Substituted thiazolamide coupled to a redox delivery system: A new γ-secretase inhibitor with enhanced pharmacokinetic profile

Article abstract of DOI:10.1039/b415090b

Inhibition of γ-secretase, one of the enzymes responsible for the cleavage of the amyloid precursor protein (APP) to produce pathogenic Aβ peptides, is an attractive approach for the treatment of Alzheimer's disease. We have designed a new γ-secretase thi

Full text of DOI:10.1039/b415090b

View Article Online / Journal Homepage / Table of Contents for this issue
A R T I C L E
Substituted thiazolamide coupled to a redox delivery system: a new
c-secretase inhibitor with enhanced pharmacokinetic profile
Younes Laras,a Gilles Que´le´ver,a Ce´drik Garino,a Nicolas Pietrancosta,a Mahmoud Sheha,b
Fre´de´ric Bihel,c Michael S. Wolfec and Jean-Louis Kraus*a
a INSERM U-623, Institut de Biologie du De´veloppement de Marseille (CNRS - INSERM -
Universite´ de la Me´diterrane´e), Laboratoire de Chimie Biomole´culaire, Faculte´ des Sciences de
Luminy, case 907, 13288 Marseille Cedex 09, France. E-mail: kraus@luminy.univ-mrs.fr;
Fax: (33) 491 82 94 16; Tel: (33) 491 82 91 41
b Medicinal Chemistry Department, Faculty of Pharmacy, Assiut University, Assiut, 71526,
Egypt
c Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, 02115, USA
Received 9th September 2004, Accepted 3rd December 2004
First published as an Advance Article on the web 10th January 2005
Inhibition of c-secretase, one of the enzymes responsible for the cleavage of the amyloid precursor protein (APP) to
produce pathogenic Ab peptides, is an attractive approach for the treatment of Alzheimer’s disease. We have designed
a new c-secretase thiazolamide inhibitor bearing a dihydronicotinoyl moiety as Redox Delivery System which allows
specific delivery of the drug to the brain. Through, on the one hand, Ab peptide production measurements by specific
in vitro assays (c-secretase Cell Free assay and Cell Based assay on HEK 293 APP transfected cells) and, on the other
hand, pharmacokinetic studies on animal models, the new inhibitor shows a good pharmacokinetic profile as well as
a potent c-secretase inhibitory activity in vitro. From the obtained results, it is expected that drug 2 will be mainly
delivered to the CNS with low diffusion in the peripheral tissues. Consequently the side effects of this c-secretase
inhibitor on the immune cells could be reduced.
This methodology increases intracranial concentrations of a
variety of compounds, including neurotransmitters, antibiotics
Introduction
Distribution of drugs to the Central Nervous System (CNS) is
one of the major problems of current therapies of brain diseases.
Difficulties in crossing the Blood Brain Barrier (BBB) often
impair the efficacy of valuable drugs. In the past few years, most
of the attempts to overcome this drawback have targeted the
amelioration of the lipophilic properties through the preparation
of prodrugs by formation of reversible linkages with suitable
groups.1,2 More recently, transporters that are involved in drug
transfer across the BBB have been molecularly identified.3
Other strategies have been investigated, such as:
and antineoplastic agents.
One of the major fields of application of CDDS is the
neurodegenerative diseases and particularly chemotherapy for
Alzheimer’s disease (AD). The available cholinergic therapies
target essentially late aspects of AD, improving temporarily the
performance of the undamaged neurones, but do not stop the
progressive mental decline. In the past years, important progress
has been made in the understanding of the pathogenic mecha-
nism of AD. New therapeutic targets have become available and
that should allow the underlying disease process to be tackled
directly. In this respect, the “amyloid hypothesis” has become
the dominant theory in this field. It is now believed that Ab
accumulation in plaques or as partial soluble filaments initiates
a pathological cascade leading to tangle formation,11 neuronal
dysfunction and possibly inflammation and oxidative damage,
with neurodegeneration and dementia as the final outcome.12
Two enzymes known as b- and c-secretases which cleave the
Ab precursor protein (APP) to yield Ab peptide are now the
potential therapeutic targets. Indeed, it is believed that lowering
Ab production will decrease the formation rate of senile plaques
in Alzheimer’s disease patients.13 Today, one of the dominant
strategies currently being pursued is the development of a drug
that inhibits the activity of b- or c-secretases and thereby reduces
the production of Ab peptide in the brain. Numerous very
potent inhibitors of c-secretase have been described so far. N-
[N-(3,5-Difluorophenacetyl)-(S)-alanyl]-(S)-phenylglycine tert-
butyl ester (DAPT) has been shown to dose-dependently reduce
Ab levels in the brains of two different strains of APP-transgenic
mice after only a single dose injection.14 A new series of
highly potent benzodiazepine c-secretase inhibitors active at
the picomolar range have been also described.15 Studies which
came out from the use of benzodiazepine c-secretase inhibitors
demonstrate their ability to chronically reduce Ab production.
Nevertheless, these studies suggest also caution in the testing
of c-secretase inhibitors in humans and indicate specific side
– liposomes targeted to the brain by exploiting receptor
mediated transcytosis systems;4
– nanoparticules for drug delivery across the BBB;5
– implantation within the brain of either genetically engi-
neered cells secreting a drug or a polymeric matrix or reservoir
containing the drug;6
– neuroproteomic approaches and gene therapy for CNS
disorders;7
– chimeric peptide technology, where a non-transportable
drug is conjugated to a BBB transport vector.8
In addition to these promising strategies, Chemical Drug
Delivery Systems (CDDS) still represent a systematic way of
targeting active biological molecules to specific target sites
or organs based on predictable enzymatic activation.9 One of
these systems called Redox Chemical Delivery Systems (RCDS)
involves the linking of a drug to a lipophilic dihydropyridine
carrier, creating a complex that after systemic administration
readily transverses the BBB because of its lipophilicity. Once
inside the brain parenchyma, the dihydropyridine moiety is
enzymatically oxidized to the corresponding ionic pyridinium
salt. The acquisition of charge has the dual effect of accelerating
the rate of systemic elimination by the kidney and bile and
trapping the drug pyridinium salt complex inside the brain.
Subsequent cleavage of the drug from the pyridinium carrier
leads to sustained drug delivery in the brain parenchyma.10
6 1 2
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 6 1 2 – 6 1 8
T h i s j o u r n a l i s
T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 5
©

View Article Online
effects that should be monitored in any future clinical trials,
mostly because Notch and the amyloid precursor protein are
cleaved by similar c-secretase.16
of pyridine as solvent for this reaction catalyses the nucleophilic
attack. It also acts as a weak base which protects the N-Boc acid-
labile function against acidolysis. During this kind of coupling
reaction activated by POCl3, we confirmed by chiral HPLC
analysis that no racemization occurred.21 The protecting group
of the N-acylated thiazolamine 4 was released according to
the conventional TFA deprotection methodology affording the
corresponding TFA salt 5 in quantitative yield. Acylation of
this deprotected amino acid 5 by 3,5-difluorophenylacetic acid
by using BOP as coupling agent led to a 53% yield of acylated
derivative 6. The nitro function was then reduced in the presence
of powdered iron in acidic conditions as the reducing system.22
The nicotinoyl analogue 8 of this aniline was isolated after N-
acylation of 7 by the hydrochloride salt of nicotinoyl chloride
in pyridine. It was then methylated by using iodomethane into
its iodide salt 1 in 39% yield for the last two steps. The methyl
pyridinium moiety was reduced into its dihydropyridine form
by using sodium dithionite in basic conditions, affording the
Bodor-like target molecule 2.23
As it has been demonstrated that DAPT severely interferes
with Notch signaling in zebrafish embryos,17 it could be hypoth-
esized that unwanted side-effects from the control of cellular
differentiation could be observed during long-term treatment
in humans with c-secretase inhibitory drugs. Moreover, several
potential liabilities seem to limit the potential promise of c-
secretase inhibitors mainly in the T cell development.18
A
possible way to overcome or to minimize these rather severe side-
effects is to find drugs that are more rapidly and more effectively
delivered to the CNS in order to reduce the concentration of the
drug in the peripheral tissues by lowering the amount of drug to
be administrated in vivo.
In this paper, we report the design, synthesis and biologi-
cal properties of new pseudopeptide thiazolamide derivatives
with c-secretase inhibitory properties. As shown in Fig. 1, a
dihydropyridine RCDS was also introduced onto these new
analogues in order to improve their BBB permeation. These
derivatives, the syntheses of which have required original
synthetic schemes, were assayed as c-secretase inhibitors in cell
free and cell based assays and their brain specific in vivo delivery
was investigated.
Biology
In vitro c-secretase cell free assay assay
The new compounds 1 and 2 were first tested for their ability
to block specifically in vitro c-secretase. This in vitro assay was
performed using solubilized c-secretase prepared essentially as
described by Li et al.24 This assay which uses solubilized c-
secretase and C100 flag as substrates reproduces both major
cleavage events (giving rise to the Ab40 and Ab42 termini) that
are ascribed to c-secretase activity in cells. It should be re-
called that to catalytically recover competent soluble c-secretase
the choice of detergent used for the membrane extraction is
critical. CHAPSO (3-[(3-cholamidopropyl)dimethylammonio]-
2-hydroxy-1-propanesulfonate) or CHAPS (3-[(3-cholamido-
propyl)dimethylammonio]-1-propanesulfonate) yield the active
enzyme. This methodology is now widely used as standard
in vitro c-secretase assay.25 Moreover, it is now well established
that c-secretase activity is catalyzed by a Presinilin-1 (PS-1)
containing macromolecular complexes.
In this in vitro assay, the new tested analogues 1, 2 and 7
exhibit very similar inhibition of Ab-peptide production (EC50
values ranging from 0.1 to 1.0 lM).
Fig. 1
In vitro cell based assay
Chemistry
The new compounds 1 and 2 were first tested for their ability to
inhibit Ab peptide production in a cell based assay according to
a well known procedure using APP transfected HEK 293 cells.26
This cell based assay confirms the efficiency of new analogues
bearing a substituted 2-amino-thiazole moiety linked to the
difluorophenylacetyl-alanyl moiety, to inhibit the production of
total Ab peptide (Ab40 + Ab42). The results are summarized
in Table 1. As it can be seen, in our assay, the inhibitory
potency for analogue 2 was found to be slightly better than
for analogue 1. The assay was validated by using DAPT as c-
secretase reference inhibitor. This result could be explained by
the lower lipophilicity of the ionic derivative 1 compare to that of
the neutral dihydropyridine 2, as illustrated by their calculated
log P values (Table 1).
The synthesis of the target molecule was performed as described
in Scheme 1. The 2-thiazolamine, hydrobromide salt, 3 was the
result of a modified conventional Hantzsch cyclocondensation
achieved from 1-bromo-4ꢀ-nitroacetophenone and thiourea in
refluxed ethanol.19 The hydrobromide salt 3 was isolated quan-
titatively and in high purity by simple filtration and it was used
for the next synthetic step without any further purification.
This thiazolamine 3 was then acylated by the N-Boc protected
amino acid (S)-alanine. Various methodologies were used to
perform this acylation step. Unfortunately, conventional peptide
coupling methods were almost completely ineffective. Indeed,
the classical BOP reagent or the DCC–HOBt coupling system
led to very poor yields in desired acylated derivative 4, 19 and
5% yields respectively. No reaction was observed even after
24 hours by using uronium coupling reagent HBTU. These
results highlight that N-acylation of such a weakly nucleophilic
heterocyclic amine is not a straightforward reaction which can
be achieved under any standard coupling conditions. The best
results were obtained with the non-classical coupling system
POCl3 in pyridine.20 This methodology led to a 77% yield
in BocAla derivative 4,21 according to a known mechanism20
involving the in situ formation of a mixed anhydride between
the amino acid and POCl3. Aminolysis of this anhydride by
the thiazolamine species 3 led to the desired compound 4. Use
BBB permeation evaluation
BBB permeation studies were performed on the dihydropyridine
derivative 2 (reduced form), its oxidized form 1 and the unsub-
stituted aniline precursor 7. These experiments were carried on
anaesthetized healthy Sprague Dawley male rats, according to a
well known methodology.27 Solutions of each compound were
freshly prepared in order to avoid any compound degradation
reaction, mainly for 2 which is sensitive to oxidation and could
regenerate its oxidized form 1. This phenomenon could be a real
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 6 1 2 – 6 1 8
6 1 3

View Article Online
Scheme 1 Conditions: (i) EtOH, reflux, 5 min, quantitative; (ii) BocAla, POCl3, pyridine, −15 ◦C to 80 ◦C, 16 h, 77%; (iii) TFA, CH2Cl2, RT, 16 h,
quantitative; (iv) 3,5-difluorophenylacetic acid, BOP, DIEA, CH2Cl2, RT, 16 h, 53%; (v) Fe, NH4Cl, H2O–EtOH (v/v 6 : 10), reflux, 16 h, 90%;
(vi) nicotinoyl chloride hydrochloride, pyridine, reflux, 18 h, 48%; (vii) CH3I, DMF, 16 h, 81%; (viii) Na2S2O4, H2O–MeOH (v/v 1 : 1), RT, 4 h, 31%.
Table 1 Cell based c-secretase inhibitory activities
Table 2 Concentration of the tested derivatives in the brain at suitable
time intervals and comparison with the corresponding concentration in
plasma
Compound
Cell based assaya ED50/lM
log Pb
2
1
7
1
2
7
1.0 0.5
0.2 0.1
5.0 (30%)c
2.93
3.96
3.46
Time (hrs) Braina,c Ratiob Braina,d Ratiob Braina Ratiob
a Potency determination for the ability of compounds to reduce total Ab
production from HEK293 cells. b Log P determinations were performed
using ACD (Advanced Chemistry Development, Inc.)/log P 1.0 base
calculations. c 30% of inhibitory activity of 7 at 5 lM.
0.25
0.50
1.00
2.00
4.00
n. q.
240
320
345
255
0
n. d.
0
0
n. d.
85
0
5.33 n. d.
12.31 80
38.33 60
1.60
8.21
24.00
n. d.
4.44 230
12.00 240
n. d.
n. d.
n. d.
125
a Concentration expressed in ng/g of brain.
problem as our aim is to study the permeation of compound 2
through the BBB according to Bodor’s concept. An unfortunate
oxidation of the dihydropyridine moiety before permeation
will lead to a lower diffusion and then will interfere with the
demonstration of the efficiency of the concept.
These freshly prepared samples were injected into rats through
the jugular vein at a dose of 20 mg Kg−1. At various time
intervals, a blood sample was withdrawn from the eyeball and
after treatment was analysed by HPLC. At the end of this
blood sampling, the animal was decapitated and the brain was
submitted to treatment before HPLC analysis. HPLC profiles of
each type of sample were compared to determine the relative
permeation of analogue 2 (reduced form) compared to its
oxidized form 1 and the aniline 7.
b
concentration of drug in brain (ng/g of brain)
concentration of drug in blood (ng/g of blood)
Ratio =
× 103.
c Total concentration of 2 and its metabolites. d Total concentration of 1
and its metabolites. n. d.: not detected. n. q.: detected but not accurately
quantified.
1 and the aniline 7 which comes from enzymatic hydrolysis
of 2. Similarly, the amount of compound 1 in the brain was
estimated as the total amount of 1 which penetrated the brain in
its intact form and its metabolite 7. These results are summarized
in Table 2 and Fig. 2a.
As 2 is susceptible to oxidation, the total amount of drug 2
in the brain was determined as the amount of dihydropyridine 2
which has penetrated the brain in its intact form, its metabolites
Rate of BBB permeation is a critical parameter. An optimal
concentration of compound 7 in the brain was reached within
2 hours (240 ng/g of brain; Table 2). In contrast, for drug
6 1 4
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 6 1 2 – 6 1 8

View Article Online
Discussion
We have synthesized a new c-secretase inhibitor, the structure of
which incorporates a thiazolamide moiety bearing a dihydron-
icotinoyl as well as a difluorophenyl-alanyl substituent. This
inhibitor was identified as lowering Ab production in intact
cells (HEK 293) as well as in a cell free system. This result indi-
cates that using the phenyl ring to introduce a substituent such
as nicotinoyl moiety improved the inhibitory activity. We also
demonstrate through in vivo experiments that the RCDS based
on Bodor’s concept inhibitor appears to be efficient, since the
obtained results are consistent with those expected. Indeed, these
studies revealed that the conjugation of the new thiazolamide to
a BBB redox chemical drug delivery system allows the delivery
of the inhibitor to the brain with improvement of the c-secretase
inhibitory activity.
As expected, the results confirmed that reduced compound 2 is
metabolized into both its oxidized form 1 and its corresponding
free anilino derivative 7.
The remarkable pharmacokinetic parameter for compound
2 is its permeation rate in the brain. Indeed, from Table 2,
it can be seen that after 30 minutes the concentration in the
brain of compound 2 was around 240 ng/g of brain, which
is 3 times higher than for its analogue 7 (85 ng/g of brain,
after 30 minutes). Besides, the RCDS shows a more cumulative
effect of compound 2 in the brain with time with maximum
concentration after about 2 hours (345 ng/g of brain) as
compared with compound 7 (240 ng/g of brain) with minimal
systemic concentration of the system. This pharmacokinetic
property is of interest since the use of this BBB-drug delivery
system optimizes the therapeutic index of compound 2 by
simultaneously increasing central nervous system drug uptake
and decreasing drug uptake in peripheral tissues. As indicated
in the introduction of the paper, the observed improved BBB
permeation property of this new c-secretase inhibitor 2 could
reduce its diffusion into the peripheral tissues and consequently
it could reduce its side-effects on the immune cells or on neural
cell differentiation.17
In conclusion, we report here the synthesis of a new thiazo-
lamide c-secretase inhibitor, the design includes a dihydronicoti-
noyl moiety as Redox Delivery System, which allows efficient
BBB permeation. Its favourable pharmacokinetic profile could
be of interest in vivo. In vivo experiments in the Tg2576 mouse
model of Alzheimer’s disease are currently under way to validate
the pharmacological potential of such compounds and to
determine which of the drug or its metabolites is the most active
c-secretase inhibitor.
Fig. 2 Biological availability of the tested compounds in brain (a), in
blood (b) and ratio between the concentrations in brain and blood (c).
Experimental
2 and its metabolites, this optimal concentration is observed
more rapidly, only after 30 minutes (240 ng/g of brain), while
during this interval of time compound 1 and its metabolites
were not detected. Indeed, it can be observed that about 1 hour
after injection, compound 1 was almost undetectable in the
brain except as its precursor 7 (80 ng/g of brain; Table 2).
Compound 1 was unable to penetrate the BBB but it was
hydrolysed into 7 which was found in a very low quantity in
the CNS compartment.
In addition to the determination of drug levels in the brain,
the concentration in the blood was also studied. As shown in
Fig. 2b, the concentration level of each tested compound in the
blood decreased gradually with time to completely disappear
from the plasma after less than 4 hours.
Analysis of the ratio between the drug level in the brain and the
drug level in the blood reinforces our findings about the better
BBB permeation of 2 compared to 1 and 7. Indeed, as shown in
Table 2 and depicted in Fig. 2c, one can notice that this ratio is
three times higher for the dihydropyridine 2 compared to that of
the pyridinium derivative 1. This ratio is even twice as high for
aniline 7 than for 1.
Chemistry
Unless otherwise noted, starting materials and reagents were
obtained from commercial suppliers and were used without
purification. All the protected amino acids and peptide coupling
reagents were purchased from Bachem, Acros, Aldrich and
Neosystem. Tetrahydrofuran (THF) was distilled over sodium
benzophenone ketyl immediately prior to use. Methylene dichlo-
ride (CH2Cl2) was distilled over P2O5 just prior to use. Pyridine
and dimethylformamide (DMF) were of anhydrous quality from
commercial suppliers (Aldrich, Acros, Carlo Erba Reagents).
Nuclear magnetic resonance spectra were recorded at 250 MHz
for 1H and 62.9 MHz for 13C on a Bru¨ker AC-250 spectrometer.
Chemical shifts are expressed as d units (part per million)
downfield from TMS (tetramethylsilane). Electro-spray mass
spectral analysis and LC-MS analysis were obtained on a Waters
Micromass ZMD spectrometer for the ES-MS analysis by direct
injection of the sample solubilized in acetonitrile. The LC-MS
analysis was carried out by using a Waters model 2690 pump
and a Waters C18 Symmetry column with a two-mobile phase
system (0.1% formic acid in water and 0.1% formic acid in
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 6 1 2 – 6 1 8
6 1 5

View Article Online
acetonitrile). Microanalyses were carried out by Service Central
d’Analyses du CNRS (Venaison, France) and were within 0.4%
of the theoretical values. Analytical thin layer chromatographies
(TLC) and preparative thin layers chromatographies (PLC) were
performed using silica gel plates 0.2 mm thick and 1 mm
thick respectively (60F254 Merck). Preparative flash column
chromatographies were carried out on silica gel (230–240 mesh,
G60 Merck). The melting points were not determined because
of the amorphous character of our peptides and derivatives.
reaction mixture was stirred for 1 h at room temperature then
once again cooled to 0 ◦C. A CH2Cl2 solution (10 mL) of
0.9 equiv. (500 mg, 1.23 mmol) of the previous trifluoroacetic
acid salt 5 and 3.0 equiv. (720 lL, 4.05 mmol) of DIEA was
added dropwise. The solution was allowed to warm and stirred
overnight at room temperature. The solvent was removed under
reduced pressure and the residue was dissolved in EtOAc
(100 mL). The organic layer was washed successively by using
H2O (50 mL), 5% aqueous citric acid (2 × 50 mL), brine
(50 mL), 5% aqueous NaHCO3 (2 × 50 mL) and brine (50 mL),
was dried over anhydrous MgSO4, filtered and concentrated
under reduced pressure. The crude residue was purified by
flash chromatography (EtOAc–hexane 1 : 1) to give the title
compound 6 (290 mg, yield: 53%) as a yellow solid. (Found:
C, 53.68; H, 3.52; N, 12.82. C20H16F2N4O4S requires C, 53.81;
2-Amino-4-(4-nitrophenyl)thiazole, hydrobromide salt, 319. 1-
Bromo-4ꢀ-nitroacetophenone (1.50 g, 1.0 equiv., 6.15 mmol)
was dissolved in refluxed ethanol (25 mL). Thiourea (0.47 g,
1.0 equiv., 6.15 mmol) was then added in one lot. The refluxed
reaction mixture was stirred and, after a few minutes, a yellow
solid appeared. The solid was filtered, washed with diethyl ether
and dried under reduced pressure to give the title compound
3 as a yellow solid (1.88 g, quantitative). (Found: C, 35.89; H,
2.83; N, 13.60. C9H7N3O2S.HBr requires C, 35.78; H, 2.67; N,
1
H, 3.61; N, 12.55%). Rf 0.36 (EtOAc–hexane 1 : 1). H NMR
3
(CDCl3, 250 MHz) d 1.48 (d, 3 H, CH3 b Ala, J = 7.0 Hz),
3.64 (s, 2 H, -CH2), 4.76 (m, 1 H, CH a Ala), 5.96 (d, 1 H, NH
3
=
Ala, J = 8.8 Hz), 6.82 (m, 3 H, Ar H), 7.36 (s, 1 H, CH–S),
13.91%). Rf 0.31 (CH2Cl2–MeOH 9 : 1). 1H NMR (DMSO-d6,
3
3
7.99 (d, 2 H, Ho−NO2, J = 9.0 Hz), 8.27 (d, 2 H, Hm−NO2, J =
9.0 Hz), 10.02 (s, 1 H, NH thiazole). 13C NMR (DMSO-d6,
62.9 MHz) d 17.6, 40.9, 48.5, 101.8 (t, 1 C, 2J = 25.3 Hz), 112.2
(d, 2 C, 2J = 24.2 Hz), 112.5, 124.1 (2 C), 126.4 (2 C), 140.4 (t,
3
=
250 MHz) d 7.47 (s, 1 H, CH–S), 8.04 (d, 2 H, Hm−NO2, J =
9.0 Hz), 8.15 (s, 2 H, NH2), 8.24 (d, 2 H, Ho−NO2, 3J = 9.0 Hz).
13C NMR (DMSO-d6, 62.9 MHz) d 107.3, 124.1 (2 C), 126.7
(2 C), 136.0, 139.6, 164.9, 169.9. MS-ES m/z 222 (M + H)+.
3
1
1 C, J = 10.0 Hz), 146.4, 146.7, 158.2, 162.1 (dd, 2 C, J =
3
2-[N-(N-tert-Butoxycarbonyl)-(S)-alanyl]amino-4-(4-nitro-
phenyl)-thiazole, 421. N-Boc-Ala (0.88 g, 1.0 equiv., 4.64 mmol)
and the previous hydrobromide salt 3 (1.40 g, 1.0 equiv.,
4.64 mmol) were dissolved in anhydrous pyridine (70 mL). The
solution was cooled to −15 ◦C and phosphorus oxychloride
(0.47 mL, 1.1 equiv., 5.10 mmol) was added dropwise under
vigorous stirring. The reaction mixture was stirred at −15 ◦C
for 30 min. The solution was allowed to ◦warm to room
temperature and then stirred overnight at 80 C. The reaction
was quenched by addition of crushed ice/water (50 mL). The
desired compound was extracted into EtOAc (3 × 50 mL). The
combined organic layers were dried over anhydrous MgSO4,
filtered and concentrated under reduced pressure. The crude
material was purified by flash chromatography to lead to the
title compound 4 as a yellow solid (1.40 g, yield: 77%). (Found:
C, 51.93; H, 5.32; N, 14.48. C17H20N4O5S requires C, 52.03;
245.4 Hz, J = 13.4 Hz), 169.1, 171.8, 181.2. MS-ES m/z 447
(M + H)+.
N-[N-(3,5-Difluorophenylacetyl)-(S)-alanyl]-N-[4-(4-amino-
phenyl)]-thiazol-2-ylamine, 7. The previous nitro derivative 6
(280 mg, 1.0 equiv., 0.62 mmol) was dissolved in a H2O–EtOH
solution (6 : 10 v/v; 15 mL). Iron (173 mg, 5.0 equiv., 3.10 mmol)
and 2.0 equiv. of NH4Cl (66 mg, 2.0 equiv., 1.24 mmol) were
then added and the reaction mixture was refluxed overnight.
The solution was cooled to room temperature, filtered and
concentrated under reduced pressure. The crude material was
purified by flash chromatography (EtOAc–hexane 1 : 1) to
provide the desired compound 7 as a yellow solid (230 mg, yield:
90%). (Found: C, 57.82; H, 4.61; N, 13.25. C20H18F2N4O2S
requires C, 57.68; H, 4.36; N, 13.45%). Rf 0.10 (EtOAc–hexane
1 : 1). 1H NMR (CDCl3, 250 MHz) d 1.45 (d, 3 H, CH3 b Ala,
3J = 7.0 Hz), 3.55 (s, 2 H, -CH2), 4.80 (m, 1 H, CH a Ala),
1
H, 5.14; N, 14.28%). Rf 0.70 (EtOAc–hexane 1 : 1). H NMR
3
6.40 (d, 1 H, NH Ala, J = 8.4 Hz), 6.70 (d, 2 H, Ho−NH2
,
3
(CD3OD, 250 MHz) d 1.44 (d, 3 H, CH3 b Ala, J = 7.3 Hz),
3
=
J = 8.7 Hz), 6.80 (m, 3 H, Ar H), 7.95 (s, 1 H, CH–S), 7.62
1.49 (s, 9 H, CH3 Boc), 4.33 (m, 1 H, CH a Ala), 5.95–6.03 (m,
3
(d, 2 H, Hm−NH2, J = 8.9 Hz), 10.04 (s, 1 H, NH thiazole),
1 H, NH Ala), 7.74 (s, 1 H, = CH–S), 8.16 (d, 2 H, Hm−NO2
,
10.96–11.07 (m, 2 H, -NH2). 13C NMR (CD3OD, 62.9 MHz) d
17.8, 42.6, 50.7, 103.1 (t, 1 C, 2J = 25.5 Hz), 105.5, 113.3 (d, 2
3J = 9.3 Hz), 8.29 (d, 2 H, Ho−NO2, 3J = 9.3 Hz), 10.82 (s, 1 H,
NH thiazole). 13C NMR (CDCl3, 62.9 MHz) d 17.4, 28.3 (3 C),
50.8, 81.5, 111.4, 124.1 (2 C), 126.6 (2 C), 140.2, 147.1, 147.7,
157.7, 158.3, 171.0. MS-ES m/z 393 (M + H)+.
2
C, J = 24.8 Hz), 116.3 (2 C), 126.2, 128.1 (2 C), 140.9 (t, 1 C,
3J = 9.7 Hz), 148.9, 151.9, 158.9, 164.4 (dd, 2 C, 1J = 246.8 Hz,
3J = 13.1 Hz), 172.7, 172.8. MS-ES m/z 417 (M + H)+.
2-(N-(S)-Alanyl)amino-4-(4-nitrophenyl)-thiazole, trifluoro-
acetic acid salt, 5. The previous N-Boc protected derivative 4
(1.0 equiv., 1.82 g, 4.64 mmol) was dissolved in CH2Cl2 (32 mL).
Trifluoroacetic acid (10.0 equiv., 3.53 mL, 46.40 mmol) was
then added and the reaction mixture was stirred overnight at
room temperature. The solvent and excess TFA were removed
under reduced pressure and the resulting solid was triturated
in Et2O. The desired trifluoroacetic acid salt 5 was isolated
quantitatively after filtration as a yellow solid (1.88 g). 1H NMR
(CDCl3, 250 MHz) d1.67 (d, 3 H, CH3 b Ala, 3J = 7.2 Hz), 4.24
N-[N-(3,5-Difluorophenylacetyl)-(S)-alanyl]-N-{4-[4-(N-
pyridyl-3-carbonyl)-aminophenyl]}-thiazol-2-ylamine, 8. The
anilino derivative 7 (100 mg, 1.0 equiv., 0.24 mmol) and the
hydrochloride salt of nicotinoylchloride (42 mg, 1.0 equiv.,
0.24 mmol) were dissolved in anhydrous pyridine (3 mL). The
reaction mixture was refluxed for 18 h. The solvent was removed
under reduced pressure. The solid residue was triturated, filtered
and dried by using successively EtOAc (20 mL), H2O (10 mL)
and Et2O (20 mL). The resulting solid was dried under reduced
pressure. The desired compound 8 was obtained as a white
solid (50 mg, yield: 48%). (Found: C, 59.98; H, 4.34; N, 13.18.
C26H21F2N5O3S requires C, 59.88; H, 4.06; N, 13.43%). Rf 0.57
(EtOAc–hexane 1 : 2). 1H NMR (DMSO-d6, 250 MHz) d 1.58 (d,
3 H, CH3 b Ala, 3J = 7.1 Hz), 3.80 (s, 2 H, -CH2), 4.75 (m, 1 H,
(m, 1 H, CH a Ala), 6.04 (d, 1 H, NH Ala, 3J = 8.4 Hz), 7.80 (s,
3
=
1 H, CH–S), 8.17 (d, 2 H, Hm NO2, J = 9.2 Hz), 8.30 (d, 2 H,
–
Ho NO2, J = 9.2 Hz) 10.92 (s, 1 H, NH thiazole). 13C NMR
3
–
(CDCl3, 62.9 MHz) d 17.3, 66.9, 113.3, 125.0 (2 C), 127.7 (2 C),
141.6, 148.5, 159.1, 169.5, 183.1. MS-ES m/z 293 (M + H)+.
=
CH a Ala), 7.30 (m, 3 H, diF Ar H), 7.81 (t, 2 H, CH–S +
N-[N-(3,5-Difluorophenylacetyl)-(S)-alanyl]-N-[4-(4-nitro-
phenyl)]-thiazol-2-ylamine, 6. 3,5-Difluorophenylacetic acid
(232 mg, 1.0 equiv., 1.35 mmol) was dissolved in freshly distilled
CH2Cl2 (30 mL) with 1.0 equiv. (652 mg, 1.35 mmol) of BOP
reagent. The reaction mixture was cooled to 0 ◦C then 1.0 equiv.
of DIEA (240 lL, 1.35 mmol) was added dropwise. The
H
5 − pyridine , 3J = 7.4 Hz), 8.11 (d, 4 H, Ar H, 3J = 6.4 Hz), 8.53
3
4
(dt, 1 H, H6 − pyridine , J = 8.0 Hz, J = 1.9 Hz), 8.85 (d, 1 H,
3
3
-NH Ala, J = 6.6 Hz), 9.00 (dd, 1 H, H4 − pyridine , J = 4.9 Hz,
4J = 1.6 Hz), 9.34 (d, 1 H, H2 − pyridine , J = 2.9 Hz), 10.78 (s,
3
1 H, NH thiazole), 12.64 (s, 1 H, NH). 13C NMR (DMSO-d6,
62.9 MHz) d 17.5, 41.0, 48.5, 101.8 (t, 1 C, 2J = 25.5 Hz), 107.2,
6 1 6
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 6 1 2 – 6 1 8

View Article Online
2
112.2 (d, 2 C, J = 24.4 Hz), 120.3 (2 C), 123.4 (2 C), 126.0
(0.5 mg ml−1) in the presence of CHAPSO. The reactions were
stopped by adding RIPA.
3
(2 C), 130.0, 130.5, 137.0, 140.5 (t, 1 C, J = 9.8 Hz), 148.6,
148.7, 152.1, 157.7, 162.0 (dd, 2 C, 1J = 245.6 Hz, 3J = 13.6 Hz),
169.2, 171.6, 181.1. MS-ES m/z 522 (M + H)+.
In vitro cell based assay. The ability to inhibit Ab peptide
production in a cell based assay was measured according to a
well known procedure using APP transfected HEK 293 cells.26
HEK 293 E cells were maintained in DMEM supplemented
with 10% fetal bovine serum, 25 lg mL−1 penicillin, 25 lg mL−1
streptomycin, and 250 lg mL−1 G 418. Cells were transfected
following specifications provided by the manufacturer (Gibco-
BRL). One day after transfection, the conditioned medium was
replaced with selection medium (media as above containing
in addition 250 lg mL−1 hygromycin B). For immunoblotting
analysis, the 24 h serum-free conditioned medium was harvested
and cellular debris was removed by centrifugation for 10 min.
The monolayers were washed with PBS, and cell lysates were
prepared by the addition of SDS–sample buffer containing
50 mM dithiothreitol. For Ab determination by ELISA, con-
fluent cultures were incubated in serum-free medium containing
0.2% BSA for 16 h. To test the effects of compounds on Ab
formation, cells were plated at confluency (1 × 106 cells cm−2) in
96-well tissue culture dishes in serum-free medium containing
0.2% BSA in the presence of the indicated concentration of
either c-secretase reference inhibitor DAPT. After 16 h, the
conditioned medium was harvested and analyzed using the HA
11 Ab-specific ELISA.
N-[N-(3,5-Difluorophenylacetyl)-(S)-alanyl]-N-(4-{4-[N-(N-
methyl)pyridinium-yl-3-carbonyl]-aminophenyl})-thiazol-2-yl-
amine, iodide salt, 1. The previous pyridino derivative 8
(1.00 g, 1.0 equiv., 1.92 mmol) was dissolved in anhydrous
DMF (50 mL). Iodomethane (2.3 mL, 20.0 equiv., 38.40 mmol)
was then added. The reaction mixture was stirred overnight at
room temperature. The solvent and excess iodomethane were
removed under reduced pressure. The solid residue was dis-
solved in MeOH (10 mL) and the pyridinium salt was
precipitated by addition of Et2O. The resulting yellow solid was
filtered to afford the desired compound 1 (1.03 g, yield: 81%).
(Found: C, 48.62; H, 3.83; N, 10.35. C27H24F2N5O3S.I requires
C, 48.88; H, 3.65; N, 10.56%). 1H NMR (DMSO-d6, 250 MHz)
3
d 1.35 (d, 3 H, CH3 b Ala, J = 7.2 Hz), 3.56 (s, 2 H, -CH2),
4.41 (m, 4 H, CH a Ala + N–CH3), 7.06 (m, 3 H, diF Ar H),
=
7.61 (s, 1 H, CH–S), 7.84–7.91 (m, 4 H, Ar H), 8.31 (dt, 1 H,
3
4
H
5 − pyridine , J = 6.1 Hz, J = 1.9 Hz), 8.62 (d, 1 H, -NH Ala,
3J = 6.5 Hz), 9.04 (d, 1 H, H6 − pyridine , J = 8.0 Hz), 9.15 (d,
3
3
1 H, H4 − pyridine , J = 6.0 Hz), 9.52 (s, 1 H, H2 − pyridine ), 10.91 (s,
1 H, NH thiazole), 12.39 (s, 1 H, NH). 13C NMR (DMSO-d6,
62.9 MHz) d 17.5, 40.9, 48.2, 48.5, 101.8 (t, 1 C, 2J = 25.7 Hz),
2
107.7, 112.4 (d, 2 C, J = 24.4 Hz), 120.4 (2 C), 126.2 (2 C),
3
BBB permeation evaluation27. Five groups, each of four
Sprague Dawley male rats of average weight of 55–65 g,
were anaesthetized with urethane. A freshly prepared solution
(50 mmol) of each compound in DMSO (diluted to 20% with
water) was injected through the jugular vein at a dose of
20 mg kg−1.
127.3 (2 C), 130.7, 140.4 (t, 2 C, J = 10.0 Hz), 137.6, 143.3,
145.7, 147.3, 160.3, 162.1 (dd, 2 C, 1J = 231.8 Hz, 3J = 13.3 Hz),
169.1, 171.6, 181.2. MS-ES m/z 268.5 (M + H)+.
N-[N-(3,5-Difluorophenylacetyl)-(S)-alanyl]-N-(4-{4-[N-(N-
methyl)dihydropyridinyl-3-carbonyl]-aminophenyl})-thiazol-2-yl-
amine, 2. The iodide salt 1 (200 mg, 1.0 equiv., 0.30 mmol)
was dissolved in a H2O–MeOH solution (1 : 1 v/v; 20 mL).
The resulting solution was degassed under a nitrogen flush at
Blood sample analysis. At time intervals (0.25, 0.50, 1.00,
2.00 and 4.00 hours) blood samples were withdrawn from the
eyeball, immediately added to previously weighed centrifuge
tubes containing acetonitrile (4 mL) and weighed to determine
the amount of blood added. The blood samples were centrifuged
at about 4000 rpm for 10 minutes and the supernatant analyzed
by HPLC.
0
◦C for 20 min. Na2CO3 (191 mg, 6.0 equiv., 1.8 mmol) and
sodium dithionite (Na2S2O4, 208 mg, 4.0 equiv., 1.20 mmol)
were then added. The reaction mixture was stirred at room
temperature for 4 h. The aqueous layer was extracted with
CH2Cl2 (3 × 5 mL). The combined organic layers were dried
over anhydrous MgSO4, filtered and concentrated under
reduced pressure. The desired compound 2 was obtained as a
yellow solid (50 mg, yield: 31%). (Found: C, 60.61; H, 4.97; N,
13.31. C27H25F2N5O3S requires C, 60.32; H, 4.69; N, 13.03%).
Brain sample analysis. The animals were decapitated and the
brains were taken, weighed and immediately homogenized with
1 mL saline and diluted with 4 mL of 5% DMSO in acetonitrile,
homogenized again and centrifuged at 4000 rpm for 10 minutes.
The supernatants were analyzed by HPLC.
1H NMR (CDCl3, 250 MHz) d 1.41 (d, 3 H, CH3 b Ala, J =
3
6.8 Hz), 2.92 (s, 3 H, N–CH3), 3.26 (s, 2 H, H4 − pyridine ), 3.51 (s,
2 H, -CH2), 4.57–4.87 (m, 2 H, CH a Ala + H5−pyridine), 5.70
HPLC analysis method. HPLC method with UV detection.
Equipment: Knauer HPLC with Marathon Plus Autosam-
pler.
Column: HPLC column C18, 4.6 mm × 25 cm, 5 lm.
Detector: UV detector at wavelength about 280 nm.
Injection volume: 50 lL.
3
4
(dd, 1 H, H6 − pyridine , J = 8.1 Hz, J = 1.3 Hz), 5.97 (d, 1 H,
NH Ala, 3J = 8.6 Hz), 6.67–6.78 (m, 3 H, diF Ar H), 7.01 (m,
=
1 H, H2 − pyridine ), 7.12 (s, 1 H, CH–S), 7.55–7.64 (m, 4 H, Ar
H), 11.03 (s, 1 H, NH thiazole), 12.27 (s, 1 H, NH). 13C NMR
(CDCl3, 62.9 MHz) d 17.2, 28.3, 38.6, 39.4, 50.6, 52.1, 102.7 (t,
2
2
Temperature: room temperature.
1 C, J = 24.5 Hz), 103.4, 110.2, 112.4 (d, 2 C, J = 26.3 Hz),
120.9 (2 C), 126.3 (2 C), 127.2 (2 C), 132.1, 138.2, 139.1, 150.8,
163.8 (d, 2 C, 1J = 237.0 Hz, 3J = 13.5 Hz), 164.2, 170.9, 172.7,
182.6. MS-ES m/z 538 (M + H)+.
Flow rate: 1 ml min−1.
Software: Eurochrom 2000.
Mobile phase: Filtered and degassed mixture of 0.2% hexane–
sulfonic acid (prepared by pH adjustment of 0.2% solution of
hexane–sulfonic acid sodium salt with 20% trifluoroacetic acid
to pH about 2.7) : acetonitrile (30 : 70).
Biology
Retention time: 1a (about 6.3 minutes), 6 (about 11.2 minute)
and 1b (about 15 minutes).
In vitro c-secretase cell free assay. The ability to block
specifically in vitro c-secretase was performed by using solu-
bilized c-secretase prepared as described by Li et al.24 A DNA
fragment encoding amino acids 596–695 of the 695-aa isoform
of APP (APP695) and the Flag sequence (DYKDDDDK)
at the C terminus was generated by PCR amplification with
suitably designed oligonucleotides and the APP695 cDNA. The
Met that serves as the translation start site is residue 596 of
APP695 (the P1 residue with respect to the b-secretase cleavage
site). C100Flag (1.7 lM) was incubated with cell membranes
Acknowledgements
INSERM (Institut National de la Sante´ et de la Recher-
che Me´dicale) is greatly acknowledged for financial support.
We are grateful to Professor Keith Dudley (Universite´ de
la Me´diterrane´e, INSERM U-623, IBDM, France) for the
preparation of the manuscript.
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 6 1 2 – 6 1 8
6 1 7

View Article Online
15 I. Churcher, K. Ashton, J. W. Butcher, E. E. Clarke, T. Harrison,
H. D. Lewis, A. P. Owens, M. R. Teall, S. Williams and J. D. J.
Wrigley, Bioorg. Med. Chem. Lett., 2003, 13, 179–183; B. Schmidt,
ChemBioChem, 2003, 4, 366–378; G. T. Wong, D. Manfra, F. M.
Poulet, Q. Zhang, H. Josien, T. Bara, L. Engstrom, M. Pinzon-Ortiz,
J. S. Fine, H. J. J. Lee, L. L. Zhang, G. A. Higgins and E. M. Parker,
J. Biol. Chem., 2004, 279, 12876–12882; T. A. Lanz, J. D. Hosley,
W. J. Adams and K. M. Merchant, J. Pharmacol. Exp. Ther., 2004,
309, 49–55.
16 W. T. Kimberly, W. P. Esler, W. J. Ye, B. L. Ostaszewski, J. Gao, T.
Diehl, D. J. Selkoe and M. S. Wolfe, Biochemistry, 2003, 42, 137–144.
17 A. Geling, H. Steiner, M. Willem, L. Bally-Cuif and C. Haass, EMBO
Rep., 2002, 3, 688–694.
18 G. T. Wong, D. Manfra, F. M. Poulet, Q. Zhang, H. Josien, T. Bara, L.
Engstrom, M. Pinzon-Ortiz, J. S. Fine, H. J. J. Lee, L. L. Zhang, G. A.
Higgins and E. M. Parker, J. Biol. Chem., 2004, 279, 12876–12882.
19 A. R. Hantzsch and J. H. Weber, Chem. Ber., 1887, 20, 3118–3132;
L. C. King and R. J. Hlavacek, J. Am. Chem. Soc., 1950, 72, 3722–
3725; B. S. Furniss, A. J. Hannaford, P. W. G. Smith and A. R.
Tatchell, in Vogel’s Textbook of Practical Organic Chemistry, 5th
edn., Longman Scientific & Technical, Harlow, UK, 1989, p. 1153;
N. Bailey, A. W. Dean, D. B. Judd, D. Middlemiss, R. Storer and S. P.
Watson, Bioorg. Med. Chem. Lett., 1996, 6, 1409–1414 and references
therein.
References
1 C. Hansch, J. P. Bjorkroth and A. Leo, J. Pharm. Sci., 1987, 76,
663–687.
2 B. Testa, P. A. Carrupt, P. Gaillard, F. Billois and P. Weber, Pharm.
Res., 1996, 13, 335–343.
3 I. Tamai and A. Tsuji, J. Pharm. Sci., 2000, 89, 1371–1388.
4 J. Huwyler, J. Yang and W. M. Pardridge, J. Pharmacol. Exp. Ther.,
1997, 282, 1541–1546.
5 J. Kreuter, Adv. Drug Delivery Rev., 2001, 47, 65–81; J. Kreuter, Curr.
Med. Chem.: Central Nervous System Agents, 2002, 2, 241–249.
6 H. Brem and P. Gabikian, J. Controlled Release, 2001, 74, 63–67.
7 C. Rohlff and C. Southan, Curr. Opin. Mol. Ther., 2002, 4, 251–
258.
8 P. L. Golden, T. J. Maccagnan and W. M. Pardridge, J. Clin. Invest.,
1997, 99, 14–18.
9 A. Misra, S. Ganesh, A. Shahiwala and S. P. Shah, J. Pharm.
Pharmaceut. Sci., 2003, 6, 252–273.
10 N. Bodor, H. H. Farag and M. E. Brewster, Science, 1981, 214, 1370–
1372; N. Bodor, L. Prokai, W. M. Wu, H. Farag, S. Jonalagadda, M.
Kawamura and J. Simpkins, Science, 1992, 257, 1698–1700.
11 J. Lewis, D. W. Dickson, W. L. Lin, L. Chisholm, A. Corral, G. Jones,
S. H. Yen, N. Sahara, L. Skipper, D. Yager, C. Eckman, J. Hardy, M.
Hutton and E. McGowan, Science, 2001, 293, 1487–1491; J. Gotz,
F. Chen, J. van Dorpe and R. M. Nitsch, Science, 2001, 293, 1491–
1495.
20 D. T. S. Rijkers, H. P. H. M. Adams, H. C. Hemker and G. I. Tesser,
Tetrahedron, 1995, 51, 11235–11250.
21 G. Que´le´ver, S. Burlet, C. Garino, N. Pietrancosta, Y. Laras and J. L.
Kraus, J. Comb. Chem., 2004, 6, 695–698.
12 D. J. Selkoe, Nature, 1999, 399, A23–A31.
13 D. L. Dominguez and B. De Strooper, Trends Pharmacol. Sci., 2002,
23, 324–330.
22 C. A. Merlic, S. Motamed and B. Quinn, J. Org. Chem., 1995, 60,
3365–3369.
14 H. F. Dovey, V. John, J. P. Anderson, L. Z. Chen, P. D. Andrieu, L. Y.
Fang, S. B. Freedman, B. Folmer, E. Goldbach, E. J. Holsztynska,
K. L. Hu, K. L. Johnson-Wood, S. L. Kennedy, D. Kholedenko, J. E.
Knops, L. H. Latimer, M. Lee, Z. Liao, I. M. Lieberburg, R. N.
Motter, L. C. Mutter, J. Nietz, K. P. Quinn, K. L. Sacchi, P. A.
Seubert, G. M. Shopp, E. D. Thorsett, J. S. Tung, J. Wu, S. Yang,
C. T. Yin, D. B. Schenk, P. C. May, L. D. Altstiel, M. H. Bender, L. N.
Boggs, T. C. Britton, J. C. Clemens, D. L. Czilli, D. K. Dieckman-
McGinty, J. J. Droste, K. S. Fuson, B. D. Gitter, P. A. Hyslop, E. M.
Johnstone, W. Y. Li, S. P. Little, T. E. Mabry, F. D. Miller, B. Ni, J. S.
Nissen, W. J. Porter, B. D. Potts, J. K. Reel, D. Stephenson, Y. Su,
L. A. Shipley, C. A. Whitesitt, T. Yin and J. E. Audia, J. Neurochem.,
2001, 76, 173–181; T. A. Lanz, C. S. Himes, G. Pallante, L. Adams, S.
Yamazaki, B. Amore and K. M. Merchant, J. Pharmacol. Exp. Ther.,
2003, 305, 864–871.
23 N. Bodor and H. H. Farag, J. Med. Chem., 1983, 26, 313–318.
24 Y. M. Li, M. T. Lai, M. Xu, Q. Huang, J. DiMuzio-Mower, M. K.
Sardana, X. P. Shi, K. C. Yin, J. A. Shafer and S. J. Gardell, Proc.
Natl. Acad. Sci. USA, 2000, 97, 6138–6143; C. McLendon, T. P. Xin,
C. Ziani-Cherif, M. P. Murphy, K. A. Findlay, P. A. Lewis, I. Pinnix,
K. Sambamurti, R. Wang, A. Fauq and T. E. Golde, FASEB J., 2000,
14, 2383–2386.
25 W. P. Esler, W. T. Kimberly, B. L. Ostaszewski, W. J. Ye, T. S. Diehl,
D. J. Selkoe and M. S. Wolfe, Proc. Natl. Acad. Sci. USA, 2002, 99,
2720–2725.
26 D. Seiffert, T. Mitchell, A. M. Stern, A. Roach, Y. Zhan and R.
Grzanna, Mol. Brain Res., 2000, 84, 115–126.
27 M. Sheha, A. Al-Tayeb, H. El-Sherief and H. Farag, Bioorg. Med.
Chem., 2003, 11, 1865–1872.
6 1 8
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 6 1 2 – 6 1 8