Molecular-modeling based design, synthesis, and activity of substituted piperidines as γ-secretase inhibitors

Article abstract of DOI:10.1016/j.bmcl.2005.02.006

Alzheimer's disease (AD) is a debilitating disease widely thought to be associated with the accumulation of beta amyloid (Aβ) in the brain. Inhibition of γ-secretase, one of the enzymes responsible for Aβ production, may be a useful strategy for the treatment of AD. Described below is a series of γ-secretase inhibitors designed from a scaffold identified by a ROCS [J. Comput. Chem. 1996, 17, 1653] search of the corporate database.

Full text of DOI:10.1016/j.bmcl.2005.02.006

Bioorganic & Medicinal Chemistry Letters 15 (2005) 1891–1894
Molecular-modeling based design, synthesis, and activity
of substituted piperidines as c-secretase inhibitors
Eric Gundersen,a,* Kristi Fan,b Kimberly Haas,b Donna Huryn,a J. Steven Jacobsen,c
Anthony Kreft,a Robert Martone,c Scott Mayer,a June Sonnenberg-Reines,c
Shaiu-Ching Sunc and Hua Zhouc
aChemical and Screening Sciences, Wyeth Research, CN 8000, Princeton, NJ 08543-8000, USA
bComputational Chemistry, Wyeth Research, CN 8000, Princeton, NJ 08543-8000, USA
cNeuroscience, Wyeth Research, CN 8000, Princeton, NJ 08543-8000, USA
Received 13 September 2004; revised 1 February 2005; accepted 2 February 2005
Abstract—AlzheimerÕs disease (AD) is a debilitating disease widely thought to be associated with the accumulation of beta amyloid
(Ab) in the brain. Inhibition of c-secretase, one of the enzymes responsible for Ab production, may be a useful strategy for the treat-
ment of AD. Described below is a series of c-secretase inhibitors designed from a scaffold identified by a ROCS [J. Comput. Chem.
1996, 17, 1653] search of the corporate database.
Ó 2005 Elsevier Ltd. All rights reserved.
AlzheimerÕs disease (AD) is a debilitating disease of the
elderly, which ultimately leads to death. The disease is
typically recognized by the gradual deterioration of
mental ability; AD is the most common form of demen-
tia—loss of memory—in the elderly.
protein (APP) by the sequential action of b- and
c-secretase. Although rare, cases of early onset of AD
have been attributed to genetic mutations in APP or in
presenilins 1 and 2, components of the c-secretase com-
plex, that lead to an overproduction of either total Ab or
its more aggregation-prone 42 amino acid isoform.6–8
Furthermore, people with DownÕs Syndrome possess
an extra copy of the chromosome that contains the gene
that encodes APP; these people have elevated Ab levels
and invariably develop AD later in life.9
AD is pathologically characterized by the presence of
two major lesions in the brain: extracellular amyloid
deposits of beta amyloid protein (Ab) in the form of pla-
ques and amyloid angiopathy; and intracellular neurofi-
brillary tangles of aggregated hyperphosphorylated tau
protein.2 Recent evidence suggests that elevated Ab lev-
els in the brain not only precede tau pathology but also
correlate with cognitive decline.3 Additional studies
have also shown that aggregated Ab is toxic to neurons
in cell culture4 and has a detrimental effect on memory.5
Together, this evidence implicates Ab in a causative role
in AD and suggests that reducing Ab levels is a viable
therapeutic strategy for the treatment of AD.
A plausible strategy for the treatment of AD suggested
from the above information is to inhibit Ab production
via the inhibition of c-secretase. Several c-secretase
inhibitors have been designed based upon the amino
acid sequence of the APP cleavage site.10 Importantly,
the c-secretase inhibitor N-[N-(3,5-difluorophenacetyl)-
L-alanyl]-S-phenylglycine tert-butyl ester (DAPT) has
recently been shown to reduce Ab protein levels in
mice brains in vivo after oral administration.11 Herein
is described a series of c-secretase inhibitors designed
by molecular modeling and exhibiting some promising
activity in our Ab40/Ab42 assay.
Ab is a 39–42 amino acid peptide that is produced from
a larger precursor protein called amyloid precursor
In our efforts to uncover new leads for the c-secretase
program, we chose to use the ROCS1 program to
find new scaffolds for lead development. The ROCS
program identifies molecules that have a similar
Keywords: AlzheimerÕs; c-Secretase inhibitors; Molecular modeling;
Piperidines.
*
Corresponding author. Tel.: +1 484 865 9202; fax: +1 484 865
6463; e-mail: gundere@wyeth.com
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.02.006

1892
E. Gundersen et al. / Bioorg. Med. Chem. Lett. 15 (2005) 1891–1894
three-dimensional shape. The similarity search requires
the input of a compound query to make comparisons
against a compound database. In our search, the degree
of shape similarity is scored by a Tanimoto coefficient.
Overlap of the hydroxamic acid of 2 with one of the aro-
matic rings in 1 (see Fig. 3) suggested the replacement of
the hydroxamic acid with a phenyl ring, satisfying the
first requirement of our pharmacophore. Secondly, the
replacement of the alkene moiety in 2 with a carbamate
added the requisite hydrogen bond acceptor. These two
modifications led us to compound 3, which we also be-
lieved would be relatively straightforward to synthesize
and would prove amenable to parallel synthesis if
needed. Minimization of 3 with the MMFF94 force field
and subsequent comparison with 1 (see Fig. 4) resulted
in good overlap while maintaining the pharmacophore
orientation and providing us with a potential new lead
for our c-secretase program.
As a starting point for the ROCS search, we used sulfon-
amide 1 (see Fig. 1), a recently reported potent inhibitor
of Ab production.12 A conformational search based on
the Monte Carlo method in Macromodel13 was per-
formed on 1. The MMFF94 force field14–18 was used
in the energy minimization steps. The low-energy con-
former of 1 was used as the query. Using a Tanimoto
coefficient cut-off of 0.7, the search of an in-house data-
base provided us with about 500 hits; one of these hits
was compound 2.
While hit 2 was inactive in our Ab40/Ab42 assay, a
sandwich ELISA employing 6E10 antibody for Ab cap-
ture and Ab40 or 42 C-terminal specific antibodies for
The hits were then analyzed to see which of them con-
tained the chemical features believed to be important
for c-secretase activity or which could be readily modi-
fied to contain those features. We have identified, in par-
ticular, three features thought to be important for c-
secretase inhibition: two hydrophobes about 4 A apart
˚
and a hydrogen bond acceptor about 6–8 A from the
hydrophobes (see Fig. 2).
˚
OMe
R1
HN
O
O
O
HO
O2S
OH
O2S
O
N
N
H
SO2
R2
N
N
Cl
Figure 3. Overlay of compound 2 with compound 1 (magenta)
generated from the ROCS software. Note the strong overlap of the
aromatic rings and the tail portions. Also, note the close proximity of
the hydroxamic acid to the aromatic hydrophobe of 1.
OEt
Cl
2
1
3-11
Figure 1. A ROCS search using the low-energy conformer of 1
identified hit 2 from the corporate database. Modification of hit 2
generated analogues 3–11.
Figure 4. Overlay of compound 3 with compound 1 (magenta); these
structures were minimized using the MMFF94 force field and then
compared by mapping the aromatic regions, the sulfones, and the
carbonyl oxygen of each molecule.
Figure 2. This figure depicts a schematic of the pharmacophore
overlayed on compound 1, shown as the low-energy conformer.

E. Gundersen et al. / Bioorg. Med. Chem. Lett. 15 (2005) 1891–1894
1893
detection,19 compound 3 gave good activity as an initial
O
N
lead (see Table 1). This result prompted the synthesis of
a small matrix of compounds to develop some SAR as
well as to validate the chemistry for a future library.
Eight additional compounds were prepared and their
activities in the Ab40/Ab42 assay are presented in Table
1. The data suggests that there is both a steric and elec-
tronic component to the structure–activity relationship.
Specifically, the data show that for R1, Cl > OMe > H
(see Table 1: compounds 3, 6, and 9). Alternatively,
the data show for R2, that Me > OMe > H, (see Table
1: compounds 3, 4, and 5).
Grignard
Reagent
HO
88% aq. HCO2H
R2
benzenethiols
65-92%
Et2O, 45-57%
OEt
N
O
O
OEt
12
13
R1
R1
oxone
S
O2S
MeOH, H2O
48-80%
Compound 3 was further tested in radiolabeled cellular
assays employing either CHO cells stably expressing a
wild-type recombinant human APP 695, or a human
APP reporter construct containing the Swedish KMto
NL mutation to determine the mechanism of action
for the series.21 The use of the wild-type construct gave
an EC50 of 1.03 lMfor A b reduction, while maintaining
constant levels of APP and aAPPs, suggesting that the
observed Ab reduction was not caused by alterations
in protein synthesis or secretion. Use of the Swedish mu-
tant reporter construct gave an EC50 of 0.53 lMfor A b
reduction, an EC50 of 0.64 lMfor elevation of the b C-
terminal fragment (Ct-99) and a 3.8-fold increase in the
level of Ct-99 over the control. Together the data con-
firms that the compound is acting via inhibition of c-
secretase.
R2
R2
N
N
O
OEt
O
OEt
14
3-11
Scheme 1. Synthetic scheme for the substituted piperidines.
that the compounds are acting via c-secretase inhibition.
However, activity exhibited by these compounds is only
moderate. WO 0208143523 describes a class of com-
pounds similar to this series for the treatment of AD
via inhibition of c-secretase. Ultimately, our findings
with this small library validated the ROCS search design
strategy. Current efforts are directed to using this tech-
nique to identify more novel scaffolds.
Compounds 3–11 were prepared by a three-step sequen-
ce outlined in Scheme 1.22 Addition of the appropriate
substituted phenyl Grignard reagent to 1-carbethoxy-
4-piperidone (12) afforded tertiary alcohols (13). Addi-
tion of the respective substituted thiol to a solution of
the alcohol in ꢀ88% formic acid gave the thioethers
(14). Oxidation of the thioethers with oxone afforded
the final products (3–11).
References and notes
1. ROCS: OpenEye Scientific Software, Inc., 3600 Cerrillos
Rd., Suite 1107 Santa Fe, NM87507, also see: Grant, J.
A.; Gallardo, M. A.; Pickup, B. T. J. Comput. Chem. 1996,
17, 1653.
2. Li, Q.; Fuller, S. J.; Beyreuther, K.; Masters, C. L. J.
Leukocyte Biol. 1999, 66, 567.
In summary a c-secretase inhibitor was designed using a
ROCS search; a small number of compounds were syn-
thesized to expand the SAR of this series. Data suggest
3. Naslund, J.; Haroutunian, V.; Mohs, R.; Davis, K. L.;
Davies, P.; Greengard, P.; Buxbaum, J. D. J. Am. Med.
Assoc. 2000, 283, 1571.
4. Small, D. H.; McLean, C. A. J. Neurochem. 1999, 73, 443.
5. Dodart, J.-C.; Mathis, C.; Ungerer, A. Rev. Neurosci.
2000, 11, 75.
6. Goate, A. M. Cell. Mol. Life Sci. 1998, 54, 897.
7. Edbauer, D.; Winkler, E.; Regula, J. T.; Pesold, B.;
Steiner, H.; Haass, C. Nat. Cell Biol. 2003, 5, 486.
8. Selkoe, D. J. Physiol. Rev. 2001, 81, 741.
Table 1. Summary of the in vitro inhibitory effects of compounds 3–11
on Ab40 and Ab42 in the ELISA assay19
Compound
R1
R2
EC50 (lM)
Ab40
EC50 (lM)
Ab42
9. Sabbagh, M. N.; Galasko, D.; Thal, L. J. AlzheimerÕs Dis.
Rev. 1998, 3, 1.
1
—
Cl
Cl
Cl
H
—
OMe
H
0.1
3.5
<0.1
5.0
3
10. Josien, H. Curr. Opin. Drug Discov. Develop. 2002, 5, 513.
11. Dovey, H. F.; John, V.; Anderson, J. P.; Chen, L. Z.; de
Saint Andrieu, P.; Fang, L. Y.; Freedman, S. B.; Folmer,
B.; Goldbach, E.; Holsztynska, E. J.; Hu, K. L.; Johnson-
Wood, K. L.; Kennedy, S. L.; Kholodenko, D.; Knops, J.
E.; Latimer, L. H.; Lee, M.; Liao, Z.; Lieberburg, I. M.;
Motter, R. N.; Mutter, L. C.; Nietz, J.; Quinn, K. P.;
Sacchi, K. L.; Seubert, P. A.; Shopp, G. M.; Thorsett, E.
D.; Tung, J. S.; Wu, J.; Yang, S.; Yin, C. T.; Schenk, D.
B.; May, P. C.; Altstiel, L. D.; Bender, M. H.; Boggs, L.
N.; Britton, T. C.; Clemens, J. C.; Czilli, D. L.; Dieckman-
McGinty, D. K.; Droste, J. J.; Fuson, K. S.; Gitter, B. D.;
Hyslop, P. A.; Johnstone, E. M.; Li, W.-Y.; Little, S. P.;
Mabry, T. E.; Miller, F. D.; Ni, B.; Nissen, J. S.; Porter,
4
5a
12.2
2.7
20.9
3.7
Increasedb
Increasedb
Me
OMe
H
6
62.7
84% at
100 lM
48.6
50.6
97.0
24.5
7
H
8
H
Me
OMe
H
Increasedb
Increasedb
Increasedb
Increasedb
9a
10
11
OMe
OMe
OMe
Me
a Compound showed cell toxicity at 100 lMas measured using the
MTS assay.19
b It is not uncommon to see increases in Ab42 with weak inhibitors. See
Ref. 20 for additional information.

1894
E. Gundersen et al. / Bioorg. Med. Chem. Lett. 15 (2005) 1891–1894
W. J.; Potts, B. D.; Reel, J. K.; Stephenson, D.; Su, Y.;
and the extracts washed with saturated NH4Cl, and brine,
dried over Na2SO4, filtered, concentrated, and dried to
give a white solid (2.08 g), which was recrystallized from
methyl-tert-butyl ether to give the desired product (1.21 g,
4.84 mmol) as a white solid, mp 154–156 °C. (ii) To a
stirred solution of the alcohol (0.20 g, 0.80 mmol) from the
previous step in 90% formic acid was added 4-meth-
oxybenzenelthiol (0.12 g, 0.88 mmol). After 18 h, the
reaction mixture was poured into saturated NaHCO3
(50 mL) and extracted with EtOAc (60 mL). The EtOAc
was washed with saturated NaHCO3 and brine, dried over
Na2SO4, filtered, and concentrated to give a gum (0.296 g).
The gum was purified by silica gel preparatory plate
chromatography, developing with 20% EtOAc/hexane to
give the desired product (0.25 g, 0.67 mmol) as a gum. (iii)
To a solution of the gum from step 2 in methanol (4 mL)
Shipley, L. A.; Whitesitt, C. A.; Yin, T.; Audia, J. E. J.
Neurochem. 2001, 76, 173.
12. Smith, D. W.; Munoz, B.; Srinivasan, K.; Bergstrom, C.
P.; Chaturvedula, P. V.; Deshpande, M. S.; Keavy, D. J.;
Lau, W. Y.; Parker, M. F.; Sloan, C. P.; Wallace, O. B.;
Wang, H. H. WO 0050391, 2000; Chem. Abstr. 2000, 133,
207678.
13. Macromodel: Schro¨dinger, 1500 S.W. First Avenue, Suite
1180 Portland, OR 97201 Mohamadi, F.; Richards, N. G.
J.; Guida, W. C.; Liskamp, R.; Lipton, M.; Caufield, C.;
Chang, G.; Hendrickson, T.; Still, W. C. J. Comput. Chem.
1990, 11, 440.
14. Halgren, T. A. J. Comput. Chem. 1996, 17, 490.
15. Halgren, T. A. J. Comput. Chem. 1996, 17, 520.
16. Halgren, T. A. J. Comput. Chem. 1996, 17, 553.
17. Halgren, T. A.; Nachbar, R. B. J. Comput. Chem. 1996,
17, 587.
18. Halgren, T. A. J. Comput. Chem. 1996, 17, 616.
19. Haugabook, S. J.; Yager, D. M.; Eckman, E. A.; Golde,
T. E.; Younkin, S. G.; Eckman, C. B. J. Neurosci. Methods
2001, 108, 171.
at 0 °C was added
a solution of oxone (0.410 g,
0.67 mmol) in water (4 mL). After addition, the bath was
removed and the reaction mixture stirred for 18 h. The
reaction mixture was diluted with water (30 mL) and
extracted with EtOAc. The extract was washed with water
and brine, dried over Na2SO4, filtered, and concentrated
to give a viscous residue (0.180 g). The residue was purified
by silica gel preparatory plate chromatography using 40%
EtOAc/hexane, to give the desired product (0.14 g,
0.35 mmol) as a white solid, dec. 138–142 °C; 1H NM R
(DMSO-d6) d 1.10 (t, J = 7.1 Hz, 3H), 1.94–2.28 (m, 2H),
2.49–2.55 (m, 2H), 3.23–3.30 (m, 2H), 3.77 (s, 3H), 3.80–
3.99 (m, 4H), 6.94 (d, J = 9.0 Hz, 2H), 7.13 (d, J = 9.0 Hz,
2H), 7.23 (dd, J = 1.2, 7.9 Hz, 2H), 7.27–7.35 (m, 3H); IR
(solid) 2980, 1690, and 1500 cmÀ1; [ES(+)], m/z 404
(M+H)+; Anal. Calcd for C21H25NO5S: C, 62.51; H,
6.24; N, 3.47 Found: C, 62.51; H, 5.96; N, 3.36.
20. Durkin, J. T.; Murthy, S.; Husten, E. J.; Trusko, S. P.;
Savage, M. J.; Rotella, D. P.; Greenberg, B. D.; Siman, R.
J. Biol. Chem. 1999, 274, 20499.
21. Jacobsen, J. S.; Spruyt, M. A.; Brown, A. M.; Saha-
srabudhe, S. R.; Blume, A. J.; Vitek, M. P.; Muenkel, H.
A.; Sonnenberg-Reines, J. J. Biol. Chem. 1994, 269, 8376.
22. The synthesis of compound 10 described here serves as a
representative example of the synthetic methodology used
in this paper. (i) To a solution of 1-carbethoxy-4-piper-
idone (1.71 g, 10 mmol) in ether (20 mL) at 0 °C was
added phenylmagnesium bromide (12 mL of a 1 Msolu-
tion in ether). The mixture was stirred for 15 min then
warmed to rt. After 2 h, the reaction mixture was
quenched with saturated NH4Cl (30 mL) then concen-
trated. The aqueous residue was extracted with EtOAc
23. Castro Pineiro, J. L.; Churcher, I.; Dinnell, K.; Harrison,
T.; Kerrad, S.; Nadin, A. J.; Oakley, P. J.; Owens, A. P.;
Shaw, D. E.; Teall, M. R.; Williams, B. J.; Williams, S.
WO 02081435, 2002; Chem. Abstr. 2002, 137, 310695.