Design, synthesis, and biological evaluation of a novel class of γ-secretase modulators with pparγ activity

Article abstract of DOI:10.1021/jm1003073

We present a novel class of dual modulators of γ-secretase and peroxisome proliferator-activated receptor γ (PPARγ) based on the structure of 2-(bis(phenethoxy)pyrimidine-2-ylthio)hexanoic acid 8 (IC ₅₀(Aβ42) = 22.8 μM, EC₅₀(PPARγ) =

Full text of DOI:10.1021/jm1003073

J. Med. Chem. 2010, 53, 4691–4700 4691
DOI: 10.1021/jm1003073
Design, Synthesis, and Biological Evaluation of a Novel Class of γ-Secretase Modulators
with PPARγ Activity
Martina Hieke,† Julia Ness,‡ Ramona Steri,† Michaela Dittrich,† Christine Greiner,§ Oliver Werz,§
Karlheinz Baumann, Manfred Schubert-Zsilavecz,† Sascha Weggen,*,‡ and Heiko Zettl*,†,^
†Institute of Pharmaceutical Chemistry, ZAFES/LiFF/Goethe University Frankfurt, Max-von-Laue-Strasse 9, D-60438 Frankfurt am Main,
Germany, ‡Molecular Neuropathology Group, Department of Neuropathology, Heinrich-Heine University Duesseldorf, Moorenstrasse 5,
D-40225 Duesseldorf, Germany, §Department of Pharmaceutical Analytics, Eberhard-Karls University Tuebingen, Auf der Morgenstelle 8,
D-72076 Tuebingen, Germany, and Pharmaceuticals Division, Preclinical Research CNS, F. Hoffmann-La Roche Ltd., CH-4070 Basel,
Switzerland. ^ Present address: ETH Zurich, Institute of Pharmaceutical Sciences, Wolfgang-Pauli-Strasse 10, CH-8093 Zurich, Switzerland.
Received March 8, 2010
We present a novel class of dual modulators of γ-secretase and peroxisome proliferator-activated
receptor γ (PPARγ) based on the structure of 2-(bis(phenethoxy)pyrimidine-2-ylthio)hexanoic acid
8 (IC50(Aβ42)=22.8 μM, EC50(PPARγ)=8.3 μM). The modulation of both targets with approved
drugs (i.e., amyloid-β 42 (Aβ42)-lowering NSAIDs for γ-secretase and glitazones for PPARγ) has
demonstrated beneficial effects in in vitro and in vivo models of Alzheimer’s disease (AD). However,
although NSAIDs and PPARγ agonists share similar structural features, no druglike compounds with
dual activities as γ-secretase modulators (GSMs) and PPARγ agonists have been designed so far. On the
basis of our initial lead structure 8, we present the structure-activity relationships (SARs) of broad
structural variations. A significant improvement was reached by the introduction of p-trifluoromethyl
substituents at the phenyl residues yielding compound 16 (IC50(Aβ42) = 6.0 μM, EC50(PPARγ) =
11.0 μM) and the replacement of the two phenyl residues of 8 by cyclohexyl yielding compound 22
(IC50(Aβ42) = 5.1 μM, EC50(PPARγ) = 6.6 μM).
Introduction
γ-secretase enzyme, a multiprotein complex that catalyzes the
final step in the cellular generation of Aβ peptides. However,
because of its essential role in the NOTCH signaling pathway,
mechanism-based toxicity with severe phenotypes in the gas-
trointestinal and hematopoietic systems has been observed in
preclinical studies of γ-secretase inhibitors6-8 and might be a
limiting factor in clinical efficacy. Intriguingly, a subset of
nonsteroidal anti-inflammatory drugs (NSAIDs) such as ibu-
profen and indomethacin were found to modulate the
γ-secretase activity in a way that allowed selective inhibition
of Aβ42 production (Figure 1).9 These compounds, termed
γ-secretase-modulators (GSMs), shifted the proteolytic pro-
cessing of APP toward higher production of shorter Aβ species
such as Aβ38 at the expense of the highly toxic Aβ42.
Importantly, impairment of NOTCH processing and signaling
was not observed.9,10 Unfortunately, these NSAID-type GSMs
suffer from low potency against γ-secretase (Figure 1), poor
brain permeability, and side effects related to inhibition of
cyclooxygenases (COX) such as gastrointestinal toxicity and
increased cardiovascular mortality.11 Consequently, NSAID-
type GSMs and derived analogues, whose consumption ap-
peared to confer a reduced risk for AD in epidemiological
studies, have produced negative or inconclusive clinical results
in patients suffering from established AD11-14 or showed
unacceptable side effects for long-term treatment. Recently,
potent GSMs with favorable pharmacological properties have
emerged.15,16
Alzheimer’s disease (ADa) is the most common age-related
cause of dementia with 26.6 million patients worldwide in
2006 and over a 100 million predicted cases by 2050.1 The
cardinal symptom of the disease is progressive memory loss
due to the degeneration of neurons and synapses in the
cerebral cortex and subcortical regions of the brain. The
neuropathology of AD is characterized by the extracellular
deposition of amyloid-β (Aβ) plaques, the formation of
intracellular neurofibrillary tangles, chronic brain inflamma-
tion, and oxidative damage.2 According to the modified
amyloid hypothesis, small soluble oligomers of Aβ peptides,
particularly those formed by the highly hydrophobic Aβ42
species, are the disease-initiating agents in AD.3 Aβ42 is a
proteolytic fragment resulting from sequential cleavage of the
amyloid precursor protein (APP) by two aspartyl proteases,
i.e., β-secretase and γ-secretase. Currently, a variety of poten-
tially disease-modifying therapeutic approaches are under
development that target the production, aggregation, or
clearance of Aβ peptides in brain (Scheme 1).4,5 Substantial
advances have been made with respect to inhibitors of the
*To whom correspondence should be addressed. For S.W.: phone,
þ49 (211) 8104506; fax, þ49 (211) 8104577; e-mail, sweggen@
uni-duesseldorf.de. For H.Z.: phone, þ41 (44) 6339113; fax, þ41 (44)
6331379; e-mail, heiko.zettl@pharma.ethz.ch.
a Abbreviations: Aβ, amyloid β peptide; AD, Alzheimer’s disease; APP,
amyloid precursor protein; COX, cyclooxygenase; GSM, γ-secretase modu-
lator; NSAIDs, nonsteroidal anti-inflammatory drugs; PPAR, peroxisome
proliferator-activated receptor; SAR structure-activity relationship.
In addition, some NSAIDs including the GSMs indometha-
cin and ibuprofen function as agonists of the peroxisome
r
2010 American Chemical Society
Published on Web 05/26/2010
pubs.acs.org/jmc

4692 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 12
Scheme 1. Amyloid Cascade
Hieke et al.
type 2 diabetes appear to have a 2-fold increased risk of
developing AD.23,24 Potential direct effects of PPARγ agonists
on brain pathology in AD include lowering of Aβ levels,
suppression of toxic proinflammatory mediators, and reduction
of cerebrovascular impairment.25-30 Recently, a phase III clini-
cal study with the brain-impermeable PPARγ agonist rosigli-
tazone failed to demonstrate efficacy in patients with established
AD.31 However, PPARγ agonists might be able to counteract
the increased AD risk in patients with type 2 diabetes, and in
fact, ongoing treatment and prevention trials for type 2 diabetes
include cognition ancillary studies.24 In this respect, a pilot study
with the PPARγ agonist pioglitazone has provided evidence for
cognitive stabilization in AD patients with concurrent type 2
diabetes.32 Notably, pioglitazone has been shown to have at least
limited blood-brain barrier permeability.33
In summary, small molecules with dual GSM and PPARγ
agonist activity could be a novel strategy for prevention of
AD, particularly in patients with type 2 diabetes. In an initial
screening of our in-house compound library, we have iden-
tified compound 8 (2-(bis(phenethoxy)pyrimidine-2-ylthio)-
hexanoic acid) that displayed a typical GSM profile and
concentration-dependently and selectively decreased Aβ42
levels with a concomitant increase in Aβ38 production (IC50-
(Aβ42) = 22.8 μM, EC50(Aβ38) = 11.3 μM, Figure 2). Addi-
tionally, compound 8 is a subtype-selective PPARγ modulator
(EC50 = 8.3 μM, 60% max activation). Here, we describe the
structure-activity relationships (SARs) of a novel class of dual
γ-secretase/PPARγ modulators based on the scaffold of com-
pound 8. Our strategy aimed at improving the GSM activity
while maintaining PPARγ agonism. Characteristic structural
features of compound 8 are a carboxylic acid headgroup and
two phenethyl residues forming a lipophilic backbone. As
conventional NSAIDs have similar structural features, we
further determined the inhibitory activity of all compounds
for COX-1 and COX-2.
Figure 1. In vitro pharmacology of γ-secretase-modulating
NSAIDs.15,47-49
proliferator-activated receptor γ (PPARγ).17,18 PPARγ be-
longs to a family of ligand-activated nuclear receptors that
act as lipid sensors, and synthetic ligands for PPARγ include
the widely prescribed antidiabetic drugs rosiglitazone and
pioglitazone.19 Evidence indicates that PPARγ agonists might
have multiple beneficial effects in AD both on core pathologi-
cal processes in brain and on peripheral factors such as serum
glucose levels and insulin sensitivity that constitute potential
risk factors for AD.20 Several studies have demonstrated that
AD is associated with perturbations in insulin metabolism, and
diet-induced insulin resistance caused increased brain Aβ levels
and plaque formation in a mouse model of AD.21 In addition, a
recent cross of an AD mouse model to leptin-deficient diabetic
mice revealed accelerated learning deficits and severe cerebro-
vascular amyloid deposition.22 Importantly, individuals with

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 12 4693
Figure 2. Lead compound 8 and its impact on Aβ38, Aβ40, and Aβ42 formation. SARs based on 8 were obtained by structural modifications
of the encircled residues.
Scheme 2. Synthetic Route of Compounds 6-25a
a Reagents and conditions: (i) 2-bromo-(R1)-ethyl acetate (1.5 equiv), thiobarbituric acid (1 equiv), TEA (1.5 equiv), DMF, 80 ꢀC, 4 h; (iia) 6a-10a
(1 equiv), R2-Hal (2.1 equiv), K2CO3 (2.3 equiv), DMF, 80 ꢀC, 3-28 h or (iib) 6a-10a (1 equiv), R2-OH (2 equiv), DEAD (2.5 equiv), TPP (2.5 equiv),
THF, room temp, 1-3 h; (iii) LiOH (5 equiv), THF/MeOH, H2O, 25-50 ꢀC, 2-24 h.
Chemistry
Potential effects on proteolytic processing of NOTCH were
investigated for the two compounds with the highest GSM
activity using a previously described reporter assay.38 Cyto-
toxicity of all compounds was determined using alamar blue
reagent.
The general synthetic procedure of the compounds is illu-
strated in Scheme 2. Presented compounds were prepared in a
three-step reaction, which has been described previously by
Koeberle et al.34 First, a nucleophilic substitution between
thiobarbituric acid and the respective R-bromoethyl ester
using dimethylformamide (DMF) as solvent and triethyl-
amine (TEA) as corresponding base was carried out. This
step was followed by the etherification of the two hydroxyl
groups, either by a Mitsunobu reaction or by a Williamson
ether synthesis. In the case of the Mitsunobu reaction, we used
the precursor, the respective alcohol, triphenylphosphine
(TPP), and diethylazodicarboxylate (DEAD) and stirred the
solution at room temperature under argon atmosphere.35
The Williamson ether synthesis was carried out with the res-
pective alkyl halogenide, K2CO3, and DMF at 80 ꢀC. Finally,
the ester group was hydrolyzed with LiOH in tetrahydrofuran
(THF) and H2O to yield final compounds 6-25.
Results
In an initial screening of our in-house-library we have
identified 2-(bis(phenethoxy)pyrimidine-2-ylthio)hexanoic acid
8, a selective PPARγ modulator that displayed the character-
istic GSM activity with IC50(Aβ42) = 22.8 μM (Figure 2).
As the first hit, compound 8 served as the structural temp-
late for all presented derivatives. Important structural fea-
tures of 8 are two 2-phenethyl residues forming a lipophilic
backbone and a carboxylic acid function as headgroup. This
acidic headgroup is an essential structural element required
for PPARγ activation39 and of NSAID-type GSMs.15,40
Therefore, we have not modified the carboxylic acid group
and focused our efforts on the remaining parts of lead
structure 8.
Biological Assays
In previous studies, we have observed that the length of the
alkyl chain of R-substituted pirinixic acid derivatives has a
high impact on PPAR activity.41,42 Hence, we systematically
varied this position by elongation and shortening of the R-n-
butyl residue as well as by introduction of a R-phenyl residue
(compounds 6-10; see Table 1). The inhibitory activity on
Aβ42 production showed clear SAR depending on the length
of the R-alkyl chain. Removal of the R-chain (compound 6,
R1 = H) caused a complete loss of GSM activity. In line with
this observation, compound 7 with the shorter ethyl residue
showed only weak GSM activity (IC50(Aβ42) = 41.6 μM). A
slight decrease of GSM activity was also observed after
elongation to n-hexyl (compound 9). In regard to PPARγ
PPAR activity of the final compounds was tested in a
cellular luciferase-based PPAR transactivation assay, which
has been described previously.36 Final compounds 6-25 were
fully characterized on all PPAR subtypes (R, γ, and δ). Since
every derivative described in this paper was inactive on
PPARR and PPARδ (tested concentration of 10 μM), the
following discussion is only focused on PPARγ activity. For
biological characterization of GSM activity, we measured the
levels of Aβ38, Aβ40, and Aβ42 peptides in cell supernatants
with an ELISA assay as described with some modifications.37
Inhibition of COX was tested in a cell-free assay using isolated
ovine COX-1 and human recombinant COX-2 enzymes.34

4694 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 12
Hieke et al.
Table 1. PPARγ Activation and γ-Secretase Modulation (Determined by Aβ42 Inhibition and Aβ38 Activation) of Compounds 6-25a
a Inactive: tested concentration of 10 μM.
activation, the length of the R-alkyl chain didnot considerably
influence theactivity(EC50 valuesof6-9 between 3 and 9 μM,
with a tendency for decreased activity with longer chains).
Phenyl-substituted 10 showed no PPARγ activity and can

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 12 4695
Figure 3. GSM activity of compounds 16 and 22: dose-response curves for Aβ40, Aβ42, and Aβ38 formation.
Figure 4. Lack of effect of compounds 16 and 22 in the NOTCH reporter gene assay.
thus be described as a conventional GSM. With an IC50 of
29.3 μM for Aβ42 inhibition, compound 10 is somewhat less
potent compared to the lead structure. Together, the initial
n-butyl chain (8) turned out to have the optimal chain length
in R-position for dual modulation of γ-secretase and PPARγ.
Interestingly, it was possible to modify the activity profile
toward a conventional GSM by the introduction of a bulky
phenyl moiety in R-position to the carboxylic acid.
On the basis of previous studies where we have assessed the
influence of the stereochemical configuration of some R-alkyl
substituted pirinixic acid derivatives on PPAR activity,43 we
decided to investigate the stereochemical impact on the biolo-
gical activity of 8. We have separated 8 into its single enantio-
mers (8c and 8d) using enantioselective preparative HPLC.44
While the (R)-enantiomer 8c (IC50(Aβ42) = 21.5 μM) showed
slightly superior activity compared to the (S)-enantiomer (8d)
(IC50(Aβ42) = 32.6 μM) on γ-secretase, the results on PPARγ
(compound 14) andmethoxy (15) substituentsincreasedGSM
activity with methyl (IC50(Aβ42) = 13.0 μM) being slightly supe-
rior to methoxy (IC50(Aβ42) = 19.3 μM). Further improvement
of Aβ42 inhibition was finally achieved by introduction of
trifluoromethyl (16) (IC50(Aβ42) = 6.0 μM, Figure 3) and
trifluoromethoxy moieties (17) (IC50(Aβ42) = 8.6 μM),
which belong to the most potent GSMs from this series.
Compounds 16 and 17 modulate PPARγ with an EC50 of
about 10 μM. In contrast to these four derivatives, introduc-
tion of nitro (18), cyano (19), and the bioisosteric replace-
ment of both benzenes by thiophene (20) caused a clear
decrease of GSM activity with IC50 values above 30 μM.
Finally, we replaced the phenyl moieties by aliphatic rings
of various sizes. Substitution with cycloheptyl (21), cyclohexyl
(22), and cyclopentyl (23) moieties improved the potency
regarding inhibition of GSM activity with IC50 values for
Aβ42 in the low micromolar range. The cyclopentyl (IC50-
(Aβ42) = 17.2 μM) and cycloheptyl (IC50(Aβ42) = 11.3 μM)
derivatives showed considerable activity. However, the most
active compound is the cyclohexyl derivative 22 (IC50(Aβ42) =
5.1 μM), which represents the most potent GSM presented in
this study (Figure 3). Further size restriction to cyclopropyl (24)
as well as ring-opening of the cyclopropyl (isopentyl-substi-
tuted 25) was detrimental with IC50 > 28 μM. Within this
series, PPARγ activity remained in the range between 4 and
11 μM (21-25). The largest cycloheptyl substituent (21) caused
a complete loss of PPARγ activity, thereby representing the
second conventional GSM besides R-phenyl substituted 10.
NOTCH Processing. Potential effects on NOTCH proces-
sing were examined for compounds 16 and 22 that displayed
the highest GSM activity with an established reporter
assay.38 Treatment of cells with the γ-secretase inhibitor
L-685,458 or replacement of the plasmid encoding NOTCH
with an empty vector caused a dramatic decrease in reporter
activity as expected (Figure 4). However, treatment of cells
with 5-40 μM compounds 16 (Figure 4A) and 22 (Figure 4B)
were vice versa (8c, 22% activation at 10 μM; 8d, EC50
=
10.4 μM). In summary, we were not able to identify any
superior biological activity of one enantiomer. Thus, the
stereocenter in the R-position of the carboxylic acid does not
cause a significant impact on γ-secretase and PPARγ activity.
Most space for structural modifications is provided by the
two ether substituents of the central pyrimidine core of lead
structure 8. First, we elongated and shortened the aliphatic
spacer between the two ether moieties and the phenyl residues
(compounds 11-13). Compound 11 (IC50(Aβ42) = 25.2 μM)
with a shorter methylene and compound 13 with a longer
butylene spacer were almost equipotent on γ-secretase com-
pared to 8, whereas the elongation to a propylene spacer (12)
led to a decreased potency. In regard to PPARγ activation, the
methylene analogue 11 showed a substantially decreased activ-
ity. The elongation to propylene and butylene spacers (12 and
13) did not cause considerable changes in PPARγ activity.
Second, we introduced different substituents in para-posi-
tion of both phenyl moieties based on lead structure 8. Methyl

4696 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 12
Hieke et al.
Table 2. COX Inhibition of Compounds 6-25a
remaining activity ( SD at 10 μM (%)
remaining activity ( SD at 10 μM (%)
compd
COX-1
COX-2
compd
COX-1
COX-2
6
7
91.8 ( 5.5
84.6 ( 5.5
91.7 ( 15.6
56.9 ( 1.4
15
16
72.8 ( 29.1
54.4 ( 23.6
IC50 = 16.9 μM
73.0 ( 11.5
94.6 ( 33.0
69.0 ( 27.7
IC50 > 40 μM
75.8 ( 8.7
8
64.2 ( 24.8
IC50 = 18.7 μM
73.9 ( 15.8
83.5 ( 12.6
61.8 ( 17.0
72.2 ( 13.3
88.9 ( 21.3
90.9 ( 28.8
IC50 > 40 μM
98.5 ( 11.1
92.8 ( 2.1
63.8 ( 16.8
86.0 ( 21.8
61.6 ( 2.4
17
8c
8d
9
18
19
20
21
22
82.1 ( 12.2
88.8 ( 16.6
76.6 ( 6.9
91.6 ( 14.6
82.6 ( 24.3
102.4 ( 16.9
50.1 ( 9.3
49.5 ( 18.5
IC50 = 10.9 μM
84.7 ( 3.5
10
11
56.1 ( 8.8
52.7 ( 11.1
IC50 = 13.4 μM
68.2 ( 6.9
97.7 ( 17.4
72.4 ( 31.3
12
13
14
78.8 ( 14.9
78.8 ( 14.0
57.2 ( 12.7
51.9 ( 11.3
53.9 ( 14.2
93.3 ( 20.0
23
24
25
87.9 ( 5.8
80.5 ( 25.5
a IC50 values were determined for compounds 8, 16, and 22.
did not reduce reporter activity, indicating that these potent
dual γ-secretase/PPARγ modulators did not affect NOTCH
processing or signaling in this concentration range (Figure 4).
Cytotoxicity. Cytotoxicity of all novel analogues was deter-
mined in the concentration range between 10 and 100 μM
using Alamar blue reagent and is summarized in the Support-
ing Information. With one exception (i.e., compound 16), none
of the compounds displayed detectable cytotoxicity up to
60 μM. Compound 16 showed 25% cytotoxicity at 60 μM
without significant effects at 40 μM, which is 6- to 7-fold above
its IC50 value for Aβ42 inhibition (6.0 μM).
which was used as a template for structural variations with a
special focus on the lipophilic parts of the molecule. Our SAR
studies revealed clear SAR in R-position to the carboxylic
acid, where the introduction of a n-butyl chain led to the
highest GSM activity. In contrast, the lipophilic backbone of
the molecule showed high tolerance toward structural modi-
fication. Introduction of different aromatic and aliphatic
moieties linked by aliphatic spacers was well tolerated and
showed activity on both γ-secretase and PPARγ. Substantial
improvement was finally achieved by the replacement of the
two phenyl residues with cyclohexyl (22) or by the attachment
of p-trifluoromethyl substituents (16). Cyclohexyl-substituted
22 showed nearly equipotent activities in the low micromolar
range with an Aβ42 IC50 of 5.1 μM and a PPARγ EC50 of
6.6 μM. The p-trifluoromethyl derivative had comparable
GSM activity with an Aβ42 IC50 of 6.0 μM, slightly weaker
PPARγactivity with an EC50 of 11 μM but the more promising
selectivity profile (IC50(COX-1) =16.9 μM, IC50(COX-2) >
40 μM). Importantly, these compounds did not impair
NOTCH processing in the noncytotoxic concentration range.
Taken together, this series represents the first class of
compounds described as dual γ-secretase/PPARγ modula-
tors. By structural modifications we were able to cover the
whole range from selective GSMs to equipotent and low
micromolar active dual γ-secretase/PPARγ modulators. A
major liability of the initially discovered GSMs in the class of
NSAIDs are COX-associated gastrointestinal and cardiovas-
cular side effects. These side effects most likely preclude the
long-term use of Aβ42-lowering NSAIDs to prevent AD.45
However, the here presented compounds show substantially
lower COX inhibition compared to Aβ42-lowering NSAIDs
such as ibuprofen (Figure 1). We propose that dual γ-secre-
tase/PPARγ modulators could provide a promising strategy
to address the increased dementia risk in patients with insulin
resistance and type 2 diabetes. Inthese individuals, dual-active
molecules might ameliorate the deleterious effects associated
with insulin resistance by activation of PPARγ and further
confront the elevated risk for AD by modulating γ-secretase
activity and lowering Aβ42 levels in brain.
COX Inhibition. Most NSAIDs-type GSMs such as in-
domethacin inhibit COX-1 and COX-2 with IC50 values in
the nanomolar range (Figure 1). Consequently, COX-related
side effects are a major obstacle for their long-term clinical
use. Thus, one of the main objectives of this study was to
develop a potent dual γ-secretase/PPARγ modulator with
substantially weaker COX inhibitory activity compared to
NSAID-type GSMs. Inhibition of isolated ovine COX-1 and
human recombinant COX-2 is summarized in Table 2 and
presented as remaining activity at 10 μM. Importantly, all novel
compounds synthesized within this study were weak COX-1
and COX-2 inhibitors with at least half-maximal remaining
activities at 10 μM. Furthermore, we have determined IC50
values for lead compound 8 as well as for the most active
compounds 16 and 22. Compound 8 displays moderate COX-1
inhibition with an IC50 of 18.7 μM and very weak COX-2
inhibition (IC50 > 40 μM). The introduction of cyclohexyl resi-
dues (compound 22: IC50(Aβ42) = 5.1 μM, EC50(PPARγ) =
6.6 μM) improved dual γ-secretase/PPARγ activity but also
caused slightly increased COX-1 inhibition and COX-2 inhibi-
tion (IC50(COX-1) = 13.4 μM, IC50(COX-2) = 10.9 μM).
p-Trifluoromethylphenyl-substituted 16 (IC50(Aβ42)=6.0 μM,
EC50(PPARγ) = 11 μM) shows weaker COX-1 inhibitory
activity (IC50=16.9 μM) while sparing inhibition of COX-2.
These results suggest an advantage of p-trifluorormethylphenyl
over cyclohexyl substitution in regard to COX selectivity. In
summary, the introduction of p-trifluoromethylphenyl residues
(16) resulted in the most favorable pharmacological profile
with superiority of GSM activity versus COX-inhibition.
Experimental Section
Discussion
Compounds and Chemistry. The structures of compounds
6-25 were confirmed by 1H NMR, 13C NMR, and mass spec-
trometry (ESI). The purities of the final compounds described
here were determined by combustion analysis and are 95% or
This study presents the successful establishment of a novel
and robust scaffold for potent dual γ-secretase/PPARγ mod-
ulators. The structural starting point was our initial hit 8,

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 12 4697
higher. Commercial chemicals and solvents were reagent grade
and used without further purification. 1H and 13C NMR spectra
were measured in DMSO-d6 or CDCl3 on a Bruker ARX 300
(1H NMR) and AC 200 E (13C NMR) spectrometer. Chemical
shifts are reported in parts per million (ppm) using tetramethyl-
silane (TMS) as internal standard. Mass spectra were obtained
on a Fissous Instruments VG Platform 2 spectrometer measur-
ing in the positive- or negative-ion mode (ESI-MS system).
Combustion analysis was performed by the Microanalytical
Laboratory of the Institute of Organic Chemistry and Chemical
Biology, Goethe University Frankfurt, on a Foss Heraeus
CHNO rapid elemental analyzer (for details see the Supporting
Information).
The general synthesis of compounds (Scheme 2) follows the
routes described recently.34 The synthetic procedure is described
representatively for compounds 8 and 16. Detailed synthesis and
analytical data of all compounds are provided in the Supporting
Information.
Step i. 2-Thiobarbituric acid (4.33 g/30 mmol, 1 equiv) was
suspended in anhydrous DMF (∼30 mL, quantum satis), and
triethylamine (4.56 g/45 mmol, 1.5 equiv) was added. Heating to
80-90 ꢀC yielded a clear solution, to which ethyl 2-bromohexa-
noate (10.05 g/45 mmol, 1.5 equiv) was added dropwise. After
being stirred for 4 h at 80 ꢀC (TLC control), the reaction mixture
was quenched with at least four parts of water and extracted
three times with ethyl acetate. Solvent evaporation of the
organic phase yielded the crude product, which was puri-
fied by column chromatography using hexane/ethyl acetate to
obtain 8a as a white solid. Yield: 55.9% (4.8 g).
Step iia. Compound 8b was prepared under Williamson
conditions. The precursor from step i (1.2 g/4.19 mmol, 1 equiv)
and (2-bromoethyl)benzene (1.63 g/8.8 mmol, 2.1 equiv) were
suspended in anhydrous DMF (quantum satis) and heated to
80-90 ꢀC. K2CO3 (1.34 g/9.72 mmol, 2.32 equiv) was added,
and the resulting suspension was stirred for 9 h. After comple-
tion of the reaction (TLC control), DMF was evaporated in
vacuo at 60 ꢀC and the remaining solid was diluted with water.
After neutralization to pH 7, the aqueous phase was extracted
two times with ethyl acetate. The combined organic fractions
were washed with brine and dried over MgSO4. Solvent eva-
poration gave the crude product, which was purified by column
chromatography using hexane/ethyl acetate to yield 8b as clear
oil. Yield: 63.7% (1.32 g).
Step iib. Compound 16b was synthesized under Mitsunobu
conditions. Thereby, 8a (0.45 g/1.6 mmol, 1 equiv), 4-trifluor-
methylphenylethanol (0.6 g/3.2 mmol, 2.1 equiv), and triphe-
nylphosphine (1.2 g/4.57 mmol, 2.5 equiv) were dissolved in
anhydrous THF and stirred under argon atmosphere with ice
bath cooling. Diethylazodicarboxylate (DEAD, 0.78 g/4.57 mmol,
2.5 equiv) diluted in 5 mL of THF was added dropwise via a
syringe, and the solution was stirred for 1 h until the reaction was
completed (TLC control). Subsequently, THF was evaporated and
the remaining residue was purified by column chromatography
using hexane/ethyl acetate to yield 16b as clear oil in 64.3% (0.64 g)
yield.
separated by column chromatography using n-hexane/ethyl
acetate. Yield: 92.2% (0.35 g).
Ethyl 2-(4,6-Dihydroxypyrimidin-2-ylthio)hexanoate (8a).
White solid, mp 173 ꢀC. 1H NMR (300.13 MHz, (CD3)2SO): δ =
0.82-0.87 (t, 3H, J = 6.9 Hz, CH3-Bu), 1.14-1.19 (t, 3H, J =
7.1 Hz, -CH3), 1.28-1.31 (m, 4H, CH2-Bu), 1.72-1.92 (m, 2H,
CH2-Bu), 4.08-4.15 (q, 2H, J = 7.1 Hz, OCH2), 4.47-4.52 (t,
1H, J = 7.1Hz, S-CH), 5.22 (s, 1H, Pyr-5H), 11.75 (s, br, 2H, Pyr-
OH). 13C NMR (75.44 MHz, (CD3)2SO): δ = 13.63 (-CH3), 13.89
(CH3-Bu), 21.59 (CH2-Bu), 28.49 (CH2-Bu), 31.27 (CH2-Bu),
46.39 (S-CH), 61.06 (OCH2), 85.62 (Pyr-C5), 167.79 (Pyr-C2),
170.86 (COO). MS (ESI-): m/e = 284.9 [M - 1]-.
Ethyl 2-(4,6-Diphenethoxypyrimidin-2-ylthio)hexanoate (8b).
Clear oil. 1H NMR (250.13 MHz, (CD3)2SO): δ = 0.82 (t, 3H,
J = 7.1 Hz, CH3-Bu), 1.05 (t, 3H, J = 7.1 Hz, -CH3), 1.26-1.34
(m, 4H, CH2-Bu), 1.74-1.91 (m, 2H, CH2-Bu), 2.98 (t, 4H, J =
7.1 Hz, Ph-CH2), 4.11 (q, 2H, J = 7.1 Hz, OCH2), 4.32-4.48 (m,
5H, S-CH2 þ Pyr-O-CH2), 5.89 (s, 1H, Pyr-5H), 7.19-7.32 (m,
10H, Ph-H). 13C NMR (75.44 MHz, (CD3)2SO): δ = 13.63
(CH3-Bu), 13.83 (-CH3), 21.60 (CH2-Bu), 28.79 (CH2-Bu),
30.72 (CH2-Bu), 34.43 (2C, Ph-CH2), 47.13 (S-CH2), 60.90
(OCH2), 67.12 (2C, Pyr-O-CH2), 86.01 (Pyr-C5), 126.34 (2C,
Ph-C4), 128,12 (4C, Ph-C2 þ -C6), 128.61 (4C, Ph-C3 þ -C5),
137.80 (2C, Ph-C1), 168.54 (Pyr-C2), 170.07 (2C, Pyr-C4 þ -C6),
171.35 (COO). MS (ESIþ): m/e = 495.2 [M þ H]þ.
Ethyl 2-(4,6-Bis(4-(trifluoromethyl)phenethoxy)pyrimidin-2-yl-
thio)hexanoate (16b). Clear oil. 1H NMR (300.13 MHz, (CD3)2-
SO): δ = 0.79-0.83 (t, 3H, J = 7.1 Hz, CH3-Bu), 1.02-1.06 (t,
3H, J = 7.0 Hz, -CH3), 1.29-1.40 (m, 4H, CH2-Bu), 1.84-1.92
(m, 2H, CH2-Bu), 3.07-3.11 (t, 4H, J = 6.6 Hz, Ph-CH2), 3.97-
4.08 (m, 2H, OCH2), 4.29-4.35 (t, 1H, J = 7.2 Hz, S-CH), 4.47-
4.53 (m, 4H, Pyr-O-CH2), 5.89 (s, 1H, Pyr-5H), 7.49-7.52
(d, 4H, J = 8.0 Hz, Ph-C2 þ -C6H), 7.63-7.66 (d, 4H, J = 8.1 Hz,
Ph-C3 þ -C5H). 13C NMR (75.44 MHz, (CD3)2SO): δ = 13.60
(CH3-Bu), 14.07 (-CH3), 21.60 (CH2-Bu), 28.79 (CH2-Bu), 30.64
(CH2-Bu), 34.15 (2C, Ph-CH2), 47.16 (S-CH), 60.88 (OCH2),
66.59 (2C, Pyr-O-CH2), 86.06 (Pyr-C5), 125.11 (4C, Ph-C3 þ -
C5), 126.13-127.35 (2C, Ph-CF3), 129.71 (4C, Ph-C2 þ -C6),
143.01 (2C, Ph-C1), 168.59 (Pyr-C2), 169.99 (2C, Pyr-C4 þ -C6),
171.31 (COO). MS (ESIþ): m/e = 631.6 [M þ 1]þ.
2-(4,6-Diphenethoxypyrimidin-2-ylthio)hexanoic Acid (8).
White solid, mp 99 ꢀC. 1H NMR (300.13 MHz, (CD3)2SO): δ =
0.81 (t, 3H, J = 7.1 Hz, CH3-Bu), 1.23-1.38 (m, 4H, CH2-Bu),
1.78-1.90 (m, 2H, CH2-Bu), 2.98 (t, 4H, J = 6.9 Hz, Ph-CH2),
4.29 (t, 1H, J = 7.2 Hz, S-CH), 4.46 (t, 4H, J = 6.9 Hz, Pyr-O-
CH2), 5.87 (s, 1H, Pyr-5H), 7.18-7.32 (m, 10H, Ph-H). 13C NMR
(75.44 MHz, (CD3)2SO): δ = 13.70 (CH3-Bu), 21.71 (CH2-Bu),
28.92 (CH2-Bu), 31.09 (CH2-Bu), 34.44 (2C, Ph-CH2), 47.61 (S-
CH2), 67.08 (2C, Pyr-O-CH2), 85.87 (Pyr-C5), 126.31 (2C, Ph-C4),
128.30 (4C, Ph-C2 þ -C6), 128.84 (4C, Ph-C3 þ -C5), 137.88 (2C,
Ph-C1), 169.01 (Pyr-C2), 170.03 (2C, Pyr-C4 þ -C6), 172.70
(COOH). MS (ESIþ): m/e = 467.0 [M þ H]þ.
2-(4,6-Bis(4-(trifluoromethyl)phenethoxy)pyrimidin-2-ylthio)-
hexanoic Acid (16). Yellow oil. 1H NMR(300.13MHz, (CD3)2SO):
δ = 0.78-0.83 (t, 3H, J = 7.1 Hz, CH3-Bu), 1.22-1.37 (m, 4H,
CH2-Bu), 1.74-1.92 (m, 2H, CH2-Bu), 3.06-3.11 (t, 4H, J = 6.6
Hz, Ph-CH2), 4.27-4.30 (t, 1H, J = 7.1 Hz, S-CH), 4.48-4.53 (t,
4H, J = 6.6 Hz, Pyr-O-CH2), 5.87 (s, 1H, Pyr-5H), 7.49-7.52 (d,
4H, J = 8.0 Hz, Ph-C2 þ -C6H), 7.63-7.66 (d, 4H, J = 8.0 Hz, Ph-
C3 þ -C5H), 12.82 (s, br, 1H, COOH). 13C NMR (75.44 MHz,
(CD3)2SO): δ= 13.64 (CH3-Bu), 21.67 (CH2-Bu), 28.92 (CH2-Bu),
30.91 (CH2-Bu), 34.16 (2C, Ph-CH2), 47.46(S-CH), 66.54 (2C, Pyr-
O-CH2), 85.97 (Pyr-C5), 124.99 (4C, Ph-C3 þ -C5), 125.14-127.74
(2C, Ph-CF3), 129.69(4C, Ph-C2 þ -C6), 143.06 (2C, Ph-C1), 168.95
(Pyr-C2), 169.97 (2C, Pyr-C4 þ -C6), 172.23 (COOH). MS (ESI-):
m/e = 601.4 [M - 1]-.
Step iii. The corresponding ester (8b (0.84 g/1.7 mmol,
1 equiv) or 16b (0.4 g/0.6 mmol, 0.6 equiv)) was dissolved in a
mixture of 5 mL of THF/10 mL of MeOH, and a solution of
LiOH H2O (3 equiv) in 3 mL of H2O was added. After the
3
mixture was stirred at 50 ꢀC until saponification was completed,
the solvent was removed and the residue was dissolved in water
(under heating; if necessary, low amounts of MeOH were
added). The solution was acidified with diluted hydrochloric acid.
The formed precipitate (8) was filtered, washed to neutrality with
water, and then washed with n-hexane. Recrystallization from
n-hexane/ethyl acetate yielded compound 8 as a white solid. Yield:
78.4% (0.62 g).
PPAR Transactivation Assay. COS7 cells were grown in
DMEM supplemented with FCS, sodium pyruvate, and penicil-
lin/streptomycine at 37 ꢀC and 5% CO2. The day before transfec-
tion, cells were seeded in 96-well plates at a density of 30 000 cells
In the case of compound 16, the precipitate formed after
addition of diluted hydrochloric acid was oily. To purify the
crude product, the residue was extracted with ethyl acetate and

4698 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 12
Hieke et al.
per well. Transient transfection was carried out by Lipofectamine
2000 reagent (Invitrogen) according to the manufacturer’s proto-
col with pFR-Luc (Stratagene), pRL-SV40 (Promega), and the
Gal4-fusion receptor plasmids (pFA-CMV-hPPAR-LBD) of the
respective subtype. At 5 h after transfection, the medium was
changed to DMEM without phenol red and FCS, containing 0.1%
DMSO and the respective concentrations of the test compounds.
Following overnight incubation with the test compounds,
cells were assayed for reporter gene activity using Dual-Glo
luciferase assay system (Promega) according to the manufac-
turer’s protocol. Luminescence was measured with a GENios
Pro luminometer (Tecan Deutschland GmbH, Crailsheim,
Germany). Each concentration of the compounds was tested
in triplicate wells, and each experiment was repeated indepen-
dently at least three times. Normalization for transfection
efficacy and cell growth was done by division of the firefly
luciferase data by renilla luciferase data resulting in relative light
units. Activation factors were obtained by dividing by DMSO
control. EC50 and standard deviation values were calculated by
mean values of at least three determinations by SigmaPlot 2001
(Systat Software GmbH, Erkrath, Germany) using a four-
parameter logistic regression. All compounds were evaluated
by comparison of the achieved maximum effect to that of the
reference compound (pioglitazone for PPARγ, GW 7647 for
PPARR, and L165,041 for PPARδ each with 1 μM).
Determination of γ-Secretase Modulator Activity. To charac-
terize the GSM activity of novel analogues, their effects on the
generation of Aβ40, Aβ42, and Aβ38 peptides were determined
in a previously described cell-based ELISA assay with modi-
fications.37 CHO cells with stable overexpression of wild type
human amyloid precursor protein and wild type human pre-
senilin-137 were maintained in DMEM supplemented with 10%
FBS and treated in 96-well plates for 24 h with increasing
concentrations of the respective compounds or DMSO vehicle.
Culture media were collected and analyzed by ELISA as follows:
monoclonal antibody IC16 raised against amino acids 1-15 of
the Aβ sequence was used as a capture antibody.46 To distin-
guish different Aβ species, C-terminal antibodies specific for
Aβ40, Aβ42, and Aβ38 labeled with horseradish peroxidase
(HRP) using the Pierce EZ-Link Plus activated peroxidase kit
(Thermo Fisher Scientific) were used for detection. The 96-well
high-binding microtiter plates were coated overnight at 4 ꢀC
with capture antibody IC16 diluted 1:250 in PBS, pH 7.2.
Capture antibody was removed, and conditioned media samples
(10 μL for detection of Aβ40, 100 μL for Aβ42, 50 μL for Aβ38)
or standard peptides were loaded. HRP-coupled detection anti-
bodies diluted in assay buffer (PBS containing 0.05% Tween-20,
1% BSA) were added to each well and incubated overnight at
4 ꢀC. Plates were washed 3 times with PBS containing 0.05%
Tween-20 and once with PBS. Then 50 μL of TMB ultrasub-
strate (Thermo Fisher Scientific) was added and incubated for
1-10 min at room temperature in the dark, and the reaction was
stopped by adding 50 μL of 2 M H2SO4. Absorbance at 450 nm
was recorded with a Paradigm microplate reader (Beckman-
Coulter). Synthetic Aβ40, Aβ42, and Aβ38 peptides (Bachem
AG) were used to generate standard curves. These Aβ peptides
were solubilized in DMSO at 1 mg/mL, aliquoted, and stored
frozen at -80 ꢀC. Immediately before use, peptides were diluted
in assay buffer to 250-3000 pg/mL. Triplicate measurements
from each drug concentration were averaged and normalized to
DMSO control condition. For calculation of IC50 values, cells
were treated with eight increasing concentrations of each com-
pound, and a nonlinear curve fit with variable slope model was
applied to the results from two to four independent experiments.
Statistics were performed using GraphPad Prism (GraphPad
Software).
DNA-binding/VP16 transactivation domain and the MH100
reporter plasmid encoding firefly luciferase under the UAS
promotor (50 ng each) using GeneJuice transfection reagent
(Merck Chemicals Ltd.). Co-transfection of the MH100 repor-
ter with empty pcDNA3 plasmid served as a negative control.
Then 5 ng of plasmid pRL-TK encoding renilla luciferase was
added to the plasmid mix to control for transfection efficiency.
At 24 h after the transfection, cells were treated for an additional
24 h with increasing concentrations of the compounds 16 and
22, 0.5 μM γ-secretase inhibitor LY-411575, or DMSO vehicle.
The cells were then lysed, and firefly and renilla luciferase
activities were quantified using the Dual Glo luciferase assay
system (Promega) and a PARADIGM microplate reader
(Beckman-Coulter). Normalization for transfection efficacy
and cell growth was achieved by dividing the firefly luciferase
values by the renilla luciferase values. Percent activation of the
reporter was then calculated by normalization of triplicate mea-
surements from each condition to the DMSO control values.
Unpaired t test analysis was performed from two independent
experiments using GraphPad Prism: ///, p < 0.001.
Cytotoxicity Assay. CHO cells with stable overexpression
of wild type human amyloid precursor protein and wild type
human presenilin-1 were seeded at low density in 96-well
plates (4000 cells/well) and cultured for 24 h. The cells
were then treated in duplicate with increasing concentrations
(0-100 μM) of the respective compounds or DMSO vehicle
for an additional 24 h. Cell viability was assessed using the
Alamar blue reagent (Invitrogen). Then 20 μL of Alamar blue
was added to cells cultured in 200 μL of medium and incubated
overnight. Absorbance was measured with a Paradigm micro-
plate reader (Beckman-Coulter) at 570 nm, using 600 nm as
the reference wavelength. The values expressed as percent
viability of vehicle control represent averages of two indepen-
dent experiments.
Determination of COX Inhibition. Inhibition of the activities
of isolated ovine COX-1 and human recombinant COX-2 was
performed as described.34 Briefly, purified COX-1 (ovine, 50
units) or COX-2 (human recombinant, 20 units) were diluted in
1 mL of reaction mixture containing 100 mM Tris buffer, pH 8, 5
mM glutathione, 5 μM hemoglobin, and 100 μM EDTA at 4 ꢀC
and preincubated with the test compounds for 5 min. Samples
were prewarmed for 60 s at 37 ꢀC, and arachidonic acid (5 μM
for COX-1, 2 μM for COX-2) was added to start the reaction.
After 5 min at 37 ꢀC, the COX product 12-HHT was extracted
and then analyzed by HPLC as described.41,47
€
Acknowledgment. We thank Dr. Jan Naslund (Karolinska
Institute, Sweden) for the NOTCH reporter plasmids,
Dr. Manfred Brockhaus (F. Hoffmann-La Roche Ltd.,
Switzerland) for carboxyl-terminus-specific Aβ antibodies,
€
€
Dr. Michael Lammerhofer (Universitat Wien, Austria) for
enantioselective separation of compound 8, and Dr. Boris
Schmidt (TU Darmstadt, Germany) for scientific advice. This
study was supported by the Competence Network Degenera-
tive Dementias of the Federal Ministry of Education (Grant
01 GI 0718 to S.W.).
Supporting Information Available: Chemical synthesis, 1H
and 13C NMR data of intermediates and final compounds, mass
spectrometry, combustion analysis, and cytotoxicity data. This
material is available free of charge via the Internet at http://
pubs.acs.org.
References
(1) Brookmeyer, R.; Johnson, E.; Ziegler-Graham, K.; Arrighi, H. M.
Forecasting the global burden of Alzheimer’s disease. Alzheimer’s
Dementia 2007, 3, 186–191.
NOTCH Reporter Assay. The NOTCH reporter assay was
performed as described.38 In brief, subconfluent CHO cells were
transiently transfected in 96-well plates with plasmid pCDNA3-
Notch-ΔE-GVP encoding truncated NOTCH fused to a Gal4
(2) Selkoe, D. J. Alzheimer’s disease: genes, proteins, and therapy.
Physiol. Rev. 2001, 81, 741–766.

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 12 4699
(3) Walsh, D. M.; Selkoe, D. J. Aβ oligomers;a decade of discovery.
J. Neurochem. 2007, 101, 1172–1184.
(4) Citron, M. Strategies for disease modification in Alzheimer’s
(24) Luchsinger, J. A. Adiposity, hyperinsulinemia, diabetes and
Alzheimer’s disease: an epidemiological perspective. Eur. J. Phar-
macol. 2008, 585, 119–129.
disease. Nat. Rev. Neurosci. 2004, 5, 677–685.
(5) Golde, T. E. Disease modifying therapy for AD? J. Neurochem.
2006, 99, 689–707.
(25) Bernardo, A.; Minghetti, L. PPAR-gamma agonists as regulators
of microglial activation and brain inflammation. Curr. Pharm. Des.
2006, 12, 93–109.
(6) Barten, D. M.; Meredith, J. E.; Zaczek, R.; Houston, J. G.;
Albright, C. F. Gamma-secretase inhibitors for Alzheimer’s dis-
ease: balancing efficacy and toxicity. Drugs R&D 2006, 7, 87–97.
(7) De Strooper, B.; Annaert, W.; Cupers, P.; Saftig, P.; Craessaerts,
K.; Mumm, J. S.; Schroeter, E. H.; Schrijvers, V.; Wolfe, M. S.;
Ray, W. J.; Goate, A.; Kopan, R. A presenilin-1-dependent
gamma-secretase-like protease mediates release of Notch intracel-
lular domain. Nature 1999, 398, 518–522.
(8) Wong, G. T.; Manfra, D.; Poulet, F. M.; Zhang, Q.; Josien, H.;
Bara, T.; Engstrom, L.; Pinzon-Ortiz, M.; Fine, J. S.; Lee, H. J.;
Zhang, L.; Higgins, G. A.; Parker, E. M. Chronic treatment with
the gamma-secretase inhibitor LY-411,575 inhibits beta-amyloid
peptide production and alters lymphopoiesis and intestinal cell
differentiation. J. Biol. Chem. 2004, 279, 12876–12882.
(9) Weggen, S.; Eriksen, J. L.; Das, P.; Sagi, S. A.; Wang, R.; Pietrzik,
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 Abeta42 independently of
cyclooxygenase activity. Nature 2001, 414, 212–216.
(10) Leuchtenberger, S.; Beher, D.; Weggen, S. Selective modulation of
Abeta42 production in Alzheimer’s disease: non-steroidal anti-
inflammatory drugs (NSAIDs) and beyond. Curr. Pharm. Des.
2006, 12, 4337–4355.
(11) Weggen, S.; Rogers, M.; Eriksen, J. NSAIDs: small molecules for
prevention of Alzheimer’s disease or precursors for future drug
development? Trends Pharmacol. Sci. 2007, 28, 536–543.
(12) Green, R. C.; Schneider, L. S.; Hendrix, S. B.; Zavitz, K. H.;
Swabb, E. Safety and Efficacy of Tarenflurbil in Subjects with Mild
Alzheimer’s Disease: Results from an 18-Month Multi-Center
Phase 3 Trial. Presented at the International Conference on
Alzheimer’s Disease, Chicago, IL, July 2008; Alzheimer’s and
Dementia; No. O3-04-01, T165.
(26) Combs, C. K.; Johnson, D. E.; Karlo, J. C.; Cannady, S. B.;
Landreth, G. E. Inflammatory mechanisms in Alzheimer’s disease:
inhibition of beta-amyloid-stimulated proinflammatory responses
and neurotoxicity by PPARgamma agonists. J. Neurosci. 2000, 20,
558–567.
(27) Heneka, M. T.; Sastre, M.; Dumitrescu-Ozimek, L.; Hanke, A.;
Dewachter, I.; Kuiperi, C.; O’Banion, K.; Klockgether, T.; Van
Leuven, F.; Landreth, G. E. Acute treatment with the PPARgam-
ma agonist pioglitazone and ibuprofen reduces glial inflammation
and Abeta1-42 levels in APPV717I transgenic mice. Brain 2005,
128, 1442–1453.
(28) Camacho, I. E.; Serneels, L.; Spittaels, K.; Merchiers, P.;
Dominguez, D.; De Strooper, B. Peroxisome proliferator-activated
receptor gamma induces a clearance mechanism for the amyloid-
beta peptide. J. Neurosci. 2004, 24, 10908–10917.
(29) Sastre, M.; Dewachter, I.; Rossner, S.; Bogdanovic, N.; Rosen, E.;
Borghgraef, P.; Evert, B. O.; Dumitrescu-Ozimek, L.; Thal, D. R.;
Landreth, G.; Walter, J.; Klockgether, T.; van Leuven, F.; Heneka,
M. T. Nonsteroidal anti-inflammatory drugs repress {beta}-secre-
tase gene promoter activity by the activation of PPAR{gamma}.
Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 443–448.
(30) Nicolakakis, N.; Aboulkassim, T.; Ongali, B.; Lecrux, C.;
Fernandes, P.; Rosa-Neto, P.; Tong, X. K.; Hamel, E. Complete rescue
of cerebrovascular function in aged Alzheimer’s disease transgenic mice
by antioxidants and pioglitazone, a peroxisome proliferator-activated
receptor gamma agonist. J. Neurosci. 2008, 28, 9287–9296.
(31) Rabiner, E. A.; Tzimopoulou, S.; Cunningham, V. J.; Jeter, B.;
Zvartau-Hind, M.; Castiglia, M.; Mistry, P.; Bird, N. P.;
Matthewsw, J.; Whitcher, B.; Nichols, T. E.; Lai, R.; Lotay, N.;
Saunders, A.; Reiman, E.; Chen, K.; Gold, M.; Matthews, P. M.
Effects of 12 months of treatment with the PPARgamma agonist
rosiglitazone on brain glucose metabolism in Alzheimer’s disease: a
18F-FDG PET study. Alzheimer’s Dementia 2009, 5, 207.
(32) Sato, T.; Hanyu, H.; Hirao, K.; Kanetaka, H.; Sakurai, H.;
Iwamoto, T. Efficacy of PPAR-gamma agonist pioglitazone in
mild Alzheimer disease. Neurobiol. Aging [Online early access]. DOI:
10.1016/j.neurobiolaging.2009.10.009. Published Online: Nov 17,
2009.
(13) in ’t Veld, B. A.; Ruitenberg, A.; Hofman, A.; Launer, L. J.;
van Duijn, C. M.; Stijnen, T.; Breteler, M. M.; Stricker, B. H.
Nonsteroidal antiinflammatory drugs and the risk of Alzheimer’s
disease. N. Engl. J. Med. 2001, 345, 1515–1521.
(14) Vlad, S. C.; Miller, D. R.; Kowall, N. W.; Felson, D. T. Protective
effects of NSAIDs on the development of Alzheimer disease.
Neurology 2008, 70, 1672–1677.
(15) Peretto, I.; La Porta, E. Gamma-secretase modulation and its
promise for Alzheimer’s disease: a medicinal chemistry perspective.
Curr. Top. Med. Chem. 2008, 8, 38–46.
(33) Maeshiba, Y.; Kiyota, Y.; Yamashita, K.; Yoshimura, Y.;
Motohashi, M.; Tanayama, S. Disposition of the new antidiabetic
agent pioglitazone in rats, dogs, and monkeys. Arzneim. Forsch.
1997, 47, 29–35.
(16) Page, R. M.; Baumann, K.; Tomioka, M.; Perez-Revuelta, B. I.;
Fukumori, A.; Jacobsen, H.; Flohr, A.; Luebbers, T.; Ozmen, L.;
Steiner, H.; Haass, C. Generation of Abeta 38 and Abeta 42 is
independently and differentially affected by FAD-associated pre-
senilin 1 mutations and gamma-secretase modulation. J. Biol.
Chem. 2008, 283, 677–683.
(17) Jaradat, M. S.; Wongsud, B.; Phornchirasilp, S.; Rangwala, S. M.;
Shams, G.; Sutton, M.; Romstedt, K. J.; Noonan, D. J.; Feller, D. R.
Activation of peroxisome proliferator-activated receptor isoforms and
inhibition of prostaglandin H(2) synthases by ibuprofen, naproxen, and
indomethacin. Biochem. Pharmacol. 2001, 62, 1587–1595.
(34) Koeberle, A.; Zettl, H.; Greiner, C.; Wurglics, M.; Schubert-
Zsilavecz, M.; Werz, O. Pirinixic acid derivatives as novel dual
inhibitors of microsomal prostaglandin E2 synthase-1 and 5-lipoxy-
genase. J. Med. Chem. 2008, 51, 8068–8076.
(35) Mitsunobu, O. The use of diethyl azodicarboxylate and triphenyl-
phosphine in synthesis and transformation of natural products.
Synthesis 1981, 1, 1–28.
(36) Zettl, H.; Steri, R.; Lammerhofer, M.; Schubert-Zsilavecz, M.
Discovery of a novel class of 2-mercaptohexanoic acid derivatives
as highly active PPARalpha agonists. Bioorg. Med. Chem. Lett.
2009, 19, 4421–4426.
(18) Lehmann, J. M.; Lenhard, J. M.; Oliver, B. B.; Ringold, G. M.;
Kliewer, S. A. Peroxisome proliferator-activated receptors alpha
and gamma are activated by indomethacin and other non-steroidal
anti-inflammatory drugs. J. Biol. Chem. 1997, 272, 3406–3410.
(19) Lehrke, M.; Lazar, M. A. The many faces of PPARgamma. Cell
2005, 123, 993–999.
(37) Czirr, E.; Leuchtenberger, S.; Dorner-Ciossek, C.; Schneider, A.;
Jucker, M.; Koo, E. H.; Pietrzik, C. U.; Baumann, K.; Weggen, S.
Insensitivity to Abeta 42-lowering non-steroidal anti-inflamma-
tory drugs (NSAIDs) and gamma-secretase inhibitors is common
among aggressive presenilin-1 mutations. J. Biol. Chem. 2007, 282,
24504–24513.
(20) Landreth, G.; Jiang, Q.; Mandrekar, S.; Heneka, M. PPARgamma
agonists as therapeutics for the treatment of Alzheimer’s disease.
Neurotherapeutics 2008, 5, 481–489.
(21) Ho, L.; Qin, W.; Pompl, P. N.; Xiang, Z.; Wang, J.; Zhao, Z.; Peng,
Y.; Cambareri, G.; Rocher, A.; Mobbs, C. V.; Hof, P. R.; Pasinetti,
G. M. Diet-induced insulin resistance promotes amyloidosis in a
transgenic mouse model of Alzheimer’s disease. FASEB J. 2004, 18,
902–904.
(22) Takeda, S.; Sato, N.; Uchio-Yamada, K.; Sawada, K.; Kunieda,
T.; Takeuchi, D.; Kurinami, H.; Shinohara, M.; Rakugi, H.;
Morishita, R. Diabetes-accelerated memory dysfunction via cere-
brovascular inflammation and Abeta deposition in an Alzheimer
mouse model with diabetes. Proc. Natl. Acad. Sci. U.S.A. 2010,
107, 7036–7041.
(23) Biessels, G. J.; Staekenborg, S.; Brunner, E.; Brayne, C.; Scheltens,
P. Risk of dementia in diabetes mellitus: a systematic review.
Lancet Neurol. 2006, 5, 64–74.
(38) Karlstrom, H.; Bergman, A.; Lendahl, U.; Naslund, J.; Lundkvist,
J. A sensitive and quantitative assay for measuring cleavage of
presenilin substrates. J. Biol. Chem. 2002, 277, 6763–6766.
(39) Farce, A.; Renault, N.; Chavatte, P. Structural insight into PPAR-
gamma ligands binding. Curr. Med. Chem. 2009, 16, 1768–1789.
(40) 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 Abeta42 production.
Nat. Med. 2005, 11, 545–550.
(41) Popescu, L.; Rau, O.; Bottcher, J.; Syha, Y.; Schubert-Zsilavecz,
M. Quinoline-based derivatives of pirinixic acid as dual PPAR
alpha/gamma agonists. Arch. Pharm. (Weinheim, Ger.) 2007, 340,
367–371.
(42) Rau, O.; Syha, Y.; Zettl, H.; Kock, M.; Bock, A.; Schubert-
Zsilavecz, M. Alpha-alkyl substituted pirinixic acid derivatives as

4700 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 12
Hieke et al.
potent dual agonists of the peroxisome proliferator activated
receptor alpha and gamma. Arch. Pharm. 2008, 341, 191–195.
(43) Zettl, H.; Dittrich, M.; Steri, R.; Proschak, O.; Rau, O.; Steinhilber,
D.; Schneider, G.; Lammerhofer, M.; Schubert-Zsilavecz, M.
Novel pirinixic acids as PPARalpha preferential dual PPARalpha/
gamma agonists. QSAR Comb. Sci. 2009, 28, 576–586.
(44) Lammerhofer, M.; Pell, R.; Mahut, M.; Richter, M.; Schiesel, S.;
Zettl, H.; Dittrich, M.; Schubert-Zsilavecz, M.; Lindner, W.
Enantiomer separation and indirect chromatographic absolute con-
figuration prediction of chiral pirinixic acid derivatives: limitations of
polysaccharide-type chiral stationary phases in comparison to chiral
anion-exchangers. J. Chromatogr., A 2010, 1217, 1033–1040.
(46) Jager, S.; Leuchtenberger, S.; Martin, A.; Czirr, E.; Wesselowski,
J.; Dieckmann, M.; Waldron, E.; Korth, C.; Koo, E. H.; Heneka,
M.; Weggen, S.; Pietrzik, C. U. Alpha-secretase mediated conver-
sion of the amyloid precursor protein derived membrane stub C99
to C83 limits Abeta generation. J. Neurochem. 2009, 111, 1369–
1382.
(47) Kato, M.; Nishida, S.; Kitasato, H.; Sakata, N.; Kawai, S.
Cyclooxygenase-1 and cyclooxygenase-2 selectivity of non-steroi-
dal anti-inflammatory drugs: investigation using human peripheral
monocytes. J. Pharm. Pharmacol. 2001, 53, 1679–1685.
(48) Demello, K. L.; Ricketts, A. P. Cox-2 Selective Carprofen for
Treating Pain and Inflammation in Dogs. US2003212123 (A1),
2003.
(45) Meinert, C. L.; McCaffrey, L. D.; Breitner, J. C. Alzheimer’s
disease anti-inflammatory prevention trial: design, methods, and
baseline results. Alzheimer’s Dementia 2009, 5, 93–104.
€
(49) Pairet, M., van Ryn, J., Eds. COX-2 Inhibitors, 1st ed.; Birkhauser:
Basel, Switzerland, 2004.