Scaffold of the cyclooxygenase-2 (COX-2) inhibitor carprofen provides Alzheimer γ-secretase modulators

Article abstract of DOI:10.1021/jm0610200

N-Sulfonylated and N-alkylated carprofen derivatives were investigated for their inhibition and modulation of γ-secretase, which is associated with Alzheimer's disease. The introduction of a lipophilic substituent transformed the COX-2 inhibitor carprofen into a potent γ-secretase modulator. Several compounds (e.g., 9p, 11f) caused selective reduction of Aβ42 and an increase of Aβ38. The most active compounds displayed activities in the low micromolar range and no effect on the γ-secretase cleavage at the ε-site.

Full text of DOI:10.1021/jm0610200

7588
J. Med. Chem. 2006, 49, 7588-7591
Scaffold of the Cyclooxygenase-2 (COX-2)
Inhibitor Carprofen Provides Alzheimer
γ-Secretase Modulators
Rajeshwar Narlawar,† Blanca I. Pe´rez Revuelta,‡
Christian Haass,‡ Harald Steiner,‡ Boris Schmidt,*,† and
Karlheinz Baumann§
Clemens Scho¨pf-Institute of Chemistry and Biochemistry,
Darmstadt UniVersity of Technology, Petersenstrasse 22,
D-64287 Darmstadt, Germany, Department of Biochemistry,
Laboratory for Alzheimer’s and Parkinson’s Disease Research,
Adolf-Butenandt-Institute, Ludwig-Maximilians-UniVersity,
Schillerstrasse 44, D-80336 Munich, Germany, and Preclinical
Research CNS, Pharmaceuticals DiVision, F. Hoffmann-La Roche
Ltd., Building 70/345, CH-4070 Basel, Switzerland
Figure 1. Schematic representation of APP processing and the
modulation of γ-secretase cleavage by NSAIDs. (A) APP is first
processed by â-secretase to generate the C-terminal fragment CTFâ,
which is subsequently cleaved within its transmembrane domain at two
principle sites, γ and ꢀ, by γ-secretase to release Aâ and AICD. (B)
Heterogeneous cleavage at the γ-site generates different Aâ species
by cleavage at the γ38, γ40, and γ42 sites. A subset of NSAIDs
increases the cleavage at the γ38 site while reducing cleavage at the
γ42 site.
ReceiVed August 24, 2006
Abstract: N-Sulfonylated and N-alkylated carprofen derivatives were
investigated for their inhibition and modulation of γ-secretase, which
is associated with Alzheimer’s disease. The introduction of a lipophilic
substituent transformed the COX-2 inhibitor carprofen into a potent
γ-secretase modulator. Several compounds (e.g., 9p, 11f) caused
selective reduction of Aâ42 and an increase of Aâ38. The most active
compounds displayed activities in the low micromolar range and no
effect on the γ-secretase cleavage at the ꢀ-site.
Scheme 1. Effect of NSAIDs (1-4) on Aâ Levels as Reported
by Weggen3 and Beher7 and the Structure of BMS-299897 (5)
Despite tremendous progress in understanding Alzheimer’s
disease (AD), there remains the challenge to develop agents
for its therapy. Approved drugs such as acetylcholinesterase
inhibitors and memantine hydrochloride offer symptomatic
treatment, but they do not address the basic pathology of the
disease: deposition of amyloid plaques and development of
neurofibrillary tangles. The metabolism of the â-amyloid
precursor protein (APP), which is cleaved by the aspartic
proteases â-secretase and γ-secretase, results in the generation
and pathological deposition of the 40 and 42 amino acid long
peptides Aâ40 and Aâ42 (Figure 1). These fragments are the
major components of amyloid fibrils.1,2
Promising results for the treatment of AD were obtained with
some cyclooxygenase-1 (COX-1) inhibitors,3-7 both in vitro and
in a prospective, population-based cohort study of 6989
patients.8 This is not a class effect because nonsteroidal anti-
inflammatory drugs (NSAIDs) (e.g., diclofenac (2-[2-(2,6-
dichlorophenyl)aminophenyl]ethanoic acid) and naproxen ((+)-
(S)-2-(6-methoxynaphthalen-2-yl)propanoic acid)) do not lower
Aâ in vitro,3,5 and neither naproxen nor rofecoxib (4-(4-
methylsulfonylphenyl)-3-phenyl-5H-furan-2-one, a COX-2 in-
hibitor) slows cognitive decline in patients with mild-to-
moderate AD.9,10 The positive clinical results are still in need
of a sound rationale and experimental validation.11 The proof
of concept is still missing. NSAIDs that modulate γ-secretase
cleavage of APP affect the distance between APP and presenilin
1, the catalytic subunit of γ-secretase.12 They seem to interfere
with substrate recognition/cleavage and shift the precision of
γ-secretase cleavage from the γ42 to the γ38 site (Figure 1) to
generate more Aâ38 and less Aâ42.3,12 Compounds with reverse
shift were reported recently, and these enhanced Aâ42 produc-
tion.13 Noncompetitive antagonism indicates an allosteric mech-
anism of action.7 Flurbiprofen ((()-2-(3-fluoro-4-phenylphenyl)-
propanoic acid, 1) (10 and 25 mg kg-1 d-1) elicits nonselective
reductions in Aâ1-40 and Aâ1-42 plasma levels but was found
to be toxic. It produced small reductions in Aâ1-40 in the cortex
at 25 mg kg-1 d-1 but did not affect Aâ levels in the
hippocampus or cerebrospinal fluid. Cyclopropylated flurbipro-
fen analogues without COX activity display improved potency
on γ-secretase inhibition.14 Contrary to previous reports, sulindac
sulfide (2-((5Z)-1-(4-(methylthio)benzylidene)-5-fluoro-2-meth-
yl-1H-inden-3-yl)acetic acid, 2) and ibuprofen (2-(4-isobutyl-
phenyl)propanoic acid, 4) were found to be neither toxic nor
efficacious at doses up to 50 mg kg-1 d-1 15
The kinetics of
.
Aâ formation in the presence of the two NSAIDs and the
displacement of an active site directed inhibitor support allo-
steric, noncompetitive modes of action of sulindac sulfide (2)6
and R-flurbiprofen (1, Scheme 1) at low concentrations.7 This
resulted in selective inhibition of Aâ42 production. However,
both NSAIDs shift their modes of action from modulation to
complete, nonselective inhibition of γ-secretase at high con-
centrations. Unfortunately, NSAID derivatives have escaped
photoaffinity labeling techniques so far, except for biotinylated
fenofibrate which labelled the C-terminal fragment of APP.16
* To whom correspondence should be addressed. Phone: (+49) 6151-
163075. Fax: (+49) 6151-163278. E-mail: schmidt_boris@t-online.de.
† Darmstadt University of Technology.
‡ Ludwig-Maximilians-University.
§ F. Hoffmann-La Roche Ltd.
10.1021/jm0610200 CCC: $33.50 © 2006 American Chemical Society
Published on Web 12/21/2006

Letters
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 26 7589
Scheme 2a
a Reagents: (a) BnBr, K2CO3, DMF, room temp, 1.5 h, 97%; (b) NaH,
RSO2Cl, THF, 0 °C to room temp, 2-8 h, 22-84%; (c) 10% Pd-C, MeOH/
EtOAc (1:1), H2, 3-18 h, 70-94%; (d) NaOH, THF/MeOH/H2O (1:1:1),
room temp, 3-12 h, 80-92%.
Figure 2. Dose response curve of carprofen (6).
Scheme 3a
In summary, just a few NSAIDs (1-4, Scheme 1) were
reported to modulate γ-secretase, and an even smaller number
of NSAIDs display confirmed modulation at physiological
concentrations. Despite these vague, sometimes contradictory
results reported for NSAID activity, we and others13,17 were
encouraged to investigate NSAIDs as scaffolds for γ-secretase
inhibitors. We commenced with the derivatization of the
carboxylic acid common to the COX-1 inhibitors. Furthermore,
we included COX-2 inhibitors that are different from the coxib
class such as carprofen ((()-2-(3-chloro-9H-carbazol-7-yl)-
propanoic acid, 6) and etodolac (2-(1,8-diethyl-1,3,4,9-tetra-
hydropyrano[3,4-b]indol-1-yl)acetic acid) because they are
structurally closer to COX-1 than to COX-2 inhibitors. Although
some of the esters and amides displayed moderate inhibition of
the cleavage by γ-secretase (Aâ38, Aâ40, Aâ42), this approach
was soon abandoned. Few of these initial derivatives displayed
increased, unselective inhibition of γ-secretase in comparison
to their parent drugs. Most of the approximately 150 NSAID
carboxylic acid derivatives (e.g., 7) resulted in loss of activity.
This indicated an important contribution of the carboxylic acid
to target affinity. After a brief investigation of sulindac
analogues we focused on carprofen, which is a COX-2 inhibitor
approved for the use in dogs, cows, and horses. The selectivity
of carprofen versus COX-2canine and COX-1canine is greater than
100:1 (COX-2canine IC50: R/S-carprofen 102 nM, R-carprofen
5.97 µM, S-carprofen 37 nM).18 Original carprofen (as isolated
from 500 mg tablets) was found to be a weak inhibitor of
γ-secretase and reduced Aâ38, Aâ40, and Aâ42 at high concen-
tration (Figure 2). This corresponds to the activity of R-methy-
lated carprofen, which was reported to inhibit Aâ42 production
by 40% at 100 µM.17 This is in contrast to the inverse
modulation of the COX-2 inhibitors that was reported recently.13
The readily accessible sulfonamides were inspired by the
γ-secretase inhibitor 5 (BMS-299897) and supported by a recent
series from Merck Sharp & Dohme.19,20 An analogue series of
carprofen N-sulfonamides was prepared as outlined in Scheme
2. The acid functionality of carprofen was protected as a benzyl
ester 7. N-Sulfonylation of 7 was carried out using NaH and a
sulfonyl chloride in THF. The benzyl group of 8 was removed
by hydrogenation (8a-m,p-r) or base hydrolysis (8n-o,s) to
give the acid 9. We expected these compounds to be inhibitors
of γ-secretase activity because of their resemblance to the
γ-secretase inhibitor 5. Serendipitously, they turned out to be
modulators of γ-secretase activity, which control the cleavage
pattern of γ-secretase. Such modulators may preserve the
cleavage of substrates like Notch; thus, we adopted our initial
objectives from inhibition to modulation. Notch-ligand interac-
tion is a highly conserved mechanism that regulates specific
cell fate during development.21,22
a Reagents: (a) NaH, RX, THF, 0 °C to room temp, 2-8 h, 30-81%;
(b) 10% Pd-C, MeOH/EtOAc (1:1), H2, 3-18 h, 70-96%; (c) NaOH,
THF/MeOH/H2O (1:1:1), room temp, 3-12 h, 80-92%.
Scheme 4a
a Reagents: (a) SnCl2, EtOH, reflux, 3 h, 90%; (b) EDCl, HOBT, biotin,
Et3N, DMF, room temp, 16 h, 65%; (c) 10% Pd-C, MeOH, H2, room temp,
18 h, 90%.
N-Alkylated carprofen derivatives were prepared (Scheme 3)
to evaluate the contribution of the sulfonamide moiety in 9r
and the most active derivative 9p, where the sulfonamide is
shielded by isopropyl substituents. The carprofen benzyl ester
7 was alkylated using NaH and RX in THF to yield the ester
10. Subsequent benzyl deprotection by hydrogenation gave the
desired N-alkylated carprofen 11. The benzyl deprotection of
10c was carried out by base hydrolysis.
The biotinylated carprofen derivative 12 was synthesized as
depicted in Scheme 4 to identify the binding partner via
immunoprecipitation.
To evaluate the compounds for their potency in modulating
γ-secretase activity, we used the Aâ liquid-phase electro-
chemiluminescence assay to measure Aâ isoforms.27 γ-Secretase
cleavage activity at the ꢀ-site was monitored by de novo
production of AICD in vitro using the previously reported
assay.25
Compounds 9a-s, 11a-f, and 12 turned out to be effective
modulators of γ-secretase. They affected the cleavage at the
γ38, γ40, and γ42 sites to a different extent and particularly
suppressed the formation of Aâ42 while enhancing the formation
of Aâ38 and thus showed the typical profile of effective NSAIDs
(see Tables 1and 2, Figures 3 and 4). Compounds 9b,p and
11a,e,f were the most potent inhibitors of Aâ42. Interestingly,
the modulatory activity was preserved in the biotinylated 12,
which may therefore be used as an affinity reagent for

7590 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 26
Table 1. Activity of Carprofen N-Sulfonamide Derivatives
Letters
IC50 (µM)
a
entry compd
R
Aâ38
Aâ40 Aâ42
133 76
>100 56
1
2
3
4
5
6
H
78
43
9a
9b
9c
9d
9e
9f
4-methylphenyl
3,5-bis(trifluoromethyl)phenyl NDb
>30
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
>20
>40
>40
ND
11
4,5-dibromothiophen-2-yl
3,5-difluorophenyl
4-biphenyl
2-fluorophenyl
3-fluorophenyl
4-fluorophenyl
2-bromophenyl
4-bromophenyl
phenyl
NDb
NDb
NDb
NDb
NDb
NDb
NDb
NDb
NDb
NDb
NDb
NDb
NDb
9.3
>40
>40
37
>40
>40
>40
40
>40
>40
>40
>40
39
>40
8.5
>40
25
>40
6
7
8
9
9g
9h
9i
10
11
12
13
14
15
16
17
18
19
20
9j
9k
9l
9m
9n
9o
9p
9q
9r
9s
4-chlorophenyl
3-chlorophenyl
3-nitrophenyl
Figure 3. Dose response curves for the most active carprofen
derivatives (Aâ % of control): (A) 11f; (B) 9p; (C) 9b; (D) 12; (9)
Aâ38; (2) Aâ40; (b) Aâ42).
4-nitrophenyl
2,4,6-triisopropylphenyl
4-n-propylphenyl
octyl
NDb
NDb
NDb
4-cyanophenyl
a EC50 values are displayed for Aâ38 except 6. The EC50 is based on the
maximum level with a slope approximating 0. b Maximum effect on Aâ38
not observed at 40 µM (except for 6, 160 µM; 9a, 200 µM; and 9b, 200
µM).
Table 2. Activity of N-Alkylated Carprofen Derivatives.
activity (µM)
EC50
IC50
IC50
entry
compd
R
Aâ38
Aâ40
Aâ42
Figure 4. Dose-response curves for carprofen derivatives (9b, 9p,
11a and 11f) on in vitro AICD generation. Results are the average of
three independent experiments. Error bars indicate the standard error
of the mean.
1
2
3
4
5
6
7
12
NDa >150
88
7.5
39
11a
11b
11c
11d
11e
11f
3-(trifluoromethoxy)benzyl
NDa
NDa
NDa
NDa
8.1
>40
>40
>40
>40
>40
>40
3-methoxybenzyl
3-nitrobenzyl
octyl
nonyl
decyl
22
6.9
3.0
2.9
affected by the compounds to various extent (Figure 4).
However, consistent with previous results,3,7 generally much
higher compound concentrations than those determined to be
modulatory were required to inhibit the ꢀ-cleavage. One of the
most active carprofen derivatives 11f was selected for evaluation
in COX-1 and COX-2 assays to rule out COX-1 or COX-2
mediated effects at the necessary concentrations for γ-
secretase modulation. The assays were performed at CEREP
(www.cerep.com) using indomethacin as a standard for COX-1
and NS398 (N-(2-(cyclohexyloxy)-4-nitrophenyl)methanesulfon-
amide) for COX-2. 11f displayed no activity on COX-1 and
only marginal activity on COX-2 at 10 µM. No toxicity was
observed at 40 µM in H4 cells except for 9p and 12, which
significantly decreased viability at 40 µM, respectively (data
not shown).
In conclusion, the introduction of a single lipophilic sub-
stituent, which may vary from arylsulfone to alkyl substituents,
turns the COX-2 inhibitor carprofen into a γ-secretase modulator
and improved potency 10-fold or more. Thus, several com-
pounds (e.g., 9p, 11f) caused the selective reduction of Aâ42
and an increase of the less aggregatory Aâ38. The most active
compounds are more potent than the best reported NSAIDs,
and they are devoid of COX-1 and COX-2 activity at the critical
concentration; thus, they do not interfere with the delicate COX-
1/COX-2 balance. Some of the sulfonamides are comparable
in potency to the best N-alkylated analogues (11b), but the 50%
increase of the tPSA (9p: 93.4 Å2) makes a penetration of the
blood-brain barrier less likely. Therefore, the N-alkylated
derivatives are favored over the sulfonamides for further
investigations. The properties of these N-alkylated lead candi-
dates are close to the range of approved drugs or preclinical
candidates except for their clogP.26 The carboxylic acid group
5.8
a Maximum effect on Aâ38 not observed at 40 µM (except for 12, 150
µM).
γ-secretase. We observed from another series that N-alkylation
of carbazole with shorter n-alkyl chains such as n-butyl and
n-hexyl diminishes the γ-secretase modulatory activity (unpub-
lished results). The observed differences for octyl, nonyl, or
decyl substitution are small and so is the difference for
alkylsulfone amides of equivalent length (see 11e versus 9r).
The sulfone amides display an increased topological polar
surface area (tPSA) from 49 to approximately 93 Å2, which
makes blood-brain barrier penetration less likely.23 The cal-
culated increase in tPSA is only partially compensated by a small
reduction of the clogP (- 0.5) in comparison to the analogue
alkyl derivatives. This favors the alkyl derivatives over the
sulfone amides for further investigation in animal models. The
N-benzylated or N-arylsulfonated derivatives benefit from
fluorinated substituents (9b, 11a) in the meta position. Ortho
fluorination as in (9f) did not alter the activity in comparison
to the parent sulfonamide (9k). The replacement of the methoxy
group in 11b by a trifluoromethoxy resulted in a 5-fold
improvement of the IC50 (Aâ42). This fluorine-derived enhance-
ment may be due to lipophilicity or polar interactions. We
speculate that the lipophilic substituent anchors the N-substituted
carprofen in the required orientation within the membrane; thus,
the maximum tolerated length should be similar to those of
natural phospholipids. The effect of the most potent compounds
on γ-secretase cleavage at the ꢀ-site was assessed by an in vitro
assay that monitors the de novo generation of the APP
intracellular domain (AICD).24,25 The generation of AICD was

Letters
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 26 7591
may interfere with uptake, but the tPSA of 11b is just 62.9 Å2.
This compares favorably to lumiracoxib ({2-[(2-chloro-6-
fluorophenyl)amino]-5-methylphenyl}acetic acid, tPSA ) 58
Å2). The more polar 11c has similar properties (tPSA ) 96.1,
clogP ) 5.83) as the carboxylic acid 5 (clogP ) 5.92, tPSA )
93.4 Å2).19 The lipophilic substituents cause amphiphilic proper-
ties of the carboxylic acids, which may interact with membranes.
However, the placement of a polar end group, as in 12,
weakened but did not reverse the modulatory effect. Again, we
favor the N-alkylated derivatives for the investigation of
potential membrane interactions, as they allow the incorporation
of phospholipids analogues and membrane disrupting fragments.
If affected at all, the ꢀ-cleavage of γ-secretase was inhibited
at much higher compound concentrations than those determined
to be modulatory at the γ-site (Figure 4). The compounds are
therefore expected to have little or no impact on γ-secretase-
mediated signaling via the AICD or via intracellular domains
of other γ-secretase substrates. However, the interaction site of
these compounds has not been identified yet. The improvement
of potency and the investigation of in vivo activity are subject
to further investigations.
(10) Reines, S. A.; Block, G. A.; Morris, J. C.; Liu, G.; Nessly, M. L.;
Lines, C. R.; Norman, B. A.; Baranak, C. C. Rofecoxib: No Effect
on Alzheimer’s Disease in a 1-Year, Randomized, Blinded, Controlled
Study. Neurology 2004, 62, 66-71.
(11) Cummings, J. L. Continuing Medical Education: Alzheimer’s
Disease. N. Engl. J. Med. 2004, 351, 110.
(12) Lleo, A.; Berezovska, O.; Herl, L.; Raju, S.; Deng, A.; Bacskai, B.
J.; Frosch, M. P.; Irizarry, M.; Hyman, B. T. Nonsteroidal Anti-
Inflammatory Drugs Lower Aâ42 and Change Presenilin 1 Conforma-
tion. Nat. Med. 2004, 10, 1065-1066.
(13) Kukar, T.; Murphy, M. P.; Eriksen, J. L.; Sagi, S. A.; Weggen, S.;
Smith, T. E.; Ladd, T.; Khan, M. A.; Kache, R.; Beard, J.; Dodson,
M.; Merit, S.; Ozols, V. V.; Anastasiadis, P. Z.; Das, P.; Fauq, A.;
Koo, E. H.; Golde, T. E. Diverse Compounds Mimic Alzheimer
Disease-Causing Mutations by Augmenting Aâ42 Production. Nat.
Med. 2005, 11, 545-550.
(14) Peretto, I.; Radaelli, S.; Parini, C.; Zandi, M.; Raveglia, L. F.; Dondio,
G.; Fontanella, L.; Misiano, P.; Bigogno, C.; Rizzi, A.; Riccardi, B.;
Biscaioli, M.; Marchetti, S.; Puccini, P.; Catinella, S.; Rondelli, I.;
Cenacchi, V.; Bolzoni, P. T.; Caruso, P.; Villetti, G.; Facchinetti,
F.; DelGiudice, E.; Moretto, N.; Imbimbo, B. P. Synthesis and
Biological Activity of Flurbiprofen Analogues as Selective Inhibitors
of â-Amyloid1-42 Secretion. J. Med. Chem. 2005, 48, 5705-5720.
(15) Lanz, T. A.; Fici, G. J.; Merchant, K. M. Lack of Specific Amyloid-
beta(1-42) Suppression by Nonsteroidal Anti-Inflammatory Drugs
in Young, Plaque-Free Tg2576 Mice and in Guinea Pig Neuronal
Cultures. J. Pharmacol. Exp. Ther. 2005, 312, 399-406.
(16) Kukar, T.; Ladd, T.; Bann, M.; Fraering, P.; Kache, R.; Fauq, A.;
Wolfe, M.; Koo, E. H.; Golde, T. E. P4-289: Mechanistic Insights
into the in Vitro and in ViVo Modulation of Aâ42 Levels and Amyloid
Pathology by NSAIDS and Related Compounds. Alzheimer’s and
Dementia 2006, 2, S601-S602.
(17) Stock, N.; Munoz, B.; Wrigley, J. D. J.; Shearman, M. S.; Beher,
D.; Peachey, J.; Williamson, T. L.; Bain, G.; Chen, W.; Jiang, X.;
St-Jacques, R.; Prasit, P. The Geminal Dimethyl Analogue of
Flurbiprofen as a Novel Aâ42 Inhibitor and Potential Alzheimer’s
Disease Modifying Agent. Bioorg. Med. Chem. Lett. 2006, 16, 2219-
2223.
(18) Ricketts, A. P.; Lundy, K. M.; Seibel, S. B. Evaluation of Selective
Inhibition of Canine Cyclooxygenase 1 and 2 by Carprofen and Other
Nonsteroidal Anti-Inflammatory Drugs. Am. J. Vet. Res. 1998, 59,
1441-1446.
(19) Barten, D. M.; Guss, V. L.; Corsa, J. A.; Loo, A.; Hansel, S. B.;
Zheng, M.; Munoz, B.; Srinivasan, K.; Wang, B.; Robertson, B. J.;
Polson, C. T.; Wang, J.; Roberts, S. B.; Hendrick, J. P.; Anderson,
J. J.; Loy, J. K.; Denton, R.; Verdoorn, T. A.; Smith, D. W.;
Felsenstein, K. M. Dynamics of â-Amyloid Reductions in Brain,
Cerebrospinal Fluid, and Plasma of â-Amyloid Precursor Protein
Transgenic Mice Treated with a γ-Secretase Inhibitor. J. Pharmacol.
Exp. Ther. 2005, 312, 635-643.
(20) Beher, D.; Bettati, M.; Checksfield, G. D.; Churcher, I.; Doughty,
V. A.; Oakley, P. J.; Quddus, A.; Teall, M. R.; Wrigley, J. D.
Preparation of Tetrahydrocarbazole-1-Alkanoic Acids for the Treat-
ment of Alzheimer’s Disease and Related Conditions. WO 2005013985.
(21) Selkoe, D.; Kopan, R. Notch and Presenilin: Regulated Intra-
membrane Proteolysis Links Development and Degeneration. Annu.
ReV. Neurosci. 2003, 26, 565-597.
Acknowledgment. We thank the DFG (B.S. and R.N.,
SPP1085 (SCHM1012-3-1/2); C.H. and H.S., SFB596), the EU
(B.S. and R.N., APOPIS (LSHM-CT-2003-503330); C.H.,
APOPIS), and the BMBF (C.H., NGFN) for support of this
work.
Supporting Information Available: Synthetic procedure and
spectral data for the tested compounds and experimental procedure
for assay. This material is available free of charge via the Internet
at http://pubs.acs.org.
References
(1) Haass, C. Take FivesBACE and the γ-Secretase Quartet Conduct
Alzheimer’s Amyloid â-Peptide Generation. EMBO J. 2004, 23,
483-488.
(2) Churcher, I.; Beher, D. γ-Secretase as a Therapeutic Target for the
Treatment of Alzheimer’s Disease. Curr. Pharm. Des. 2005, 11,
3363-3382.
(3) Weggen, S.; Eriksen, J. L.; Das, P.; Sagi, S. A.; Wang, R.; Pietrizik,
C. U.; Findlay, K. A.; Smith, T. E.; Murphy, M. P.; Bulter, T.; Kang,
D. E.; Marquez-Sterling, N.; Golde, T. E.; Koo, E. H. A Subset of
NSAIDs Lower Amyloidogenic Aâ42 Independently of Cyclooxy-
genase Activity. Nature 2001, 414, 212-216.
(4) Weggen, S.; Eriksen, J. L.; Sagi, S. A.; Pietrzik, C. U.; Golde, T. E.;
Koo, E. H. Aâ42-Lowering Nonsteroidal Anti-Inflammatory Drugs
Preserve Intramembrane Cleavage of the Amyloid Precursor Protein
(APP) and ErbB-4 Receptor and Signaling through the APP Intra-
cellular Domain. J. Biol. Chem. 2003, 278, 30748-30754.
(5) Eriksen, J. L.; Sagi, S. A.; Smith, T. E.; Weggen, S.; Das, P.;
McLendon, D. C.; Ozols, V. V.; Jessing, K. W.; Zavitz, K. H.; Koo,
E. H.; Golde, T. E. NSAIDs and Enantiomers of Flurbiprofen Target
γ-Secretase and Lower Aâ42 in Vivo. J. Clin. InVest. 2003, 112, 440-
449.
(6) Takahashi, Y.; Hayashi, I.; Tominari, Y.; Rikimaru, K.; Morohashi,
Y.; Kan, T.; Natsugari, H.; Fukuyama, T.; Tomita, T.; Iwatsubo, T.
Sulindac Sulfide Is a Noncompetitive γ-Secretase Inhibitor That
Preferentially Reduces Aâ42 Generation. J. Biol. Chem. 2003, 278,
18664-18670.
(7) Beher, D.; Clarke, E. E.; Wrigley, J. D. J.; Martin, A. C. L.; Nadin,
A.; Churcher, I.; Shearman, M. S. Selected Non-Steroidal Anti-
Inflammatory Drugs and Their Derivatives Target γ-Secretase at a
Novel Site: Evidence for an Allosteric Mechanism. J. Biol. Chem.
2004, 279, 43419-43426.
(8) in ’t Veld, B. A.; Ruitenberg, A.; Hofman, A.; Launer, L. J.; van
Duijn, C. M.; Stijnen, T.; Breteler, M. M. B.; Stricker, B. H. C.
Nonsteroidal Antiinflammatory Drugs and the Risk of Alzheimer’s
Disease. N. Engl. J. Med. 2001, 345, 1515-1521.
(9) Aisen, P. S.; Schafer, K. A.; Grundman, M.; Pfeiffer, E.; Sano, M.;
Davis, K. L.; Farlow, M. R.; Jin, S.; Thomas, R. G.; Thal, L. J. Effects
of Rofecoxib or Naproxen vs Placebo on Alzheimer Disease
Progression: A Randomized Controlled Trial. J. Am. Med. Assoc.
2003, 289, 2819-2826.
(22) Kopan, R.; Ilagan, M. X. γ-Secretase: Proteasome of the Membrane?
Nat. ReV. Mol. Cell Biol. 2004, 5, 499-504.
(23) Ertl, P.; Rohde, B.; Selzer, P. Fast Calculation of Molecular Polar
Surface Area as a Sum of Fragment-Based Contributions and Its
Application to the Prediction of Drug Transport Properties. J. Med.
Chem. 2000, 43, 3714-3717.
(24) Sastre, M.; Steiner, H.; Fuchs, K.; Capell, A.; Multhaup, G.; Condron,
M. M.; Teplow, D. B.; Haass, C. Presenilin-Dependent γ-Secretase
Processing of â-Amyloid Precursor Protein at a Site Corresponding
to the S3 Cleavage of Notch. EMBO J. 2001, 2, 835-841.
(25) Larbig, G.; Zall, A.; Schmidt, B. Inhibitors Designed for Presenilin
1 Utilizing by Means of Aspartic Acid Activation. HelV. Chim. Acta
2004, 87, 2334-2340.
(26) Boehm, H.-J., Schneider, G., Eds.; Virtual Screening for BioactiVe
Molecules; Wiley-VCH: Weinheim, Germany, 2000.
(27) Brockhaus, M.; Grunberg, J.; Rohrig, S.; Loetscher, H.; Wittenburg,
N.; Baumeister, R.; Jacobsen, H.; Haass, C. Caspase-Mediated
Cleavage Is Not Required for the Activity of Presenilins in
Amyloidogenesis and NOTCH Signaling. NeuroReport 1998, 9,
1481-1486.
JM0610200