Part 1: Notch-sparing γ-secretase inhibitors: The identification of novel naphthyl and benzofuranyl amide analogs

Article abstract of DOI:10.1016/j.bmcl.2016.03.040

γ-Secretase is one of two proteases directly involved in the production of the amyloid β-peptide (Aβ), which is pathogenic in Alzheimer's disease. Inhibition of γ-secretase to suppress the production of Aβ should not block processing of one of its alterna

Full text of DOI:10.1016/j.bmcl.2016.03.040

Bioorganic & Medicinal Chemistry Letters 26 (2016) 2129–2132
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Part 1: Notch-sparing c-secretase inhibitors: The identification
of novel naphthyl and benzofuranyl amide analogs
Dai Lu, Han-Xun Wei, Jing Zhang, Yongli Gu, Pamela Osenkowski, Wenjuan Ye, Dennis J. Selkoe,
⇑
Michael S. Wolfe, Corinne E. Augelli-Szafran
Laboratory for Experimental Alzheimer Drugs (LEAD), Center for Neurologic Diseases, Harvard Medical School and Brigham and Women’s Hospital, 77 Avenue Louis Pasteur,
Harvard Institutes of Medicine, Boston, MA 02115, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
c
-Secretase is one of two proteases directly involved in the production of the amyloid b-peptide (Ab),
which is pathogenic in Alzheimer’s disease. Inhibition of -secretase to suppress the production of Ab
should not block processing of one of its alternative substrates, Notch1 receptors, as interference with
Notch1 signaling leads to severe toxic effects. In the course of our studies to identify -secretase inhibi-
tors with selectivity for APP over Notch, 1 [3-(benzyl(isopropyl)amino)-1-(naphthalen-2-yl)propan-1-
one] was found to inhibit -secretase-mediated Ab production without interfering with -secretase-
Received 2 January 2016
Revised 10 March 2016
Accepted 11 March 2016
Available online 12 March 2016
c
c
c
c
Keywords:
Alzheimer’s disease
mediated Notch processing in purified enzyme assays. As 1 is chemically unstable, efforts to increase
the stability of this compound led to the identification of 2 [naphthalene-2-carboxylic acid benzyl-iso-
propyl-amide] which showed similar biological activity to compound 1. Synthesis and evaluation of a ser-
ies of amide analogs resulted in benzofuranyl amide analogs that showed promising Notch-sparing
secretase inhibitory effects. This class of compounds may serve as a novel lead series for further study
in the development of -secretase inhibitors.
c
-Secretase inhibitors
Naphthyl amides
Notch processing
Ab production
c-
c
Ó 2016 Elsevier Ltd. All rights reserved.
Alzheimer’s disease (AD) is characterized by neurodegeneration
and progressive deterioration of memory and cognitive abilities. At
present, there is no cure or effective treatment for this disease, only
several approved drugs for alleviating certain AD symptoms. Over
the past 20 years, advances in deciphering AD pathology have
revealed that AD is a protein-misfolding disorder,1 as aggregation
of the amyloid b-peptide (Ab) in the brain is a central event in
AD pathology.2 Ab proteins are produced naturally inside the
human brain as proteolytic products of the amyloid precursor pro-
production, prevention of Ab aggregation and promotion of Ab
clearance.5
Suppressing Ab production through inhibition of
has been aggressively pursued as a potential disease-modifying
approach. However, -secretase is a multifunctional protease with
many substrates including the essential cell-signaling Notch recep-
tors.6 Hence, identification of
-secretase inhibitors or modulators
c-secretase
c
c
that can lower Ab production in general or Ab42 in particular with
minimal effects on Notch signaling (particularly that of Notch1)
has become one of the most prominent challenges in the pursuit
of AD therapeutics.3 While much progress has been made toward
tein (APP) through sequential cleavages mediated by b- and c-sec-
retases. Secreted Ab varies in length ranging from 37 to 43 amino
acids,3 of which the 42 amino acid peptide (Ab42) is especially
prone to aggregation when over-produced or normal clearance is
interrupted.2 Ab dimers and higher order assemblies can initiate
a series of cellular events that can ultimately cause neuronal dys-
functions and the onset and progression of AD.4 Based on the Ab
hypothesis of AD-pathogenesis, several disease-modifying
approaches have been proposed, including suppression of Ab
Ab42-lowering
c
c
-secretase modulators,7–9 various ‘Notch-sparing’
-secretase inhibitors have been reported,8,10–13 and even for these
the degree of selectivity for APP versus Notch1 is often unclear.
Thus, there is a keen need to identify new structures with this
important substrate-selective inhibitory property as demonstrated
in comparable biochemical assays.
In search of such compounds, one of our earliest hits was naph-
thylaminoalkyl ketone 1.14 Unfortunately, 1 is unstable and subject
to degradation by a retro-Michael-addition to give the correspond-
ing naphthyl vinyl ketone and N-benzylisopropylamine. One
attempt to seek more stable analogs of 1 was to generate its amide
counterpart 2 (Fig. 1), with the ethylene linker between the
carbonyl and amine nitrogen removed. This compound 2 showed
⇑
Corresponding author at present address: Chemistry Department, Drug Dis-
covery Division, Southern Research, 2000 Ninth Avenue South, Birmingham, AL
35205, United States. Tel.: +1 205 581 2305.
E-mail address: caugelli-szafran@southernresearch.org (C.E. Augelli-Szafran).
http://dx.doi.org/10.1016/j.bmcl.2016.03.040
0960-894X/Ó 2016 Elsevier Ltd. All rights reserved.

2130
D. Lu et al. / Bioorg. Med. Chem. Lett. 26 (2016) 2129–2132
Table 3
O
O
Naphthyl amide analogs with substituted phenyl rings
N
N
R3
O
R4
R5
1
2
N
Figure 1. Naphthyl aminoalkyl ketone 1 and naphthyl amide 2.
R6
Entry
R3
R4
R5
R6
Ab40 (%)
Notch1
inhibitiona
processingb
Table 1
Naphthyl amides
17
18
19
20
21
22
23
24
25
26
27
28
H
OH
H
OH
H
OH
H
OH
H
H
H
H
H
H
H
H
F
F
H
H
CN
H
OMe
OMe
Cl
Cl
F
F
H
H
OMe
OMe
H
CN
H
Cl
Cl
F
F
F
58
41
34
49
21
27
47
33
31
31
35
19
No change
No change
No change
No change
n.t.
O
N
R1
F
n.t.
H
H
H
H
OMe
H
Inhibition
No change
n.t.
n.t.
No change
No change
Entry
R1
Ab40 (%) inhibitiona
Notch1 processingb
2
3
4
5
6
i-Pr
H
CH3
t-Bu
Cyclic-Pr
58
0
15
52
37
No change
n.t.
n.t.
No change
No change
H
N
29
H
H
H
38
Inhibition
n.t. = not tested; see References and notes sections for assay descriptions.19
a
a,b
Compounds were tested at 100
l
M. Inhibitory effects on Ab40 production were
See Table 1 notes. n.t. = not tested.
recorded as a percentage in comparison with DMSO control.
b
Compounds were tested at 100 lM. Effects on Notch processing were recorded
(Western blot) in comparison with DMSO control.
Table 4
Benzofuranoyl amide analogs
R8
O
Table 2
N
O
Aryl amide analogs replacing the 2-naphthyl group
R9
R10
R7
O
Ar
N
Entry R7
R8
R9 R10 Ab40 (%) inhibitiona Notch1 processingb
Entry
2
Ar
Ab40 (%) inhibitiona
Notch1 processingb
No change
30
31
32
33
34
35
36
H
H
H
OMe
OMe
OMe
OEt
H
H
H
Cl
H
40
51
0
75
49
23
58
No change
No change
n.t.
No change
No change
No change
No change
Cl
Cl
Cl
F
Cl
Cl
Me Cl
58
H
H
H
H
Cl
Cl
F
OH
7
67
No change
Cl
8
9
10
85
n.t.
a,b
See Table 1 notes. n.t. = not tested.
Inhibition
a reasonable profile to begin synthesizing analogs. A series of these
amide analogs were synthesized as illustrated in Tables 1–4.
These aryl amides were synthesized by the methods illustrated
in Scheme 1. Alkylation of readily available amines (i.e., isopropyl
or cyclopropyl amine) with various substituted benzyl halides
was carried out at 0 °C to room temperature for 6–8 h. Excess
amine was then removed in vacuo. The crude product contained
ꢀ10–15% of the undesired dialkylated product which was easily
removed by an aqueous–organic partitioning at pH 8–8.5. The
desired mono-alkylated amine (II) was obtained by subsequent
extraction at pH 11–12. Coupling of amine (II) or commercially
available amine with either acyl chlorides or aryl acids yielded
the desired target amide analogs (III).
10
27
n.t.
11
12
14
40
n.t.
O
No change
O
N
13
0
n.t.
CH3
14
15
16
50
33
50
Inhibition
No change
Inhibition
N
O
OH
N
These amides were then evaluated for their inhibitory effects on
Ab40 production from purified human
binant APP-based substrate using a specific ELISA.16–18 Effects on
-secretase processing of a comparable recombinant Notch1-based
c
-secretase15 and a recom-
Cl
c
substrate were examined by Western blot.14 Compounds with
>50% inhibition in the Ab40 ELISA were typically evaluated for
their effects on Notch processing. Since our goal in the early stages
of our program was to identify viable chemical leads, testing our
new analogs at a high concentration in search for at least 50% inhi-
bition of Ab production without effect on Notch1 cleavage led to
some novel compounds and chemical series that are now being
presented here and in subsequent manuscripts.
a,b
See Table 1 notes. n.t. = not tested.
similar biological activity to 1. Evaluation of various drug-like
properties (e.g., solubility, LogD, plasma protein binding,
permeability, and human and rodent microsomal stability—see
Supplemental data) of 2 was completed and the results indicated

D. Lu et al. / Bioorg. Med. Chem. Lett. 26 (2016) 2129–2132
2131
R1
R2
a: R1=Cl, R2=Cl, R3=H, X=Cl, R= i-Pr
b: R1=OMe, R2=H, R3=OMe, X=Br, R= i-Pr
c: R1=F, R2=F, R3=H, X=Br, R= i-Pr
d: R1=H, R2=OMe, R3=H, X=Br, R= i-Pr
e: R1=Cl, R2=F, R3=H, X= Br, R= i-Pr
f: R1=F, R2=H, R3=F, X=Br, R= i-Pr
g: R1=H, R2=CN, R3=H, X=Br, R= i-Pr
h: R1=CN, R2=H, R3=H, X=Br, R= i-Pr
i: R1=H, R2=H, R3=H, X=Br, R= cyclo-Pr
X
RNH2
+
R3
(I) a-i
(a) 3 eq amine (i-Pr alcohol)
CH2Cl2, 00C to rt, 6-8 h;
Extraction at pH 8-8.5;
Extraction at pH 11-12;
80-85%
(b) ArCOCl,Et3N/THF
O
R1
00C to rt, 75-85%
R1
R2
HN
R
or commercial
amine
Ar
N
R
R2
c) ArCOOH, HATU, HOAt,
DIEA, DMF, rt, 29-49%
R3
(II)
R3
(III)
Or ArCONR'R''
(Entry 2-20)
Scheme 1. Preparation of aryl amides.
Initial variations of the isopropyl group of 1 revealed that sub-
stitution on this nitrogen is essential for activity (e.g., 3 vs 2;
Table 1). Moreover, the size of this R1 seems important since 5
(t-butyl) and 6 (cyclopropyl) maintained inhibition of Ab40 pro-
duction (52% and 37%, respectively) while a methyl-substituted
analog (4) was less active (15%).
These preliminary modifications suggested that at least the size
of an iso-propyl group is preferred for the inhibition of Ab40 pro-
duction. Analogs 2, 5, and 6 also showed no effect on Notch pro-
cessing. This was encouraging data and convinced us to then
turn our attention to replacing the naphthyl group with other aryl
groups (Table 2) and investigating various substitution patterns on
the phenyl ring (Table 3).
The results suggested some structural features that may be nec-
essary for exhibiting Notch1-sparing properties and inhibition of
Ab40 production. Notably, when the naphthyl group was replaced
with a phenyl ring, activity was markedly reduced (8, 10%; Table 2).
However, when a 4-t-butyl group was introduced on the phenyl
ring, activity on the inhibition of APP processing was enhanced
(9, 85%), but the desired Notch1-sparing property was lost. These
results suggest that a bulky group or larger ring system may be
needed to make a tight interaction with the binding site. Introduc-
ing a hydroxyl group on aromatic rings adjacent to the amide car-
bonyl group (e.g., 7 and 15) likely leads to the formation of an
intramolecular hydrogen bond between the hydroxyl and the
amide carbonyl, a notion supported by 1H NMR. Two sets of signals
were seen for compounds without the OH, consistent with two
amide rotomeric forms, as typically observed by NMR. In contrast,
only one set of signals was observed for those analogs with the OH,
consistent with conformation restriction due to H-bonding.
From the data shown in Table 2, it appeared that better activity
was observed when the aryl (Ar) group was larger in size consist-
ing of two fused rings or had larger substituents on a ring. And, in
general, fused bicyclic aryl groups appear to be preferred for selec-
tivity versus Notch1-processing.
Additionally, replacing a chlorine substituent with a fluorine sub-
stituent (35) for R9 reduced activity (33 vs 35). It is encouraging
to note that benzofuranyl amides consistently exhibited Notch-
sparing properties while displaying promising inhibitory effects
on
c
-secretase-mediated Ab production.
This early-stage SAR study of naphthyl and benzofuranyl amide
analogs led to diaryl amides that were found to inhibit Ab40 pro-
duction mediated by -secretase. In particular, the naphthalene-
c
2-yl and benzofuran-2-yl amides displayed Notch1-sparing prop-
erties. Additional Ab42 cellular data was obtained for many of
the compounds that exhibited >50% inhibition in the gamma secre-
tase assay (see Supplemental data). Showing weak activity allowed
us to search diligently for a modified diaryl amide core that exhib-
ited increased potency. Thus, converting the amide to a sulfon-
amide and exploring various substation patterns gave analogs
such as 37, 38, and 39, just a few of the examples synthesized
(Fig. 2).
In summary, it was from the discovery of this amide series that
led to sulfonamide analogs (that were potent against gamma-sec-
retase as well as in cells against Ab40 and Ab42). Some of these
analogs (e.g., 39) were later tested in vivo in animals from which
these results will be reported. These findings have the potential
to lead to a treatment for Alzheimer’s disease that ultimately
may show differentiation from previously failed clinical
candidates.
The results reported here were generated while our laboratories
continued to search for additional templates that could be used for
developing Notch1-sparing
c-secretase inhibitors. Findings of
O
F
O
S
O
N
N
F
F
CO2CH3
2
37
58% at 100 μM, Aβ40
γ -secretase inhibition
IC50 = 0.58 μM, Aβ40
γ -secretase inhibition
Analogs with various substitution patterns (R4, R5, and R6) on
the phenyl group were studied and the results are shown in Table 3.
Maintaining a di-chloro substitution pattern at R9 and R10 with R8
equal to hydrogen seemed to yield the better activity profile.
The findings from the above studies were incorporated into
some of the benzofuranyl analogs shown in Table 4.
In this group of compounds, 3,4-dihalogenated analogs also
exhibited good inhibitory effects on Ab production. Interestingly,
introducing a methyl group at the 3-position (R8) on the benzofu-
ran ring completely abolished the activity (32 vs 31). This observa-
tion could suggest that the orientation of the amide bound with
respect to the benzofuran ring may be critical for activity.
O
O
F
O
O
O
O
S
S
N
N
F
F
Cl
Et
39
Et
CO2H
γ
IC
IC
-secretase inhibition:
38
Cellular assays:
50 = 0.36 μM, Aβ40
IC50 = 0.12 μM, Aβ40
50 = 0.40 μM, Aβ42
= 0.13 μM, Aβ42
IC50
Aβ 42 cellular assay:
IC50 = 10 μM, Notch
IC
50 = 0.62 μM, Aβ42
Figure 2. Evolution of Amide 2.

2132
D. Lu et al. / Bioorg. Med. Chem. Lett. 26 (2016) 2129–2132
8. Imbimbo, B. P.; Giardina, G. A. Curr. Top. Med. Chem. 2011, 11, 1555.
9. Oehlrich, D.; Berthelot, D. J.; Gijsen, H. J. J. Med. Chem. 2011, 54, 669.
10. Kreft, A. F.; Martone, R.; Porte, A. J. Med. Chem. 2009, 52, 6169.
11. Olson, R. E.; Albright, C. F. Curr. Top. Med. Chem. 2008, 8, 17.
12. Pissarnitski, D. Curr. Opin. Drug Disc. Dev. 2007, 10, 392.
13. Imbimbo, B. P. Curr. Top. Med. Chem. 2008, 8, 54.
additional identified templates of chemical series are reported in
subsequent manuscripts.
Acknowledgments
14. Fraering, P. C.; Ye, W.; LaVoie, M. J.; Ostaszewski, B. L.; Selkoe, D. J.; Wolfe, M. S.
J. Biol. Chem. 2005, 280, 41987.
15. Fraering, P. C.; Ye, W.; Strub, J. M.; Dolios, G.; LaVoie, M. J.; Ostaszewski, B. L.;
van Dorsselaer, A.; Wang, R.; Selkoe, D. J.; Wolfe, M. S. Biochemistry 2004, 43,
9774.
16. Esler, W. P.; Kimberly, W. T.; Ostaszewski, B. L.; Ye, W.; Diehl, T. S.; Selkoe, D. J.;
Wolfe, M. S. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 2720.
17. Kimberly, W. T.; Esler, W. P.; Ye, W.; Ostaszewski, B. L.; Gao, J.; Diehl, T.; Selkoe,
D. J.; Wolfe, M. S. Biochemistry 2003, 42, 137.
We acknowledge Jian Chen and Katherine M. Brogan for their
assistance with the biological testing. We acknowledge the Alzhei-
mer Drug Discovery Foundation, the Harvard NeuroDiscovery Cen-
ter and members of the Center for Neurologic Diseases for funding
support.
Supplementary data
18. Xia, W.; Zhang, J.; Ostaszewski, B. L.; Kimberly, W. T.; Seubert, P.; Koo, E. H.;
Shen, J.; Selkoe, D. J. Biochemistry 1998, 37, 16465.
19. In vitro
from Chinese Hamster ovary cells stably overexpressing all four components.15
APP-based -secretase substrate (C100-Flag) and Notch-based substrate
c-secretase activity assay: Human c-secretase complex was purified
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2016.03.
040.
c
(N100-Flag) was expressed in Escherichia coli and purified as previously
described.16,17 The tested compounds were dissolved in DMSO, and then
prepared with the final concentration of 100
CHAPSO/HEPES (pH 7.0) with final DMSO concentration less than 1%. The
proteolytic reaction mixtures containing purified -secretase complex, 0.08%
lM in the reaction buffer of 0.2%
References and notes
c
1. Selkoe, D. J. Cold Spring Harb. Perspect. Biol. 2011, 3.
2. Karran, E.; Mercken, M.; De Strooper, B. Nat. Rev. Drug Disc. 2011, 10, 698.
3. Wolfe, M. S. Neurotherapeutics 2008, 5, 391.
4. Galimberti, D.; Scarpini, E. Front. Biosci. (School Ed.) 2011, 3, 252.
5. Mangialasche, F.; Solomon, A.; Winblad, B.; Mecocci, P.; Kivipelto, M. Lancet
Neurol. 2010, 9, 702.
6. Wolfe, M. S. Semin. Cell Dev. Biol. 2009, 20, 219.
7. Pettersson, M.; Kauffman, G. W.; am Ende, C. W.; Patel, N. C.; Stiff, C.; Tran, T. P.;
Johnson, D. S. Expert Opin. Ther. Pat. 2011, 21, 205.
phosphatidylcholine, 0.02% phosphatidylethanol-amine, tested compounds,
and substrate C100-Flag or N100-Flag were then incubated at 37 °C for 4 h. All
reactions were stopped by adding SDS to a final concentration of 0.5%. The
reactions employing the substrate of C100-Flag were evaluated for inhibition
of Ab40 production by ELISA using commercial human Ab40 antibodies
(Invitrogen),18 whereas the reactions using the substrate of N100-Flag were
analyzed by Western blots for effects on Notch1 processing as previously
reported.14