Discovery of selective nonpeptidergic neuropeptide FF2 receptor agonists

Article abstract of DOI:10.1021/jm9011998

We report the discovery and initial characterization of a novel class of selective NPFF2 agonists. HTS screening using R-SAT, a whole cell based functional assay, identified a class of aryliminoguanidines as NPFF1 and NPFF2 ligands. Subsequent optimizatio

Full text of DOI:10.1021/jm9011998

J. Med. Chem. 2009, 52, 6511–6514 6511
DOI: 10.1021/jm9011998
Chart 1
Discovery of Selective Nonpeptidergic
Neuropeptide FF2 Receptor Agonists
Gilles Gaubert,† Fabio Bertozzi,† Nicholas M. Kelly,†
Jan Pawlas,† Audra L. Scully,‡ Norman R. Nash,‡
Luis R. Gardell,‡ Jelveh Lameh,‡ and Roger Olsson*,†
†ACADIA Pharmaceuticals AB, Medeon Science Park, S-205 12
Malmo, Sweden, and ‡ACADIA Pharmaceuticals Inc.,
3911 Sorrento Valley Boulevard, San Diego, California 92121
Received August 11, 2009
Abstract: We report the discovery and initial characterization of
a novel class of selective NPFF2 agonists. HTS screening using
R-SAT, a whole cell based functional assay, identified a class of
aryliminoguanidines as NPFF1 and NPFF2 ligands. Subsequent
optimization led to molecules exhibiting selective NPFF2 agonistic
activity. Systemic administration showed that selective NPFF2
agonists (1 and 3) are active in various pain models in vivo, whereas
administration of a nonselective NPFF1 and NPFF2 agonist (9)
increases sensitivity to noxious and non-noxious stimuli.
published a year later, thiazoleguanidine derivatives were
described as NPFF1 antagonists, but no data regarding their
NPFF2 activity were provided (Chart 1).6
The octapeptide neuropeptide FF (NPFF,a FLFQPQRF-
NH2) is generally considered to be a pain modulating pep-
tide,1 but it also displays potent effects on the cardiovascular
system.2 Its effects on nociception are believed to be mediated
in the spinal cord and brain, while its cardiovascular effects
may be centrally or peripherally mediated.2 These distinct
actions are believed to be mediated via interaction with two
specific G-protein-coupled receptors (GPCRs), NPFF1 and
NPFF2, which are present in the central nervous system
(CNS) and peripherally.3 In the CNS, NPFF2s have been
found in the most external layers of the dorsal horn of the
spinal cord, the parafascicular thalamic nucleus, laterodorsal
thalamic nucleus, and presubiculum of the hippocampus,
whereas NPFF1s have been detected in septal, thalamic,
and hypothalamic areas but not in the spinal cord.4 Hence,
identifying selective agonists and antagonists of NPFF1 and
NPFF2 would not only be useful inelucidating these receptors
specific physiological roles but also provide leads for ther-
apeutically interesting molecules. In addition, nonpeptidergic,
metabolically stable ligands would be preferable as tool
compounds, compared topeptides based on the natural ligand
NPFF, because they would not be subject to peptidolytic
degradation, which could have been the cause of the transient
effects previously observed of NPFF.2
Here, we report the discovery of selective NPFF2 agonists
and initial characterization of their structure-activity rela-
tionships (SARs) and in vivo effects in pain models.
A chemical library containing 250 000 small druglike mo-
lecules was screened for agonist activity using receptor selec-
tion and amplification technology (R-SAT) assays. R-SAT is
a functional cell-based assay that allows one to monitor
receptor-dependent proliferative responses of various recep-
tor classes, including GPCRs and nuclear receptors.7
Aryliminoguanidines (compounds 1-11, Chart 1) were
prepared in moderate to excellent yields (48-95%) by react-
ing the corresponding aldehydes or ketones with aminogua-
nidine hydrochloride in ethanol under microwave irradiation
in a sealed vial (130 °C, 12 min, pressure increased up to 4 bar)
or conventional heating (70-90 °C, 18 h) (Scheme 1). With
conventional heating, the reactions involving ketones re-
quired slightly higher temperatures (90 °C, 18 h). Commer-
cially available aldehydes and ketones were used in the
reactions except for aldehydes 12-15, required for the pre-
paration of 7-11 (Scheme 2). Aryliodide 16 was synthesized
following an aromatic Finkelstein methodology as reported
by Klapars et al.8 from the bromide analogue. A combination
of copper iodide, sodium iodide, and ligand L1 was used in the
reaction, which afforded an excellent yield of 95%. Aldehydes
12-15 were prepared using a modified Heck cross-coupling
procedure based largely on the method published by McClure
et al.9 with different substrates. In our hands, for our
specific substrates, the air stable ligand L2 (i.e., cyclohexyl
JohnPhos)10 proved to be a superior catalyst than the oxida-
tion-prone tricyclohexylphosphine used in the original report.
The yields of the coupling reactions were moderate
(45-55%). Notably, aldehyde 15 could not be purified com-
pletely (it was contaminated with approximately 5% of the
starting aromatic iodide) and was used in the last step without
further purification.
To date, small nonpeptidergic ligands for NPFF1 and
NPFF2 have been primarily described in patents. Notably,
substituted quinolino- or quinazolinoguanidines that interact
agonistically or antagonistically with both receptors or antag-
onistically with NPFF1 and agonistically with NPFF2 were
reported in a 2003 patent (Chart 1).5 In another patent,
*To whom correspondence should be addressed. Phone:
þ46406013400. Fax: þ46406013401. E-mail: roger@acadia-pharm.com.
a Abbreviations: CNS, central nervous system; GPCR, G-protein-
coupled receptor; ip, intraperitoneal; MPE, maximal possible effect;
NPFF, neuropeptide FF; NPFF1, neuropeptide FF1 receptor; NPFF2,
neuropeptide FF2 receptor; R-SAT, receptor selection and amplifica-
tion technology; SAR, structure-activity relationship; SNL, spinal
nerve ligation.
Compounds 1-11 area representativeset of the mostactive
compounds selected from approximately 500 compounds
synthesized throughout the project. These compounds were
r
2009 American Chemical Society
Published on Web 10/05/2009
pubs.acs.org/jmc

6512 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21
Gaubert et al.
Scheme 1a
Scheme 2a
a Reaction conditions: microwave irradiation in a sealed vial, 130 °C,
12 min; or conventional heating, 70-90 °C, 18 h.
first tested in functional R-SAT assays, for in vitro agonist
and antagonist activities. Dose-response curves were gener-
ated by testing the compounds in seven concentrations, each
in triplicate, at human NPFF1 and NPFF2 receptors
(Table 1). To simplify the SAR characterization, the arylimi-
noguanidines were divided into two subclasses; substituted
phenyliminoguanidines (1-6) and 5-aryl substituted-five-
membered heteroaromatic iminoguanidines (7-11). In the
first subclass (1-6) the active and NPFF2 selective com-
pounds bore inductive electron withdrawing substituents in
the 3 and 4 positions such as chloro, bromo, or trifluorometh-
yl groups. If there was no substitutents or if alkyl, aryl, amino,
or ether substituents were present at these positions, the
compounds displayed low partial agonistic activity, a 10-fold
lower potency, and lower selectivity toward the NPFF2
receptor. Moreover, the presence of additional substituents
at the 2-position, the 6-position, or both was generally detri-
mental for NPFF2 agonistic activity. For example, guanabenz
(Chart 2), a drug used for treating hypertension, showed no
activity at NPFF2 or NPFF1. Guanabenz is described as an
R2-selective adrenergic agonist; therefore, 1 was tested at all
the R-adrenergic receptors, but no crossover activity was
observed. Dichloro compound 1 was also N-methylated at
the iminoguanidine in all permutations, but all the resulting
analogues showed little or no activity. In addition, the 5-HT4
partial agonist tegaserod (Chart 2), used for treating irritable
bowel syndrome, was tested at NPFF2 and was found to be
inactive (Table 1). Compounds 4 and 6 (acetophenone de-
rivatives) showed similar potency to the compounds originat-
ing from aldehydes 1 and 3 but with slightly less efficacy,
especially 3 compared to 6. However, compounds originating
from the aldehyde 1, acetophenone 4, and benzophenone 5
showed similar activities at NPFF2, but the phenyl in 5
increased the activity at NPFF1, resulting in 5 being equipo-
tent at NPFF2 and NPFF1. Members of the second subclass
of the aryliminoguanidines (7-11) are 5-aryl substituted-five-
membered heteroaromatic iminoguanidines. In general, these
compounds exhibited improved NPFF2 agonist properties,
but at the cost of selectivity toward NPFF1 (in accordance
with the effects of the phenyl group introduced in the benzo-
phenone 5 of the first subclass). Compounds 9-11 showed
that similar molecules such as furan, thiophene, or N-methyl-
pyrrole derivatives had similar activities in R-SAT. Moreover,
only compounds with ortho-substitution on the phenyl moi-
ety were the most active compounds (e.g., 5-phenylfuranyli-
minoguanididine the nonsubstituted analogue of 7 showed
>10-fold lower potency), and the most active compounds 8
and 9 have an isopropyl or trifluoromethyl substituent,
respectively. These observations imply that the restriction in
the conformation could be beneficial for NPFF1 and NPFF2
agonistic activity.
a Reaction conditions: (a) PdCl2, L2, Bu4NBr, AcOK, DMF, 110 °C,
10 h, 45-55% yield of 12-15; (b) CuI, L1, NaI, dioxane, 110 °C, 20 h.
Table 1. In Vitro Activities of NPFF2 Agonists Using R-SATa
NPFF2
NPFF1
% efficacy
compd
pEC50
% efficacy
pEC50
nd
1
2
3
4
5
6
7
8
9
6.0 ( 0.2
5.9 ( 0.1
6.3 ( 0.2
6.3 ( 0.1
6.1 ( 0.1
6.4 ( 0.2
7.0 ( 0.2
6.7 ( 0.1
7.0 ( 0.4
6.8 ( 0.1
6.9 ( 0.1
68 ( 10
90 ( 15
114 ( 28
64 ( 4
63 ( 7
66 ( 6
90 ( 6
70 ( 7
73 ( 13
55 ( 6
72 ( 7
na
18 ( 7
na
<5.5
46 ( 0
na
6.1 ( 0.1
46 ( 7
na
7.1 ( 0.2
6.8 ( 0.1
7.4 ( 0.2
6.9 ( 0.1
6.5 ( 0.1
55 ( 6
62 ( 7
55 ( 6
44 ( 6
36 ( 7
na
10
11
guanabenz
tegaserod
na
na
a nd, not determined. na, not active at 10 mM. R-SAT was performed
as described in ref 7. Values represent the mean ( SEM of at least three
independent experiments.
showed full agonistic activity at human NPFF2 and minimal
activity at NPFF1. Compound 1 displayed partial efficacy at
NPFF2 and was inactive at NPFF1. Compound 1 was further
tested as an antagonist at NPFF1 and was found to be
inactive. Biaryl 9 demonstrated full efficacy at NPFF2 and
partial efficacy at NPFF1 (Table 1). These results were
confirmed in a secondary assay, cAMP inhibition. The selec-
tive NPFF2 agonist 3 showed full efficacy (89%) at NPFF2
On the basis of the initial R-SAT data, three compounds
(1, 3, and 9) were selected for further evaluation. Compound 3

Letter
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21 6513
described method,12 and a significant reduction in the paw
withdrawal threshold was interpreted as the presence of tactile
allodynia. The NPFF2 full agonist 3 completely attenuated
SNL-induced hypersensitivity at 30mg/kg, whereas treatment
with the partial agonist 1 yielded an inverted U-shaped
dose-response curve, displaying a maximum effect at
10 mg/kg. However, the nonselective NPFF1 and NPFF2
agonist 9 dose dependently potentiated the SNL-induced
hypersensitivity in rats (Figure 2). Furthermore, 9 exerted
tactile allodynia in sham-operated rats, an effect that was
completely blocked by the selective NPFF1 antagonist dan-
sylPQRamide (30 mg/kg, ip administration).
The stability in human and rat liver microsomes was
analyzed for 1. The compound showed moderate to low
clearance (less than 30% liver blood flow) in human and rat
Figure 1. Selective NPFF2 agonists 1 and 3 attenuate carrageenan-
induced thermal hypersensitivity, % MPE (% maximum possible
effect), ip administration.
microsomes, 26 and 18 μL/(min mg), respectively, indicating
3
that 1 is a stable compound. The in vitro data were corrobo-
rated by the result from a pharmacokinetic study, in which F
= 43% and T1/2 = 2.5 h in rat were obtained for 1. Moreover,
experimentally physicochemical properties of 1 (pKa = 8.4,
log D = 1.8, and log P = 2.8) all appeared to be suitable for a
potential drug.
In conclusion, we have described the discovery and char-
acterization of the first small molecule nonpeptidic selective
NPFF2 agonists. Compounds 1 and 3 showed dosedependent
attenuation of hypersensitivity in acute inflammatory and
neuropathic pain rat models. In contrast, the NPFF1 and
NPFF2 agonist 9 dose-dependently potentiated SNL-induced
hypersensitivity and induced tactile allodynia in sham-oper-
ated rats. The identification of potent, selective, and metabo-
lically stable small-molecule NPFF2 agonists will provide
useful pharmacological tools for elucidating the complex
physiological functions of the NPFF receptors.
Figure 2. Selective NPFF2 agonists 1 and 3 attenuate and non-
selective NPFF2 agonist 9 produces SNL-induced mechanical
hypersensitivity.
Acknowledgment. We thank May-Britt Nielsen Kary for
providing microsomal data, Paula Jacobi and William Keefe
for assistance with developing NPFF1 and NPFF2 RSAT
assays, and Abhisheck Bajpai and Derek Nguyen for RSAT
data.
Chart 2
Supporting Information Available: Experimental details and
characterization data for all compounds. This material is avail-
able free of charge via the Internet at http://pubs.acs.org.
and only weak agonist activity (42% at the highest dose) at
NPFF1. The nonselective 9 was fully efficacious at both
NPFF receptors in cAMP assays, showing 114% and 93%
efficacy at NPFF2 and NPFF1, respectively.
The selective NPFF2 agonists 1 and 3 were tested in a rat
model of acute inflammatory pain, and both compounds
displayed dose dependent inhibition of carrageenan induced
thermal hypersensitivity measured using a metal plate main-
tained at 52 °C in “hot plate test” (Figure 1). The result from
the acute inflammatory pain model corroborated the in vitro
data; hence, the partial agonist 1 displayed a partial inhibition
whereas the full agonist 3 showed maximum response
(Figure 1).
Three compounds (1, 3, and 9) were tested in a rat model of
neuropathic pain developed by Kim and Chung11 (Figure 2).
This model requires the ligation of the L5 and L6 spinal nerves
distal to the dorsal root ganglion. At 10-14 days following
spinal nerve ligation (SNL) surgery, the rats were reassessed
for their response thresholds to mechanical stimuli with a
series of calibrated von Frey filaments. Paw withdrawal thres-
holds to probing were determined according to a previously
References
(1) (a) Panula, P.; Kalso, E.; Nieminen, M.-L.; Kontinen, V.; Brendt,
A.; Pertovaara, A. Neuropeptide FF and modulation of pain. Brain
Res. 1999, 848, 191–196. (b) Roumy, M.; Zajac, J.-M. Neuropeptide
FF. Pain and analgesia. Eur. J. Pharmacol. 1998, 345, 1–11.
(2) (a) Allard, M.; Labroucha, S.; Nosjean, A.; Laguzzi, R. Mecha-
nism underlying the cardiovascular response to peripheral admin-
istration of NPFF in the rat. J. Pharmacol. Exp. Ther. 1995, 274,
577–583. (b) Nijsen, M. J. M. A.; de Ruiter, G. J. W.; Kasbergen, C. M.;
Hoogerhout, P.; de Wildt, D. J. Relevance of the C-terminal Arg Phe
sequence in g2-melanocyte-stimulating hormone (g2-MSH) for indu-
cing cardiovascular effects in conscious rats. Br. J. Pharmacol. 2000,
131, 1468–1474. (c) Roth, B.; Disimone, J.; Majane, E. A.; Yang, H. Y.
T. Elevation of arterial-pressure in rats by 2 new vertebrate peptides
FLFQPQRF-NH2 and AGEGLSSPFWSLAAPQRF-NH2 which are
immunoreactive to FMRF-NH2 antiserum. Neuropeptides 1987, 10,
37–42. (d) Thiemermann, C.; Al-Damluji, S.; Hecker, M. V. FMRF-
amide and L-Arg-L-Phe increase blood pressure and heart rate in the
anesthetized rat by central stimulation of the sympathetic nervous
system. Biochem. Biophys. Res. Commun. 1991, 175, 318–324.
(3) (a) Elshourbagy, N. A.; Ames, R. S.; Fitzgerald, L. R.; Foley, J. J.;
Chambers, J. K.; Szekeres, P. G.; Evans, N. A.; Schmidt, D. B.;
Buckley, P. T.; Dytko, G. M.; Murdock, P. R.; Milligan, G.;
Groarke, D. A.; Tan, K. B.; Shabon, U.; Nuthulaganti, P.; Wang,
D. Y.; Wilson, S.; Bergsma, D. J.; Sarau, H. M. Receptor for the

6514 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 21
Gaubert et al.
pain modulatory. neuropeptides FF and AF is an orphan G
protein-coupled receptor. J. Biol. Chem. 2000, 275, 25965–25971.
(b) Kotani, M.; Mollereau, C.; Detheux, M.; Le Poul, E.; Brezillon, S.;
Vakili, J.; Mazarguil, H. A.; Vassart, G.; Zajac, J.-M.; Parmentier, M.
Functional characterization of a human receptor for neuropeptide FF
and related peptides. Br. J. Pharmacol. 2001, 133, 138–144.
5490–5493. (d) Lund, B. W.; Eeg, A.; Piu, F.; Gauthier, N. K.; Begtrup,
M.; Hacksell, U.; Olsson, R. Design, synthesis and structure-activity
analysis of isoform-selective retinoic acid β ligands. J. Med. Chem.
2009, 52, 1540–1545. (e) Pettersson, H.; Bulow, A.; Ek, F.; Jensen, J.;
Ottesen, L. K.; Fejzic, A.; Ma, J.-N.; Del Tredeci, A. L.; Currier, E. A.;
Gardell, L. R.; Tabatabaei, A.; Craig, D.; McFarland, K.; Ott, T. R.; Piu,
F.; Burstein, E. S.; Olsson, R. Synthesis and evaluation of dibenzothia-
zepines: a novel class of selective cannabinoid-1 receptor inverse
agonists. J. Med. Chem. 2009, 52, 1975–1982.
(4) Gouarderes, C.; Quelven, I.; Mollereau, C.; Mazarguil, H.; Rice,
S. Q. J.; Zajac, J. M. Quantitative autoradiographic distribution of
NPFF1 neuropeptide FF receptor in the rat brain and comparison
with NPFF2 receptor by using [125I]YVP and [125I]EYF as selective
radioligands. Neuroscience 2002, 115, 349–361.
(8) Klapars, A.; Buchwald, S. L. Copper-catalyzed halogen exchange
in aryl halides: an aromatic Finkelstein reaction. J. Am. Chem. Soc.
2002, 124, 14844–14845.
(5) Kawakami, J. K.; Konkel, M. J.; Boteju, L. W.; Wetzel, J. M.;
Noble, S. A.; Wan, H. Patent WO 03026667A1, 2003.
(9) McClure, M. S.; Glover, B.; McSorley, E.; Millar, A.; Osterhout,
M. H.; Roschangar, F. Regioselective palladium-catalyzed aryla-
tion of 2-furaldehyde. Org. Lett. 2001, 3, 1677–1680.
(10) Old, D. W.; Wolfe, J. P.; Buchwald, S. L. A highly active catalyst
for palladium-catalyzed cross-coupling reactions: room-tempera-
ture Suzuki couplings and amination of unactivated aryl chlorides.
J. Am. Chem. Soc. 1998, 120, 9722–9723.
(11) Kim, S. H.; Chung, J. M. An experimental model for peripheral
neuropathy produced by segmental spinal nerve ligation in the rat.
Pain 1992, 50, 355–363.
(12) Chaplan, S. R.; Bach, F. W.; Pogrel, J. W.; Chung, J. M.; Yaksh, T.
L. Quantitative assessment of tactile allodynia in the rat paw.
J. Neurosci. Methods 1994, 53, 55–63.
(6) Caroff, E.; Steger, M.; Valdenaire, O.; Fecher, A.; Breu, V.;
Hilpert, K.; Fretz, H.; Giller, T. Patent WO 04083218A1, 2004.
(7) (a) Lund, B. W.; Piu, F.; Gauthier, N. K.; Eeg, A.; Currier, E.;
Sherbukhin, V.; Brann, M. R.; Hacksell, U.; Olsson, R. Discovery
of a potent, orally available, and isoform-selective retinoic acid β2
receptor agonist. J. Med. Chem. 2005, 48, 7517–7519. (b) Piu, F.;
Gauthier, N. K.; Olsson, R.; Currier, E. A.; Lund, B. W.; Croston, G. E.;
Hacksell, U.; Brann, M. R. Identification of novel subtype selective
RAR agonists. Biochem. Pharmacol. 2005, 71, 156–162. (c) Seitzberg,
J. G.; Knapp, A. E.; Lund, B. W.; Bertozzi, S. M.; Currier, E. A.; Ma, J.-
N.; Sherbukhin, V.; Burstein, E. S.; Olsson, R. Discovery of potent and
selective small-molecule PAR-2 agonists. J. Med. Chem. 2008, 51,