Macrocyclic inhibitors of β-secretase: Functional activity in an animal model

Article abstract of DOI:10.1021/jm060884i

A macrocyclic inhibitor of β-secretase was designed by covalently cross-linking the P1 and P3 side chains of an isophthalamide-based inhibitor. Macrocyclization resulted in significantly improved potency and physical properties when compared to the initia

Full text of DOI:10.1021/jm060884i

J. Med. Chem. 2006, 49, 6147-6150
6147
development of a series of macrocyclic active site inhibitors of
BACE-1. Most noteworthy, compound 19 was shown to reduce
Aâ levels in a murine model following iv administration.
Macrocyclic Inhibitors of â-Secretase: Functional
Activity in an Animal Model
Shawn J. Stachel,*,† Craig A. Coburn,†
Sethu Sankaranarayanan,‡ Eric A. Price,‡ Beth L. Pietrak,‡
Qian Huang,‡ Janet Lineberger,‡ Amy S. Espeseth,‡
Lixia Jin,§ Joan Ellis,§ M. Katharine Holloway,|
Sanjeev Munshi, Timothy Allison, Daria Hazuda,‡
Adam J. Simon,‡ Samuel L. Graham,† and Joseph P. Vacca†
Department of Medicinal Chemistry, Biological Chemistry,
Molecular Systems and Structural Biology Merck Research
Laboratories, Post Office Box 4, West Point, PennsylVania 19486
Figure 1. Representative potent BACE-1 inhibitors.
ReceiVed July 26, 2006
5-Substituted isophthalamides, as represented by structures
1 and 2 (Figure 1), emerged as early lead classes that were
investigated for inhibition of BACE-1.8 Both inhibitors pos-
sessed desirable properties of both good in vitro (IC50 ) 15
Abstract: A macrocyclic inhibitor of â-secretase was designed by
covalently cross-linking the P1 and P3 side chains of an isophthalamide-
based inhibitor. Macrocyclization resulted in significantly improved
potency and physical properties when compared to the initial lead
structures. More importantly, these macrocyclic inhibitors also displayed
in vivo amyloid lowering when dosed in a murine model.
and 7 nM, respectively) and good cell-based potencies (IC50
)
29 and 20 nM, respectively) against BACE-1. However,
problems associated with their pharmacokinetic properties
rendered them less than ideal candidates for advancement. A
main concern was that 1 and 2 were found to be substrates of
P-glycoprotein (P-gp) transport and both exhibited low apparent
permeability in wild-type porcine LLC-PK1 cells. While iv
administration of either 1 or 2 showed no effect on Aâ levels
in a murine model, intracranial administration of 1 succeeded
in reducing Aâ levels by 50-70%.9 This result demonstrated
that inhibition of BACE-1 can reduce Aâ levels in vivo, but to
be clinically relevant, the compound must be CNS penetrant.
To achieve greater CNS penetration, we sought to reduce the
peptidic character of our inhibitors while maintaining potency.
Macrocyclization is a well-precedented technique to enhance
potency by conformational preorganization, and in many cases,
the physical properties of the inhibitors were also improved.10
We envisaged that, if macrocyclization were successful in
enhancing potency, reduction of molecular weight and removal
of some hydrogen bonding elements might be possible. Several
research groups have recently reported similar macrocyclization
strategies for BACE inhibitors, albeit with limited success.11
In our case, examination of the X-ray crystallographic data of
112 complexed in the active site of BACE-1 revealed the close
spatial proximity of the lipophilic P1 and P3 groups and thus
directed efforts toward the design of macrocycles that would
link these groups in an isophthalamide-based inhibitor.
We chose 4 as a simplified core to test whether macrocy-
clization would provide an increase in potency due to a reduction
in conformational freedom (Figure 2). We were gratified to
discover that while 4 was not a high affinity ligand for BACE-1
(IC50 ) 2.9 µM) it was vastly superior to the corresponding
seco variant 3 (IC50 > 100 µM). It should be noted that
compound 4 was designed as a model system and thus lacks
several potency enhancing features paramount for high binding
affinity.
Alzheimer’s disease (ADa) is a debilitating neurodegenerative
disease characterized by progressive cognitive decline, inevitably
leading to incapacitation and death. Although researchers have
uncovered a variety of genetic mutations that result in early
onset AD over the past 20 years, our current understanding of
genetic predisposition accounts for <10% of AD cases.1 The
majority of AD cases are sporadic in nature, and the risk factor
increases in an age-dependent manner. It is generally recog-
nized that â-amyloid plaques and neurofibrillary tangles are
the key pathological features of the disease, although the spe-
cific roles of these agents in disease progression remain
debatable. However, it is well-established that the principle
component of amyloid plaque is a neurotoxic peptide frag-
2
ment of the amyloid precursor protein (APP), namely, Aâ40-42
.
As such, many therapeutic strategies are founded on the ability
to inhibit catabolic pathways of APP that result in Aâ forma-
tion.3
Proteolytic processing of APP is primarily accomplished
through the activity of three discrete proteases (R-, â-, and
γ-secretase).4 The initial cleavage event is performed by either
R-secretase or â-secretase (BACE-1) releasing the APP
ectodomain. The remaining membrane-bound C-terminal do-
mains are subsequently cleaved by γ-secretase, resulting in the
nonamyloidogenic P3 peptide (R-cleavage product) or the
amyloidogenic Aâ40-42 (â-cleavage product). γ-Secretase in-
hibitors have been demonstrated to inhibit the production of
Aâ in vivo, however, their development as therapeutic agents
remains tenuous due to the small therapeutic window that results
from untoward effects on notch processing.5 In contrast,
inhibitors of BACE-1 are predicted to be well-tolerated, as
evidenced by viable knockout mice.6 As such, BACE-1 inhibi-
tion is considered an attractive therapeutic target for the
treatment and prevention of AD.7 In this letter, we describe the
* To whom correspondence should be addressed. Phone: 216-652-2273.
Fax: 215-652-3971. E-mail: shawn_stachel@merck.com.
† Department of Medicinal Chemistry.
‡ Department of Biological Chemistry.
§ Department of Drug Metabolism.
| Department of Molecular Systems.
Department of Structural Biology.
a Abbreviations: AD, Alzheimer’s disease; APP, amyloid precursor
protein; BACE-1, beta-site APP cleaving enzyme 1; APP-YAC, amyloid
precursor protein-yeast artificial chromosome.
Figure 2. Macrocyclization results in increased potency.
10.1021/jm060884i CCC: $33.50 © 2006 American Chemical Society
Published on Web 09/01/2006

6148 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 21
Letters
Scheme 1a
Figure 3.
functionality acts mainly to orient the P3 ligand, and the amide
hydrogen bonding contacts do not significantly contribute to
the binding energy. With this in mind, we posited that the
conformational constraint induced by the macrocycle would
allow for excision of the amide bond without abrogating
potency. It was also hoped that removal of the amide bond
would result in improved physical properties, namely, cell-
permeability. To test this hypothesis, we synthesized the
corresponding alkyl-linked 15-membered macrocycle 11, but
were disappointed by a greater than 34-fold loss in intrinsic
potency, IC50 ) 1100 nM (Figure 3).
a Reaction conditions: (a) m-tyrosine ethyl ester BOP, DIPEA, CH2Cl2,
86%; (b) tert-butyl(2-iodoethyl)carbamate, K2CO3, DMF, 75%; (c) Pearl-
man’s catalyst, MeOH, 100%; (d)TFA, CH2Cl2, 100%; (e) BOP, DIPEA,
CH2Cl2, 37%; (f) LiBH4, THF, 51%; (g) SO3-pyridine, TEA, DMSO; (h)
(S)-2-amino-N-isobutylbutanamide, NaBH3CN, MeOH, 60%.
Recognizing that the 15-membered macrocyclic ring size for
the bisamide-containing macrocycle 10 may not be optimal for
the carbon analogue, we synthesized the ring-contracted 14-
membered macrocycle 12 for comparison. We were gratified
to see that potency was recovered and 12 exhibited a potency
of 90 nM when assayed against BACE-1. Even more impor-
tantly, activity for 12 in the cell-based assay (IC50 ) 1139 nM)
as well as apparent permeability (Papp ) 19 × 10-6 cm/s) were
also improved when compared with 10. With these encouraging
results, we employed some potency enhancing modifications
to the sulfonamide and P1′ regions of the inhibitor learned from
previous studies culminating in the synthesis of macrocycle 19
as a single diastereomer.14
An efficient synthesis of the alkyl macrocycle 19 was devised
as depicted in Scheme 2. The synthesis began with the iodoarene
13 that we recently reported as an intermediate in alternative
series of inhibitors.13 Allylation of 13 under Stille conditions
and subsequent hydrolylsis of the methyl ester provide benzoic
acid 14. Carboxylic acid 14 was coupled with (S)-m-allyl
phenylalanine 15 using standard conditions to provide the
macrocyclic precursor 16. Macrocyclization of 16 was ac-
complished through the use of second-generation Grubbs
catalyst15 20 to provide macrocycle 17 as a single regioisomer.
The metathesis reaction was high yielding and resulted in
exclusive formation of the E-isomer, as confirmed by 1H NMR.
Hydrogenation of the olefin followed by further elaboration of
the macrocycle was done in a manner similar to 10. Reduction
of the ester using LiBH4 provided alcohol 18, and Parik-
Doering oxidation followed by reductive amination with N-
isobutyl-L-norleucinamide afforded the desired alkyl-linked
macrocycle 19.
Based on the encouraging results of the model system, we
focused on incorporating a prime side element to enhance
binding. We chose to incorporate a ψ[CH2NH]-reduced amide
isostere as present in 2, because we have previously shown that
such isosteres are not only highly active pharmacophores for
BACE inhibition, but also improved cell permeability in similar
series.8b
The synthesis of macrocycle 10 began with a coupling
reaction between isophthalamide 5 and m-tyrosine methyl ester
to produce phenol 6 (Scheme 1). Alkylation of the phenol using
tert-butyl(2-iodoethyl)carbamate afforded the differentially pro-
tected macrocyclic precursor 7 in 75% yield. Deprotection of
the benzyl ester, followed by removal of the amino BOC
protecting group, afforded seco-acid 8 in near quantitative yield.
seco-Acid 8 was subjected to macrocyclization mediated by
BOP reagent under high dilution conditions to provide 9 in
modest yield. The ester was reduced to the alcohol 4 with
lithium borohydride and subsequently oxidized to the corre-
sponding aldehyde using Parikh-Doering conditions. Reductive
amination of the corresponding aldehyde with (S)-2-amino-N-
isobutylbutanamide afforded macrocyclic inhibitor 10.
Macrocycle 10, as a 1:1 mixture of diastereomers at the
tyrosinyl center, was assayed for BACE-1 activity and exhibited
an IC50 ) 32 nM. This represents a 90-fold increase in potency
due to addition of the prime-side binding domain. It should be
noted that macrocycle 10 is only 4-fold less-potent than 2 and
lacks the P3 phenyl substituent that has been shown to contribute
upward of 90-fold in potency. Unfortunately, when 10 was
evaluated in the cell-based assay, there was a significant loss
in potency (IC50 ) 5400 nM) when compared with 2. It was
also suspected that 10 was still a severe P-gp substrate, but due
to a low apparent permeability (0.5 × 10-6 cm/s), its suscep-
tibility could not be accurately determined. Despite the ability
to concomitantly increase potency and trim the molecular weight
of the inhibitor, it was apparent that to succeed in lowering Aâ
in vivo it would be necessary to improve the cell permeability
of our inhibitors.
When assayed against BACE-1 enzyme, macrocycle 19
displayed an IC50 ) 4 nM. This represents a greater than 8-fold
increase in intrinsic activity when compared to compound 10.
Even more impressive, in the cell-based assay, 19 displayed an
IC50 ) 76 nM, representing a 70-fold improvement over 10 in
functional activity. Most gratifying, the apparent permeability
of 19 compared with 10 was significantly improved (Papp ) 13
× 10-6 cm/s). With the improved permeability and reduced P-gp
In a previous communication, we demonstrated that the P2/
P3 amide is noncritical for BACE-1 activity.13 The amide

Letters
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 21 6149
Scheme 2a
Figure 5. Brain compound levels versus Aâ levels.
in Aâ levels, and a statistically significant reduction was
observed when brain concentrations of >700 nM were achieved
(Figure 5).
In summary, we have developed a series of macrocyclic
inhibitors of BACE by covalently cross-linking the P1 and P3
side-chains of an isophthalamide-based inhibitor. Macrocycliza-
tion resulted in significantly improved physical properties when
compared to the initial lead structure. Macrocyclic inhibitors
such as 19 also displayed in vivo amyloid lowering when dosed
in a murine model.
a Reaction conditions: (a) Bu3SnAllyl, Pd(PPh3)4, 82%; (b) NaOH,
MeOH/THF, 83%; (c) 15, BOP, DIPEA, 88%; (d) Grubbs2, CH2Cl2, 50
°C, 84%; (e) Pd/C, H2, MeOH, 93%; (f) LiBH4, THF, 99%; (g) SO3-
pryidine, TEA, DMSO; (h) N-isobutyl-L-norleucinamide, NaBH3CN, MeOH,
71%, 2 steps.
Supporting Information Available: Experimental procedures
and compound characterization data. This material is available free
of charge via the Internet at http://pubs.acs.org.
References
(1) (a) Selkoe, D. J. Translating cell biology into therapeutic advances
in Alzheimer’s disease. Nature 1999, 399A, 23. (b) Goate, A.;
Chartier-Harlin, M. C.; Mullan, M.; Brown, J.; Crawford, F.; Findani,
L.; Giuffra, L.; Haynes, A.; Irving, N.; James, L.; Mant, R.; Newton,
P.; Rooke, K.; Roques, P.; Talbot, C.; Pericak, V.; Roses, A.;
Williamson, R.; Rossor, M.; Owen, M.; Hardy, J. Segregation of
missense mutation in the amyloid presursor protein gene with familial
Alzheimer’s disease. Nature 1991, 349, 523-527. (c) Hardy, J. The
Alzheimer family of diseases: many etiologies, one pathogenesis?
Proc. Natl. Acad. Sci. U.S.A. 1997, 349, 704-706.
(2) Sinha, S.; Lieberburg, I. Cellular mechanisms of â-amyloid production
and secretion. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 11049-11053.
(3) Beher, D.; Graham, S. Protease inhibitors as potential disease-
modifying therapeutics for Alzheimer’s disease. Expert Opin. InVest.
Drugs 2005, 14, 1385-1409.
(4) Selkoe, D. J. Amyloid â-protein and the genetics of Alzheimer’s
disease. J. Biol. Chem. 1996, 271, 18295-18298.
Figure 4. In vivo Aâ40 lowering of 100 mpk iv does of 19 at 1 and
3 h.
susceptibility (BA/AB ) 5.5 mdr1a), we dosed macrocycle 19
at 100 mpk in the APP-YAC16 mouse model to obtain a more
relevant physiological determination of compound levels in the
brain versus plasma. The ratio of compound concentration
between the brain and the plasma was determined to be 20-
30%, and compound concentrations of 19 in the brain were
measured at ∼1100 nM. With these encouraging results, we
undertook a more robust study to evaluate efficacy. APP-YAC
mice were administered a 100 mpk iv bolus of 19 and euthanized
after 1 and 3 h, respectively, to determine functional efficacy.
Analysis of the DEA brain extracts demonstrated a robust
decrease in Aâ(40)17,18 levels of 25% when compared to vehicle.
To our knowledge, this is the first reported BACE-1 inhibitor
to demonstrate efficacy from an iv administration.19 In the study,
the Aâ reduction was statistically significant in the 1 h grouping
compared to the vehicle (Figure 4). However, by the 3 h
timepoint, reduction of Aâ had begun to abate and only a 10%
reduction was measurable in the group.20 Analysis of the
concentration of 19 in the brain correlated well with reduction
(5) (a) Searfoss, G. H.; Jordan, W. H.; Calligaro, D. O.; Galbreath, E.
J.; Schirtzinger, L. M.; Berridge, B. R.; Gao, H.; Higgins, M. A.;
May, P. C.; Ryan, T. P. Adipsin: A biomarker of gastrointestinal
toxicity mediated by a functional γ-secretase inhibitor. J. Biol. Chem.
2003, 278, 46107-46116. (b) Milano, J.; McKay, J.; Dagenais, C.;
Foster-Brown, L.; Pognan, F.; Gadient, R.; Jacobs, R. T.; Zacco, A.;
Greenberg, B.; Ciaccio, P. J. Modulation of notch processing by
γ-secretase inhibitors causes intestinal goblet cell metaplasia and
induction of genes known to specify gut secretory lineage differentia-
tion. Toxicol. Sci. 2004, 82, 341-358.
(6) (a) Cai, H.; Wang, Y.; McCarthy, D.; Wen, H.; Borchelt, D. R.; Price,
D. L.; Wong, P. C. BACE1 is the major beta-secretase for generation
of Abeta peptides by neurons. Nat. Neurosci. 2001, 4, 233-234. (b)
Roberds, S. L.; Anderson, J.; Basi, G.; Bienkowski, M. J.; Branstetter,
D. G.; Chen, K. S.; Freedman, S.; Frigon, N. L.; Games, D.; Hu, K.;
Johnson-Wood, K.; Kappenman, K. E.; Kawabe, T.; Kola, I.; Kuehn,
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.; Sihna, S.; Svensson, K. A.; Tatsuno, G.; Tintrup, H.; Wijsman,
J.; Wright, S.; McConlogue, L. BACE knockout mice are healthy
despite lacking the primary â-secretase activity in the brain. Hum.
Mol. Genet. 2001, 10, 1317-1324.
(7) Thompson, L. A.; Bronson, J. J.; Zusi, F. C. Progress in the discovery
of BACE inhibitors. Curr. Pharm. Des. 2005, 11, 3383-3404.

6150 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 21
Letters
(8) (a) Stachel, S. J.; Coburn, C. A..; Steele, T. G.; Jones, K. G.;
Loutzenhiser, E. F.; Gregro, A. R.; Rajapakse, H. A.; Lai, M.-T.;
Crouthamel, M.-C.; Xu, M.; Tugusheva, K.; Lineberger, J. E.; Pietrak,
B. L.; Espeseth, A. S.; Shi, X.-P.; Chem-Dodson, E.; Holloway, M.
K.; Munshi, S.; Simon, A. J.; Lawrence, K.; Vacca, J. P. Structure-
Based Design of Potent and Selective Cell-Permeable Inhibitors of
Human â-Secretase (BACE-1). J. Med. Chem. 2004, 47, 6447-6450.
(b) Coburn, C. A.; Stachel, S. J.; Jones, K. G.; Steele, T. G.; Rush,
D. M.; DiMuzio, J.; Pietrak, B. L.; Lai, M.-T.; Huang, Q.; Lineberger,
J.; Jin, L.; Munshi, S.; Holloway, M. K.; Espeseth, A.; Simon, A. J.;
Hazuda, D.; Graham, S. L.; Vacca, J. P. BACE-1 inhibition by a
series of Ψ[CH2NH] reduced amide isosteres Bioorg. Med. Chem.
Lett. 2006, 16, 3635-3638.
(9) Simon, A. J.; Crouthamel, M. C.; Wu, G.; Price, E. A.; Shi, X. P.;
Tugusheva, K.; Stachel, S.; Steele, T.; Coburn, C.; Ellis, J.; Jin, L.;
Sankaranarayanan, S. BACE1 inhibition after ICV infusion of a
potent, cell-permeable small molecule inhibitor: effects of APP
metabolism. Presented at 2005 AD/PD meeting, Sorrento, Italy, 2005.
(10) (a) Tsantrizos, Y. S.; Bolger, G.; Bonneau, P.; Cameron, D. R.;
Goudreau, N.; Kukolj, G.; LaPlante, S. R.; Llinas-Brunet, M.; Nar,
H.; Lamarre, D. Macrocyclic inhibitors of the NS3 protease as
potential therapeutic agents of hepatitis C virus infection. Angew.
Chem., Int. Ed. 2003, 42, 1356-1360. (b) Hu, X.; Nguyen, K. T.;
Verlinde, C. L. M. J.; Hol, W. G. J.; Pei, D. Structure-based design
of a macrocyclic inhibitor for peptide deformylase. J. Med. Chem.
2004, 47, 4941-4949.
(11) (a) Rojo, I.; Martin, J. A.; Broughton, H.; Timm, D.; Erickson, J.;
Yang, H. C.; McCarthy, J. Macrocyclic peptidomimetic inhibitors
of â-secretase (BACE): First X-ray structure of a macrocyclic
peptidomimetic-BACE complex. Bioorg. Med. Chem. Lett. 2006,
16, 191-195. (b) Hannessian, S.; Yang, G.; Rondeau, J.-M.;
Neumann, U.; Betschart, C.; Tintelnot-Blomley, M. Structure-based
design and synthesis of macroheterocyclic peptidomimetic inhibitors
of the aspartic protease b-site amyloid precursor protein cleaving
enzyme (BACE) J. Med. Chem. 2006, 49, 4544-4567.
(12) The PDB file for BACE-1/inhibitor 1 complex has been deposited
with the Protein Data Bank (PDB identifier 2B8L).
(13) Stachel, S. J.; Coburn, C. A.; Steele, T. G.; Crouthamel, M.-C.;
Pietrak, B. L.; Lai, M.-T.; Holloway, M. K.; Munshi, S. K.; Graham,
S. L.; Vacca, J. P. Bioorg. Med. Chem. Lett. 2006, 16, 641-
644.
(14) Brady, S. F.; Singh, S.; Crouthamel, M.; Holloway, M. K.; Coburn,
C. A.; Garsky, V. M.; Bogusky, M.; Pennington. M. W.; Vacca,
J. P.; Hazuda, D.; Lai, M. Rational design and synthesis of
selective BACE-1 inhibitors. Bioorg. Med. Chem. Lett. 2004, 14,
601-604.
(15) Scholl, M.; Ding, S.; Lee, C. W.; Grubbs, R. H. Synthesis and activity
of a new generation ruthenium-based olefin metahesis catalyst
coordinated with 1,3-dimesityl-4,5-dihydroimidazol-2-ylidene ligands.
Org. Lett. 1999, 1, 953-956.
(16) 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. Nat.
Genet. 1993, 5, 22-29.
(17) Best, B. T.; Jay, M. T.; Out, F.; Ma, J.; Nadin, A.; Ellis, S.; Lewis,
H. D.; Pattison, C.; Reilly, M.; Harrison, T.; Sherman, M. S.;
Williamson, T. L.; Atack, J. R. J. Quantitative measurement of
changes in amyloid-â(40) in the rat brain and cerebrospinal fluid
following treatment with the γ-secretase inhibitor LY-411575 [N2-
[(2S)-2-)3,5-difluorophenyl)-2-hydroxyethanoyl-N1-[(7S)-5-methyl-
6-oxo-6,7-dihydro-5H-dibenzo[b,d]azepin-7-yl]-l-alaninamide] Phar-
macol. Exp. Ther. 2005, 313, 902-908.
(18) Savage, M. J.; Trusko, S. P.; Howland, D. S.; Pinsker, L. R.; Mistretta,
S.; Reaume, A. G.; Greenberg, B. D.; Siman, R.; Scott, R. W.
Turnover of amyloid â-protein in mouse brain and acute reduc-
tion of its level by phorbol ester J. Neurosci. 1998, 18, 1743-
1752.
(19) Biological data of a closely related analogue in this class was recently
disclosed at the 10th International conference for Alzheimer’s disease
and related disorders. Sankaranarayanan, S.; Price, E. A.; Wu, G.;
Crouthamel, M. C.; Ellis, J. D.; Jin, L.; Stachel, S. J.; Graham, S.
L.; Vacca, J. P.; Coburn, C. A.; Simon, A. J. Presented at the 10th
International conference for Alzheimer’s disease and related disorders,
Madrid, Spain, July, 2006; poster #3050.
(20) The turnover of Aâ is reported to be approximately 2 h. This is in
accord with the rapid recovery when compound levels fall to
subefficacious levels.
JM060884I