Synthesis and in vivo evaluation of cyclic diaminopropane BACE-1 inhibitors

Article abstract of DOI:10.1016/j.bmcl.2011.06.116

The synthesis, evaluation, and structure-activity relationships of a set of related constrained diaminopropane inhibitors of BACE-1 are described. The full in vivo profile of an optimized inhibitor in both normal and P-gp deficient mice is compared with data generated in normal rats.

Full text of DOI:10.1016/j.bmcl.2011.06.116

Bioorganic & Medicinal Chemistry Letters 21 (2011) 6909–6915
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and in vivo evaluation of cyclic diaminopropane BACE-1 inhibitors
Lorin A. Thompson a, , Jianliang Shi a, Carl P. Decicco b, Andrew J. Tebben b, Richard E. Olson a,
⇑
Kenneth M. Boy a, Jason M. Guernon a, Andrew C. Good c, Ann Liauw a, Changsheng Zheng d,
,
Robert A. Copeland e, , Andrew P. Combs d, , George L. Trainor b, Daniel M. Camac b, Jodi K. Muckelbauer b,
Kimberley A. Lentz a, James E. Grace a, Catherine R. Burton a, Jeremy H. Toyn a, Donna M. Barten a,
Jovita Marcinkeviciene f, , Jere E. Meredith a, Charles F. Albright a, John E. Macor a
a Bristol-Myers Squibb Research and Development, 5 Research Parkway, Wallingford, CT 06492, USA
b Bristol-Myers Squibb Research and Development, Route 206 and Province Line Road, Princeton, NJ 08543, USA
c Genzyme Pharmaceuticals, 153 Second Avenue, Waltham, MA 02451, USA
d Incyte Pharmaceuticals, Rte. 141 & Henry Clay Road, Wilmington, DE 19880, USA
e Epizyme, Inc., 840 Memorial Drive, Cambridge, MA 02139, USA
f Novartis Instutite for Biomedical Research, 9 Cambridge Center, Cambridge, MA 02142, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Available online 2 July 2011
The synthesis, evaluation, and structure–activity relationships of a set of related constrained diaminopro-
pane inhibitors of BACE-1 are described. The full in vivo profile of an optimized inhibitor in both normal
and P-gp deficient mice is compared with data generated in normal rats.
Ó 2011 Elsevier Ltd. All rights reserved.
Keywords:
BACE-1 inhibitors
Alzheimer’s disease
BACE-1 crystallography
P-gp knockout mice
COOH
COOH
Alzheimer’s disease (AD) is the leading cause of dementia, and
is currently believed to affect 4 million Americans and up to 30
million people worldwide.1 In addition to the devastating burden
of the disease on patients and their families, estimates of the finan-
cial cost of the disease in the US alone range up to $100 billion
annually.2 Current treatments include acetylcholinesterase inhibi-
tors and the NMDA blocker memantine, however these therapies
treat disease symptoms and do not significantly affect the underly-
ing progression of the disease. While the full etiology of the disease
is still a matter of intense research, current data suggest that olig-
omers of the 42-amino acid form of the beta amyloid peptide (Abe-
ta1-42) are central to the disease process.3 The amyloid peptides
are excised from the amyloid precursor protein (APP) by the
sequential action of b-secretase4–8 (typically referred to as BACE-
1) to form the N-terminus, followed by a series of cleavages by
CONH2
O
OH
O
H
H
N
H
H
N
N
N
COOH
H2N
N
N
H
H
Me
O
O
O
O
1, OM99-2
BACE1 IC50 = 10 nM
P2
P1'
O
OH
H
H
H2N
N
N
N
H
O
O
P1
2
P3
BACE1 IC50 = 88 nM
c
-secretase to form the C-terminus.9 Inhibitors of both of these
enzymes are being actively investigated for the treatment of AD.
This letter details the discovery and SAR of a series of cyclic diami-
nopropane BACE-1 inhibitors. The disclosure of the hydroxyethyl-
ene isostere-based OM99-2 (Fig. 1) by Ghosh and colleagues10 as
a 10 nM inhibitor of BACE-1, combined with the publication of
O
OH
H
H
N
H2N
N
N
O
O
3, Constrained P2-P3 group
BACE1 IC50 = 141 nM
⇑
Corresponding author.
E-mail address: lorin.thompson@bms.com (L.A. Thompson).
Present address.
Figure 1. OM99-2 and peptidic BACE-1 inhibitors.
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.06.116

6910
L. A. Thompson et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6909–6915
ligand efficiency. This strategy was immediately successful, with
the acetylated lactam diaminopropanes 5 and 6 (Fig. 2) giving
2–5 nM inhibition in the enzyme assay with nearly equal potency
in cellular assays for blocking Abeta production (see Ref. 16 for as-
say details and Ref. 17 for synthesis). Compound 5 was successfully
tritiated and served as the ligand for the development of an effi-
cient radioligand displacement assay for BACE-1 activity.18 Having
achieved nanomolar cellular potency, compound 6 was profiled for
brain Abeta 1–40 reduction in the rat. The compound was dosed
intraperitoneally (ip) to maximize exposure, and the compound
caused a dose-dependent inhibition in plasma Ab production, with
an ED50 of 30 mpk (Fig. 3). Drug levels at this dose were 3700 nM,
and the fraction unbound in rat plasma was 1.1%, giving 40 nM free
OH
H
H
N
N
N
O
O
O
F
4, Elan and Pfizer
F
O
OH
H
N
H
N
O
N
exposure, or approximately 10-fold coverage of the cellular EC50
.
N
O
H
R
O
There was no effect on brain Ab up to 100 mpk, and brain exposure
at that dose was only 113 nM total drug, giving a brain/plasma ra-
tio of <0.05, indicating that there was essentially no drug in the
brain.
R
5: R = H, BMS-599240 6: R = F
The lack of brain penetration for compound 6 was not surpris-
ing, given it’s relatively high polar surface area (PSA) of 114 Å2
and high molecular weight (mw = 679 Da). Although compound 6
had good overall permeability, there was the possibility of a mild
P-gp effect as demonstrated by faster basolateral to apical passage
across Caco cell monolayers (A–B = 75 nm/s, B–A = 114 nm/s, ra-
tio = 1.5). In order to lower the molecular weight of the inhibitors,
we planned to introduce an additional ring constraint into the mol-
ecule followed by truncation of the nonprime side. To support this
effort, we determined the crystal structure of compound 5 bound
to BACE-1 which is shown in Figure 4 in green.19 The compound
binds in a manner similar to that reported for related diaminopro-
panes,15 with the hydroxyl group engaging Asp 32 and the amino
group engaging Asp 228. Additionally, the added potency of the
P2 homophenylalanine residue was explained by an aryl edge
interaction with the sidechain of Arg 235. With this information
in hand, several constrained structural variants were modeled.
We were drawn to the possibility of creating a ring at the carbon
next to the prime-side amine of the isostere, creating a novel cyclic
constraint in the isostere itself.20 In this regard, the tetrahydroiso-
quinoline (THIQ) 14a modeled particularly well (shown in yellow
in Fig. 4). The relative position of the amine was unchanged, and
the extra size of the rings appeared well accommodated in the pep-
tide-binding cleft of the enzyme. The model predicted that the
BACE1 IC50 = 5 nM
Cell IC50 = 11 nM
BACE1 IC50 = 2 nM
Cell IC50 = 3 nM
Figure 2. Lactam diaminopropanes.
the crystal structure of BACE-1 complexed with OM99-2,11 gave a
solid starting point to begin a structure-based design effort. We be-
gan with the construction of truncated peptide amides related to
OM99-2. We quickly determined that most of the activity of the
parent compound could be maintained using P3–P20 pentapeptide
amides, including structure 2 (Fig. 1, BACE-1 IC50 = 88 nM) where
homophenylalanine was used as a preferred residue at P2. We then
set out to rigidify the scaffold hoping to gain activity that would
allow truncation elsewhere in the molecule. Focusing on the non-
prime side, we utilized the well-precedented formation of
c
-lactams, a strategy which we had used successfully in the design
of inhibitors of tumor necrosis factor-
a
converting enzyme.12 Ap-
plied to the BACE P2–P3 region, this strategy resulted in compound
3, a Freidinger-type lactam13 with similar activity to the peptide
lead (BACE-1 IC50 = 141 nM).
As we considered how to further reduce the peptide character
of compound 3, a patent application appeared describing less pep-
tidic BACE-1 inhibitors built on the cis-2-hydroxy-1,3-diaminopro-
pane aspartyl protease isostere (compound 4, Fig. 2).14,15 It was
immediately clear that we could exchange the hydroxyethylene
isostere for the diaminopropane core. The resulting compounds
would be less peptidic, lower in molecular weight, and have higher
Brain
Plasma and Brain Abeta 1-40
Compound 6 at 5 Hrs
Plasma
125
100
75
50
25
0
*p<0.05
**p<0.01
*
**
0
0.1
0.3
1
3
10
30
100
Dose (mg/kg, i.p.)
Figure 4. X-ray crystal structure of compound 5 (green) and modeled tetrahydro-
quinoline inhibitor (yellow). Electron density is shown as a mesh contoured around
the ligand. Flap residues are removed for clarity.
Figure 3. Plasma and brain Ab lowering with compound 6. For assay details, see
Ref. 16.

L. A. Thompson et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6909–6915
6911
compound would be more potent when the newly-formed stereo-
genic center was in the (R)-configuration. There also appeared to
be room off the THIQ 7-position where substituents could access
the space occupied by the methoxybenzyl group in inhibitors like
5. Upon examining the stereochemical configuration of this pro-
posed inhibitor, it was clear that the requisite stereochemistry
could be obtained from the result of an Evans0 diastereoselective
syn-aldol reaction, a reliable and versatile technology.
The required scaffold was constructed by reacting the Z-boron
enolate of the acyl oxazolidinone 8 with the aldehydes 9 (see
Ref. 21 for synthesis details) to produce the aldol adducts 10a–f
in good yield (Scheme 1). The chiral auxiliary was removed under
standard conditions, and in a step precendented by the synthesis of
cytoxazone,22 the resulting acid was subjected to Curtuis rear-
rangement forming the cyclic oxazolidinones 12a–f in modest
yield. We found it difficult to hydrolyze the oxazolidinone selec-
tively in the presence of the Boc group, so the BOC group was ex-
changed for a PMB in high yield, at which time saponification
provided the desired scaffolds 13a–f (Table 1) in good yield. Scaf-
folds 13b and 13f which are diastereomeric at the THIQ were pre-
pared using identical chemistry, but starting with racemic starting
materials 9 followed by a final separation by HPLC at the end of the
synthesis.
Final compounds were then prepared by simple amide coupling
of the amines 13a–f with the known lactam headgroup23 to pro-
vide, after acidic deprotection, compounds 14a–f (Scheme 2). We
were gratified to see potent activity at BACE-1 from the con-
strained isostere inhibitor 14a, with an IC50 = 54 nM in the enzyme
assay and only a twofold loss in activity in the cellular assay
(110 nM, Table 1). As predicted by the modeling, the diastereomer
with the THIQ attached to the chain in the (R)-configuration was
around 10-fold more potent than the (S)-diastereomer 14b. Chang-
ing the lactam P3 substituent from sec-butyl (14a) to isobutyl (i.e.,
2-methylpropyl, 14c) resulted in a modest increase in affinity. An
additional set of analogs was made starting with the phenol 14g.
While the phenol itself was poorly tolerated (IC50 = 870 nM), re-
lated ethers were considerably better, with the best activity in
the series coming from the methyl ether 14e, with a cellular
IC50 = 20 nM. The crystal structure of BACE-1 in complex with com-
pound 14c was obtained using similar conditions as for the analy-
sis with compound 5 (Fig. 5).25 The overall binding mode was
nearly identical to that predicted by the initial modeling (compare
Fig. 4 and 5), with only a small deviation in the position of the THIQ
nitrogen atom binding to aspartate 228.
Having identified the constrained THIQ scaffold, we investi-
gated its activity with other known P2–P3 groups (Table 2). These
compounds were prepared using appropriate carboxylic acids and
coupled to amine 13a in the manner shown in Scheme 2. Using the
N,N-dipropylisophthalate headgroup26 led to a loss in potency,
with both diastereomeric THIQs (15a and 15b) having micromolar
O
OH Boc
N
O
O
O
Boc
N
a
Xp
F
N
H
O
F
X
X
Bn
F
10a-f
9
8
F
Ph
O
OH PMB
N
O
O
OH Boc
N
O
Boc
N
OH
H
N
H2N
H
HN
AcHN
a, b
c
b
N
HO
N
F
F
X
F
X
F
X
O
X
11a-f
12a-f
F
F
F
13a-f
14a-f
F
OH PMB
N
H2N
Scheme 2. Reagents and conditions: (reported yields are for X = H) (a) lactam,
HATU, DIEA, rt 85%; (b) 4 M HCl dioxane/water, rt (90%).
F
X
d,e,f
13a-f
F
Scheme 1. Reagents and conditions: (reported yields are for X = H) (a) (n-Bu)2BOTf,
DIEA, THF, ꢀ78 °C, 78%; (b) LiOOH, THF/water, 0 °C to rt (c) DPPA, PhCH3 32% over 2
steps; (d) 50% TFA/DCM, rt; (e) PMB–CHO, NaBH(OAc)3, DCM, rt 80% over 2 steps; (f)
50% NaOH, EtOH, 60 °C, 87%.
Table 1
Tetrahydroisoquinoline lactams
Compound Substituent (X)
IC50 enzyne
(nM)
IC50 HEKswa
(nM)
14a
14b
14c
H
54
600
8
110
n.d.
10
H, (S) THIQ diastereomer
H (isobutyl group on
lactam)
14d
14e
14f
On-Pr
OMe
OMe, (S) THIQ
diastereomer
OH
17
4
140
140
20
n.d.
14 g
870
n.d.
Figure 5. X-ray crystal structures of compound 5 (green) and 14c (yellow). Electron
density is shown as a mesh contoured around the ligand. Flap residues are removed
for clarity.
a
Activity assayed in HEK 293 cells according to Ref. 24 Enzyme data are n = 2,
cellular data are n = 4.

6912
L. A. Thompson et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6909–6915
Table 2
Activity of selected biaryl succinyl amides
OH
H
N
H
N
P2-P3 cap
F
15a-e
F
Compound
P2–P3 cap
IC50 enzyne (nM)
IC50 HEKswa (nM)
15a
1100
1600
n.d.
N
O
O
O
N
O
15b
n.d.
n.d.
Diastereomeric THIQ
N
O
N
N
15c
15d
1200
Bu
O
Ph
O
1000
1600
1400
720
n-Bu
N
N
O
O
O
n-Bu
15e
a
Activity assayed in HEK 293 cells according to Ref. 16. Enzyme data are n = 2, cellular data are n = 4.
potency. We recently reported27 on the use of the azaindolyl urea
group (compound 15c), however this combination was also only
modestly potent with the THIQ scaffold. In other work recently re-
OTBS
OTBS
Boc
Boc
CBZHN
N
CBZHN
N
b
a
F
F
ported, we prepared a series of truncated lactam scaffolds intended
to reduce the molecular weight of the P2–P3 headgroup.28 As
shown for compounds 15d and 15e, however, the THIQ scaffold re-
F
OH
OMs
16
17
mained modestly potent (1–2
l
M, see Table 2). At this point, we
F
F
OTBS
OTBS
Boc
Boc
N
decided to vary the type of heterocycle used in the construction
of the constrained isostere. This work is detailed in the accompany-
ing article,28 where SAR and molecular modeling on a number of
related scaffolds led to the discovery that a similar scaffold based
on a substituted pyrrolidine provided potent compounds with
H2N
CBZHN
N
c, d
F
O S
O
S
R
R
18a-c
the
c
-lactam headgroup. Accordingly, we profiled analogs in these
F
F
18,19a; R = propyl
related series with additional headgroups and pyrrolidine substitu-
tion. In particular, we were interested in making compounds
where the pyrrolidine 4-substituent was a sulfone, as the S@O dou-
ble bond was pointed toward Tyr 198 in a manner that suggested a
hydrogen bond could be made.
18,19b; R = phenyl
18,19c; R = 4-fluorobenzyl
Scheme 3. Reagents and conditions: (yields reported are for R = n-pr): (a) DEAD,
PPh3,MsOH, THF/toluene, 60 °C, 94%; (b) R-SH, NaH, DMF, rt, 67%; (c) Oxone, MeOH/
water, 64% + 28% sulfoxide; (d) H2 (50 psi), 10%Pd/C, MeOH, 99%.
We started with the pyrrolidine intermediate 16, which was
synthesized by aldol chemistry in a manner analogous to the
chemistry reported in Scheme 1 (Scheme 3).28 Our existing SAR
suggested that pyrrolidine substituents at the 4-position with the
existing (R)-configuration were likely to be more potent, so we em-
ployed a double inversion protocol to return the original stereo-
chemistry. The alcohol 16 was converted to the (S)-mesylate
under Mitsunobu conditions, and then simple displacement with
thiols provided intermediates 18a–c. The thiols were then oxidized
to the sulfone using oxone follwed by deprotection to the free
amines 19a–c by hydrogenolysis of the CBz group. The diastereo-
meric propyl sulfone used in the synthesis of compound 20a was

L. A. Thompson et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6909–6915
6913
Table 3
Pyrrolidine sulfide and sulfones
OH
OH
H
N
H
N
H
H
N
N
P2-P3 cap
F
P2-P3 cap
R1
R1
F
21a-f
20a
F
F
Compound
P2–P3 cap
R1 group
IC50 HEKsw (nM)
Ph
O
AcHN
AcHN
AcHN
21a
S-propyl
15
N
O
Ph
O
21b
20a
SO2-propyl
5
N
N
O
Ph
O
SO2-propyl (S)-sulfone diastereomer
45
O
N
O
21ca
SO2-propyl
68
12
N
O
O
Ph
O
AcHN
AcHN
21d
SO2-phenyl
N
N
O
Ph
O
21e
21f
SO2-4-fluorobenzyl
SO2-propyl
3
O
O
>330
N
O
Activity assayed in HEK 293 cells according to Ref. 24 n = 4 in all cases except for compound 21e where n = 2.
a
For synthesis of this P2–P3 group, see Ref. 29.
synthesized by simple mesylation of 16 followed by identical thiol
displacement and oxidation. The amines 19a–c were then con-
verted to final products using a variety of P2–3 caps using the same
chemistry as reported earlier (Scheme 2) to provide compounds
21a–f and 20a (Table 3).
The cellular activity of the sulfones and related compounds is
shown in table 3. The parent propyl thiol 21a (synthesized by
deptotecting and coupling 18a) was a potent inhibitor with a cel-
lular IC50 = 15 nM. In accordance with the hypothesis of targeting
Tyr 198, oxidation to the sulfone 21b provided a modest increase
in potency to 5 nM. As expected, the diastereomeric (S)-sulfone
20a was approximately 10-fold less potent. The combination of
an optimized isophthalate headgroup with the propyl sulfone
(21c, IC50 = 68 nM) was not as potent as the lactam sulfone (21b).
Additional sulfones prepared from thiophenol or 4-fluorobenzylth-
iol (21d,e) had similar potency to the propyl compound, but had
larger molecular weights and correspondingly lower ligand effi-
ciencies and were not pursued further. The poor activity of the
truncated lactam 21f was surprising given the good activity of this
P2–P3 group in our related pyrrolidine ether series.28
Due to its cellular potency (5 nM), compound 21b was chosen
as a representative of this new series to evaluate further. This
compound was evaluated in the bi-directional Caco-2 model
and showed apical to basolateral permeability of <15 nm/s versus

6914
L. A. Thompson et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6909–6915
basolateral to apical permeability of 58 nm/s (efflux ratio = >4).
The data predicted the compound is a P-gp substrate. Compound
21b is only modestly protein bound, with a 24% unbound fraction
in mouse plasma and a 30% unbound fraction in rat plasma. In or-
der to deconvolute the potential P-gp effect from passive penetra-
tion of 21b into the brain we profiled the compound both in
normal mice and also in P-gp (MDR1aꢀ/ꢀ) deficient mice, and
this data has already been disclosed.16
We also fully profiled the compound in rats to ensure that both
rodent species gave similar data. Figure 6 shows the peripheral
Abeta1-40 lowering effect of compound 21b dosed s.c. (70% PEG-
400/30% water vehicle) in both wt mice (1, 3, 10, 30, and
100 mpk) and rats (3, 10, and 100 mpk) as a function of the free
plasma drug concentration. In general, the data in the rat were con-
sistent with those from the wt mouse.
brain barrier penetration in the absence of P-gp, and P-gp was
the main factor limiting efficacy of this compound.
In summary, we have developed a series of cyclic diaminopro-
pane isosteres that can be combined with a number of different
P2–P3 headgroups to provide potent BACE-1 inhibitors. Combina-
tions of the pyrrolidine-based isostere substituted with a 4-sulfone
group provided compounds that were highly potent in both en-
zyme and cellular assays and were effective at lowering plasma
Abeta in normal rodents and both plasma and brain Abeta in
P-gp deficient mice. These data demonstrate that in the absence of
P-gp, the scaffold shows good intrinsic brain penetrance. Continued
work will be necessary to remove the P-gp liability and create ana-
logs that are effective in lowering brain Ab in normal animals and
this work is ongoing.
In the rat, all doses were effective peripherally. As shown in Ta-
ble 4, the 3 mpk dose provided an 160 nM total (37 nM free) drug
concentration and led to 72% Abeta 1–40 lowering (28% of vehicle
remaining). Higher doses provided good escalation of drug expo-
sure and led to efficacy that appeared maximal at 10–15% of vehi-
cle control. Brain drug levels, however, did not rise proportionally
to the peripheral levels, with the 100 mpk dose providing total
brain drug levels of only 260 nM, giving a Brain/plasma ra-
tio = 0.05, suggesting only blood volume contamination. As is
shown in Table 4, there was no brain Abeta 1–40 lowering at any
dose.
These data in the rat are consistent with the data we published
for compound 21b in both the normal and P-gp KO mouse (com-
pound 21b is referred to as compound A in this reference).16
Peripheral Abeta lowering was similar in both the wild-type and
the P-gp KO mice, with a calculated EC50 of 12 nM free drug, a va-
lue that provides approximately a 2.5-fold coverage of the cellular
IC50 of 5 nM. There was, however, no brain efficacy up to 100 mpk.
In P-gp knockout (KO) mice, compound 21b produced robust Abeta
1–40 lowering in both plasma and brain, with similar calculated
ED50s in both compartments (brain = 50 mpk, plasma = 30 mpk).16
This data demonstrated that compound 21b had excellent blood/
Acknowledgments
The authors appreciate the contributions of the following per-
sons. Maria Pierdomenico, Kelli Jones, and Rudy Krause conducted
the rat and mouse in vivo experiments, while Tracey Fiedler and Ja-
son Corsa executed the accompanying Abeta assays. Carol Krause
and Cathy Kieras performed our HEKsw assays. Paul Morin and
Vidhyashankar Ramamurthy generated the BACE constructs used
in crystallographic studies. Lisa Kopcho generated BACE-1 enzyme
assay data.
References and notes
All new compounds reported herein gave satisfactory analytical
data including parent mass, proton NMR spectral integrity, and
high purity by HPLC characterization.
1. Selkoe, D. J. Ann. Intern. Med. 2004, 140, 627.
2. Plosker, G. L.; Keating, G. M. Dis. Manag. Health Out. 2004, 12, 55.
3. Walsh, D. M.; Selkoe, D. J. J. Neurochem. 2007, 101, 1172.
4. Lin, X.; Koelsch, G.; Wu, S.; Downs, D.; Dashti, A.; Tang, J. Proc. Natl. Acad. Sci.
U.S.A. 2000, 97, 1456.
5. Hussain, I.; Powell, D.; Howlett, D. R.; Tew, D. G.; Meek, T. D.; Chapman, C.;
Gloger, I. S.; Murphy, K. E.; Southan, C. D.; Ryan, D. M.; Smith, T. S.; Simmons, D.
L.; Walsh, F. S.; Dingwall, C.; Christie, G. Mol. Cell. Neurosci. 1999, 14, 419.
6. Yan, R.; Bienkowski, M. J.; Shuck, M. E.; Miao, H.; Tory, M. C.; Pauley, A. M.;
Brashler, J. R.; Stratman, N. C.; Mathews, W. R.; Buhl, A. E.; Carter, D. B.;
Tomasselli, A. G.; Parodi, L. A.; Heinrikson, R. L.; Gurney, M. E. Nature 1999, 402,
533.
7. Sinha, S.; Anderson, J. P.; Barbour, R.; Basi, G. S.; Caccavello, R.; Davis, D.; Doan,
M.; Dovey, H. F.; Frigon, N.; Hong, J.; Jacobson-Croak, K.; Jewett, N.; Keim, P.;
Knops, J.; Lieberburg, I.; Power, M.; Tan, H.; Tatsuno, G.; Tung, J.; Schenk, D.;
Seubert, P.; Suomensaari, S. M.; Wang, S.; Walker, D.; Zhao, J.; McConlogue, L.;
John, V. Nature 1999, 402, 537.
8. 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.
9. Takami, M.; Nagashima, Y.; Sano, Y.; Ishihara, S.; Morishima-Kawashima, M.;
Funamoto, S.; Ihara, Y. J. Neurosci. 2009, 29, 13042.
Compound 21b Plasma Efficacy
200
150
100
50
WT mouse
Rat
10. Ghosh, A. K.; Shin, D.; Downs, D.; Koelsch, G.; Lin, X.; Ermolieff, J.; Tang, J. J. Am.
Chem. Soc. 2000, 122, 3522.
11. Hong, L.; Koelsch, G.; Lin, X.; Wu, S.; Terzyan, S.; Ghosh, A. K.; Zhang, X. C.;
Tang, J. Science (Washington, DC) 2000, 290, 150.
12. Duan, J. J. W.; Chen, L.; Wasserman, Z. R.; Lu, Z.; Liu, R.-Q.; Covington, M. B.;
Qian, M.; Hardman, K. D.; Magolda, R. L.; Newton, R. C.; Christ, D. D.; Wexler, R.
R.; Decicco, C. P. J. Med. Chem. 2002, 45, 4954.
0
-1
0
1
2
3
4
Log Free Plasma Conc (nM)
13. Freidinger, R. M. J. Med. Chem. 2003, 46, 5553.
Figure 6. Plasma efficacy of compound 21b in mice and rats.
14. Maillard, M.; Hom, R.; Gailunas, A.; Jagodzinska, B.; Fang, L. Y.; John, V.;
Freskos, J. N.; Pulley, S. R.; Beck, J. P.; Tenbrink, R. E. WO Patent Appl. 2002/
002512, 2002.
15. 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.
16. Meredith, J. E., Jr.; Thompson, L. A.; Toyn, J. H.; Marcin, L.; Barten, D. M.;
Marcinkeviciene, J.; Kopcho, L.; Kim, Y.; Lin, A.; Guss, V.; Burton, C.; Iben, L.;
Polson, C.; Cantone, J.; Ford, M.; Drexler, D.; Fiedler, T.; Lentz, K. A.; Grace, J. E.,
Jr.; Kolb, J.; Corsa, J.; Pierdomenico, M.; Jones, K.; Olson, R. E.; Macor, J. E.;
Albright, C. F. J. Pharmacol. Exp. Ther. 2008, 326, 502.
Table 4
Pharmacokinetics of compound 21b in rats
Dose
Plasma
total
Plasma Abeta
1–40 (% Veh)
Brain
level
(nM)
Brain Abeta
1–40 (% Veh) plasma
Brain/
(nM)
3 mpk
10 mpk
100 mpk 5800
160
740
28
10
15
46
78
260
85
92
91
0.32
0.11
0.05
17. Decicco, C. P.; Tebben, A. J.; Thompson, L. A.; Combs, A. P. U.S. Patent 7,557,
2004, 137, B2.

L. A. Thompson et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6909–6915
6915
18. Iben, L. G.; Kopcho, L.; Marcinkeviciene, J.; Zheng, C.; Thompson, L. A.; Albright,
C. F.; Toyn, J. H. Eur. J. Pharmacol. 2008, 593, 10.
25. Co-crystals of BACE-1 were prepared using the Phe-statine analog of OM99-2.
The crystals were soaked in 1 mM compound 14c in a stabilizing buffer
containing 35% PEG8K, 0.2 M ammonium sulfate, 0.1 M sodium cacodylate at
pH 6.2 for 7 days. Coordinates for the complex of BACE-1 with compound 14c
have been deposited in the Protein Data Bank (www.rcsb.org) under PBD code
3SKG.
26. Hom, R. K.; Gailunas, A. F.; Mamo, S.; Fang, L. Y.; Tung, J. S.; Walker, D. E.; Davis,
D.; Thorsett, E. D.; Jewett, N. E.; Moon, J. B.; John, V. J. Med. Chem. 2004, 47, 158.
27. Marcin, L. R.; Higgins, M. A.; Zusi, F. C.; Zhang, Y.; Dee, M. F.; Parker, M. F.;
Muckelbauer, J. K.; Camac, D. M.; Morin, P. E.; Ramamurthy, V.; Tebben, A. J.;
Lentz, K. A.; Grace, J. E.; Marcinkeviciene, J. A.; Kopcho, L. M.; Burton, C. R.;
Barten, D. M.; Toyn, J. H.; Meredith, J. E.; Albright, C. F.; Bronson, J. J.; Macor, J.
E.; Thompson, L. A. Bioorg. Med. Chem. Lett. 2011, 21, 537.
19. Co-crystals of BACE-1 were prepared using the Phe-statine analog of OM99-2.
The crystals were soaked with 1 mM compound 5 in a stabilizing buffer
containing 35% PEG8 K, 0.2 M ammonium sulfate, 0.1 M sodium cacodylate at
pH 6.2 for 9 days. Coordinates for the complex of BACE-1 with compound 5
have been deposited in the Protein Data Bank (www.rcsb.org) under PDB code
3SKF.
20. This concept was not reported in the literature at the time of this work, but it
has subsequently appeared in the literature. See Iserloh, U. et al, J. Bioorg. Med.
Chem. Lett. 2008, 18, 414 and references cited therein.
21. Thompson, L. A.; Boy, K. M.; Shi, J.; Macor, J. E.; Good, A. C.; Marcin, L. R. U.S.
Patent Appl. 2008/0153868, 2008.
22. Carter, P.; LaPorte, J.; Scherle, P.; DeCicco, C. Bioorg. Med. Chem. Lett. 2003, 13,
1237.
23. Thompson, L. A.; Boy, K. M.; Shi, J.; Macor, J. E. U.S. Patent 7,388,007, 2008,
B2
24. Higaki, J. N.; Chakravarty, S.; Bryant, C. M.; Cowart, L. R.; Harden, P.; Scardina, J.
M.; Mavunkel, B.; Luedtke, G. R.; Cordell, B. J. Med. Chem. 1999, 42, 3889.
28. Boy, K. M.; Guernon, J. M.; Wu, Y.-J.; Zhang, Y.; Gerritz, S. W.; Toyn, J. H.;
Meredith, J. E.; Barten, D. M.; Burton, C. R.; Albright, C. F.; Good, A. C.;
Muckelbauer, J. K.; Camac, D. M.; Grace, J. E.; Lentz, K. A.; Olson, R. E.; Macor, J.
E.; Thompson, L. A. Bioorg. Med. Chem. Lett. 2010, accompanying article.
29. Thompson, L. A.; Boy, K. M.; Shi, J.; Macor, J. E. U.S. Patent Appl. 2006/046984,
2006.