Chemical modification of the naphthoyl 3-position of JWH-015: In search of a fluorescent probe to the cannabinoid CB2 receptor

Article abstract of DOI:10.1016/j.bmcl.2005.05.049

In silico modelling was used to guide the positioning of the fluorescent dye NBD-F on the cannabinoid CB₂ receptor agonist JWH-015. While the ultimate fluorescent conjugate lost extensive binding affinity to the cannabinoid CB₂ receptor, affinity and efficacy studies on the naphthoyl 3-position modified precursor molecules have provided new insight into structure-activity relationships associated with this position.

Full text of DOI:10.1016/j.bmcl.2005.05.049

Bioorganic & Medicinal Chemistry Letters 15 (2005) 3758–3762
Chemical modification of the naphthoyl 3-position of JWH-015:
In search of a fluorescent probe to the cannabinoid CB2 receptor
Andrew S. Yates,a Stephen W. Doughty,a David A. Kendallb and Barrie Kellama,*
aSchool of Pharmacy, Centre for Biomolecular Sciences, University of Nottingham, University Park, Nottingham NG7 2RD, UK
bSchool of Biomedical Sciences, University of Nottingham Medical School, Nottingham NG7 2UH, UK
Received 21 January 2005; revised 12 May 2005; accepted 17 May 2005
Available online 29 June 2005
Abstract—In silico modelling was used to guide the positioning of the fluorescent dye NBD-F on the cannabinoid CB2 receptor
agonist JWH-015. While the ultimate fluorescent conjugate lost extensive binding affinity to the cannabinoid CB2 receptor, affinity
and efficacy studies on the naphthoyl 3-position modified precursor molecules have provided new insight into structure–activity rela-
tionships associated with this position.
ꢀ 2005 Elsevier Ltd. All rights reserved.
Investigation of receptor–ligand interactions is one of
the principal tools employed in pharmacological re-
search. Conventional methods, for example, the use of
radioactively labelled ligands, still contribute a major as-
pect of this activity. However, these approaches are
being increasingly replaced by fluorescence-based tech-
niques,1 allowing researchers to develop and exploit as-
says at the single cell (confocal microscopy)2 and single
molecule (fluorescence correlation spectroscopy, FCS)
levels.3 To fully exploit this technology, there remains
a key requirement for the development of bioactive fluo-
rescent probes for the receptor of interest. As such, we
have recently demonstrated that fluorescently labelled
small molecule ligands for the G-protein-coupled hu-
man adenosine A1-receptor can be used to study their
binding to the receptor in single living cells.4 In
continuation of our studies, we were eager to explore
the potential to develop fluorescent molecules for
non-biogenic amine-activated GPCRs and we selected
the human cannabinoid CB2 receptor as our target.
is mainly located within the CNS and is responsible
for many mind-altering effects of cannabis use.7 hCB2
receptor systems have been shown, either by agonist-in-
duced attenuation or by receptor up-regulation, to be
important mediators in numerous diseases, including
multiple sclerosis,8 malignant disease9 and neuropathic
pain.10 These diseases are not only severely debilitating
but also offer limited pharmacotherapeutic choices
available for treatment.
A high affinity fluorescent hCB2 receptor ligand would
allow a greater understanding of the hCB2 receptorÕs
pharmacology and its role in disease states through the
use of modern fluorescence-based techniques that have
been previously highlighted. Our efforts were initially
concentrated towards conjugating the selective hCB2
receptor agonist JWH-015 1 with a fluorescent dye.
JWH-015 1 was chosen due to its nanomole affinity to
hCB2 receptors, high hCB2/hCB1 selectivity, extensive
SAR and rapid synthetic accessibility.11 The latter factor
was key to us rejecting synthetically complex com-
pounds, such as JWH-133, L-759633 and HU-308, as
chemical starting points, despite their displaying a better
CB2 selectivity. In contrast to our approach taken dur-
ing the development of fluorescent adenosine receptor li-
gands,4 we were also keen to explore whether in silico
modelling of JWH-015 1 bound within the hCB2 recep-
tor, coupled with de novo drug design, would aid in the
identification of a suitable position upon 1 that could
tolerate the incorporation of a bulky fluorescent dye.
Therefore, alongside the key advantages listed above,
JWH-015 1 possessed a number of benefits with regard
The human cannabinoid CB2 receptor (hCB2), first dis-
covered in 1993,5 has been implicated as a potential tar-
get in a number of diseases.6 The hCB2 receptor is found
predominantly on the immune cells of the periphery.
The other known class of cannabinoid receptors (CB1)
Keywords: Cannabinoid receptor; GPCR; Homology model; Fluores-
cent ligand.
*
Corresponding author. Tel.: +44 115 9513026; fax: +44 115
9513412; e-mail: barrie.kellam@nottingham.ac.uk
0960-894X/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.05.049

A. S. Yates et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3758–3762
3759
to the specific modelling strategy we planned to imple-
ment. More specifically, JWH-015 1 displays a signifi-
cant amount of structural homology to Win 55,212-2
2, a molecule which has been extensively studied by
molecular modelling and mutational analysis tech-
niques.12 We therefore envisaged docking Win 55,212-
2 2 into an homology model of the hCB2 receptor using
previously published data concerning Win 55,212-2 2
binding. Then, the structural similarities between
JWH-015 1 and Win 55,212-2 2 could be utilised to
superimpose common structural motifs (i.e., the naph-
thoyl rings) of the two molecules, to generate a rational
model where JWH-015 is bound to the hCB2 receptor.
agonist.12 The receptor–ligand complex was energy min-
imized and subjected to a 100 ps molecular dynamics
simulation using the CHARMM force field.14 Our mod-
el demonstrated similar receptor–ligand aromatic stack-
ing interactions observed in previous hCB2 models and
indicated that the aromatic amino acids Phe 197, Phe
117 and Trp 194 interacted with Win 55,212-2 2, as pre-
viously reported by Song et al.12 Structural alignment,
by superimposing the naphthoyl moieties of both
JWH-015 1 and Win 55,212-2 2, deletion of the latter
and further simulation produced a ligand–receptor mod-
el where JWH-015 1 was located within the binding
pocket (Fig. 2). Growing points for the addition of sub-
stituents onto JWH-015 1 were determined (Fig. 1),
based on previous SAR,11b,c,d and the molecule was
subjected to de novo design using the program
LigBuilder.15
An in silico model of the hCB2 receptor was therefore
˚
constructed by homology modelling to the 2.8 A X-ray
crystal structure of bovine rhodopsin.13 The 3-naph-
thoylindole cannabinoid agonist Win 55,212-2
2
(Fig. 1) was docked into the model utilising the previ-
ously published data to guide the positioning of the
The results of this exercise indicated a number of posi-
tions upon JWH-015 1 that could potentially tolerate a
bulky substituent, especially the 3-position of the naph-
thyl ring substituent. When we examined the structures
generated by LigBuilder, one molecule that scored
highly in its predicted hCB2 receptor binding affinity
possessed a substituent that closely resembled a known
fluorescent dye (i.e., 7-nitrobenzoxadiazole, NBD). This
molecule 3 was therefore chosen as our lead compound
(Fig. 3).
Synthesis of the fluorescent congener was accomplished
using modifications to a procedure by Bell et al.16 The 2-
methyl-1-n-propylindole derivative 5 was prepared by
N-1 alkylation of commercially available 2-methylindole
4 using standard methodology (Scheme 1).
Meanwhile, commercially available 3-nitro-1,8-naphtha-
lic anhydride 6 was selectively decarboxylated, via the
Figure 1. Development of the fluorescent hCB2 agonist. + Indicates a
grow point used in LigBuilder.
Figure 2. Molecular model of hCB2. Inset shows detail of JWH-015 in the active site together with key interacting aromatic amino acids.

3760
A. S. Yates et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3758–3762
to keep the overall molecular size as close to that of
the lead molecule 3 as possible and because this type
of linker had been previously shown to completely ne-
gate the PET effect.19 Therefore, N-Boc-protected gly-
cine was coupled to 9 using standard carbodiimide
chemistry. Acidolysis of the Boc group of 10 exposed
the primary amine which, when coupled to the amine
reactive NBD fluoride, gave the fluorescent congener
12.20
Figure 3. The lead molecule generated by the LigBuilder programme.
The hCB2 receptor affinity for compounds 8, 9, 11 and
12 was measured by a competitive displacement assay
against [3H]CP55, 940 (a high affinity cannabinoid ago-
nist) on membranes prepared from CHO cells expressing
the hCB2 receptor.21 The results of this assay are given
in Table 1.
mercuric salt of the di-acid, to give the desired 3-nitro-
naphthoic acid.17 The meta-substitution of this com-
1
1
pound was confirmed by mp, H and H–13C COSY
NMR spectroscopy.18 The 3-nitronaphthoic acid was
then converted to its corresponding acid chloride 7 using
SOCl2. Lewis acid (AlCl3) promoted electrophilic substi-
tution at the C-3 position of 4, affording the nitro deriv-
ative of JWH-015 8. Functional group transformation
of the nitro group to the aryl amine 9 took place using
catalytic hydrogenation. While the two low yielding
reactions to give the acid chloride 7 (22%) and the 3-
aroylindole 8 (19%) are unoptimised, both are compara-
ble with previous literature preparations (see Refs. 17b
and 11, respectively).
Table 1. Results of in vitro assays using the hCB2 receptor
a
Compound Ki CB2 (nM)
% [35S]GTP-c-S
binding from basal
at 1 lMc
number
JWH-015
1
36 (31–42) (lit.b 14
5)
142 ( 19)*
134 ( 8)*
151 ( 22)*
Not tested
8
143 (56–364)
191 (59–622)
420 (359–492)
9
11
12
25% displacement at 10 lM Not tested
It is recognised that only non-fluorescent derivatives are
formed when NBD fluoride is reacted with aromatic
amines, due to a photoinduced electron transfer (PET)
mechanism.19 It is therefore necessary to insert a short
functionalised linker between the aryl amine and the
NBD fluorochrome. A short amide linker was chosen
a Results in parentheses are 95% confidence intervals for three indi-
vidual experiments run in triplicate.
b Result from Ref. 11c.
c Results in parentheses are the standard error of the mean for three
individual experiments run in triplicate.
* P < 0.05 compared to basal binding (100%).
Scheme 1. Regents: (a) 1-bromopropane, NaH, DMF (97%); (b) (i) NaOH, water, HgO. (ii) SOCl2 (22%); (c) 5, DCM, AlCl3 (19%); (d) Pd/C, EtOH,
H2 (66%); (e) Boc-glycine, DCC, DMAP, DCM (67%); (f) MeOH, AcCl (50%); (g) NBD-F, MeCN, NaHCO3 buffer 8.3 (46%).

A. S. Yates et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3758–3762
3761
The results show that the naphthyl 3-position of JHW-
015 1 is tolerant to limited chemical modification. As
the steric bulk at this position increases in this series
of molecules, affinity towards the hCB2 receptor de-
creases accordingly. Loss of affinity is greatest (>250-
fold) for the bulky NBD fluorescent conjugate 12, which
showed a limited displacement of [3H]CP55, 940 from
the receptor at 10 lM. The two compounds, which
retained the greatest affinity for the hCB2 receptor (com-
pounds 8 and 9), were subsequently tested in a
[35S]GTP-c-S-binding assay to assess if they retained
the agonist profile of the original lead molecule
JWH-015 1. The results in Table 1 show that com-
pounds 1, 8 and 9 were able to significantly increase
[35S]GTP-c-S binding above basal levels when tested at
a concentration of 1000 nM, indicating agonist proper-
ties of all three compounds. None of the three com-
pounds were able to significantly raise [35S]GTP-c-S
binding above the basal level when used at 10 nM con-
centration. ANOVA was unable to detect any significant
statistical difference between 1, 8 and 9 at a concentra-
tion of 1000 nM.
the hydrophobic binding pocket, which was to accom-
modate the NBD moiety, may not be as accessible (or
as large) as originally predicted by our homology model.
An alternative hypothesis could argue that the addi-
tional linker atoms produced potential conformations
of the final molecule that create unfavourable steric
clashes with the receptor structure.
However, it was rewarding to observe that the non-fluo-
rescent 3-substituted naphthoyl precursor molecules 8, 9
and 11 did show that affinity to the hCB2 receptor could
still be retained as long as the 3-substituent was kept
small. In addition, modifications using nitro 8 and ami-
no 9 groups led to compounds that demonstrated ago-
nist behaviour when used at 1000 nM. To the best of
our knowledge, this is the first reported modification
to the 3-naphthoyl position of JWH-015 and is therefore
an important addition to the established SAR of the in-
dole-based cannabinoid ligands. It is also interesting to
note that the recently reported indole-based hCB2 ago-
nist AM124110b,23 has a 5-iodo-3-nitro benzoyl substitu-
ent at the 3-position of an aminoalkylindole, which adds
to the suggestion that the 3-aryl position within the 3-
arylindoles may be an important extra binding pocket
within the hCB2 receptor.
Notwithstanding its low affinity, the fluorescent conju-
gate 12 was assessed in confocal microscopy experiments
to observe if the compound exhibited any discernable
membrane binding to CHO cells expressing the hCB2
receptor.21 When conjugate 12 (100 nM) was applied
to live CHO cells expressing the hCB2 receptor in
Hepes-buffered saline at 22 ꢁC, a slight membrane bind-
ing was observed, with the majority of fluorescence
occurring from cytosolic accumulation, presumably as
a consequence of the high lipophilicity of the compound
and therefore rapid cellular uptake.
Acknowledgments
We wish to thank Tim Self and Diane Hall for their kind
help in the fluorescence confocal imaging and cell cul-
ture, respectively, and the BBSRC (A.S.Y.) for the
funding.
In conclusion, in silico studies have suggested that the 3-
naphthoyl position of the hCB2 receptor agonist JWH-
015 1 could potentially tolerate conjugation to a bulky
fluorescent dye. Photophysical requirements dictated
the insertion of an additional 3-atom spacer between
JWH-015 and NBD to ensure that the ultimate molecule
would indeed be fluorescent. It has been previously
established that non-fluorescent species are formed
when NBD is conjugated directly to an aromatic amine.
The mechanism of this fluorescence quenching has been
investigated and attributed to Photoinduced Electron
Transfer (PET).19,22 Importantly, it was observed that
an amide group placed between the aromatic amine
and NBD prevented PET from occurring, even when
the chain length was reduced to an acetyl (two carbon)
spacer.22 If a simple alkyl linker was used, the chain
length required to prevent PET was found to be longer
(nine carbon atoms). We therefore decided to install a
short amine functionalised acetyl linker to separate the
aromatic amine and NBD moieties and to keep the over-
all size of the molecule as close to 3 as possible. Chem-
istry, that incorporated substitution at the 3-position,
was utilised to afford the NBD conjugate 12 in 7 steps.
Unfortunately, this compound showed a >250-fold loss
in affinity to the hCB2 receptor and could not be used
successfully in its intended application of fluorescence-
based pharmacology assays. The decreased binding
affinity of the glycyl derivative 11 may indicate that
References and notes
1. Daly, C. J.; McGrath, J. C. Pharmacol. Ther. 2003, 100,
101.
2. (a) Mackenzie, J. F.; Daly, C. J.; Pediani, J. D.; McGrath,
J. C. J. Pharm. Exp. Ther. 2000, 294, 434; (b) Madsen, B.
W.; Beglan, C. L.; Spivak, C. E. J. Neurosci. Meth. 2000,
97, 123.
3. (a) Wohland, T.; Friedrich, K.; Hovius, R.; Vogel, H.
Biochemistry 1999, 38, 8671; (b) Pick, H.; Preuss, A. K.;
Mayer, M.; Wohland, T.; Hovius, R.; Vogel, H. Biochem-
istry 2003, 42, 877.
4. (a) Briddon, S. J.; Cordeaux, Y.; Middleton, R. J.;
Weinstein, J.; George, M. W.; Kellam, B.; Hill, S. J. Proc.
Natl. Acad. Sci. U.S.A. 2004, 101, 4673; (b) Briddon, S. J.;
Middleton, R. J.; Yates, A. S.; George, M. W.; Kellam, B.;
Hill, S. J. Faraday Discuss. 2004, 126, 197.
5. Munro, S.; Thomas, K.; Abu-Shaar, M. Nature 1993, 365,
61.
6. (a) Molina-Holgado, E.; Guaza, C.; Borrell, J.; Molina-
Holgado, F. Biodrugs 1999, 12, 317; (b) Goutopoulos, A.;
Makriyannis, M. Pharmacol. Ther. 2002, 95, 103.
7. (a) Pertwee, R. G. Pharmacol. Ther. 1997, 74, 129; (b)
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.
8. (a) Arevalo-Martin, A.; Vela, J.; Molina-Holgado, E.;
Borrell, J.; Guaza, C. J. Neurosci. 2003, 23, 2511; (b)
Croxford, J.; Miller, S. J. Clin. Invest. 2003, 111, 1231; (c)

3762
A. S. Yates et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3758–3762
Baker, D.; Pryce, G.; Croxford, J. L.; Brown, P.; Pertwee,
R. G.; Huffman, J. W.; Layward, L. Nature 2000, 404, 84.
16. Bell, M. R.; DÕAmbra, T. E.; Kumar, V.; Eissenstat, J. L.
H.; Wetzel, J. R.; Rosi, D.; Philion, R. E.; Daum, S. J.;
Hlasta, D. J.; Kullnig, R. K.; Ackerman, J. H.; Haubrich,
D. R.; Luttinger, D. A.; Baizman, E. R.; Miller, M. S.;
Ward, S. J. J. Med. Chem. 1991, 34, 1099.
17. (a) Leuck, G. J.; Perkins, P. R.; Whitmore, F. C. J. Chem.
Soc. 1929, 51, 1832; (b) Duffy, K. J.; Darcy, M. G.;
Delorme, E.; Dillon, S. B.; Eppley, D. F.; Erickson-Miller,
C.; Giampa, L.; Hopson, C. B.; Huang, Y.; Keenan, R.
M.; Lamb, P.; Leong, L.; Liu, N.; Miller, S. G.; Price, A.
T.; Rosen, J.; Shah, R.; Shaw, T. N.; Smith, H.; Stark, K.
C.; Tian, S.; Tyree, C.; Wiggal, K. J.; Zhang, L.; Luengo,
J. I. J. Med. Chem. 2001, 44, 3730.
9. (a) McKallip, R.; Lombard, C.; Fisher, M.; Martin, B.;
Ryu, S.; Grant, S.; Nagarkatti, P.; Nagarkatti, M. Blood
2002, 100, 627; (b) Sanchez, C.; Ceballos, M.; Gomez del
Pulgar, T.; Rueda, D.; Corbacho, C.; Velasco, G.; Galve-
Roperh, I.; Huffman, J.; Ramon y Cajal, S.; Guzman, M.
Cancer Res. 2001, 61, 5784.
10. (a) Zhang, J.; Hoffert, C.; Vu, H.; Groblewski, T.; Armad,
S.; OÕDonnell, D. Eur. J. Neurosci. 2003, 17, 2750; (b)
Ibrahim, M.; Deng, H.; Zvonok, A.; Cockayne, D.; Kwan,
J.; Mata, H.; Vanderah, T.; Lai, J.; Porreca, F.; Makri-
yannis, A.; Malan, T. Proc. Natl. Acad. Sci. U.S.A. 2003,
100, 10529.
11. (a) Huffman, J. W.; Dai, D.; Martin, B. R.; Compton, D.
R. Bioorg. Med. Chem. Lett. 1994, 4, 563; (b) Huffman, J.
W. Curr. Med. Chem. 1999, 6, 705; (c) Aung, M. M.;
Griffin, G.; Huffman, J. W.; Wu, M. J.; Keel, C.; Yang, B.;
Showalter, V. M.; Abood, M. E.; Martin, B. R. Drug
Alcohol Depend. 2000, 60, 133; (d) Huffman, J. W. Curr.
Pharm. Des. 2000, 6, 1323.
12. (a) Song, Z. H.; Slowey, C. A.; Hurst, D. P.; Reggio, P. H.
Mol. Pharmacol. 1999, 56, 834; (b) Gouldson, P.; Calan-
dra, B.; Legoux, P.; Kerneis, A.; Rinaldi-Carmona, M.;
Barth, F.; Fur, G.; Ferrara, P.; Shire, D. Eur. J.
Pharmacol. 2000, 401, 17; (c) Chin, C. H.; Murphy, J.
W.; Huffman, J. W.; Kendall, D. A. J. Pharmacol. Exp.
Ther. 1999, 291, 837; (d) Shire, D.; Calandra, B.; Bouab-
oula, M.; Barth, F.; Rinaldi-Carmona, M.; Casellas, P.;
Ferrara, P. Life Sci. 1999, 65, 627; (e) Reggio, P. H.; Basu-
Dutt, S.; Barnett-Norris, J.; Castro, M. T.; Hurst, D. P.;
Seltzman, H. H.; Roche, M. J.; Gilliam, A. F.; Thomas, B.
F.; Stevenson, L. A.; Pertwee, R.; Abood, M. E. J. Med.
Chem. 1998, 41, 5177; (f) Xie, X. Q.; Han, X. H.; Chen, J.
Z.; Eissenstat, M. A.; Makriyannis, A. J. Med. Chem.
1999, 42, 4021; (g) Huffman, J. W.; Mabon, R.; Wu, M. J.;
Lu, J.; Hart, R.; Hurst, D. P.; Reggio, P. H.; Wiley, J. L.;
Martin, B. R. Bioorg. Med. Chem. 2003, 11, 539.
13. (a) Palczewski, K.; Kumasaka, T.; Hori, T.; Behnke, C.;
Motoshima, H.; Fox, B.; Trong, I.; Teller, D.; Okada, T.;
Stenkamp, R.; Yamamototo, M.; Miyano, M. Science
2000, 289, 739; (b) Sakmar, T. Curr. Opin. Cell Biol. 2002,
14, 189.
18. mp 250–252 ꢁC (lit.17b mp 255–257 ꢁC). 1H NMR
(DMSO-d6) d 13.89 (1H, br s, COOH), 9.22 (1H, d, J
2.4 Hz, 4-H), 8.93 (1H, d, J 8.6 Hz, 5-H), 8.75 (1H, d, J
2.4 Hz, 2-H), 8.43 (1H, d, J 8.6 Hz, 8-H), 7.97–7.89 (2H,
m, aromatic). 13C NMR (DMSO-d6) d 166.1 (4ꢁ, C@O
acid), 144.0 (4ꢁ), 135.1 (4ꢁ), 133.6 (4ꢁ), 132.8 (4ꢁ), 131.5
(CH), 130.6 (CH), 128.8 (CH), 128.3 (CH), 126.3 (CH),
123.2 (CH). m/z (ESMS-) 217.0 (Mꢀ). Calcd for
C11H7NO4 217.04 (Mꢀ). vmax (KBr) cmꢀ1: 3500 (OAH),
3068, 2641, 1701 (C@O), 1597, 1529, 1453, 1412, 1336,
1287, 1253, 1203, 1102, 915, 796, 760, 687, 500.
19. Onoda, M.; Uchiyama, S.; Santa, T.; Imai, K. Lumines-
cence 2002, 17, 11.
20. 12 1H NMR (DMSO-d6) d 10.54 (1H, br s), 9.45 (1H, br s),
8.55 (1H, d, J 8.9 Hz) 8.41 (1H, d, J 2.0 Hz) 7.95 (1H, d, J
8.2 Hz), 7.79 (1H, d, J 8.3 Hz), 7.66 (1H, d, J 2.0 Hz),
7.60–7.48 (2H, m), 7.41–7.32 (1H, m), 7.20–7.10 (1H, m),
7.01–6.92 (2H, m), 6.47 (1H, m), 4.45 (2H, br s, amide-
CH2), 4.21 (2H, t, J 7.3 Hz, indole N-CH2), 2.45 (3H, s,
CH3-indole). 1.81 (2H, sextet, J 7.2 Hz, CH2-propyl), 0.91
(3H, t, J 7.3 Hz, CH3-propyl). m/z (ESMSꢀ) 561.1 (Mꢀ),
Calcd for C31H26N6O5 562.2; (TOFESꢀ) 562.1962 (Mꢀ),
Calcd for C31H26N6O5 562.1965 (M). Analytical RP-
HPLC (Vydac reverse phase C8 column (150 · 4.6 mm),
flow rate of 1 mL/min and UV detection at 254 nm);
gradient 35% ! 100% MeCN(aq) over 30 min. One major
peak at Rt = 13.64 min.
21. We gratefully acknowledge the kind donation of this cell
line by GlaxoSmithKline, UK.
22. Uchiyama, S.; Santa, T.; Imai, K. Analyst 2000, 125, 1839.
23. (a) Malan, T. P., Jr.; Ibrrahim, M. M.; Deng, H.; Liu, Q.;
Mata, H. P.; Vanderah, T.; Porreca, F.; Makriyannis, A.
Pain 2001, 93, 239; (b) Nackley, A. G.; Makriyannis, A.;
Hohmann, A. G. Neuroscience 2003, 119, 747.
14. Brooks, B. R.; Bruccoleri, R. E.; Olafson, B. D.; States, D.
J.; Swaminathan, S.; Karplus, M. J. Comp. Chem. 1983, 4,
187.
15. Wang, R.; Gao, Y.; Lai, L. J. Mol. Model. 2000, 6, 498.