Synthesis and NK1/NK2 binding activities of a series of diacyl-substituted 2-arylpiperazines

Article abstract of DOI:10.1016/S0960-894X(02)00645-5

The synthesis and binding affinity for hNK₁ and hNK₂ receptors of a series of diacyl substituted 2-aryl piperazines are described. SAR evaluation led to the racemic derivative 11g as an apparent dual inhibitor. Chiral chromatographic separation of 11g led to the observation that NK₁ activity was shown by one enantiomer (13a) and NK₂ activity was shown by the other enantiomer (13b). X-ray crystallographic analysis of the crystalline di-BOC derivative of the NK₂ active piperazine (15) showed that the 2R configuration was associated with NK₂ activity. Further derivatization indicated that dual NK₁/NK₂ activity could be built into the 2R series.

Full text of DOI:10.1016/S0960-894X(02)00645-5

Bioorganic & Medicinal Chemistry Letters 12 (2002) 3161–3165
Synthesis and NK1/NK2 Binding Activities of a Series of
Diacyl-Substituted 2-Arylpiperazinesy
David J. Blythin,a,* Xiao Chen,a John J. Piwinski,a Neng-Yang Shih,a Ho-Jane Shue,a
John C. Anthesb and Andrew T. McPhailc
aChemical Research Department, Schering-Plough Research Institute, 2015 Galloping Hill Road, Kenilworth, NJ 07033, USA
bDepartment of Allergy and Immunology, Schering-Plough Research Institute, 2015 Galloping Hill Road, Kenilworth, NJ 07033, USA
cPaul M. Gross Chemical Laboratory, Duke University, Durham, NC 27708, USA
Received 16 April 2002; accepted 24 July 2002
Abstract—The synthesis and binding affinity for hNK1 and hNK2 receptors of a series of diacyl substituted 2-aryl piperazines are
described. SAR evaluation led to the racemic derivative 11g as an apparent dual inhibitor. Chiral chromatographic separation of
11g led to the observation that NK1 activity was shown by one enantiomer (13a) and NK2 activity was shown by the other enan-
tiomer (13b). X-ray crystallographic analysis of the crystalline di-BOC derivative of the NK2 active piperazine (15) showed that the
2R configuration was associated with NK2 activity. Further derivatization indicated that dual NK1/NK2 activity could be built into
the 2R series.
# 2002 Elsevier Science Ltd. All rights reserved.
The family of neurokinins, or tachykinins, consists of
three related neuropeptides (Substance P, neurokinin A,
and neurokinin B) that possess the same C-terminal
sequence. They interact to different degrees with three
G-protein linked receptors, known as NK1, NK2, and
NK3, but each has selectivity for one of them. Substance
P shows greatest affinity for NK1, neurokinin A for
NK2, and neurokinin B for NK3 receptors.2
approach is illustrated in work from, for instance, our
laboratories6,7 and from AstraZeneca.8
Our own work was based, in part, on the 1-phenyl
ethylenediamine fragment present in the early Pfizer
lead, CP-99,994, a potent NK1 antagonist.9 By incor-
porating this potential pharmacophore into a piperazine
ring the general target 1 was designed, as shown in Fig-
ure 1.10 The orientation of the two pairs of aryl rings in
the two structures is very similar.
Recent interest in dual NK1/NK2 antagonists has been
based in part on the expectation that a drug showing
these activities would affect favorably several undesir-
able elements of bronchial asthma. Substance P (NK1
agonist) is known to be involved in the extravasation
and consequent inflammation of the airways, and neu-
rokinin A (NK2 agonist) appears to be involved in
bronchoconstriction and cough.3,4 Because of this
anticipated activity in asthma there have been numerous
efforts to design potent dual NK1/NK2 antagonists that
have resulted in a number of structural leads. These
series were developed mainly from existing NK1 or NK2
antagonist leads as described in the review by Ger-
spacher and von Sprecher.5 More recently, this
We chose the unsubstituted phenyl (2a) and the 3,4-
dichlorophenyl (2b) piperazines as representative of
structures present in several neurokinin antagonists.
The key intermediates needed for these syntheses were
ySee ref. 1
*Corresponding author. Fax: +1-908-740-7305; e-mail: david.blythin@
spcorp.com
Figure 1. Design of potential piperazine antagonists of the neuro-
kinins.
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00645-5

3162
D. J. Blythin et al. / Bioorg. Med. Chem. Lett. 12 (2002) 3161–3165
the 2-arylpiperazines. Two different routes were used to
prepare them (Scheme 1). Unsubstituted 2-phenylpiper-
azine 2a was prepared from 2-chloropyrazine 3 b y a
nickel-catalyzed coupling with phenyl Grignard
reagent11 to produce 2-phenylpyrazine 4. Catalytic
reduction produced 2a in about 70% overall yield. We
anticipated that a similar route would not be suitable
for preparing the 3,4-dichlorophenylpiperazine 2b
because of potential dechlorination in the catalytic
reduction step. We chose to use the method described
by Roderick et al.12 starting from methyl 3,4-dichloro-
phenylacetate 5 to produce 2b in three steps. The pro-
ducts (2a, b) were racemic and our early work was
carried out with these racemates.
tion to produce 6 followed by low-temperature
acylation at the 1-position giving 7 and, finally, sub-
stitution of the bromine by a series of amines to yield
the products 8a–j. Some products (e.g., 8a) showed a
moderate level of NK1 activity but no NK2 activity at
the highest level tested (1 mM). The use of 4-amino-1-
benzylpiperidine produced 8i and 8j, which were the
first targets to show a significant level of dual NK1/NK2
binding affinity. Retaining the 4-amino-1-benzylpiper-
idine unit allowed us to examine the effect of substitu-
tion at the 4-position of the piperazine.
Synthesis of this series is outlined in Scheme 3. Initial
protection of the more reactive 4-position of the 2-aryl
piperazine with (BOC)2O led to intermediates 9a and
9b. This was followed by incorporation of the bromo-
acetyl unit and then the 4-amino-1-benzylpiperidine, as
described before (Scheme 2) to produce intermediates
Our initial plan was to retain a basic nitrogen at the 4-
position of the piperazine ring, to mimic the basic sub-
stituent at the 3-position in CP-99,994 (Fig. 1). Incor-
poration into 1 of substituted amines in which the
substituents, R1 and R2, resemble those present in
known NK1 and/or NK2 antagonists, led to the series of
compounds shown in Table 1. The synthesis of these
compounds is shown in Scheme 2. This involved selec-
tive alkylation of 2a and 2b at the less hindered 4-posi-
Table 1. Discovery of the initial leads
a
Compd
X
–NR1R2
NK1aKi (nM) NK2 Ki (nM)
8a
8b
H
Cl
13
32
>1000
>1000
8c
8d
H
Cl
37
110
>1000
>1000
8e
8f
H
Cl
48
148
>1000
>1000
8g
8h
H
Cl
48
28
>1000
>1000
8i
8j
H
Cl
17
30
300
175
aNK1 and NK2 Binding assays: Binding data are the average of two or
three independent determinations. Receptor binding assays were per-
formed on membrane preparations containing recombinant human
NK1 or NK2 receptors in CHO cells. [3H]Sar SP and [3H]NKA were
used as the ligands for the NK1 and NK2 receptor assays respectively,
at the experimentally derived Kd values. Ki values were obtained using
the Cheng and Prusoff equation.13
Scheme 1. (a) (Ph2PCH2)2NiCl2, PhMgBr, THF (95%); (b) Pd(OAc)2,
H2, CH3CO2H, 50 psi, 4 h (75%); (c) NBS; (d) H2N–CH2–CH2–NH2;
(e) LAH, Et2O.
Scheme 2. (a) 3,5-(CF3)2C6H3–CH2Br, Et3N, CH2Cl2, À78 ꢀC to rt; (b) BrCH2COBr, Et3N, CH2Cl2, À78 ꢀC; R1R2N–H, Et3N, CH2Cl2.
Scheme 3. (a) (BOC)2O, MeOH, À78 ꢀC; (b) BrCH2COBr, (i-Pr)2NEt, CH2Cl2; (c) 4-amino-1-benzylpiperidine; (d) 4 M HCl in dioxane; (e) R-CH2–
Br(Cl); (f) R-CO2H, HOBT, Et3N, DEC.

D. J. Blythin et al. / Bioorg. Med. Chem. Lett. 12 (2002) 3161–3165
Table 2. Effect of substitution at the 4-position of the piperazine ring
3163
towards optimizing the desired enantiomer. 2-(3,4-
Dichlorophenyl)piperazine 2b was converted to its
N-acetyl-l-leucine salt which was separated by fractional
crystallization (Scheme 4).14 In fact, both enantiomers,
14a and 14b, could be obtained pure using this salt
without the need to use N-acetyl-d-leucine to obtain the
second enantiomer. Conversion of the pure enantiomers
to the same compounds, previously obtained by chiral
HPLC (13a and 13b), confirmed by bioassay which
enantiomer bound to which neurokinin receptor.
Compd
X
Y
R
NK1aKi NK2aKi
(nM)
(nM)
11a
11b
11c
11d
11e
H
H
O¼
–CH2–[3,5-(CF3)2–C6H3]
136
135
>1000
>1000
>1000
124
H,H
Cl O¼ –CH2–[3,4,5–(CH3O)3–C6H2] >1000
Cl O¼
Cl O¼
–[3,4,5–(CH3O)3–C6H2]
–[3,5–(CF3)2–C6H3]
112
5.3
17
10a and 10b after removal of the BOC protecting group.
Alkylation or acylation produced the series of com-
pounds 11a–e shown in Table 2. Very limited activity
was seen with phenylacetyl derivatives 11a–c whereas
the 3,5-disubstituted benzoyl derivative 11e appeared to
show enhanced potency at both the NK1 and NK2
receptors.
To establish the absolute configuration of the NK2
binding series the relevant piperazine enantiomer (14b)
was converted to its crystalline di-BOC derivative 15
(Scheme 4) which was subjected to X-ray structure
determination using the anomalous dispersion
method.20 Results showed that the NK2-active enantio-
mer possessed the 2R configuration. Therefore, the NK1
active series was produced from the 2S isomer. A stereo
view of 15, with the methyl groups of the BOC group
removed for clarity, is shown in Figure 2.
We inferred from these results that diacylpiperazines,
such as 11e, constituted the best lead. We carried out an
optimization of the acid group by synthesizing numer-
ous aromatic amide derivatives of which a selection is
shown in Table 3. This indicated that the 3,5-dimethyl-
benzoyl derivative 11g gave optimum binding potency.
The benzoyl derivative 11f showed good NK2 binding
potency but lacked significant NK1 activity confirming
the need for the 3- and 5- substituents on the aromatic
ring. However, highly polar substituents on the ring are
not favored (e.g., 11k).
Evaluation of the data obtained on the pure enantio-
meric products (Table 4) suggested that introducing
NK1 activity into the 2R series might be more feasible
than attempting to build NK2 activity into the 2S series.
NK2 activity in 12a and 13a is extremely weak (Ki>200
nM) and would need to be increased almost two orders
Table 4. Separation of enantiomers of 11e and 11g
Next, we investigated the influence of absolute stereo-
chemistry on activity. Our initial assumption was that
activity would reside in one enantiomer with the other
being much less potent, or inactive. Compounds 11e
and 11g were separated into their corresponding enan-
tiomers (12a, 12b and 13a, 13b) by chiral HPLC on a
Daicel Chiralpak AD1 column. Testing of the enantio-
mers showed that, contrary to our expectations, NK1
activity resided in one enantiomer (12a and 13a) and
NK2 activity resided in the other (12b and 13b), as
shown in Table 4. The absolute stereochemistry was
determined as shown below. At this time we confirmed,
using appropriate functional assays, that the racemates
or enantiomers were antagonists, although their potency
was rather low (Table 4).
Compd NK1 Ki (nM) NK2 Ki (nM) pA2a,c NK1 pA2 NK2
b,c
11e
12a
12b
11g
13a
13b
5.3
1.8
48
1.8
0.9
25
17.2
270
8.3
4.9
219
3.4
7.1
6.5
8.2
6.5
aThe functional NK1 bioassay was performed in isolated guinea pig
vas deferens (gpvd) induced to contract by electrical field stimulation
(EFS). Inhibition of the SP-induced NK1-mediated enhancement of
the EFS-induced sympathetic neurogenic contractions of gpvd was
used as a measure of antagonist activity.15,16
bThe functional NK2 bioassay was performed using NKA-induced
airway contractions of isolated hamster trachea (ht). The inhibition of
NKA-induced airway contraction was used as a measure of NK2-
receptor antagonist activity.17,18
cAntagonists were applied to the baths 30 min before the application
of rising cumulative concentrations of SP (gpvd) or NKA (ht).
Apparent pA2 values were estimated from the magnitude of the
antagonist-induced rightward parallel shifts in the agonist concentra-
tion–response curves, using the method of Furchgott19 and assuming
competitive kinetics and equilibrium conditions.
In order to determine which enantiomer possessed
which biological activity we resolved an early inter-
mediate. The 2-arylpiperazines were the obvious choice
because all subsequent reactions could be directed
Table 3. Modifications to the 4-aroyl substituent in 11 (X=Cl;
Y=O)
Compd
R
NK1 Ki (nM)
NK2 Ki (nM)
11f
11g
11h
11i
11j
11k
11l
–C6H5
–[3,5–(CH3)2–C6H3]
–[3,5–F2–C6H3]
–[3,5–(CH3O)2–C6H3]
–[3,5–Cl2–C6H3]
–[3,5–(OH)2–C6H3]
–[3,5–Br2–C6H3]
93
1.8
23
3.4
4.9
22
27
7.1
7.7
>1000
3.2
14
>1000
14
Figure 2. Stereo view of the 2R-piperazine unit (methyls of BOC
groups removed for clarity).

3164
D. J. Blythin et al. / Bioorg. Med. Chem. Lett. 12 (2002) 3161–3165
Scheme 4. (a) N-Acetyl-l-leucine salt; recrystallize from MeOH/EtOAc, liberate free base, recrystallize free base from hexanes to >99% ee; (b)
(BOC)2O; MeOH; À78 ꢀC to rt; recrystallize from EtOH.
Scheme 5. (a) 3,5-(CH3)2C6H3–CO2H, HOBT, Et3N, DEC, 0 ꢀC; (b) BrCH2COBr, (i-Pr)2NEt, CH2Cl2, À78 ꢀC; (c) 4-Amino-N1-BOC-piperazine,
(i-Pr)2NEt, CH2Cl2, (d) 4 M HCl in dioxane; (e) Y-CH2–Br, Et3N, CH2Cl2; (f) Y-CHO, NaBH3CN, AcOH, MeOH.
of magnitude to be useful. However, NK1 activity in 12b
and 13b is quite respectable (Ki<50 nM) and only needs
to be improved by about one order of magnitude to
produce a useful dual antagonist. An additional
requirement in each case, of course, is to retain binding
to the original receptor. To test this hypothesis we chose
to modify the group attached to the 4-aminopiperidine
unit of 13b. This was carried out as shown in Scheme 5
in which the more reactive 4-position of 14b was cou-
pled with 3,5-dimethylbenzoic acid giving 16 followed
by low temperature acylation with bromoacetyl bro-
mide to produce 17. Displacement of the activated
halogen in 17 with N1-BOC-protected 4-aminopiper-
idine followed by deprotection produced the free amine.
This amine was then alkylated either directly with alkyl
halides or by reductive alkylation to produce a series of
alkylated derivatives.
(15 nM for 18a compared to 25 nM for 13b). However,
when the piperidine nitrogen was acylated with pro-
tected amino-acid derivatives, followed by deprotection
(Scheme 6), we found another slight increase in binding
potency (11 nM for 20c) in one compound (Table 6).
We have shown that, in the 2-(3,4-dichloro-
phenyl)piperazine series, NK1 activity resides in the 2S
enantiomer and NK2 activity resides in the 2R enantio-
mer. Furthermore, dual NK1/NK2 antagonists can be
produced in the 2R series with a substituted 4-amino-
piperidine substituent. This template is being used as a
basis to explore the potential of the diacylpiperazines as
dual antagonists.
Table 5. Modification of the substituent on the piperidine nitrogen
Compd
Y
NK1 Ki (nM) NK2 Ki (nM)
A selection of the substituents synthesized is shown in
Table 5 (18a–d). The unsubstituted piperidine 19
showed very little NK1 activity and even NK2 activity
was reduced compared to 13b. While it was possible to
retain NK2 activity by appropriate substitution, only
slight improvement in NK1 binding activity was found
19
No substituent on N
4-CH3CONH–C6H4—
2-Imidazolyl
3-Pyrrolyl
4-(2-CH3CONH)–thiazolyl
215
15
20
1.3
18a
18b
18c
18d
92
35
52
3.0
1.3
0.9

D. J. Blythin et al. / Bioorg. Med. Chem. Lett. 12 (2002) 3161–3165
3165
Table 6. Amino acid derivatives in the 2R series
Shih, N.-Y.; Piwinski, J. J. Bioorg. Med. Chem. Lett. 2000, 10,
2329.
Compd
Y
NK1 Ki (nM)
NK2 Ki (nM)
7. Ting, P. C.; Lee, J. F.; Anthes, J. C.; Shih, N.-Y.; Piwinski,
J. J. Bioorg. Med. Chem. Lett. 2000, 10, 2333.
8. Bernstein, P. R.; Aharony, D.; Albert, J. S.; Andisik, D.;
Barthlow, H. G.; Bialecki, R.; Davenport, T.; Dedinas, R. F.;
Dembofsky, B. T.; Koether, G.; Kosmider, B. J.; Kirkland,
K.; Ohnmacht, C. J.; Potts, W.; Rumsey, W. L.; Shen, L.;
Shenvi, A.; Sherwood, S.; Stollman, D.; Russell, K. Bioorg.
Med. Chem. Lett. 2001, 11, 2769.
20a
20b
20c
20d
20e
20f
S-Phg
R-Phg
S-Phe
R-Phe
S-Tyr
S-Trp
143
42
11
21
32
70
16
1.1
1.1
3.2
7.0
28
9. Desai, M. C.; Lefkowitz, S. L.; Thadeio, P. F.; Longo,
K. P.; Snider, R. M. J. Med. Chem. 1992, 35, 4911.
10. After this project was begun a PCT application issued to
Fujisawa Pharmaceutical Co., Ltd., Japan WO97/22597 cov-
ering certain 2-aryl-substituted piperazines as neurokinin
antagonists. As far as we are aware, no literature publications
have appeared describing their work.
11. Godek, D. M.; Rosen, T. J. WO 91/18878. Chem. Abstr.
1992, 116, 106106.
12. Roderick, W. R.; Platte, H. J.; Pollard, C. B. J. Med.
Chem. 1966, 9, 181.
13. Cheng, Y. C.; Prusoff, W. H. Biochem. Pharmacol. 1973,
22, 3099.
14. We thank Dr. K. McCormick for developing the resolu-
tion methodology.
Scheme 6. (a) N-BOC-Z-CO2H, HOBT, Et3N, DEC; (b) 4 M HCl in
dioxane.
15. Watson, S. P. Br. J. Pharmacol. 1983, 80, 205.
16. Hall, J. M.; Morton, I. K. M. Br. J. Pharmacol. 1991, 102,
511.
17. Maggi, C. A.; Patacchini, R.; Robero, P.; Mali, A. Eur. J.
Pharmacol. 1989, 166, 435.
Acknowledgements
The authors would like to thank Mr. W. Tom for
developing the procedure described in reference 12, Mr.
C. Rizzo and Mr. S. Tozzi for the in vitro functional
assay data, Mr. Z. Zhan and Mr. C. Richard for the
NK1 and NK2 binding data, and Dr. J. Kaminski and
Ms. L. Weber for help with the X-ray structure repre-
sentation.
18. Ellis, J. L.; Undem, B. J.; Kays, J. S.; Ghanekar, S. V.;
Barthlow, H. G.; Buckner, C. K. J. Pharmacol. Exper. Ther.
1993, 267, 95.
19. Furchgott, R. F. Pharmacol. Rev. 1955, 7, 183.
20. Compound 15, mp 153–154 ꢀC, [a]23.9=À53.9ꢀ (3.54 mg
in 2 mL MeOH), was crystallized from 95% EtOH. Crystal
data: C20H28Cl2N2O4, M=431.46, orthorhombic, space
˚
˚
˚
group P212121, a=14.118(2) A, b=26.022(5) A, c=6.058(1) A,
V=2226(1) A , Z=4, Dcalcd=1.287 g cmÀ3, m(Cu Ka radiation,
3
˚
l=1.5418 A)=28.9 cmÀ1, crystal size: 0.03ꢁ0.06ꢁ0.60 mm.
˚
Intensity data (2640 non-equivalent reflections, ymax=75ꢀ)
were recorded on an Enraf-Nonius CAD4 diffractometer. The
crystal structure was solved by direct methods. Full-matrix
least-squares refinement of atomic positional and thermal
parameters (anisotropic C, Cl, N, O, fixed H contributions)
converged (max. shift:esd=0.03) at R=0.045 (Rw=0.058)
over 1446 reflections with I>2.0s(I). Crystallographic data
(excluding structure factors) have been deposited with the
Cambridge Crystallographic Data Centre, deposition number
CCDC 175702. Copies of the data can be obtained, free of
charge, on application to CCDC, 12 Union Road, Cambridge
CB2 1EZ, UK (fax: +44–1223–336033; e-mail: deposit@ccdc.
cam.ac.uk).
References and Notes
1. Presented at the 221st National Meeting of the American
Chemical Society, San Diego, CA, April 1–5, 2001; MEDI
244.
2. Regoli, D.; Boudon, A.; Fauchere, J.-L. Pharmacol. Rev.
1994, 46, 551.
3. Advenier, C.; Lagente, V.; Boichot, E. Eur. Respir. J. 1997,
10, 1892.
4. Chapman, R. W.; Hey, J. A.; McLeod, R.; Minnicozzi, M.;
Rizzo, C. A. Drug News Perspect. 1998, 11, 480.
5. Gerspacher, M.; von Sprecher, A. Drugs Future 1999, 24, 883.
6. Reichard, G. R.; Ball, Z. T.; Aslanian, R.; Anthes, J. C.;