Design and synthesis of hydroxyethylamine (HEA) BACE-1 inhibitors: Structure-activity relationship of the aryl region

Article abstract of DOI:10.1016/j.bmcl.2010.08.070

The structure-activity relationship of the prime region of hydroxyethylamine BACE inhibitors is described. Variation in the aryl linker region with 5- and 6-membered heterocycles provided compounds such as 33 with improved permeability and reduced P-gp liability compared to benzyl amine analog 1.

Full text of DOI:10.1016/j.bmcl.2010.08.070

Bioorganic & Medicinal Chemistry Letters 20 (2010) 6034–6039
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Design and synthesis of hydroxyethylamine (HEA) BACE-1 inhibitors:
Structure–activity relationship of the aryl region
Gary D. Probst a, Simeon Bowers a, , Jennifer M. Sealy a, Brian Stupi a, Darren Dressen a,
⇑
Barbara M. Jagodzinska a, Jose Aquino a, Andrea Gailunas a, Anh P. Truong a, Luke Tso a, Ying-Zi Xu a,
Roy K. Hom a, Varghese John a, Jay S. Tung a, Michael A. Pleiss a, John A. Tucker a, Andrei W. Konradi a,
Hing L. Sham a,b, Jacek Jagodzinski b, Gergely Toth c, Eric Brecht c, Nanhua Yao c, Hu Pan c, May Lin c,
Dean R. Artis c, Lany Ruslim d, Michael P. Bova d, Sukanto Sinha d, Ted A. Yednock a,b,c,d,e, Shawn Gauby e,
Wes Zmolek e, Kevin P. Quinn e, John-Michael Sauer e
a Department of Medicinal Chemistry, Elan Pharmaceuticals, 180 Oyster Point Boulevard, South San Francisco, CA 94080, United States
b Department of Process and Analytical Chemistry, Elan Pharmaceuticals, 180 Oyster Point Boulevard, South San Francisco, CA 94080, United States
c Department of Molecular Design, Elan Pharmaceuticals, 180 Oyster Point Boulevard, South San Francisco, CA 94080, United States
d Department of Biology, Elan Pharmaceuticals, 1000 Gateway Boulevard, South San Francisco, CA 94080, United States
e Department of Lead Finding, Drug Disposition, and Safety Evaluation, Elan Pharmaceuticals, 800 Gateway Boulevard, South San Francisco, CA 94080, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 7 July 2010
Revised 11 August 2010
Accepted 12 August 2010
Available online 19 August 2010
The structure–activity relationship of the prime region of hydroxyethylamine BACE inhibitors is
described. Variation in the aryl linker region with 5- and 6-membered heterocycles provided compounds
such as 33 with improved permeability and reduced P-gp liability compared to benzyl amine analog 1.
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
Alzheimer’s disease
Beta-secretase inhibitor
BACE-1
HEA
Hydroxyethylamine
Alzheimer’s disease (AD) is the most prevalent form of demen-
tia and is characterized pathologically by the formation of extracel-
lular insoluble amyloid plaques and intraneuronal fibrillary tangles
in the brain.1 The amyloid plaques are composed of b-amyloid
(Ab)2 peptide which is generated by cleavage of the b-amyloid pre-
cursor protein (b-APP) by two aspartic acid proteases referred to as
substituents fill the S10 and S20 pockets, respectively. The aryl lin-
ker occupies a region between S10 and S20 with Gly34 and Tyr198
on the ‘top’ side and Thr72, Thr71 and Pro70 at the bottom right
side. The lower portion of the benzylamine is exposed to solvent.
During our systematic exploration of the prime region of BACE7,
the SAR of the aryl region of the inhibitors was targeted for study.
Our previously reported HEA inhibitors such as 1 suffered from low
b-secretase (BACE-1) and c-secretase. BACE-1 knockout mice are
healthy, fertile3 and are unable to generate Ab in the brain4 sug-
gesting that BACE-1 would be an attractive therapeutic target for
the development of disease modifying treatments for AD.5
S1
During the course of our program to develop BACE-1 inhibitors
based on the hydroxyethylamine (HEA) scaffold6 we discovered
that aryl cyclohexyl amines provided excellent potency against
BACE-1.7 The general binding mode of this series of BACE-1 inhib-
itors is depicted in Figure 1. The hydroxyl and protonated amino
group bind to Asp228 and Asp32, respectively, while the difluoro-
aryl occupies the S1 pocket and the cyclohexyl and tert-butyl
F
S1'
S2'
F
O
N
H
N
H
3
1
OH
5
1
aryl linker
⇑
Corresponding author. Tel.: +1 650 616 2630.
E-mail address: simeon.bowers@elan.com (S. Bowers).
Figure 1. Binding mode of the HEA’s.
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.08.070

G. D. Probst et al. / Bioorg. Med. Chem. Lett. 20 (2010) 6034–6039
6035
Table 1
F
BACE inhibition data and cellular activity for compounds 1–17
F
I
a, b
c, d
H2N
F
H2N
BocHN
F
I
O
I
2
3
4
5
O
3R1
F
N
H
N
1
H
F
OH
5
R2
O
e, f, g
N
H
N
a
b
c
Compds R1
R2
BACE IC50
(nM)
,
Cat D IC50
(nM)
,
Cell ED50
(nM)
,
H
OH
I
7
1
H
H
H
>10,000
47
na
24
>1000
17
6
tert-
Butyl
neo-
Pentyl
tert-
Butyl
tert-
Butyl
tert-
Butyl
tert-
Butyl
tert-
Butyl
tert-
Butyl
tert-
Butyl
tert-
Butyl
tert-
Butyl
tert-
Butyl
Scheme 1. Reagents and conditions: (a) (i) BnEt3NCl, ICl, H2O, then (ii) 2, CaCO3,
CH2Cl2, MeOH, 18 h, 40 °C; (b) t-butyl nitrite, DMF, reflux, 10 min; (c) (i)
cyclohexanone, t-butyl sulfinamide, Ti(OEt)4, THF, 2 h, room temperature, (ii)
8
9
H
F
2100
46
125
24
440
9
i
n-BuLi then 3 AlMe3, toluene, ꢀ78 °C; (d) HCl, MeOH, dioxane; (e) 5, DIPEA, PrOH,
18 h, 70 °C; (f) 4 N HCl in dioxane, 0.5 h; (g) Ac2NOCH3, CH2Cl2, Et3N, 18 h.
10
6
Br
I
174
223
180
77
42
435
185
240
68
55
permeability and high P-gp efflux and our goal was to develop new
inhibitors which possessed cellular activity, were highly permeable
and were able to penetrate the blood–brain barrier. It was envi-
sioned that the introduction of substituents and heteroatoms into
the aryl linker would enable modulation of the pharmacokinetic
properties of the inhibitors and enable this goal to be achieved. Ini-
tial exploration focused on substituted benzene analogs which
could be prepared using a route outlined in Scheme 1.7 For exam-
ple, the versatile aryl iodide 6 was prepared by addition of an aryl
lithium to the N-sulfinyl imine derived from cyclohexanone and
tert-butyl sulfinamide followed by hydrolysis of the resulting sul-
finamide to give cyclohexylamine 4. Alkylation of epoxide 5 fol-
lowed by deprotection and acetylation gave the desired
compounds. In the case of aryl iodide 6, further functionalization
using cross-coupling methodology allowed expansion of the SAR
at the 5-position.
11
12
13
14
15
16
17
Me
67
OMe
OH
47
170
318
93
90
na
NH2
105
12
25
NHSO2Me
C(O)Me
4-Br
na
160
6900
45
33
1500
2000
a
b
c
See Ref. 10.
See Ref. 11.
See Ref. 12.
As previously reported7 the tert-butyl analog 1 was identified as
one of the more potent inhibitors in this series being significantly
more potent than the neo-pentyl analog 8 (Table 1). These alkyl
chains point into the S20 region of BACE-1 and are crucial to provid-
ing potency as illustrated with mono-substituted benzyl amine
analog 7. Exploration of the SAR on the 5-position with halogens
indicated that larger substituents resulted in a loss of potency in
the order F < Br < I. Interestingly the methoxy analog 12 had simi-
lar potency to 1 while the methyl analog 11 and phenol 13 were
three-fold less potent. Aniline 14 lost seven-fold potency compared
to 1 but methylsulfonamide 15 gained back some potency and was
only two-fold less active against BACE-1 than 1. Substitution at the
4-position was not tolerated as shown with 17. Generally, these
compounds were all more potent against Cat-D over BACE-1.
In addition to exploring the SAR of the benzene linker, a series
of heterocyclic aryl linkers was also prepared. Thiophenes are well
known as phenyl isosteres8 and a series of thiophene regioisomers
was synthesized with various alkyl substituents to fill the S20 pock-
et. One of the general synthetic routes used to access these com-
pounds is outlined in Scheme 2.9 Metal–halogen exchange of a
commercially available di-bromothiophene followed by quenching
with cyclohexanone gave the tertiary alcohols which were con-
verted into amines upon treatment with azidotrimethylsilane
and BF3OEt2. Reduction of the resulting azide with LiAlH4 gave
amine 20 which was elaborated into the HEA using the standard
protocol to give the key synthetic intermediate 21. The P20 substit-
uents could be conveniently installed using cross-coupling meth-
odology, such as Stille, Negishi, and Sonogashira couplings, to
provide the compounds shown in Table 2.
Br
S
b, c
a
d, e, f
HO
Br
H2N
F
Br
Br
S
S
18
19
20
F
F
F
g
O
O
N
H
N
N
H
N
Br
H
H
OH
21
S
OH
22
S
Scheme 2. Reagents and conditions: (a) cyclohexanone, t-BuLi, THF 1 h, ꢀ78 °C; (b)
i
TMS-N3, BF3OEt2, Et2O, reflux, 1.5 h; (c) LiAlH4, Et2O, 2 h; (d) 5, DIPEA, PrOH, 18 h,
70 °C; (e) 4 N HCl in dioxane, 0.5 h; (f) Ac2NOCH3, CH2Cl2, Et3N, 18 h; (g)
IZnCH2C(CH3)3, Pd2dba3ꢁCHCl3, DTBP, THF, reflux, 20 h.
The 2,5-disubstituted thiophenes proved to be less active than
the 2,4-disubstituted analogs. For example, the most potent 2,5-
disubstituted thiophene, 24 (BACE IC50 = 4.0
magnitude less potent than the corresponding 2,4-disubstituted
thiophenes 25 (BACE IC50 = 0.32 M) and 26 (BACE IC50 = 0.35 M).
lM) was an order of
l
l
Further exploration of the SAR of the S20 region indicated that tert-
butyl 27 and neo-pentyl 22 were the most potent analogs.
Since it was possible to replace the benzene ring with a thio-
phene ring and achieve similar BACE-1 inhibition both in the
enzyme and cellular assays, the SAR of this region was expanded

6036
G. D. Probst et al. / Bioorg. Med. Chem. Lett. 20 (2010) 6034–6039
Table 2
O
BACE inhibition data and cellular activity for compounds 22–29
O
a
H
N
OH
H2N
BocNH
BocNH
F
F
O
O
38
36
b
37
F
O
F
N
N
Ar
O
H
N
H
H2N
OH
c, d, e
N
N
H
O
N
a
b
c
H
Compds Ar
BACE IC50
(nM)
,
Cat D IC50
(nM)
,
Cell ED50
(nM)
,
OH
O
40
39
S
S
23
24
>30,000
4010
4420
411
1330
1036
Scheme 4. Reagents and conditions: (a) EDCI, HOAt, Et3N, CH2Cl2, rt; (b) POCl3,
105 °C; (c) 5, DIPEA, iPrOH, 18 h, 70 °C; (d) 4 N HCl in dioxane, 0.5 h; (e) Ac2NOCH3,
CH2Cl2, Et3N, 18 h.
25
26
320
42
<80
S
I
e, f, g
c, d
a, b
44
352
104
40
32
<80
10
NH
H2N
S
N
N
N
N
42
N
27
22
41
43
S
S
131
9
na
Br
j, k
e, f, g
h, i
48
NH
H2N
HO
28
29
4316
1740
276
228
757
107
N
N
N
S
S
N
N
45
47
46
Scheme 5. Reagents and conditions: (a) H2SO4, t-BuOH, 100 °C; (b) iPrMgCl then
cyclohexanone, 0 °C; (c) NaN3, TFA, CH2Cl2 rt; (d) Pd-C, H2, MeOH; (e) 5, DIPEA,
iPrOH, 18 h, 70 °C; (f) 4 N HCl in dioxane, 0.5 h; (g) Ac2NOCH3, CH2Cl2, Et3N, 18 h;
(h) n-BuLi then cyclohexanone ꢀ78 °C to 0 °C; (i) NaH, neo-pentyl iodide, DMF,
50 °C; (j) NaN3, TFA, CH2Cl2; (k) Pd(OH)2, H2, MeOH.
a
b
c
See Ref. 10.
See Ref. 11.
See Ref. 12.
by the preparation of a more diverse series of heterocyclic aryl
linkers with particular attention paid to the effect on the pharma-
cokinetic properties of the compounds.
Isoxazoles and 1,2,3-triazoles were prepared by cycloaddition of
the commercially available alkyne 30 with aldoximes and alkyl
azides, respectively (Scheme 3). Oxazoles were prepared from ami-
no acids using the Robinson–Gabriel oxazole synthesis (Scheme 4)
and pyrazoles were prepared by the addition of a metalated-pyra-
zole into cyclohexanone followed by elaboration into the cyclo-
hexylamines (Scheme 5).
order of magnitude less potent (Table 3). This is in contrast to the
phenyl series where the tert-butyl analog (1) is significantly more
potent than neo-pentyl analog (8). Interestingly, in the thiophene
series the neo-pentyl (22) and tert-butyl (27) analogs were equipo-
tent suggesting that subtle differences in the bond lengths and
bond angles between the five- and six-membered rings places
the P20 substituent in optimal or sub-optimal positions within
the pocket. Furthermore, the tert-butyl substituted oxazoles (40
and 49) showed dramatic differences in potency with the tert-butyl
substituent in the 5-position (49) being almost three orders of
magnitude more potent than the analog with the P20 substituent
at the 4-position (40). Interestingly, the S20 pocket proved to be
able to accommodate groups as large as adamantyl as demon-
strated with triazole 35. Pyrimidine 59 lost 50-fold potency com-
pared to phenyl analog 1.
In this series of compounds bearing heterocyclic aryl linkers,
the analogs with a neo-pentyl P20 substituent had potency compa-
rable to the 1 while the tert-butyl analogs were approximately an
F
In addition to exploring the SAR of the aryl region of the cyclo-
hexylamine based inhibitors, a range of cyclopropyl amines bear-
ing various aryl and heteroaryl linkers were prepared. In general,
F
O
d, e, f
a, b, c
N
H
N
H
the synthesis of these analogs relied upon the conversion of a suit-
R
Z
H2N
H2N
R
Z
13
OH
X
N
ably substituted heterocyclic nitrile to a cyclopropyl amine
fol-
X
N
lowed by elaboration into the HEA as depicted in Scheme 6.14
30
31
As with the cyclohexyl amine series, the phenyl linker com-
bined with a tert-butyl P20 substituent proved to be the most po-
tent combination (54, IC50 = 170 nM) (Table 4) although it was
three-fold less active than its cyclohexyl counterpart 1. This differ-
ence may be attributed to the less effective filling of the S10 pocket
and/or due to differences in the conformation of the rings. Replace-
ment of the benzene ring with pyridine gave a moderate loss in
potency.
32 X = O Z = C R = tBu
33 X = O Z = C R = neopentyl
34 X = N Z = N R = CH2TMS
35 X = N Z = N R = adamantyl
Scheme 3. Reagents and conditions: (a) (Boc)2O, CH2Cl2 X days, rt; (b) for 32 and 33
(i) RCHO, hydroxylamine hydrochloride, pyridine, ethanol, (ii) NCS, Et3N, CH2Cl2; for
34 and 35 RN3; (c) 4 N HCl in dioxane, 0.5 h; (d) 5, DIPEA, iPrOH, 18 h, 70 °C; (e) 4 N
HCl in dioxane, 0.5 h; (f) Ac2NOCH3, CH2Cl2, Et3N, 18 h.

G. D. Probst et al. / Bioorg. Med. Chem. Lett. 20 (2010) 6034–6039
6037
Table 3
Table 4
BACE inhibition data and cellular activity for compounds 32–49
BACE inhibition data and cellular activity for compounds 54–58
F
F
F
F
O
O
N
N
N
Ar
N
Ar
H
H
H
H
OH
OH
a
b
c
a
b
c
Compds Ar
BACE IC50
(nM)
,
Cat D IC50
(nM)
,
Cell ED50
(nM)
,
Compds Ar
BACE IC50
(nM)
,
Cat D IC50
(nM)
,
Cell ED50
(nM)
,
32
4457
139
263
14
3711
88
54
55
56
140
341
640
290
379
850
51
720
229
O
N
33
O
N
O
49
40
1580
310
760
N
O
N
N
13,7800
12,170
>10,000
57
58
780
720
530
310
N
44
48
N
N
1840
210
140
29
235
23
N
N
1050
8700
N
O
a
b
c
See Ref. 10.
See Ref. 11.
See Ref. 12.
N
N
34
35
59
154
392
12
8
6
N
N
N
N
Si
na
348
CH in this position is crucial for activity. Examination of the crystal
structure of 53 (Fig. 2) clearly shows the CH in the 3-position of the
thiophene engaging in a CHꢁ ꢁ ꢁO15 non-classical hydrogen bond
with the carbonyl of Gly34 at a distance of 3.2 Å. This interaction
explains the observed SAR in this region, the addition of heteroat-
oms such as N or O causes a repulsive interaction with Gly34 C@O
resulting in reduced BACE-1 potency.
2430
245
N
N
a
b
c
See Ref. 10.
See Ref. 11.
See Ref. 12.
The in vitro derived pharmacokinetic properties for a selected
number of inhibitors are shown in Table 6. In the cyclohexylamine
series, the permeability was poor with benzene as the aryl linker
but replacement with both five-membered and six-membered het-
erocyclic aromatic rings gave compounds with generally good per-
meability. For example, isoxazole 33 had much higher
permeability (180 nm/s) and lower P-gp liability compared to ben-
zylamine 1 (4 nm/s). In general, the cyclopropylamine analogs
have increased permeability and reduced P-gp liability compared
to the cyclohexylamines as exemplified by comparing analog 54
to compound 1. Interestingly, thiazole 64 proved not to be a sub-
strate for P-gp.
NC
OHC
S
c, d H2N
a, b
e, f, g
53
Br
Br
S
S
50
51
52
Scheme 6. Reagents and conditions: (a) hydroxylamine hydrochloride, EtOH, (b)
Ac2O; (c) neo-pentylzinc iodide, PdCl2[(o-Tol3P)]2, THF; (d) EtMgBr, Ti(OiPr)4,
BF3OEt2, Et2O; (e) 5, DIPEA, iPrOH, 18 h, 70 °C; (f) 4 N HCl in dioxane, 0.5 h; (g)
Ac2NOCH3, CH2Cl2, Et3N, 18 h.
The pKa of a subset of these compounds was measured (Table 6)
revealing a good correlation between the basicity of the cycloalkyl
amine and the permeability of the compounds. Moving from a
cyclohexyl amine (1) to a cyclopropyl amine (54) reduces the pKa
by 1.6 units resulting in a neutral molecule in the assay medium
and allowing greater permeability through the lipid bi-layer.7b
Similar effects were observed by replacing the phenyl group with
heterocycles, allowing the pyrimidine 59 (pKa = 6.7) to have much
higher permeability (Papp = 265 nm/s) compared to 1 (pKa = 8.2,
Papp = 4 nm/s) but unfortunately this improvement in permeability
did not have any effect on the P-gp efflux ratio of these com-
pounds.16 The log P of the inhibitors was also measured and it indi-
cates that a high rate of permeability is only obtained when the
log P falls in the range of 3.0–4.3 and it declines dramatically when
the log P of the compound is greater than 4.6.
As observed with the cyclohexylamine series, the position of the
sulfur atom in the thiophenes proved to be important in
determining the BACE-1 activity of the inhibitors (Table 5). 2,4-
Disubstituted thiophenes were 3–6 fold more potent that their
2,5-disubstituted analogs. A screen of oxazole (prepared as in
Scheme 4, using amino cyclopropyl carboxylic acid in place of
amino cyclohexyl carboxylic acid) and thiazole aryl linkers was
performed but this did not provide any compounds with sub-
micromolar potency (Table 5). As with the cyclohexylamine series,
placing the nitrogen atom of the heterocycle in between the cyclo-
alkyl and the P20 substituent resulted in a dramatic loss in potency
with nitrogen having a more deleterious effect than oxygen. Pyra-
zole 67 and triazole 69 were the most potent analogs in this series
and had activity similar to that of thiophene 53 indicating that a

6038
G. D. Probst et al. / Bioorg. Med. Chem. Lett. 20 (2010) 6034–6039
Table 5
Table 6
BACE inhibition data and cellular activity for compounds 53–64
Pharmacokinetic properties of selected inhibitors
F
F
F
F
O
N
O
n
N
N
N
Ar
Ar
H
H
H
H
OH
OH
a
b
c
c
a
Compds Ar
BACE IC50
(nM)
,
Cat D IC50
(nM)
,
Cell ED50
(nM)
,
Compds
n
Ar
log P pKa
Papp (nm/s) %
P-gp
recovery
effluxb
S
60
53
1620
470
1720
395
1475
346
1
33
59
4
4
4
4.8
4.2
3.7
8.2
6.4
6.7
4 nm/s 7%
19
3.5
20
180 nm/s 100%
265 nm/s 112%
S
O
N
O
N
N
62
63
4900
3470
>10,000
>10,000
N
O
S
N
N
42,880
>200,000
54
57
64
1
1
1
4.6
4.2
4.3
6.6
6.3
na
50 nm/s 27%
135 nm/s 94%
113 nm/s 54%
1.3
na
1
64
65
67
68
69
9360
3050
346
3110
1950
720
>10,000
3850
68
N
S
N
N
S
N
N
N
N
N
N
67
1
3.0
7.0
106 nm/s 67%
11
N
870
3080
200
430
a
b
c
See Ref. 17.
See Ref. 18.
See Ref. 19.
450
725
N
N
a
b
c
See Ref. 10.
See Ref. 11.
See Ref. 12.
previously reported compounds. The addition of heterocycles into
the aryl linker region of the inhibitors lowered the pKa and log P of
the compounds into a range that is amenable to high rates of per-
meability. Subsequent efforts to address the remaining issues with
these compounds will be reported in due course.
Acknowledgements
We thank Nancy Jewett, Jill Labbe, Jim Miller, Colin Lorentzen,
C. Yin, Michael S. Dappen, David Quincy, and Lee Latimer for their
contributions to this work.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2010.08.070.
References and notes
1. Ferri, C.; Prince, M.; Brayne, C.; Brodaty, H.; Fratiglioni, L.; Ganguli, M.; Hall, K.;
Hasegawa, K.; Hendrie, H.; Huang, Y. Lancet 2005, 366, 2112.
2. Hardy, J.; Selkoe, D. J. Science 2002, 297, 353.
Figure 2. Crystal structure of truncated (57–453) human BACE-1 bound to 53 in
green (2.2 Å resolution). The PBD deposition code is 3NSH. See supplemental
material for experimental details.
3. (a) Luo, Y.; Bolon, B.; Kahn, S.; Bennett, B. D.; Babu-Khan, S.; Denis, P.; Fan, W.;
Kha, H.; Zhang, J.; Gong, Y.; Martin, L.; Louis, J.-C.; Yan, Q.; Richards, W.; Citron,
M.; Vassar, R. Nat. Neurosci. 2001, 4, 231; (b) Cai, H.; Wang, Y.; McCarthy, D.;
Wen, H.; Borchelt, D. R.; Price, D. L.; Wong, P. C. Nat. Neurosci. 2001, 4, 233; (c)
Roberds, S. L.; Anderson, J.; Basi, G.; Bienkowski, M.; Branstetter, D. G.; Chen, K.
S.; Freedman, S. B.; Frigon, N. L.; Games, D.; Hu, K.; Johnson-Wood, K.;
Kappenman, K. E.; Kawabe, T. T.; Kola, I.; Keuhn, R.; Lee, M.; Liu, W.; Motter, R.;
Nichols, N. F.; Power, M.; Robertson, D. W.; Schenk, D.; Schoor, M.; Shopp, G.
In conclusion, optimization of the aryl linker region of a series of
HEA BACE-1 inhibitors resulted in cell potent compounds with im-
proved permeability and reduced P-gp liability compared to our

G. D. Probst et al. / Bioorg. Med. Chem. Lett. 20 (2010) 6034–6039
6039
M.; Shuck, M. E.; Sinha, S.; Svensson, K. A.; Tatsuno, G.; Tintrup, H.; Wijsman, J.;
Wright, S.; McConlogue, L. Hum. Mol. Genet. 2001, 10, 1317.
100 mM sodium acetate buffer pH 4.5 to a 1.8 nM working concentration. A
biotinylated peptide Cathepsin- substrate was labeled at the C-terminus of
D
4. Luo, Y.; Bolon, B.; Damore, M. A.; Fitzpatrick, D.; Liu, H.; Zhang, J.; Yan, Q.;
Vassar, R.; Citron, M. Neurobiol. Dis. 2003, 14, 81.
5. Schmidt, B.; Baumann, S.; Braun, H. A.; Larbig, G. Curr. Top. Med. Chem. 2006, 6,
377.
the peptide with an Oregon Green fluorophore. Compounds were serially
diluted three-fold in DMSO at a 100ꢂ concentration and then subsequently
diluted 33ꢂ with 100 mM sodium acetate buffer pH 4.5. Compound was added
to Cathepsin-D (0.6 nM final concentration) for 30 min before the addition of
6. Maillard, M. C. J. Med. Chem. 2007, 50, 776.
peptide substrate to initiate the reaction. The reaction was allowed to proceed
for 110 min at 37 °C and then quenched by the addition of streptavidin in
200 mM sodium phosphate buffer pH 7.5. The amount of fluorescence
polarization in the well was measured using a LJL Analyst (Perkin Elmer). If
7. (a) Varghese, J.; Maillard, M.; Tucker, J.; Aquino, J.; Hom, R.; Tung, J.; Dressesn,
D.; Shah, N.; Neitz, R. J. WO2005087215, 2005.; (b) Sealy, J. M.; Truong, A. P.;
Tso, L.; Probst, G. D.; Aquino, J.; Hom, R. K.; Jagodzinska, B. M.; Dressen, D.;
Wone, D. W. G.; Brogley, L.; John, V.; Tung, J. S.; Pleiss, M. A.; Tucker, J. A.;
Konradi, A. W.; Dappen, M. S.; Tóth, G.; Pan, H.; Ruslim, L.; Miller, J.; Bova, M. P.;
Sinha, S.; Quinn, K. P.; Sauer, J.-M. Bioorg. Med. Chem. Lett. 2009, 19, 6386.
8. Lima, L. M.; Barreiro, E. J. Curr. Med. Chem. 2005, 12, 23.
Cathepsin-D cleaves the peptide substrate then polarization will be reduced.
12. (a) Dovey, H. F. et al J. Neurochem. 2001, 76, 173. in which the HEK 293 cells
were modified to express the Swedish mutation of APP FAD isoform; (b)
Johnson-Wood, K. et al Proc. Natl. Acad. Sci. U.S.A. 1997, 94, 1550; (c) Seubert, P.
et al Nature 1992, 359, 325.
9. Varghese, J.; Hom, R.; Sealy, J.; Aquino, J.; Probst, G.; Tung, J.; Fang, L.
WO2005070407, 2005.
13. Bertus, P.; Szymoniak, J. J. Org. Chem. 2003, 68, 7133.
10. Compounds serially diluted in DMSO were added to recombinant BACE
purified from E. coli in 100 mM sodium acetate buffer containing 0.001%
Tween-20 at pH 4.5 and allowed to incubate for 20 min at room temperature. A
biotinylated peptide substrate, based on the Swedish mutant APP sequence,
containing an Oregon Green moiety at the C-terminus, was added to initiate
the reaction which was allowed to proceed for 3 h at 37 °C. The reaction was
quenched by the addition of a five-fold volume excess of 100 mM sodium
14. Hom, R.; Toth, G.; Probst, G.; Bowers, S.; Truong, A. WO2007047306, 2007.
15. Toth, G.; Bowers, S. G.; Truong, A. P.; Probst, G. Curr. Pharm. Des. 2007, 13, 3476.
16. Raub, T. J. Mol. Pharm. 2006, 3, 3.
17. Compounds (5 lM) in mHBSS (pH 7.4) were incubated with MDCK II cell
monolayers for 120 min at 37 °C. Samples were taken from apical and
basolateral chambers, and analyzed using LC/MS/MS.
18. Compounds (5 lM) in mHBSS (pH 7.4) were incubated with MDR1-MDCK cell
phosphate buffer pH = 7.4 containing
streptavidin. The extent of fluorescence polarization was measured using a
LJL Analyst.
a
1.5
l
M
final concentration of
monolayers for 120 min at 37 °C with and without a P-gp inhibitor. Samples
were taken from apical (A) and basolateral (B) chambers, and analyzed using
LC/MS/MS. The efflux ratio was determined by dividing the rate of the A to B
direction with and without a P-gp inhibitor.
11. The Cathepsin-
D
was obtained from Sigma (cat. #C8696). Cathepsin-
D was first
dissolved with water to a 2
lM concentration and then subsequently with
19. Determined by capillary electrophoresis.