Rational design and synthesis of selective BACE-1 inhibitors

Article abstract of DOI:10.1016/j.bmcl.2003.11.061

An effective approach for enhancing the selectivity of β-site amyloid precursor protein cleaving enzyme (BACE 1) inhibitors is developed based on the unique features of the S1′ pocket of the enzyme. A series of low molecular weight (<600) compounds were s

Full text of DOI:10.1016/j.bmcl.2003.11.061

Bioorganic & Medicinal Chemistry Letters 14 (2004) 601–604
Rational design and synthesis of selective BACE-1 inhibitors§
Stephen F. Brady,a Satendra Singh,b Ming-Chih Crouthamel,c M. Katharine Holloway,a
Craig A. Coburn,a Victor M. Garsky,a Michael Bogusky,a Michael W. Pennington,b
Joseph P. Vacca,a Daria Hazudac and Ming-Tain Laic,*
aDepartment of Medicinal Chemistry, Merck Research Laboratories, West Point, PA 19486, USA
bBachem Bioscience Inc., King of Prussia, PA 19406, USA
cDepartment of Biological Chemistry, Merck Research Laboratories, West Point, PA 19486, USA
Received 28 September 2003; revised 20 November 2003; accepted 20 November 2003
Abstract—An effective approach for enhancing the selectiv0ity of b-site amyloid precursor protein cleaving enzyme (BACE 1) inhi-
bitorsisdeveloped based on the unique featuresof the S1
pocket of the enzyme. A series of low molecular weight (<600) com-
pounds were synthesized with different moieties at the P10 position. The selectivity of BACE 1 inhibitors versus cathepsin D and
renin wasenhanced 120-fold by replacing the hydrophobic propyl group with a hydrophilic propionic acid group.
# 2003 Elsevier Ltd. All rights reserved.
b-site amyloid precursor protein cleaving enzyme
(BACE 1, also called memapsin 2, Asp 2), a type 1
membrane bound aspartyl protease, has been demon-
strated to be the major b-secretase responsible for the
b-site cleavage of amyloid precursor protein (APP) to
generate a C-99 fragment.1 The C-99 peptide is subse-
quently processed by g-secretase to release Ab40/42
peptides, which are the major components of amyloid
plaques, a hallmark of Alzheimer’s Disease (AD).2 Tis-
sue distribution studies have shown that BACE 1 is
predominantly expressed in the brain and pancreas,
where the protein in pancreasisan inactive BACE 1
spliced form.3 BACE 1 deficient mice have been shown
to be phenotypically normal and fertile.4 More impor-
tantly, Ab40 and 42 are not detectable in either BACE 1
deficient mouse brain or embryonic cortical neurons.1,4
These unique features render BACE 1 an ideal ther-
apeutic target for AD.
have shown good enzymatic7 and cellular potency,8 but
the selectivity versus related aspartyl proteases has been
poor.7a BACE 1 selectivity is important since the target
enzyme resides in the brain, where closely related ubi-
quitousenzymescathepsin D (Cat D) and renin can be
found. BACE 1 inhibitorswith poor selectivity against
these enzymes are likely to cause undesired side effects.
To address this concern, we compared the X-ray struc-
9
turesof renin, Cat D,10 and BACE 111 and noticed
significant differences in the length and sequence of the
loop defining the S10/S30 pocket (Fig. 1). The BACE 1
loop is truncated (like the fungal aspartyl proteases)12
and primarily hydrophilic, whereasthe renin and Cat D
loopsare extended and primarily hydrophobic. In
addition, Arg235 (numbering based on the mature
form), which resides between the P10 and P2 regions, is
unique to BACE 1 (Val233 in Cat D and Ser222 in
renin) and may contribute to the preference for a P10
Asp in the BACE 1 substrate APP.13 Based on this
analysis, it seemed reasonable that the selectivity of a
BACE 1 inhibitor could be greatly enhanced by
increasing the polarity of the side chain at the P10 posi-
tion.
Consequently, since the discovery of BACE 1,5 intensive
efforts have focused on the synthesis of the enzyme
inhibitors that possess appropriate properties for drug
development, such as good potency, small size, and high
selectivity.6 A number of statine-containing compounds
To test this hypothesis, sid0e chains with different pola-
ritieswere placed at the P1 position of a BACE 1 inhi-
bitor discovered via targeted screening of compounds
similar to known BACE 1 inhibitors (Sta-Val14 and
§Supplementary data associated with this article can be found at doi:
10.1016/j.bmcl.2003.11.061
* Corresponding author. Tel.: +215-652-5038; fax: +215-652-0264;
e-mail: mintain_lai@merck.com
0960-894X/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2003.11.061

602
S. F. Brady et al. / Bioorg. Med. Chem. Lett. 14 (2004) 601–604
OM99-27), but different from Pepstatin A which does
not inhibit BACE 1 at high concentration (>1 mM). As
shown in Scheme 1, the protected intermediates 7, 8,
and 10 were synthesized starting from lactones 1 and
2.7,15 Alkylation of lactones 1 or 2 with allyl iodide in
the presence of LHMDS proceeded in good yield and
afforded the predominantly kinetically controlled trans
isomers, 3 or 4 in 72–79% yield. The other isomer (cis)
wasalso formed but only in <5% yield. The main side
product in thisreaction wa,s however, the dialkylated
lactone, which presumably formed as a result of aggre-
gation of anion.16
Lactones 3 and 4 were converted to primary alcohols 5
and 6, respectively, in 70–73% yield via hydroboration
with catecholborane in the presence of Wilkinson’s cat-
alyst and subsequent oxidation using H2O2 under neutral
conditions.17 Hydroboration in the absence of the rho-
dium catalyst gave poor yield due to competitive reduc-
tion of the g-lactone to form a diol intermediate.18
Lactone 3 washydrolyzed with aq LiOH and silylated
with TBSCl in the presence of imidazole to afford 7, the
synthetic precursor of compound 12. Similarly, lactone
5 wassilylated and then hydrolyzed to give compound
8, the requisite intermediate for the synthesis of inhibi-
tors 14,15. Lactone 6 was subjected to a Sharpless oxi-
dation in 76% yield to provide lactone 9, which was
selectively protected as a tert-Bu ester in 50% yield to
furnish lactone 10,19,20 the intermediate required for the
synthesis of 17 (Scheme 2).
Intermediates 7, 8, and 10 were subjected to standard
deprotection and amide coupling proceduresto yield
compounds 12–17 with molecular weightslesthan 600.
In the synthesis of analogue 12 the final hydrogenolytic
cleavage of the benzyl ester after coupling of valeric acid
to compound 11 proceedswith reduction of the isde
chain allyl double bond. In the route to analogue 14 the
overall yield wassignificantly lowered by the formation
of lactone in the amidation of intermediate 8 with iso-
butyl amine. In addition, the use of HCl in ethyl acetate
in conversion of 13 resulted in formation about 50% of
the O-acetyl by-product, which ultimately afforded
analogue 15. The problem of lactone formation was
circumvented in the synthesis of 16 by direct aminolysis
of the lactone precursor 10 with isobutyl amine, aspartic
acid being incorporated asitsNa-Fmoc derivative. The
free amine generated upon treatment of 16 with piper-
idine could be coupled with valeric acid without the
need for isolation, and treatment with TFA removed
both the Boc and t-butyl ester moieties to afford analo-
gue 17. Compounds 14 and 17 also represent the two
most functionalized hydroxyethylene isosteres at the P10
position reported to date.
Figure 1. (a) Sequence comparison of the BACE 1, Cathepsin D, and
renin S10/S30 loop derived from an overlay of the crystal structures
1FKN, 1LYB, and 1HRN, respectively. Numbering is taken from the
crystal structure. Note the 6-residue deletion in BACE 1. (b) Structure
comparison of S10/S30 region of BACE 1 (yellow), Cathepsin D
(magenta), and human renin (cyan). Arg235 (BACE 1) ishsown in
ball-and-stick format. Models of 3 low-energy poses for 17 which vary
in the orientation of the P10 side-chain are illustrated bound in the
BACE 1 active site.
The ability of these four compounds to inhibit BACE 1,
Cat D, and renin wasevaluated ( Table 1). A soluble
truncated form of BACE 1 (1–460) wasexpreessd in
insect cells using Baculovirus and was then purified
from a concanavalin A-Sepharose followed by a Sta-Val
affinity column.21 A fluorescence resonance energy
transfer (FRET) substrate containing the Swedish
sequence (KMDA!NLDA) wasemployed for mon-
itoring the activity of the three enzymes.22 The substrate
appearsto be cleaved efficiently by the enzymesat their
optimum pH. Ki (app) valueswere calculated based on
the initial rate of substrate cleavage in the absence or
presence of inhibitors at different concentrations.
Scheme 1. Synthesis of the intermediates of BACE 1 inhibitors.
Reagentsand conditions: (a) LHMDS, THF, ꢀ78 ꢁC, allyl iodide, 30
min; (b) Rh(PPh3)3, THF, catechol borane, rt, 90 min; (c) 50 mM
KH2PO4, THF/EtOH, pH 7.2, 30% H2O2, rt, 12 h; (d) MeCN, CCl4,
NaIO4, RuCl3, H2O, 2 h; (e) H2SO4, isobutylene, CH2Cl2/dioxane, 48
h; (f) 1N LiOH, 16 h; (g) TBSCl, imidazole, DMF, rt, 24 h; (h) TBSCl,
imidazole, DMF, rt, 12 h; (i) NaOH/dioxane/H2O 30 min.
Ashsown in Table 1, compound 12 displays greater
potency for inhibiting both Cat D and renin than BACE

S. F. Brady et al. / Bioorg. Med. Chem. Lett. 14 (2004) 601–604
603
Scheme 2. Synthesis of BACE 1 inhibitors. Reagents and conditions: (a) isobutyl amine/DMF, EDT–HoBt, NMM, 20 h; (b) HCl/EtOAc, ꢀ20 ꢁC to
0 ꢁC, 20 min; (c) Na-Boc-l-aspartica-benzyl ester/DMF, EDT–HOBt, NMM, 20 h; (d) isovaleric acid/DMF, EDT–HOBt, NMM, 20 h; (e) H2-10%
Pd(C)/EtOH, 2 h; (f) isobutylamine/THF, 2-hydroxypyridine, 72 h; (g) Na-Fmoc-l-aspartic a-t-butyl ester/DMF, EDT–HOBt, NMM, 20 h; (h)
30% piperidine/DMF, 10 min.
1. The addition of an o-hydroxyl group to the P10 side-
chain of 12 yieldsinhibitor 14 and results in a 36-fold
increase in potency against BACE 1 but only a 2-fold
increase in potency versus Cat D and renin. Interest-
ingly, the acetyl derivative (15) maintainsBACE 1
potency but shows approximately 5-fold increase in the
potency for Cat D and renin, thus diminishing its selec-
tivity. Furthermore, replacement of the hydroxyl moiety
with a carboxylic acid functional group (17) enhances
Cat D selectivity from 6-fold to 47-fold as well as renin
selectivity from 6-fold to 44-fold while maintaining
BACE 1 potency. Overall, the selectivity of this series of
BACE 1 inhibitorsagainst Cat D and renin isenhanced
approximately 120-fold by simply replacing the hydro-
phobic propyl group with a hydrophilic propionic acid
group at the P10 position. These results suggest that the
selectivity of 17 isattributable to the carboxylic acid at
P10 and that it may have a favorable interaction with
Arg235 or the more hydrophilic environment of the
truncated S10/S30 loop of BACE 1 (Fig. 1).
compounds will display good selectivity against pepsin
and cathepsin E because they are in the same family as
Cat D and renin, which have the extended S10/S30 loop
and Thr replacing the Arg235.3
This study demonstrates that selective BACE 1 inhibi-
torscan be achieved based on the unique featuresof the
BACE 1 S10 pocket. Thisknowledge hsould facilitate
the discovery of small, potent, and highly selective
BACE 1 inhibitors.
Acknowledgements
The authorsthank MsEmily R. Payne for NMR
determination and Dr. Charles W. Ross III for mass
spectroscopic analysis. The authors are also grateful to
Drs. Elizabeth Chen-Dodson and Xiao-Ping Shi for
providing BACE 1, and Yue-ming Li for advising on
the FRET assay.
Interestingly, these modifications did not offer any
selectivity against BACE 2 (a homologue of BACE 123).
This is not surprising because BACE 2 also possesses
the truncated S10/S30 loop aswell asthe Arg232 between
P10 and P2 pockets. On the other hand, we expect these
References and notes
1. Cai, H.; Wang, Y.; McCarthy, D.; Wen, H.; Borchelt,
D. R.; Price, D. L.; Wong, P. Nature Neuroscience 2001,
4, 233.
Table 1. Selectivity and potency of BACE 1 inhibitors
2. Selkoe, D. J. Trends Cell Biol. 1998, 8, 447.
3. (a) 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, Yi.; Fisher, S.; Fuller, J.;
Edenson;, S.; Lile, J.; Jarosinki, M. A.; LeonaBiere, A.;
Curran, E.; Burgess, T.; Louis, J.-C.; Collins, F.; Treanor,
J.; Rogers, G.; Citron, M. Science 1999, 286, 735. (b)
Vassar, R. J. Mol. Neuroscience 2001, 17, 157. (c) Bod-
endorf, U.; Fischer, F.; Bodian, D.; Multhaup, G.;
Paganette, P. J. Biol. Chem. 2001, 276, 12019.
4. Luo, Y.; Bolon, B.; Kahn, S.; Bennett, B. D.; Babu-Khan,
S.; Dents, P.; Fan, Wei.; Kha, H.; Zhang, J.; Gong, Y.;
Marting, L.; Louis, J.-C.; Yan, Q.; Richards, W. G.;
Citron, M.; Vassar, R. Nature Neuroscience 2001, 4, 231.
R1
BACE 1 Ki (mM) Cat D Ki (mM) Renin Ki (mM)
12
–CH3ꢀ
14 CH2OH
15 CH2OAc
69.5ꢂ9.3
1.9ꢂ0.1
2.7ꢂ0.4
1.7ꢂ0.1
28.0ꢂ3.2
12.5ꢂ0.2
2.2ꢂ0.2
24.4ꢂ0.8
10.9ꢂ0.8
1.9ꢂ0.1
17
CO2H
79.7ꢂ6.7
73.6ꢂ8.2

604
S. F. Brady et al. / Bioorg. Med. Chem. Lett. 14 (2004) 601–604
5. Citron, M. J. Neurosci. Res. 2002, 70, 373 and references
cited therein.
6. Ghosh, A. K.; Hong, L.; Tang, J. Curr. Med. Chem. 2002,
9, 1135.
7. (a) Ghosh, A. K.; Shin, D.; Downs, D.; Koelsch, G.; Lin,
X.; Ermolieff, J.; Tang, J. J. Am. Chem. Soc. 2000, 122,
3522. (b) Ermolieff, J.; Loy, J. A.; Koelsch, G.; Tang, J. J.
Biochemistry 2000, 39, 12450. (c) Turner, R. T.; Koelsch,
G.; Hong, L.; Castenheira, P.; Ghosh, A.; Tang, J. J.
Biochemistry 2001, 40, 10001.
N.; Hong, J.; Jacobson-Croak, K.; Hweyt, 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.
15. (a) Fray, A. H.; Kaye, R. L.; Kleinman, E. F. J. Org.
Chem. 1986, 51, 4828. (b) Brewer, M.; Rich, D. Org. Lett.
2001, 3, 945.
16. Williams, R. M.; Im, M. N. J. Am. Chem. Soc. 1991, 113,
9276.
8. Horn, R. K.; Fang, L. y.; Mamo, S.; Tung, J. S.; Guinn,
A. C.; Walker, D. E.; Davis, D. L.; Gailunas, A. F.;
Thorsett, E. D.; Sinha, S.; Knops, J. E.; Jewett, N. E.;
Anderson, J. E.; John, V. J. Med. Chem. 2003, 46, 1799.
9. Tong, L.; Lamarre, D.; Pilote, L.; LaPlante, S.; Anderson,
P. C.; Jung, G. J. Mol. Biol. 1995, 250, 211.
10. Baldwin, E. T.; Bhat, T. N.; Gulnik, S.; Hosur, M. V.;
Sowder, R. C., II; Cavhau, R. E.; Collins, J.; Silva, A. M.;
Erickson, J. W. Proc. Natl. Acad. Sci. 1993, 90, 6796.
11. Hong, L.; Koelsch, G.; Lin, X.; Wu, S.; Terzyan, S.;
Ghosh, A. K.; Zhang, X. C.; Tang, J. Science 2000, 290,
150.
17. (a) Evans, D. A.; Fu, G. C.; Hoveyda, A. H. J. Am.
Chem. Soc. 1992, 114, 6671. (b) Evans, D. A.; Fu, G. C.;
Anderson, B. A. J. Am. Chem. Soc. 1992, 114, 6679.
18. (a) Brown, H. C.; Knights, E. F.; Scouten, C. G. J. Am.
Chem. Soc. 1974, 96, 7765. (b) Brown, H. C.; Kulkarni,
S. U.; Rao, C. G. Synthesis 1980, 151. (c) Kabalka, G. W.;
Shoup, T. M.; Goudgaon, N. M. J. Org. Chem. 1989, 54,
5930.
19. Carlsen, P. H. J.; Katsuki, T.; Martin, V. S.; Sharpless,
K. B. J. Org. Chem. 1981, 4, 3936.
20. Bodanszky, A. The Practice of Peptide Synthesis; Springer
Verlag: Berlin-Heidelberg, 1994; pp 38–39.
12. See, for example, the PDB structures for Endothia pepsin
(1ER8), Rhizopus pepsin (2APR), and penicillopepsin
(3APP).
13. Sauder, J. M.; Arthur, J. W.; Dunbrack, R., Jr. J. Mol.
Biol. 2000, 300, 241.
21. Shi, X.-P.; Chen, E.; Yin, K.-C.; Na, S.; Garsky, V. M.;
Lai, M.-T.; Li, Y.-M.; Platchek, M.; Register, B.; Sar-
dana, M. K.; Tang, M.-J.; Thiebeau, J.; Wood, T.; Shafer,
J. A.; Gardell, S. J. J. Biol. Chem. 2001, 276, 10366.
22. The detailed assay protocol will be published elsewhere.
23. Saunders, A. J.; Kim, T.-W.; Tanzi, R. E.; Fan, W.; Ben-
nett, B. D.; Babu-Khans, et al. Science 1999, 286, 1255a.
14. Sinha, S.; Anderson, J. P.; Barbour, R.; Basl, G. S.; Cac-
cavello, R.; Davis, D.; Doan, M.; Dovey, H. F.; Frigon,