Cyanobacterial peptides as a prototype for the design of potent β-secretase inhibitors and the development of selective chemical probes for other aspartic proteases

Article

pubs.acs.org/jmc

Cyanobacterial Peptides as a Prototype for the Design of Potent

β‑Secretase Inhibitors and the Development of Selective Chemical

Probes for Other Aspartic Proteases

Yanxia Liu,^†,⊥Wei Zhang,^‡,†,⊥Li Li,^§,⊥Lilibeth A. Salvador,^†Tiantian Chen,^§Wuyan Chen,^§

Kevin M. Felsenstein,^∥Thomas B. Ladd,^∥Ashleigh R. Price,^∥Todd E. Golde,^∥Jianhua He,^#Yechun Xu,*^,§

Yingxia Li,*^,‡and Hendrik Luesch*^,†

^†Department of Medicinal Chemistry, University of Florida, Gainesville, Florida 32610, United States

^‡School of Pharmacy, Fudan University, Shanghai 201203, China

^§Drug Discovery and Design Center, Shanghai Institute of Materia Medica, Chinese Academy of Sciences, 555 Zuchongzhi Road,

Shanghai 201203, China

^∥Department of Neuroscience, Center for Translational Research in Neurodegenerative Disease, University of Florida, Gainesville,

Florida 32610, United States

^#Shanghai Institute of Applied Physics, Chinese Academy of Sciences, Shanghai 201800, China

S

* Supporting Information

ABSTRACT: Inspired by marine cyanobacterial natural

products, we synthesized modiﬁed peptides with a central

statine-core unit, characteristic for aspartic protease inhibition.

A series of tasiamide B analogues inhibited BACE1, a

therapeutic target in Alzheimer’s disease. We probed the

stereospeciﬁcity of target engagement and determined addi-

tional structure−activity relationships with respect to BACE1

and related aspartic proteases, cathepsins D and E. We

cocrystallized selected inhibitors with BACE1 to reveal the

structural basis for the activity. Hybrid molecules that combine

features of tasiamide B and an isophthalic acid moiety-

containing sulfonamide showed nanomolar cellular activity.

Compounds were screened in a series of rigorous comple-

mentary cell-based assays. We measured secreted APP ectodomain (sAPPβ), membrane bound carboxyl terminal fragment

(CTF), levels of β-amyloid (Aβ) peptides and selectivity for β-secretase (BACE1) over γ-secretase. Prioritized compounds

showed reasonable stability in vitro and in vivo, and our most potent inhibitor showed eﬃcacy in reducing Aβ levels in the rodent

brain.

type 1 (BACE1), which is the β-secretase implicated in the

generation of the amyloid β-peptide (Aβ) of Alzheimer’s

disease (AD), we found here that the related cyanobacterial

compound tasiamide B (2)^4,5possesses broader speciﬁcity and

also inhibits BACE1. Grassystatins and tasiamide B (2) contain

the characteristic statine-core (γ-amino-β-hydroxy acid) unit

that confers the aﬃnity to the active site of certain aspartic

proteases; however, the statine core in grassystatins is Leu-

derived (Sta) and in tasiamide B (2) is phenylalanine-derived

(phenylstatine). Certain synthetic aspartic protease inhibitors

comprising a phenylstatine unit have previously been shown to

inhibit BACE1.⁶We recently achieved the total synthesis of

grassystatin A,⁷tasiamide B (2)⁵and the originally proposed

(epimeric) structure of tasiamide B (1),⁵which allows us to

INTRODUCTION

■

Cyanobacteria are proliﬁc producers of diverse secondary

metabolites, many of which are linear or cyclic modiﬁed

peptides and depsipeptides.¹These compounds appear to have

a propensity to inhibit various proteases with diﬀerential

selectivity and may provide guidance for the development of

therapeutic protease inhibitors or selective chemical probes. For

example, cyclodepsipeptide-based serine protease inhibitors

with speciﬁcity for elastase, chymotrypsin or trypsin are

commonly produced by many terrestrial and marine cyanobac-

teria.²We also recently reported the discovery and character-

ization of grassystatins A and B, marine-derived linear

depsipeptides that potently inhibit the aspartic proteases

cathepsins D and E, with selectivity for cathepsin E.³

Grassystatin C is a truncated, linear peptide of lower potency

but similar protease selectivity.³While grassystatins were unable

to inhibit the aspartic protease β-site APP Cleaving Enzyme

Received: November 5, 2012

Published: November 26, 2012

10749

dx.doi.org/10.1021/jm301630s | J. Med. Chem. 2012, 55, 10749−10765

Journal of Medicinal Chemistry

Article

characterize their biological activities more rigorously and

deconvolute or selectively enhance certain overlapping

activities. As part of this ongoing research, we engaged in

eﬀorts to design improved analogues with BACE1 inhibitory

activity and with the ultimate goal of obtaining a molecule with

potent cellular activity and in vivo eﬃcacy.

BACE1 is a key factor in the proteolytic formation of the Aβ

peptide, a major component of plaques in the brains of AD

patients, and inhibition of this enzyme has emerged as a major

strategy for pharmacologic intervention in AD.⁸Aβ is generated

upon sequential action of BACE1 and γ-secretase to proteolyti-

cally cleave the amyloid precursor protein (APP).⁹The scission

of APP by BACE1 liberates two cleavage products, viz. secreted

APP ectodomain (sAPPβ) and membrane bound carboxyl

terminal fragment (CTF), C99, which is further processed by γ-

secretase (presenilins) to generate the C-terminus of the Aβ

peptide (Aβ40 and Aβ42) and APP intracellular domain

(Figure 1). BACE1 competes for APP with α-secretase

synthesize other phenylstatine containing peptides 6−16

(Table 1) and directly compare them with compound 5 and

with our previously published tasiamide B-like structures,

compounds 1−4.⁵

In order to prepare compounds 5−16, several building

blocks shown in Scheme 1 were synthesized ﬁrst. Commercially

available amino acids or other simple chemicals were coupled

with other residues, as contained in the designed target

molecules. The combination of EDC and HOAt as coupling

reagents gave satisfactory results. With these building blocks

and other commercially available amino acids in hand, the

designed compounds were prepared smoothly, as described in

Schemes 2−4. Removal of the benzyl ester protecting group of

17 gave the corresponding carboxylic acid, which was coupled

with the amine released from 25, yielding the protected

tetrapeptide 37. Treatment of this compound with TFA

followed by coupling with the amine derivate from building

block 29 gave the linear heptapeptide 39. The desired target

compound 5 was achieved smoothly after hydrogenation of 39

in methanol. A similar strategy yielded compound 8 by using

other building blocks (17, 26, and 36). Amidation of carboxylic

acid 5 with benzyl amine or isopropyl amine provided

compound 6 or 7 in 76% yield (Scheme 2). Hydrogenation

of compound 25 (Scheme 3) yielded the corresponding free

amine, which was coupled with Cbz-Val-OH, giving tripeptide

43 in 82% yield. Treatment of 43 and 33 with TFA released the

corresponding free carboxylic acid and amine, which were

connected under the assistance of HATU/HOAt, resulting in

the desired heptapeptide 9. Removal of the Cbz group of 9

provided compound 10. Blocking the free amine of 10 with the

Boc group gave compound 11. Replacement of the Val-MeGln

residues in N-terminus of 9 with building blocks bearing

aromatic sulfanilamide structures (22/23),¹⁶led to the desired

peptides 12/13. Acetylation of the free amine derivate from

building block 24 (Scheme 4) gave compound 49. Removal of

the protecting group of 49 and coupled with the free amine

deviate from 30 or 33, yielded tetrapeptide 16 or 14,

respectively. Similarly, hexapeptide 15 was obtained smoothly

over three steps using building blocks 27 and 33.

Compound 8, lacking N-methyl groups for the Phe and Gln

residues compared with 2, showed dramatically reduced activity

against BACE1, indicating the importance of the tertiary amide

functionalities. Since N-terminally truncated compounds 14

and 15 containing a secondary amide for Gln and an N-Me-Phe

residue near the C-terminus have similar activity as 8, it appears

that the N-Me-Gln residue is particularly important for BACE1

inhibition of this series. Compound 9, an N-terminally

truncated Cbz analogue of 2 (Figure 2), appeared to be a

potent broad-spectrum aspartic protease inhibitor in vitro, with

potent activity against BACE1. BACE1 selectivity was slightly

enhanced by replacing Cbz with Boc on the N-terminus

(compound 11), while the free amine (compound 10) showed

only 4-fold reduced activity against BACE1, and activities

against cathepsin E and D were reduced by 30−137-fold.

Compounds 12 and 13 (Figure 2) possessing isophthalic acid

residues with sulfonamide functionalities at the Gln position (in

2) retained signiﬁcant activity against BACE1, and we expected

these tasiamide B−isophthalic acid dipeptide hybrid molecules

to possess better cellular activity (see below). The activity is

consistent with previous reports that isophthalic acid derivatives

at the S2 position possess potent BACE1 inhibitory activity.^6,16

Other modiﬁcations or further truncation on the N-terminus

Figure 1. APP processing by secretases.

(metalloproteases ADAM9, 10 and/or 17), which diverts

away from the potentially amyloidogenic Aβ genesis pathway

by generating sAPPα and C83. The amyloid cascade hypothesis

predicts that lowering Aβ in the brain should be disease

modifying,¹⁰and the molecular pathways that aﬀect Aβ levels

have been major targets in eﬀorts to develop therapeutic agents

for AD.¹¹BACE1 has been receiving considerable attention as a

therapeutic target¹²since BACE1 knockout mice are viable¹³as

compared to embryonic lethal presenilin (γ-secretase) knock-

outs.¹⁴Partial reduction of BACE1 in vivo leads to diminished

AD-like pathology in transgenic mouse models.¹⁵

RESULTS AND DISCUSSION

■

Synthesis and Structure−Activity Relationship (SAR)

Studies in Vitro. We have previously reported the lack of

activity of grassystatins A−C against BACE1, while these

compounds potently inhibit two other aspartic proteases,

cathepsins D and E, and to diﬀerent extents.³The central

statine-type unit is crucial for the aspartic protease inhibitory

activity. Tasiamide B (2), which we recently synthesized,⁵

leading to the revision of the originally proposed structure (1,

L-Phe),⁴contains a diﬀerent statine unit, namely, phenylstatine.

In light of the grassystatin data, we tested tasiamide B (2)

against these three aspartic proteases in enzymatic in vitro

assays and found that 2 inhibited BACE1 at sub-micromolar

concentrations, but also lost the diﬀerential activity and potency

against cathepsins D and E (Table 1). While the diastereomeric

compounds 3 (^L-Phe-D-Ala) and 4 (D-Phe) lost their aspartic

protease inhibitory activity almost completely, tasiamide B

analogue 5 lacking the prolyl methylester at the C-terminus

relative to 2 not only restored but also improved on all

activities, including BACE1 inhibition. This led us to design and

10750

dx.doi.org/10.1021/jm301630s | J. Med. Chem. 2012, 55, 10749−10765

Journal of Medicinal Chemistry

Article

Table 1. Structures of Aspartic Protease Inhibitors and Inhibition of Enzymatic Activities

a

b

c

IC₅₀positive control (β-secretase inhibitor IV) 250 nM. IC₅₀positive control (pepstatin A) 0.693 nM. IC₅₀positive control (pepstatin A) 0.446

d

nM. Binding eﬃciency index, BEI = (pIC₅₀)/(molecular weight).

led to reduced activities against all or a subset of the three

proteases tested (Table 1).

BACE1−Inhibitor Crystal Structure Analyses. To inter-

rogate the structural basis for activity and SAR, we determined

10751

dx.doi.org/10.1021/jm301630s | J. Med. Chem. 2012, 55, 10749−10765

Journal of Medicinal Chemistry

Article

a

Scheme 1. Synthesis of Building Blocks

a

Reagents and conditions: (a) H-Val-OBn, EDC, HOAt, DIEA, DCM, 99%; (b) NaOH, THF/MeOH/H₂O, 64% for 20, 97% for 21; (c) (R)-1-

phenylethanamine, EDC, HOAt, DIEA, DCM, 57% for 22, 39% for 23; (d) H₂, Pd−C, EtOAc; (e) Cbz-MeGln(Trt)−OH, EDC, HOAt, DIEA,

THF, 64%; (f) Cbz-Gln(Trt)−OH, EDC, HOAt, DIEA, DCM, 65%; (g) Ac₂O, MeOH, 84%; (h) 4 M HCl-EtOAc; (i) Boc-Ala-OH, EDC, HOAt,

DIEA, DCM, 74%; (j) Boc-Leu-OH, EDC, HOAt, DIEA, DCM, 69%; (k) K₂CO₃, MeI, DMF, 99%; (l) Boc-Me-D-Phe-OH, EDC, HOAt, DIEA,

DCM, 96%; (m) Cbz-Ala-OH, EDC, HOAt, DIEA, DCM, 81%; (n) Boc-Leu-OH, EDC, HOAt, DIEA, DCM, 96%; (o) Boc-^D-Phe-OH, EDC,

HOAt, DIEA, DCM, 84%; (p) Boc-Ala-OH, EDC, HOAt, DIEA, DCM, 86%; (q) Boc-Leu-OH, EDC, HOAt, DIEA, DCM, 70%.

a

Scheme 2. Synthesis of Compounds 5−8

a

Reagents and conditions: (a) LiOH, THF/MeOH/H₂O; (b) Pd/C, EtOAc; (c) HATU, HOAt, DIEA, THF, 45% for 37, 80% for 40; (d) TFA,

DCM; (e) 4 M HCl-EtOAc; (f) HATU, HOAt, DIEA, 56% for 39, 20% for 42; (g) H₂, Pd−C, MeOH; (h) BnNH₂, EDC, HOAt, NaHCO₃, 76%;

(i) isopropyl amine, EDC, HOAt, NaHCO₃, 76%.

the crystal structures of compounds 2, 5, and 11−13 in

complex with BACE1. The crystalline complexes of BACE1−2,

BACE1−5, and BACE1−11 were obtained by soaking the

compounds into the orthorhombic apo BACE1 crystals (space

group, C222₁), which diﬀracted to 2.08, 1.59, and 1.77 Å,

respectively (Table S1). We were able to obtain the cocrystals

́

10752

dx.doi.org/10.1021/jm301630s | J. Med. Chem. 2012, 55, 10749−10765

Journal of Medicinal Chemistry

Article

a

Scheme 3. Synthesis of Compounds 9−13

a

Reagents and conditions: (a) (i) Pd−C, H₂, EtOAc; (ii) Cbz-Val-OH, EDC, HOAt, DIEA, DCM, 82% over two steps; (b) TFA, DCM; (c) HATU,

HOAt, DIEA, DCM, 64%; (d) Pd−C, H₂, MeOH, 83%; (e) Boc₂O, DCM, 60%; (f) NaOH, THF/MeOH/H₂O; (g) Pd−C, H₂, MeOH; (h) EDC,

HOAt, DIEA, DCM, 81% for 45, 70% for 46; (i) HATU, HOAt, DIEA, DCM, 69% for 12, 60% for 13.

a

Scheme 4. Synthesis of Compounds 14−16

a

Reagents and conditions: (a) (i) Pd−C, H₂, EtOAc; (ii) Ac₂O, MeOH, 83% over two steps; (b) TFA; (c) HATU, HOAt, DIEA, THF, 86% over

three steps for 14; 75% over three steps for 15; 98% over three steps for 16.

of compounds 12 and 13 with the enzyme by mixing the

solutions of the BACE1−12 and BACE1−13 complexes with

1.0 M ammonium sulfate/100 mM sodium citrate, pH 5.0.

These crystals belonged to the P2₁2₁2₁space group and

of residues P70, T72, G230, and T232 with the compound

were seen in all ﬁve complexes. In some structures, more than

one H-bond formed between the residues and the compound.

For example, in the complex of BACE1−2, three H-bonds

formed between T232 and compound 2. Residues such as

N233 also interact with the compound via a water molecule,

which can be found in the LIGPLOT plots (Figures 4 and S1).

Due to the low diﬀraction resolution, water molecules were not

visualized in the complex structure of BACE1−13. Residues

Y71, F108, W115, and I126 interact with the compounds

mainly through hydrophobic interactions, and Y71 in particular

always contributes the most among all the residues’ hydro-

phobic interaction with the compound in each complex (Table

S2). The total number of paired atoms which formed

hydrophobic interactions between the enzyme and the

compound are 30, 24, 36, 51, and 57 in complex structures

of BACE1/2, BACE1/5, BACE1/11, BACE1/12, and BACE1/

13, respectively. Therefore, compounds 12 and 13 form more

hydrophobic interactions with the enzyme than the other three

compounds do because of the two additional benzene rings

́

diﬀracted to 1.8 and 3.06 Å for BACE1−12 and BACE1−13,

respectively (Table S1). In general, the compounds ﬁt very well

into the binding groove between the C-terminal and N-terminal

lobes of BACE1 where the two catalytic residues D32 and

D228 are located (Figure 3). The hairpin loop in the N-

terminal lobe known as the ‘ﬂap’ (residues 67−75) interacted

directly with the bound compounds and thus adopted a close

conformation, which is highly similar to the conformation seen

in the ﬁrst crystal structure of BACE1 in complex with peptoid

OM99-2.¹⁷The binding position of the ﬁve compounds within

the enzyme is very similar since the superimposition of the ﬁve

compounds after ﬁtting the Cα atoms of the protein shows that

the compounds were overlapped (Figure 3). The program

LIGPLOT¹⁸was used to examine the interactions between the

compound and BACE1. Table 2 lists all the residues which

formed the direct hydrogen bonds (H-bonds) with the

compounds in the ﬁve complexes. The H-bond interactions

10753

dx.doi.org/10.1021/jm301630s | J. Med. Chem. 2012, 55, 10749−10765

Journal of Medicinal Chemistry

Article

Table 2. The Number of H-Bonds between the Inhibitor and

Residues in the Crystal Structures of BACE1 in Complex

with Compounds 2, 5, 11, 12, and 13

2

5

11

12

13

G11

1

2

1

G34

1

2

1

2

1

2

P70

1

2

1

T72

Q73

Y198

K224

G230

T231

T232

N233

S325

2

1

3

1

3

1

2

1

that regard, we found that 9 is extremely potent against

cathepsin D (IC₅₀78.3 pM) and 9-fold selectivity for cathepsin

D over cathepsin E, a reverse selectivity trend. This selectivity is

even more pronounced for 11 (15-fold) at the expense of some

slight potency loss against cathepsins. This compound was also

our most potent BACE1 inhibitor in our in vitro enzymatic

assays (IC₅₀48.8 nM). Both compounds 9 and 11 are

characterized by their carbamate moiety on the N-terminus.

When we substituted the two N-terminal residues with a

modiﬁed dipeptide containing an isophthalic acid-based

sulfonamide functionality at the Gln position (12 and 13),

we retained all aspartic protease inhibitory activity. The N-

methylated sulfonamide had slightly better activity against

BACE1; however, activity against cathepsin D and particularly

cathepsin E was also enhanced. Truncation of the N-terminus

resulting in compounds 14 and 15 increased the selectivity for

cathepsins D and E over BACE1; in fact 14 which is the

simplest analogue showed 69-fold selectivity for cathepsin D

over cathepsin E and >140-fold selectivity over BACE1.

Removal of the Pro residue on the C-terminus (compound

16) reduced the inhibitory eﬀects on cathepsin D without

dramatically changing the activity against cathepsin E. Thus, the

Figure 2. Fragments in compounds 9−13.

(Table 2), which may explain why these compounds have high

inhibition activity to BACE1 in the cell.

Activity against Cathepsins D and E. For all analogues

we also assessed the selectivity index for cathepsins D/E, which

may provide the basis for application of certain compounds as

chemical tools to probe cathepsin D or cathepsin E function.

For example, we had previously reported that the natural

product grassystatin A shows 30-fold selectivity for cathepsin

E,³and our synthetic compound showed the same selectivity

trend (50-fold preference for cathepsin E).⁷Most notably in

Figure 3. Molecular surface of the binding pocket for compounds 2 (A), 5 (B), 11 (C), 12 (D), and 13 (E) in their complex structures with BACE1.

(F_o− F_c) diﬀerence electron-density maps contoured at 3.0 σ for all ﬁve crystal structures. (F) Superimposition of ﬁve crystal structures of BACE1−

2, BACE1−5, BACE1−11, BACE1−12, and BACE1−13 by ﬁtting the Cα atoms of the protein. Five inhibitors 2 (green), 5 (cyan), 11 (magenta),

12 (yellow), and 13 (light brown) were displayed as sticks.

10754

dx.doi.org/10.1021/jm301630s | J. Med. Chem. 2012, 55, 10749−10765

Journal of Medicinal Chemistry

Article

(Figure S2)²⁴increased the levels of C83 and C99 since both

fragments are substrates of γ-secretase (Figure 1).

Measurement of Aβ Production Downstream of β-

and γ-Secretase Activity. Next we measured secreted

amounts of Aβ peptides formed upon sequential action of β-

and γ-secretases in CHO 2B7 cells (clonal expression of

APP695 wt). Speciﬁcally, we measured total Aβ and

individually Aβ40 and Aβ42 upon treatment with the most

promising compounds based on HPLAP cellular and BACE1

enzyme inhibitory activity. Cellular activity in CHO cells was

validated for six compounds (Figure 6). Compounds 2, 9, 10

and 11 showed activity in the micromolar range (Figure 6) and

therefore exhibited approximately 10−100-fold lower potency

than in the enzymatic BACE1 assays. Compounds 12 and 13,

which together with 10 showed the best activity in the HPLAP

cellular assay for sAPPβ reduction at 10 μM (Figure 7), were

even active in the sub-micromolar range with IC₅₀values near

100 and 10 nM, respectively. These values for 12 and 13 closely

correspond to the IC₅₀values for the in vitro BACE1 inhibition

(Table 1), suggesting that both compounds have excellent cell

penetrating ability.

Selectivity for β-Secretase over γ-Secretase. Our

previous assays were unable to eliminate the possibility that

the BACE1 inhibitors are additionally γ-secretase inhibitors.

Speciﬁcally, sAPPβ levels are independent of γ-secretase action,

while the observed inhibition of Aβ production is the net result

of the sequential action of both β- and γ-secretase. Given that

β- as well as γ-secretase have aspartic protease activity, and

considering the observed eﬀects on cathepsins D and E, it was

quite possible that our aspartic protease inhibitors may not have

been able to diﬀerentiate both secretases. To speciﬁcally test for

γ-secretase inhibition, we used H4 cells overexpressing the BRI-

C99 fusion protein and treated them with our compounds (10

μM). The BRI-C99 fusion protein utilizes a fusion between the

BRI2 (ITM2B) portion and the terminal 99 amino acids of

APP.²⁵This construct does not require β-secretase for

processing but is processed by furin and furin-like proteases

to produce C99, which is equivalent to the β-secretase derived

fragment from APP. Lack of accumulation with inhibitor

treatment (Figure 8) indicates that these compounds are not γ-

secretase inhibitors,²⁶which would lead to an intensity increase

of the C99 fragment, as seen for the positive control LY411575

(Figure 8).

Stability and in Vivo Activity. The in vitro plasma and

cellular stabilities of 12 and 13, together with identiﬁcation of

the major biotransformation products, were assessed using

HPLC-MS. Both compounds showed reasonable stability and

were primarily converted to their corresponding acids 51 and

52 (Figure 9A) in both mouse serum (Figure 9B) and when

incubated with H4 cell lysates (Figure 9C). Notably, acids 51

and 52 retained nanomolar BACE1 inhibitory activity (IC₅₀

56.3 and 369 nM), suggesting that at least intracellular ester

hydrolysis would not be detrimental to activity. Conversely, one

might expect that the cellular activity would be reduced if

hydrolysis occurs before cell penetration; e.g, the similar acid 5

(Table 1) displayed potent activity against the enzyme but not

in a cellular context (see Figure 5C). However, the ester

hydrolysis occurred only slowly in vitro; 12 and 13 possessed a

presumably suﬃcient half-life of several hours (Figure 9B,C).

To assess the potential to cross the blood−brain barrier (BBB),

we tested 12 and 13 in the parallel artiﬁcial membrane

permeability assay (PAMPA) using a ﬁlter-immobilized BBB

lipid formulation. Compound 12 gave P_e(× 10⁻⁶cm/s) of 10.7

Figure 4. LIGPLOT representations of the interactions of inhibitor 11

with the enzyme in the crystal structure complex. Dashed lines

represent hydrogen bonds, and spiked residues form hydrophobic

interactions with 11.

selectivity of tasiamide B-based protease inhibitors can be tuned

through the editing or modiﬁcation of both termini.

Measurement of Immediate Downstream Eﬀects of

Cellular BACE1 Inhibition. A human placental alkaline

phosphatase-amyloid precursor protein (HPLAP-APP) fusion

protein¹⁹was used to quantitatively measure secreted APP

(sAPP). Control BACE1 inhibitors II (BI−II)²⁰and IV (BI−

IV)²¹(Figure S2) alone did not aﬀect total sAPP (Figure 5B)

due to the well-documented shift from β- to α-cleavage since α-

and β-secretases both compete for the APP substrate.²²In

order to distinguish sAPPβ from total sAPP, transfected cells

were treated with 25 μM of the metalloprotease inhibitor

Batimastat (BB-94, Figure S2)²³along with the various

inhibitors in order to speciﬁcally quantitate eﬀects on sAPPβ

by inhibiting α-secretase activity (Figure 5). The production of

sAPPβ, which corresponds to the total sAPP in the presence of

Batimastat (BB-94, Figure 5B), was further reduced by BI−II

and BI−IV (Figure 5B,C) and also by many test compounds

(Figure 5C). Strongest reduction in sAPPβ at the 10 μM

concentration was observed for compound 13 followed by 12,

10 and then 2.

We then tested the most active compounds from the reporter

assay to determine eﬀects on endogenous levels of the CTF,

C99. Consistent with the reporter assay, compounds 10, 12 and

13 prevented the generation of C99, without aﬀecting C83

levels. These data clearly indicate that these compounds are

BACE1 inhibitors with cellular activity and do not aﬀect α-

secretase (Figure 6). The γ-secretase inhibitor LY411575

10755

dx.doi.org/10.1021/jm301630s | J. Med. Chem. 2012, 55, 10749−10765

Journal of Medicinal Chemistry

Article

Figure 5. Reporter assay to detect cellular β-secretase activity. (A) Assay design. Our HPLAP-APP is processed by α- and β-secretase to generate

HPLAP-sAPPα and HPLAP-sAPPβ, respectively. By inhibiting α-secretase with batimastat, only HPLAP-sAPPβ is generated and then quantitated.

(B) Validation of detection of sAPPβ. Cells were treated for 16 h in serum-free media and sAPP was quantitated from the media by measuring

HPLAP activity. Addition of β-secretase inhibitor II or IV was at a ﬁnal concentration 10 μM. Only cells treated with BB-94 can diﬀerentiate the

sAPPβ from total sAPP. Statistical analysis was by one-way Anova followed by Dunnett’s multiple comparison test. (C) Transfected cells were

simultaneously treated with BB-94 and the individual compounds at 10 μM for 16 h and assayed for HPLAP activity as in (B). The results of the

individual compounds are shown. All compounds except 15 show statistically signiﬁcant inhibition p < 0.01. Compounds 10, 12, and 13 were

identiﬁed as the most potent in the set against the β-secretase activity.

peptides, modiﬁed peptides and peptidomimetics possess P_e

values of 0.6−82.5 × 10⁻⁶cm/s in a gastrointestinal tract model

system.^27,28Evaluation of the in vivo stability of 12 and 13 from

mice treated with 30 mg/kg by intraperitoneal (i.p.) injection

showed that after 3 h both compounds are present in the

plasma together with lower levels of their corresponding acids

51 and 52 (Figure 9D). The same acute treatment led to a

signiﬁcant decrease in rodent Aβ40 for treatment with 13 but

not with 12 (Figure 9E), which correlates with the 10-fold

higher potency of 13 compared with 12 in the secreted Aβ

assay using CHO 2B7 cells (Figure 7). In order to rationalize

the observed in vivo Aβ40 levels for 12 and 13, we determined

the concentrations of these compounds in the cerebellum and

cortex of test animals using HPLC-MS (Figure 9F). Relative to

the plasma levels, 2.5% of 12 and 5.8% of 13 are delivered to

the brain (Figure 9F). These values are about 2−4-fold lower

compared to those reported for other BACE1 inhibitors such as

GRL-8234, where approximately 10% brain penetration was

reported.²⁹The high molecular weight and number of

hydrogen bonding groups are possible contributors to the

lower levels of 12 and 13 in the brain. Nonetheless, because of

the potent activity of 13, it is able to signiﬁcantly decrease Aβ40

levels in vivo. The detected concentration of 13 in the brain is

around 20−40-fold in excess of its cellular IC₅₀for Aβ40 and

Aβ42 secretion by ELISA (Figure 7). In contrast, brain

concentration of 12 corresponds to only three times its cellular

IC₅₀. Other BACE1 inhibitors previously demonstrated to

reduce in vivo Aβ40 levels in cortical neurons typically showed

greater than 10-fold excess of their cellular IC₅₀in brain

tissues.^8,29−31

Figure 6. CTF immunoblot analysis. H4 cells were treated for 16 h

with BACE1 inhibitor IV (BI−IV, 10 μM), γ-secretase inhibitor

LY411575 (1 μM), or with compounds 10, 12 and 13 (10 μM) or

solvent control, and cell lysates analyzed. BI−IV and the prioritized

test compounds prevented the generation of C99, indicating cellular

BACE1 inhibition, while LY411575 led to accumulation of both

fragments.

1.41 and 8.58 0.47 at pH 5.0 and 7.4, respectively, which

corresponds to intermediate BBB permeability compared with

controls (Table 3). We failed to accurately determine the

permeability coeﬃcient for 13, which may be in part due to

possible irreversible adsorption of the compound to the lipid-

coated immobilized ﬁlter, giving variable concentration in the

acceptor well. Small molecule transport across the BBB can

occur through both passive and active modes, whereas PAMPA

only accounts for the contribution of the former. It is possible

that 12, 13 and related congeners may be transported across

the BBB in vivo via both mechanisms, thus only partially

accounted for by this permeability assay, but additional

processes cannot be ruled out either. Compared to other

peptidomimetics, compound 12 has intermediate permeability

in a BBB system based on the permeability coeﬃcient; other

10756

dx.doi.org/10.1021/jm301630s | J. Med. Chem. 2012, 55, 10749−10765

Journal of Medicinal Chemistry

Article

Figure 7. Secreted β-amyloid (Aβ) assay. CHO 2B7 cells (clonal expression of APP695 wt) were incubated for 16 h with compound and analyzed by

direct sandwich Aβ ELISA. Control compounds: 10 μM BACE inhibitor II, 10 μM BACE inhibitor IV, and 1 μM LY411575 γ-secretase inhibitor. All

treatments were performed in triplicate. Means and SEM are shown.

lography, and a multiplexed disease-relevant molecular proﬁling

platform to characterize potential lead compounds and

chemical probes.

EXPERIMENTAL SECTION

■

Synthesis. General Experimental Procedures. Solvents were

puriﬁed by standard methods. TLCs were carried out on Merck 60

Figure 8. Immunoblot analysis for γ-secretase activity. H4 cells

overexpressing the BRI-C99 fusion were treated for 16 h with the

indicated compounds (10 μM). Lysates were separated by SDS-PAGE

and probed for the amounts of C99 generated by furin-like proteases

(instead of β-secretase). C99 is accumulated upon inhibition of γ-

secretase with LY411575 (100 nM).

F₂₅₄silica gel plates and visualized by UV irradiation or by staining

with iodine absorbed on silica gel, ninhydrin solution or with aqueous

acidic ammonium molybdate solution as appropriate. Flash column

chromatography was performed on silica gel (200−300 mesh,

Qingdao, China). Petroleum ether used as eluting solvent in this

paper was the fraction of boiling range 60−90 °C. Optical rotations

were measured using a JASCO P-1020 digital polarimeter. NMR

spectra were recorded on a Varian 400 MHz spectrometer. Chemical

shifts are reported in parts per million (ppm), relative to the signals

We identiﬁed a cyanobacterial metabolite as a BACE1 inhibitor,

due to the solvent. Data are described as follows: chemical shift,

CONCLUSION

■

prepared a series of related aspartic protease inhibitors with

multiplicity (s = singlet, d = doublet, t = triplet, q = quartet, br =

diﬀerential selectivities against BACE1 and cathepsins D and E,

and obtained inhibitor-bound BACE1 crystal structure

complexes to determine the structural basis for BACE1

inhibition. We combined the natural product scaﬀold with

structural features at the S2 position that are known to increase

cellular activity against BACE1. This merger led us to hybrid

molecules that showed inhibitory activity in a variety of assays,

in which we probed the eﬀects on APP processing. Among APP

processing secretases, the resulting compounds speciﬁcally

aﬀected β-secretase. Hybrids 12 and 13 were fairly stable in

vitro and in vivo, and compound 13 showed eﬃcacy in

reducing Aβ levels in a preliminary acute treatment study. This

work exempliﬁes the power of a drug discovery platform

combining natural products chemistry, chemical synthesis,

structure-based design and protein−inhibitor complex crystal-

broad, m = multiplet), coupling constants (Hz), integration, and

assignment. Mass spectra were recorded on a Q-Tof Ultima Global

mass spectrometer. Compound purity of ≥95% was conﬁrmed by

HPLC.

Compound 37. A cold solution (ice bath) of 17 (86.0 mg, 0.31

mmol) in THF-MeOH (4:1, V/V, 5.0 mL) was treated with a solution

of LiOH (1 M in H₂O, 2.0 mL). After being stirred at rt for 3 h, the

mixture was recooled to 0 °C and adjusted to pH = 3 with 1 M HCl,

then extracted with EtOAc (30 mL × 3). The combined organic layers

were washed with brine (15 mL × 2), dried (Na₂SO₄), and

concentrated in vacuo to give the free carboxylic acid as a white

solid, which was used without further puriﬁcation.

A solution of 25 (127.1 mg, 0.20 mmol) in EtOAc (10 mL) was

hydrogenated for 2 h at 1 atm in the presence of a catalytic amount of

Pd/C. The solution was ﬁltrated and concentrated to give a colorless

oil (amine), which was used without further puriﬁcation.

10757

dx.doi.org/10.1021/jm301630s | J. Med. Chem. 2012, 55, 10749−10765

Journal of Medicinal Chemistry

Article

Figure 9. Stability and eﬃcacy studies. (A) Structures of prioritized compounds 12 and 13 and their corresponding primary metabolites, ester

hydrolysis products 51 and 52, respectively. (B, C) Plasma and cellular stabilities of 12 and 13 were assessed by HPLC-MS following extraction of

the compounds and their biotransformation products. (B) In vitro mouse serum stability of 12 and 13 showed a time-dependent decrease in

concentration and accompanied by an increase in concentration of the corresponding ester hydrolysis product 51 or 52. Data points depict mean

SEM (n = 2). (C) Cellular stability of 12 and 13 assessed by incubation with whole-cell lysate of H4 cells (1.0 mg/mL). Data points depict mean

SEM (n = 2). (D−F) Acute treatment of CF-1 mice. (D) Compound pairs 12/51 and 13/52 were detected in plasma of mice treated with 30 mg/kg

of 12 or 13 (i.p. injection) after 3 h. Graphs illustrate mean + SEM (n = 6). (E) Brain concentrations of Aβ40 determined by sandwich ELISA,

measured 3 h after i.p. injection with 10 mg/kg of LY411575 or 30 mg/kg of 12 and 13, following diethylamine extraction. (F) In vivo brain levels of

12 and 13 (after 3 h) determined by HPLC-MS. Percentage of brain absorption is indicated (relative to plasma levels). Graphs for D−F illustrate

mean + SEM (n = 6 per group). Diﬀerences from the vehicle-treated group: *p < 0.05 or ***p < 0.001 by one-way Anova followed by Dunnett’s

post test.

Table 3. Permeability Coeﬃcient (P_e, × 10⁻⁶cm/s)

(m, 1H, Ahppa NCH), 4.04 (m, 1H, La α-H), 3.94 (m, 1H, Ahppa

CHOH), 3.77 and 3.65 (br, 1 H, Ahppa OH), 2.89 (m, 1H, Ahppa H-

5a), 2.88 and 2.45 (s, 3H, N-CH₃), 2.87 (m, 1H, Ahppa δ-HaHb), 2.77

(m, 1H, Ahppa δ-HaHb), 2.51 (m, 1H, Gln γ-HaHb), 2.33−2.18 (m,

3H, Gln γ-HaHb + Ahppa α-H₂), 1.97 (m, 2H, Gln β-HaHb + Val β-

H), 1.72 (m, 1 H, Gln β-HaHb), 1.51 and 1.27 (d, 3H, J = 6.8 Hz, La

β-H₃), 1.42 and 1.37 (s, 9H, C(CH₃)₃), 1.27, 0.99, 0.94, and 0.80 (m,

6H, Val γ-H × 6); ESI-MS (m/z): 911.5 [M + H]⁺, 933.5 [M + Na]⁺

(Calcd 910.5).

Compound 5. To a stirred solution of 37 (71.2 mg, 0.078 mmol) in

dichloromethane (5 mL) was added TFA (2 mL). The mixture was

stirred at room temperature for 1 h and then concentrated under

reduced pressure to provide crude 38.

Determined using the BBB Parallel Artiﬁcial Membrane

Permeability Assay (mean SD)

compound

12

pH 7.4/7.4_sink

pH 5.0/7.4_sink

8.58 0.47

10.7 1.41

13

not determined

17.30 0.02

not determined

16.00 0.94

0.12 0.20

a

corticosterone

b

theophylline

0.09 0.10

a

b

Highly BBB permeable control. Impermeable control.

To the cold solution (ice bath) of the above carboxylic acid and

Building block 29 (47.63 mg, 0.086 mmol) in 4 M HCl/EtOAc (2

mL) was stirred at r.t. for 1 h, and then concentrated to dryness to give

the corresponding amine, which was used without further puriﬁcation.

To the cold solution (ice bath) of the amine obtained above and the

crude 38 in THF were added HATU (60.8 mg, 0.16 mmol), HOAt

(21.2 mg, 0.16 mmol), and DIEA (38.7 μL, 0.23 mmol). The mixture

was stirred at 0 °C for 2 h, and then at r.t. for another 26 h. The

solvent was removed under reduced pressure; the residue was

dissolved in EtOAc (70 mL), washed with 10% citric acid (10 mL ×

2), 5% NaHCO₃(10 mL × 2) and brine (10 mL × 2), dried (Na₂SO₄)

and concentrated in vacuo. The residue was puriﬁed by chromato-

graphy on silica gel (DCM/MeOH = 30:1 → 20:1) to give the

protected heptapeptide 39 (46.0 mg, 56%).

amine in THF were added PyAOP (153.3 mg, 0.29 mmol), HOAt

(40.1 mg, 0.29 mmol), and DIEA (97.2 μL, 0.59 mmol). The mixture

was stirred at 0 °C for 2 h and then at room temperature for 26 h.

After most of the solvent was removed at reduced pressure, the residue

was dissolved in EtOAc (100 mL) and washed with 10% citric acid (10

mL × 2), 5% NaHCO₃(10 mL × 2), and brine (10 mL × 2). After

drying over Na₂SO₄, the organic layer was concentrated, giving crude

37, which was puriﬁed by preparative TLC (petroleum ether/EtOAc =

1:1) (80.0 mg, 45%). [α]_D²⁰= −70.90 (c = 1.00, CHCl₃); H NMR

1

(CDCl₃, 400 MHz) δ: 7.90 and 6.46 (d, 1 H, J = 9.4 Hz, Ahppa NH),

7.15−7.41 (m, 26 H, Ar-H × 25 + Val NH), 7.06 (s, 1 H, Gln δ-NH),

4.95 and 4.70 (m 1H, Gln α-H), 4.63 (m, 1H, PhCHaHbO), 4.51 (t,

1H, J = 6.9 Hz, Val α-H), 4.50 and 4.32 (m, 1H, PhCHaHbO), 4.23

10758

dx.doi.org/10.1021/jm301630s | J. Med. Chem. 2012, 55, 10749−10765

Journal of Medicinal Chemistry

Article

A solution of 39 (41.9 mg, 0.04 mmol) in MeOH (5 mL) was

hydrogenated overnight in the presence of a catalytic amount of Pd/C.

The solution was ﬁltrated, concentrated and puriﬁed by preparative

TLC to give 5 (33.1 mg, 95.4%) as a white solid. [α]²_D²= −73.13 (c =

3H, J = 6.3 Hz, Ala β-CH₃); ¹³C NMR (CDCl₃, 100 MHz) δ: 176.7,

175.9, 174.1, 173.5, 172.6, 171.9, 170.1, 168.6, 137.9, 137.1, 129.0,

128.7, 128.3, 126.5, 126.4, 69.6, 68.5, 57.0, 55.3, 54.3, 53.7, 52.0, 45.6,

41.5, 40.9, 37.7, 34.0, 31.6, 31.1, 30.8, 30.6, 30.0, 29.7, 24.6, 23.7, 23.0,

22.9, 22.4, 22.3, 21.8, 20.7, 19.5, 17.7, 16.3; ESI-MS (m/z): 909.5 [M

+ H]⁺, 931.5 [M + Na]⁺(Calcd 908.5).

1

0.34, CHCl₃); H NMR (CD₃OD, 400 MHz) two rotamers δ: 7.13−

7.25 (m, 10 H, Ar-H × 10), 5.36 and 5.24 (m, 1H, Phe α-H), 4.73 and

4.68 (m, 1H, Gln α-H), 4.60 and 4.55 (m, 1H, Ala α-H), 4.35 (m, 1H,

Val α-H), 4.23 (m, 1H, Ahppa CHOH), 4.16 (q, 1H, J = 7.2 Hz, La α-

H), 4.10 (m, 1H, Ahppa NCH), 3.46 (d-like, 1H, J = 10.7 Hz, Phe β-

Ha), 3.00 (m, 2H, Phe β-Hb + Ahppa δ-Ha), 2.82 (m, 1H, Ahppa δ-

Hb), 2.53 (m, 1H, Ahppa α-Ha), 2.45 (m, 1H, Ahppa α-Hb), 2.35 (m,

1H, Gln γ-Ha), 2.17 (m, 1H, Gln γ-Hb), 1.91−2.04 (m, 2H, Leu β-Ha

+ Leu γ-H), 1.75 (m, 1H, Leu β-Hb), 1.61−1.70 (m, 2H, Val β-H +

Gln β-Ha), 1.55 (m, 1H, Gln β-Hb), 1.38 and 1.48 (d, 3H, J = 6.7 Hz,

La CH₃), 0.90−1.35 (m, 12H, Val γ-H × 6 + Leu δ-H × 6), 0.87 (d,

3H, J = 7.6 Hz, Ala β-CH₃); ¹³C NMR (CD₃OD, 100 MHz) δ: 178.5,

178.1, 177.8, 177.6, 175.4, 175.2, 175.0, 174.7, 174.6, 174.2, 174.1,

173.6, 172.0, 170.6, 139.9, 139.7, 139.5, 130.4, 130.3, 129.9, 129.7,

129.5, 129.4, 129.3, 127.5, 127.4, 71.3, 70.2, 69.1, 69.0, 61.0, 60.0, 57.9,

56.3, 55.8, 55.5, 55.3, 54.2, 53.2, 47.3, 46.8, 41.3, 41.2, 40.7, 38.7, 38.2,

36.2, 36.0, 32.5, 32.3, 32.0, 31.8, 30.8, 29.1, 26.6, 25.8, 25.3, 23.7, 23.5,

21.8, 21.7, 21.5, 20.2, 19.9, 18.9, 18.0, 17.2, 16.6; ESI-MS (m/z): 868.5

[M + H]⁺, 890.5 [M + Na]⁺(Calcd 867.5).

Compound 8. Similar as described for compound 5. [α]²_D⁰= −81.45

(c = 0.23, CHCl₃); ¹H NMR (DMSO-d₆, 400 MHz) δ: 8.21 (d, 1 H, J

= 7.8 Hz, NH), 8.17 (d, 1 H, J = 6.7, 9.0 Hz, Phe-αH), 7.90 (m, 2H,

NH × 2), 7.67 (d, 1 H, J = 7.8 Hz, Phe-NH), 7.46 (d, 1 H, J = 9.4 Hz,

NH), 7.24 (s, 1H, Gln-δCNHa), 7.11−7.22 (m, 10H, ArH × 10), 6.77

(s, 1H, Gln-δCNHb), 5.99 (d, 1 H, J = 5.0 Hz, Ala-NH), 5.37 (d, 1 H,

J = 3.7 Hz, Gln-NH), 5.04 (q, 1 H, J = 6.3 Hz, Ahppa-NH), 4.21 (m,

4H, Val-αH + La-αH + Leu-αH), 3.96 (m,1H, Ala-αH), 3.87 (m, 3H,

Gln-αH + Ahppa-CHOH + Phe-αH), 2.77 (dd, 1H, J = 7.0, 13.4 Hz,

Ahppa-δHa), 2.68 (m, 2H, Pro-βH₂), 2.48 (m, 1H, Ahppa-δHb),

1.87−2.09 (m, 4H, Ahppa-αH₂+ Pro-βH₂), 1.77 (m, 1H, Val-βH),

1.57−1.68 (m, 4H, Pro-δH₂+ Gln-γH₂), 1.41 (m, 1H, Leu-γH), 1.23

(m, 2H, Leu-βH₂), 1.07 (d, 3H, J = 6.7 Hz, Ala-βCH₃), 0.81 (d, 3H, J

= 7.0 Hz, La-CH₃), 0.59−0.70 (m, 12H, Val-γH × 6 + Leu-δH × 6);

¹³C NMR (DMSO-d₆, 100 MHz) δ: 174.2, 173.8, 172.9, 172.2, 171.6,

170.9, 170.6, 169.3, 139.1, 137.2, 129.4, 129.2, 129.0, 128.1, 126.5,

67.5, 58.6, 56.4, 51.8, 47.9, 46.5, 31.4, 28.6, 28.0, 24.1, 23.2, 21.6, 21.3,

19.3, 18.4, 17.7; ESI-MS (m/z): 951.5 [M + H]⁺, 973.5 [M + Na]⁺

(Calcd 950.5).

Compound 6. To the cold solution (ice bath) of 5 (20.6 mg, 0.031

mmol) and benzylamine (4.3 μL, 0.037 mmol) in THF (5 mL) were

added EDC (11.9 mg, 0.062 mmol), HOAt (8.5 mg, 0.062 mmol), and

DIEA (14.8 μL, 0.093 mmol). The mixture was stirred at 0 °C for 2 h,

and then at r.t. for another 16 h. The solvent was removed under

reduced pressure; the residue was dissolved in EtOAc (70 mL) and

washed with 10% citric acid (10 mL × 2), 5% NaHCO₃(10 mL × 2)

and brine (10 mL × 2), dried (Na₂SO₄) and concentrated in vacuo.

The residue was puriﬁed by TLC (DCM/MeOH = 12:1) to give

Compound 43. A solution of 25 (205.9 mg, 0.26 mmol) in EtOAc

(15 mL) was hydrogenated for 2 h at 1 atm in the presence of a

catalytic amount of Pd/C. The solution was ﬁltrated and concentrated

to give the free amine, which was used without further puriﬁcation.

To the cold solution (ice bath) of the amine obtained above and

Cbz-Val-OH in DCM (20 mL) was added EDC (151.1 mg, 0.79

mmol), HOAt (107.2 mg, 0.79 mmol), and DIEA (208.7 μL, 1.31

mmol). The mixture was stirred at 0 °C for 2 h, and then at r.t.

overnight. The reaction mixture was diluted with EtOAc (100 mL)

and washed with 10% citric acid (10 mL × 2), 5% NaHCO₃(10 mL ×

2) and brine (10 mL × 2), dried (Na₂SO₄) and concentrated in vacuo,

1

compound 6 (17.2 mg, 76%). [α]²_D⁰= −52.99 (c = 0.67, CHCl₃); H

NMR (CDCl₃, 400 MHz) δ: 7.57 (d-like, 1H, Ala NH), 7.45 (d-like,

1H, Val NH), 7.40 (d, 1H, J = 5.9 Hz, BnNH), 7.07−7.28 (m, 17 H,

Ar-H × 15 + Leu NH + Ahppa NH), 6.61 and 6.70 (s, 2H, Gln δ-

CONH₂), 5.59 (dd, 1 H, J = 5.5, 11.0 Hz, Phe α-H), 4.92 and 5.02 (m,

1H, Gln α-H), 4.52 (q, 1H, J = 6.7, Ala α-H), 4.45 (dd, 1H, J = 6.7,

14.9 Hz, NHCHaHbPh), 4.37 (m, 2H, Leu α-H + Val α-H), 4.29 (dd,

1H, J = 5.5, 14.9 Hz, NHCHaHbPh), 4.29 (m, 1H, Ahppa NCH) 4.13

(q, 1H, J = 5.9 Hz, La α-H), 4.02 (m, 1H, Ahppa CHOH), 3.40 (dd,

1H, J = 5.9, 13.5 Hz, Phe β-Ha), 2.94 (m, 5H, Phe β-Hb + Phe N-CH₃

+ Ahppa δ-Ha), 2.80 (dd, 1H, J = 9.3, 13.7 Hz, Ahppa δ-Hb), 2.74 (s,

3H, Gln N-CH₃), 2.34 (m, 3H, Ahppa α-H₂+ Gln γ-Ha), 2.04 (m, 2H,

Gln β-Ha + Gln γ-Hb), 1.87 (m, 2H, Gln β-Hb + Val β-H), 1.53 (m,

1H, Leu γ-H), 1.47 (m, 2H, Leu β-H₂), 1.36 (d, 1H, J = 6.7 Hz, La−

CH₃), 0.92 (d, 3H, J = 7.0 Hz, Val γ-H₃), 0.89 (d, 3H, J = 6.7 Hz, Val

γ-H₃), 0.84 (d, 3H, J = 6.3 Hz, Leu δ-H₃), 0.79 (d, 3H, J = 6.3 Hz, Leu

δ-H₃), 0.76 (d, 3H, J = 6.7 Hz, Ala β-CH₃); ¹³C NMR (CDCl₃, 100

MHz) δ: 176.6, 175.9, 174.0, 172.6, 171.8, 170.1, 169.7, 138.4, 137.9,

136.9, 129.0, 128.7, 128.5, 128.4, 127.7, 127.1, 126.6, 126.4, 69.7, 68.6,

67.9, 56.9, 55.4, 54.3, 53.7, 52.0, 45.6, 43.4, 40.9, 40.2, 37.7, 34.2, 31.1,

30.9, 30.6, 30.0, 25.6, 24.6, 23.7, 23.0, 21.6, 20.7, 19.5, 17.7, 16.6; ESI-

MS (m/z): 957.5 [M + H]⁺, 979.5 [M + Na]⁺(Calcd 956.5).

puriﬁed by TLC to give compound 43 (190.8 mg, 82%). [α]_D²⁰

=

−93.76 (c = 0.14, CHCl₃); ¹H NMR (CDCl₃, 400 MHz) two

rotamers δ: 7.60 and 6.34 (d, 1 H, J = 8.6 Hz, Ahppa NH), 7.12−7.40

(m, 25 H, Ar-H × 25), 7.00 and 6.84 (s, 1 H, Gln δ-NH), 5.44 (m, 1H,

Val NH), 5.22 and 5.01 (d, 1H, J = 12.5 Hz, PhCHaHb), 5.09 and

4.88 (d, 1H, J = 12.5 Hz, PhCHaHb), 4.91 and 4.93 (m, 1H, Val α-H),

4.65 and 3.94 (d-like, 1H, J = 9.8 Hz, Ahppa CHOH), 4.25 and 4.11

(m, 1H, Ahppa NCH), 3.64 and 3.67 (s-like, 1H, Gln α-H), 2.71−2.94

(m, 2H, Phe β-H₂), 2.28−2.52 (m, 4H, Gln γ-H₂+ Ahppa α-H₂),

t

1.76−1.98 (m, 3H, Gln β-H₂+ Val β-H), 1.42 (s, 9H, Bu), 0.99 and

0.83 (d, 3H, J = 6.3 Hz, Val γ-CH₃), 0.92 and 0.80 (d, 3H, J = 6.3 Hz,

Val γ-CH₃); ¹³C NMR (CDCl₃, 100 MHz) two rotamers δ: 173.3,

172.6, 172.3, 171.0, 170.9, 170.7, 169.5, 168.5, 144.7, 144.6, 137.9,

137.7, 136.1, 135.6, 129.2, 128.7, 128.5, 128.1, 128.0, 127.9, 126.9,

126.5, 126.3, 81.7, 81.2, 70.4, 69.0, 67.6, 66.9, 66.7, 60.4, 58.9, 56.3,

56.0, 55.4, 54.1, 53.3, 40.9, 40.4, 39.5, 38.0, 37.6, 33.2, 31.2, 30.8, 30.7,

28.5, 28.0, 25.1, 23.3, 19.7, 19.2, 18.5, 17.3, 14.2; ESI-MS (m/z): 883.3

[M + H]⁺, 905.4 [M + Na]⁺(Calcd 882.5).

Compound 7. Prepared using the same method described for 6, but

Compound 9. To a stirred solution of 43 (88.8 mg, 0.101 mmol) in

dichloromethane (2 mL) was added TFA (2 mL). The reaction

mixture was stirred at room temperature for 1 h and then concentrated

under reduced pressure to give the corresponding free carboxylic acid

(44), which was used without further puriﬁcation.

To a stirred solution of 33 (69.35 mg, 0.121 mmol) in

dichloromethane (2 mL) was added TFA (2 mL). The reaction

mixture was stirred at room temperature for 1 h and then concentrated

under reduced pressure to give the corresponding amine, which was

used in the next step without further puriﬁcation.

To the cold solution (ice bath) of the carboxylic acid 44 and amine

obtained above in DCM (10 mL) were added HATU (76.8 mg, 0.202

mmol), HOAt (27.5 mg, 0.20 mmol), and DIEA (64.2 μL, 0.404

mmol). The mixture was stirred at 0 °C for 2 h, and then at r.t. for

another 24 h. The solvent was removed under reduced pressure; the

residue was dissolved in EtOAc (100 mL) and washed with 10% citric

replacing the benzylamine with isopropyl amine. Yield 76%. [α]_D²⁰

=

1

−62.97 (c = 0.64, CHCl₃); H NMR (CDCl₃, 400 MHz) δ: 7.63 (d,

1H, J = 5.9 Hz, Ala NH), 7.47 (d, 1H, J = 7.8 Hz, Val NH), 7.33 (d,

1H, J = 7.3 Hz, Leu NH), 7.24 (d, 1H, J = 4.4 Hz, Ahppa NH), 7.16

(m, 10 H, Ar-H × 10), 6.67 and 6.78 (s, 2H, Gln δ-CONH₂), 6.64 (d,

1H, J = 7.8 Hz, Isp NH), 5.50 (dd, 1 H, J = 5.7, 10.7 Hz, Phe α-H),

5.06 (m, 1H, Gln α-H), 4.50 (m, 1H, Ala α-H), 4.40 (m, 2H, Leu α-H

+ Val α-H), 4.26 (q, 1H, J = 6.6 Hz, Ahppa NCH), 4.14 (q, 1H, J = 6.8

Hz, La α-H), 0.4.03 (m, 2H, Ahppa CHOH + Isp NCH), 3.40 (dd, 1

H, J = 5.4, 15.1 Hz, Phe β-Ha), 3.00 (s, 3H, Phe N-CH₃), 2.82−2.96

(m, 3H, Phe β-Hb + Ahppa δ-H₂), 2.38 (m, 2H, Ahppa α-H₂), 2.16

(m, 2H, Gln β-Ha + Gln γ-H), 1.90 (m, 2H, Gln β-Hb + Val β-H),

1.59 (m, 2H, Leu β-H₂), 1.51 (m, 1H, Leu γ-H), 1.37 (d, 1H, J = 6.3

Hz, La CH₃), 1.12 (d, 3H, J = 6.3 Hz, NCHCH₃), 1.10 (d, 3H, J = 6.8

Hz, NCHCH₃), 0.89 (m, 12H, Val γ-H₃× 2 + Leu δ-H₃× 2), 0.79 (d,

10759

dx.doi.org/10.1021/jm301630s | J. Med. Chem. 2012, 55, 10749−10765

Journal of Medicinal Chemistry

Article

acid (10 mL × 2), 5% NaHCO₃(10 mL × 2) and brine (10 mL × 2),

dried (Na₂SO₄) and concentrated in vacuo, puriﬁed by chromatog-

raphy on silica gel (DCM: MeOH = 60:1 → 20:1) to give compound

9 (67.1 mg, 64%). [α]_D²⁰= −49.15 (c = 0.47, CHCl₃); ¹H NMR

(CDCl₃, 400 MHz) δ: 7.00−7.34 (m, 21 H, Ar-H × 15 + NH × 6),

5.61 (dd, 1 H, J = 6.7, 9.0 Hz, Phe α-H), 5.06 (m, 2H, CO₂CH₂Ph),

4.93 (m, 1 H, Gln α-NH), 4.69 (m, 2H, Val α-H + Ala α-H), 4.38 (m,

3H, Pro α-H + Leu α-H + Ahppa NCH), 4.04 (m, 1H, Ahppa

CHOH), 3.73 (s, 3 H, Pro OCH₃), 3.38 (m, 1H, Pro δ-Ha), 3.21 (m,

2H, Pro δ-Hb + Phe β-Ha), 3.00 (s, 3 H, Phe N-CH₃), 2.87−2.95 (m,

5H, Ahppa δ-H × 2 + Phe β-Hb + Gln N-Me), 2.44 (m, 1H, Ahppa α-

Ha), 2.34 (m, 2H, Gln β-H × 2), 1.83−2.19 (m, 8H, Ahppa α-Hb +

Pro β-H × 2 + Pro γ-H × 2 + Gln γ-H × 2 + Val β-H), 1.58 (m, 1H,

Leu γ-H), 1.47 (m, 1H, Leu β-H × 2), 0.81−0.96 (m, 12H, Val γ-H ×

6 + Leu δ-H × 6 + Ala β-CH × 3); ¹³C NMR (CDCl₃, 100 MHz) δ:

173.4, 172.7, 172.4, 172.2, 171.7, 168.1, 156.5, 138.2, 136.6, 136.1,

129.5, 129.0, 128.8, 128.6, 128.5, 128.2, 127.9, 126.7, 126.4, 69.3, 67.0,

59.2, 56.1, 55.5, 52.3, 51.9, 46.9, 45.1, 40.9, 40.7, 36.9, 34.7, 31.6, 30.7,

30.4, 28.8, 25.3, 24.7, 23.9, 22.9, 21.8, 19.6, 17.3, 17.2; ESI-MS (m/z):

1041.6 [M + H]⁺, 1063.5 [M + Na]⁺(Calcd 1040.6).

Compound 45. To a cold solution (ice bath) of compound 22

(95.3 mg, 0.25 mmol) in THF/MeOH (4:1, 5.0 mL) was added a

solution of NaOH (0.8 M in H₂O, 5.0 mL). The mixture was stirred at

r.t. for 3 h before adjusting to pH = 2 and extracted with EtOAc (30

mL × 3). The combined organic layer was washed with brine (15 mL

× 2), dried (Na₂SO₄), and concentrated in vacuo to give the free

carboxylic acid as a white solid, which was used without further

puriﬁcation.

A solution of compound 24⁵(104.0 mg, 0.23 mmol) in EtOAc (10

mL) was hydrogenated for 5 h at 1 atm in the presence of a catalytic

amount of Pd/C. The solution was ﬁltrated and concentrated to give

the free amine, which was used without further puriﬁcation.

To the cold solution (ice bath) of the acid and amine obtained

above in DCM (15 mL) were added EDC (95.9 mg, 0.50 mmol),

HOAt (68.1 mg, 0.50 mmol), DIEA (119.2 μL, 0.75 mmol). The

mixture was stirred at 0 °C for 2 h, and then at r.t. overnight. The

reaction mixture was diluted with EtOAc (100 mL) and washed with

10% citric acid (10 mL × 2), 5% NaHCO₃(10 mL × 2) and brine (10

mL × 2), dried (Na₂SO₄) and concentrated in vacuo. The residue was

puriﬁed by chromatography on silica gel (petroleum ether: EtOAc =

Compound 10. A solution of compound 9 (54.2 mg, 0.052 mmol)

in MeOH (10 mL) was hydrogenated for 24 h at 1 atm in the presence

of a catalytic amount of Pd/C. The mixture was ﬁltrated, concentrated

and puriﬁed by TLC to give compound 10 (39.4 mg, 83%) as a white

5:1) to give compound 45 (124.6 mg, 81%) as a white solid. [α]_D²⁰

=

1

−51.62 (c = 1.00, CHCl₃); H NMR (CDCl₃, 400 MHz) δ: 8.91 (s,

1H, SO₂NH), 8.01, 7.97, and 7.95 (s, 3H, Ar-H × 3), 7.12−7.40 (m,

10H, Ar-H × 10), 7.06 (d, 1H, J = 9.0 Hz, BnNH), 7.00 (d, 1H, J = 7.8

Hz, Ahppa NH), 5.35 (m, 1H, PhCH), 4.40 (q, J = 8.4 Hz, Ahppa

NCH), 4.13 (d, 1H, J = 7.3 Hz, CHOH), 3.96 (s, 1H, CHOH), 3.01

(m, 2H, Ahppa δ-H × 2), 2.90 (s, 3H, SO₂Me), 2.44 (m, 2H, Ahppa α-

solid. [α]²_D⁰= −29.85 (c = 0.068, CHCl₃); H NMR (CD₃OD, 400

1

MHz) δ: 7.16−7.31 (m, 10 H, Ar-H × 10), 5.66 (dd, 1 H, J = 5.9, 9.4

Hz, Phe α-H), 5.04 (m, 1H, Gln α-H), 4.75 (m, 1H, Pro α-H), 4.67

(m, 1H, Ala α-H), 4.42 (t, 1H, J = 7.2 Hz, Val α-H), 4.34 (dd, 1 H, J =

6.3, 9.0 Hz, Leu α-H), 4.22 (m, 1H, Ahppa NCH), 4.09 (m, 1H,

Ahppa CHOH), 3.72 (s, 3 H, Pro OCH₃), 3.66 (m, 1H, Pro δ-Ha),

3.44−3.51 (m, 3H, Pro δ-Hb + Gln γ-H₂), 3.13 (dd, 1 H, J = 6.3, 14.1

Hz, Phe β-Ha), 3.06 (s, 3 H, Phe N-CH₃), 2.93 (m, 2H, Ahppa δ-Ha +

Phe β-Hb), 2.78 (m, 1H, Ahppa δ-Hb), 2.13−2.42 (m, 5H, Ahppa α-H

× 2 + Pro β-H × 2 + Gln β-Ha), 1.95 (m, 1H, Gln β-Hb), 1.85 (m,

3H, Pro γ-H × 2 + Val β-H), 1.68 (m, 1H, Leu γ-H), 1.52 (m, 2H, Leu

β-H × 2), 0.88−1.01 (m, 15H, Val γ-H × 6 + Leu δ-H × 6 + Ala β-

CH₃); ¹³C NMR (CD₃OD, 100 MHz) two rotamers δ: 188.5, 177.7,

177.6, 175.7, 175.5, 174.3, 174.2, 173.9, 173.8, 173.6, 171.8, 171.6,

170.7, 170.3, 139.9, 139.7, 138.3, 138.2, 130.8, 130.4, 130.2, 129.6,

129.4, 129.2, 126.7, 127.4, 127.2, 73.2, 72.1, 70.9, 69.9, 62.2, 60.9, 60.3,

59.5, 58.3, 57.6, 57.3, 57.0, 56.1, 55.3, 53.2, 53.0, 52.6, 46.5, 41.9, 41.6,

38.4, 35.5, 33.1, 32.9, 32.7, 32.1, 31.9, 31.3, 31.1, 30.8, 30.0, 28.5, 26.3,

25.8, 25.2, 23.6, 21.9, 20.6, 20.3, 20.2, 18.2, 16.9, 16.8; ESI-MS (m/z):

907.5 [M + H]⁺, 929.5 [M + Na]⁺(Calcd 906.5).

t

H × 2), 1.60 (d, 3H, J = 6.7 Hz, PhCHCH₃), 1.40 (s, 9H, Bu); ¹³C

NMR (CDCl₃, 100 MHz) δ: 172.7, 166.1, 165.2, 142.8, 138.8, 137.8,

135.9, 135.6, 129.4, 128.7, 128.5, 127.5, 126.5, 126.3, 122.4, 122.1,

121.4, 81.8, 67.7, 54.8, 49.7, 39.6, 39.5, 38.1, 34.4, 28.0, 21.9; ESI-MS

(m/z): 610.3 [M + H]⁺, 632.2 [M + Na]⁺(Calcd 609.3).

Compound 12. To a stirred solution of 45 (54.2 mg, 0.089 mmol)

in dichloromethane (2 mL) was added TFA (2 mL). The reaction

mixture was stirred at room temperature for 1 h and then concentrated

under reduced pressure to give the corresponding carboxylic acid (47),

which was used in the next coupling without further puriﬁcation.

To a stirred solution of 33 (61.4 mg, 0.11 mmol) in dichloro-

methane (2 mL) was added TFA (2 mL). The reaction mixture was

stirred at room temperature for 1 h and then concentrated under

reduced pressure to give the corresponding amine, which was used

without further puriﬁcation.

To the cold solution (ice bath) of the carboxylic acid (47) and

amine obtained above in DCM (10 mL) were added HATU (67.7 mg,

0.18 mmol), HOAt (48.5 mg, 0.18 mmol), and DIEA (56.6 μL, 0.36

mmol). The mixture was stirred at 0 °C for 2 h, and then at r.t.

overnight. The reaction mixture was diluted with EtOAc (100 mL)

and washed with citric acid (10%, 10 mL × 2), 5% NaHCO₃(10 mL ×

2) and brine (10 mL × 2), dried (Na₂SO₄) and concentrated in vacuo.

After puriﬁcation by column chromatography (DCM/MeOH =

60:1→20:1), compound 12 (61.9 mg, 69%) was obtained as a white

solid. [α]²_D¹= −61.70 (c = 0.91, CHCl₃); ¹H NMR (CDCl₃, 400 MHz)

δ: 9.04 (s, 1H, SO₂NH), 8.02, 7.91, and 7.83 (s, 3H, Ar-H × 3), 7.56

(s-like, 1H, Ahppa NH), 7.05−7.33 (m, 18H, NH x 3 + Ar-H × 15),

5.63 (dd, 1H, J = 7.0, 8.6 Hz, Phe α-H), 5.30 (m, 1H, PhCH), 4.80 (s-

like, 1H, Ala α-H), 4.41 (m, 2H, Leu α-H + Pro α-H), 4.24 (s-like, 1H,

Ahppa NCH), 4.13 (s-like, 1H, CHOH), 3.65 (s, 3 H, Pro OCH₃),

3.26 (s-like, 1H, Pro δ-Ha), 3.26 (m, 1H, Pro δ-Hb), 3.15 (dd, 1H, J =

6.1, 14.3 Hz, Phe β-Ha), 3.00 (s, 3H, Phe N-CH₃), 2.95 (m, 2H,

Ahppa δ-H × 2), 2.92 (dd, 1H, J = 9.7, 14.2 Hz, Phe β-Hb), 2.84 (s,

3H, SOOMe), 2.14−2.38 (m, 5H, Leu β-Ha + Pro γ-H × 2 + Ahppa α-

H × 2), 1.86 (m, 1H, Leu β-Hb), 1.79 (m, 2H, Pro β-H × 2), 1.49 (d,

4H, PhCHCH₃+ Leu γ-H), 0.81 (d, 3H, J = 4.7 Hz, Ala β-H x 3), 0.77

(m, 6H, Leu δ-H × 6); ¹³C NMR (CDCl₃, 100 MHz) δ: 172.6, 172.2,

171.8, 169.9, 168.8, 166.5, 165.4, 143.3, 138.5, 138.0, 136.6, 136.0,

135.3, 129.4, 129.3, 128.6, 128.5, 128.2, 127.2, 126.6, 126.5, 126.1,

122.9, 122.1, 121.4, 69.6, 60.9, 59.2, 55.6, 52.2, 51.8, 49.7, 46.9, 45.3,

41.3, 39.4, 37.7, 34.7, 33.8, 31.9, 31.6, 30.4, 29.7, 29.6, 29.5, 29.3, 28.8,

26.2, 25.3, 24.6, 22.8, 22.7, 21.9, 17.2, 14.1; ESI-MS (m/z): 1010.4 [M

+ H]⁺, 1032.4 [M + Na]⁺(Calcd 1009.5).

Compound 11. To the cold solution (ice bath) of compound 10

(23.0 mg, 0.025 mmol) in DCM (5 mL) was added Boc₂O (6.7 mg,

0.031 mmol). The mixture was stirred at rt for 2 h, the solvent

evaporated and product puriﬁed by chromatography (DCM/MeOH =

40:1 → 20:1) to give compound 11 (15.2 mg, 60%) as a white solid.

1

[α]²_D⁰= −52.43 (c = 0.58, CHCl₃); H NMR (CDCl₃, 400 MHz) δ:

7.43 (d, 1H, J = 7.8 Hz, Leu NH), 7.34 (d, 1H, J = 7.8 Hz, Ahppa

NH), 7.16−7.26 (m, 10 H, Ar-H × 10), 7.08 (d, 1H, J = 7.0 Hz, Ala

NH), 6.13 and 6.24 (s, 2H, Gln δ-CONH × 2), 5.62 (dd, 1 H, J = 6.7,

9.4 Hz, Phe α-H), 5.13 (d, 1H, J = 8.2 Hz, Va NH), 4.95 (m, 1H, Gln

α-H), 4.69 (m, 1H, Ala α-H), 4.43 (dd, 1 H, J = 6.5, 8.2 Hz, Pro α-H),

4.32 (m, 2H, Leu α-H + Val α-H), 4.04 (s-like, 2H, Ahppa CHOH +

Ahppa NCH), 3.72 (s, 3 H, Pro OCH₃), 3.41 (m, 1H, Pro δ-Ha), 3.22

(m, 2H, Phe β-Ha + Pro δ-Hb), 3.01 (s, 3 H, Phe N-CH₃), 2.95 (m,

3H, Phe β-Hb + Gln γ-H × 2 + Pro β-Ha), 2.46 (m, 1H, Ahppa α-Ha),

2.18−2.34 (m, 4H, Ahppa α-Hb + Gln γ-H × 2, Pro γ-Ha), 2.08 (m,

3H, Gln β-Ha + Pro β-Hb + Leu β-Ha), 1.94 (m, 2H, Gln β-Hb + Pro

γ-Ha), 1.84 (m, 3H, Val β-H + Pro γ-Hb + Ahppa δ-H × 2), 1.60 (m,

t

1H, Leu γ-H), 1.42 (s, 9H, Bu), 0.85−0.95 (m, 12H, Val γ-H x 6 +

Leu δ-H × 6), 0.82 (d, 3H, J = 6.7 Hz, Ala β-CH₃); ¹³C NMR (CDCl₃,

100 MHz) δ: 175.0, 173.7, 172.6, 172.4, 172.2, 171.7, 170.8, 168.2,

156.0, 138.3, 136.6, 129.5, 129.1, 128.6, 128.3, 128.2, 126.6, 126.4,

79.9, 69.4, 59.2, 55.9, 55.5, 54.4, 52.3, 51.9, 46.9, 45.1, 40.8, 37.1, 34.7,

31.7, 30.7, 30.6, 30.4, 28.8, 28.4, 28.3, 25.3, 24.7, 24.2, 23.0, 21.9, 19.6,

18.5, 17.3, 17.2; ESI-MS (m/z): 1007.5 [M + H]⁺, 1029.6 [M + Na]⁺

(Calcd 1006.6).

10760

dx.doi.org/10.1021/jm301630s | J. Med. Chem. 2012, 55, 10749−10765

Journal of Medicinal Chemistry

Article

Compound 13. Obtained in the same way as described for 12. [α]_D²⁰

= −71.20 (c = 0.59, CHCl₃); H NMR (CDCl₃, 400 MHz) δ 8.17,

34.6, 31.6, 30.3, 29.7, 28.8, 25.3, 24.7, 23.3, 22.8, 21.9, 17.5; ESI-MS

(m/z): 708.4 [M + H]⁺, 730.3 [M + Na]⁺(Calcd 707.4).

1

7.95, and 7.88 (s, 3H, Ar-H × 3), 7.57 (s-like, 1H, Ahppa-NH), 7.13−

7.31 (m, 16H, PheCHMeNH + ArH × 15), 6.87 (d, 1H, J = 6.7 Hz,

Ala-NH), 6.71 (d, 1H, J = 7.0 Hz, Leu-NH), 5.59 (dd, 1H, J = 6.3, 9.4

Hz, Phe-αH), 5.26 (m, 1H, Ph−CH), 4.83 (m, 1H, Ala-αH), 4.44 (t,

1H, J = 7.4 Hz, Pro-αH), 4.00 (m, 1H, Leu-αH), 4.17 (q, 1H, J = 8.1

Hz, Ahppa-NCH), 4.06 (s-like, 1H, CHOH), 3.67 (s, 3 H, Pro-

OCH₃), 3.41 (m, 1H, Pro-δHa), 3.30 (s, 3H, SO₂NMe), 3.24 (m, 1H,

Pro-δHb), 3.13 (dd, 1H, J = 6.1, 14.5 Hz, Phe-βHa), 2.89−3.08 (m,

3H, Ahppa-δH₂+ Phe β-Hb), 3.01 (s, 3H, Phe-NCH₃), 2.81 (s, 3H,

SMe), 2.19 (m, 2H, Pro-βHa + Ahppa-αHa), 1.74−2.00 (m, 4H,

Ahppa-αHb + Pro-βHb + Leu-βH₂), 1.48 (d, 3H, J = 6.7 Hz, Ph−

CHCH₃), 1.48 (d, 3H, J = 6.7 Hz, Phe-NCH₃), 1.47 (m, 1H, Leu-γH),

0.89 (d, 3H, J = 7.1 Hz, Ala-βCH₃), 0.82 (d, 6H, J = 5.9 Hz, Leu-

δCH₃); ¹³C NMR (CDCl₃, 100 MHz) δ 191.0, 172.6, 172.5, 171.7,

168.2, 165.9, 164.8, 143.2, 141.9, 138.0, 136.5, 136.2, 135.0, 129.4,

129.2, 128.6, 128.2, 128.0, 127.2, 126.7, 126.6, 126.1, 59.2, 55.6, 52.2,

51.6, 49.7, 46.9, 45.3, 41.5, 37.9, 35.3, 34.7, 31.6, 30.4, 29.7, 29.6, 28.8,

25.3, 24.6, 22.9, 21.9, 21.7, 17.4; ESI-MS (m/z): 1024.5 [M + H]⁺,

1046.4 [M + Na]⁺(Calcd 1023.5).

Compound 15. Obtained similarly as described for the preparation

of 14. [α]²_D⁰= −19.66 (c = 1.00, CHCl₃); H NMR (CDCl₃, 400

1

MHz) δ: 8.00 (d, 1H, J = 6.7 Hz, Ala-NH), 7.76 (s-like, 1H, Leu-NH),

7.40 (d, 1H, J = 7.4 Hz, Ahppa-NH), 7.11−7.26 (m, 10H, Ar-H₅× 2),

6.17 and 6.14 (s, 2H, Gln-CONH₂), 5.35 (d, 1 H, J = 7.8 Hz, Phe-αH),

5.10 (d, 1H, J = 5.9 Hz, Gln-NH), 4.70 (m, 1H, Gln-αH), 4.45 (m,

1H, Pro-αH), 4.29 (m, 2H, Leu-αH + Ahppa-NCH), 4.15 (m, 1H,

Ahppa-CHOH), 3.73 (s, 3H, Pro-COOCH₃), 3.16 (m, 3H, Pro-δHa +

Ahppa-δHaHb + Phe-βHa), 3.00 (s, 3H, Phe-NCH₃), 2.94 (m, 2H,

Phe-βHb + Ahppa-δHaHb), 2.56 (dd, 1H, J = 7.2, 14.5 Hz, Ahppa-

αHaHb), 2.38 (m, 1H, Pro-δHb), 2.23 (m, 3H, Ahppa-αHaHb + Gln-

γH₂), 1.98 (s, 3H, COCH₃), 1.85 (m, 3H, Gln-βH₂+ Leu-βHb), 1.61

(m, 1H, Leu-γH), 0.89 (d, 3H, J = 6.3 Hz, Leu-δCH₃), 0.86 (d, 3H, J =

6.3 Hz, Leu-δCH₃), 0.73 (d, 3H, J = 6.7 Hz, Ala-βCH₃); ¹³C NMR

(CDCl₃, 100 MHz) δ: 175.5, 173.0, 172.9, 172.6, 172.5, 171.3, 171,2,

168.1, 139.0, 136.7, 129.5, 129.2, 128.3, 128.1, 126.6, 126.4, 69.3, 59.4,

56.1, 53.5, 52.2, 46.9, 44.5, 41.1, 40.8, 36.8, 34.5, 31.2, 30.5, 29.6, 28.8,

26.7, 25.4, 24.7, 23.0, 22.8, 21.8, 16.9; ESI-MS (m/z): 836.5 [M + H]⁺,

858.4 [M + Na]⁺(Calcd 835.5).

Compound 49. A solution of compound 24 (67.7 mg, 0.15 mmol)

in EtOAc (10 mL) was hydrogenated for 5 h at 1 atm in the presence

of a catalytic amount of Pd/C. The solution was ﬁltrated and

concentrated to give the free amine, which was dissolved in MeOH

(10 mL), and then treated with Ac₂O (266 μL, 15.2 mmol). The

mixture was stirred at 0 °C for 1 h and then at rt for 2 h. After that, the

solvent was removed and the residue was subjected to column

chromatography (petroleum ether/EtOAc = 2:1) to give compound

49 (38.6 mg, 83%) as a colorless oil. [α]²_D⁰= −62.38 (c = 0.046,

Compound 16. Obtained similarly as described for the preparation

of 14. [α]²_D⁰= −4.80 (c = 1.00, CHCl₃); ¹H NMR (CDCl₃, 400 MHz)

δ: 7.13−7.26 (m, 10H, Ar-H₅× 2), 6.98 (d, 1H, J = 7.4 Hz, Ala-NH),

6.57 (d, 1H, J = 8.2 Hz, Leu-NH), 6.25 (d, 1H, J = 9.0 Hz, Ahppa-

NH), 5.33 (dd, 1 H, J = 4.7, 11.7 Hz, Phe-αH), 4.69 (m, 1H, Ala-αH),

4.38 (m, 1H, Leu-αH), 4.02 (m, 2H, AhppaCHOH + NCH), 3.74 (s,

3H, Pro-COOCH₃), 3.40 (dd, 1H, J = 5.1, 14.9 Hz, Phe-βHa), 3.00

(dd, 1H, J = 12.1, 14.9 Hz, Phe-βHb), 2.92 (d, 2H, J = 7.4 Hz, Ahppa-

δH₂), 2.85 (s, 3H, Phe-NMe), 2.40 (dd, 1H, J = 9.4, 14.9 Hz, Ahppa-

αHaHb), 2.26 (dd, 1H, J = 3.5, 15.3 Hz, Ahppa-αHaHb), 1.95 (s, 3H,

COMe), 1.45−1.59 (m, 3H, Leu-βH₂+ Leu-γH), 0.89 (d, 6H, J = 6.3

Hz, Leu-δCH₃), 0.80 (d, 3H, J = 6.7 Hz, Ala-βCH₃); ¹³C NMR

(CDCl₃, 100 MHz) δ: 173.0, 172.3, 170.9, 170.7, 170.6, 138.1, 136.3,

129.3, 128.6, 128.4, 126.9, 126.4, 68.1, 58.0, 54.4, 52.5, 51.8, 50.7, 45.5,

41.4, 39.9, 37.8, 34.7, 32.3, 31.6, 29.7, 24.7, 23.2, 22.9, 21.9, 19.9; ESI-

MS (m/z): 611.3 [M + H]⁺, 633.3 [M + Na]⁺(Calcd 610.3).

Compound 51. Compound 12 (500 μg) was dissolved in 1 mL of

the mixture of THF/MeOH/H₂O (4:1:2, V:V:V) and treated with

LiOH monohydrate (200 μg) at 0 °C. After TLC showed that all the

starting material was consumed, the reaction mixture was diluted,

adjusted to pH 5, and extracted with EtOAc. The combined EtOAc

layer was dried over Na₂SO₄, ﬁltered, and concentrated in vacuo. The

residue was puriﬁed by RP-HPLC (Phenomenex Synergi 4μ, 250 × 10

mm, 1.0 mL/min, gradient elution with 80% to 100% MeOH−H₂O

over 20 min), giving 253 μg (estimated by UV) of the corresponding

carboxylic acid 51 (t_R7.75 min). HRESIMS calcd for C₅₂H₆₄N₇O₁₁S,

[M − H]⁻994.4390, found 994.4411.

1

CHCl₃); H NMR (CDCl₃, 400 MHz) δ: 7.22−7.32 (m, 5H, Ar-H₅),

5.92 (d, 1H, J = 9.0 Hz, NH), 4.05 (q, 1H, J = 8.2 Hz, NCH), 3.97 (d-

like, 1H, J = 9.4 Hz, Ahppa CHOH), 2.91 (d, 2H, J = 7.8 Hz, Ahppa δ-

H₂), 2.43 (dd, 1H, J = 10.6, 17.2 Hz, Ahppa α-Ha), 2.28 (dd, 1H, J =

t

2.3, 17.2 Hz, Ahppa α-Hb), 2.00 (s, 3H, COCH₃), 1.42 (s, 9H, Bu);

¹³C NMR (CDCl₃, 100 MHz) δ: 173.0, 169.9, 138.0, 129.4, 128.5,

126.5, 81.6, 66.8, 53.8, 39.4, 38.2, 28.0, 23.4; ESI-MS (m/z): 308.2 [M

+ H]⁺, 330.2 [M + Na]⁺(Calcd 307.2).

Compound 14. To a stirred solution of 49 (16.4 mg, 0.053 mmol)

in dichloromethane (1 mL) was added TFA (1 mL). The reaction

mixture was stirred at room temperature for 1 h and then concentrated

under reduced pressure to give the corresponding carboxylic acid 50.

To a stirred solution of 33 (31.0 mg, 0.054 mmol) in

dichloromethane (2 mL) was added TFA (2 mL). The reaction

mixture was stirred at room temperature for 1 h and then concentrated

under reduced pressure to give the corresponding amine.

To the cold solution (ice bath) of the amine and carboxylic acid 50

obtained above in THF (8 mL) was added HATU (41.0 mg, 0.11

mmol), HOAt (14.7 mg, 0.11 mmol), and DIEA (33.7 μL, 0.21

mmol). The mixture was stirred at 0 °C for 2 h, and then at r.t.

overnight. After concentration under reduced pressure, the residue was

dissolved in EtOAc (50 mL) and washed with citric acid (10%, 10 mL

× 2), 5% NaHCO₃(10 mL × 2) and brine (10 mL × 2), dried

(Na₂SO₄) and concentrated in vacuo, puriﬁed by chromatography on

silica gel (DCM/MeOH = 50:1 → 20:1) to give the compound 14

(32.7 mg, 86%) as a colorless oil. [α]²_D⁰= −15.1 (c = 0.19, CHCl₃); ¹H

NMR (CDCl₃, 400 MHz) δ: 7.14−7.28 (m, 10H, Ar-H₅× 2), 6.81 (d,

1H, J = 7.8 Hz, Ala-NH), 6.53 (d, 1H, J = 7.8 Hz, Leu-NH), 6.29 (d,

1H, J = 9.0 Hz, Ahppa-NH), 5.55 (dd, 1 H, J = 6.3, 9.4 Hz, Phe-αH),

4.71 (m, 1H, Ala-αH), 4.45 (dd, 1H, J = 6.3, 7.8 Hz, Pro-αH), 4.32 (m,

2H, Leu-αH), 4.01 (m, 2H, AhppaCHOH + NCH), 4.01 (m, 2H, Leu-

αH + Ahppa-NCH), 3.73 (s, 3H, Pro-COOCH₃), 3.36 (m, 1H, Phe-

βHa), 3.19 (dd, 2H, J = 6.3, 14.9 Hz, Pro-δH₂), 2.94 (m, 3H, Phe-βHb

+ Ahppa-δH₂), 2.41 (dd, 1H, J = 9.4, 15.3 Hz, Ahppa-αHaHb), 2.26

(dd, 1H, J = 3.5, 15.3 Hz, Ahppa-αHaHb), 2.22 (m, 1H, Pro-δHa),

1.81−1.94 (m, 3H, Pro-γH₂+ Pro-βHa), 1.52−1.58 (m, 3H, Leu-βH₂

+ Leu-γH), 0.87 (d, 6H, J = 6.3 Hz, Leu-δCH₃), 0.80 (d, 3H, J = 7.0

Hz, Ala-βCH₃); ¹³C NMR (CDCl₃, 100 MHz) δ: 172.4, 172.3, 171.2,

170.7, 168.1, 138.2, 136.6, 129.5, 129.2, 128.9, 128.4, 128.2, 126.6,

126.4, 68.1, 59.3, 55.6, 54.6, 52.2, 51.8, 46.9, 45.0, 41.2, 39.8, 37.8,

Compound 52. Following the same procedure as described for

compound 51 but using compound 13 (640 μg) as starting material

gave 107 μg (estimated by UV) of carboxylic acid 52 (t_R6.75 min).

HRESIMS calcd for C₅₃H₆₆N₇O₁₁S [M − H]⁻1008.4541, found

1008.4547.

Protein Crystallography. Protein Puriﬁcation and Crystalliza-

tion. A detailed description of the production of recombinant human

BACE1 has been provided in our previous publication.³²Brieﬂy,

BACE1 was expressed in E. coli as inclusion bodies that were then

denatured and refolded into the active monomer. A cDNA fragment

encoding BACE1 residues 43−454 was cloned into pET28a with a

TEV protease cleavage site following a six-residue His-Tag added at

the N-terminus. After refolding, nickel beads (Ni Sepharose High

Performance, Amersham Biosciences, Uppsala, Sweden) were used to

concentrate the protein. The eluted BACE1 was then injected into a

124 mL Hiload Superdex75 column from which it was eluted with 1

mM DTT/0.5 M urea/150 mM NaCl/20 mM Tris-HCl, pH 7.5. In

order to obtain crystals with diﬀerent protein packing patterns, two

mutations, K75A and E77A, were introduced into BACE1 together.

The active BACE1 monomer in the elution buﬀer was concentrated

to 8−10 mg/mL. Cocrystallizations of compounds 12 and 13 with

wild type BACE1 were carried out by mixing the solution of the

10761

dx.doi.org/10.1021/jm301630s | J. Med. Chem. 2012, 55, 10749−10765

Journal of Medicinal Chemistry

Article

protein−ligand complex with an equal volume of a precipitant

solution, 1.0 M ammonium sulfate/100 mM sodium citrate, pH 5.0.

The protein−ligand complex was prepared by adding the compound

to the protein solution to reach a ﬁnal concentration of 1 mM ligand.

Crystallization of apo mutated BACE1 was performed by mixing equal

volumes of the protein stock solution and the precipitant (1.7 M

Li₂SO₄/100 mM HEPES, pH 7.5). To obtain a crystalline complex

with compounds 2, 5, and 11, a 100 mM solution of the compound in

DMSO was diluted 20-fold, to 5 mM, in the precipitant solution so as

to generate a soaking drop. The crystal was transferred into the

soaking drop and left for 24 h prior to data collection. All the

crystallization utilized the vapor diﬀusion method in hanging drops.

The perﬂuoropolyether, PFO-X175/08 (Hampton Research), was

used as cryoprotectant for all the crystals.

Structure Determination and Reﬁnement. Data were collected at

100 K on beamline BL17U (at wavelength 0.9793 Å) at the Shanghai

Synchrotron Radiation Facility (SSRF) (Shanghai, China) for the

soaked and the cocrystallized structures. The data were processed with

the HKL2000³³and XDS³⁴software packages, and the structures were

then solved by molecular replacement, using the CCP4 program

MOLREP.³⁵The search model used for the crystals was the apo wild

type BACE1 structure (pdb code 3tpl). The structures were reﬁned

using the CCP4 program REFMAC5³⁵combined with the simulated-

annealing protocol implemented in the program PHENIX.³⁶With the

aid of the program Coot,³⁷ligand, water molecules, and others were

ﬁtted into the initial F_o−F_cmaps. The complete statistics, as well as the

quality of the solved structures, are shown in Table S1.

Pharmacology and Biology. Aspartic Protease Enzymatic

Assays. In vitro assays to determine inhibitory activity against

BACE1, cathepsin D and cathepsin E were carried out as described.³

Enzymes BACE1 (R&D Systems, Minneapolis, MN), cathepsin D

(Enzo Life Sciences, Farmingdale, NY) and cathepsin E (R&D

systems) and corresponding substrates were prepared freshly in their

respective reaction buﬀers. The reaction buﬀer for BACE1 contained

100 mM NaOAc (pH 4.0) and the reaction buﬀer for cathepsins D

and E contained 100 mM NaOAc/100 mM NaCl (pH 3.5). 1%

DMSO was added into the buﬀer before use. Enzymes were added

(ﬁnal concentrations are 30 μg/mL for BACE1, 2 μg/mL for cathepsin

D, 0.05 μg/mL for cathepsin E) followed by compounds dissolved in

DMSO at 12 diﬀerent concentrations (3-fold dilutions starting from

10 μM ﬁnal assay concentration, up to ≤2% DMSO). Subsequently,

substrates (ﬁnal concentration of 10 μM) were added to initiate the

reaction (substrate for BACE1: Mca-Ser-Glu-Val-Asn-Leu-Asp-Ala-

Glu-Phe-Arg-Lys(Dnp)-Arg-Arg-NH₂(R&D Systems); substrate for

cathepsins D and E: Mca-Gly-Lys-Pro-Ile-Leu-Phe-Phe-Arg-Leu-Lys-

(Dnp)-^D-Arg-NH₂(Enzo Life Sciences); Mca = (7-methoxycoumarin-

4-yl)acetyl; Dnp = 2,4-dinitrophenyl). The enzyme activities were

detected as a time-course measurement of the increase in ﬂuorescence

signal (Ex 320 nm, Em 405 nm) from ﬂuorescently labeled peptide

substrate for 120 min at room temperature.

min intervals to determine pNPP hydrolysis rate to para-nitrophenol,

which was expressed as an increase in absorbance units per minute.

For ease of analysis, a single time point within the linear range of the

reaction was selected for the data analysis. Statistical analysis was by

one-way Anova followed by Dunnett’s multiple comparison test.

CTF Analysis. H4 cells were plated at 3 × 10⁵per well in six well

dishes and allowed to adhere for 4−6 h in DMEM supplemented with

10% fetal calf serum at 37 °C. The medium was replaced with

Ultraculture (Lonza), to which test compounds were added to a ﬁnal

concentration of 10 μM and the incubation was allowed to proceed for

an additional 16 h. The medium was removed, the cells were washed

twice with phosphate buﬀered saline (PBS) and after the removal of

the last wash, 50 μL of phospholysis buﬀer (50 mM Tris-HCl pH 7.4,

1 mM EDTA, 150 mM NaCl, 1% NP40, 0.25% sodium deoxycholate)

was added to the center of the plate to initiate lysis. Using a cell

scraper, the buﬀer was spread across the plate, and the resulting lysate

was transferred to a 1.5-mL microcentrifuge tube. Nuclei were

removed by pelleting in a microfuge for 30 s with the supernatant

transferred to a fresh tube. Protein concentrations were determined by

the bicinchoninic acid (BCA) method according to the manufacturer’s

instructions (Thermo Scientiﬁc). The samples were equated for

protein concentration and diluted into 4× tricine sodium dodecyl

sulfate sample buﬀer, denatured by heating at 95 °C for 5 min, and

electrophoresed on 10−20% tricine polyacrylamide gels (BioRad).

The proteins were transferred onto PVDF and blocked overnight. The

blots were incubated with A8717, a rabbit polyclonal antibody, to the

last 20 amino acids of APP at a 1:2000 dilution (Sigma). This was

followed by incubation with a goat-anti rabbit IR antibody at a 1:20000

dilution. Blots were developed using the LiCor Odyssey Imager.

Secreted β-Amyloid (Aβ) Assay. CHO 2B7 cells (clonal expression

of APP695 wt) were plated in 96-well plates in Opti-MEM with 1%

serum. At 90% conﬂuency cells were treated with positive controls (10

μM BACE1 inhibitor IV, 10 μM BACE1 inhibitor II, 1 μM LY411575

γ-secretase inhibitor) or test compounds (1% DMSO ﬁnal) and

incubated for 16 h. Supernatants were collected and analyzed by direct

sandwich Aβ ELISA. Compound-treated media were captured onto

ELISA plates with immobilized c-terminal speciﬁc Mabs to Aβ1−40

(13.1.1) or Aβ1−42 (2.1.3) and then detected with Aβ1−16 HRP-

conjugated Mab (AB5-HRP). Total Aβ was measured with AB5 and

detected with 4G8-HRP. ELISAs were developed with 3,3′,5,5′-

tetramethylbenzidine (TMB), stopped with o-phosphoric acid

solution, and absorbance was measured at 562 nm on a Molecular

Devices Flexstation.

Measurement of Cellular γ-Secretase Inhibition. H4 cells

overexpressing the Bri-C99 fusion protein were cultured in 12-well

plates (400000/well) in Opti-MEM media with 1% serum and treated

for 16 h with BACE1 inhibitor IV (1 μM), LY411575 γ-secretase

inhibitor (100 nM) or test compounds (10 μM). Cells were lysed in

TBS with 1% Triton X100 detergent and loaded onto a 12% bis-tris

gel. Proteins were blotted onto sequencing grade 0.2 μm PVDF,

probed with CT-20 antibody (Sigma 8717), scanned and analyzed on

LiCor Odyssey.

BBB PAMPA. Permeability was assayed according to the

manufacturer’s instruction (pION Inc., Woburn, MA). In brief, 10

μM of the test samples were prepared in an aqueous buﬀer containing

2.5% Prisma HT buﬀer (P/N 110151), 5% acetonitrile and 0.5%

DMSO, with pH of 5.0 and 7.4. Aliquots of the prepared test samples

were added to donor wells of the PAMPA sandwich (P/N 110163).

Immobilized ﬁlters of the acceptor plate were coated with 5 μL of

pION artiﬁcial membrane forming BBB-1 lipid solution (P/N 110672)

and the acceptor wells were ﬁlled with the Brain Sink Buﬀer (P/N

110674). The donor and acceptor plates were assembled to create the

sandwich and left to incubate for 4 h under a humidiﬁed environment

at room temperature. The absorbance of the donor, acceptor, and

reference wells were scanned at 250−350 nm using a SpectraMax M5

plate reader (Molecular Devices, Sunnyvale CA). The permeability

coeﬃcient P_ewas calculated using the PAMPA Explorer Command

Software version 3.5 from pION. Assays were conducted in triplicate

HPLAP-APP Reporter Assay.¹⁹In a 96-well tissue culture plate,

stably transfected H4 (human neuroglioma) cells were seeded at a

density of 2 × 10⁴cells/well in 100 μL of Dulbecco’s Modiﬁed Eagle’s

Medium (DMEM) which was supplemented with fetal calf serum to

ﬁnal concentration of 10% and allowed to attach for approximately 6 h

at 37 °C. Following the incubation period, the cells were washed with

100 μL of serum free media (Ultraculture, Lonza), and 100 μL of fresh

serum free media was added to the cells which contained either test

compounds, β-secretase inhibitor II or IV (Calbiochem) or the

appropriate vehicle controls. Transfected cells were simultaneously

treated with BB-94 (Calbiochem) and the individual compounds at 10

μM. Incubation was allowed to proceed for 16 h at 37 °C and samples

were assayed for HPLAP activity to quantitate sAPPβ. Speciﬁcally, 50

μL of culture medium was transferred to a new 96-well assay plate and

mixed with an equal volume of HPLAP assay solution containing 1

mM MgCl₂, 10 mM L-homoarginine, and 1 M diethanolamine, pH 9.8.

It was empirically determined that heating the samples to 65 °C for 30

min to inhibit other phosphatase activities was not necessary. The

reaction was started by the addition of 5 μL of 200 mM para-

nitrophenyl phosphate (pNPP). Absorbance was read at 410 nm at 1

and results are expressed as mean

SD. Theophylline and

10762

dx.doi.org/10.1021/jm301630s | J. Med. Chem. 2012, 55, 10749−10765

Journal of Medicinal Chemistry

Article

corticosterone were used as controls for low and high BBB

permeability, respectively.

extraction steps were pooled, the acetonitrile evaporated under

nitrogen and further partitioned with ethyl acetate. The organic

layer was collected, dried and reconstituted in 50 μL of MeOH. A 10

μL-aliquot was injected for HPLC-MS analysis.

Acute Treatment Study in CF-1 Mice. All compounds were

prepared in a vehicle of 5% DMSO, 15% Solutol HS 15 (BASF,

Ludwigshafen, Germany), 10% EtOH and 65% water. CF-1 mice

(Charles River Laboratories, Wilmington, MA) were used at 8 weeks

of age. Mice (25−35 g body weight) were administered either 10 mg/

kg of LY411575 or 30 mg/kg of 12 and 13 by intraperitoneal injection,

and euthanized by CO₂asphyxiation 3 h postdosing. After extraction

of the brain, olfactory bulbs and hindbrain were removed and the

cerebral hemispheres were cut into two pieces for Aβ analysis. The

cerebellum was removed for metabolite analysis. All tissue samples

were weighed, frozen in liquid nitrogen, and stored at −80 °C until

analysis. Aβ analysis was carried out essentially as previously described

with antibody 13.1.1 used for the analysis of rodent Aβ40.³⁸All

procedures were conducted according to NIH guidelines for animal

care and approved by the University of Florida Institutional Animal

Care and Use Committees.

Preparation of Plasma Samples. Whole blood was collected by

cardiac puncture using a 22 gauge needle attached to a 1 mL syringe,

as described.³⁹Brieﬂy, the sample was transferred to a 1.3 mL

microtube containing potassium EDTA and placed on ice after gentle

inversion 8−10 times to mix the anticoagulant additive with the blood.

Tubes were maintained upright on ice until centrifugation within 30

min of collection. The samples were centrifuged (4−10 °C) for 20 min

at 1300g. After centrifugation the plasma layer was removed and

aliquoted into fresh microfuge tubes, ﬂash frozen, and placed at −80

°C until needed.

HPLC-MS Monitoring of 12 and 13 and Corresponding Acid

Metabolites 51 and 52. Analysis of unmetabolized 12 and 13 and

their corresponding acid products in plasma and cell lysates was done

using HPLC-MS with multiple reaction monitoring (MRM) in

negative ionization. A 10 μL-aliquot of the reconstituted methanolic

solution was injected and chromatographic separation was done on an

Onyx C18 monolithic column (Phenomenex, 3.0 × 100 mm) using a

stepwise gradient of 0.1% aqueous formic acid (solvent A) and 0.1%

formic acid in MeOH (solvent B). Elution was started using 70%

solvent B and increased to 100% solvent B after 5 min and further

maintained in 100% solvent B for another 10 min. The retention times

and corresponding MRM pair (t_R, min; MRM) for the compounds are

as follows: 12 (3.2; 1008.4/492.1), 51 (2.8; 994.4/492.1); 13 (3.4;

1022.4/506.2), 52 (3.1; 1008.4/506.2), 8 (2.2; 949.5/447.3).

Compound- and solvent-dependent parameters were initially opti-

mized using syringe infusion. Compound-dependent parameters were

as follows. 12: declustering potential (DP) −115, entrance potential

(EP) −9.0, collision energy (CE) −59, collision exit potential (CXP)

−15, collision entrance potential (CEP) −43; 51: DP −95, EP −8.0,

CE −70, CXP −15, CEP −8; 4−67: DP −140, EP −9.0, CE −38,

CXP −12, CEP −46; 52: DP −125, EP −10, CE −72, CXP −4.0, CEP

−52; 8: DP −80, EP −7.0, CE −72, CXP − 7.0, CEP −43. Solvent-

dependent parameters were as follows: CUR 30, CAD Medium, IS

−4500, TEM 450, GS1 60, GS2 60.

For brain analysis (cerebellum and cortex), monitoring for 12 was

performed according to the procedure for plasma stability. Compound

13 was monitored in the positive ionization mode using the same

chromatographic condition and using 8 as internal standard. The

retention times and corresponding MRM pair (t_R, min; MRM) by

positive ionization for 8 and 13 were as follows: 13 (3.4; 1024.4/

895.1), 8 (2.2; 951.5/675.7). Compound-dependent parameters were

as follows: 13: DP 80, EP 9.0, CE 24, CXP 42, CEP 30; 8: DP 95 EP

10, CE 37, CXP 34, CEP 30. Solvent dependent-parameters were as

follows: CUR 30, CAD Medium, IS 5000, TEM 450, GS1 60, GS2 60.

Quantitation of 12 and 13 and Corresponding Metabolites 51

and 52. Calibration curves for 12, 51, 13 and 52 were generated by

least-squares linear regression analysis of standard solutions. Analyte

peak area and internal standard peak area ratio was plotted against the

analyte concentration and internal standard concentration ratio, and

the concentrations of 12, 13 and corresponding acids 51 and 52 at

each time point in the in vitro plasma and cellular experiments and in

vivo plasma experiment were determined through interpolation. All

calculations were done using Analyst 1.4.2 Quantitation Mode.

Analytical Chemistry. Materials. HPLC-MS was done on a 3200

QTRAP (Applied Biosystems) equipped with a Shimadzu UFLC

System. Mouse serum (S7273) was purchased from Sigma Aldrich.

Sample Preparation. Stock solutions (1 mg/mL) of 12, 13, 51, and

52 were prepared in absolute EtOH. Aliquots were diluted in MeOH

to give a 40 μg/mL solution and standard solutions with concentration

of 25, 12.5, 2.5, 1.25, 0.25, 0.125 μg/mL were prepared from the serial

dilution of the 40 μg/mL working solution. Stock solution (1 mg/mL)

of the internal standard compound 8 was prepared in absolute ethanol,

which was subsequently used to prepare a 100 μg/mL solution in

MeOH. An aliquot of the 100 μg/mL solution of compound 8 was

diluted with EtOAc to give the working internal standard solution

(100 ng/mL).

In Vitro Plasma Stability of 12 and 13. To evaluate the in vitro

plasma stability of 12 and 13 and to monitor the formation of the

corresponding acid metabolites 51 and 52, 10 μL of the 25 μg/mL

solution of 12 and 13 were separately incubated at 37 °C with 100 μL

of mouse serum for the following durations: 0, 1, 2, 4, 8, 12, and 24 h.

Reactions were quenched and metabolites were extracted at the end of

the incubation period by adding 400 μL of EtOAc and followed by 200

μL of 8 in EtOAc (100 ng/mL). The reaction mixtures were left to

shake on a plate mixer (750 rpm) for 5 min at room temperature and

then further centrifuged at 16000g for 10 min. The EtOAc layer was

collected, dried under nitrogen and reconstituted in 50 μL of MeOH.

In Vivo Plasma Levels of 12 and 13. 100 μL-aliquots of plasma

sample from test animals treated with 12 and 13 for 3 h (i.p. injection,

see above) were extracted with EtOAc and spiked with the internal

standard (8) using the same procedure as for the in vitro plasma

stability.

In Vitro Cellular Stability of 12 and 13. H4 cell lysates were diluted

with 0.1 M phosphate buﬀer (pH 7.4) to give a 1 mg/mL protein

concentration and volume of 100 μL. To diluted cell lysates were

added 10 μL of 12 and 13 (12.5 μg/mL) and incubated at 37 °C for 0,

1, 2, 4, 8, 12, and 24 h. Extraction of 12 and 13 as well as the

corresponding acid metabolites 51 and 52 from H4 cell lysates were

performed according to the same procedure for the in vitro plasma

stability.

In Vivo Brain Levels of 12 and 13. Cerebellum and cortex from test

animals were homogenized in acetonitrile (1:10 w:v) spiked with the

internal standard and incubated on ice for 10 min before centrifugation

at 16000g (15 min at 4 °C). The supernatant was collected and the

pellet was re-extracted with acetonitrile. Supernatants from both

ASSOCIATED CONTENT

■

S

* Supporting Information

Preparation of building blocks 22, 23, 26, 27, 29, 30, and 34−

1

36, copies of H NMR and ¹³C NMR of compounds 5−16,

Supplementary Figure S1 with LIGPLOT representations,

Figure S2 with structures of control secretase inhibitors, Tables

S1 and S2 with X-ray data collection/reﬁnement statistics and

hydrophobic interactions. This material is available free of

charge via the Internet at http://pubs.acs.org.

Accession Codes

PDB ID codes: 4DV9, 4FGX, 3UQP, 4DVF, and 3UQR.

AUTHOR INFORMATION

■

Corresponding Author

*(Y.X.) phone, +86-21-50801267; fax, +86-21-50807188; e-

mail, ycxu@mail.shcnc.ac.cn. (Y.L.) phone, +86-21-51980127;

fax, +86-21-51980127; e-mail, liyx417@fudan.edu.cn. (H.L.)

phone, +1-352-273-7738; fax, +1-352-273-7741; e-mail,

luesch@cop.u ﬂ .edu.

10763

dx.doi.org/10.1021/jm301630s | J. Med. Chem. 2012, 55, 10749−10765

Journal of Medicinal Chemistry

Article

(5) Sun, T.; Zhang, W.; Zong, C.; Wang, P.; Li, Y. Total synthesis

and stereochemical reassignment of tasiamide B. J. Pept. Sci. 2010, 16,

364−374.

Author Contributions

^⊥Contributed equally to this work.

Notes

(6) (a) Kortum, S. W.; Benson, T. E.; Bienkowski, M. J.; Emmons, T.

L.; Prince, D. B.; Paddock, D. J.; Tomasselli, A. G.; Moon, J. B.;

LaBorde, A.; TenBrink, R. E. Potent and selective isophtalamide S2

hydroxyethylamine inhibitors of BACE1. Bioorg. Med. Chem. Lett.

2007, 17, 3378−3383. (b) Cumming, J. N.; Le, T. X.; Babu, S.;

Carroll, C.; Chen, X.; Favreau, L.; Gaspari, P.; Guo, T.; Hobbs, D. W.;

Huang, Y.; Iserloh, U.; Kennedy, M. E.; Kuvelkar, R.; Li, G.; Lowrie, J.;

McHugh, N. A.; Ozgur, L.; Pan, J.; Parker, E. M.; Saionz, K.; Stamford,

A. W.; Strickland, C.; Tadesse, D.; Voigt, J.; Wang, L.; Wu, Y.; Zhang,

L.; Zhang, Q. Rational design of novel, potent piperazinone and

imidazolidinone BACE1 inhibitors. Bioorg. Med. Chem. Lett. 2008, 18,

3236−3241.

The authors declare no competing ﬁnancial interest.

ACKNOWLEDGMENTS

■

This study was supported by the University of Florida College

of Pharmacy (H.L.), by the “100 Talents Project” of CAS (to

Y.X.), the Shanghai Pujiang Program (Grant No.

10PJ1412000), and the National Natural Science Foundation

of China (Grant No. 91013010, 81172975, 21172233 and

81001392). W.Z. thanks the China Scholarship Council for a

Visiting Research Fellowship.

(7) Yang, S.; Zhang, W.; Ding, N.; Lo, J.; Liu, Y.; Clare-Salzler, M. J.;

Luesch, H.; Li, Y. Total synthesis of grassystatin A, a probe for

cathepsin E function. Bioorg. Med. Chem. 2012, 20, 4774−4780.

(8) Ghosh, A. K.; Brindisi, M.; Tang, J. Developing β-secretase

inhibitors for treatment of Alzheimer’s disease. J. Neurochem. 2012,

120 (Suppl. 1), 71−83.

(9) Zheng, H.; Koo, E. H. Biology and pathophysiology of the

amyloid precursor protein. Mol. Neurodegener. 2011, 6, 27−43.

(10) Hardy, J. A.; Higgins, G. A. Alzheimer’s disease: the amyloid

cascade hypothesis. Science 1992, 256, 184−185.

(11) Galimberti, D.; Scarpin, E. Alzheimer’s disease: from patho-

genesis to disease-modifying approaches. CNS Neurol. Disord. Drug

Targets 2011, 10, 163−174.

(12) (a) Ghosh, A. K.; Bilcer, G.; Hong, L.; Koelsch, G.; Tang, J.

Memapsin 2 (beta-secretase) inhibitor drug, between fantasy and

reality. Curr. Alzheimer Res. 2007, 4, 418−422. (b) Durham, T. B.;

Shepherd, T. A. Progress toward the discovery and development of

efficacious BACE inhibitors. Curr. Opin. Drug Discovery Dev. 2006, 9,

776−791.

ABBREVIATIONS USED

■

Ac₂O, acetic anhydride; Aβ, β-amyloid; BI−II, BACE1 inhibitor

II; BI−IV, BACE1 inhibitor IV; BBB, blood−brain barrier;

BCA, bicinchoninic acid; CAD, collision activated dissociation;

CE, collision energy; CTF, carboxyl terminal fragment; CEP,

collision entrance potential; CXP, collision exit potential; CUR,

curtain gas; DIEA, N,N-diisopropylethylamine; DMEM,

Dulbecco’s Modiﬁed Eagle’s medium; DP, declustering

potential; EDC, 1-ethyl-3-(3-dimethylaminopropyl)-

carbodiimide hydrochloride; EP, entrance potential; GS1, gas

1; GS2, gas 2; HATU, 2-(7-aza-1H-benzotriazole-1-yl)-1,1,3,3-

tetramethyluronium hexaﬂuorophosphate; HIs, hydrophobic

interactions; HOAt, 1-hydroxy-7-azabenzotriazole; HPLAP,

human placental alkaline phosphatase; HRP, horseradish

peroxidase; i.p., intraperitoneal; IS, ionspray voltage; MRM,

multiple reaction monitoring; PAMPA, parallel artiﬁcial

membrane permeability assay; pNPP, para-nitrophenyl phos-

phate; PVDF, polyvinylidene ﬂuoride; PyAOP, (7-azabenzo-

triazol-1-yloxy)tripyrrolidinophosphonium hexaﬂuorophos-

phate; sAPPβ, secreted APP ectodomain; sAPP, secreted

APP; SSRF, Shanghai Synchrotron Radiation Facility; TEM,

temperature; TMB, 3,3′,5,5′-tetramethylbenzidine

(13) (a) Willem, M.; Garratt, A. N.; Novak, B.; Citron, M.;

Kaufmann, S.; Rittger, A.; DeStrooper, B.; Saftig, P.; Birchmeier, C.;

Haass, C. Control of peripheral nerve myelination by the β-secretase

BACE1. Science 2006, 314, 664−666. (b) Dominguez, D.; Tournoy, J.;

Hartmann, D.; Huth, T.; Cryns, K.; Deforce, S.; Serneels, L.;

Camacho, I. E.; Marjaux, E.; Craessaerts, K.; Roebroek, A. J. M.;

Schwake, M.; D’Hooge, R.; Bach, P.; Kalinke, U.; Moechars, D.;

Alzheimer, C.; Reiss, K.; Saftig, P.; DeStrooper, B. Phenotypic and

biochemical analyses of BACE1- and BACE2-deficient mice. J. Biol.

Chem. 2005, 280, 30797−30806.

REFERENCES

■

(1) Tan, L. T. Filamentous tropical marine cyanobacteria: a rich

source of natural products for anticancer drug discovery. J. Appl.

Phycol. 2010, 22, 659−676.

(14) Shen, J.; Bronson, R. T.; Chen, D. F.; Xia, W.; Selkoe, D. J.;

Tonegawa, S. Skeletal and CNS defects in presenilin-1-deficient mice.

Cell 1997, 80, 629−639.

(2) (a) Chlipala, G. E.; Mo, S.; Orjala, J. Chemodiversity in

freshwater and terrestrial cyanobacteria - a source for drug discovery.

Curr. Drug Targets 2011, 12, 1654−1673. (b) Carmeli, S.; Grach-

Pogrebinsky, O.; Gesner-Apter, S.; Raveh, A.; Lifshitz, M.; Zafrir, E.

Biologically active cyanobacteria metabolites - Ecological and

biomedical perspectives. Planta Med. 2008, 74. (c) Taori, K.; Paul,

V. J.; Luesch, H.; Kempopeptins, A and B, serine protease inhibitors

with different selectivity profiles from a marine cyanobacterium,

Lyngbya sp. J. Nat. Products 2008, 71, 1625−1629. (d) Kwan, J. C.;

Taori, K. T.; Paul, V. J.; Luesch, H. Lyngbyastatins 8−10, elastase

inhibitors with cyclic depsipeptide scaffolds isolated from the marine

cyanobacterium Lyngbya semiplena. Mar. Drugs 2009, 7, 528−538.

(e) Matthew, S.; Ratnayake, R.; Becerro, M. A.; Ritson-Williams, R.;

Paul, V. J.; Luesch, H. Intramolecular modulation of serine protease

inhibitor activity in a marine cyanobacterium with antifeedant

properties. Mar. Drugs 2010, 8, 1803−1816.

(15) McConlogue, L.; Buttini, M.; Anderson, J. P.; Brigham, E. F.;

Chen, K. S.; Freedman, S. B.; Games, D.; Johnson-Wood, K.; Lee, M.;

Zeller, M.; Liu, W.; Motter, R.; Sinha, S. Partial reduction of BACE1

has dramatic effects on Alzheimer plaque and synaptic pathology in

APP transgenic mice. J. Biol. Chem. 2007, 282, 26326−26334.

(16) Ghosh, A. K.; Kumaragurubaran, N.; Hong, L.; Kulkarni, S. S.;

Xu, X.; Chang, W.; Weerasena, V.; Turner, R.; Koelsch, G.; Bilcer, G.;

Tang, J. Design, synthesis, and X-ray structure of potent memapsin 2

(β-secretase) inhibitors with isophthalamide derivatives as the P₂−P₃-

ligands. J. Med. Chem. 2007, 50, 2399−2407.

(17) Hong, L.; Koelsch, G.; Lin, X.; Wu, S.; Terzyan, S.; Ghosh, A.

K.; Zhang, X. C.; Tang, J. Structure of the protease domain of

memapsin 2 (β-secretase) complexed with inhibitor. Science 2000, 290,

150−153.

(18) Wallace, A. C.; Laskowski, R. A.; Thornton, J. M. LIGPLOT: a

program to generate schematic diagrams of protein-ligand interactions.

Protein Eng. 1995, 8, 127−134.

(3) Kwan, J. C.; Eksioglu, E. A.; Liu, C.; Paul, V. J.; Luesch, H.;

Grassystatins, A.-C. potent cathepsin E inhibitors that reduce antigen

presentation. J. Med. Chem. 2009, 52, 5732−5747.

(4) Williams, P. G.; Yoshida, W. Y.; Moore, R. E.; Paul, V. J. The

isolation and structure elucidation of tasiamide B, a 4-amino-3-

hydroxy-5-phenylpentanoic acid containing peptide from the marine

cyanobacterium Symploca sp. J. Nat. Prod. 2003, 66, 1006−1009.

(19) Felsenstein, K. M.; Hunihan, L. W.; Roberts, S. B. Altered

cleavage and secretion of a recombinant β-APP bearing the Swedish

familial Alzheimer’s disease mutation. Nat. Genet. 1994, 6, 251−256.

(20) Abbenante, G.; Kovacs, D. M.; Leung, D. L.; Craik, D. J.; Tanzi,

R. E.; Fairlie, D. P. Inhibitors of β-amyloid formation based on the β-

10764

dx.doi.org/10.1021/jm301630s | J. Med. Chem. 2012, 55, 10749−10765

Journal of Medicinal Chemistry

Article

secretase cleavage site. Biochem. Biophys. Res. Commun. 2000, 268,

133−135.

crystal structures and molecular dynamics simulations. Acta Crystallogr.

D Biol. Crystallogr. 2012, 68, 13−25.

(33) Otwinowski, Z.; Minor, W. Processing of X-ray diffraction data

collected in oscillation mode. Methods Enzymol. 1997, 276, 307−326.

(34) Kabsch, W. XDS. Acta Crystallogr. D Biol. Crystallogr 2010, 66,

125−132.

(35) Collaborative-Computational-Project-Number-4. The CCP4

suite: programs for protein crystallography. Acta Crystallogr. D Biol.

Crystallogr. 1994, 50, 760−763.

(36) Adams, P. D.; Grosse-Kunstleve, R. W.; Hung, L. W.; Ioerger, T.

R.; McCoy, A. J.; Moriarty, N. W.; Read, R. J.; Sacchettini, J. C.;

Sauter, N. K.; Terwilliger, T. C. PHENIX: building new software for

automated crystallographic structure determination. Acta Crystallogr. D

Biol. Crystallogr. 2002, 58, 1948−1954.

(21) 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.; Chen-Dodson, E.; Holloway, M. K.;

Munshi, S.; Simon, A. J.; Kuo, L.; Vacca, J. P. Structure-based design of

potent and selective cell-permeable inhibitors of human β-secretase

(BACE-1). J. Med. Chem. 2004, 47, 6447−6450.

(22) Colombo, A.; Wang, H.; Kuhn, P. H.; Page, R.; Kremmer, E.;

Dempsey, P. J.; Crawford, H. C.; Lichtenthaler, S. F. Constitutive α-

and β-secretase cleavages of the amyloid precursor protein are partially

coupled in neurons, but not in frequently used cell lines. Neurobiol. Dis.

2012, 49C, 137−147.

(23) (a) Hooper, N. M.; Trew, A. J.; Parkin, E. T.; Turner, A. J. The

role of proteolysis in Alzheimer’s disease. Adv. Exp. Med. Biol. 2000,

477, 379−390. (b) Parvathy, S.; Hussain, I.; Karran, E. H.; Turner, A.

J.; Hooper, N. M. Alzheimer’s amyloid precursor protein α-secretase is

inhibited by hydroxamic acid-based zinc metalloprotease inhibitors:

similarities to the angiotensin converting enzyme secretase.

Biochemistry 1998, 37, 1680−1685.

(24) (a) Wu, J.; Tung, J. S.; Thorsett, E. D.; Pleiss, M. A.; Nissen, J.

S.; Neitz, J.; Latimer, L. H.; John, V.; Freedman, S. Preparation of

cycloalkyl, lactam, lactone and related compounds for inhibiting β-

amyloid peptide release and/or its synthesis. PCT Int. Appl. WO

9828268 A2 19980702, 1998. (b) 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.; 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.

(25) Lewis, P. A.; Piper, S.; Baker, M.; Onstead, L.; Murphy, M. P.;

Hardy, J.; Wang, R.; McGowan, E.; Golde, T. E. Expression of BRI-

amyloid β peptide fusion proteins: a novel method for specific high-

level expression of amyloid β peptides. Biochim. Biophys. Acta 2001,

1537, 58−62.

(37) Emsley, P.; Cowtan, K. Coot: model-building tools for

molecular graphics. Acta Crystallogr. D Biol. Crystallogr. 2004, 60,

2126−2132.

(38) (a) Lanz, T. A.; Schachter, J. B. Solid-phase extraction enhances

detection of beta-amyloid peptides in plasma and enables Aβ

quantification following passive immunization with Aβ antibodies. J.

Neurosci. Methods 2008, 169, 16−22. (b) Lanz, T. A.; Schachter, J. B.

Demonstration of a common artifact in immunosorbent assays of brain

extracts: development of a solid-phase extraction protocol to enable

measurement of amyloid-beta from wild-type rodent brain. J. Neurosci.

Methods 2006, 157, 71−81. (c) Yohrling, G. J.; Felsenstein, K. M.;

Conway, K. A.; Zupa-Fernandez, A.; Brenneman, D. E.; Arnold, H. M.

A comparative analysis of brain and plasma Aβ levels in eight common

non-transgenic mouse strains: validation of a specific immunoassay for

total rodent Aβ. Curr. Alzheimer Res. 2007, 4, 297−303. (d) Moore, B.

D.; Chakrabarty, P.; Levites, Y.; Kukar, T. L.; Baine, A. M.; Moroni, T.;

Ladd, T. B.; Das, P.; Dickson, D. W.; Golde, T. E. Overlapping profiles

of Aβ peptides in the Alzheimer’s disease and pathological aging

brains. Alzheimer's Res. Ther. 2012, 4, 18.

(39) Hoff, J. Methods of blood collection in the mouse. Lab Anim.

2000, 29, 47−53.

(26) Lanz, T. A.; Hosley, J. D.; Adams, W. J.; Merchant, K. M.

Studies of Aβ pharmacodynamics in the brain, cerebrospinal fluid, and

plasma in young (plaque-free) Tg2576 mice using the γ-secretase

inhibitor 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-ala-

ninamide (LY-411575). J. Pharmacol. Exp. Ther. 2004, 309, 49−55.

(27) Li, C.; Nair, L.; Liu, T.; Li, F.; Pichardo, J.; Agrawal, S.; Chase,

R.; Tong, X.; Uss, A. S.; Bogen, S.; Njoroge, F. G.; Morrison, R. A.;

Cheng, K.-C. Correlation between PAMPA permeability and cellular

activities of hepatitis C virus protease inhibitors. Biochem. Pharmacol.

2008, 75, 2286−1197.

(28) Fujikawa, M.; Ano, R.; Nakao, K.; Shimizu, R.; Akamatsu, M.

Relationships between structure and high-throughput screening

permeability of diverse drugs with artificial membranes: application

to prediction of Caco-2 cell permeability. Bioorg. Med. Chem. 2005, 13,

4721−4732.

(29) Chang, W.; Huang, X.; Downs, D.; Cirrito, J. R.; Koelsch, G.;

Holtzma, D. M.; Ghosh, A. K.; Tang, J. β-secretase inhibitor GRL-

8234 rescues age-related cognitive decline in APP transgenic mice.

FASEB J. 2011, 25, 775−784.

(30) Stachel, S. J. Progress towards the development of a viable

BACE-1 inhibitor. Drug Dev. Res. 2009, 70, 101−110.

(31) Stanton, M. G.; Stauffer, S. R.; Gregro, A. R.; Steinbeiser, M.;

Nantermet, P.; Sankaranarayanan, S.; Price, E. A.; Wu, G.;

Crouthamel, M.; Ellis, J.; Lai, M.; Espeseth, A. S.; Shi, X.; Jin, L.;

Colussi, D.; Pietrak, B.; Huang, Q.; Xu, M.; Simon, A. J.; Graham, S.

L.; Vacca, J. P.; Selnick, H. Discovery of isonicotinamide derived β-

secretase inhibitors: In vivo reduction of β-amyloid. J. Med. Chem.

2007, 50, 3431−3433.

(32) Xu, Y.; Li, M. J.; Greenblatt, H.; Chen, W.; Paz, A.; Dym, O.;

Peleg, Y.; Chen, T.; Shen, X.; He, J.; Jiang, H.; Silman, I.; Sussman, J.

L. Flexibility of the flap in the active site of BACE1 as revealed by

10765

dx.doi.org/10.1021/jm301630s | J. Med. Chem. 2012, 55, 10749−10765

Article Doi

Cyanobacterial peptides as a prototype for the design of potent β-secretase inhibitors and the development of selective chemical probes for other aspartic proteases

DOI: 10.1021/jm301630s

Source and publish data:

Authors:

Article abstract of DOI:10.1021/jm301630s

Full text of DOI:10.1021/jm301630s

Products guided by the article

R&D Labs maybe for 1415034-19-3

Relevant to this article

Hot Product