Structure-based design of novel dihydroisoquinoline BACE-1 inhibitors that do not engage the catalytic aspartates

Article abstract of DOI:10.1016/j.bmcl.2013.01.103

The structure-activity relationship of a series of dihydroisoquinoline BACE-1 inhibitors is described. Application of structure-based design to screening hit 1 yielded sub-micromolar inhibitors. Replacement of the carboxylic acid of 1 was guided by X-ray

Full text of DOI:10.1016/j.bmcl.2013.01.103

Bioorganic & Medicinal Chemistry Letters 23 (2013) 2181–2186
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Structure-based design of novel dihydroisoquinoline BACE-1
inhibitors that do not engage the catalytic aspartates
a
a
a
a
a
Simeon Bowers a, , Ying-zi Xu , Shendong Yuan , Gary D. Probst , Roy K. Hom , Wayman Chan ,
Andrei W. Konradi a, Hing L. Sham a, Yong L. Zhu b, Paul Beroza b, Hu Pan b, Eric Brecht b, Nanhua Yao b,
Julie Lougheed b, Danny Tam b, Zhao Ren b, Lany Ruslim b, Michael P. Bova b, Dean R. Artis b
⇑
a Department of Chemical Sciences, Elan Pharmaceuticals, 180 Oyster Point Boulevard, South San Francisco, CA 94080, USA
b Department of Molecular Design, Elan Pharmaceuticals, 180 Oyster Point Boulevard, South San Francisco, CA 94080, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The structure–activity relationship of a series of dihydroisoquinoline BACE-1 inhibitors is described.
Application of structure-based design to screening hit 1 yielded sub-micromolar inhibitors. Replacement
of the carboxylic acid of 1 was guided by X-ray crystallography, which allowed the replacement of a key
water-mediated hydrogen bond. This work culminated in compounds such as 31, which possess good
BACE-1 potency, excellent permeability and a low P-gp efflux ratio.
Received 16 November 2012
Revised 16 January 2013
Accepted 22 January 2013
Available online 4 February 2013
Ó 2013 Elsevier Ltd. All rights reserved.
Keywords:
Alzheimer’s disease
Beta-secretase
BACE-1 inhibitor
Alzheimer’s disease (AD) is the leading cause of dementia and is
characterized by a slow progressive loss of memory and cognitive
ability.1 Pathologically, AD manifests itself by the formation of
extracellular insoluble amyloid plaques composed of -amyloid
(A) peptide.2 Ab-peptides are generated by the proteolytic cleavage
of the b-amyloid precursor protein (b-APP) by two aspartic acid
OH
HN
1
8
BACE-1 IC50 = 27.2 uM
O
N
6
3
proteases, referred to as b-secretase (BACE-1) and
c-secretase,
1
respectively.3 BACE-1 knockout mice are healthy, and fertile,4 but
because they lack compensatory activity, are unable to generate
A in the brain,5 which suggests that BACE-1 would be an attractive
therapeutic target for the development of disease modifying treat-
ments for AD.6 Consequently, BACE-1 inhibitors have been inten-
sely investigated by many laboratories as potential AD modifying
therapeutics.7,8
Figure 1. Screening hit.
with a biotinylated small PEG linker, gave an approximately 10-
fold boost in assay sensitivity.9 Approximately 2000 hits from a
500 lM HTS screen of 500,000 compounds were crystallographi-
cally screened to identify a BACE-1 inhibitor complex with a novel
structure. Of the compounds identified in this screen, dihydroiso-
quinoline 1 (Fig. 1) was intriguing.
Although it had modest potency against BACE-1
(IC50 = 27.2 lM), the crystal structure revealed an interesting bind-
ing mode (Fig. 2). The (S)-enantiomer of compound 1 occupies the
non-prime region of BACE-1 and interestingly, does not interact
with the catalytic aspartates or the catalytic water molecule. The
carboxylic acid of compound 1 engages in one direct hydrogen
bond with BACE-1 to the backbone NH of Thr293 and also engages
in a water-mediated interaction with the backbone NH of Asn294.
Furthermore, the carboxylic acid also engages in an intramolecular
hydrogen bond with the amine at the 1-position of the dihydroiso-
quinoline. The phenyl ring points towards the S3 pocket and could
As part of a new high throughput screen/high throughput crys-
tallization campaign (HTS/HTX), we sought to develop an assay for
compound binding that would be more sensitive than our fluores-
cence polarization functional screen and be better suited to a high-
concentration screening approach. We reasoned that an Alpha-
Screen format with an avidity component, utilizing a probe that
bound to the active site and did not displace the central water be-
tween the two aspartic acids, would provide the needed character-
istics.8j A satisfactory probe was developed from a member of the
hydantoin class of BACE-1 inhibitors, which, when functionalized
⇑
Corresponding author.
E-mail address: simeongbowers@gmail.com (S. Bowers).
0960-894X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2013.01.103

2182
S. Bowers et al. / Bioorg. Med. Chem. Lett. 23 (2013) 2181–2186
Asn294
3.1Å
Thr293
2.5Å
2.7Å
Asp289
2.5Å
2.9Å
2.7Å
2.8Å
Figure 4. The small hydrophobic pocket at the 6-position of the dihydroisoquin-
oline ring. The center of the pocket is 3.1 Å from the ring.
3.5Å
Asp93
OH
O
SMe
N
OH
HN
a
b
c
O
N
R
R
R
R
Figure 2. Crystal structure of compound
1 in green bound to BACE-1 (1.9 Å
4 R = Me
5 R = Cl
2
6 R = Me
7 R = Cl
3
resolution). The catalytic aspartates are shown behind the inhibitor and do not
directly interact with it. The catalytic water molecule is also shown. The hydrogen
bond network of the inhibitor’s carboxylate group, two backbone NH groups, and a
water molecule is shown with dotted lines along with the inhibitor’s intramolecular
hydrogen bond. The PDB deposition code is 4I11. For experimental conditions see
Ref. 10.
Scheme 1. Reagents and conditions: (a) MeMgBr, THF, À78 °C, 67%; (b) MeSCN,
H2SO4, rt, 54%; (c) L-Phe, DMSO, 120 °C, 30%.
Table 1
SAR of the 6-position
possibly provide an opportunity to improve the potency of the
inhibitors.
a
Compound design was guided by molecular dynamics simula-
tion. The Molecular Mechanics-Possion Boltzmann Surface Area
(MM-PBSA) method was used to estimate binding energies for
new analogs.11 Binding energies were calculated for compounds
that were synthesized early in the project, and this yielded a corre-
lation with in vitro potency (Fig. 3). Subsequently, the regression
could be used to estimate the potency of designed ligands; those
that had more favorable calculated binding energies were priori-
tized for synthesis.
Compounds
R
BACE-1 IC50 (lM)
1b
6
7
H
Me
Cl
27.2
5.41
2.84
a
See Ref. 9.
Compound is racemic.
b
Table 2
SAR of heterocyclic carboxylic acid replacements
Molecular modeling suggested that a small pocket in the region
of the 6-position of the dihydroisoquinoline ring could accommo-
date a small hydrophobic substituent such as methyl or chlorine
HN
R1
N
R2
R1
R2
H
BACE-1 IC50
27.2
(lM)
a
Compounds
1
COOH
H
N
8
H
638.6
184.9
N
N
N
N
N
O
9
Cl
N
O
10
11
Cl
Cl
Cl
339.7
137.1
>250
N
N
N
N
N
12
a
See Ref. 9.
Figure 3. Correlation between calculated binding energy and potency.

S. Bowers et al. / Bioorg. Med. Chem. Lett. 23 (2013) 2181–2186
2183
H
N
a
b
c
HN
O
H
N
N
NH2
NOH
O
BocHN
O
N
BocHN
BocHN
CN
N
O
Cl
13
14
15
16
Scheme 2. Reagents and conditions: (a) NH2OH (aq), EtOH, rt, 86%; (b) CDI, THF, reflux, 81%; (c) (i) TFA, CH2Cl2, rt, (ii) 5, DMSO, 100 °C, 15%.
(Fig. 4). These analogs were readily prepared by the addition of
methyl magnesium bromide to an appropriate aryl propanone 2.
Subsequent treatment with methyl thiocyanate and sulfuric acid
cles were prepared that could present hydrogen bond acceptors in
the same manner as the carboxylic acid. The results of these initial
efforts are documented in Table 2.
gave the key intermediates 4 and 5, which were heated with
phenylalanine to give compounds 6 and 7 (Scheme 1).
The BACE-1 alpha assay results for these analogs are shown in
Table 1. Gratifyingly, both the 6-methyl analog (6) and 6-chloro
analog (7) possessed improved potency against BACE-1 compared
to the screening hit 1.
Although the addition of substituents to the 6-position of the
dihydroisoquinoline ring provided significant gains in potency,
the major challenge remained the replacement of the carboxylic
acid to increase the chances of CNS penetration.12 As mentioned
previously, this moiety interacts with the backbone NH of Thr293
and, via a water molecule, the backbone NH of Asn294. Further-
more, it also engages in an intramolecular hydrogen bond with
the amine at the 1-position of the dihydroisoquinoline. Successful
replacement of the carboxylic acid would likely require these key
interactions. With this in mind, a series of acid replacements was
designed. Tetrazoles are well known as carboxylic acid isosteres12b
and were the first replacements to be prepared. Additionally, a
small number of both five-membered and six-membered heterocy-
L
-
To our disappointment, this initial search for a carboxylic acid
replacement was not successful and the most potent analog,
pyrimidine 11 was fivefold less potent than the carboxylic acid 1.
At this juncture, we began to reconsider the key water-mediated
interaction between the carboxylic acid and the backbone NH of
Asn294. As an alternative to interacting with the water molecule,
it could be possible to design an acid replacement that would dis-
place the water molecule and allow the inhibitor to interact di-
rectly with the protein. To explore this idea,
a series of
heterocycles bearing exocyclic hydrogen bond acceptors was pre-
pared. Five-membered heterocycles with exocyclic oxygen atoms
were prepared as shown in Scheme 2. Treatment of hydroxyami-
dine 14 with CDI13 gave the intermediate oxadiazolone 15, which
was deprotected and reacted with 5 to give the final compound 16.
H
N
O
H
N
HN
a
b
O
OMe
NH
BocHN
N
BocHN
N
N
Cl
22
27
28
O
a
b
O
H2N
H
N
N
H
NH2
O
BocHN
BocHN
BocHN
N
H
NH2
H
H2N
O
N
O
c
d
OH
O
O
HN
H
N
17
18
19
N
BocHN
CO2H
BocHN
N
O
N
N
Cl
H
29
30
31
N
HN
c
H
N
N
NH
Scheme 5. Reagents and conditions: (a) 2-aminobenzoic acid, MeOH, reflux, 62%;
(b) (i) TFA, CH2Cl2, rt, (ii) 5, DMSO, 100 °C, 19%; (c) 4-aminonicotinamide, EDCI,
HOBt, Et3N, CH2Cl2, 79%; (d) (i) NaOH (aq), EtOH, 60 °C, 96%, (ii) TFA, CH2Cl2, rt, (iii)
5, DMSO, 100 °C, 23%.
O
N
N
NH
Cl
20
21
Scheme 3. Reagents and conditions: (a) EDCI, HOBt, Et3N, CH2Cl2, rt, 75%; (b) NaOH,
H2O, reflux, 51%; (c) (i) TFA, CH2Cl2, rt, (ii) 5, DMSO, 120 °C, 10%.
H
N
O
O
HN
H
N
H
N
b
a
O
O
O
N
a
BocHN
BocHN
BocHN
b
N
Cl
OEt
ONa
NH2
NH
OMe
NH
22
N
N
BocHN
BocHN
Cl
BocHN
Cl
N
25
32
33
13
23
24
H
N
O
H
N
HN
H
N
H
N
d
c
H
N
O
c
d
O
O
HN
N
BocHN
BocHN
CN
N
N
N
N
N
I
Cl
25
34
35
Cl
25
26
Scheme 6. Reagents and conditions: (a) NCS, AcOH, rt, 90%; (b) 5, DMSO, 100 °C,
12%; (c) NIS, CHCl3, rt, 67%; (d) (i) Pd(PPh3)4, Zn(CN)2, DMF, 110 °C, microwave,
40 min, 56%, (ii) TFA, CH2Cl2, rt, (iii) 5, DMSO, 100 °C, 15%.
Scheme 4. Reagents and conditions: (a) NaOMe, MeOH, 50 °C; (b) NH4Cl, MeOH,
50 °C; (c) EtOH, H2O, 91%, (3-steps); (d) (i) TFA, CH2Cl2, rt, (ii) 5, DMSO, 100 °C, 18%.

2184
S. Bowers et al. / Bioorg. Med. Chem. Lett. 23 (2013) 2181–2186
Triazolones were prepared by cyclization of semicarbazide 19,14
which was elaborated to the final analog using standard methods
(Scheme 3).
Table 3
SAR of acid replacements
A number of six-membered and bicyclic heterocycles were also
prepared using methods shown in Schemes 4 and 5. Pyrimidinones
were prepared via cyclo-condensation of amidine 23 with sodium
2-ethoxycarbonyl ethanolate (24)15 to provide compound 26.
Quinazolinone 28 was readily prepared by condensation of imi-
date 22 with 2-aminobenzoic acid, while pyridopyrimidone 31 was
prepared by cyclization of amide 30 (Scheme 5).
Further modification of pyrimidone 25 was achieved by haloge-
nation to give intermediates 32 and 34 (Scheme 6). Aryl iodide 34
was converted into nitrile 35 via palladium catalyzed cross
coupling.
In an attempt to remove the protic nature of these heterocyclic
acid replacements, the exocyclic oxygen was incorporated into a
bicyclic ring system. Sonogashira coupling of ethynyltrimethylsi-
lane with iodide 34, followed by subsequent removal of the TMS
group and heat-mediated cyclization, gave furopyrimidine analog
39 (Scheme 7).
For the carboxylic acid replacement to be successful, it must not
only provide equal or greater potency than the acid but must also
allow the inhibitor to reach its target. This requires the compound
to be highly permeable and possess a low P-gp efflux in order to
cross the blood brain barrier.17 The SAR of the carboxylic acid
replacements, along with their permeability and P-gp efflux ratios
derived from MDR-MDCK cells, are shown in Table 3. Gratifyingly,
oxadiazolone 16 proved to be an effective replacement for the car-
HN
R
N
Cl
a
b
#
R
BACE-1 IC50
pKa
Papp nm/s
%recoveryc
P-gp
l
M
effluxc
7
COOH
2.84
2.1
4.0
nd
nd
H
N
26 nm/s
75%
16
0.75
47
O
N
N
O
H
N
23 nm/s
72%
21d
26
3.37
2.71
7.1
6.3
60
O
O
NH
H
N
231 nm/s
60%
2.6
N
N
H
N
O
O
28
7.23
7.2
Low recovery
—
H
N
196 nm/s
51%
31
33
0.17
0.56
nd
nd
1.9
5.5
N
N
N
H
N
boxylic acid (IC50 = 0.75
potency compared to acid 7. Similarly, triazolone 21 also had good
potency (IC50 = 3.37 M), albeit slightly attenuated compared to
lM) providing a fourfold improvement in
O
213 nm/s
65%
Cl
O
l
H
N
16. The crystal structure of BACE-1 in complex with compound
16 was determined to 1.8 Å resolution as shown in Figure 5. This
structure revealed that the exocyclic oxygen effectively displaces
the water molecule found in the structure of 1 and interacts di-
rectly with the backbone NH of Asn294 at 3.1 Å distance. Further-
more, the contact with Thr293 is maintained.
Unfortunately, oxadiazolone 16 had a low pKa (4.0) indicating
that the compound would exist as a zwitterion at physiological
pH and may have a lower chance of CNS penetration compared
to a basic molecule.12a Compound 16 had low passive permeability
(26 nm/s) and a high P-gp efflux ratio (47), thereby limiting its po-
tential as an acid replacement for a CNS target. Although triazolone
21 possesses a more reasonable pKa (7.1), it still suffers from low
permeability (23 nm/s) and a very high P-gp efflux ratio (60),
which also limits its usefulness as an acid replacement.
73 nm/s
75%
35
39
40
0.47
73.7
0.59
nd
nd
nd
39
nd
nd
N
N
N
CN
O
N
nd
nd
H
N
O
I
a
b
c
See Ref. 9.
Determined by capillary electrophoresis
See Ref. 16.
Compound is racemic; nd; not determined.
d
H
H
N
H
N
b
N
O
a
O
O
I
BocHN
BocHN
BocHN
N
N
N
TMS
34
36
37
c
d
N
O
HN
N
O
BocHN
N
N
N
Cl
38
39
Scheme 7. Reagents and conditions: (a) ethynyltrimethylsilane, PdCl2(PPh3)2, CuI,
Et3N, microwave, 120 °C, 20 min; (b) K2CO3, MeOH; (c) MeOH, 120 °C, 45% (3-
steps); (d) (i) TFA, CH2Cl2, rt, (ii) 5, DMSO, 100 °C, 12%.
Figure 5. Crystal structure of compound 16 in green bound to BACE-1 (1.8 Å
resolution). The PDB deposition code is 4HZT. For experimental conditions see Ref.
10.

S. Bowers et al. / Bioorg. Med. Chem. Lett. 23 (2013) 2181–2186
2185
For the six-membered heterocyclic acid replacement, pyrimidi-
none 26 had similar potency (BACE-1 IC50 = 2.71 M), compared to
Molecular modeling also indicated that larger, bicyclic rings
could be accommodated and, indeed, quinazoline 28 proved that
l
acid 7, but interestingly, it also possessed a significantly lower P-gp
efflux ratio and higher passive permeability compared to 16 and
21. Molecular modeling suggested that there would be sufficient
space on the 5-position of the pyrimidine ring to place a substitu-
ent. Indeed, chlorine, iodine and cyano pyrimidinones 33, 40 and
35 all had improved potency compared to 26, with the cyanopyr-
imidone 35 being sixfold more potent than the carboxylic acid.
The crystal structure of nitrile 35 with BACE-1 was determined
to 1.8 Å resolution (Fig. 6). This structure revealed that the inhibi-
tor’s pyrimidinone ring interacts with Thr293 and Asn294 in a sim-
ilar manner as oxadiazolone 16, but its nitrile group forms a new
interaction with the side-chain of Ser-386 at 3.0 Å distance.
this was the case (IC50 = 7.2
lM). Since 28 had poor permeability
and low solubility (20
l
M),16 it was envisioned that the addition
of heteroatoms to the benzo ring would improve the pharmacoki-
netic properties of the inhibitor. Molecular modeling indicated that
nitrogen, placed at the 6-position of the ring, could form further
interactions with BACE-1. This was confirmed with the synthesis
of pyrido[4,3-d]pyrimidin-4(3H)-one 31. Compound 31 gained sig-
nificant potency compared to 35 and, furthermore, had improved
solubility (57 lM), high permeability (196 nm/s) and low P-gp ef-
flux (efflux ratio = 1.9). The crystal structure of 31 with BACE-1 was
also solved (Fig. 7), and, as observed in the structure of pyrimidine
35, the interactions between the acid replacement and Thr293 and
Asn294 are maintained. Additionally, the nitrogen in the 6-position
of the pyridine ring forms a water-mediated hydrogen bond with
the side-chain of Ser-186.
Finally, we attempted to remove the protic nature of the hetero-
cyclic acid replacements with furanopyrimidine 39. To our disap-
pointment, this endeavor was not successful, and compound 39
lost an order of magnitude in potency compared to acid 7.
In conclusion, guided by structure-based design, we were able
to rapidly modify the initial screening hit 1 to gain significant
improvements in potency against BACE-1. Key to the continued
development of this series of analogs was the replacement of the
carboxylic acid. Molecular modeling aided in the design of hetero-
cycles bearing exocyclic hydrogen bond acceptors that effectively
replaced a water mediated hydrogen bond. Compound 31 proved
to be the most promising analog in this series with good potency,
low P-gp efflux and high permeability, all hallmarks of a CNS pene-
trant small molecule. The continued development of this series of
compounds will be reported in due course.
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.
3. Sinha, S.; Liberburg, I. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 11049.
4. (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.
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.
Figure 6. Crystal structure of compound 35 in green bound to BACE-1 (1.8 Å
resolution). The PDB deposition code is 4I0Z. For experimental conditions see Ref.
10.
5. Luo, Y.; Bolon, B.; Damore, M. A.; Fitzpatrick, D.; Liu, H.; Zhang, J.; Yan, Q.;
Vassar, R.; Citron, M. Neurobiol. Dis. 2003, 14, 81.
6. Schmidt, B.; Baumann, S.; Braun, H. A.; Larbig, G. Curr. Top. Med. Chem. 2006, 6,
377.
7. Probst, G. D.; Xu, Y. Z. Expert Opin. Ther. Pat. 2012, 5, 511.
8. (a) Maillard, M. C.; Hom, R. K.; Benson, T. E.; Moon, J. B.; Mamo, S.; Bienkowski,
M.; Tomasselli, A. G.; Woods, D. D.; Prince, D. B.; Paddock, D. J.; Emmons, T. L.;
Tucker, J. A.; Dappen, M. S.; Brogley, L.; Thorsett, E. D.; Jewett, N.; Sinha, S.;
John, V. J. Med. Chem. 2007, 50, 776; (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; (c) Truong, A.
P.; Toth, G.; Probst, G. D.; Sealy, J. M.; Bowers, S.; Wone, D. W. G.; Dressen, D.;
Hom, R. K.; Konradi, A. W.; Sham, H. L.; Wu, J.; Peterson, B. T.; Ruslim, L.; Bova,
M. P.; Kholodenko, D.; Motter, Ruth N.; Bard, F.; Santiago, P.; Ni, H.; Chian, D.;
Soriano, F.; Cole, T.; Brigham, E. F.; Wong, K.; Zmolek, W.; Goldbach, E.; Samant,
B.; Chen, L.; Zhang, H.; Nakamura, D. F.; Quinn, K. P.; Yednock, T. A.; Sauer, J.-M.
Bioorg. Med. Chem. Lett. 2010, 20, 6231; (d) Probst, G. D.; Bowers, S.; Sealy, J. M.;
Stupi, B.; Dressen, D.; Jagodzinska, B. M.; Aquino, J.; Gailunas, A.; Truong, A. P.;
Tso, L.; Xu, Y.-Z.; Hom, R. K.; John, V.; Tung, J. S.; Pleiss, M. A.; Tucker, J. A.;
Konradi, A. W.; Sham, H. L.; Jagodzinski, J.; Toth, G.; Brecht, E.; Yao, N.; Pan, H.;
Lin, M.; Artis, D. R.; Ruslim, L.; Bova, M. P.; Sinha, S.; Yednock, T. A.; Gauby, S.;
Zmolek, W.; Quinn, K. P.; Sauer, J.-M. Bioorg. Med. Chem. Lett. 2010, 20, 6034; (e)
Truong, A. P.; Probst, G. D.; Aquino, J.; Fang, L.; Brogley, L.; Sealy, J. M.; Hom, R.
K.; Tucker, J. A.; John, V.; Tung, J. S.; Pleiss, M. A.; Konradi, A. W.; Sham, H. L.;
Figure 7. Crystal structure of compound 31 in green bound to BACE-1 (2.1 Å
resolution). The PDB deposition code is 4I10. For experimental conditions see Ref.
10.

2186
S. Bowers et al. / Bioorg. Med. Chem. Lett. 23 (2013) 2181–2186
Compounds at 10 mM DMSO solution were added to
Dappen, M. S.; Toth, G.; Yao, N.; Brecht, E.; Pan, H.; Artis, D. R.; Ruslim, L.; Bova,
a
soaking solution
M. P.; Sinha, S.; Yednock, T. A.; Zmolek, W.; Quinn, K. P.; Sauer, J.-M. Bioorg.
Med. Chem. Lett. 2010, 20, 4789; (f) Mandal, M.; Zhu, Z.; Cumming, J. N.; Liu, X.;
Strickland, C.; Mazzola, R. D.; Caldwell, J. P.; Leach, P.; Grzelak, M.; Hyde, L.;
Zhang, Q.; Terracina, G.; Zhang, L.; Chen, X.; Kuvelkar, R.; Kennedy, M. E.;
Favreau, L.; Cox, K.; Orth, P.; Buevich, A.; Voigt, J.; Wang, H.; Kazakevich, I.;
McKittrick, B. A.; Greenlee, W.; Parker, E. M.; Stamford, A. W. J. Med. Chem.
2012, 55, 9331; (g) Malamas, M. S.; Erdei, J.; Gunawan, I.; Turner, J.; Hu, Y.;
Wagner, E.; Fan, K.; Chopra, R.; Olland, A.; Bard, J.; Jacobsen, S.; Magolda, R. L.;
Pangalos, M.; Robichaud, A. J. J. Med. Chem. 2010, 53, 1146; (h) Zhu, Z.; Sun, Z.-
Y.; Ye, Y.; Voigt, J.; Strickland, C.; Smith, E. M.; Cumming, J.; Wang, L.; Wong, J.;
Wang, Y.-S.; Wyss, D. F.; Chen, X.; Kuvelkar, R.; Kennedy, M. E.; Favreau, L.;
Parker, E.; McKittrick, B. A.; Stamford, A.; Czarniecki, M.; Greenlee, W.; Hunter,
J. C. J. Med. Chem. 2010, 53, 951; (i) May, P. C.; Dean, R. A.; Lowe, S. L.; Martenyi,
F.; Sheehan, S. M.; Boggs, L. N.; Monk, S. A.; Mathes, B. M.; Mergott, D. J.;
Watson, B. M.; Stout, S. L.; Timm, D. E.; LaBell, E. S.; Gonzales, C. R.; Nakano, M.;
Jhee, S. S.; Yen, M.; Ereshefsky, L.; Lindstrom, T. D.; Calligaro, D. O.; Cocke, P. J.;
Hall, D. G.; Friedrich, S.; Citron, M.; Audia, J. E. J. Neurosci. 2011, 31, 6507; (j)
Brodney, M. A.; Barreiro, G.; Ogilvie, K.; Hajos-Korcsok, E.; Murray, J.; Vajdos, F.;
Ambroise, C.; Christoffersen, C.; Fisher, K.; Lanyon, L.; Liu, J. H.; Nolan, C. E.;
Withka, J. M.; Borzilleri, K. A.; Efremov, I.; Oborski, C. E.; Varghese, A.; O’Neill, B.
R. J. Med. Chem. 2012, 55, 9224.
consisting of 100 mM sodium acetate pH 6.6, 10% PEG 8000, 5% ethylene glycol,
5 mM zinc chloride and 10%DMSO. Apo crystals were transferred to soaking
solution and incubated overnight at room temperature. 80 Individual
compounds were soaked with apo BACE-1 crystals in one batch. 80 Soaked
crystals in one batch were frozen directly into liquid nitrogen, data sets were
collected and processed automatically by the Rigaku X-ray Homelab Highflex
system with an ACTOR auto sampler. Electron density was analyzed and
interpreted manually.
11. Kollman, P. A.; Massova, I.; Reyes, C.; Kuhn, B.; Huo, S.; Chong, L.; Lee, M.; Lee,
T.; Duan, Y.; Wang, W.; Donini, O.; Cieplak, P.; Srinivasan, J.; Case, D. A.;
Cheatham, T. E., 3rd Acc. Chem. Res. 2000, 33, 889.
12. (a) Gleeson, P. M. J. Med. Chem. 2008, 51, 817; (b) Meanwell, N. A. J. Med. Chem.
2011, 54, 2529.
13. Mangette, J. E.; Johnson, M. R.; Le, V.-D.; Shenoy, R. A.; Roark, H.; Stier, M.;
Belliotti, T.; Capiris, T.; Guzzo, P. R. Tetrahedron 2009, 65, 9536.
14. Kandeel, K. A.; Youssef, A. S. A.; Abou-Elmagd, W. S. I.; Hashem, A. I. J.
Heterocycl. Chem. 2006, 43, 957.
15. Mylari, B. L.; Oates, P. J.; Beebe, D. A.; Brackett, N. S.; Coutcher, J. B.; Dina, M. S.;
Zembrowski, W. J. J. Med. Chem. 2001, 44, 2695.
16. For details on in vitro pharmacokinetic assays, see: Truong, A. P.; Aubele, D. A.;
Probst, G. D.; Neitzel, M. L.; Semko, C. M.; Bowers, S.; Dressen, D.; Hom, R. K.;
Konradi, A. W.; Sham, H. L.; Garofalo, A. W.; Keim, P. S.; Wu, J.; Dappen, M. S.;
Wong, K.; Goldbach, E.; Quinn, K. P.; Sauer, J.-M.; Brigham, E. F.; Wallace, W.;
Nguyen, L.; Hemphill, S. S.; Bova, M. P.; Basi, G. Bioorg. Med. Chem. Lett. 2009,
19, 4920.
9. Ren, Z.; Tam, D.; Xu, Y.-Z.; Wone, D.; Yuan, S.; Jobling, M.; Cheung, H.; Chen, X.;
Rudolph, D.; Sham, H.; Artis, D. R.; Bova, M. P. J. Biomol. Screen. submitted for
publication.
10. Crystallography conditions: BACE-1 protein from construct pQE708e was
concentrated to 10 mg/ml in 100 mM borate pH 8.5. Apo crystals were
17. Wager, T. T.; Chandrasekaran, R. Y.; Hou, X.; Troutman, M. D.; Verhoest, P. R.;
Villalobos, A.; Will, Y. ACS Chem. Neurosci. 2010, 1, 420.
grown at 4 °C in 1
lL sitting drops with a 1:1 (v/v) ratio of protein to reservoir,
a solution of 9% PEG 8000, 100 mM sodium acetate pH 5.3 and 10 mM ZnCl2.