2-Aryl indole NK1 antagonists: Optimisation of the amide substituent

Article abstract of DOI:10.1016/S0960-894X(01)00616-3

The in vivo properties of a series of 2-arylindole NK₁ antagonists have been improved, by modification of the amide substituent. The 1-(2-methoxyphenyl)piperazine amide was identified as a major area of metabolism in the lead compound 1. Replacement of this amine moiety by a 4-benzyl-4-hydroxypiperidine resulted in a compound 18 with reduced clearance and improved central duration of action.

Full text of DOI:10.1016/S0960-894X(01)00616-3

Bioorganic & Medicinal Chemistry Letters 11 (2001) 3031–3034
2-Aryl Indole NK1 Antagonists: Optimisation of the Amide
Substituent
Duncan Shaw,a,* Gary G. Chicchi,c Jason M. Elliott,a Marc Kurtz,c
Denise Morrison,a Mark P. Ridgill,a Nicola Szeto,a Alan P. Watt,a Angela R. Williamsb
and Christopher J. Swaina
aDepartment of Medicinal Chemistry, Merck Sharp & Dohme Research Laboratories, Neuroscience Research Centre,
Terlings Park, Eastwick Road, Harlow, Essex CM20 2QR, UK
bDepartment of Pharmacology, Merck Sharp & Dohme Research Laboratories, Neuroscience Research Centre, Terlings Park,
Eastwick Road, Harlow, Essex CM20 2QR, UK
cDepartment of Biochemistry, Merck Research Laboratories, 126 E. Lincoln Ave, Rahway, NJ 07065, USA
Received 11 June 2001; revised 24 July 2001; accepted 7 September 2001
Abstract—The in vivo properties of a series of 2-arylindole NK1 antagonists have been improved, by modification of the amide
substituent. The 1-(2-methoxyphenyl)piperazine amide was identified as a major area of metabolism in the lead compound 1.
Replacement of this amine moiety by a 4-benzyl-4-hydroxypiperidine resulted in a compound 18 with reduced clearance and
improved central duration of action. # 2001 Published by Elsevier Science Ltd.
Neurokinin 1 (NK1) antagonists have long been descri-
bed as potential therapeutic agents and numerous small
molecule ligands have been described. The progress in
this area has been recently reviewed.1
To investigate whether the high clearance rate was due
to hepatic metabolism, a rat liver microsome study was
conducted.5 This showed that the primary routes of
metabolism were initial O-demethylation of the aryl
methyl ether and N-dephenylation of the arylpiperazine
group. There is extensive literature precedent for this
process in other related 2-methoxyphenylpiperazine
systems (Scheme 1).6
We have recently reported our studies of the optimisa-
tion of the indole core7 which lead to the identification
of the N-methyl-5-chloro-2-(4-chlorophenyl)indole,
exemplified by 2 which showed improved affinity for the
hNK1 receptor (IC50 0.28Æ0.14 nM).
The 2-arylindole 1 was identified as an interesting lead
compound from screening of a combinatorial library,
prepared using the chemistry described by Hutchins et
al.2 The profile of this lead was promising in that it
showed good hNK1 binding (hNK1 IC50=1.0Æ0.20
nM) and modest in vivo activity in the gerbil foot tap-
ping assay3 (46% inhibition at 3 mg/kg iv at the 0 h.
time point).
The pharmacokinetic profile of 1 in rats showed great
scope for improving the activity of this lead in vivo
(Table 1). Although the compound distributes out of
body water, it is compromised by an extremely high
clearance rate.4 This leads to a short half life and low
oral bioavailability (3%).
The
desired
N-methyl-5-chloro-2-(4-chlorophenyl)
indole acid was synthesised using the chemistry depicted
in Scheme 2. This allowed access to a wide range of
amides by using a simple peptide coupling modified to
be applicable to rapid analogue synthesis.
*Corresponding author. Fax: +44-1279-440390;
e-mail: duncan_shaw@merck.com
0960-894X/01/$ - see front matter # 2001 Published by Elsevier Science Ltd.
PII: S0960-894X(01)00616-3

3032
D. Shaw et al. / Bioorg. Med. Chem. Lett. 11 (2001) 3031–3034
Modification of the amide substituent began with
investigation of piperazine amides in order to find
potential replacements for the 2-methoxyphenyl group.
Table 2. Variation of biological properties with the introduction of
aryl piperazine substituents
The initial results of this study were encouraging; rea-
sonable binding affinity was achieved with a range of
lipophilic substituents (Table 2). For example, the ben-
zyl piperazine amide 3 was equipotent with the lead
structure 2. Closer examination of the data identified a
requirement for a lipophilic substituent of clearly defined
size. The preferred size of this substituent was well illu-
strated by the fact that whilst isopropyl 4 appeared to
be too small to give good binding affinity, cyclopropyl-
methyl 5 exhibited good potency. The cyclohexylmethyl
6 substituent appeared to be slightly too large for the
lipophilic pocket, in contrast to cyclohexyl 7 which gave
high affinity. Another feature was that alkyl substituents
containing quaternary carbon atoms gave moderate
affinity (e.g., neo-pentyl 8) whilst branched alkyl groups
gave good binding (e.g., 3-methylbutyl 9). The N-methyl-
phenylglycinamide 10 exhibited extremely good binding
affinity whereas the N-phenylglycinamide 11 was much
weaker; again this was consistent with a lipophilic
pocket which does not tolerate a polar NH group.
R
R1
hNK1
IC50 (nM)a
R
R1
hNK1
IC50 (nM)a
2
Me 0.28 (Æ0.14)
3
Me 0.30 (Æ0.14)
4
6
8
Me
13 (Æ6)
5
7
Me 0.78 (Æ0.29)
Me 0.35 (Æ0.18)
Me 0.34 (Æ0.06)
Me 2.9 (Æ1.8)
Me 2.0 (Æ1.4)
9
Table 1. Pharmacokinetic profile of 1 in rat
10 Me 0.13 (Æ0.08)
11
H
30 (Æ3.2)
iv dose
po dose
Half life
1.5 mg/kg
1.5 mg/kg
1.4 h
aDisplacement of [125I]-labelled substance P from the cloned hNK1
receptor expressed in CHO cells (n=3).8
Mean residence time
Plasma clearance rate
Volume of distribution (steady state)
Oral bioavailability
Cmax
0.9Æ0.1 h
179Æ75 mL/min/kg
9.5Æ4.3 L/kg
3Æ2%
2Æ1 ng/mL
1.1Æ0.9 h
Tmax
Table 3. Variation of in vivo activity in the gerbil foot tapping assay
with the introduction of piperazine substituents
R
hN
Gerbil foKot tapping %inhibitionb
1
IC50 (nM)a
t=0 h (mg/kg iv) t=2 h (mg/kg iv)
3
7
0.30 (Æ 0.14)
0.13 (Æ0.08)
100@3
17@3
Scheme 1.
100@3
87@1
7@3
6@1
10 0.35 (Æ0.18)
aDisplacement of [125I]-labelled substance P from the cloned hNK1
receptor expressed in CHO cells (n=3).8
bInhibition of foot-tapping by iv administration of test compound
immediately prior to icv infusion of GR73632. The duration of foot-
tapping was recorded for 5 min and is expressed as a percentage inhi-
bition of values observed in vehicle-treated animals.3
Scheme 2. (i) Chlorobenzene, AlCl3; (ii) 4-chlorophenylhydrazine,
EtOH; (iii) TFA; (iv) NaH, MeI, THF; (v) KOH, MeOH.

D. Shaw et al. / Bioorg. Med. Chem. Lett. 11 (2001) 3031–3034
3033
As illustrated in Table 3, some of these modifications of
the piperazine substituent gave an improved in vivo
profile at t=0 h in the gerbil assay. This indicated that
good brain penetration had been achieved. The effects,
however, were of short duration because activity
returned to baseline at t=2 h suggesting that rapid
clearance of these compounds remained a problem.
acting as hydrogen bond acceptors were good
piperazine replacements in terms of binding affinity.
Another example of this is the 4-(N-phenyl)piperidine
16 which had moderate affinity whereas the corre-
sponding propionamide 17 (a potential hydrogen bond
acceptor) was highly potent, and brain penetrant
although it exhibited poor duration of action. The exact
position of the heteroatom also had a profound effect
on NK1 binding. The 4-benzyl-4-hydroxy piperidine 18
was an excellent replacement for the N-benzylpiperazine
but the 4-hydroxymethyl-4-phenyl piperidine 19 showed
decreased binding affinity.
We reasoned that this deficiency was probably due to
metabolic N-dealkylation of the piperazine and we
therefore began to consider further the replacement of
the piperazine ring itself.
At this point, it was noted that good affinity was
retained when the N-benzylpiperazine was replaced with
a 4-N-benzylpiperazin-3-one group 12 (Table 4). This
illustrated that it was not necessary to have a basic
nitrogen atom to achieve good binding. However,
exchanging the nitrogen of the piperazine with a carbon
atom to give the corresponding 4-benzylpiperidine 13
resulted in a significant reduction in binding affinity
suggesting the possible loss of a hydrogen bond. Fur-
thermore, both the 4-acetyl-4-phenylpiperidine 14 and
4-fluoro-4-benzylpiperidine 15 with groups capable of
The best replacement identified for the benzylpiperazine
in terms of in vivo activity was the 4-benzyl-4-hydroxy-
piperidinol 18, which was potent in vivo at t=0 h and
showed good activity at the 2-h time point. 18 was the
first compound in this series to achieve good central
duration of action in this assay (Table 5).
Table 5. Variation of in vivo properties in the gerbil foot tapping
assay with the introduction of non-piperazine amide substituents
Table 4. Variation of binding affinities with the introduction of var-
ious non piperazine amide substituents
HNR2
hNK1
IC50 (nM)a
Gerbil foot tapping %inhibitionb
t=0 h (mg/kg iv) t=2 h (mg/kg iv)
HNR2
R1
hNK1
IC50 (nM)a
HNR2
R1
hNK1
IC50 (nM)a
17 0.16 (Æ 0.11)
100@3
50@3
12 Me
13 Me
0.39 (Æ0.08)
2.2 (Æ 0.1)
18 0.21 (Æ0.01) ID50=0.3 mg/kg ID50=0.9 mg/kg
aDisplacement of [125I]-labelled substance P from the cloned hNK1
receptor expressed in CHO cells (n=3).8
14 Me 0.53 (Æ 0.14)
16 Me 1.3 (Æ 0.26)
18 Me 0.21 (Æ0.01)
15 Me 0.55 (Æ 0.36)
bInhibition of foot-tapping by iv administration of test compound
immediately prior to icv infusion of GR73632. The duration of foot-
tapping was recorded for 5 min and is expressed as a percentage inhi-
bition of values observed in vehicle-treated animals. The ID50 was
calculated by non-linear least-squares regression analysis of mean
data.3
17 Me 0.16 (Æ 0.11)
Table 6. Pharmacokinetic profile of 18 in rat
iv dose
po dose
Half life
1 mg/kg
1 mg/kg
3.1 h
Mean residence time
Plasma clearance rate
Volume of distribution (steady state)
Oral bioavailability
Cmax
1.8Æ0.3 h
31Æ2 ml/min/kg
3.5Æ0.7 L/kg
3Æ1%
19
H
55 (Æ 11)
4Æ1 ng/mL
1.2Æ0.8 h
aDisplacement of [125I]-labelled substance P from the cloned hNK1
receptor expressed in CHO cells (n=3).8
Tmax

3034
D. Shaw et al. / Bioorg. Med. Chem. Lett. 11 (2001) 3031–3034
The pharmacokinetic properties of 18 in rat (Table 6),
showed some interesting differences from that of the
original lead 1. Since the clearance of 18 had been
reduced to 50% of liver blood flow, although the
volume of distribution has decreased this still leads to a
modest increase in half life. Disappointingly, however,
the oral bioavailability of 18 (3%) was not improved
relative to 1. It is not known whether this is due to
metabolism, poor absorption or both. Absorption is
possibly compromised for this compound due to poor
aqueous solubility (<0.1 mg/mL) (as the molecule has
no basic centre). The replacement of the indole core
with an azaindole offered a potential solution to this
problem.7a
References and Notes
1. Seward, E. M.; Swain, C. J. Exp. Opin. Ther. Pat. 1999, 9,
571.
2. Hutchins, S. M.; Chapman, K. T. J. Tetrahedron Lett.
1996, 37, 4869.
3. Rupniak, N. M. J.; Tattersall, F. D.; Williams, A. R.;
Rycroft, W.; Carlson, E. J.; Cascieri, M. A.; Sadowski, S.; Ber,
E.; Hale, J. J.; Mills, S. G.; MacCoss, M.; Seward, E.;
Huscroft, I.; Owen, S.; Swain, C. J.; Hill, R. G.; Hargreaves,
R. J. Eur. J. Pharmacol. 1997, 326, 201.
4. This clearance figure is greater than the liver blood flow
rate in rat, suggesting that hepatic metabolism may be augmented
by significant systemic metabolism (i.e., lung and kidney).
5. 1 was incubated at 10 mM concentration with rat liver
microsomes and a S9 preparation in conditions suitable for
oxidative metabolism. Samples were analyzed by LC–UV and
LC–MS to identify metabolic pathways.
In this study we were able to identify and decrease the
metabolic liabilities of 1, and identify compound 18
which showed improved iv pharmacokinetics.
6. Enereille, A.; Pognat, J. F.; Lartigue-Mattei, C.; Chabard,
J. L.; Busch, N.; Berger, J. A. Eur. J. Drug Metab. Pharmaco-
kinet. 1991, 16, 161.
7. (a) Cooper, L.; Chicchi, G. G.; Dinnell, K.; Elliott, J. M.;
Hollingworth, G. J.; Kurtz, M. M.; Locker, K. L.; Morrison,
D.; Shaw, D. E.; Tsao, K.-L.; Watt, A. P.; Williams, A. R.;
Swain, C. J. Bioorg. Med. Chem. Lett. 2001, 11, 1233. (b)
Dinnell, K.; Chicchi, G. G.; Dhar, M. J.; Elliott, J. M.; Hol-
lingworth, G. J.; Kurtz, M. M.; Ridgill, M. P.; Rycroft, W.;
Tsao, K.-L.; Williams, A. R.; Swain, C. J. Bioorg. Med. Chem.
Lett. 2001, 11, 1237.
By alteration of the amide substituent we were able to
demonstrate that considerable variation in this area of
the molecule was tolerated. The requirement for a lipo-
philic group of defined size and a hydrogen bond donor
in a specific region of space was identified. By looking at
the in vivo properties of the more potent compounds we
were able to select for improved activity in the gerbil
foot-tapping assay.3 This allowed us to identify 18, which
showed high binding affinity good brain penetration and
significantly improved central duration of action.
8. Cascieri, M. A.; Ber, E.; Fong, T. M.; Sadowski, S.; Bansal,
A.; Swain, C.; Seward, E.; Frances, B.; Burns, D.; Strader,
C. D. Mol. Pharmacol. 1992, 42, 458.