The discovery and synthesis of JNJ 31020028, a small molecule antagonist of the Neuropeptide y Y2 receptor

Article abstract of DOI:10.1016/j.bmcl.2011.06.136

A series of small molecules based on a chemotype identified from our compound collection were synthesized and tested for binding affinity (IC ₅₀) at the human Neuropeptide Y Y₂ receptor (NPY Y ₂). Six of the 23 analogs tes

Full text of DOI:10.1016/j.bmcl.2011.06.136

Bioorganic & Medicinal Chemistry Letters 21 (2011) 5552–5556
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
The discovery and synthesis of JNJ 31020028, a small molecule antagonist
of the Neuropeptide Y Y2 receptor
⇑
Devin M. Swanson , Victoria D. Wong, Jill A. Jablonowski, Chandra Shah, Dale A. Rudolph, Curt A. Dvorak,
Mark Seierstad, Lisa K. Dvorak, Kirsten Morton, Diane Nepomuceno, John R. Atack, Pascal Bonaventure,
Timothy W. Lovenberg, Nicholas I. Carruthers
Johnson & Johnson Pharmaceutical Research & Development, L.L.C., 3210 Merryfield Row, San Diego, CA 92121, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of small molecules based on a chemotype identified from our compound collection were synthe-
sized and tested for binding affinity (IC50) at the human Neuropeptide Y Y2 receptor (NPY Y2). Six of the
23 analogs tested possessed an NPY Y2 IC50 6 15 nM. One member of this series, JNJ 31020028, is a selec-
tive, high affinity, receptor antagonist existing as a racemic mixture. As such a synthetic route to the
desired enantiomer was designed starting from commercially available (S)-(+)-mandelic acid.
Ó 2011 Elsevier Ltd. All rights reserved.
Received 15 April 2011
Revised 17 June 2011
Accepted 20 June 2011
Available online 18 July 2011
Keywords:
Neuropeptide Y Y2 antagonist
Neuropeptide Y Y2 receptor
Neuropeptide Y
Neuropeptide Y (NPY) consists of 36 amino acid residues and is
the most abundant peptide in the brain. Discovered in 1982,1 it is
widely distributed in the central and peripheral nervous system
and is highly conserved across species.2 The high abundance and
localization of NPY suggests a role in a variety of physiological pro-
cesses including food intake, anxiety, hormone release, and
memory.3
The NPY Y2 receptor is one of five identified G- protein coupled
receptors that bare its name (Y1, Y2, Y4, Y5, and y6). Unlike other
NPY receptors, Y2 is a pre- and postsynaptic receptor that regulates
the synthesis and release of NPY.4 Thus antagonism of the Y2 recep-
tor is thought to elevate levels of NPY in the synapse and be ben-
eficial for a variety of CNS disorders. Consistent with this, Y2
knockout mice exhibit reduced anxiety-like behavior compared
with wild type.5 In addition, the Y2 receptor has attracted interest
as a drug target for its role in obesity,6 bone formation,7 and
alcohol intake.8
A number of small molecule NPY Y2 antagonists have now been
reported in the literature. Bristol–Meyers Squib (1),9 GlaxoSmithK-
line (2),10 The Scripps Research Institute (3),11 and our own labora-
tory (4, 5)12–14 have all described potent, selective, and functional
antagonists at the receptor (Fig. 1). These compounds show im-
proved brain penetration, solubility, and synthetic accessibility
over the NPY peptidomimetic, BIIE3-0246,15 which while a potent
Y2 receptor ligand suffers from a complex structure and high
molecular weight, limiting its therapeutic potential.
We previously described the discovery, synthesis, in vivo and
in vitro characterization of the small molecule Y2 antagonist, JNJ
5207787 (4).12,13 This was followed by the in vivo and in vitro char-
acterization of the potent and selective brain penetrant, JNJ
31020028 (5).14 Herein we describe the synthesis, SAR, and evolu-
tion of the chemotype that led to JNJ 31002028 as well as a chiral
synthesis of the molecule starting from commercially available
(S)-(+)-mandelic acid.
To identify new Y2 chemotypes a pharmacophore directed
screen of our corporate compound collection was initiated. Thus
pharmacophore models were constructed using a set of known
Y2 ligands. These models included select small molecules and Neu-
ropeptide Y analogs.16 Compounds from our corporate collection
were subsequently identified via pharmacaphore based screening.
One series that emerged from these efforts is represented by com-
pound 6 (Fig. 2). This compound was originally prepared as part of
our microsomal triglyceride transfer protein (MTP) program and is
Abbreviations: DCM, dichloromethane; HOBt, 1-hydroxy-benzotriazole; EDCI,
1-ethyl-3-(3-dimethylaminopropyl)carbodiimide; BOC, butoxycarbonyl; MeOH,
methanol; K2CO3, potassium carbonate; Na2CO3, sodium carbonate; DMF,
dimethylformamide; Et3N, triethylamine; Et2O, ether; HCl, hydrochloride; NBS, N-
bromosuccinimide; AIBN, azobisisobutyronitirile; CCl4, carbon tetrachloride; DCC,
N,N0-dicyclohexylcarbodiimide; EtOAc, ethyl acetate; MsCl, mesyl chloride;
NaHCO3, sodium bicarbonate; NaOH, sodium hydroxide; Na2SO4, sodium sulfate;
TFA, trifluoroacetic acid; THF, tetrahydrofuran; DIPEA, diisopropylethylamine.
⇑
Corresponding author. Tel.: +1 858 320 3306; fax: +1 858 784 3063.
E-mail address: dswanso1@its.jnj.com (D.M. Swanson).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.06.136

D. M. Swanson et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5552–5556
5553
O
S
H
O
N
Cl
N
S
N
N
H
N
N
HO
N
S
NMe2
Ph
O
O
1
2
3
BMS ( )
2 IC50 = 450 nM
GSK ( )
2 Ki = 25 nM
SF-11 ( )
Y2 IC50 = 200 nM
Y
Y
H
F
N
N
N
O
N
N
N
N
N
O
O
O
CN
4
5
JNJ 5207787 ( )
Y2 IC50 = 100 nM
JNJ 31020028 ( )
Y2 IC50 = 6 nM
Figure 1. Previously reported small molecule Y2 antagonists.
With our chemotype validated, we chose to focus our chemistry
efforts on three regions of the molecule: the methyl ester, the cen-
tral phenyl core, and the biaryl amide. These regions were chosen
for their synthetic accessibility and potential issues associated
with the existing functionality. For example, the methyl ester is a
likely metabolic liability, the biaryl amide a hindrance to solubility,
and the 4-pyridine a potential cytochrome P450 (CYP) inhibitor.
To achieve variability at the three regions of the molecule a gen-
eral synthesis was designed (Scheme 1). Starting from 1-fluoro-4-
nitrobenzenes of interest (7) we installed N-BOC piperidine with
an SNAr reaction using DMF and base to provide intermediate 8.
Depending on the substitution at the 2-position of the nitroben-
zene (where X = H, F, CN, or Me) this reaction may require elevated
temperatures. For example, the more electron withdrawing 1,2-di-
fluoro-4-nitrobenzene proceeded readily at room temperature
H
N
O
N
N
O
N
screening lead (6)
Y2 IC50 = 240 nM
O
Figure 2. Neuropeptide Y Y2 screening lead.
comprised of a piperidine-phenyl core flanked by a phenyl glycine
and biaryl amide.17,18 With an NPY Y2 IC50 <500 nM this series of-
fered a promising start for medicinal chemistry efforts.
H
X
N
NO2
b, c or d
H
X
N
N
R1
X
N
N
R1
X
F
NO2
a
e, f
O
O
O
N
O
N
N
X = H, F, CN, or Me
O
O
7
8
9
12
O
R2
e, f
H
N
X
N
NO2
X
N
R1
b, c or d
O
N
N
O
R2
O
R2
11
12
Scheme 1. Reagents and conditions: general synthesis. (a) N-BOC piperidine, K2CO3, DMF, rt to 60 °C, 2 to 6 h; (b) Pd/C, H2, EtOH, 24 h; (c) R1COOH, EDCI, HOBt, DCM, rt, 16 h;
(d) R1COCl, Et3N, DCM, 16 h; (e) 2 N HCl in Et2O, MeOH, 16 h; (f) PhCHBrCOR2 (10), Na2CO3, DMF, rt, 6 h.

5554
D. M. Swanson et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5552–5556
Table 2
Receptor binding affinity of Y2 analogs20
Br
a
b
Br
Cl
H
N
X
N
R
O
NR
O
O
Cl
O
N
R = H or Et
10
O
NH
Scheme 2. Reagents and conditions: synthesis of 2-bromo-2-phenyl acetamides
(10). (a) NBS, AIBN, CCl4, 80 °C, 8 h; (b) HNEtR, DIPEA, DCM, 0 °C, 4 h.
R
X
F
NPY Y2 IC50 (nM)
35 22
while 1-fluoro-2-methyl-4-nitrobenzene required 6 h of heating at
60 °C.
22
23
24
25
From intermediate 8 we had the option to install the amide (R1)
or phenyl glycine (R2) portion of the molecule. If we reduced the
NO2 group first, we coupled the resulting primary amine with
our commercially available acid of interest to give intermediate
CN
F
11
4
9. BOC deprotection of 9 followed by reaction of our
a-bromoke-
42 21
N
N
tone (10) with our liberated secondary amine gave our final ana-
logs represented by 12. The
a-bromoketones utilized are either
CN
30 11
26
27
F
370 45
Table 1
Receptor binding affinity of Y2 analogs20
CN
150
6
H
X
N
N
R
O
N
Table 3
Receptor binding affinity of Y2 analogs20
O
O
H
N
X
N
R
R
X
NPY Y2 IC50 (nM)
240 60
O
N
6
H
N
N
O
N
13
14
15
16
17
18
19
F
37
5
R
X
F
NPY Y2 IC50 (nM)
H
530 100
52 24
(JNJ31020028)
5
6
2
N
F
28
29
30
31
CN
F
37
10
8
9
4
CN
Me
H
25 12
100 32
220 53
40 33
CN
F
4
2
N
N
6
N
F
32
33
34
CN
F
200
9
20
21
H
F
530 16
165 20
50 20
CN
13
1

D. M. Swanson et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5552–5556
5555
the use of common intermediates makes this route advantageous
for generation of SAR.
F, CN, Me >> H
Twenty three analogs were synthesized and screened for bind-
ing activity at the human receptor (Tables 1–3).20 In agreement
with the SAR detailed around the GlaxoSmithKline compound
(2),10 it was found that substitution off the central phenyl core
(where X = F, CN, or Me) provides a significant improvement in
Y2 affinity (13, 15–17, 19, 21). This appears to be a steric interac-
tion versus an electronic effect with the methyl, fluoro, and nitrile
substituted compounds all showing improved affinity over their
unsubstituted counterparts (where X = H) (6, 14, 18, 20, 28).
Replacement of the methyl ester with an alkyl amide was also
explored (Tables 2 and 3). Thus the ethyl amides (22–27) and
diethyl amides (5, 28–34) were favored over their methyl ester
counterparts (6, 13–21). For example, the methyl ester 16 saw a
twofold improvement in affinity for the receptor when replaced
with an amide (23, 30) while the methyl ester 21 saw a threefold
improvement in affinity when the ester was replaced (26, 33).
Comparing the alkyl amides, the diethyl amide was found to be
optimal providing improved Y2 affinity over the ethyl amide in
most cases (24 vs 5, 25 vs 31).
1,11-biphenyl,
3-phenylpyridine, and
2-ethylbutane
H
N
X
O
N
N
are tolerated
N
O
O
methylester < ethylamide < diethylamide
Figure 3. SAR Summary.
commercially available or made in two steps from phenyl-acetyl
chloride (Scheme 2).19
If we chose to install our phenyl glycine (R2) portion of the mol-
ecule first we removed the BOC protecting group from intermedi-
ate 8, reacted our
a-bromoketone (10) to give intermediate 11,
In the biaryl amide region of the molecule, three substituents
were found to be tolerated. This included the biphenyls (14–17,
22, 23, 28–30), 3-phenyl pyridines (18, 19, 24, 25, 5, 31), and 2-eth-
ylbutanes (20, 21, 26, 27, 32–34). While the aryl amides
then reduced the NO2 group and coupled our acid of interest to
provide the final compounds represented by 12. Overall, the intro-
duction of structural variability in the last step of the synthesis and
H
H
F
N
F
N
O
O
N
N
N
N
N
N
N
N
O
O
(R)-(-)-JNJ 31020028 (5R)
Y2 IC50 = 9 2 nM
(S)-(+)-JNJ 31020028 (5S)
Y2 IC50 = 14 3 nM
Figure 4. JNJ 31020028 enantiomers.
O
CH3
O
OH
O
S
OH
a
c
b
OH
N
O
OH
OH
+
O
O
O
N
O
O
N
O
N
O
O
(S)(+) Mandelic Acid
35
36
37
38
39
O
H
N
CH3
H
N
F
O
S
F
d
O
O
+
N
O
N
N
N
N
)
O
N
HN
N
O
39
40
5R
(R)-(-)-JNJ 31020028 (
Scheme 3. Reagents and conditions: chiral synthesis of JNJ 31020028. (a) DCC, EtOAc, rt, 4 h; (b) diethylamine (25% in H2O), THF:H2O (1:1), 16 h, rt; (c) MsCl, Et3N, DCM, 0 °C;
(d) Na2CO3, DMF, 50 °C, 4 h, 97% ee by chiral HPLC.

5556
D. M. Swanson et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5552–5556
maintained similar potency across the series the 2-ethylbutanes
were slightly less active at the receptor but with the benefit of a re-
duced molecular weight. Thus while the 2-ethylbutane 34 is less
active than the aryl amide 31 it has a molecular weight below
500 Da, an important criterion for a CNS drug. Ultimately, com-
pound 34 was not pursued further due to cytochrome P450 inhibi-
References and notes
1. Tatemoto, K. Proc. Natl. Acad. Sci. U.S.A. 1982, 79, 5485.
2. Blomqvist, A. G.; Herzog, H. Trends Neurosci. 1997, 20, 294.
3. Kaga, T.; Fujimiya, M.; Inui, A. Peptides 2001, 22, 501.
4. Gehlert, D. R.; Beavers, L. S.; Johnson, D.; Gackenheimer, S. L.; Schober, D. A.;
Gadski, R. A. Mol. Pharmacol. 1996, 49, 224.
5. Redrobe, J. P.; Dumont, Y.; Herzog, H.; Quirion, R. Behav. Brain Res. 2003, 141,
251.
tion (>75% inhibition at CYP2C9 and 2D6 @10 lM).
Overall, six of the 23 analogs tested had an NPY Y2 IC50 6 15 nM
(5, 23, 29–31, 34), correlating to a 15-fold improvement over our
screening lead. Replacement of the methyl ester with an alkyl
amide and substitution off the central core led to both improve-
ments in Y2 affinity and removal of microsomal triglyceride trans-
fer protein (MTP) activity. Three replacements for the 4-pyridyl
phenyl were found to be tolerated including biphenyl, 3-phenyl
pyridine, and 2-ethylbutane. A summary of the SAR for the three
regions of the molecule can be seen in Figure 3.
One member of this series, JNJ 31020028 (5), was previously re-
ported by our labs to be a potent and selective Y2 antagonist that
crosses the blood brain barrier and occupies the receptor after sub-
cutaneous (s.c.) dosing.14 Further, JNJ 31020028 possessed a clean
CYP profile and good solubility at acidic pH. As such it was profiled
in a variety of preclinical animal models to elucidate the pharma-
codynamic effect(s) of a small molecule Y2 antagonist. To address
6. Sato, N.; Ogino, Y.; Mashiko, S.; Ando, M. Expert Opin. Ther. Patents 2009, 19,
1401.
7. Baldock, P. A.; Sainsbury, A.; Couzens, M.; Enriquez, R. F.; Thomas, G. P.;
Gardiner, E. M.; Herzog, H. J. Clin. Invest. 2002, 109, 915.
8. Rimondini, R.; Thorsell, A.; Heilig, M. Neurosci. Lett. 2005, 375, 129.
9. Andres, C. J.; Zimanyi, I. A.; Deshpande, M. S.; Iben, L. G.; Grant-Young, K.;
Mattson, G. K.; Zhai, W. Bioorg. Med. Chem. Lett. 2003, 13, 2883.
10. Lunniss, G. E.; Barnes, A. A.; Barton, N.; Biagetti, M.; Bianchi, F.; Blowers, S. B.;
Caberlotto, L.; Emmons, A.; Holmes, I. P.; Montanari, D.; Norris, R.; Walters, D.
J.; Watson, S. P. Bioorg. Med. Chem. Lett. 2009, 19, 4022.
11. Brothers, S. P.; Saldanha, S. A.; Spicer, T. P.; Cameron, M.; Mercer, B. A.; Chase,
P.; McDonald, P.; Wahlestedt, C.; Hodder, P. S. Mol. Pharmacol. 2010, 77, 46.
12. Jablonowski, J. A.; Chai, W.; Li, X.; Rudolph, D. A.; Murray, W. V.; Youngman, M.
A.; Dax, S. L.; Nepomuceno, D.; Bonaventure, P.; Lovenberg, T. W.; Carruthers,
N. I. Bioorg. Med. Chem. Lett. 2004, 14, 1239.
13. Bonaventure, P.; Nepomuceno, D.; Mazur, C.; Lord, B.; Rudolph, D. A.;
Jablonowski, J. A.; Carruthers, N. I.; Lovenberg, T. W. J. Pharmacol. Exp. Ther.
2004, 308, 1130.
14. Shoblock, J. R.; Welty, W.; Nepomucneo, D.; Lord, B.; Aluisio, L.; Fraser, I.;
Motley, T. M.; Sutton, S. W.; Morton, K.; Galici, R.; Atack, J. R.; Dvorak, L.;
Swanson, D. M.; Carruthers, N. I.; Dvorak, C.; Lovenberg, T. W.; Bonaventure, P.
Psychopharmacology 2010, 208, 265.
any issues associated with profiling
a racemic mixture, JNJ
15. Doods, H.; Gaida, W.; Wieland, H. A.; Dollinger, H.; Schnorrenberg, G.; Esser, F.;
Engel, W.; Eberlein, W.; Rudolf, K. Eur. J. Pharmacol. 1999, 384, R3.
16. Seierstad, M.; Bonaventure, P.; Dvorak, L.; Lord, B.; Miller, K.L.; Motley, S. T.;
Nepomuceno, D.; Chai, W.; Dvorak, C. A.; Jablonowski, J.; Rudolph, D. A.; Shah,
C. R.; Swanson, D. M.; Wong, V. D.; Axe, F. U.; Lovenberg, T. W.; Carruthers, N. I.
Abstracts of Papers, 234th ACS National Meeting, Boston, MA, August 19–23,
2007; USA.
17. Meerpoel, L.; Roevens, P.; Walter, M.; Backx, L.J.J.; Van der Veken, L.J.E.;
Viellevoye, M. PCT Int. Appl. WO 2002020501, A2 20020314, 2002.
18. Meerpoel, L.; Viellevoye, M. PCT Int. Appl. WO 2002081460, A1 20021017,
2002.
19. Dvorak, C. A.; Swanson, D. M.; Wong, V. D. PCT Int. Appl. WO 2009006185, A1
20090108, 2009.
20. NPY Y2 binding affinity was determined as detailed in Bonaventure et al. J.
Pharmacol. Exp. Ther. 2004 308, 1130. IC50 values are the mean of 3–10
determinations followed by SEM. All compounds were characterized by 1H
NMR and MS.
31020028 was resolved by chiral HPLC and the enantiomers were
shown to have similar Y2 affinity (Fig. 4).21 Next, we designed a
representative chiral synthesis starting from commercially avail-
able (S)-(+)-mandelic acid (35) (Scheme 3).
The N-hydroxysuccinimide ester (37) is formed using
N-hydroxysuccinimide (36) and DCC in EtOAc.22 The hydroxy-
succinimide ester (37) is then displaced with aqueous diethylamine
in a mixture of THF and H2O to give 38. Reaction of the alcohol with
mesyl chloride and base in DCM at lowered temperatures provided
the desired mesylate 39. Finally, an SN2 reaction of our chiral mesy-
late (39) with our secondary amine (40) in DMF and base at 50 °C
gave the inversion product, (R)-(ꢀ)-JNJ 31020028 (5R), in 97% ee.23
In summary, a pharmacophore directed screen of our corporate
compound collection led to the identification of promising screen-
ing lead 6. Medicinal chemistry efforts examined three areas of the
molecule that were synthetically accessible and potentially prob-
lematic. Thus 23 analogs were synthesized and tested for binding
affinity at the human Neuropeptide Y Y2 receptor. Six of the ana-
logs tested had an IC50 6 15 nM (5, 23, 29–31, 34). One member
of this series, JNJ 31020028 (5), was recognized as a useful pharma-
cological tool and access to the pure enantiomer was desired. As
such a chiral synthesis was designed starting from a member of
the chiral pool, (S)-(+)-mandelic acid.
21. Racemic JNJ 31020028 was separated by chiral HPLC using a Diacel 1A column
(chiral technologies) with a mobile phase of 10% propanol:90% hexanes with a
flow rate of 1.5 mL/min. Fraction 1, (ꢀ)-enantiomer with retention time of
25.9 min. Optical rotation ½a D20
ꢀ26.6 (c 1.2 MeOH). Fraction 2, (+)-enantiomer
ꢁ
with retention time of 31.4 min. Optical rotation ½a D20
ꢁ
+26.2 (c 1.2 MeOH)
found using a Perkin Elmer Model 341 Polarimeter.
22. Kolassa, J.; Miller, M. J. Org. Chem. 1987, 52, 4978.
23. % Ee was determined by chiral HPLC comparison of the resolved enantiomers.