Aryl sulfones: A new class of γ-secretase inhibitors

Article abstract of DOI:10.1016/j.bmcl.2004.12.017

The development of a novel series of 4-aryl, 4-phenylsulfonyl cyclohexananone-derived γ-secretase inhibitors for the potential treatment of Alzheimer's disease is described.

Full text of DOI:10.1016/j.bmcl.2004.12.017

Bioorganic & Medicinal Chemistry Letters 15 (2005) 2685–2688
Aryl sulfones: a new class of c-secretase inhibitors
Martin Teall,a,* Paul Oakley,a Timothy Harrison,a Duncan Shaw,a
a
a
a
a
b
´
Euan Kay, Jason Elliott, Ute Gerhard, Jose L. Castro, Mark Shearman,
Richard G. Ballc and Nancy N. Tsouc
aDepartment of Medicinal Chemistry, Merck Sharp & Dohme Research Laboratories, The Neuroscience Research Centre,
Terlings Park, Eastwick Road, Harlow, Essex CM20 2QR, UK
bDepartment of Molecular and Cellular Neuroscience, Merck Sharp & Dohme Research Laboratories,
The Neuroscience Research Centre, Terlings Park, Eastwick Road, Harlow, Essex CM20 2QR, UK
cDepartment of Medicinal Chemistry, Merck Research Laboratories, Merck & Co., Inc., PO Box 2000, Rahway, NJ 07065, USA
Received 27 September 2004; revised 23 November 2004; accepted 8 December 2004
Available online 12 April 2005
Abstract—The development of a novel series of 4-aryl, 4-phenylsulfonyl cyclohexananone-derived c-secretase inhibitors for the
potential treatment of AlzheimerÕs disease is described.
ꢀ 2004 Elsevier Ltd. All rights reserved.
AlzheimerÕs disease (AD) is a progressive neurodegenera-
tive disease characterized by memory loss and cognitive
decline. AD is the most common form of dementia and
affects up to 50% of the population over 85 years of
age.1 The amyloid-b (Ab) peptide is the major compo-
nent of extracellular proteinaceous plaques found in
AlzheimerÕs brains, and soluble Ab has been shown to
be neurotoxic in vitro.2 Cleavage of amyloid precursor
protein (APP) by both b- and c-secretases3 releases
Ab peptides and it is these cleavage events that are
thought to play a key role in the neurodegenerative
pathways responsible for the progressive cognitive de-
cline observed in AD patients. Inhibition of the prot-
eases responsible for the unfavourable cleavage of
APP therefore becomes an attractive pharmaceutical
point of intervention, which could alter the underlying
pathophysiology4 of the disease rather than act as a pal-
liative treatment.
bon analogues for synthesis.7,8 Thus 4-chlorobenzeneth-
iol was reacted with 2,5-difluorobenzyl bromide
followed by oxidation of the sulfide to give the interme-
diate sulfone (5). This could be readily alkylated with
substituted bromides, and the esters reduced with lith-
ium aluminium hydride to give the alcohols (Scheme 1).
The unsubstituted sulfone (5) was a weak c-secretase
inhibitor as measured in a whole cell inhibition assay
using SH-SY5Y cells,6 however a significant improve-
ment in potency was observed as the size of the lipo-
philic group at the benzylic position was increased
(e.g., 6–7). Introduction of the ester (8) or the alcohol
(9) gave a slight reduction in potency, but potency was
re-established when the homologated alcohol was intro-
duced (10) (Table 2).
To explore the effects of restricting the conformational
freedom of the two aryl groups and to remove the chiral
centre, a geminally-fused cyclohexane ring was intro-
duced. The sulfone (5) underwent a double Michael addi-
tion followed by a Dieckmann cyclization to give the b-
keto ester (11),9 which underwent decarboxylation under
Krapcho conditions to give the 4-aryl-4-phenylsulfonyl
cyclohexanone (12, Scheme 2). The cyclohexanone was
reduced with NaBH4 to give a separable mixture of cis
and trans alcohols (13 and 14). The sulfone (5) also
underwent double alkylation with 1,5-diiodopentane to
give the unsubstituted cyclohexane (15).10
A recent published patent application5 exemplified a ser-
ies of sulfonamides (e.g., 1–4, Table 1) that inhibit the
production of Ab peptides in vitro.6
Based on the simplicity and relatively high potency of
these molecules we targeted the structurally novel car-
*
Corresponding author. Tel.: +1 279 440407; fax: +1 279 440390;
e-mail: martin_teall@merck.com
0960-894X/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.12.017

2686
M. Teall et al. / Bioorg. Med. Chem. Lett. 15 (2005) 2685–2688
Table 1. c-Secretase inhibition for sulfonamides 1–4
Entry
R
IC50 (nM)6
1
2
3
H
>10,000
1900
170
Me
iPr
4
27
IC50 concentration to inhibit production of 50% of Amyloid (Ab)
peptide.
Scheme 2. Reagents and conditions: (i) KOtBu, methyl acrylate, 60%;
(ii) NaCl/DMSO–H2O, 150 ꢁC, 30%; (iii) NaBH4, 80%; (iv) I(CH2)5I,
NaH, 29%.
Scheme 1. Reagents and conditions: (i) (a) Et3N, ClArSH, 90%, (b) 3-
t
chloroperoxybenzoic acid; (ii) R–Br, BuOK, 40–90% for 6, 7 and 8;
(iii) lithium aluminium hydride, THF, 36% for 9 and 10.
Table 3. c-Secretase inhibition for cyclic sulfones
Table 2. c-Secretase inhibition for sulfones 5–10
Entry
Cyclohexyl substitution
IC50 (nM)6
104
11
Entry
R
IC50 (nM)6
5
6
7
II
4408
2472a
70a
Me
iPr
12
13
14
15
22
10
8
253b
304b
35b
9
385
3
10
a Racemate.
b Racemic diastereoisomers (1:1).
The cyclohexyl keto ester (11) had improved potency
relative to the unsubstituted sulfone 5 (Tables 2 and
3). Removal of the ester gave the ketone (12) and pro-
vided a further increase in potency (IC50 22 nM). The
cis alcohol (13) retained potency, while the trans alcohol
(14) was a weaker inibitor. Intriguingly, the unsubsti-
tuted cyclohexane (15) proved to be the most potent
analogue prepared (IC50 3 nM).
2,5-difluoraryl group in either the axial or equatorial
position.
Oxidation of norbornene was accomplished with ruthe-
nium chloride and sodium periodate, and the resulting
bis-acid (16)11 reduced with lithium aluminium hydride
to give the diol (17). This was activated as the bis-mesyl-
ate (18) and used in a double alkylation with the sulfone
(5) to give a separable mixture of the bicyclic analogues
19 and 20 in low yield (24% and 4%, respectively).
The structures were confirmed by 1H NMR NOE
To further investigate the conformation of 15, two bicy-
clic analogues were synthesized (Scheme 3) that hold the

M. Teall et al. / Bioorg. Med. Chem. Lett. 15 (2005) 2685–2688
Table 4. c-Secretase inhibition for cyclic sulfons 15, 21–24
2687
O
O
S
R1
R3
R2
Entry
R1
R2
R3
IC50 (nM)5
15
21
22
23
24
F
F
H
H
F
F
H
F
H
F
Cl
Cl
Cl
Cl
H
3
55
249
461
98
clearly depicted the 2,5-difluoroaryl group in the axial
position (Fig. 1).14
To investigate the importance of the halogen substitu-
ents, a series of sulfones 21–24 were prepared (Table
4) following the general procedures outlined in Schemes
1 and 2.
Scheme 3. Reagents and conditions: (i) RuCl3, NaIO4, CHCl3–H2O,
rt, 70%; (ii) LiAlH4, THF, 65%; (iii) MeSO2Cl, Et3N, CH2Cl2, 0 ꢁC,
92%; (iv) KOtBu, NaH, THF, 5, 28%.
A decrease in potency was observed with removal of the
5-fluoro (21), the 2-fluoro substituent (22) or both (23).
Removal of the 4-chloro substituent in the aryl sulfone
ring, as in analogue (24) also resulted in a decrease in
potency (IC50 98 nM). All substituents are therefore
contributing to the high potency of the c-secretase inhibi-
tor 15.
experiments12. In analogue 19 the ortho-fluorophenyl
proton was shown to be in close proximity to the C-2
equatorial proton of the cyclohexane and the proton
on the ethylene bridge. For analogue 20 NOEÕs were ob-
served from the ortho-fluorophenyl proton to both the 2-
axial and 2-equatorial protons and the ortho-chlorophen-
yl proton also showed an NOE to the proton on the
ethylene bridge, indicating that the SO2Ar group exists
in the axial orientation (Scheme 3). In analogue 15
NOEÕs were observed between the ortho-fluorophenyl
proton and the 2-equatorial and 3-axial protons on the
cyclohexane. Taken together, these experiments suggest
that the more potent bicyclic analogues (19 IC50 12 nM)
and 15 (IC50 3 nM) possess the 2,5-difluoroaryl group in
the axial position. This suggestion is supported by the
work of Eliel and Manoharan13 who showed that the
larger phenyl group in 1-methyl-1-phenylcyclohexane
prefers to exist in the energetically favourable axial posi-
tion. An X-ray structure was obtained compound 15
In conclusion, we have described a series of novel, aryl
sulfone c-secretase inhibitors, which possesses a simple
pharmacophore, and requires only a three step synthesis
to prepare potent compounds (e.g., 15, IC50 3 nM). The
contribution of the halogen aromatic substituents was
investigated, with the 2,5-difluorophenyl and 4-chloro-
phenyl groups found to be essential to maintain high
potency. Finally, using NMR experiments, analogy with
structurally defined analogues (19 and 20), and an X-ray
it was concluded that the 2,5-difluoroaryl group resides
in the axial position in the active conformer of 15.
Acknowledgements
We thank Jonathan Wrigley, Dirk Beher, Earl Clarke
and Huw Lewis for their support in obtaining the data.
References and notes
1. Smith, D. W. WO 50391, 2000, CAN 133:207678; Hardy,
J.; Selkoe, D. Science 2002, 297, 353; Moore, C. L.; Wolfe,
M. S. Exp. Opin. Ther. Patents 1999, 9, 135; Jhee, S.;
Shiovitz, T.; Crawford, A. W.; Cutler, N. R. Exp. Opin.
Invest. Drugs 2001, 10, 593; Conde, S. Exp. Opin. Ther.
Patents 2002, 12, 503; Dominguez, D. I.; De Strooper, B.;
Annaert, W. Amyloid 2001, 8, 124.
2. Gong, Y.; Chang, L.; Viola, K. L.; Lacor, P. N.; Lambert,
M. P.; Finch, C. E.; Krafft, G. A.; Klein, W. L. Proc. Natl.
Acad. Sci. U.S.A. 2003, 100, 10417.
3. Price, D. L.; Sisodia, S. S. Ann. Rev. Neurosci. 1998, 21,
4790; Selkoe, D. J. Trends Cell Biol. 1998, 8, 447; Bayer, T.
Figure 1. X-ray crystal structure of compound 15.

2688
M. Teall et al. / Bioorg. Med. Chem. Lett. 15 (2005) 2685–2688
A.; Cappei, R.; Masters, C. L.; Beyreuther, K.; Multhaup,
G. Mol. Psychiat. 1999, 4, 524.
9. Sagami Chem. Res. Center. JP 56026847, CAN 95:150191;
Yashimoto, M.; Ishida, N.; Kishida, Y. Chem. Pharm.
Bull. 1972, 20, 2137.
4. Wu, J.; Tung. J. S. WO 28268, 1998, CAN 129:122870;
Audia, J.; Dressman, B. A. WO 66934, 1999, CAN
132:64531; 67219, 1999, CAN 132:64532; Zaczek, R.;
Olson, R. E. WO 28331, 2000, CAN 133:779; Kreft, F. A.
WO Patent 057252, 2002, CAN 137:125077; WO
103660A1, 2003, CAN 140:41906; Asberom, T.; Gusik,
H. S. WO 014075, 2003, CAN 138:18749; Josien, H.;
Cladder, J. W. WO 013527, 2003, CAN 138:170082.
5. Smith, D. W.; Munoz, B. WO 50391, 2000, CAN
133:207678.
6. Clarke, E. E.; Shearman, M. S. J. Neurosci. Methods 2000,
102, 61, All IC50 data shown are the geometric mean of
minimum of three independent results. Cell viability is
measured by a colorimetric cell proliferation assay (Cell-
Titre 96TMAQ assay, Promega).
10. Sulfone (5, 2.30 g, 7.61 mmol) and 1,5-diiodopentane
(1.24 mL, 8.37 mmol) in dimethylformamide (30 mL) were
treated with sodium hydride (0.67 g, 16.7 mmol, 60 % in
oil) portionwise over 30 min. Reaction was stirred for 8 h
at room temperature, diluted with water (60 mL) and
products extracted into ethyl acetate (3 · 60 mL). The
combined organic phase was washed with saturated
sodium chloride (30 mL), dried over magnesium sulfate
and evaporated. The product was obtained by purification
on silica gel (BDH) eluting with hexane–ethyl acetate
mixtures (0–10%) to obtain product (0.81 g, 29%) plus
uncyclized intermediate.
11. Ben, R. N.; Breau, L.; Bensimon, C.; Durst, T. Tetrahe-
dron 1994, 50, 6061.
7. Teall, M. R.; Churcher, I.; Dinnell, K.; Harrison, T.
Patent application WO 018543, 2003, CAN 137:310695.
8. Subsequent to this work the Bayer (Hendrix, M.; Bauman,
K. WO 059335, 2003, CAN 139:133337) and Daiichi
(Yasukochi, T.; Ito, M. WO 055850, 2003, CAN
139:101146) groups have published patent applications
exemplifying similar derivatives.
12. Samples were dissolved in chloroform-d (100% D). All
experiments were acquired on a Bruker AV 500 MHz
NMR instrument equipped with a 3 mm triple resonance
probe (1H/19F/13C) with Z gradients. A temperature of
300 K was maintained throughout the data acquisition.
13. Eliel, E.; Manoharan, M. J. Org. Chem. 1981, 46, 1959.
14. CCDC deposition number for 15 251386.