
testing in an animal model of urinary incontinence.
In conclusion, by using a stepwise medicinal chemistry
Detrusor overactivity, which is characterized by involun- approach, we have developed a novel ultra-potent TRPV1
tary bladder contractions and loss of urine, is a major cause agonist that, by reducing the C-fiber input, appears to sig-
of urinary incontinence, a common health problem for both nificantly reduce idiopathic detrusor overactivity associated
genders. Anticholinergic drugs are still considered as the with bladder incontinence. This compound and its analogs,
first-line therapy for treating overactive bladder, although by lacking the potential tumorigenicity of resiniferatoxin, are
their clinical use is limited by their well known side effects. likely to represent a safer alternative to “resiniferoids” to
Inhibiting afferent C-fiber pathway from the bladder by in- treat this disorder. Finally, we have described two more
travesical administration of vanilloid compounds such as potential applications of the further chemical modification of
resiniferatoxin or capsaicin is an attractive option and might PhAR, consisting of the possible future development of either
represent a possible alternative to the anticholinergic drugs. novel cannabinoid CB2 receptor ligands or potent TRPV1
The effects of these two compounds has already been inves- antagonists. Further studies will be required to assess the
tigated on patients with spinal cord injures (Giannantoni et efficacy in the clinic of PhAR and then the potential to de-
al., 2002) and also on patients with idiopathic detrusor over- velop novel analgesics and anti-inflammatory agents from
activity (Silva et al., 2002). In these patients, detrusor over- PhAR-derived TRPV1 antagonists and/or CB2 agonists.
activity was abolished or reduced for 2 to 3 months from
bladder instillation, revealing that C-fibers seem to have an
important role in the generation of this bladder dysfunction.
Acknowledgments
We are grateful to M. G. Cascio and P. Cavallo for valuable
Here, we used anesthetized rats with ligature-intact partial
assistance.
urethral obstruction as a well established animal model of
this disorder. Our results show that this procedure, in agree-
ment with published results (Steers and De Groat, 1988),
leads to a significant increase in bladder weight, capacity,
frequency of micturition, and to the development of a detru-
sor overactivity in the majority of the animals. These effects
were similar to those observed in humans (Groutz et al.,
2000). In our experimental conditions, a bladder instillation
of PhAR was able to abolish spontaneous bladder contrac-
tions in anesthetized rats, and similarly, to resiniferatoxin, it
was able to reduce frequency of micturition. Both compounds
exhibited a very long duration of action, more than 4 to 6 h
(not shown). This is actually an advantage compared with the
short duration of action (less than 1 h) of other compounds
such as the ATP-sensitive potassium channel openers (Fabiyi
et al., 2003), which are claimed to be an alternative to the
anticholinergic drugs.
Since olvanil and other N-acyl-vanyllamides have been
reported to interact with proteins of the endocannabinoid
system, including the cannabinoid CB1 receptor, FAAH and
the putative AMT (Di Marzo et al., 2002b), we decided to test
the activity of PhAR and its analogs on these proteins. How-
ever, in most cases, the compounds were weakly active or
inactive in these assays with the notable exception of PhAR
which, unlike other N-acyl-vanyllamides, exhibited a signif-
References
Appendino G, Harrison S, De Petrocellis L, Daddario N, Bianchi F, Schiano Moriello
A, Trevisani M, Benvenuti F, Geppetti P, and Di Marzo V (2003) Halogenation of
a capsaicin analogue leads to novel vanilloid TRPV1 receptor antagonists. Br J
Pharmacol 139:1417–1424.
Bevan S, Hothi S, Hughes G, James IF, Rang HP, Shah K, Walpole CS, and Yeats JC
(1992) Capsazepine: a competitive antagonist of the sensory neurone excitant
capsaicin. Br J Pharmacol 107:544–552.
Birder LA, Nakamura Y, Kiss S, Nealen ML, Barrick S, Kanai AJ, Wang E, Ruiz G,
De Groat WC, Apodaca G, et al. (2002) Altered urinary bladder function in mice
lacking the vanilloid receptor TRPV1. Nat Neurosci 5:856–860.
Brand L, Berman E, Schwen R, Loomans M, Janusz J, Bohne R, Maddin C, Gardner
J, Lahann T, Farmer R, et al. (1987) NE-19550: a novel, orally active anti-
inflammatory analgesic. Drugs Exp Clin Res 13:259–265.
Caterina MJ, Leffler A, Malmberg AB, Martin WJ, Trafton J, Petersen-Zeitz KR,
Koltzenburg M, Basbaum AI, and Julius D (2000) Impaired nociception and pain
sensation in mice lacking the capsaicin receptor. Science (Wash DC) 288:306–313.
Chan CL, Facer P, Davis JB, Smith GD, Egerton J, Bountra C, Williams NS, and
Anand P (2003) Sensory fibres expressing capsaicin receptor TRPV1 in patients
with rectal hypersensitivity and faecal urgency. Lancet 361:385–391.
Chung KF and Chang AB (2002) Therapy for cough: active agents. Pulm Pharmacol
Ther 15:335–338.
Contassot E, Tenan M, Schnuriger V, Pelte MF, and Dietrich PY (2004) Arachidonyl
ethanolamide induces apoptosis of uterine cervix cancer cells via aberrantly ex-
pressed vanilloid receptor-1. Gynecol Oncol 93:182–188.
Davis JB, Gray J, Gunthorpe MJ, Hatcher JP, Davey PT, Overend P, Harries MH,
Latcham J, Clapham C, Atkinson K, et al. (2000) Vanilloid receptor-1 is essential
for inflammatory thermal hyperalgesia. Nature (Lond) 405:183–187.
Di Marzo V, Blumberg PM, and Szallasi A (2002a) Endovanilloid signaling in pain.
Curr Opin Neurobiol 12:372–379.
Di Marzo V, Griffin G, De Petrocellis L, Brandi I, Bisogno T, Williams W, Grier MC,
Kulasegram S, Mahadevan A, Razdan RK, et al. (2002b) A structure/activity
relationship study on arvanil, an endocannabinoid and vanilloid hybrid. J Phar-
macol Exp Ther 300:984–991.

570
Appendino et al.
Doly S, Fischer J, Salio C, and Conrath M (2004) The vanilloid receptor-1 is ex-
pressed in rat spinal dorsal horn astrocytes. Neurosci Lett 357:123–126.
Fabiyi AC, Gopalakrishnan M, Lynch JJ 3rd, Brioni JD, Coghlan MJ, and Brune ME
(2003) In vivo evaluation of the potency and bladder-vascular selectivity of the
ATP-sensitive potassium channel openers (Ϫ)-cromakalim, ZD6169 and WAY-
133537 in rats. BJU Int 91:284–290.
Giannantoni A, Di Stasi SM, Stephen RL, Navarra P, Scivoletto G, Mearini E, and
Porena M (2002) Intravesical capsaicin versus resiniferatoxin in patients with
detrusor hyperreflexia: a prospective randomized study. J Urol 167:1710–1714.
Gilabert R and McNaughton P (1997) Enrichment of the fraction of nociceptive
neurones in cultures of primary sensory neurones. J Neurosci Methods 71:191–
198.
Groutz A, Blaivas JG, and Chaikin DC (2000) Bladder outlet obstruction in women:
definition and characteristics. Neurourol Urodyn 19:213–220.
Gunthorpe MJ, Benham CD, Randall A, and Davis JB (2002) The diversity in the
vanilloid (TRPV) receptor family of ion channels. Trends Pharmacol Sci 23:183–
191.
receptor subtype 1 in cutaneous sensory nerve fibers, mast cells and epithelial cells
of appendage structures. Exp Dermatol 13:129–139.
Steers WD and De Groat WC (1988) Effect of bladder outlet obstruction on micturi-
tion reflex pathways in the rat. J Urol 140:864–871.
Sterner O and Szallasi A (1999) Novel natural vanilloid receptor agonists: new
therapeutic targets for drug development. Trends Pharmacol Sci 20:459–465.
Szallasi A (2002) Vanilloid (capsaicin) receptors in health and disease. Am J Clin
Pathol 118:110–121.
Szallasi A and Appendino G (2004) Vanilloid receptor TRPV1 antagonists as the next
generation of painkillers. Are we putting the cart before the horse? J Med Chem
47:2717–2723.
Szallasi A and Fowler CJ (2002) After a decade of intravesical vanilloid therapy: still
more questions than answers. Lancet Neurol 1:167–172.
Tognetto M, Amadesi S, Harrison S, Creminon C, Trevisani M, Carreras M, Matera
M, Geppetti P, and Bianchi A (2001) Anandamide excites central terminals of
dorsal root ganglion neurons via vanilloid receptor-1 activation. J Neurosci 21:
1104–1109.
Tympanidis P, Casula MA, Yiangou Y, Terenghi G, Dowd P, and Anand P (2004)
Increased vanilloid receptor VR1 innervation in vulvodynia. Eur J Pain 8:129–
133.
Inoue K, Koizumi S, Fuziwara S, Denda S, Inoue K, and Denda M (2002) Functional
vanilloid receptors in cultured normal human epidermal keratinocytes. Biochem
Biophys Res Commun 291:124–129.
Karai L, Brown DC, Mannes AJ, Connelly ST, Brown J, Gandal M, Wellisch OM,
Neubert JK, Olah Z, and Iadarola MJ (2004) Deletion of vanilloid receptor 1-ex-
pressing primary afferent neurons for pain control. J Clin Investig 113:1344–1352.
Luo H, Cheng J, Han JS, and Wan Y (2004) Change of vanilloid receptor 1 expression
in dorsal root ganglion and spinal dorsal horn during inflammatory nociception
induced by complete Freund’s adjuvant in rats. Neuroreport 15:655–658.
Malan TP Jr, Ibrahim MM, Lai J, Vanderah TW, Makriyannis A, and Porreca F
(2003) CB2 cannabinoid receptor agonists: pain relief without psychoactive effects?
Curr Opin Pharmacol 3:62–67.
Mezey E, Toth ZE, Cortright DN, Arzubi MK, Krause JE, Elde R, Guo A, Blumberg
PM, and Szallasi A (2000) Distribution of mRNA for vanilloid receptor subtype 1
(VR1) and VR1-like immunoreactivity, in the central nervous system of the rat and
human. Proc Natl Acad Sci USA 97:3655–3660.
Nilius B, Vriens J, Prenen J, Droogmans G, and Voets T (2004) TRPV4 calcium entry
channel: a paradigm for gating diversity. Am J Physiol Cell Physiol 286:C195–
C205.
Rami HK, Thompson M, Wyman P, Jerman JC, Egerton J, Brough S, Stevens AJ,
Randall AD, Smart D, Gunthorpe MJ, et al. (2004) Discovery of small molecule
antagonists of TRPV1. Bioorg Med Chem Lett 14:3631–3634.
Veldhuis WB, van der Stelt M, Wadman MW, van Zadelhoff G, Maccarrone M, Fezza
F, Veldink GA, Vliegenthart JF, Bar PR, Nicolay K, et al. (2003) Neuroprotection
by the endogenous cannabinoid anandamide and arvanil against in vivo excitotox-
icity in the rat: role of vanilloid receptors and lipoxygenases. J Neurosci 23:4127–
4133.
Wahl P, Foged C, Tullin S, and Thomsen C (2001) Iodo-resiniferatoxin, a new potent
vanilloid receptor antagonist. Mol Pharmacol 59:9–15.
Walker KM, Urban L, Medhurst SJ, Patel S, Panesar M, Fox AJ, and McIntyre P
(2003) The VR1 antagonist capsazepine reverses mechanical hyperalgesia in mod-
els of inflammatory and neuropathic pain. J Pharmacol Exp Ther 304:56–62.
Witte DG, Cassar SC, Masters JN, Esbenshade T, and Hancock AA (2002) Use of a
fluorescent imaging plate reader—based calcium assay to assess pharmacological
differences between the human and rat vanilloid receptor. J Biomol Screen 7:466–
475.
Yamaji K, Sarker KP, Kawahara K, Iino S, Yamakuchi M, Abeyama K, Hashiguchi
T, and Maruyama I (2003) Anandamide induces apoptosis in human endothelial
cells: its regulation system and clinical implications. Thromb Haemostasis 89:875–
884.
Yiangou Y, Facer P, Dyer NH, Chan CL, Knowles C, Williams NS, and Anand P
(2001) Vanilloid receptor 1 immunoreactivity in inflamed human bowel. Lancet
357:1338–1339.
Rashid MH, Inoue M, Bakoshi S, and Ueda H (2003) Increased expression of va-
nilloid receptor 1 on myelinated primary afferent neurons contributes to the
antihyperalgesic effect of capsaicin cream in diabetic neuropathic pain in mice.
J Pharmacol Exp Ther 306:709–717.
Silva C, Ribeiro MJ, and Cruz F (2002) The effect of intravesical resiniferatoxin in
patients with idiopathic detrusor instability suggests that involuntary detrusor
contractions are triggered by C-fiber input. J Urol 168:575–579.
Stander S, Moormann C, Schumacher M, Buddenkotte J, Artuc M, Shpacovitch V,
Brzoska T, Lippert U, Henz BM, Luger TA, et al. (2004) Expression of vanilloid
Address correspondence to: Vincenzo Di Marzo, Endocannabinoid Re-
search Group, Institute of Biomolecular Chemistry, Consiglio Nazionale delle
Ricerche, Via Campi Flegrei 34, Comprensorio Olivetti, Bldg. 70, 80078, Poz-
zuoli (NA), Italy. E-mail: vdimarzo@icmib.na.cnr.it