Diphenylsalfone muscarinic antagonists: Piperidine derivatives with high M2 selectivity and improved potency

Article abstract of DOI:10.1016/S0960-894X(00)00437-6

Piperidine analogues of our previously described piperazine muscarinic antagonists are described. Piperidine analogues show a distinct structure-activity relationship (SAR) that differs from comparable piperazines. Compounds with high selectivity and impr

Full text of DOI:10.1016/S0960-894X(00)00437-6

Bioorganic & Medicinal Chemistry Letters 10 (2000) 2209±2212
Diphenylsulfone Muscarinic Antagonists: Piperidine Derivatives
with High M2 Selectivity and Improved Potency
William Billard, Herbert Binch, III, Kristen Bratzler, Lian-Yong Chen,
Gordon Crosby, Jr., Ruth A. Du€y, Sundeep Dugar, Jean Lachowicz, Robert McQuade,
Pradeep Pushpavanam, Vilma B. Ruperto, Lisa A. Taylor and John W. Clader*
Schering-Plough Research Institute, 2015 Galloping Hill Road, Kenilworth, NJ 07033, USA
Received 5 June 2000; accepted 27 July 2000
AbstractÐPiperidine analogues of our previously described piperazine muscarinic antagonists are described. Piperidine analogues show a
distinct structure±activity relationship (SAR) that di€ers from comparable piperazines. Compounds with high selectivity and improved
potency for the M2 receptor have been identi®ed. The lead compound, 12b, increases acetylcholine release in vivo. Compounds of this class
may be useful for the treatment of cognitive disorders such as Alzheimer's disease (AD). # 2000 Elsevier Science Ltd. All rights reserved.
Introduction
We have previously described a series of piperazine
diphenylsulfoxides typi®ed by 1, that are potent and
selective antagonists of the M2 receptor.6 As a part of
our on-going follow-up to this discovery, we now
describe our work in a related piperidine series of M2
antagonists, 2. Despite the obvious structural similari-
ties, the piperidine series shows a distinct structure±
activity relationship (SAR) which di€ers from the
piperazine series in many respects but has nonetheless
produced compounds with improved potency and high
selectivity for the M2 receptor.
Alzheimer's disease (AD) is a neurodegenerative disease
characterized by a profound cognitive impairment that
progresses eventually to an inability to function inde-
pendently and ultimately to death. One of the consistent
®ndings in brains of AD patients is loss of cholinergic
markers, including levels of acetylcholine (ACh).1 The
cholinergic approach to treatment of AD involves
counteracting this loss in cholinergic activity by pharma-
cologic intervention to increase cholinergic transmission.
Currently available cholinergic therapy accomplishes
this by inhibition of acetylcholinesterase, the major enzyme
responsible for the degradation of ACh.2 Cholinesterase
inhibitors such as Aricept1 have shown moderate e-
cacy in AD patients but also produce dose-limiting peri-
pheral cholinergic side e€ects.3 Acetylcholine levels can
also be increased by blocking central presynaptic M2
receptors. Centrally acting muscarinic M2 antagonists
may improve cognition without the side e€ects asso-
ciated with other cholinergic approaches provided there
is sucient selectivity for the M2 receptor over other
muscarinic receptors, particularly the M1 receptor,
which mediates the cognitive e€ects of acetylcholine.4,5
Very few compounds with the requisite selectivity are
known, and many of these do not cross the blood±brain
barrier e€ectively.
Chemistry7
Piperidine analogues of 1 were prepared as shown
in Scheme 1. Aldol condensation of compound 38
*Corresponding author. Tel.: +1-908-740-3478; fax: +1-980-740-4563;
e-mail: john.clader@spcorp.com
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00437-6

2210
W. Billard et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2209±2212
followed by reduction of the resulting ole®n provided 4.
This was oxidized to the corresponding sulfoxide 5 or
sulfone 6 using either one equivalent or excess sodium
perborate.
Methylidine, ketone, and methylated analogues were
prepared as shown in Scheme 2. Fluorophenyl ketone 8
was alkylated with cyclohexenyl bromide followed by
reduction to give 9. Reaction with the sodium salt of
Scheme 1. Synthesis of piperidine analogues of 1: (a) 1-cyclohexyl-4-piperidinone, NaOEt/EtOH, re¯ux; (b) Mg/MeOH; (c) 1.05 equiv
.
NaBO3 4H2O; (d) 2.7 equiv NaBO3 4H2O.
.
Scheme 2. Synthesis of methylidine, ketone, and methylated analogues of 1: (a) 3-bromocyclohexene; (b)ꢀH2 Pd±C 95%; (c) 4-methoxythiophenol,
.
.
NaH 75%; (d) 1.04 equiv NaBO3 4H2O 92%; (e) 2.7 equiv NaBO3 4H2O 55%; (f) Tebbe reagent, THF, 0 C, 51%; (g) MeMgBr/CeCl3; (h) Et3SiH,
TFA.

W. Billard et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2209±2212
2211
4-methoxythiophenol converted 9 to a diarylsul®de
ketone, which served as a common intermediate for
other compounds as shown.
less selective, but unlike in the piperazine series, the
decrease in selectivity is due almost entirely to an
increase in potency at M1. Ketone analogues 14a and
14b show a similar preference for one sulfoxide enan-
tiomer. The preferred enantiomer in the ketone series
14b is only slightly less selective than 12b but is about
60Â less potent. Sulfones 13 and 15 are consistently less
selective than sulfoxides.
Racemic sulfoxides were resolved via HPLC using a
Chiracel OJ column.10
Results
The e€ect of 12b on acetylcholine release in rat brain
was investigated using microdialysis as previously
described for 1b6,13 (Fig. 1). After ip administration, 12b
caused a rapid and sustained increase in acetylcholine
release comparable to what was observed with 1b. Peak
acetylcholine levels were reached about 30 min after
dosing, and signi®cant increases over baseline were
maintained for at least 2 h. Peak acetylcholine levels
were signi®cantly lower after oral administration and
were achieved about 2 h after dosing, suggesting slow
absorption of compound. However, acetylcholine levels at
2 h were comparable after either po or ip administration.
Our initial e€orts focused on direct analogues of 1 and
related compounds in the piperidine series (Table 1). In
earlier work, we observed that both the benzylic nitrile
and the (S)-sulfoxide of 1 were important for potency
and selectivity for the M2 receptor. These e€ects did not
carry over to the analogous piperidine series. For
instance, (R)-sulfoxide 1a is signi®cantly less selective
than its (S)-sulfoxide isomer 1b due to a small loss in M1
potency but a larger loss in M2 potency for the (R)-
enantiomer versus the (S)-isomer. By contrast, the sulf-
oxide enantiomers of the corresponding piperidine 5a
and 5b show comparable potency and selectivity, and
both are less potent and selective than 1b. Changing the
benzylic nitrile to a methyl group 10 signi®cantly
increases potency in the piperidine series but has only a
modest impact on potency in the piperazine series, 7. In
all cases, benzylic methyl derivatives are less selective
than benzylic nitriles, and sulfones are less selective than
sulfoxides, although in the piperidine series, sulfone 6 is
more potent than either sulfoxide enantiomer.
In summary, we have identi®ed a series of piperidines
that are potent inhibitors of the muscarinic M2 receptor
Table 2. Methylidine and ketone derivatives
We investigated a number of alternative substituents at
the benzylic position, motivated in part by a desire to
limit the chirality of compounds. Among these, the
most promising was a series of methylidine and related
ketone derivatives (Table 2). The methylidine com-
pounds are 10±100Â more potent than the benzylic
methyl or nitrile analogues but display SARs which
are similar to the piperazine series. Thus, (S)-sulfox-
ide 12b is ꢁ50Â more potent than 1b but shows
comparable selectivity versus the M1 receptor. The
corresponding (R)-sulfoxide isomer 12a is signi®cantly
Muscarinic binding Ki (nM)
Compound
Y
X
S
M1
M2
M3
M4
M5
NT
11
CH2
CH2 SO (R)
CH2 SO (S)
CH2
O
O
4.4
0.68
2.14
2.8
0.37
0.03
0.057
0.17
10.8
3.58
3.07
NT
0.13
0.38
0.48
20.1
27.8
13.76
NT
12a
12b
13
14a
14b
15
0.09 NT
0.32 NT
0.33 NT
15.2
17.9
SO2
SO (R)
69.5
NT
NT
SO (S) 103.8
SO2 31.09
Table 1. Piperidine analogues of 1
O
9.05 NT
Muscarinic binding Ki (nM)a
Compound
X
Y
R
M1
M2
M3
M4
M5
1a
1b
2
SO (R)
SO (S)
SO2
N
N
N
CN 253.3 17.3
CN 111.9
CN 71.8
NT 18.4
2.78 29.4 14.8 308.6
7.3 6.8 18.6 NT
NT
5a
5b
6
7
10
SO (R) CH CN 136.3
SO (S) CH CN 113.1
SO2 CH CN 16.4
7.14 58.8 45.05 NT
7.67 37.04 12.74 NT
1.3
1.0
NT NT
0.97 2.2
NT
NT
1.66
SO
SO
N CH3
CH CH3
6.0
0.53 0.14 0.16 0.07
Figure 1. E€ect of 12B on ACh release from striatum of conscious rat
following po and ip administration.
aAssay procedures are described in refs 11 and 12.

2212
W. Billard et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2209±2212
and show up to 38Â selectivity versus the M1 receptor.
Compound 12b increases acetylcholine release in rats,
indicating that this class of compounds could be useful
in treating the cognitive de®cit of Alzheimer's disease.
As both 1b and 12b show signi®cant activity at the M3
receptor, our on-going follow-up in both series is
focused on improving selectivity versus the other mus-
carinic receptors.
6. Kozlowski, J. A. Bioorg. Med. Chem. Lett. in press.
7. All new compounds were characterized using 1H NMR and
FABHRMS and showed spectral properties consistent with
the assigned structure. All compounds submitted for biologi-
cal testing were chromatographically homogeneous based on
either TLC or HPLC.
8. Petit, L.; Sonnier, M. Bull. Soc. Chim. Fr. 1974, 477.
9. Herdon, J. L.; Ismaiel, A.; Ingher, S. P.; Teitler, M.; Glen-
non, R. A. J. Med. Chem. 1992, 26, 4903.
10. Sulfoxide enantiomers were separated using a preparative
Chiracel OJ column, typically eluting with 10±20% 2-propa-
nol/hexane. The ®rst eluting enantiomer, designated (a), was
tentatively assigned the (R) stereochemistry by analogy to the
order of elution of 1b and its enantiomer, whose stereo-
chemistry was unambiguously assigned via asymmetric synth-
esis as described in ref 6. Although the assignment of
stereochemistry via HPLC elution order is tentative, in cases
where signi®cant di€erences in activity were seen, the more
selective enantiomer was consistently the one assigned the (S)
stereochemistry.
Acknowledgements
We are deeply indebted to Drs. A. K. Ganguly, Michael
Green, Joel Berger, William Greenlee, and Catherine
Strader for many helpful discussions and their con-
tinued support of this project. We also thank Dr.
Wayne Vaccaro for the procedure for synthesis of
methylidine derivatives using the Tebbe reagent.
11. Muscarinic receptor binding was determined using cloned
human M1±M5 receptors as previous described (ref 12). Com-
petition binding experiments were performed using 3H-QNB
as the radioligand. IC50 values were determined in triplicate
and Ki values were derived from IC50 values using the Cheng±
Pruso€ equation.
References and Notes
1. Ladner, C. J.; Lee, J. M. J. Neuropathol. Exp. Neurol. 1998,
57, 719.
2. For a review of cholinergic approaches to the treatment of AD,
see Clader, J. W. Curr. Opin. Drug Discov. Develop. 1999, 2, 311.
3. Nordberg, A.; Svensson, A.-L. Drug Safety 1998, 19, 465.
4. Doods, H. N. Drugs Fut. 1995, 20, 157.
12. Lachowicz, J. E.; Lowe, D.; Du€y, R. A.; Ruperto, V.;
Taylor, L. A.; Guzik, H.; Brown, J.; Berger, J. G.; Tice, M.;
McQuade, R.; Kozlowski, J.; Clader, J.; Strader, C. D.; Mur-
golo, N. Life Sciences 1999, 64, 535.
5. Quirion, R.; Wilson, A.; Rowe, W.; Aubert, I.; Doods, H.;
Parent, A.; White, N.; Meany, M. J. J. Neurosci. 1995, 15, 1455.
13. Billard, W.; Binch, H.; Crosby, G.; McQuade, R. D. J.
Pharm. Exper. Therap. 1995, 273, 273.