The design and synthesis of novel NK1/NK2 dual antagonists

Article abstract of DOI:10.1016/S0960-894X(00)00463-7

Functional probing of the backbone of the Sanofi NK₂ antagonist SR 48968 has resulted in the discovery of two new classes of NK₁/NK₂ dual antagonists: the diamine class and the oxime class. The addition of the amino or the oxime functional group results in the reversal of the stereochemical preference of the NK₂ receptor. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0960-894X(00)00463-7

Bioorganic & Medicinal Chemistry Letters 10 (2000) 2329±2332
The Design and Synthesis of Novel NK1/NK2 Dual Antagonists
Gregory A. Reichard,a,* Zachary T. Ball,b Robert Aslanian,a John C. Anthes,a
Neng-Yang Shiha and John J. Piwinskia
aSchering-Plough Research Institute, 2015 Galloping Hill Rd. Kenilworth, NJ 07033-1300, USA
bDepartment of Chemistry, Stanford University, Stanford, CA 94305-5080, USA
Received 9 June 2000; accepted 7 August 2000
AbstractÐFunctional probing of the backbone of the Sano® NK2 antagonist SR 48968 has resulted in the discovery of two new
classes of NK1/NK2 dual antagonists: the diamine class and the oxime class. The addition of the amino or the oxime functional
group results in the reversal of the stereochemical preference of the NK2 receptor. # 2000 Elsevier Science Ltd. All rights reserved.
Design Strategy
Introduction
The neurokinin oligopeptides, substance P (SP) and
neurokinin A (NKA) bind to the neurokinin receptors,
NK1 and NK2, respectively, with moderate selectivity.1
These peptides have been implicated in bronchoconstric-
tion, vasodilation, smooth muscle contraction, edema,
and neurogenic in¯ammation.2 The pathophysiological
e€ects mediated by either the NK1 or NK2 receptor in
asthmatics are related to pulmonary dysfunction.3
Damage to airway epithelium leads to the local release of
SP and NKA from sensory pulmonary C ®bers resulting
in the various pathophysiological responses related to
asthma (bronchospasm, edema, in¯ammatory cell
recruitment, plasma extravasation, and mucous gland
secretion).4 Additionally, lung tissues derived from
asthmatics overexpress both NK1 and NK2 receptors.5
Considerable structure±activity relationship (SAR) work
around NK1 antagonists had been reported in the litera-
ture by P®zer,6 Merck,7 Rhone Poulenc,8 Ciba-Geigy,9
Sano®,10 and Lilly.11 It appeared from the literature that
there was substantial structural variability among the
known NK1 antagonists. In contrast, the SAR infor-
mation for selective NK2 antagonists was extremely
limited. Only one nonpeptidic selective NK2 antagonist
discovered by Sano® (SR 48968) had been reported with
limited SAR disclosed.12
The rational design of antagonists with dual activity is
an emerging ®eld in medicinal chemistry. When our
program was initiated, the only known dual NK1/NK2
antagonist was FK 224, a cyclic peptide discovered by
Fujisawa with binding IC50=37 and 72 nM for the NK1
and NK2 receptors, respectively.13 Although there were
selective NK1 and NK2 nonpeptidic antagonists repor-
ted by several groups, there were no dual nonpeptidic
antagonists in the literature in the neurokinin area at
the commencement of our e€ort.
Antagonists of either SP or NKA have been shown to have
bene®cial e€ects in asthma models; namely, inhibition of
microvascular leakage by NK1 antagonists and reduc-
tion of bronchospasm by NK2 antagonists.2 Since inhi-
bition of either receptor carries potential bene®ts for
asthma, the simultaneous blockade of both receptors
may lead to a new, more ecacious approach to asthma
therapy. Thus, the design and synthesis of single digit or
lower nanomolar dual NK1/NK2 antagonists was the
major focus of our e€orts at the onset of the program.
During the course of our investigation several groups have
also looked at the feasibility of designing dual NK1/NK2
antagonists. Both Hoechst Marion Roussel14 (MDL
105,212) and Merck15 (spiroindolines) have worked from
SR 48968 to build in impressive dual activity.
The strategy described herein focuses on the develop-
ment of dual antagonists by imparting NK1 activity to
the sca€old of a known active NK2 antagonist. The ®rst
generation e€ort was to understand the e€ects of sys-
tematically appending various functional groups to the
*Corresponding author. Tel.: +1-908-740-3522; fax: +1-908-740-
7152; e-mail: gregory.reichard@spcorp.com
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00463-7

2330
G. A. Reichard et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2329±2332
carbon backbone of SR 48968 to probe this structure
for building points that the NK2 receptor would tolerate.
This functionality was then used to construct branched
dual antagonists by the incorporation of an NK1 phar-
macophore into the NK2 antagonist backbone.
Alkylation of the base sensitive hydroxy lactones with
silver oxide and 3,5-bis(tri¯uoromethyl)benzyl bromide
gave lactone ether 7 in 71%. Reduction of the lactone
to the lactol using DIBAL-H, followed by reductive
amination and Jones oxidation provided ketone 8 in
>70% over three steps. Reductive amination with
glycineamide provided a mixture of diastereomeric
diamines 9.
Acyclic NK1 antagonists7 developed at Merck were con-
sidered to be model pharmacophores to incorporate into
backbone substituted analogues of the NK2 antagonist SR
48968. Various modi®cations of the functional appendages
were planned in order to optimize dual activity.
Separation of the diastereomers of diamine 9 and sub-
sequent biological evaluation in NK1 and NK2 receptor
binding assays revealed the major diastereomer of 9 to
be a modest dual inhibitor with Kis of 4 and 34 nM,
respectively.16 Separation of this diastereomer by chiral
HPLC provided the enantiomeric pair 11 and 12. Bio-
logical evaluation showed that 11 retained the dual NK1/
NK2 binding with Kis of 9 and 34 nM, respectively.
Synthesis of optically pure 11 revealed the relative and
absolute con®guration to be R,R.17
Synthesis and Binding Results
A stereo-divergent synthesis depicted in Scheme 1 was
originally pursued in order to assess the biological
activity of all four possible diastereomers of the pro-
posed dual antagonist 2. Thus, conversion of 3,4-
dichlorocinnamic acid to the corresponding cinnamyl
acetate followed by Ireland±Claisen rearrangement
proceeded in 85% yield over four steps to give acid 6.
Subsequent epoxidation with m-CPBA followed by
cyclization with Amberlyst 15 provided a 1:1 mixture of
diastereomeric hydroxy lactones in 81% over two steps.
Biological evaluation of the S,S isomer 12 resulted in
the identi®cation of a potent NK1 selective antagonist.
The racemic trans diamine 13 proved to be an even more
potent NK1 selective antagonist with subnanomolar
binding.
Scheme 1. Reagents and conditions: (a) Cs2CO3, MeI; (b) DIBAL-H; (c) acetyl chloride; pyridine, DMAP; (d) KHMDS, TESCl, THF, Á, 3 h; (e)
m-CPBA; (f) Amberlyst 15; (g) Ag2O, 3,5-bis(tri¯uoromethyl)benzyl bromide, DMF, 23 ꢀC, 48 h; (h) 4-hydroxy-4-phenylpiperidine, NaBH3CN,
.
2,2,2-Tri¯uorethanol, 3 A mol. sieves; (i) Jones; (j) gly-NH2 HCl, NaBH3CN, MeOH.

G. A. Reichard et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2329±2332
2331
Resolution of the enantiomers of oxime 14 by chiral
HPLC followed by methylation provided the optically
pure methyloximes 18 and 19. Subsequent biological
evaluation revealed 19 to be a novel dual antagonist
with Kis of 20 and 7 nM for the NK1 and NK2 recep-
tors, respectively. Synthesis of 19 from optically pure
intermediates con®rmed the absolute stereochemistry to
be the R con®guration.17
Discussion
The disappointing 30% yield of the reductive amination
of ketone 8 prompted further studies to optimize the
transformation of 8 to diamine 9. Since oximes are
readily transformed into amines, and they represent an
interesting moiety from an SAR perspective, we decided
to utilize this group in this series. Thus, treatment of
ketone 8 with both hydroxylamine and methoxylamine
provided the corresponding oximes 14 through 17.
Although all attempts to reduce these oximes failed,
subsequent biological evaluation of these intermediates
revealed a new class of neurokinin dual antagonists. In
all cases, regardless of the oxime geometry, the NK1
activity is similar (i.e., Ki=25 nM). Although the NK2
activity varies as a function of oxime geometry and
substitution, the Z-methyloxime, 17 shows very promis-
ing NK2 binding with a Ki=21 nM.
Utilizing SR 48968, a selective NK2 antagonist, as a
sca€old, we have designed two unique classes of NK1/
NK2 dual antagonists, diamines and oximes. It is very
interesting to note that the usual stereochemical pref-
erence of known selective NK2 receptor antagonists is
the S enantiomer. This is not only the case for SR 48968,
but also for the dual antagonists reported by both Merck
and HMR during the course of our research. In fact, the
design of MDL 105,212 utilized molecular modeling
studies which showed very good overlap of the 3,4-
dichlorophenyl ring, the benzamide aromatic ring, a
hydrogen bond acceptor14 and the piperidine nitrogen
with SR 48968. Correspondingly, Merck's dual antago-
nists contain no functional backbone modi®cations to
SR 48968; hence it is reasonable to assume the overlap
of these groups with SR 48968 is very good.
In our work we have found that in modifying the back-
bone of the Sano® antagonist, the stereochemical pref-
erence can be signi®cantly a€ected by functional groups
adjacent to the benzylic center. In the case of our origi-
nal designed amine branched antagonists, dual activity
was imparted to the Sano® sca€old by the introduction of
the Merck NK1 pharmacophore. Unexpectedly, the intro-
duction of this unit in a stereospeci®c manner reverses the
stereopreference for the NK2 receptor. In this series, no
isomers containing the S stereochemistry, show any
appreciable anity for the NK2 receptor. The stereo-
chemistry does not appreciably in¯uence the NK1 binding.
In the case of the oxime derivatives, the introduction of
the achiral oxime functional group adjacent to the
benzylic center also reverses the stereopreference for the
NK2 receptor such that the dual active compounds

2332
G. A. Reichard et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2329±2332
contain the R stereochemistry as well. Similarly, for this
class, the NK1 activity is relatively una€ected by the
stereochemistry of the benzylic center.
4. (a) Joos, G. F.; Pauweis, R. A.; van der Straeten, M.; Bull
Eur. Physiopathol. Respir. 1988, 23, 619. (b) Nadel, J. A.;
Borson, D. B. Am. Respir. Dis. 1991, 143, S33. (c) Martins, M.
A.; Shore, S. A.; Drazen, J. M. Int. Arch. Allergy Appl.
Immunol. 1991, 94, 325. (d) Solway, J.; Le€, A. R. J. Appl.
Physiol. 1991, 71, 2077.
5. Joos, G. F.; Van Schoor, J.; Kips, J. C.; Pauwels, R. A.
Am. J. Respir. Crit. Care Med. 1996, 153, 1781.
6. Desai, M.; Lefkowitz, S. L.; Thadeio, P. F.; Longo, K. P.;
Snider, R. M. J. Med. Chem. 1992, 35, 4911.
7. Williams, B.; Teall, M.; McKenna, J.; Harrison, T.; Swain,
C. J.; Cascieri, M.; Sodowski, S.; Strader, C.; Baker, R.
Bioorg. Med. Chem. Lett. 1994, 4, 1903.
8. Archard, D.; Truchon, A.; Peyronel, J. F. Bioorg. Med.
Chem. Lett. 1994, 4, 669.
9. Ofner, S.; Hausser, K.; Schilling, W.; Vassout, A.; Veen-
stra, S. J. Bioorg. Med. Chem. Lett. 1996, 6, 1623.
10. Emonds-Alt, X.; Doutremepuich, J. D.; Healulme, M.;
Neliat, G.; Santucci, V.; Steinberg, R.; Vilain, P.; Bichon, D.;
Ducoux, J. P.; Proietto, V.; van Brock, D.; Soubrie, P.; Le
Fur, G.; Breliere, J. C. Eur. J. Pharmacol. 1993, 250, 403.
11. Hipskind, P. A.; Howbert, J. J.; Bruns, R. F.; Cho, S. Y.;
Crowell, T. A.; Foreman, M. M.; Gehlert, D. R.; Iyengar, S.;
Johnson, K. W.; Krushinski, J. H.; Li, D. L.; Lobb, K. L.;
Mason, N. R.; Muehl, B. S.; Nixon, J. A.; Phebus, L. A.;
Regoli, D.; Simmons, R. M.; Threkeld, P. G.; Waters, D. C.;
Girtter, B. D. J. Med. Chem. 1996, 39, 736.
12. Emonds-Alt, X.; Proietto, V.; Van Broeck, D.; Vilain, P.;
Advenier, C.; Neliat, G.; Le Fur, G.; Breliere, J. C. Bioorg.
Med. Chem. Lett. 1993, 3, 925.
13. Murai, M.; Morimoto, H.; Maeda, Y.; Kiyotoh, S.; Nish-
ikawa, M.; Fuji, T. J. Pharmacol. Exp. Ther. 1992, 262, 403.
14. Burkholder, T. P.; Kudlacz, E. M.; Le, T.-B.; Knippen-
berg, R. W.; Shatzer, S. A.; Maynard, G. D.; Webster, M. E.;
Horgan, S. W. Bioorg. Med. Chem. Lett. 1996, 6, 951.
15. Shah, S. K.; Hale, J. J.; Qi, H.; Miller, D. J.; Dorn, C. P.,
Jr.; Mills, S. G.; Sadowski, S. J.; Cascieri, M. A.; Metzger, J.
M. Book of Abstracts, 212th National Meeting of the ACS,
Orlando, FL American Chemical Society: Washington, DC;
MEDI-136.
In summary, we have discovered two novel classes of
NK1/NK2 dual antagonists. The novelty of the oxime
class of compounds in particular, has prompted further
SAR investigation of this class of compounds, the results
of which will be the subject of future publications.
16. Receptor binding assays were performed 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 Kds. Kis were obtained according
to the Cheng and Prusso€ equation.
References and Notes
1. Buck, S. H.; Burcher, E. Trends Pharmacol. Sci. 1986, 7, 65.
2. Watling, K. J.; Guard, S. Neurotransmissions 1992, 8, 1.
3. (a) Barnes, P. J. Lancet 1986, 1, 242. (b) Barnes, P. J.;
Chung, F. K.; Page, C. P. Pharmacol. Rev. 1988, 40, 49.
17. Manuscript in preparation.