Discovery and X-ray crystallographic analysis of a spiropiperidine iminohydantoin inhibitor of β-secretase

Article abstract of DOI:10.1021/jm800914n

A high-throughput screen at 100 μM inhibitor concentration for the BACE-1 enzyme revealed a novel spiropiperidine iminohydantoin aspartyl protease inhibitor template. An X-ray cocrystal structure with BACE-1 revealed a novel mode of binding whereby the in

Full text of DOI:10.1021/jm800914n

J. Med. Chem. 2008, 51, 6259–6262
6259
Discovery and X-ray Crystallographic
Analysis of a Spiropiperidine Iminohydantoin
Inhibitor of ꢀ-Secretase‡
James C. Barrow,*,† Shaun R. Stauffer,† Kenneth E. Rittle,†
Phung L. Ngo,† ZhiQiang Yang,† Harold G. Selnick,†
Samuel L. Graham,† Sanjeev Munshi,4
Georgia B. McGaughey,§ M. Katharine Holloway,§
Adam J. Simon,| Eric A. Price,| Sethu Sankaranarayanan,|
Dennis Colussi,| Katherine Tugusheva,| Ming-Tain Lai,|
Amy S. Espeseth,| Min Xu,| Qian Huang,| Abigail Wolfe,|
Beth Pietrak,| Paul Zuck, Dorothy A. Levorse,#
Daria Hazuda,| and Joseph P. Vacca†
Figure 1. Piperidine aspartyl protease inhibitors
Scheme 1. General Synthesis of Inhibitors
Departments of Medicinal Chemistry, Molecular Systems, Structural
Biology, Automated Biotechnology, and Alzheimer’s Research,
Merck Research Laboratories, West Point, PennsylVania 19486
ReceiVed July 22, 2008
Abstract: A high-throughput screen at 100 µM inhibitor concentration
for the BACE-1 enzyme revealed a novel spiropiperidine iminohydan-
toin aspartyl protease inhibitor template. An X-ray cocrystal structure
with BACE-1 revealed a novel mode of binding whereby the inhibitor
interacts with the catalytic aspartates via bridging water molecules.
Using the crystal structure as a guide, potent compounds with good
brain penetration were designed.
Treatments for Alzheimer’s disease, especially those that can
slow or reverse neurodegeneration, represent a significant unmet
medical need. Incidence of the disease increases with age and
is thus growing in prevalence due to the aging population. As
a consequence, Alzheimer’s disease is forecasted to have a
significant increased healthcare burden.1 Abnormal production
and/or clearance of a small peptide termed Aꢀ has been
implicated in the pathophysiology of Alzheimer’s disease from
genetic and other studies, and agents that modulate Aꢀ produc-
tion represent an attractive strategy in the search for novel
therapeutic agents.2 The Aꢀ peptide arises from processing of
the amyloid precursor protein (APPa) by several proteases, first
by ꢀ-secretase followed by γ-secretase. On the basis of these
observations, ꢀ-secretase (BACE-1) is a promising drug target
for disease-modifying therapy and has attracted significant
attention from the medicinal chemistry community.3
of central Aꢀ lowering following peripheral dosing of com-
pounds.5 Given the high value of this target, many approaches
to finding new leads with better brain-penetration prospects have
been investigated. Modification of the traditional transition-state
isosteres has proved successful in several instances, and
fragment-based approaches are now beginning to bear new
leads.6 High-throughput screening (HTS) of large collections
of diverse compounds is often successful at identifying leads
for medicinal chemistry programs; however, for practical
reasons, the screens are often done at 10 µM or lower
concentration of test compound, thereby missing weak hits.
Given the low hit-rate of our initial BACE-1 HTS screen, a
second screen was performed at higher test compound concen-
tration (100 µM) whereby the substrate was immobilized to
allow for extensive washing to minimize artifacts.7
BACE-1 is a challenging target because it is an aspartyl
protease with a site of action inside the central nervous system.
Experience from other aspartyl protease drug targets such as
renin and HIV protease led to transition-state isosteres as first-
generation inhibitors.4 Despite good potency, these initial
compounds had poor brain penetration and there are few reports
From this effort, compound 1 emerged as a weak BACE-1
inhibitor with a potency of 22 µM in the isolated enzyme assay
but inactive in a cell-based assay (Figure 1).8 Despite the weak
activity, this spiropiperidine structure looked promising as a lead.
The overall molecular scaffold is quite different than traditional
transition-state isosteres and more similar to piperidine-based
renin inhibitors (such as 2) discovered by Roche more than a
decade ago.9 The binding mode of 1 was not clear from docking
the compound into known crystal structures of BACE-1 or
models generated from renin complexed with 2. Furthermore,
attempts to soak or cocrystallize the compound with BACE-1
were unsuccessful, thus it was deemed that a more potent and
soluble compound would be needed.
‡
A PDB file for the BACE-1/inhibitor 16 complex (PDB identifier
3E3W) has been deposited with the Protein Data Bank.
* To whom correspondence should be addressed. Phone: 215-652-4780.
Fax: 215-652-7310. E-mail: james_barrow@merck.com.
†
Department of Medicinal Chemistry.
Department of Structural Biology.
Department of Molecular Systems.
Department of Alzheimer’s Research.
4
§
|
Department of Automated Biotechnology.
#
Department of Materials Characterization.
a Abbreviations: APP, amyloid precursor protein; BACE-1, ꢀ-site
amyloid precursor protein cleaving enzyme or ꢀ-secretase; HTS, high-
throughput screening; PGP, P-glycoprotein.
10.1021/jm800914n CCC: $40.75
2008 American Chemical Society
Published on Web 09/24/2008

6260 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 20
Table 1. N-Aryl Variations
Letters
a
b
no.
R2
BACE-1 IC50 (µM)
sAPPꢀ (µM)
9
H
>100
NT
NT
10
11
12
13
CH2CH2CH(CH3)3
benzyl
3-fluorophenyl
4-acetamidophenyl
>100
>100
11 ( 0.98
8 ( 0.22
NT
32 c
21 ( 8.7
a IC50 determinations were performed as described in ref 8. b Cell-based
assay as described in ref 8. NT ) not tested. c Average of two determinations.
Table 2. Piperidine Modifications
a
b
no.
R3
BACE-1 IC50 (µM)
sAPPꢀ (µM)
PGPd
14
H
>100
NT
NT
3.4
NT
19
12 benzyl
15 4-fluorophenyl
16 4-acetamidobenzyl
11 ( 0.98
>100
32 c
NT
8.2 ( 1.7
2.8 ( 0.12
a IC50 determinations were performed as described in ref 8. b Cell-based
assay as described in ref 8. NT ) not tested. c Average of two determinations.
d B:A/A:B ratio in MDR1 transfected cells (see ref 16).
Synthesis of these compounds can be achieved by an Ugi
four component coupling with piperidone 3, an isonitrile 4,
amines 5, and KOCN in modest yield as shown in Scheme 1.10
Use of the Boc-protected piperidone 3 allows for subsequent
Boc removal and alkylation and acylation of the piperidine to
give inhibitors 9-18.
Figure 2. Two views of the key hydrogen bond interactions between
16 and BACE-1, a three-dimensional view of the protein with the
catalytic aspartates and bridging water molecules highlighted, and a
two-dimensional representation. Heavy-atom distances are shown in
Å.
Initial studies with a variety of commercially available
isonitriles to explore the exocyclic amine substituent revealed
that the original cyclohexyl was superior to benzyl or smaller
alkyl groups (data not shown) and thus the cyclohexyl was used
for subsequent studies.
of aspartyl protease inhibitors displace the catalytic water
molecule upon binding,13 although there are recent reports of a
novel inhibitor from these laboratories14 and a weak fragment15
interacting with BACE-1 via a water molecule with the catalytic
aspartates.
Significant variations of the N-aryl substituent were not
tolerated as shown in Table 1.8 Alkyl and benzyl were inactive;
however, some substitutions of the phenyl resulted in improved
potency such as 3-fluoro and 4-acetamido. Importantly, these
changes brought measurable cell-based activity and the shift
between enzyme potency and cell-based activity was only 3-fold.
With other scaffolds, large discrepancies between enzyme
potency and cell-based activity are common, and the results in
Table 1 gave further encouragement to pursue this lead.
Using the potency-enhancing 3-fluorophenyl, changes to the
piperidine alkyl group were examined as shown in Table 2.
Deviations from the original benzyl substitution resulted in a
loss in potency, and acylation also diminished activity. Extensive
empirical exploration of substituted benzyl groups failed to
deliver significant gains in potency; however, the 4-acetamido
substituted benzyl compound 16 was one of the few that offered
an improvement in solubility without compromising activity,
and cell-based activity was now 8.2 µM.
The most notable feature of the 16-BACE-1 complex is the
bridging water molecules between the catalytic aspartates and
the basic piperidine and exocyclic NH of the inhibitor (Figure
2), in contrast to the Roche renin inhibitor 2, where the
piperidine interacts directly with the aspartates. This unantici-
pated mode of binding for 16 explains the lack of success with
initial modeling using previously defined structures and defines
a new mode of aspartyl protease inhibition. The catalytic water
molecule bridges the aspartates to the basic nitrogen of the
piperidine. A second water is within hydrogen-bonding distance
of Asp228, Gly34, and the exocyclic nitrogen of the inhibitor.
Because the structure is at 1.8 Å resolution, the locations of
the hydrogen atoms cannot be specified. The exocyclic NH is
likely a hydrogen bond donor because pKa measurements of
16 determined by potentiometric titration indicate pKa values
of 2.41 for the iminohydantoin and 6.79 for the piperidine.
As a result of the intervening water molecule, the inhibitor
sits higher in the active site and the flexible hairpin loop of
residues from Val69 to Trp76 called the “flap” cannot fully close
down as it does in traditional structures (Figure 3). There is a
hydrogen bond formed between the iminohydantoin carbonyl
and the flap backbone NH of Gln73, and the raised flap allows
the Tyr71 side chain to shift down into the traditional P1 pocket
and form new hydrogen bonds with the structural water network
This combination of potency and solubility allowed for
successful determination of the X-ray cocrystal structure of 16
with the BACE-1 enzyme at 1.8 Å resolution,11 revealing a new
mode of aspartyl protease inhibition. Proteases in this class use
two interacting aspartic acid residues to activate a water
molecule that attacks the scissile amide bond.12 Most classes

Letters
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 20 6261
group (Figure 4, compound 18) resulted in a 10-fold increase
in BACE-1 potency. This can be rationalized based on the X-ray
structure of compound 16 in that a model of the ortho-methyl
substituted phenyl group demonstrated that it more effectively
filled the S3 pocket compared to the unsubstituted phenyl
compound 17. Despite the modest potency in the cell-based
assay, this was a key compound because it was not a PGP
substrate (B:A/A:B transport ratio 1.8) and had good brain levels
(brain concentration 1.7 µM, brain/plasma ratio 0.44) after a
30 mg/kg ip dose to mice. After a 100 mg/kg dose of 18 to
transgenic mice expressing human APP under the control of a
yeast artificial chromosome vector,17 a statistically nonsignificant
reduction in brain Aꢀ40 of 8% was observed. Plasma and brain
levels of 18 were high at 16 and 14 uM, respectively; however,
when adjusted for mouse plasma protein binding (9% unbound),
free compound concentration is well below the 5.2 µM cellular
IC50, which may explain the less than robust result. Nevertheless,
the good brain penetration observed in this study with this new
class of BACE-1 inhibitors is encouraging.
Figure 3. View of the X-ray structure of inhibitor 16 in the BACE-1
active site. The S3 site of the enzyme is highlighted in green.
In summary, a high-throughput screening campaign has
identified a novel, nonpeptidic template for aspartyl protease
inhibition with good potential for brain penetration. While
cellular potency needs improvement for in vivo efficacy, this
class of compounds represent a good starting point for further
development.
Acknowledgment. We thank C. Ross and J. Murphy for
mass spectral analysis and J. Hochman and N. Pudvah for PGP
experiments.
Supporting Information Available: Experimental procedures,
characterization of new compounds, and X-ray structure coordinates.
This material is available free of charge via the Internet at http://
pubs.acs.org.
Figure 4. Designed BACE-1 inhibitors.
References
rather than Trp76. This Tyr71 motion and the raised flap forms
a new pocket just under the flap, and the fluorophenyl ring of
the inhibitor tucks into this space, thereby explaining the tight
SAR in this region. The cyclohexyl group fills the S1′ pocket,
and the acetamido NH makes a hydrogen bond to the Phe108
backbone carbonyl. Most notably, 16 does not fill either the S2
or S3 sites of the enzyme.
Further profiling of 16 revealed no selectivity for BACE-2
(IC50 ) 1.5 µM) but excellent selectivity versus other aspartyl
proteases (IC50 > 100 µM versus renin and cathepsin D).
Despite the basic nitrogen, significant hERG binding was not
observed (IC50 > 10 µM).
Given the promising cellular potency and overall selectivity
versus other enzymes, further optimization was undertaken with
focus toward occupying the S2 or S3 site of BACE-1.
Traditionally, appropriate substituents in these sites contribute
significantly to potency; therefore, initial efforts were aimed at
growing the molecule in these directions.
We chose to focus on filling the S3 site by substituting the
benzyl group. Despite the extra hydrogen bond interaction with
the backbone carbonyl of Phe108, the 4-acetamido group did
not give a significant boost in potency (16 vs 12, Table 2) and
was considered a liability in terms of brain penetration as
compound 16 showed a high P-glycoprotein (PGP) transport
ratio (B:A/A:B ratio of 19) with low permeability.16 Therefore,
this substituent was removed and S3 access was attempted from
the 3-position of the benzyl ring. While an unsubstituted
3-phenyl group did not offer any benefit on its own (compound
17, Figure 4), it served as a good scaffold to orient groups in
the S3 pocket from the meta position. Addition of a methyl
(1) (a) Nussbaum, R. L.; Ellis, C. E. Alzheimer’s Disease and Parkinson’s
Disease. N. Engl. J. Med. 2003, 348, 1356–1364. (b) Selkoe, D. J.
Alzheimer Disease: Mechanistic Understanding Predicts Novel Thera-
pies. Ann. Intern. Med. 2004, 140, 627–638. (c) Mattson, M. P.
Pathways towards and away from Alzheimer’s disease. Nature 2004,
430, 631–639.
(2) (a) Goedert, M.; Spillantini, M. G. A Century of Alzheimer’s Disease.
Science 2006, 314, 777–781. (b) Cummings, J. L. Alzheimer’s Disease.
N. Engl. J. Med. 2004, 351, 56–67. (c) Cole, S. L.; Vassar, R. BACE-1
Structure and Function in Health and Alzheimer’s Disease. Curr.
Alzheimer Res. 2008, 5, 100–120.
(3) (a) Guo, T.; Hobbs, D. W. Development of BACE-1 Inhibitors for
Alzheimer’s Disease. Curr. Med. Chem. 2006, 13, 1811–1829. (b)
John, V.; Beck, J. P.; Bienkowski, M. J.; Sinha, S.; Heinrikson, R. L.
Human ꢀ-secretase (BACE) and BACE inhibitors. J. Med. Chem.
2003, 46, 4625–4630.
(4) Hills, I. D.; Vacca, J. P. Progress toward a practical BACE-1 inhibitor.
Curr. Opin. Drug DiscoVery DeV. 2007, 10, 383–391.
(5) (a) Stanton, M. G.; Stauffer, S. R.; Gregro, A. R.; Steinbeiser, M.;
Nantermet, P. G.; Sankaranarayanan, S.; Price, E. A.; Wu, G.;
Crouthamel, M.-C.; Ellis, J.; Lai, M.-T.; Espeseth, A. S.; Shi, X.-P.;
Jin, L.; Colussi, D.; Pietrak, B.; Huang, Q.; Xu, M.; Simon, A. J.;
Graham, S. L.; Vacca, J. P.; Selnick, H. G. Discovery of isonicoti-
namide derived ꢀ-secretase inhibitors: In vivo reduction of ꢀ-amyloid.
J. Med. Chem. 2007, 50, 3431–3433. (b) Hussain, I.; Hawkins, J.;
Harrison, D.; Hille, C.; Wayne, G.; Cutler, L.; Buck, T.; Walter, D.;
Demont, E.; Howes, C.; Naylor, A.; Jeffrey, P.; Gonzalez, M. I.;
Dingwall, C.; Michel, A.; Redshaw, S.; Davis, J. B. Oral administration
of a potent and selective non-peptidic BACE-1 inhibitor decreases
ꢀ-cleavage of amyloid precursor protein and amyloid-ꢀ production in
vivo. J. Neurochem. 2007, 100, 802–809.
(6) (a) Murray, C. W.; Callaghan, O.; Chessari, G.; Cleasby, A.; Congreve,
M.; Frederickson, M.; Hartshorn, M. J.; McMenamin, R.; Patel, S.;
Wallis, N. Application of Fragment Screening by X-ray Crystal-
lography to ꢀ-Secretase. J. Med. Chem. 2007, 50, 1116–1123. (b)
Congreve, M.; Aharony, D.; Albert, J.; Callaghan, O.; Campbell, J.;
Carr, R.; Chessari, G.; Cowan, S.; Edwards, P. D.; Fredrickson, M.;
McMenamin, R.; Murray, C. W.; Patel, S.; Wallia, N. Application of

6262 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 20
Letters
Fragment Screening by X-ray Crystallography to the Discovery of
Aminopyridines as Inhibitors of ꢀ-Secretase. J. Med. Chem. 2007,
50, 1124–1132. (c) Geschwindner, S.; Olsson, L.-L.; Albert, J. S.;
Deinum, J.; Edwards, P. D.; deBeer, T.; Folmer, R. H. A. Discovery
of a Novel Warhead against ꢀ-Secretase through Fragment-Based Lead
Generation. J. Med. Chem. 2007, 50, 5903–5911.
(10) Chung, J. L.; Garvey, D. S. Aza-spirocyclic compounds that enhance
cholinergic neurotransmission. US patent 5,221,675, 1993.
(11) A PDB file for the BACE-1/inhibitor 16 complex (PDB identifier
3E3W) has been deposited with the Protein Data Bank (http://
www.rcsb.pdb). Purification of BACE-1 is described in the following:
Sardana, V.; Xu, B.; Zugay-Murphy, J.; Chen, Z.; Sardana, M.; Darke,
P.; Munshi, S.; Kuo, L. C. A general procedure for the purification of
human ꢀ-secretase expressed in E. coli. Protein Expression Purif. 2004,
34, 190-196.
(12) Northrop, D. B. Follow the protons: a low-barrier hydrogen bond
unifies the mechanisms of the aspartic proteases. Acc. Chem. Res. 2001,
34, 790–797.
(7) A biotinylated BACE substrate peptide containing the NFEV cleavage
site (Shi, X. P.; Tugusheva, K.; Bruce, J. E.; Lucka, A,; Chen-Dodson,
E.; Hu, B. H.; Wu, G. X.; Price, E.; Register, R. B.; Lineberger, J.;
Miller, R.; Tang, M. J.; Espeseth, A.; Kahana, J.; Wolfe, A.;
Crouthamel, M. C.; Sankaranarayanan, S. Simon, A.; Chen, L.; Lai,
M. T.; Pietrak, B.; DiMuzio, J. ; Li, Y. M.; Xu, M.; Huang, Q.; Garsky,
V.; Sardana, M. K.; Hazuda, D. J. Novel mutations introduced at the
beta-site of amyloid beta protein precursor enhance the production of
amyloid beta peptide by BACE-1 in vitro and in cells J. Alzheimer’s
Dis. 2005, 7, 139-148) was immobilized to a black, 384-well
microtiter plate that had been coated with streptavidin. After removal
of the unbound peptide, test compounds and BACE enzyme (0.5 nM
final) was added to the plate. Reaction proceeded for 40 min at RT.
Reaction mixture was removed and replace with a solution containing
5 nM of europium-labeled antibody (Perkin Elmer 1244-301), which
recognized the NF-COOH of the cleaved product peptide. After 30
min, the plate was washed 3× and europium enhancement solution
(Perkin Elmer 4001-0010) was added and europium counts quantified
on a VictorV plate reader.
(13) Eder, J.; Hommel, U.; Cumin, F.; Martoglio, B.; Gerhartz, B. Aspartic
proteases in drug discovery. Curr. Pharm. Des. 2007, 13, 271–285.
(14) Coburn, C. A.; Stachel, S. J.; Li, Y.-M.; Rush, D. M.; Steele, T. G.;
Chen-Dodson, E.; Holloway, M. K.; Xu, M.; Huang, Q.; Lai, M.-T.;
DiMuzio, J.; Crouthamel, M.-C.; Shi, X.-P.; Sardana, V.; Chen, Z.;
Munshi, S.; Kuo, L.; Makara, G. M.; Annis, D. A.; Tadikonda, P. K.;
Nash, H. M.; Vacca, J. P.; Wang, T. Identification of a Small Molecule
Nonpeptide Active Site ꢀ-Secretase Inhibitor That Displays a Non-
traditional Binding Mode for Aspartyl Proteases. J. Med. Chem. 2004,
47, 6117–6119.
(15) Kuglstatter, A.; Stahl, M.; Peters, J.-U.; Huber, W.; Stihle, M.;
Schlatter, D.; Benz, J.; Ruf, A.; Roth, D.; Enderle, T.; Hennig, M.
Tyramine fragment binding to BACE-1. Bioorg. Med. Chem. Lett.
2008, 18 1304-1307.
(8) For a description of the in vitro assays, see: Pietrak, B. L.; Crouthamel,
M.-C.; Tugusheva, K.; Lineberger, J. E.; Xu, M.; DiMuzio, J. M.;
Steele, T.; Espeseth, A. S.; Stachel, S. J.; Coburn, C. A.; Graham,
S. L.; Vacca, J. P.; Shi, X.-P.; Simon, A. J.; Hazuda, D. J.; Lai, M.-T.
Biochemical and cell-based assays for characterization of BACE-1
inhibitors Anal. Biochem. 2005, 342, 144-151.
(16) Hochman, J. H.; Pudvah, N.; Qiu, J.; Yamazaki, M.; Tang, C.; Lin,
J.; Prueksaritanont, T. Interactions of Human P-Glycoprotein with
Simvastatin, Simvastatin Acid, and Atorvastatin. Pharm. Res. 2004,
21, 1686–1691.
(9) Oefner, C.; Binggeli, A.; Breu, V.; Bur, D.; Clozel, J. P.; D’Arcy, A.;
Dorn, A.; Fischli, W.; Gru¨ninger, F.; Gu¨ller, R.; Hirth, G.; Ma¨rki,
H. P.; Mathews, S.; Mu¨ller, M.; Ridley, R. G.; Stadier, H.; Vieira, E.;
Wilhelm, M.; Winkler, F. K.; Wostl, W. Renin inhibition by substituted
piperidines: A novel paradigm for the inhibition of monomeric aspartic
proteinases? Chem. Biol. 1999, 6, 127–131.
(17) Lamb, B. T.; Sisodia, S. S.; Lawler, A. M.; Slunt, H. H.; Kitt, C. A.;
Kearns, W. G.; Pearson, P. L.; Price, D. L.; Gearhart, J. D. Introduction
and expression of the 400 kilobase amyloid precursor protein gene in
transgenic mice. Nat. Genet. 1993, 5, 22–30.
JM800914N