Effect of helical conformation and side chain structure on γ-secretase inhibition by β-peptide foldamers: Insight into substrate recognition

Article abstract of DOI:10.1021/jm301306c

Substrate-selective inhibition or modulation of the activity of γ-secretase, which is responsible for the generation of amyloid-β peptides, might be an effective strategy for prevention and treatment of Alzheimer's disease. We have shown that helical β-pe

Full text of DOI:10.1021/jm301306c

Article
pubs.acs.org/jmc
Effect of Helical Conformation and Side Chain Structure on
γ‑Secretase Inhibition by β‑Peptide Foldamers: Insight into Substrate
Recognition
Yuki Imamura,† Naoki Umezawa,*,† Satoko Osawa,‡ Naoaki Shimada,§ Takuya Higo,§
Satoshi Yokoshima,§ Tohru Fukuyama,§ Takeshi Iwatsubo,‡,∥,⊥ Nobuki Kato,† Taisuke Tomita,*,‡,∥
and Tsunehiko Higuchi*,†
†Department of Bioorganic-Inorganic Chemistry, Graduate School of Pharmaceutical Sciences, Nagoya City University, 3-1
Tanabe-dori, Mizuho-ku, Nagoya, Aichi, Japan
‡Department of Neuropathology and Neuroscience, Graduate School of Pharmaceutical Sciences, The University of Tokyo, 7-3-1
Hongo, Bunkyo-ku, Tokyo, Japan
§Department of Synthetic Natural Products Chemistry, Graduate School of Pharmaceutical Sciences, The University of Tokyo, 7-3-1
Hongo, Bunkyo-ku, Tokyo, Japan
∥Core Research for Evolutional Science and Technology (CREST), Japan Science and Technology Agency, Tokyo, Japan
⊥Department of Neuropathology, Graduate School of Medicine, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo, Japan
S
* Supporting Information
ABSTRACT: Substrate-selective inhibition or modulation of
the activity of γ-secretase, which is responsible for the
generation of amyloid-β peptides, might be an effective
strategy for prevention and treatment of Alzheimer’s disease.
We have shown that helical β-peptide foldamers are potent and
specific inhibitors of γ-secretase. Here we report identification
of target site of the foldamers by using a photoaffinity probe.
The photoprobe directly and specifically labeled the N-
terminal fragment of presenilin 1, in which the initial substrate
docking site is predicted to be located. We also optimized the
foldamer structure by preparing a variety of derivatives and
obtained two highly potent foldamers by incorporation of a
hydrophilic and neutral functional group into the parent
structure. The class of side chain functional group and the position of incorporation were both important for γ-secretase-
inhibitory activity. The substrate selectivity of the inhibitory activity was also quite sensitive to the class of side chain group
incorporated.
endoproteolytic fragments of PS (NTF and CTF, respectively).
A series of chemical−biological experiments indicated that
reported GSIs can be classified into three categories based on
the main enzymatically functional sites,5 namely, the catalytic
site (transition-state analogue),9,10 the initial substrate docking
site (helical peptide),11,12 and the transition path or allosteric
site.13,14
Conventional GSIs, however, have little therapeutic potential.
γ-Secretase cleaves a wide variety of other substrates in addition
to APP, including APP-like proteins 1 and 2, N- and E-
cadherins, ErbB4, and Notch receptors.15,16 Proteolysis of the
Notch transmembrane domain by γ-secretase is an essential
part of its signaling pathway, and blocking of this process by
GSIs causes severe adverse effects, including gastrointestinal
INTRODUCTION
■
Several lines of evidence suggest that aggregation and
deposition of amyloid-β peptides (Aβ) underlie the patho-
genesis of Alzheimer’s disease (AD).1 Aβ is a proteolytic
fragment of amyloid precursor protein (APP) formed via
sequential cleavages by two proteases, β- and γ-secretases. γ-
Secretase determines the C-terminal length of Aβ, which in
turn impacts on the aggregation properties of Aβ; the C-
terminally longer Aβ42 is the initially deposited and most
aggregation-prone species and is linked to the pathogenesis of
AD.2 Thus, γ-secretase has been a prime target for drug
discovery, and many γ-secretase inhibitors (GSIs) have been
developed.3−7
γ-Secretase is a complex of four integral membrane proteins,
namely, presenilin (PS), nicastrin, Aph-1, and Pen-2.8 PS serves
as the catalytic subunit of γ-secretase, with a pair of catalytic
aspartate residues being located on the N- and C-terminal
Received: June 30, 2012
Published: January 23, 2013
© 2013 American Chemical Society
1443
dx.doi.org/10.1021/jm301306c | J. Med. Chem. 2013, 56, 1443−1454

Journal of Medicinal Chemistry
Article
toxicity and immunosuppression.17,18 Thus, recent efforts in
drug discovery have focused on identification of γ-secretase
modulator and Notch-sparing GSIs, which would influence Aβ
formation without affecting Notch processing.3−7 γ-Secretase
modulators do not change total Aβ production but rather shift
the spectrum of generated Aβ species toward shorter forms,
which are more soluble and less pathogenic. On the other hand,
Notch-sparing GSIs would inhibit cleavage of APP in a
substrate-selective manner, blocking the production of all Aβs
while allowing Notch proteolysis to occur.
of foldamer 1 derivatives with various side chain functional
groups. Our findings provide information on substrate
recognition by γ-secretase and are expected to be useful for
the design of substrate-selective GSIs.
RESULTS AND DISCUSSION
■
Shape of Helix. β-Peptides, which are oligomers of β-amino
acids, represent an attractive unnatural scaffold because they
adopt predictable helical conformations and resist proteolysis.26
Our γ-secretase-inhibitory β-peptide foldamers were composed
of (S,S)-2-aminocyclopentanecarboxylic acid (ACPC, Figure
1a) and form a stable 12-helix. The β-peptide 12-helix is
defined by CO(i)→H−N(i+3) hydrogen bonds that involve
12 atoms, and its formation is promoted by the use of β-
residues with a five-membered ring constraint. Among the
foldamers we synthesized, the (S,S)-ACPC dodecamer 1
showed the most potent inhibitory activity. The β-peptide
12-helix is a mimic of the α-helices seen in conventional
peptides and proteins.27,28 The internal hydrogen bond
orientation and macrodipole are analogous to those of α-
peptide helices (α-helix and 310-helix) but different from those
of other β-peptide helices.21 The β-peptide 12-helix repeats
approximately every 2.5 residues.21 The rise-per-turn of β-
peptide 12-helix is 5.4 Å, which is about the same as the pitch of
the α-helix and slightly smaller than that of the 310-helix (5.8
Å).27−29 Thus, the β-peptide 12-helix seems to be the most
appropriate candidate for an α-helix mimic, but nevertheless,
other types of helical motif might also be potential GSIs. To
examine this idea, we designed and synthesized a series of
oligoproline peptides that are known to form a well-defined
left-handed polyproline II (PPII) helix in polar solvents.27
Polyproline display helical conformations in the absence of
hydrogen bonding; PPII contains three residues per turn,
aligning every third ring on the same face of the helix with a
pitch of approximately 10 Å per turn.27
Recently, we have identified several foldamers as GSIs.12
Foldamers are sequence-specific oligomers of non-natural
building blocks that adopt well-defined secondary and tertiary
structures.19−21 Helical β-peptide foldamers can mimic the
structure of a transmembrane domain of C99, which is a direct
γ-secretase substrate, and as a result, they potently inhibit γ-
secretase activity.12 Foldamer 1 is the most potent GSI among
the foldamers we have synthesized (Figure 1b). Despite its
Figure 1. Structures of (a) (S,S)-ACPC and (b) foldamer 1.
simple structure, foldamer 1 did not inhibit membrane-bound
metalloproteases ADAM9, -10, and -17 or signal peptide
peptidase (SPP). Foldamers showed moderate substrate
selectivity in a fashion similar to that of Notch-sparing GSIs.
We considered that foldamer 1 is a potential lead compound
for the development of APP-cleaving γ-secretase activity-
specific inhibitors.
As substrate-based inhibitors that target the initial substrate
docking site, helical peptides containing multiple helix-inducing
Aib (2-aminoisobutyric acid) residues have been reported.22,23
We found that foldamers compete with Aib-containing peptide
for binding to PS1 NTF, where the initial substrate docking site
is located, suggesting that the foldamers may also target the
initial binding site or allosterically affect the structure of PS1.
Both Aib-containing L- and D-peptides potently inhibit γ-
secretase activity, but in either case structural modifications that
disrupt the helical conformation result in a dramatic reduction
of inhibitory potency. A similar tendency was observed in the
case of foldamers.12 However, no systematic investigation on
the effect of side chain functional groups has been reported.
The helical peptides contained only a few different amino acids,
such as valine, isoleucine, and threonine.22,23 In general, short
linear peptides tend not to form a stable helix because of their
high intrinsic flexibility. Although the Aib residue is known to
stabilize helical structure, formation of a highly stable helix is
still difficult.24,25 Thus, it would be problematic to incorporate
side chain functional groups at specific positions of helical
peptides. However, since even short foldamers have strong
tendency to form a stable helix, helical foldamers should be an
appropriate platform to examine the effect of side chain
functional groups.
The synthesized oligoprolines are shown in Figure 2a.
Oligomers of L- and D-proline were both synthesized. Circular
dichroism (CD) spectra were measured in methanol (Figure
2b). The spectra of L-peptides 2−4 were all similar, showing
minima at 205 nm and maxima at 228 nm, which are typical of
Here, we examined in detail the properties of foldamers as
GSIs by synthesizing a series of compounds related to 1 in
order to identify the optimum shape of the helix, the optimum
length of the foldamer, and the enzyme target site of 1. We also
report the design, synthesis, and γ-secretase-inhibitory activity
Figure 2. (a) Structures of oligoproline peptides. Peptides 2−4 are
composed of L-proline, and 5−7 are composed of D-proline. (b) CD
spectra of oligoproline peptides in methanol. The molar ellipticity [Θ]
values have been normalized for peptide concentration and the
number of backbone amide groups.
1444
dx.doi.org/10.1021/jm301306c | J. Med. Chem. 2013, 56, 1443−1454

Journal of Medicinal Chemistry
Article
the left-handed PPII helix.30 The increase in intensity at 228
nm with addition of further proline residues indicates that the
population of the PPII helical state increases as the peptide
length increases. D-Peptides 5−7 had mirror-image CD spectra
compared with their enantiomers 2−4 within experimental
error, corresponding to a right-handed PPII helix.
The inhibitory potency of these oligoproline peptides was
analyzed using an in vitro γ-secretase assay (Table 1).31 None
of the oligoprolines showed inhibitory activity, suggesting that
12-helical structure is important for γ-secretase inhibition.
addition of further ACPC residues indicates that the population
of 12-helical state increases; i.e., the stability of this helical state
increases as the length of the foldamer is increased. Foldamers
1 (12mer) and 9 (11mer) showed similar spectra, suggesting
that the 11mer is sufficiently long to achieve a high population
of the 12-helical state. In order to clarify the inhibitory potency
precisely, the IC50 values were measured using an in vitro
assay.31 As shown in Table 2, foldamer 1 was the most potent
Table 2. Inhibitory Potency of Foldamers toward γ-Secretase
in in Vitro Assay (n = 3)
Table 1. Inhibitory Potency of Oligoproline Peptides toward
γ-Secretase in in Vitro Assay (n = 3)
IC50 (nM)
compd
no. of β-amino acids
Aβ40
Aβ42
IC50 (nM)
1
8
9
12
10
11
8.81 1.05
2280 118
244 20.0
9.93 2.14
5210 259
225 8.62
compd
Aβ40
Aβ42
2
3
4
5
6
7
>3.00 × 104
>3.00 × 104
>3.00 × 104
>3.00 × 104
>3.00 × 104
>3.00 × 104
>3.00 × 104
>3.00 × 104
>3.00 × 104
>3.00 × 104
>3.00 × 104
>3.00 × 104
inhibitor, although foldamers 1 and 9 showed comparably high
population of 12-helical state. We concluded that foldamer 1 is
an appropriate lead structure for further modification.
Direct Identification of the Target Site. It is widely
accepted that (1) PS fragments represent the proteolytically
active molecule, (2) PS fragments form an initial substrate
docking site that is distinct from the catalytic site of γ-secretase,
and (3) substrates are transferred from the docking site to the
catalytic site through a transition path.14 Pep.11-Bt, a
photoactivatable helical peptide containing Aib, is known to
label PS1 NTF under photoirradiation (see Supporting
Information for chemical structure).11 The labeling of PS1
NTF by pep.11-Bt was completely abolished by preincubation
with foldamer 1, suggesting that 1 shares the same binding site
with pep.11, namely, the initial substrate docking site.12
However, we could not exclude the possibility that 1
allosterically altered the conformation of the pep.11-Bt binding
site. Therefore, we set out to directly identify the target site by
employing a photoactivatable analogue of 1.
Length of Foldamer. Next, we examined the most
appropriate length of 12-helical foldamer. We have reported
that foldamer 1 showed stronger activity than ACPC 9mer, but
we did not evaluate ACPC 10mer and 11mer. Thus, foldamers
8 and 9 were synthesized and their inhibitory activity was
examined (Figure 3a). Fmoc-(S,S)-ACPC-OH was efficiently
To create the photoprobe, we modified foldamer 1 by
incorporating 4-benzoyl-L-phenylalanine (Bpa) at the N-
terminus. We also installed biotin via a linker module at the
N terminus to generate 10 (Figure 4a). The side chain of the
Bpa residue is benzophenone, a commonly used photoreactive
moiety that is covalently inserted into the closest C−H bond
(within 3 Å) upon irradiation at 350 nm.36 Biotin allows
isolation of the labeled species with avidin-based beads.
Synthesis of 10 was performed by a solid-phase method, and
the product was purified by reverse-phase HPLC. The CD
spectrum of 10 confirmed that it retained the right-handed 12-
helical structure, although the population of 12-helix was
decreased compared with 1 (see Supporting Information). Also,
foldamer 10 retained a significant γ-secretase-inhibitory activity,
although the activity was weaker than that of 1 (see Supporting
Information).
CHAPSO-solubilized cell lysates were incubated with
foldamer 10 in the presence or absence of the parent foldamer
1 and irradiated at 350 nm. The photolabeled proteins were
pulled down with streptavidin beads and subjected to
immunoblotting. As shown in Figure 4b, foldamer 10
specifically labeled PS1 NTF in the same manner as did
pep.11-Bt, and the labeling was subject to competition by an
excess of the parent foldamer 1. In addition, we never observed
the labeling of PS1 CTF, which is a direct molecular target of
dipeptidic GSI DAPT-based photoprobe DAP-BpB (Figure
Figure 3. (a) Structure of foldamers 8 (10mer) and 9 (11mer). (b)
CD spectra of foldamers 8 and 9 in methanol. The molar ellipticity
[Θ] values have been normalized for peptide concentration and the
number of backbone amide groups.
prepared from ethyl 2-oxocyclopentanecarboxylate according to
the literature.32 The foldamers were prepared by solid-phase
methods analogous to the standard Fmoc solid-phase peptide
synthesis and purified by preparative reverse-phase HPLC (for
details, see Supporting Information). The folding propensities
of these foldamers in methanol were examined by CD
spectroscopy (Figure 3b). All foldamers, including foldamer
1, displayed a maximum at 222 nm and a minimum at 204−205
nm, which are characteristic of right-handed 12-helical β-
peptides.33−35 The increase in intensity at 222 nm with
1445
dx.doi.org/10.1021/jm301306c | J. Med. Chem. 2013, 56, 1443−1454

Journal of Medicinal Chemistry
Article
Figure 4. (a) Structure of 10, which incorporates biotin and benzophenone. (b−d) Photoaffinity labeling experiment with 10 and competition assay
in the presence of 1. Labeling by pep.11-Bt (b, d) or DAP-BpB (c) with/without parent compound (pep.11 or DAPT, respectively) is shown as
control experiments. Antibodies used are shown under the panels. Nonspecific interaction of Pen-2 with both photoprobes was observed.
4c).14 In contrast, we did not observe specific labeling of other
γ-secretase components (i.e., PS1 CTF, nicastrin, Aph-1, and
Pen-2; Figure 4b,d). We also investigated the labeling of signal
peptide peptidase (SPP), an intramembrane cleaving protease
that shares catalytic YD/GxGD motifs with PS.37 We and
others have reported that transition state analogue-type GSIs,
as well as potent dipeptidic GSIs such as DBZ but not DAPT,
cross-inhibit SPP.13,38,39 It is noteworthy that leucine-rich Aib-
containing helical peptides, although their sequence is different
from that of pep.11, cross-inhibited γ-secretase and SPP
activities.40,41 In fact, pep.11-Bt also bound to SPP in our
assay. However, no specific photolabeling of SPP by photop-
robe 10 was observed (Figure 4d), in good accordance with our
previous finding that foldamer 1 did not inhibit SPP.12 Taken
together, these results strongly indicate that foldamer 1 directly
targets the initial substrate docking site in PS.
Modification of Foldamer 1. Foldamer 1 seemed the
most appropriate candidate for structural modification.
Although the cyclopentane unit in ACPC residue is hydro-
phobic, the chemical structure of 1 is extremely simple and
contains no significant functional group. We envisioned that
replacement of the ACPC residue with a β3-amino acid with a
side chain functional group might improve the inhibitory
potency and/or substrate selectivity. The strong folding
propensity of β-peptide foldamer was expected to ensure that
the overall conformation would be maintained after the
incorporation of functional group(s), which is not necessarily
the case with α-peptides; short α-peptides normally do not
form stable helices. Precise design and rapid solid-phase
synthesis are feasible with the β-peptide foldamer scaffold.
Since β-peptide foldamers are not susceptible to proteolytic
degradation,26,42 foldamer 1 should be a good lead compound
not only for development of chemical tools for biomedical
research but also for candidate pharmaceuticals active in vivo.
We designed foldamers 11−25 by replacing an ACPC
residue with a linear β3-amino acid bearing a proteinogenic side
chain functional group (Figure 5a). ACPC oligomers are
known to maintain their 12-helical conformation even when
some of the rigid ACPC residues are replaced with linear,
flexible β3-amino acids.29,43 We chose five different β3-amino
acids with representative side chain functional groups, namely,
β3-hSer, β3-hAsn, β3-hGlu, β3-hLys, and β3-hLeu. These β3-
amino acids contain neutral and hydrophilic (β3-hSer, β3-hAsn),
negatively charged (β3-hGlu), positively charged (β3-hLys), and
hydrophobic (β3-hLeu) functional groups. To clarify the
appropriate position of replacement, we synthesized foldamer
1 in which the residue at the third (X3), sixth (X6), or ninth
(X9) position was replaced with one of the above β3-amino
acids.
Fmoc-β3-amino acids, which are suitably protected for solid-
phase synthesis, were prepared enantiospecifically from the
corresponding Fmoc-α-amino acids via methodology developed
by Seebach et al.,44 as modified by Muller et al.45 The obtained
̈
compounds showed spectroscopic data identical to those
reported in the literature.46−48 All the foldamers were prepared
by the solid-phase method and purified by preparative reverse-
phase HPLC (for details, see Supporting Information).
The folding propensities of these foldamers were examined
by means of CD spectroscopy. The CD spectra in methanol are
shown in Figure 5b. All foldamers displayed characteristic
spectra of right-handed 12-helical β-peptides.33−35 Foldamers
11−25 showed quite similar CD spectra, indicating that the
overall populations of 12-helix were similar. The intensity at
222 nm is slightly diminished for 11−25 relative to 1,
suggesting that replacement of one cyclopentyl residue by an
acyclic residue slightly decreased the stability of the 12-helix.
The inhibitory potency of foldamers 11−25 at 100 nM was
measured by in vitro assay using recombinant substrate (Figure
1446
dx.doi.org/10.1021/jm301306c | J. Med. Chem. 2013, 56, 1443−1454

Journal of Medicinal Chemistry
Article
Figure 5. Structure, CD spectra, and γ-secretase-inhibitory activity of foldamers 11−25. (a) Chemical structure. (b) CD spectra in methanol. The
molar ellipticity [Θ] values have been normalized for oligomer concentration and the number of backbone amide groups. (c) Inhibition of γ-
secretase activity by foldamers at 100 nM in an in vitro assay (n = 3).
5c).31 Most foldamers exhibited γ-secretase-inhibitory activity;
foldamers 11 (X3 = β3-hSer) and 12 (X3 = β3-hAsn) showed
much stronger activity than the parent foldamer 1, while the
other foldamers showed comparable or weaker activity. In
general, substitution at X3 (foldamers 11−15) tended to
increase the inhibitory activity, while substitution at position X6
(foldamers 16−20) or X9 (foldamers 21−25) tended to
decrease it. The activity of foldamers with charged residues (β3-
hGlu; 18 and 23, β3-hLys; 19 and 24) at position X6 or X9 was
greatly reduced, although substitution at X3 had only a minimal
effect (13 and 14). This is consistent with the hypothesis that
the foldamers are recognized by γ-secretase as transmembrane
mimetics via hydrophobic interaction at the initial substrate
docking site. Thus, charged residues at position X6 or X9 would
be inappropriate for the interaction. In the case of β3-hLeu, the
inhibitory activity was moderately reduced, and the effect was
independent of the substitution position (15, 20, 25).
Among foldamers 11−25, 11 and 12, which contain β3-hSer
and β3-hAsn, respectively, showed potent inhibitory activity.
This result suggests that a neutral and hydrophilic functional
group at X3 is preferable for strong γ-secretase inhibition. Thus,
we designed and synthesized foldamers 26−34, which contain
β3-hThr, β3-hTyr, β3-hGln at positions X3, X6, and X9 (Figure
6a). As shown in Figure 6b, the CD spectra of these foldamers
were similar to those of foldamers 11−25 and are characteristic
of a right-handed 12-helix. The signal intensities were very
1447
dx.doi.org/10.1021/jm301306c | J. Med. Chem. 2013, 56, 1443−1454

Journal of Medicinal Chemistry
Article
Figure 6. Structure, CD spectra, and γ-secretase-inhibitory activity of foldamers 26−34. (a) Chemical structure. (b) CD spectra in methanol. The
molar ellipticity [Θ] values have been normalized for oligomer concentration and the number of backbone amide groups. (c) Inhibition of γ-
secretase activity by foldamers at 100 nM in an in vitro assay (n = 3).
similar, suggesting that the foldamers 11−34 have similar
populations of 12-helix. The inhibitory potency of foldamers
26−34 at 100 nM was measured by means of in vitro assay
using recombinant substrate (Figure 6c).31 Most of the
foldamers showed weaker γ-secretase-inhibitory activity than
the parent foldamer 1 except for 28 (X3 = β3-hGln).
Modification at the X3 position with β3-hGln resulted in
relatively potent activity. Both 28 and 12 (X3 = β3-hAsn)
contain a carboxamide functional group, suggesting that the
presence of a neutral and hydrophilic carboxamide group is
important for the activity. Further, 12 showed stronger activity
than 28, which may suggest that a smaller side chain group is
favorable for the inhibitory activity. Despite the similarity of the
β3-hThr and β3-hSer side chain functional groups, foldamer 26
(X3 = β3-hThr) showed only modest γ-secretase-inhibitory
activity, while foldamer 11 (X3 = β3-hSer) showed potent
activity. The additional methyl group in the β3-hThr side chain
might cause steric repulsion at the interface of γ-secretase. This
result also suggests that a smaller functionality is preferable for
effective inhibition of γ-secretase activity.
right-handed 12-helix. The inhibitory potency of foldamers 35−
42, along with 11 and 12, was measured at 100 nM (Figure 7c).
Foldamers 11 and 12, which contain β3-hSer and β3-hAsn at
position X3, respectively, showed potent inhibitory activity.
This result confirmed that the X3 position is the most favorable
position to modify.
Since the X3 position seemed appropriate for modification,
we further designed and synthesized four additional foldamers
with substitution at X3 (43−46, Figure 8a). β-Amino acids with
a relatively small functional group were chosen. β3-hCys was
selected for 43, since the thiol group is polar, though it is prone
to be oxidized. Also, a hydrophilic ring-constrained β-amino
acid, Fmoc-trans-4-aminotetrahydrofuran-3-carboxylic acid
((R,R)-AFC),49 was selected in anticipation of both hydro-
philicity and strong induction of 12-helical structure. In
addition, we chose β3-hGly and β3-hAla, which have no side
chain functional group and a methyl side chain functional
group, respectively. Glycine and alanine are small residues
frequently found at the contact interfaces between trans-
membrane helices.50 The foldamers 43−46 were synthesized
and purified in the same manner. The synthesis of Fmoc-(R,R)-
AFC-OH is described in Supporting Information.49
In order to clarify the best position for modification, we have
synthesized foldamers 35−42 with the replacement of ACPC
with β3-hSer or β3-hAsn on positions X1, X2, X4, X5 (Figure 7a).
As shown in Figure 7b, the CD spectra of these foldamers were
similar to those of foldamers 11−34 and are characteristic of a
Foldamers 43−46 each formed a stable 12-helix in methanol,
as demonstrated by the CD spectra; the intensities at 222 nm
were quite similar to those of foldamers 11−42 except for 44
1448
dx.doi.org/10.1021/jm301306c | J. Med. Chem. 2013, 56, 1443−1454

Journal of Medicinal Chemistry
Article
Figure 7. Structure, CD spectra, and γ-secretase-inhibitory activity of foldamers 35−42. (a) Chemical structure. (b) CD spectra in methanol. The
molar ellipticity [Θ] values have been normalized for oligomer concentration and the number of backbone amide groups. (c) Inhibition of γ-
secretase activity by foldamers at 100 nM in an in vitro assay (n = 3).
(Figure 8b). As expected, foldamer 44 showed strong CD
intensity comparable to that of foldamer 1, suggesting that its
12-helix conformation is more stable than those of the other
modified foldamers. The inhibitory potency of foldamers 43−
46 at 100 nM was measured by means of in vitro assay (Figure
8c). All the foldamers showed potent γ-secretase-inhibitory
activity comparable to or even stronger than that of foldamer 1.
Interestingly, foldamers 45 (X3 = β3-hGly) and 46 (X3 = β3-
hAla) showed slightly more potent activity than foldamer 1,
even though the population of 12-helix was decreased and they
contain no polar side chain. This result indicates that a small
functional group at the X3 position is superior to ACPC in
terms of efficient inhibition of γ-secretase.
In order to clarify the inhibitory potency more precisely, the
IC50 values of potent foldamers were measured using an in vitro
assay. Foldamers 11 (X3 = β3-hSer), 12 (X3 = β3-hAsn), and 45
(X3 = β3-hGly) were selected from the initial screening. As
shown in Table 3, foldamers 11 and 12 were the most potent
inhibitors; they showed similar IC50 values and were more than
10-fold more potent than the lead foldamer 1. Foldamer 45
showed weaker inhibitory activity but was still more potent
than foldamer 1. The IC50 values for foldamer 1 were slightly
different from the values reported previously.12 We have
experienced that IC50 is somewhat varied (75−150%) using a
different lot of the recombinant substrate for in vitro assay by
unknown reason. However, the difference in the potency of
modified peptides compared to that of foldamer 1 is almost
similar. To make adequate comparison throughout the
manuscript, we have shown exact values of IC50.
Next, the inhibitory activity of the potent foldamers was
measured using a cell-based assay. As noted previously, there is
serious concern about adverse effects caused by GSIs that
inhibit the release of NICD, a signaling molecule in the Notch
pathway.3−7 Thus, a means to spare Notch-cleaving γ-secretase
activity is considered mandatory for development of AD
therapeutics. We established a HEK293-based reporter cell line
stably expressing APP carrying the Swedish mutation, truncated
notch, notch intracellular domain (NICD) driven TP1-firefly
luciferase, and EGFP (NLNTK cell line).51 Secretion of Aβ
from NLNTK cells in the presence of foldamers 1, 11, and 12
was quantified by ELISA. Simultaneously, notch cleavage was
evaluated in terms of firefly luciferase activity in NLNTK cells.
As shown in Table 4, foldamers 11 (X3 = β3-hSer) and 12 (X3 =
β3-hAsn) showed greater activity than foldamer 1 (X3
=
ACPC). Both foldamers 11 and 12 contain a side chain
functional group capable of hydrogen bond formation. The
hydrogen bond donating nature might be important for their
potent activities. Foldamer 1 preferentially inhibited NICD
over Aβ production in NLNTK cells, which is a different
selectivity from that which we found previously.12 This
difference is probably due to the difference of assay method
and cell type. Intriguingly, foldamer 11 inhibited Aβ production
10-fold more potently than did 1, though it retained inhibitory
potency toward NICD. Although foldamer 1 showed
preferential inhibition on NICD generation, simple incorpo-
ration of β3-hSer at X3 position increased inhibitory potency
against Aβ without affecting NICD inhibition. This result
indicates that β3-hSer caused a substrate-specific effect
1449
dx.doi.org/10.1021/jm301306c | J. Med. Chem. 2013, 56, 1443−1454

Journal of Medicinal Chemistry
Article
Figure 8. Structure, CD spectra, and γ-secretase-inhibitory activity of foldamers 43−46. (a) Chemical structures. (b) CD spectra in methanol. The
molar ellipticity [Θ] values have been normalized for oligomer concentration and the number of backbone amide groups. (c) Inhibition of γ-
secretase activity by foldamers at 100 nM in an in vitro assay (n = 3).
Table 3. Inhibitory Potency of Foldamers toward γ-Secretase
in in Vitro Assay (n = 3)
IC50 (nM)
compd
X3
Aβ40
Aβ42
1
ACPC
8.81 1.05
9.93 2.14
11
12
45
β3-hSer
β3-hAsn
β3-hGly
0.402 0.0300
0.555 0.120
2.47 0.368
0.830 0.111
0.880 0.0950
7.47 2.04
Table 4. Inhibitory Potency of Foldamers in HEK293/
NLNTK Cell Assay (n = 3)
IC50 (μM)
NICD/Aβ42 IC50
compd
X3
IAβ40
Aβ42
NICD
ratio
1
ACPC
1.18
0.626
0.0204
0.0413
0.00316
0.0326
0.478
11
12
β3-hSer
β3-hAsn
0.0842
0.0846
0.0864
0.0684
0.0462
regarding γ-secretase inhibition. In contrast, 12 was 10-fold
more potent than 1 for both Aβ and NICD generation. We also
examined the effect of the foldamers on human H4 glioma cells,
which express endogenous APP and Notch1 proteins (Figure
9). We did not observe any change in expression of either APP
holoprotein or Notch1 in response to treatment with GSIs. In
contrast, significant accumulation of APP CTF was detected,
suggesting that foldamers inhibited the γ-secretase activity
without affecting the expression of γ-substrates. Moreover, 11
and 12 showed more potent inhibitory activity than did 1.
Finally, to analyze the mode of inhibition by 11 and 12, we
Figure 9. Effect of foldamers on the expression levels of endogenous γ-
substrates in H4 glioma cells. Cells were treated with the indicated
compounds for 24 h. Antibodies used for immunoblot analysis are
shown under the panels.
examined these foldamers as competitors of the labeling of PS1
NTF by photoprobe 10 (Figure 10b). The biotinylation of PS1
NTF was almost completely blocked by preincubation with
foldamers 11 and 12 in a similar fashion to that by foldamer 1,
suggesting that these foldamers share the same binding site.
1450
dx.doi.org/10.1021/jm301306c | J. Med. Chem. 2013, 56, 1443−1454

Journal of Medicinal Chemistry
Article
Figure 10. (a) Structures of sulfonamide-type Notch-sparing GSIs, begacestat 47, and avagacestat 48. (b) Cross-competition assay using 47 in a
photoaffinity labeling experiment with 10 and pep.11-Bt.
These data support our in vitro finding that incorporation of β3-
hSer at the X3 position of foldamers 1 provides increased
substrate selectivity as well as increased inhibitory potency.
A series of sulfonamide/sulfone-containing small-molecular
compounds have been identified as Notch-sparing GSIs.6,7
These compounds predominantly inhibit Aβ production rather
than Notch signaling in vivo, while they show similar potency
against Aβ and NICD in in vitro γ-secretase assay.52 Among
Notch-sparing GSIs, begacestat 47 and avagacestat 48 have
been tested in clinical trials.6,7 However, the molecular
mechanism whereby Notch-sparing GSIs show substrate
selectivity remains unknown. To test whether the foldamer
binding site in PS1 NTF is functionally correlated with the
Notch-sparing effect of sulfonamide-type GSI, we tested the
effect of begacestat (47) on photoaffinity labeling. Inhibitory
activity of 47 was confirmed by in vitro γ-secretase assay (IC50
for Aβ40, 0.203 μM; IC50 for Aβ42, 0.414 μM). We found that
the labeling of PS1 NTF by photoprobe 10 was significantly,
but incompletely, reduced by preincubation with 47, suggesting
that the binding of Notch-sparing GSI allosterically affected the
structure of the foldamer binding site (Figure 10). Surprisingly,
however, 47 did not compete with the labeling of PS1 NTF by
pep.11-Bt, a substrate-mimetic helical peptide-type GSI-based
photoprobe. These data suggest that, in contrast to Aib-
containing helical peptides, ACPC-based foldamers target not
only the helix recognition region in the initial substrate docking
site but also the domain involved in the mechanism of substrate
selectivity.
Notch-sparing small-molecular GSIs targeting the initial
substrate docking site.
CONCLUSION
■
The results presented here and in our previous paper12 clearly
demonstrate the potential utility of 12-helical foldamers as
GSIs. We have shown that compounds with the polyproline II
helix lack γ-secretase-inhibitory activity, highlighting the
importance of helix shape (i.e., the 12-helix) for the activity.
As for length, ACPC 12mer seemed appropriate. The target site
of foldamer 1 was directly identified as the initial substrate
docking site by means of a photoaffinity labeling study, using
photoprobe 10. Further detailed examination of the binding site
of photoprobe 10 is expected to provide protein-structural
information about the initial substrate binding site in PS.
In order to improve the inhibitory potency and/or substrate
selectivity of foldamer 1, we prepared various derivatives by
replacement of an ACPC residue with a linear β3-amino acid.
Various side chain functional groups were incorporated at
diverse positions, and we identified foldamers 11 (X3 = β3-
hSer) and 12 (X3 = β3-hAsn) as potent GSIs. Intriguingly,
drastic position-specific effects were observed for incorporation
of β3-hSer, β3-hAsn, β3-hGlu, β3-hLys, and β3-hGln; i.e.,
substitution at the X3 position is favorable for inhibitory
activity, whereas the X6 and X9 positions are unfavorable. These
results suggest that the foldamers are recognized at two binding
sites of PS in position-specific manner. Notably, our cysteine-
based structural analysis revealed that several TMDs of PS may
have dual functions, dependent on topological location; the
luminal side of TMD1, -6, and -9 is implicated in substrate
binding via hydrophobic interaction.53−55 In contrast, the
cytosolic side of these TMDs directly faces the hydrophilic
environment within the membrane. Thus, the X6/9 positions in
foldamers would bind with the luminal side of these TMDs.
Moreover, small, neutral, and polar side chain groups at the X3
position were found to be favorable for potent activity. In
addition, a significant difference in inhibitory potencies toward
As previously described, γ-secretase cleaves a wide variety of
membrane proteins with no clear consensus sequence.16
However, the results presented here indicate that the position
and structure of side chain functional groups do in fact
drastically affect the activity. Further studies to define the
binding site of the photoprobe 10 in detail may provide insight
into the molecular mechanism of the substrate selectivity of γ-
secretase, as well as provide a basis for rational development of
1451
dx.doi.org/10.1021/jm301306c | J. Med. Chem. 2013, 56, 1443−1454

Journal of Medicinal Chemistry
Article
from excess cold Et2O and isolated by centrifugation. The crude
oligoproline peptides and β-peptides thus obtained were purified by
RP-HPLC.
CD Spectra. Dry peptide/foldamer samples were weighed on a
microanalytical balance and dissolved in an appropriate amount of
HPLC-grade methanol. Sample cells of 2 mm path length were used.
Data were collected on a Jasco J-725 spectropolarimeter at 25 °C. Data
were converted to mean residue ellipticity (deg cm2 dmol−1) according
to the following equation:
APP- and Notch-cleaving activities was observed between 11
and 12, suggesting that the binding site of the X3 residue in the
foldamers is critical to the substrate selectivity of γ-secretase. In
support of this notion, a known Notch-sparing GSI, begacestat,
decreased the binding of the foldamer to PS1 NTF whereas it
did not compete with the interaction between Aib-based helical
peptide and PS1. Nevertheless, a more detailed analysis of the
binding mode of these foldamers should be clarified further.
Structural analysis of the γ-secretase using the foldamers may
throw light on the molecular mechanism of substrate
recognition.
ψMr
[Θ] =
100lc
where ψ is the CD signal in degrees, Mr is the molecular weight
divided by the number of chromophores, l is the path length in
decimeters, and c is the concentration in g/mL.
In summary, our results demonstrate the potential value of
foldamer-based inhibitors of γ-secretase. Through optimization,
we have identified some of the substrate recognition properties
of γ-secretase, which might offer clues for understanding the
substrate selectivity and for developing Notch-sparing GSIs.
The foldamer-based approach might be useful to explore
substrate recognition of various intramembrane proteases.
γ-Secretase Assay and Photoaffinity Labeling. γ-Secretase-
inhibitory activity of oligoproline peptides and foldamers was
measured using in vitro or cell-based assay as previously
described.12,31,51 H4 glioma cells were purchased from ATCC and
maintained in DMEM with 10% FBS and Penicillin/Streptomycin.
Photoaffinity labeling experiments using foldamer 10 as well as pep.11-
Bt were performed as previously described.12,14 N-[N-(3,5-Difluor-
ophenacetyl)-L-alanyl]-(S)-phenylglycine tert-butyl ester (DAPT) was
synthesized as previously reported.56 Begacestat (39) was prepared
according to the literature with slight modifications.57,58 L-685,458
(Peptide Institute), pep.11 (BEX CO., LTD.), and pep.11-Bt (BEX
CO., LTD.) were purchased from the indicated vendors. Anti-PS1NT
and SPPCT for detection of PS1 NTF and SPP, respectively, were
kindly provided by Drs. Gopal Thinakaran (The University of
Chicago) and Todd Golde (University of Florida). Anti-nicastrin
(N1660) (Sigma), anti-Aph-1aL (O2C2) (Covance), anti-APP C
terminus (APPc) (Immuno-Biological Laboratories), anti-Notch C
terminus (EP1238Y) (Epitomics), anti-NICD (D3B8) (Cell Signaling
Technology), and anti-α-tubulin (DM1A) (Sigma) antibodies were
purchased from the indicated vendors. anti-Pen-2 PNT3 was reported
previously.59
EXPERIMENTAL SECTION
■
General. All reagents and solvents were of the highest commercial
quality and were used without purification. Fmoc-trans-2-amino-
cyclopentanecarboxylic acid (Fmoc-ACPC-OH),32 Fmoc-β3-amino
acids,44−48 were synthesized according to the cited references.
Synthesis of Fmoc-(R,R)-AFC-OH is described in Supporting
Information. Flash column chromatography was performed on silica
gel (Fuji Silysia Chemical Ltd., BW-300) using forced flow. Thin layer
chromatography (TLC) was performed on Merck precoated plates
(silica gel 60 F254, 0.25 mm), and bands were visualized by
fluorescence quenching under UV light or by staining with potassium
1
permanganate or ninhydrin. H NMR and 13C NMR spectra were
recorded on a JEOL LNM-LA500 at 500 and 125 MHz, respectively.
Matrix-assisted laser desorption ionization-time-of-flight mass spec-
trometry (MALDI-TOF MS) was done with a Shimadzu AXIMA-
CFR-NC using α-cyano-4-hydroxycinnamic acid as the matrix.
Preparative RP-HPLC and analytical RP-HPLC were performed
using a Shimadzu SPD-M10AVP variable-wavelength UV detector.
Inertsil ODS-3 (10 mm × 250 mm, GL Science) and Inertsil ODS-3
(4.6 mm × 250 mm, GL Science) columns were employed for
chromatographic and analytical separations, respectively. The purities
of oligoproline peptides and β-peptide foldamers were >95%, as
measured by comparing peak areas in analytical HPLC traces at 220
nm (see Supporting Information).
ASSOCIATED CONTENT
* Supporting Information
■
S
Synthesis of Fmoc-(R,R)-AFC-OH, identification of oligopro-
line peptides and foldamers, CD spectra and γ-secretase-
inhibitory activity of photoprobe 10, and chemical structure of
pep.11-Bt. This material is available free of charge via the
Internet at http://pubs.acs.org.
Synthesis of Oligoproline Peptides and β-Peptide Fol-
damers. Oligoproline peptides and β-peptides were synthesized
with standard Fmoc solid-phase methods. Fmoc-NH-SAL-PEG resin
(0.24 mmol/g, 100−200 mesh, 1% DVB) was employed for all peptide
synthesis. For a typical 10 μmol scale synthesis, 42 mg of Fmoc-NH-
SAL-PEG resin was swollen for 15 min in DMF.
Coupling Cycles. Amounts of 3 equiv of Fmoc-amino acid and 3
equiv of 2-(1H-benzotriazol-1-yl)-1,1,3,3-tetramethyluronium hexa-
fluorophosphate (HBTU) and 3 equiv of 1-hydroxybenzotriazole
monohydrate (HOBt·H2O) were dissolved in 300 μL of DMF, and 6
equiv of N,N-diisopropylethylamine (DIEA) was added. The solution
was added to resin bearing the N-deprotected peptide or β-peptide.
The resin was agitated for 1−3 h.
Fmoc Deprotection Cycles. Fmoc deprotection was accomplished by
adding to the resin 1.0 mL of 20% (v/v) piperidine in DMF and
rocking for 30 min.
Acetylation. Acetylation of peptides was conducted for 2 h, by
addition of 0.5 mL of 2:2:1 (v/v/v) Ac2O/DMF/Et3N to resin bearing
the final desired N-deprotected peptide or β-peptide sequence.
Cleavage. Cleavage of all oligoproline peptides and β-peptides from
the resin was accomplished by shaking the resin in a solution of
trifluoroacetic acid (TFA)/triisopropylsilane (TIPS)/H2O, 95:2.5:2.5
(0.5 mL), for 2 h. The resin was removed by filtration and rinsed with
additional TFA. The combined filtrate was concentrated under a
stream of Ar. Oligoproline peptides and β-peptides were precipitated
AUTHOR INFORMATION
Corresponding Author
*For N.U.: phone, +81-52-836-3438; e-mail, umezawa@phar.
nagoya-cu.ac.jp. For T.T.: phone, +81-3-5841-4868; e-mail,
taisuke@mol.f.u-tokyo.ac.jp. For T.H.: phone, +81-52-836-
3435; e-mail, higuchi@phar.nagoya-cu.ac.jp.
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported in part by Grants-in-Aid for Scientific
Research (A) (T.H.), (C) (N.U.), Grants-in-Aid for Young
Scientists (S) (T.T.) from the Japan Society for the Promotion
of Science (JSPS), and grants from the Targeted Proteins
Research Program of the Japan Science and Technology
Agency (JST) (T.I., T.T.), and Takeda Memorial Foundation
(N.U.). Y.I. is a research fellow of JSPS.
REFERENCES
■
(1) Holtzman, D. M.; Morris, J. C.; Goate, A. M. Alzheimer’s disease:
the challenge of the second century. Sci. Transl. Med. 2011, 3, 77sr1.
1452
dx.doi.org/10.1021/jm301306c | J. Med. Chem. 2013, 56, 1443−1454

Journal of Medicinal Chemistry
Article
(2) Iwatsubo, T.; Odaka, A.; Suzuki, N.; Mizusawa, H.; Nukina, N.;
Ihara, Y. Visualization of Aβ42(43) and Aβ40 in senile plaques with
end-specific Aβ monoclonals: evidence that an initially deposited
species is Aβ42(43). Neuron 1994, 13, 45−53.
(3) Tomita, T. Secretase inhibitors and modulators for Alzheimer’s
disease treatment. Expert Rev. Neurother. 2009, 9, 661−679.
(4) De Strooper, B.; Vassar, R.; Golde, T. The secretases: enzymes
with therapeutic potential in Alzheimer disease. Nat. Rev. Neurol. 2010,
6, 99−107.
(5) Wolfe, M. S. γ-Secretase inhibitors and modulators for
Alzheimer’s disease. J. Neurochem. 2012, 120 (Suppl. 1), 89−98.
(6) Kreft, A. F.; Martone, R.; Porte, A. Recent advances in the
identification of γ-secretase inhibitors to clinically test the Aβ oligomer
hypothesis of Alzheimer’s disease. J. Med. Chem. 2009, 52, 6169−6188.
(7) Oehlrich, D.; Berthelot, D. J.-C.; Gijsen, H. J. M. γ-Secretase
modulators as potential disease modifying anti-Alzheimer’s drugs. J.
Med. Chem. 2011, 54, 669−698.
(8) Takasugi, N.; Tomita, T.; Hayashi, I.; Tsuruoka, M.; Niimura, M.;
Takahashi, Y.; Thinakaran, G.; Iwatsubo, T. The role of presenilin
cofactors in the γ-secretase complex. Nature 2003, 422, 438−441.
(9) Li, Y. M.; Xu, M.; Lai, M. T.; Huang, Q.; Castro, J. L.; DiMuzio-
Mower, J.; Harrison, T.; Lellis, C.; Nadin, A.; Neduvelil, J. G.; Register,
R. B.; Sardana, M. K.; Shearman, M. S.; Smith, A. L.; Shi, X. P.; Yin, K.
C.; Shafer, J. A.; Gardell, S. J. Photoactivated γ-secretase inhibitors
directed to the active site covalently label presenilin 1. Nature 2000,
405, 689−694.
(10) Xu, M.; Lai, M.-T.; Huang, Q.; DiMuzio-Mower, J.; Castro, J. L.;
Harrison, T.; Nadin, A.; Neduvelil, J. G.; Shearman, M. S.; Shafer, J. A.;
Gardell, S. J.; Li, Y.-M. γ-Secretase: characterization and implication for
Alzheimer disease therapy. Neurobiol. Aging 2002, 23, 1023−1030.
(11) Kornilova, A. Y.; Bihel, F.; Das, C.; Wolfe, M. S. The initial
substrate-binding site of γ-secretase is located on presenilin near the
active site. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 3230−3235.
(12) Imamura, Y.; Watanabe, N.; Umezawa, N.; Iwatsubo, T.; Kato,
N.; Tomita, T.; Higuchi, T. Inhibition of γ-secretase activity by helical
β-peptide foldamers. J. Am. Chem. Soc. 2009, 131, 7353−7359.
(13) Fuwa, H.; Takahashi, Y.; Konno, Y.; Watanabe, N.; Miyashita,
H.; Sasaki, M.; Natsugari, H.; Kan, T.; Fukuyama, T.; Tomita, T.;
Iwatsubo, T. Divergent synthesis of multifunctional molecular probes
to elucidate the enzyme specificity of dipeptidic γ-secretase inhibitors.
ACS Chem. Biol. 2007, 2, 408−418.
(14) Morohashi, Y.; Kan, T.; Tominari, Y.; Fuwa, H.; Okamura, Y.;
Watanabe, N.; Sato, C.; Natsugari, H.; Fukuyama, T.; Iwatsubo, T.;
Tomita, T. C-terminal fragment of presenilin is the molecular target of
a dipeptidic γ-secretase-specific inhibitor DAPT (N-[N-(3,5-difluor-
ophenacetyl)-L-alanyl]-S-phenylglycine t-butyl ester. J. Biol. Chem.
2006, 281, 14670−14676.
(15) Haapasalo, A.; Kovacs, D. M. The many substrates of presenilin/
γ-secretase. J. Alzheimer’s Dis. 2011, 25, 3−28.
(16) Beel, A. J.; Sanders, C. R. Substrate specificity of γ-secretase and
other intramembrane proteases. Cell. Mol. Life Sci. 2008, 65, 1311−
1334.
(17) 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.
(18) Wong, G. T.; Manfra, D.; Poulet, F. M.; Zhang, Q.; Josien, H.;
Bara, T.; Engstrom, L.; Pinzon-Ortiz, M.; Fine, J. S.; Lee, H.-J. J.;
Zhang, L.; Higgins, G. A.; Parker, E. M. Chronic treatment with the γ-
secretase inhibitor LY-411,575 inhibits β-amyloid peptide production
and alters lymphopoiesis and intestinal cell differentiation. J. Biol.
Chem. 2004, 279, 12876−12882.
(19) Gellman, S. H. Foldamers: a manifesto. Acc. Chem. Res. 1998,
31, 173−180.
(20) Seebach, D.; Beck, A. K.; Bierbaum, D. J. The world of β- and γ-
peptides comprised of homologated proteinogenic amino acids and
other components. Chem. Biodiversity 2004, 1, 1111−1239.
(21) Cheng, R. P.; Gellman, S. H.; DeGrado, W. F. β-Peptides: from
structure to function. Chem. Rev. 2001, 101, 3219−3232.
(22) Das, C.; Berezovska, O.; Diehl, T. S.; Genet, C.; Buldyrev, I.;
Tsai, J.-Y.; Hyman, B. T.; Wolfe, M. S. Designed helical peptides
inhibit an intramembrane protease. J. Am. Chem. Soc. 2003, 125,
11794−11795.
(23) Bihel, F.; Das, C.; Bowman, M. J.; Wolfe, M. S. Discovery of a
subnanomolar helical D-tridecapeptide inhibitor of γ-secretase. J. Med.
Chem. 2004, 47, 3931−3933.
(24) Karle, I. L.; Balaram, P. Structural characteristics of α-helical
peptide molecules containing Aib residues. Biochemistry 1990, 29,
6747−6756.
(25) Balaram, P. Non-standard amino acids in peptide design and
protein engineering. Curr. Opin. Struct. Biol. 1992, 2, 845−851.
(26) Porter, E. A.; Weisblum, B.; Gellman, S. H. Mimicry of host-
defense peptides by unnatural oligomers: antimicrobial β-peptides. J.
Am. Chem. Soc. 2002, 124, 7324−7330.
(27) Crisma, M.; Formaggio, F.; Moretto, A.; Toniolo, C. Peptide
helices based on α-amino acids. Biopolymers 2006, 84, 3−12.
(28) Barlow, D. J.; Thornton, J. M. Helix geometry in proteins. J. Mol.
Biol. 1988, 201, 601−619.
(29) Choi, S. H.; Guzei, I. a; Spencer, L. C.; Gellman, S. H.
Crystallographic characterization of 12-helical secondary structure in
β-peptides containing side chain groups. J. Am. Chem. Soc. 2010, 132,
13879−13885.
(30) Rabanal, F.; Ludevid, M. D.; Pons, M.; Giralt, E. CD of proline-
rich polypeptides: application to the study of the repetitive domain of
maize glutelin-2. Biopolymers 1993, 33, 1019−1028.
(31) Takahashi, Y.; Hayashi, I.; Tominari, Y.; Rikimaru, K.;
Morohashi, Y.; Kan, T.; Natsugari, H.; Fukuyama, T.; Tomita, T.;
Iwatsubo, T. Sulindac sulfide is a noncompetitive γ-secretase inhibitor
that preferentially reduces Aβ42 generation. J. Biol. Chem. 2003, 278,
18664−18670.
(32) LePlae, P. R.; Umezawa, N.; Lee, H. S.; Gellman, S. H. An
efficient route to either enantiomer of trans-2-aminocyclopentanecar-
boxylic acid. J. Org. Chem. 2001, 66, 5629−5632.
(33) Appella, D. H.; Christianson, L. A.; Klein, D. A.; Powell, D. R.;
Huang, X.; Barchi, J. J.; Gellman, S. H. Residue-based control of helix
shape in β-peptide oligomers. Nature 1997, 387, 381−384.
(34) Applequist, J.; Bode, K. A.; Appella, D. H.; Christianson, L. A.;
Gellman, S. H. Theoretical and experimental circular dichroic spectra
of the novel helical foldamer poly[(1R,2R)-trans-2-aminocyclopenta-
necarboxylic acid]. J. Am. Chem. Soc. 1998, 120, 4891−4892.
(35) Appella, D. H.; Christianson, L. A.; Klein, D. A.; Richards, M.
R.; Powell, D. R.; Gellman, S. H. Synthesis and structural
characterization of helix-forming β-peptides: trans-2-aminocyclopenta-
necarboxylic acid oligomers. J. Am. Chem. Soc. 1999, 121, 7574−7581.
(36) Dorman, G.; Prestwich, G. D. Benzophenone photophores in
biochemistry. Biochemistry 1994, 33, 5661−5673.
(37) Fluhrer, R.; Haass, C. Signal peptide peptidases and γ-secretase:
cousins of the same protease family? Neurodegener. Dis. 2007, 4, 112−
116.
(38) Nyborg, A. C.; Jansen, K.; Ladd, T. B.; Fauq, A.; Golde, T. E. A
signal peptide peptidase (SPP) reporter activity assay based on the
cleavage of type II membrane protein substrates provides further
evidence for an inverted orientation of the SPP active site relative to
presenilin. J. Biol. Chem. 2004, 279, 43148−43156.
(39) Weihofen, A.; Lemberg, M. K.; Friedmann, E.; Rueeger, H.;
Schmitz, A.; Paganetti, P.; Rovelli, G.; Martoglio, B. Targeting
presenilin-type aspartic protease signal peptide peptidase with γ-
secretase inhibitors. J. Biol. Chem. 2003, 278, 16528−16533.
(40) Sato, T.; Nyborg, A. C.; Iwata, N.; Diehl, T. S.; Saido, T. C.;
Golde, T. E.; Wolfe, M. S. Signal peptide peptidase: biochemical
properties and modulation by nonsteroidal antiinflammatory drugs.
Biochemistry 2006, 45, 8649−8656.
(41) Sato, T.; Ananda, K.; Cheng, C. I.; Suh, E. J.; Narayanan, S.;
Wolfe, M. S. Distinct pharmacological effects of inhibitors of signal
peptide peptidase and γ-secretase. J. Biol. Chem. 2008, 283, 33287−
33295.
1453
dx.doi.org/10.1021/jm301306c | J. Med. Chem. 2013, 56, 1443−1454

Journal of Medicinal Chemistry
Article
(42) Frackenpohl, J.; Arvidsson, P. I.; Schreiber, J. V.; Seebach, D.
The outstanding biological stability of β- and γ-peptides toward
proteolytic enzymes: an in vitro investigation with fifteen peptidases.
ChemBioChem 2001, 2, 445−455.
(43) LePlae, P. R.; Fisk, J. D.; Porter, E. A.; Weisblum, B.; Gellman,
S. H. Tolerance of acyclic residues in the β-peptide 12-helix: access to
diverse side-chain arrays for biological applications. J. Am. Chem. Soc.
2002, 124, 6820−6821.
Zhang, M.; Zhou, H.; Jacobsen, J. S. Discovery of begacestat, a Notch-
1-sparing γ-secretase inhibitor for the treatment of Alzheimer’s disease.
J. Med. Chem. 2008, 51, 7348−7351.
(58) Keese, R.; Hinderling, C. Efficient synthesis of (S)-methyl
hexafluorovalinate. Synthesis 1996, 695−696.
(59) Isoo, N.; Sato, C.; Miyashita, H.; Shinohara, M.; Takasugi, N.;
Morohashi, Y.; Tsuji, S.; Tomita, T.; Iwatsubo, T. Aβ42 over-
production associated with structural changes in the catalytic pore of γ-
secretase: common effects of Pen-2 N-terminal elongation and
fenofibrate. J. Biol. Chem. 2007, 282, 12388−12396.
(44) Guichard, G.; Abele, S.; Seebach, D. Preparation of N-Fmoc-
protected β2- and β3-amino acids and their use as building blocks for
the solid-phase synthesis of β-peptides. Helv. Chim. Acta 1998, 81,
187−206.
(45) Muller, A.; Vogt, C.; Sewald, N. Synthesis of Fmoc-β-
̈
homoamino acids by ultrasound-promoted Wolff rearrangement.
Synthesis 1998, 837−841.
(46) Gademann, K.; Kimmerlin, T.; Hoyer, D.; Seebach, D. Peptide
folding induces high and selective affinity of a linear and small β-
peptide to the human somatostatin receptor 4. J. Med. Chem. 2001, 44,
2460−2468.
(47) Lelais, G.; Micuch, P.; Josien-Lefebvre, D.; Rossi, F.; Seebach, D.
Preparation of protected β2- and β3-homocysteine, β2- and β3-
homohistidine, and β2-homoserine for solid-phase syntheses. Helv.
Chim. Acta 2004, 87, 3131−3159.
́
(48) Wilczynska, D.; Kosson, P.; Kwasiborska, M.; Ejchart, A.; Olma,
A. Synthesis and receptor binding of opioid peptide analogues
containing β3-homo-amino acids. J. Pept. Sci. 2009, 15, 777−782.
(49) Bunnage, M. E.; Davies, S. G.; Roberts, P. M.; Smith, A. D.;
Withey, J. M. Asymmetric synthesis of the cis- and trans-stereoisomers
of 4-aminopyrrolidine-3-carboxylic acid and 4-aminotetrahydrofuran-
3-carboxylic acid. Org. Biomol. Chem. 2004, 2, 2763−2776.
(50) Liu, W.; Crocker, E.; Zhang, W.; Elliott, J. I.; Luy, B.; Li, H.;
Aimoto, S.; Smith, S. O. Structural role of glycine in amyloid fibrils
formed from transmembrane α-helices. Biochemistry 2005, 44, 3591−
3597.
(51) Ohki, Y.; Higo, T.; Uemura, K.; Shimada, N.; Osawa, S.;
Berezovska, O.; Yokoshima, S.; Fukuyama, T.; Tomita, T.; Iwatsubo,
T. Phenylpiperidine-type γ-secretase modulators target the trans-
membrane domain 1 of presenilin 1. EMBO J. 2011, 30, 4815−4824.
́ ́
(52) Chavez-Gutierrez, L.; Bammens, L.; Benilova, I.; Vandersteen,
A.; Benurwar, M.; Borgers, M.; Lismont, S.; Zhou, L.; Cleynenbreugel,
S.; Van; Esselmann, H.; Wiltfang, J.; Serneels, L.; Karran, E.; Gijsen,
H.; Schymkowitz, J.; Rousseau, F.; Broersen, K.; De Strooper, B. The
mechanism of γ-secretase dysfunction in familial Alzheimer disease.
EMBO J. 2012, 31, 2261−2274.
(53) Sato, C.; Takagi, S.; Tomita, T.; Iwatsubo, T. The C-terminal
PAL motif and transmembrane domain 9 of presenilin 1 are involved
in the formation of the catalytic pore of the γ-secretase. J. Neurosci.
2008, 28, 6264−6271.
(54) Watanabe, N.; Image Image, I. I.; Takagi, S.; Tominaga, A.;
Image Image, I.; Tomita, T.; Iwatsubo, T. Functional analysis of the
transmembrane domains of presenilin 1: participation of trans-
membrane domains 2 and 6 in the formation of initial substrate-
binding site of γ-secretase. J. Biol. Chem. 2010, 285, 19738−19746.
(55) Takagi, S.; Tominaga, A.; Sato, C.; Tomita, T.; Iwatsubo, T.
Participation of transmembrane domain 1 of presenilin 1 in the
catalytic pore structure of the γ-secretase. J. Neurosci. 2010, 30,
15943−15950.
(56) Kan, T.; Tominari, Y.; Rikimaru, K.; Morohashi, Y.; Natsugari,
H.; Tomita, T.; Iwatsubo, T.; Fukuyama, T. Parallel synthesis of DAPT
derivatives and their γ-secretase-inhibitory activity. Bioorg. Med. Chem.
Lett. 2004, 14, 1983−1985.
(57) Mayer, S. C.; Kreft, A. F.; Harrison, B.; Abou-Gharbia, M.;
Antane, M.; Aschmies, S.; Atchison, K.; Chlenov, M.; Cole, D. C.;
Comery, T.; Diamantidis, G.; Ellingboe, J.; Fan, K.; Galante, R.;
Gonzales, C.; Ho, D. M.; Hoke, M. E.; Hu, Y.; Huryn, D.; Jain, U.; Jin,
M.; Kremer, K.; Kubrak, D.; Lin, M.; Lu, P.; Magolda, R.; Martone, R.;
Moore, W.; Oganesian, A.; Pangalos, M. N.; Porte, A.; Reinhart, P.;
Resnick, L.; Riddell, D. R.; Sonnenberg-Reines, J.; Stock, J. R.; Sun, S.-
C.; Wagner, E.; Wang, T.; Woller, K.; Xu, Z.; Zaleska, M. M.; Zeldis, J.;
1454
dx.doi.org/10.1021/jm301306c | J. Med. Chem. 2013, 56, 1443−1454