Synthesis and structure-activity relationships of oxime neurokinin antagonists: Discovery of potent arylamides

Article abstract of DOI:10.1016/S0960-894X(01)00709-0

The structural modification of the benzylic ether region of oxime 1 has resulted in the identification of several novel aryl amides as selective or dual NK₁/NK₂ antagonists.

Full text of DOI:10.1016/S0960-894X(01)00709-0

Bioorganic & Medicinal Chemistry Letters 12 (2002) 141–145
Synthesis and Structure–Activity Relationships of Oxime
Neurokinin Antagonists: Discovery of Potent Arylamides
Neng-Yang Shih,* Margaret Albanese, John C. Anthes, Nicholas I. Carruthers,y
Cheryl A. Grice,y Ling Lin, Pietro Mangiaracina, Gregory A. Reichard, John Schwerdt,
Vera Seidl, Shing-Chung Wong and John J. Piwinski
Schering-Plough Research Institute, 2015 Galloping Hill Rd, Kenilworth, NJ 07033-1300, USA
Received14 August 2001; accepted10 October 2001
Abstract—The structural modification of the benzylic ether region of oxime 1 has resultedin the identification of several novel aryl
amides as selective or dual NK1/NK2 antagonists. # 2002 Elsevier Science Ltd. All rights reserved.
Synthesis
Substance P (SP) andneurokinin A (NKA) which bind
to the neurokinin receptors, NK1 andNK 2, respectively,
have been implicatedin many pathophysiological effects
relatedto pulmonary dysfunction. 1 Since antagonists of
either SP or NKA carries potential benefits for asthma,
concomitant blockade of both receptors may lead to a
new approach to asthma therapy. The administration of
We identified the oxime alcohol 5 as a versatile synthetic
intermediate for the preparation of the modifications of
the ether region. The synthesis of 5 (Scheme 1) employs
carboxylic acid 2 preparedfrom 3,4-dichlorocinnamic
acid.2 Standard coupling and reduction give rise to the
amine olefin 3. Subsequent dihydroxylation followed by
selective silylation andoxidation affordthe TBS ketone
4. Oxime formation followedby deprotection then pro-
vides the desired oxime alcohol 5 as a single isomer.
This intermediate is functionalized appropriately using
standard chemistry to access the analogues in Table 1.
a single entity with both NK1 andNK affinity is one
2
approach to achieve such simultaneous antagonism.
The discovery and optimization of potent dual NK1/
NK2 antagonists is an active area in medicinal chem-
istry. We recently reportedthe design andsynthesis of a
new class of oxime antagonists that are equipotent at
the NK1 andNK receptors.2 Towardthe further opti-
2
mization of this series, we now report the structure–
activity relationships (SAR) involvedin the modifica-
tion of the benzylic ether region of oxime 1 (Fig. 1).
Figure 1.
*Corresponding author. Tel.: +1-908-740-3530; fax: +1-908-740-
7152; e-mail: neng-yang.shih@spcorp.com
yCurrent address: The R. W. Johnson Pharmaceutical Research Insti-
tute, 3210 MerryfieldRow, San Diego, CA 92121, USA.
Scheme 1. Reagents and conditions: (a) EDC, 4-hydroxyl-4-phenyl-
piperidine, CH2Cl2; (b) LAH/THF; (c) OsO4/NMO; (d) TBSCl/NEt3/
CH2Cl2; (e) PDC/CH2Cl2; (f) NH2OMe/NaOAc, aq EtOH; (g) TBAF.
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00709-0

142
N.-Y. Shih et al. / Bioorg. Med. Chem. Lett. 12 (2002) 141–145
The amine-basedanalogues (Tables 2 and3) are pre-
paredby oxidation of oxime alcohol 5, followedby
standard reductive amination. Functionalization to
those tabulatedanalogues in Tables 2 and3 is straight-
forward. The carbogenic analogues (entries 27 and 28,
Table 2) require a different route (Scheme 2).
with the a-chloro-Weinreb amide. Conversion to the
oxime, followed by formation of the aldehyde and
reductive amination provides the carbogenic analogues
(entries 27 and28).
Discussion
We initially set out to explore the tolerance of both
receptors for different types of aryl and alkyl ethers.
The parent unsubstitutedbenzyl ether (entry 2) shows
a significant loss of NK1 binding,4 however, the NK2
binding is retained. Positional modulation of a single
methyl group indicates a preference, albeit diminished,
for dual activity with the 3-substituted analogue (entry
4). Alkyl substituents in the 4-position leads to NK2
selective compounds. In fact, bulky alkyl ethers
(entries 7 and8) show excellent selective NK binding
2
with the adamantyl analogue completely selective for
the NK2 over the NK1 receptor. Larger aryl andhet-
eroaryl ethers continue the trendof NK
selectivity
2
(entries 9–13). However, the pyridyl, oxadiazole, and
fluoronaphthyl derivatives (entries 14–16) show modest
dual activity. Focusing on 3,5-disubstituted-phenyl5
modifications led to a consistent series of dual active
analogues. The smaller difluoro substitution (entry
17) have diminished NK1 affinity comparedto the
dimethyl, dichloro, and dibromo derivatives (entries
18–20). Both receptors seem to accommodate the 3,5-
Scheme 2. Reagents andconditions: (a) 2 equiv LHMDS @ ꢀ78 ꢁC;
then, add 7, warm to 23 ꢁC; (b) LHMDS, Me(MeO)N(CO)CH2Cl; cat.
KI; (c) NH2OMe HCl/pyridine; (d) DIBAL-H/THF; (e) 4-hydroxyl-4-
phenylpiperidine, NaBH3CN, MeOH.
Ketone 8 is accessed via the addition of the dianion of
3,4-dichlorophenylacetic acid to the appropriate acid
chloride 73 followedby decarboxylation. Two-carbon
homologation of ketone 8 is performedvia alkylation
dimethoxy and 3,4,5-tri-methoxy substituents found in
6
reporteddual antagonists. However, at this point in
our SAR survey, there was no clear advantage for
Table 1. Structure–activity relationships for alkyl andaryl ether analogues
Compd
Ki (nM) or %I (1 mM) Compd
Ki (nM) or %I (1 mM) Compd
Ki (nM) or %I (1 mM)
NK1
NK2
NK1
NK2
NK1
NK2
1
2
3
4
5
6
7
8
25
21
9
2%
38
17
18
19
20
21
22
23
24
113
42
33%
34%
140
40
136
14
9
10
11
12
13
14
15
16
37%
23%
29%
48%
115
12
20
17
21
43
1
26
10
6
15
8
32
48%
6%
33
18
28
10
130
4
53
21%
0%
13
7
100
96
70
21
19%

N.-Y. Shih et al. / Bioorg. Med. Chem. Lett. 12 (2002) 141–145
143
this substitution pattern. The di-t-butyl analogue
(entry 24) reveals the lack of tolerance for large alkyl
substitution at the 3,5-positions at the NK1 receptor
binding site.
Table 2. Structure–activity relationships for 3,5-bis(trifluoro-
methyl)phenyl analogues
Compd
Ki (nM) or %I (1 mM)
NK1
NK2
With 3,5-substitutedaryl ethers representing a promis-
ing dual active series, it was interesting to investigate
alternate linkages to the ether unit (Fig. 2). The 3,5-
bis(trifluoromethyl)phenyl group was heldconstant and
several heteroatom andcarbogenic linkers were pre-
pared(Table 2).
1
25
21
25
26
27
28
29
30
31
32
33
34
35
36
168
75
25
157
332
51%
31%
118
100
20
92
23
Figure 2.
81
39
We discovered that a heteroatom is necessary but not
sufficient along a five- or six-atom chain between the
two aryl rings in order to maintain dual activity. The
addition of a carbonyl group between the ether oxygen
20%
36%
50
andthe aromatic ring (entry 25) diminishes NK bind-
1
ing yet retains NK2 affinity. Methylation of the oppos-
ing carbon (entry 26) diminshes binding at both
receptors. The all carbon analogue (entry 27) is also less
active at both receptors. Shortening the carbon chain to
four atoms between the aryl rings (entry 28), surpris-
ingly restores NK1 activity, but diminishes NK2 activ-
ity. By reinstalling the ether linkage for a four-atom
analogue (entry 29) the same preference for the NK1
receptor is observed, however, the binding is dimin-
ished. The six-atom saturated ether chain restores some
NK2 activity, however, NK1 binding is poor for the
unsaturatedanalogue (entries 30 and31). Interestingly,
the allyl ether with a seven-atom chain (entry 32)
restores NK2 affinity. The secondary amine (entry 33)
suffers reduced NK2 binding, however, by switching to
the methyl amine derivative (entry 34) the dual activity
is maintained. The corresponding retroamide (entry 35)
is very active in the NK1 assay, however, NK2 binding is
poor. The sulfonamide (entry 36) shows very poor activ-
ity at both receptors. The carbamate (entry 37) looked
promising, as did the urea (entry 38), however, stability
problems upon exposure to low pH temperedinterest in
this series. The more stable reversedcarbamate (entry
39) shows excellent dual activity. The corresponding
amide analogues (entries 40 and 41) improves the dual
activity to single digit affinity at both receptors. From
our study of the different linkages between the oxime
130
22
42
6
167
286
435
37
32
35
38
39
5
6
8
11
andthe 3,5-bis(trifluoromethyl)phenyl moiety, the NK
andNK receptors are highly sensitive to this region.
Although several linkages have been identified which
modulate the relative selectivity for the NK1 or NK2
receptors, the amide is the most promising linker identified
for dual active antagonists.
1
40
41
2
4
2
9
2

144
N.-Y. Shih et al. / Bioorg. Med. Chem. Lett. 12 (2002) 141–145
Table 3. Structure–activity relationships for amide analogues
Compd
Ki (nM)
Compd
Ki (nM)
Compd
Ki (nM)
NK1
NK2
NK1
NK2
NK1
NK2
41
4
9
46
9
6
51
1
2
42
43
44
45
33
58
10
12
54
4
47
48
49
50
2
2
2
7
3
1
6
4
52
53
54
55
5
13
5
2
3
8
4
2
7
2
Subsequent second-generation exploration of the SAR
of the amide linked series via parallel and targeted
synthesis results in a number of potent NK1/NK2 dual
antagonists (Table 3). Utilization of the ortho-methoxy-
phenyl substitution pattern (entry 42) present in the
Pfizer series of NK1 antagonists7 results in diminished
activity in our oxime class. The N-methyl derivative
(entry 43) shows improvedNK 2 binding affinity. Several
analogues which contain an alternate aryl group for the
3,5-disubstituted phenyl represent novel replacements
for this group not only for selective NK antagonists, but
for dual NK antagonists as well. The thiophene and
aminonaphthyl amides have surprisingly good affinity
for both receptors (entries 44 and45). A similar active
series was foundfor the pyridine class (entries 46–50).
Aminopyridines (entries 47 and 48) are particularly
potent dual NK antagonists. Finally, several 3,5-di-
substituted benzamides are potent dual antagonists
(entries 51–55). The 3,5-dichlorophenyl substitution is
very promising in the NK binding assays; and many
second generation derivatives of these compounds dis-
play excellent biological activity in more advanced
assays (data not shown).
dual activity. The generality of the 3,5-disubstituted
phenyl substitution has been confirmed, in both the
ether andamide series, andseveral novel aryl groups
outside of this substitution pattern have been discovered
which have excellent affinity for both receptors.
Acknowledgements
The authors wish to thank Mr. C. RichardandMr. Z.
Zhan for performing the NK binding studies.
References and Notes
1. (a) Barnes, P. J. Lancet 1986, 1, 242. (b) Barnes, P. J.;
Chung, F. K.; Page, C. P. Pharmacol. Rev. 1988, 40, 49.
2. Reichard, G. A.; Ball, A. T.; Aslanian, R.; Anthes, J. C.; Shih,
N.-Y.; Piwinski, J. P. Bioorg. Med. Chem. Lett. 2000, 10, 2329.
3. 3,5-Bis(trifluoromethyl)phenylpropionic acid( n=1) was pre-
paredfrom 3,5-bis(trifluoromethyl)benzyl bromide using the 2
carbon homologation methodby Meyers: Meyers, A. I.; Nabeya,
A.; Adickes, H. W.; Politzer, I. R.; Malone, G. R.; Kovelesky, A.
C.; Nolen, R. L.; Portnoy, R. C. J. Org. Chem. 1973, 38, 36.
Further homologation to 3,5-bis(trifluoromethyl)phenylbutyric
acid( n=2) was performedby tosylation, idsplacement with
KCN followed by standard acidic hydrolysis.
4. 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 usedas
the ligands for the NK1 andNK 2 receptor assays respectively,
at the experimentally derived Kd’s. Ki’s were obtainedaccord-
ing to the Cheng andPrussoff equation.
Summary
A thorough SAR study of the benzylic ether region of
our leadoxime dual NK 1/NK2 antagonist revealedsev-
eral modifications, which results in more potent selec-
tive and dual antagonists. A study of different linkers
between the oxime andthe terminal aryl group identi-
fiedthe amide moiety as an improvement over the ether
in the original leadoxime 1 with regardto maximizing

N.-Y. Shih et al. / Bioorg. Med. Chem. Lett. 12 (2002) 141–145
145
5. The 3,5-disubstituted pharmacophore for the NK1 receptor,
discovered by Merck, has been reported in a number of benzyl
ether NK1 receptor antagonists, see: 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.
6. Burkholder, T. P.; Kudlacz, E. M.; Le, T.-B.; Knippenberg,
R. W.; Shatzer, S. A.; Maynard, G. D.; Webster, M. E.;
Horgan, S. W. Bioorg. Med. Chem. Lett. 1996, 6, 951.
7. Desai, M. C.; Lefkowitz, S. L.; Thadeio, P. F.; Longo,
K. P.; Snider, R. L. J. Med. Chem. 1992, 35, 4911.