Structure-based design of cycloamide-urethane-derived novel inhibitors of human brain memapsin 2 (β-secretase)

Article abstract of DOI:10.1016/j.bmcl.2004.10.084

A series of novel macrocyclic amide-urethanes was designed and synthesized based upon the X-ray crystal structure. Inhibitors containing 14- to 16-membered rings exhibited low nanomolar inhibitory potencies against memapsin 2. A series of novel macrocycli

Full text of DOI:10.1016/j.bmcl.2004.10.084

Bioorganic & Medicinal Chemistry Letters 15 (2005) 15–20
Structure-based design of cycloamide–urethane-derived
novel inhibitors of human brain memapsin 2 (b-secretase)
Arun K. Ghosh,a,* Thippeswamy Devasamudram,a,b Lin Hong,b,c
Christopher DeZutter,a Xiaoming Xu,a Vajira Weerasena,b Gerald Koelsch,b,c
Geoffrey Bilcera,b and Jordan Tangc,d
aDepartment of Chemistry, University of Illinois at Chicago, 845 West Taylor Street, Chicago, IL 60607, USA
bZapaq Inc., 755 Research Parkway, Oklahoma City, OK 73104, USA
cProtein Studies Program, Oklahoma Medical Research Foundation, University of Oklahoma Health Science Center,
Oklahoma City, OK 73104, USA
dDepartment of Biochemistry and Molecular Biology, University of Oklahoma Health Science Center, Oklahoma City,
OK 73104, USA
Received 17 September 2004; revised 27 October 2004; accepted 28 October 2004
Available online 21 November 2004
Abstract—A series of novel macrocyclic amide-urethanes was designed and synthesized based upon the X-ray crystal structure of
our lead inhibitor (1, OM99-2 with eight residues) bound to memapsin 2. Ring size and substituent effects have been investigated.
Cycloamide–urethanes containing 14- to 16-membered rings exhibited low nanomolar inhibitory potencies against human brain
memapsin 2 (b-secretase).
Ó 2004 Elsevier Ltd. All rights reserved.
˚
crystal structure of 1-bound memapsin 2 at 1.9A resolu-
tion9 (Fig. 1).
The proteolytic enzyme memapsin
2
(b-secretase,
BACE-1) is one of two proteases that cleave the b-
amyloid precursor protein (APP) to produce the 40–42
residue amyloid-b peptide (Ab) in the human brain, a
key event in the progression of AlzheimerÕs disease
(AD).1,2 Because of its involvement in an early stage
of the cascade of events leading to the pathogenesis of
AD, inhibition of memapsin 2 has become an excellent
target for AD.3–5 We recently designed a number of po-
tent memapsin 2 inhibitors incorporating a nonhydroly-
zable Leu-Ala hydroxyethylene dipeptide isostere.6
Potent memapsin 2 inhibitors incorporating other tran-
sition-state isosteres have also been reported recently.7
Our first generation memapsin inhibitors were designed
based upon memapsin 2 specificity studies.8 One such
inhibitor is OM99-2 (1), which has shown a Ki value
of 1.6nM for human memapsin 2.6a In order to obtain
specific ligand-binding site interactions in the active site
of memapsin 2, we subsequently determined the X-ray
This crystal structure provided important molecular in-
sight, which may aid the design of novel and selective
O
CO2H
H2N
CH3
O
O
OH
H
N
H
N
H
N
N
H
H2N
O
O
HN
O
H
N
Ph
HO2C
1 (OM-99-2)
O
CO2H
O
N
H
H
N
OH Me
H
N
HN
( )n
O
H
N
O
O
O
HN
O
O
Ph
2a (n = 1); 2b (n = 2);
2c (n = 3)
Keywords: AlzheimerÕs disease; Inhibitor; Memapsin 2; Secretase.
*
Corresponding author. Tel.: +1 312 996 9672; fax: +1 312 996
1547; e-mail: arunghos@uic.edu
Figure 1.
0960-894X/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.10.084

16
A. K. Ghosh et al. / Bioorg. Med. Chem. Lett. 15 (2005) 15–20
inhibitors of memapsin 2. We are particularly intrigued
by a number of key interactions involving P1–P4 side
chains in the S1–S4 subsites.9 It appears that the P1-
Leu and the P3-Val fill the hydrophobic pocket effec-
tively. Furthermore, the P2-Asn is involved in a hydro-
gen bond with Arg-235. Interestingly, the P4-Glu is
hydrogen bonded to P2-Asn and possibly assists the for-
mation of the P2-Asn hydrogen bond with Arg-235. Our
preliminary modeling studies indicate that a suitably
functionalized 15- or 16-membered cycloamide–
urethane could effectively hydrogen bond with Arg-235
as well as effectively fill in the hydrophobic pockets in
the S2–S4 binding sites. Herein, we report our prelimi-
nary investigation that led to the design and synthesis
of cycloamide–urethane-derived inhibitors of memapsin
2. Cyclic inhibitors containing a 16-membered ring
turned out to be more potent than the acyclic counter-
part. A 16-membered macrocycle containing a trans-ole-
fin has shown the best memapsin 2 inhibitory potency
with a Ki value of 14nM. Various cycloamide–urethanes
were conveniently prepared by efficient ring-closing
metathesis utilizing GrubbsÕ catalyst.10
by ring-closing olefin metathesis of acyclic terminal ole-
fins 6a–c followed by saponification of the resulting
esters. Olefin metathesis was expected to provide a
cis/trans mixture.11 Various terminal olefins 6a–c were
synthesized from commercially available12 N-(tert-but-
oxycarbonyl)-L-aspartic acid-4-benzylester 7. We also
planned to prepare the acyclic inhibitors by coupling
of the 6a–c-derived carboxylic acids and the correspond-
ing amine derived from Leu-Ala dipeptide 5. This ena-
bled us to examine inhibitory potencies of various
acyclic and cyclic inhibitors.
Synthesis of various cyclic carboxylic acid containing
Val-Asn dipeptides 4a–c is shown in Scheme 1. Treat-
ment of known 4-benzylester 7 with KHCO3 and methyl
iodide in DMF for 65h provided methyl ester 8 in near
quantitative yield. Hydrogenolysis of the benzyl ester
over 10% Pd–C, followed by coupling of the resulting
acid with allylamine in the presence of 1-[3-(dimethyl-
amino) propyl]-3-ethyl-carbodiimide hydrochloride
(EDCI), diisopropylethylamine (DIPEA), and 1-hydr-
oxybenzotriazole (HOBt) afforded allyl amide 9 in 74%
yield for the two-step sequence. Synthesis of various
urethanes 10a–c was carried out with valine methyl ester
by a one pot two-step procedure. Thus, a mixture of
As outlined in Figure 2, we planned to synthesize vari-
ous macrocyclic inhibitors from the corresponding cyclic
olefins 3a–c by hydrogenation of the olefins followed by
removal of the hydroxyl protecting group. Cyclic deriv-
atives 3a–c were obtained by coupling cyclic dipeptide
acids 4a–c and the amine derived from known dipeptide
isostere 5.6a Cyclic dipeptide acids 4a–c can be derived
O
KHCO3, MeI
DMF
BnO
BocHN
8
7
OMe
(99%)
O
2a-c
1. H2, Pd/C
O
(74%)
2. EDCI, HOBt
DIPEA
H2N
OMe
O
O
TBS
1. alcohol,
(COCl2)3
2. aq LiOH
N
O
Me
N
(64-68%)
H
N
H
N
H
H
OMe
BocHN
( )n
O
HN
H
N
O
O
O
9
( )n
O
O
O
HN
O
H
N
O
1. TFA,CH2Cl2
OH
Ph
(60-65%)
2. 10a-c, EDCI,
HOBt, DIPEA
O
3a-c
10a (n =1), 10b (n=2),
10c (n=3)
O
TBS
HN
O
O
Me
N
H
H
H
Boc-N
N
OH
( )n
O
O
( )n
O
HN
H
6a (n = 1)
6b (n = 2)
6c (n = 3)
H
N
N
OMe
+
O
O
N
HN
O
O
H
O
O
O
4a-c
Ph
5
Cl2(PCy3)(IMes)Ru=CHPh
CH2Cl2, 42°C
(63-73%)
O
O
4a (n = 1)
4b (n = 2)
4c (n = 3)
O
N
N
H
HN
Aq. LiOH
(99%)
H
OMe
OMe
BnO
BocHN
( )n
O
HN
( )n
O
H
N
H
N
OH
O
O
O
O
O
11a-c
O
O
6a-c
7
Figure 2.
Scheme 1.

A. K. Ghosh et al. / Bioorg. Med. Chem. Lett. 15 (2005) 15–20
17
Table 1. Structure and potencies of memapsin 2 inhibitors
valine methyl ester hydrochloride and HunigÕs base in
CH2Cl2 was slowly added to a solution of triphosgene
in CH2Cl2. The resulting isocyanate was then reacted
with allyl alcohol, 3-buten-1-ol, or 4-penten-1-ol for
24h to afford the corresponding urethanes.13 Saponifica-
tion of the methyl esters with aqueous lithium hydroxide
furnished the valine derivatives 10a–c (64–68% yield).
Entry Compound
Ki (nM)
O
OH Me
N
H
HN
H
N
H
N
1
2
26.6
87.2
H
N
O
O
O
HN
O
O
12a
OH Me
3a
O
Exposure of 9 to trifluoroacetic acid (20% in CH2Cl2) ef-
fected the removal of the Boc group. To the resulting
TFA salt, HunigÕs base was added to liberate the free
amine. This was coupled with the valine derivatives
10a–c under standard peptide coupling conditions
(EDCI, HOBt, DIPEA) to afford olefins 6a–c in 60–
65% yield. Ring-closing metathesis of dienes 6a–c pro-
ceeded smoothly in the presence of GrubbsÕ second gen-
eration catalyst14 to afford macrocyclic trans-olefins as
the major products in very good yields (63% for 11a;
68% for 11b and 73% for 11c).15 Saponification of
11a–c with aqueous LiOH afforded the acids 4a–c in
near quantitative yields.
Ph
O
N
H
N
H
N
H
HN
H
N
O
O
O
O
HN
O
O
O
Ph
O
OH Me
O
N
H
H
N
H
N
3
217.6
HN
H
O
N
HN
O
Ph
O
2a
O
O
Synthesis of acyclic and cyclic inhibitors are shown in
Scheme 2. Selective removal of the Boc-group from
the known silyl ether 5 by treatment with trifluoroacetic
acid (20% in CH2Cl2) at 23°C for 30min provided the
corresponding amine. Coupling of the resulting amine
with acids derived from aqueous LiOH promoted
OH Me
N
H
N
H
N
H
4
61.4
HN
H
N
O
O
O
HN
O
O
12b
O
Ph
O
N
OH Me
H
N
H
N
H
O
5
28.4
HN
H
N
N
H
OH Me
O
O
H
N
H
N
O
HN
O
3b
O
HN
( )n
O
O
H
N
Ph
O
O
HN
O
O
O
12a-c
O
Ph
N
OH Me
H
N
H
H
O
N
O
1. 6a-c-derived acid,
EDCI, HOBt, DIPEA (53-55%)
2. Aq. HF (35-40%)
6
108.2
HN
O
O
O
HN
TFA
CH2Cl2
HN
O
O
2b
5
Ph
4a-c, EDCI,
HOBt, DIPEA (36-40%)
O
OH Me
N
H
HN
OR Me
H
N
H
N
H
H
N
H
HN
N
N
O
7
112.1
( )n
O
H
N
H
N
O
O
O
O
HN
O
O
12c
HN
O
O
O
O
Ph
Ph
NH
R=TBS: 13a (n=1), 13b (n=2), 13c (n=3)
R=H: 3a (n=1), 3b (n=2), 3c (n=3)
Aq. HF
(60-63%)
O
OH Me
O
H
H
N
N
8
14.2
HN
1. H2, Pd-C
H
(25-28%)
O
O
O
N
2. Aq. HF
HN
O
Ph
3c
O
O
O
N
OH Me
H
N
H
N
H
NH
HN
( )n
O
H
N
OH Me
O
O
O
H
N
H
N
HN
O
O
9
25.1
HN
O
H
Ph
O
O
N
HN
O
O
2a (n=1), 2b (n=2), 2c (n=3)
2c
O
Ph
Scheme 2.

18
A. K. Ghosh et al. / Bioorg. Med. Chem. Lett. 15 (2005) 15–20
saponification of esters 6a–c in the presence of EDCI
and HOBt provided the corresponding coupled products
in 53–55% yields. Removal of the silyl ether with aque-
ous HF in THF yielded the acyclic inhibitors 12a–c in
35–40% yield. For synthesis of cyclic inhibitors, 5-de-
rived amine was coupled with cyclic dipeptides 4a–c as
described above to afford the corresponding silyl ethers
13a–c in modest yield (36–40%). Removal of the silyl
ether furnished 14- to 16-membered mono-unsaturated
macrocyclic inhibitors 3a–c in a typical 60–63% yield.
Macrocyclic 14- to 16-membered saturated inhibitors
were prepared by catalytic hydrogenation of olefins
13a–c, followed by deprotection of silyl ether with aque-
ous HF in THF to afford inhibitors 2a–c in modest
yields (25–28% for the two steps).
the inhibitor-bound crystal structure of the protein–lig-
and complexes. This effort led to the determination of
the three-dimensional structure of 2c-bound memapsin
17
˚
2 by X-ray diffraction at a resolution of 2.8A.
A
stereoview of the inhibitor-bound structure is shown in
Figure 3. As can be seen, the (S)-hydroxyl group of
inhibitor 2c hydrogen bonds with the catalytic aspar-
tates (Asp-32 and Asp-228) of memapsin 2. The P1-
isobutyl side chain and the P01-methyl moieties of 2c
effectively fill in the S1 and S01-regions, respectively, in
the enzyme active site. The P02-inhibitor carbonyl hydro-
˚
gen bonds with Tyr-198 at a distance of 2.41A. The
binding properties of the P2-cycloamide–urethane is
quite intriguing. While the valine side chain and the
urethane side chain fill in the hydrophobic pocket
effectively, the asparagine carbonyl of 2c is within
˚
The inhibitory potencies of various acyclic and cyclic
memapsin 2 inhibitors are shown in Table 1. Inhibitory
potencies of various inhibitors against recombinant
memapsin 2 were measured using our previously de-
scribed assay protocol.16 As can be seen, the size of
the alkyl chains of the acyclic inhibitors is important
for potency. Furthermore, ring size of the mono-unsatu-
rated and saturated macrocyclic rings is critical to the
observed inhibitory activities. Both mono-unsaturated
and saturated 14-membered inhibitors (3a and 2a) have
shown significantly higher inhibitory potencies com-
pared to acyclic inhibitor 12a. It appears inhibitor 3b
with a 15-membered unsaturated ring is more suitable
than the 14-membered inhibitor 3a. As mentioned ear-
lier, based upon the inhibitor-bound crystal structure,
it appeared that a 16-membered ring with a P2-asparag-
ine derivative could effectively interact with the Arg-235
residue as well as fill in the S2–S3 sites. Indeed, the cor-
responding 16-membered mono-unsaturated and satu-
rated inhibitors (3c and 2c) are significantly more
potent than their acyclic counterpart with respective Ki
values of 14 and 25nM against memapsin 2.
hydrogen bonding distance (2.56A between the heavy
atoms) to the Arg-235 of the memapsin 2. In order to
analyze the protein–ligand interactions in greater preci-
sion for further improvements of this class of inhibitor
with structure-based design, higher resolution crystal
structures will be attempted.
Memapsin 1 (BACE-2) is the closest homologous pro-
tease to memapsin 2. Some circumstantial evidence indi-
cates that memapsin 1 is linked to the pathogenesis of
AlzheimerÕs disease, but this has not been established
conclusively.19 If it is shown that memapsin 1 is not in-
volved in the disease pathogenesis, then selectivity over
memapsin 1 is desirable. Conversely, if involvement of
memapsin 1 is proven, along with memapsin 2, it may
be a therapeutic target. In any event, we have examined
inhibitory properties of inhibitors 2c and 3c against
memapsin 1 as well. Inhibitors 2c and 3c inhibited
memapsin 1 with Ki values of 4.5 and 31nM, respec-
tively. It appears that the E-olefin in inhibitor 3c is
responsible for its slight selectivity over memapsin 1.
Interestingly, saturated inhibitor 2c is 5-fold more po-
tent for memapsin 1 than memapsin 2. This cross inhibi-
tion of memapsin 1 is not totally unexpected since
memapsin 1 also cleaves the b-secretase site of APP
To gain further insight into the molecular binding prop-
erties of inhibitors 2c and 3c, we attempted to resolve
Figure 3. Inhibitor 2c (magenta) bound X-ray crystal structure of memapsin 2.

A. K. Ghosh et al. / Bioorg. Med. Chem. Lett. 15 (2005) 15–20
19
and APP with Swedish mutations.20 A structure-based
design strategy may aid the design of inhibitors with
enhanced selectivity over memapsin 1. We have also
examined cellular inhibition of memapsin 2 by 2c in
Chinese hamster cells. It has shown an average cellular
IC50 value of 3.9 1.2lM (n = 2) compared to IC50
value of 45 5lM for OM-99-2 (Ki = 1.2nM).21
Walker, D. E.; Davis, D.; Thorsett, E. D.; Jewett, N. E.;
Moon, J. B.; Varghese, J. J. Med. Chem. 2004, 47, 158.
8. (a) Ermolieff, J.; Loy, J. A.; Koelsch, G.; Tang, J.
Biochemistry 2000, 39, 12450; (b) Turner, R.; Koelsch,
G.; Hong, L.; Castenheira, P.; Ghosh, A. K.; Tang, J.
Biochemistry 2001, 40, 10001.
9. Hong, L.; Koelsch, G.; Lin, X.; Wu, S.; Terzyan, S.;
Ghosh, A. K.; Zhang, X. C.; Tang, J. Science 2000, 290,
150.
In conclusion, structure-based design of cycloamide–
urethane functionality at P2–P3 resulted in a novel series
of memapsin 2 inhibitors. These cyclic ligands are de-
signed to allow effective hydrogen bonding with specific
residues in the S2-active site. Interestingly, 16-membered
mono-unsaturated and saturated inhibitors (3c and 2c)
are the most potent inhibitors. A preliminary inhibitor-
bound mempasin 2 structure of 2c indicated that the
asparagine carbonyl of 2c is within hydrogen bonding
distance to the Arg-235 of memapsin 2. This interaction
is apparent in the X-ray structure of OM-99-2-bound
memapsin 2. Further investigations including specific
nonpeptidal ligand design from these preliminary stud-
ies are currently underway.
10. Grubbs, R. H.; Chang, S. Tetrahedron 1998, 54,
4413.
11. Ghosh, A. K.; Swanson, L. M.; Hussain, K. A.; Cho, H.;
Walters, D. E.; Holland, L.; Buthod, J. Bioorg. Med.
Chem. Lett. 2002, 12, 1993.
12. Available from Aldrich Chemical Co., Milwaukee, WI.
13. N,N0-Disuccinimidyl carbonate promoted alkoxycarbonyl-
ation protocols also provided similar results, see: (a)
Ghosh, A. K.; Duong, T. T.; McKee, S. P. Tetrahedron
Lett. 1991, 32, 4251; (b) Ghosh, A. K.; Duong, T. T.;
McKee, S. P.; Thompson, W. J. Tetrahedron Lett. 1992,
33, 2781.
14. Chatterjee, A. K.; Morgan, J. P.; Scholl, M.; Grubbs, R.
H. J. Am. Chem. Soc. 2000, 122, 3783, and references cited
therein.
15. Olefin metathesis typically provided a 4:1 mixture of trans/
cis isomers for 14- to 16-membered rings. The cis/trans
mixture has been separated by silica gel chromatography
using 25% ethyl acetate and hexanes as the eluent. The
olefin geometry of the major trans-isomers was established
by extensive decoupling experiments. The coupling con-
stants of the olefinic protons (Jab = 15–16Hz) indicated
that the trans-cycloamides are formed.
Acknowledgements
Financial support by the National Institutes of Health
(AG 18933) is gratefully acknowledged. J.T. is the
holder of the J.G. Puterbaugh Chair in Biomedical Re-
search at the Oklahoma Medical Research Foundation.
The authors thank Mr. Debasis Manna for experimental
assistance.
16. Memapsin 2 inhibition was measured using recombinant
enzyme produced from E. coli expression as described
in Ref. 5a. A fluorogenic substrate Arg-Glu (EDANS)-
Glu-Val-Asn-Leu-Asp-Ala-Glu-Phe-Lys
(Dabcyl)-Arg
was used with 0.47lM of the enzyme in 0.1M Na-
acetate + 5% dimethylsulfoxide, pH4.5 at 37°C. The
excitation wavelength was 350nm and the emission
wavelength was 490nm. Details of this assay will be
published in due course.
References and notes
1. (a) Selkoe, D. J. Nature 1999, 399A, 23; (b) Selkoe, D.
Physiol. Rev. 2001, 81, 741.
2. Vassar, R.; Citron, M. Neuron 2000, 27, 419.
3. (a) Roggo, A. Curr. Top. Med. Chem. 2002, 2, 359; (b)
Olson, R. E.; Thompson, L. A. Ann. Rep. Med. Chem.
2000, 35, 31.
17. The protein–ligand X-ray structure of 2c-bound memapsin
2 has been deposited in PDB with the access code: 1XS7.
Memapsin 2/inhibitor 2c crystal was soaked in the mother
liquor plus 20% (v/v) glycerol and quickly frozen under a
cryogenic nitrogen gas stream. Diffraction data of a single
crystal was recorded on a Mar 345 image plate mounted
on a MSC-Rigaku RU-300 X-ray generator with Osmic
focusing mirrors. Data were integrated and merged using
previously described protocols.9,16 The crystal form was
4. Vassar, R.; Bennett, B. D.; Babu-Khan, S.; Kahn, S.;
Mendiaz, E. A.; Denis, P.; Teplow, D. B.; Ross, S.;
Amarante, P.; Loeloff, R.; Luo, Y.; Fisher, S.; Fuller, J.;
Edenson, S.; Lile, J.; Jarosinski, M. A.; Biere, A. L.;
Curran, E.; Burgess, T.; Louis, J. C.; Collins, F.; Treanor,
J.; Rogers, G.; Citron, M. Science 1999, 286, 735.
5. (a) Lin, X.; Koelsch, G.; Wu, S.; Downs, D.; Dashti, A.;
Tang, J. Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 1456; (b)
Sinha, S.; Lieberburg, I. Proc. Natl. Acad. Sci. U.S.A.
1999, 96, 11049.
6. (a) Ghosh, A. K.; Shin, D.; Downs, D.; Koelsch, G.; Lin,
X.; Ermolieff, J.; Tang, J. J. Am. Chem. Soc. 2000, 122,
3522; (b) Ghosh, A. K.; Bilcer, G.; Harwood, C.;
Kawahama, R.; Shin, D.; Downs, D.; Hussain, K. A.;
Hong, L.; Loy, J. A.; Nguyen, C.; Koelsch, G.; Ermolieff,
J.; Tang, J. J. Med. Chem. 2001, 44, 2865.
7. (a) Shuto, D.; Kasai, S.; Kimura, T.; Liu, P.; Koushi, H.;
Takashi, H.; Shibakawa, S.; Hayashi, Y.; Hattori, C.;
Sazbo, B.; Isiura, S.; Kiso, Y. Bioorg. Med. Chem. Lett.
2003, 13, 4273; (b) Tung, J. S.; Davis, D. L.; Anderson, J.
P.; Walker, D. E.; Mamo, S.; Jewett, N. E.; Hom, R. K.;
Sinha, S. J. Med. Chem. 2002, 45, 259; (c) Hom, R. K.;
Gailunas, A. F.; Mamo, S.; Fang, L. Y.; Tung, J. S.;
˚
determined to be monoclinic with a resolution of 2.8A.
The unit cell parameters are a = 85.8, b = 87.8, c = 130.6,
b = 89.8°.
The structure was determined by molecular replacement
implemented with the program AmoRe using the previ-
ously determined memapsin 2 structure (PDB ID: 1M4H)
as a search model.18 Rotation and translation functions
followed by the rigid-body refinement with data from 15
˚
to 3.5A resolution in space group P21 gave unambiguous
solutions for the four memapsin 2 molecules in the
asymmetric unit. A random selection of 8% of reflections
was set aside as the test set for cross-validation during the
refinement. The refined model has well defined electron
density for the inhibitor and its corresponding structure
was built into the active site. The four molecules in the
crystallographic asymmetric unit have essentially identical
structures.
18. Hong, L.; Turner, R. T., III; Koelsch, G.; Shin, D.;
Ghosh, A. K.; Tang, J. Biochemistry 2002, 41, 10963.

20
A. K. Ghosh et al. / Bioorg. Med. Chem. Lett. 15 (2005) 15–20
19. Yan, R.; Bienkowski, M. J.; Shuck, M. E.; Miao, H.;
Troy, M. C.; Pauley, A. M.; Brashier, J. R.; Stratman, N.
C.; Mathewa, W. R.; Buhl, A. E.; Carter, D. B.;
Tomasselli, A. G.; Parodi, L. A.; Heinrikson, R. L.;
Gurney, M. E. Nature 1999, 402, 533.
20. (a) Bennett, B. D.; Babu-khan, S.; Loeloff, R.; Louis, J. C.;
Curran, E.; Citron, M.; Vasser, R. J. Biol. Chem. 2000,
275, 20647; (b) Farzan, M.; Schnitzler, C. E.; Vasilieva,
N.; Leung, D.; Choe, H. Proc. Natl. Acad. Sci. U.S.A.
2000, 97, 9712.
21. Cells transfected with the human APP Swedish mutation
were cultured essentially as described previously.22 Con-
ditioned media was collected following a 24h incubation
with compound OM99-2 in 0.4% DMSO and assayed for
Ab40 by sandwich ELISA (BioSource International) for
determination of IC50.
22. Chang, W. P.; Koelsch, G.; Wong, S.; Downs, D.; Da, H.;
Weerasena, V.; Gordon, B.; Devasamudram, T.; Bilcer,
G.; Ghosh, A. K.; Tang, J. J. Neurochem. 2004, 89,
1409.