Inhibition of amyloid ?-peptide production by blockage of ?-secretase cleavage site of amyloid precursor protein

Inhibition of amyloid b-peptide production by blockage ofb-secretase cleavage site of amyloid precursor protein

Chan Hyun Na,*,� Sang Hee Jeon,* Guangtao Zhang,� Gary L. Olson� and Chi-Bom Chae*,�

*Department of Life Science, Division of Molecular and Life Sciences, Pohang University of Science and Technology, Pohang, Korea

�Institute of Biomedical Science and Technology, Department of Biomedical Science and Technology, Konkuk University, Seoul, Korea

�Provid Pharmaceuticals Inc., North Brunswick, New Jersey, USA

Abstract

Amyloid b-peptide (Ab) is implicated as the major causative

agent in Alzheimer’s disease (AD). Ab is produced by the

processing of the amyloid precursor protein (APP) by BACE1

(b-secretase) and c-secretase. Many inhibitors have been

developed for the secretases. However, the inhibitors will

interfere with the processing of not only APP but also of other

secretase substrates. In this study, we describe the develop-

ment of inhibitors that prevent production of Ab by specific

binding to the b-cleavage site of APP. We used the hydro-

pathic complementarity (HC) approach for the design of short

peptide inhibitors. Some of the HC peptides were bound to the

substrate peptide (Sub W) corresponding to the b-cleavage

site of APP and blocked its cleavage by recombinant human

BACE1 (rhBACE1) in vitro. In addition, HC peptides specific-

ally inhibited the cleavage of Sub W, and not affecting other

BACE1 substrates. Chemical modification allowed an HC

peptide (CIQIHF) to inhibit the processing of APP as well as

the production of Ab in the treated cells. Such novel APP-

specific inhibitors will provide opportunity for the development

of drugs that can be used for the prevention and treatment of

AD with minimal side effects.

Keywords: b-amyloid, Alzheimer’s disease, amyloid precur-

sor protein, APP-specific inhibitor, BACE1, hydropathic com-

plementarity.

J. Neurochem. (2007) 101, 1583–1595.

Alzheimer’s disease (AD) is a neurodegenerative disorderthat currently affects nearly 2% of the population inindustrialized countries, and the incidence of the AD willincrease threefold within the next 50 years (Vickers et al.2000; Mattson 2004). The central tenet of the amyloidhypothesis of AD is that amyloid b-peptide (Ab) is the majorcausative agent of the disease process (Selkoe 1994; Doveyet al. 2001; Sommer 2002; Mattson 2004). Ab is generatedfrom the amyloid precursor protein (APP) by endoproteolyticprocessing by BACE1 and c-secretase (Sinha and Lieberburg1999; De Strooper and Annaert 2000). Therefore, productionof Ab can be blocked by the inhibition of secretases, andmany inhibitors for these enzymes have been developed(Olson et al. 2001; John et al. 2003; Cumming et al. 2004;Harrison et al. 2004).

However, APP is not the only substrate for secretases. Tillnow, many BACE1 substrates other than APP have beendiscovered including sialyltransferase ST6Gal1 (Kitazumeet al. 2001, 2003), the adhesion protein P-selectin glycopro-tein ligand-1 (Lichtenthaler et al. 2003), b subunits ofvoltage-gated sodium channels (Wong et al. 2005),

APP-like proteins (Li and Sudhof 2004), and the type IIIisoform of the epidermal growth factor-like factor neuregulin1 (type III-NRG1) (Willem et al. 2006). Especially, type III-NRG1 is known to be involved in myelination and correctbundling of axons of nerve cells. Therefore, inhibitors for theenzymes will also block the normal processing of othersubstrates of the enzymes, thus leading to various side effects(Kitazume et al. 2001, 2003; Ikeuchi and Sisodia 2003;

Received October 10, 2006; revised manuscript received December 11,2006; accepted December 15, 2006.Address correspondence and reprint requests to Chi-Bom Chae,

Institute of Biomedical Science and Technology, Konkuk University,Seoul 143-710, Korea. E-mail: [email protected] used: AD, Alzheimer’s disease; APP, amyloid precursor

protein; Ab, amyloid b-peptide; BCA, bicinchoninic acid; CTFb, theC-terminal fragment of APP generated by BACE1; FRET, fluorescenceresonance energy transfer; f-Sub M, 7-methoxycoumarin-4-acetyl-SEV-NLDAEFRK(Dnp)-RR-NH2; HC, hydropathic complementarity; HEK293 cells, human embryonic kidney 293 cells; MTT, 3-[4,5-dimethyl-thiazol-2-yl]-2,5-diphenyltetrazolium bromide.PVDF, polyvinylidenedifluoride; rhBACE1, recombinant human BACE1.

Journal of Neurochemistry, 2007, 101, 1583–1595 doi:10.1111/j.1471-4159.2006.04441.x

� 2007 The AuthorsJournal Compilation � 2007 International Society for Neurochemistry, J. Neurochem. (2007) 101, 1583–1595 1583

Lichtenthaler et al. 2003; Periz and Fortini 2004; Willemet al. 2006). If a molecule is developed that can bind andprotect the cleavage site of APP from secretase, the sideeffects may be minimized. In this report, we focused on theb-cleavage site of APP. A molecule that can specifically bindto the b-cleavage site of APP will probably inhibit theproduction of Ab without affecting the processing of otherBACE1 substrates. A similar approach was demonstratedwith the antibody that binds to the b-cleavage site of APP(Arbel et al. 2005). However, delivery of the antibody intothe cell as well as brain may be difficult.

We used the hydropathic complementarity (HC) approach(Blalock and Smith 1984; Bost et al. 1985; Heal et al. 2002)to design short peptides that bind to the b-cleavage site ofAPP. According to the HC hypothesis, the peptide sequencesthat are predicted from the codons read from the non-codingstrand of a gene either in the 5¢ fi 3¢ or 3¢ fi 5¢ directiontend to bind to the protein derived from the coding strand.Several examples demonstrate the successful application ofthis approach. Antagonists were developed for ACTH,ribonuclease S peptide, c-Raf protein, fibronectin, insulin,a-chain of fibrinogen, and angiogenin (Bost et al. 1985; Shaiet al. 1987; Brentani et al. 1988; Knutson 1988; Fassinaet al. 1989; Gho and Chae 1997; Heal et al. 1999, 2002).

We used a 10 amino acid region of either wild type APP(Sub W) or APP that contains Swedish mutations on the leftside of the b-cleavage site (Sub M) as the target for designingHC inhibitory peptides. The cleavage site of b-secretase waslocated in the center of the substrate peptides.

Some of the HC peptides were bound to both Sub M andSub W, and also inhibited cleavage of the substrate byb-secretase. In addition, HC peptides specifically inhibitedthe cleavage of Sub W, while not significantly affecting otherBACE1 substrates.

A chemically modified HC peptide inhibited processingof APP as well as the production of Ab from the treatedcells. Thus, APP-targeted inhibitor described in this reportpresents a new opportunity for the development of drugs forthe prevention and treatment of AD with minimal sideeffects.

Experimental procedures

Peptide synthesis

All the non-labeled and amidated peptides were synthesized with

purity better than 95% by A&Pep Co., Inc. (Choong Nam, Korea).

All the HC peptides were amidated. The peptides labeled with biotin

at the N-terminus were synthesized by Peptron, Inc. (Daejeon,

Korea). The purity and identity of the peptides were verified by high

performance liquid chromatography (HPLC) and mass spectrometry.

APP b-secretase inhibitor and b-secretase inhibitor IV were

purchased from Calbiochem (Darmstadt, Germany). Chemically

modified HC peptides were provided by Provid Pharmaceuticals Inc.

(North Brunswick, NJ, USA).

Cell culture

Human embryonic kidney (HEK) 293 cells stably transformed with

the gene for APP695 (HEK293-APP) were used for studies on the

inhibitory activity of HC peptides. HEK293-APP cells were

generously supplied by Dr T.W. Kim (Columbia University, NY,

USA). HEK293-APP cells were cultured in Dulbecco’s modified

Eagle’s medium (DMEM) (Invitrogen, Carlsbad, CA, USA) with

10% fetal bovine serum (HyClone, Logan, UT, USA) and 300 lg/mL of geneticin (Invitrogen) in a humidified atmosphere of 5% CO2,

95% air at 37�C. The cells were subcultured after trypsinization, and

the medium was changed every 2–3 days.

Binding of BACE1 substrates to HC peptides

Reacti-BindTM maleic anhydride-activated polystyrene plate (Pierce,

Rockford, IL, USA) was coated with 50 lL of HC peptide

(200 lmol/L in distilled water) by chemical coupling for 3 h at

25�C, and washed thrice with distilled water and thrice with

phosphate-buffered saline (PBS) containing 0.05% tween 20 (PBST).

The plate was blocked with blocking buffer (0.5% gelatin in PBS) for

1 h at 25�C. After the blocking buffer was discarded, biotinylated

BACE1 substrates (200 lmol/L) in the blocking buffer were applied

on the plate and incubated for 2.5 h at 25�C. Each well was washed

thrice with PBST. The bound biotinylated BACE1 substrates were

detected by incubation with streptavidin–horseradish peroxidase (Str-

HRP, 125 mU/mL in blocking buffer) for 2 h at 25�C. Color reactionwas carried out with 50 lL of 3,3¢,5,5¢-tetramethylbenzidine liquid

substrate (Sigma–Aldrich, St Louis, MO, USA). After stopping the

reaction by the addition of an equal volume of 1 NHCl, absorbance at

450 nm was read in an automated ELISA reader (EL 312e, Bio-Tek

Instruments, Winooski, VT, USA).

Assay for inhibitory activity of HC peptides using fluorescence

resonance energy transfer system in vitro

In this assay system, fluorescence resonance energy transfer (FRET)

technology was used. A Swedish mutant APP substrate (f-Sub M)

with a fluorescent donor and a proprietary quenching acceptor

(7-methoxycoumarin-4-acetyl-SEVNLDAEFRK(Dnp)-RR-NH2)

was purchased from R&D Systems, Inc. (Minneapolis, MN, USA).

The donor fluorescence energy is significantly quenched by the

acceptor. Upon cleavage of the substrate by rhBACE1 (R&DSystems,

Inc.), the fluorescence donor is separated from the quenching group,

restoring the full fluorescence yield of the donor. f-SubM (10 lmol/L)

was pre-incubated with HC peptide in the assay buffer (0.1 mol/L

NaOAc, pH 4.0) for 2 h at 25�C. After pre-incubation, the substrateand HC peptide mixture in the assay buffer was transferred to a

FluoroNuncTM 96 well white plate (Nunc, Roskilde, Denmark), and

rhBACE1 (35 nmol/L) was added. The time-dependent emission of

fluorescence (excitation at 320 nm, emission at 405 nm) was

monitored in a Molecular Devices SpectraMax Gemini EM fluores-

cence reader (Sunnyvale, CA, USA) for 1 h at 37�C.

Assay for inhibitory activity of HC peptides on Sub W using

HPLC system

Reversed-phase HPLC was also used to analyze the products of

enzyme action in vitro. After 25 lmol/L of SubWand 750 lmol/L of

each HC peptide were pre-incubated for 2 h at 25�C, 2 lmol/L (final

concentration) of BACE1 was added to the mixture. The mixture was

incubated further for 12 h at 25�C. The cleavage products were

1584 C. H. Na et al.

Journal Compilation � 2007 International Society for Neurochemistry, J. Neurochem. (2007) 101, 1583–1595� 2007 The Authors

separated on a C18 reversed-phase HPLC column (Grace VyDac,

Hesperia, CA, USA) using the New Agilent 1200 Series HPLC

system (Santa Clara, CA, USA) (Kitazume et al. 2003). The sample,

applied to the column equilibrated in 0.1% trifluoroacetic acid (TFA)

in double distilled water, was then eluted with a gradient of 0–70% of

0.1% TFA in acetonitrile for 40 min. The elution rate was 1 mL/min.

The cleavage products were identified at 215 nm and quantitated by

integrating the area under each peak.

Assay for inhibitory activity of PV557 on the cleavage of various

BACE1 substrates

PV557 was pre-incubated with Sub W (25 lmol/L), Sub-ST6Gal1

peptide (25 lmol/L, DYEALTLQAKEFQMPKSQE), or Sub 28

peptide (25 lmol/L, SGISYEVESRHDW), for 2 h at 25�C. Themixture containing Sub W was treated with 2 lmol/L of rhBACE1

for 12 h at 25�C. The mixture containing Sub-ST6Gal1 was treated

with 1 lmol/L of rhBACE1 for 12 h at 25�C. The mixture containing

Sub 28 was treated with 11 nmol/L of rhBACE1 for 12 h at 25�C.Under these reaction conditions, roughly 50% of each BACE1

substrate was cleaved in the absence of the HC peptide. The cleavage

products were separated on a C18 reversed-phase HPLC column as

described above.

Measurement of CTFb level

HEK293-APP cells were plated on 6-well culture plate (Nunc,

Roskilde, Denmark) coated with poly-D-lysine (Sigma-Aldrich).

When the confluency of the cells reached 90%, the cells were washed

with PBS once, and the HC peptide in serum-free DMEM was added

to the cells. After incubation for 9 h in a humidified CO2 incubator, the

cells were lysed in the following solution [10 mmol/L Tris–HCl, pH

7.4, 150 mmol/L NaCl, 1% Triton X-100, 0.25% Nonidet P-40,

2 mmol/L EDTA supplemented with the protease inhibitor mixture

(Sigma-Aldrich)] and scraped with a cell scraper. The lysed cells were

centrifuged at 12 000 g for 10 min at 4�C. The protein in the super-

natant was determined by bicinchoninic acid (BCA) assay (Pierce)

(Smith et al. 1985). After being heated in boiling water, the protein

sample (150 lg) in lithium dodecyl sulfate sample buffer (Invitrogen)

was loaded on 4–12% bis-tris NuPAGE gel (Invitrogen). After

electrophoresis, the proteins in the gel were electrophoretically

transferred onto a polyvinylidene difluoride (PVDF) membrane at

100 mA for 80 min (Ida et al. 1996). The blottedmembranewas fixed

with 0.2% glutaraldehyde in PBS for 45 min at 25�C and treated for

5 min in boiling PBS. Full-length APP and CTFb (the C-terminal

fragment of APP generated by BACE1) were detected by treating the

membrane with 0.5 lg/mL of anti-Ab N-terminal 6E10 antibody

(Signetlabs Inc., Dedham,MA,USA) followed by incubationwith 0.2

lg/mL of sheep anti-mouse antibody conjugated with HRP (Amer-

sham Biosciences Ltd, Uppsala, Sweden). The developed film was

scanned, and the densities of the APP and CTFb bands were deter-

mined byMultiGauge program (Fuji Photo FilmCo. Ltd, Tokyo, Japan).

Measurement of Ab level

For the measurement of secreted Ab level, the culture media from

the peptide-treated cells (see above) were harvested and centrifuged

at 3500 g for 10 min at 4�C. The amounts of Ab40 and Ab42 in the

supernatant were determined with human Ab40 and Ab42 immuno-

assay kits (Signal SelectTM, BioSource, Camarillo, CA, USA)

according to the instructions provided by the company.

For the measurement of intracellular Ab level, cells were

collected from each well by using a cell scraper, centrifuged at

3000 g for 2 min, washed with PBS, and resuspended in 100 lL of

70% formic acid, followed by 10 s sonication. The solution was

then centrifuged at 100 000 g for 20 min at 4�C to remove insoluble

material, and supernatant was collected and neutralized with 1.9 mL

of 1 mol/LTris (pH 9.0). All samples were analyzed for their protein

concentration by using BCA assay (Pierce). The amounts of Ab40

and Ab42 in the supernatant were determined with human Ab40 andAb42 immunoassay kits (Signal SelectTM, BioSource) according to

the instructions provided by the company.

Cytotoxicity assay

Colorimetric MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetra-

zolium bromide) reduction method was used to determine the

cytotoxicity of PV557 on cells. After treatment with an inhibitor,

1/10 volume of 5 mg/mL MTT (Sigma-Aldrich) was added to the

cells, and the cells were incubated for 3 h. At the end of the

incubation period, the medium was removed and the converted dye

in the cells was solubilized with solibilization solution (10% SDS,

25% dimethylformamide) Then, absorbance at 570 nm was meas-

ured with a Molecular Devices SpectraMax microplate reader.

Assay for stability of PV557 in serum

Blood from animals (9-week-old male SD Rats) was collected in

sterile centrifuge tubes, allowed to clot at 4�C, and then centrifuged

twice at 1500 g for 10 min at 4�C and 20 000 g for 20 min at 4�C.The supernatant (serum) was filtered with a sterile filter (0.22 lm,

Millipore Co., Medford, MA, USA) and stored in small aliquots at

)70�C until use. For estimating the stability of PV557 in serum,

PV557 (25 lg in 50 lL PBS) was incubated with 50 lL of filtered

S E F C I Q I H F R

S E V N L D A E F R

R L H L D L R L K A

S E F C I H L H F R

APP

S E V K M D A E F R

R L H F Y L R L K A

S E F C I Q I H F R

Swedish mutant typeof APP

S E V N L D A E F R

R L H L D L R L K A

S E F C I H L H F R

S E V K M D A E F R

R L H F Y L R L K A

Sub W

c-Sub W

Sub W-c

Sub M

c-Sub M

Sub M-c

Complementarypeptide

Complementarypeptide

Fig. 1 Design of HC peptide using hydropathic complementary

approach. The non-coding strands of DNA sequences corresponding to

the 10 amino acids surrounding the b-cleavage site of wild-type and

Swedish mutant type of APP were read either in 5¢ fi 3¢ or 3¢ fi 5¢directions, and codons were predicted. Sub W: Wild type APP substrate;

c-Sub W: The HC peptide sequence read from Sub W DNA in 5¢ fi 3¢direction; Sub W-c: The HC peptide sequence read from Sub W DNA in

3¢ fi 5¢ direction; Sub M: Swedish mutant APP substrate; c-Sub M: The

HC peptide sequence read from Sub M DNA in 5¢ fi 3¢ direction; Sub M-

c: The HC peptide sequence read from Sub M DNA in 3¢ fi 5¢ direction.

APP-specific inhibitor of b-secretase 1585

� 2007 The AuthorsJournal Compilation � 2007 International Society for Neurochemistry, J. Neurochem. (2007) 101, 1583–1595

rat serum at 37�C for the indicated time. The incubation mixture was

fractionated on a C18 reversed-phase HPLC column (Grace VyDac)

using the New Agilent 1200 Series HPLC system (Yoo et al. 2005).The sample, applied to the column equilibrated in 0.1% trifluoro-

acetic acid (TFA) in double distilled water, was then eluted with a

gradient of 0–70% of 0.1% TFA in acetonitrile for 40 min. The

elution rate was 1 mL/min. PV557 peak was identified at 215 nm

and quantitated by integrating the area under the peak.

Stability of PV557 when exposed to cells

HEK293-APP cells were plated on 6 well culture plate (Nunc,

Roskilde, Denmark) coated with poly-D-lysine (Sigma-Aldrich).

When the confluency of the cells reached 90%, the cells were

washed with PBS twice, and 25 lmol/L of PV557 in serum-free

DMEM was added to the cells. After the incubation for indicated

time, cells were added with 1/10 volume of 10 · cell solubilization

buffer [250 mmol/L Tris, pH 8.0, 5% Triton X-100, 5% Nonidet P-

40 supplemented with the 10 · protease inhibitor mixture (Sigma-

Aldrich)] and the remaining attached cells were collected using a

cell scraper. The mixture containing lysed cells and culture media

was sonicated for 30 s, and diluted with 2 mL of 1 · cell

solubilization buffer and filtered with a sterile filter (0.22 lm,

Millipore Co.). The filtered sample was fractionated on a C18

reversed-phase HPLC column (Grace VyDac) using the New

Agilent 1200 Series HPLC system. The sample, applied to the

column equilibrated in 0.1% trifluoroacetic acid (TFA) in double

distilled water, was then eluted with a gradient of 0–70% of 0.1%

TFA in acetonitrile for 40 min. The elution rate was 1 mL/min.

PV557 peak was identified at 215 nm and quantitated by integrating

the area under the peak.

Results

APP targeted inhibitors were designed based on the HC

approach

The first step of Ab generation from APP is the cleavage ofAPP at the b-site by BACE1. If a molecule that binds to thecleavage site is developed, such a molecule will most likelyinhibit the cleavage of APP by blocking the approach ofBACE1. We designed decapeptides that are complementaryto the 10 amino acids derived from the b-cleavage site ofAPP, as shown in Fig. 1. The decapeptide (SEVKM/DAE-FR) (Vassar 2002) containing the b-cleavage site of the wildtype of APP was designated as Sub W, and the corresponding

(a)

(b)

(c)

Fig. 2 Binding and inhibitory activity of HC peptides for Sub M. (a)

Binding of Sub M to HC peptides. HC peptides were chemically cou-

pled on the wells of a microtiter plate, and biotin-labeled Sub M was

applied to the HC peptide-coated wells. The binding was detected with

Str-HRP as described in Experimental procedures. (b) Effect of HC

peptides on the cleavage of f-Sub M by rhBACE1. HC peptides

(1 mmol/L) were pre-incubated with f-Sub M (10 lmol/L) for 2 h at

25�C and cleaved with rhBACE1 as described in Experimental pro-

cedures. (c) Binding both of Sub M and Sub W to four HC peptides.

HC peptides were chemically coupled on the wells of a microtiter plate,

and biotin-labeled either Sub M or Sub W was applied to the HC

peptide-coated wells. The binding was detected with Str-HRP as

described in Experimental procedures. These experiments are a

representative of several independent experiments in duplicate. Val-

ues shown are mean ± range.



DNA sequence was used for prediction of HC peptides. Thepeptide read in the 5¢ fi 3¢ direction from the non-codingstrand was designated as c-Sub W (SEFCIHLHFR), and thepeptide sequence read in the 3¢ fi 5¢ direction was designa-ted as Sub W-c (RLHFYLRLKA), respectively. The deca-peptides corresponding to the b-cleavage site of the Swedishmutant type of APP was designated as Sub M (SEVNL/DAEFR) (Lannfelt et al. 1993; Vassar 2002), and the two

HC decapeptide sequences derived from the non-codingstrand of Sub M DNA were designated as c-Sub M(SEFCIQIHFR) and Sub M-c (RLHLDLRLKA), respect-ively (Fig. 1).

HC peptides bound to substrate peptides and inhibited

cleavage by BACE1

The wild-type APP substrate is a poor substrate for BACE1in vitro. On the other hand, the Swedish mutant APP isreadily cleaved by the enzyme (as much as 47 times morethan the wild-type APP substrate) in vitro (Tomasselli et al.2003). Therefore, we initially focused on the HC peptidesthat correspond to Sub M. In order to investigate if the HCpeptides bind to the substrate peptide, the peptides werechemically coupled on the surface of a microtiter plate tominimize variation caused by the different coating efficiencyof each peptide, and the Sub M labeled with biotin (bio-SubM) was applied. c-Sub M, but not Sub M-c, showed bindingactivity for Sub M (Fig. 2a).

For cleavage of Sub M, we employed the fluorescenceassay described in Experimental Procedures. The decapep-tide substrate containing a fluorescence group and a quencheron either side of the molecule is commonly used for in vitroassay of BACE1 (Gruninger-Leitch et al. 2002; Shuto et al.2003). Interestingly, c-Sub M that bound Sub M inhibited thecleavage of Sub M by BACE1. Consistent with the result ofbinding to the substrate peptide, c-Sub M but not Sub M-c,inhibited the cleavage of Sub M (Fig. 2b). Approximately50% inhibition was obtained at a concentration of c-Sub Mthat was 100-fold higher than that of the substrate.

Most AD patients have wild-type APP (Lannfelt et al.1993; Ruiz-Opazo et al. 2004). Therefore, the activity of theHC peptide on Sub W has a very important meaning. Thus, 4HC peptides were tested for their binding activity to bothsubstrates. c-Sub M showed rather higher binding activity toSub W than c-Sub W did (Fig. 2c). This result suggests thatc-Sub M can be used both for Sub M and Sub W in studieson the inhibition of their cleavages.

Activities of deletion types of HC peptides

In order to investigate theminimal core sequence necessary forinhibition of cleavage of Sub M by HC peptides, we made

(a)

(b)

(c)

Fig. 3 Inhibitory activities of deletion types of HC peptides for Sub M.

(a) Amino acids of c-Sub M were deleted from either side of c-Sub

M. (b) These deletion types of HC peptides were tested for their

inhibitory activity for the cleavage of f-Sub M by rhBACE1. HC peptides

(1 mmol/L) were preincubated with f-Sub M (10 lmol/L) for 2 h at 25�Cand cleaved with rhBACE1 as described in Experimental procedures.

The concentration of APP b-secretase inhibitor was 30 lmol/L. (c)

c-Sub M DC1 was further deleted from the N-terminus. HC peptides

(0.5 mmol/L) were preincubated with f-Sub M (10 lmol/L) for 2 h at

25�C and cleaved with rhBACE1 as described in Experimental proce-

dures. The experiments shown are representative of three independent

experiments. Values shown are mean ± range.



serial deletions from either ends of c-Sub M, as shown inFig. 3a. When one amino acid was deleted from the N-terminus of c-Sub M, inhibitory activity was increased about

twofold (from 45% to 90%). Deletion of an additional aminoacid (c-Sub M DN2) rather decreased inhibitory activity.Deleting 3 amino acids from the N-terminus (c-Sub M DN3)completely abolished the inhibitory activity. Deleting 1 aminoacid from the C-terminus (c-Sub M DC1) increased theinhibitory activity twofold (from 45% to 98%). Deleting anadditional amino acid decreased the inhibitory activity to 25%.On the other hand, deleting 3, 4, and 5 amino acids from the C-terminus increased the inhibitory activity two times (greaterthan 90%) as much as deleting 1 amino acid from the C-terminus. We noted, however, that these three deletion formsaggregated during the enzyme assay. A known APP b-secretase inhibitor (30 lmol/L, KTEEISEVN-Statin-VAEF;(Sinha et al. 1999)) also inhibited cleavage of SubM (Fig. 3b).

Since c-Sub M DC1 showed the highest inhibitory activity,and other C-terminal deletion types of HC peptides had atendency to aggregate during the assay, we decided to focuson c-Sub M DC1 for further analysis. Serial deletion of c-SubM DC1 from the N-terminus showed that up to 3 amino acids(c-Sub M DN3C1; CIQIHF) can be removed withoutsacrificing the inhibitory activity too much (Fig. 3c). Basedon the above studies, CIQIHF is considered the coresequence. CIQIHF showed about 50% inhibition of cleavageof Sub M at inhibitor/substrate ratio of 10 (data not shown).

Inhibition of cleavage of Sub W by CIQIHF

As mentioned above, most AD patients have the wild-typeAPP sequence. Therefore, it was important to investigate ifCIQIHF also binds to Sub W and inhibits cleavage byBACE1. As shown in Fig. 4a, both c-Sub M and CIQIHF (c-Sub M DN3C1) bound to Sub W. The decapeptide c-Sub Mshowed higher binding activity than hexamer CIQIHF. Itshould be noted that the shorter peptide fixed on the plasticsurface may be less favorable than the decapeptide forbinding to the biotin-labeled decapeptide substrate. There-fore, the binding assay used in this report will give us onlyqualitative information. For studies on the inhibitory activityof CIQIHF, non-labeled Sub W was used, since there is nocommercially available Sub W that is labeled with fluores-cence. Cleavage of Sub W was analyzed by HPLC. CIQIHFalso inhibited cleavage of Sub W by BACE1. About 30%inhibition was observed at the inhibitor/substrate ratio of 50(Fig. 4b). For this assay, 57-fold higher amount of enzyme(2 lmol/L) had to be used compared to the cleavage of SubM due to the poor cleavage of Sub W compared with Sub M.

Structural modification of an HC inhibitor

We found that HC peptides, including the shortest CIQIHF,do not enter into the cells (data not shown). Therefore,structural modification of an HC peptide was required fordelivery of the peptides into cells. We chose c-Sub M DN3C1(CIQIHF) for structural modifications.

Alanine scanning mutations of CIQIHF showed thatcysteine is important for the inhibitory activity (data not

(a)

(b)

Fig. 4 Binding to and inhibition of cleavage of Sub W by c-Sub M

DN3C1 (CIQIHF). (a) Binding between HC peptide and bio-Sub W. HC

peptides were chemically coupled to the surface of a microtiter plate,

and biotin-labeled Sub W was applied. The binding of bio-Sub W was

determined as described in Experimental procedures. This experiment

was repeated twice and results are means of duplicate determinations

in one of the two experiments. (b) Inhibition of cleavage of Sub W by

BACE1. Sub W (25 lmol/L) was pre-incubated with HC peptides

(750 lmol/L) for 2 h at 25�C and cleaved with rhBACE1 (2 lmol/L) for

12 h at 25�C. The cleavage rate was detected with reversed-phase

HPLC as described in Experimental procedures. This experiment was

repeated twice. Values shown are mean ± range.



shown). As a first step, the N-terminus of c-Sub M DN3C1was modified. To allow an HC peptide to pass through thecell membrane, adding lipophilicity would be helpful(Scherrmann 2002). Therefore, either a carbon chain or abenzene ring was added to the N-terminus maintaining thethiol group. The 6 compounds listed in Table 1 showed

inhibitory activity for cleavage of Sub M. Although theinhibitory activity for cleavage of Sub M by BACE1 wasonly one third of the starting structure, only PV557showed inhibitory activity for the cleavage of APP in thecells, as will be shown in Fig. 5. PV557 contains 6-aminohexanoic acid linked to the N-terminus of c-Sub M

Table 1 Structural modification of c-Sub M DN3C1

−

−

−

−

−

+

−

Activity in the cells

80

120

71

144

260

295

95

IC50 (μmol/L)

PV731

PV723

PV700

PV699

PV698

PV557

c-Sub M ΔN3C1

Name

IQIHF-NH2

IQIHF-NH2

IQIHF-NH2

IQIHF-NH2

IQIHF-NH2

IQIHF-NH2

IQIHF-NH2

Structure

O

SH

NO

O

SH

N

O

SH

CH3

O

O

SH

O

SH

N

O

SH

H

N

O

SH

NH2

O

The N-terminus of c-Sub M DN3C1 (CIQIHF) was modified either by substituting or adding various chemicals. Their IC50 for inhibition of cleavage of

Sub M was determined by fluorescence assay as described in Experimental procedures. The concentration of f-Sub M was 10 lmol/L. The

inhibition of release of Ab and accumulation of CTFb in HEK293-APP cells by modified HC peptides were carried out as described in Experimental

procedures. Results of IC50 are the mean of two repeats.



DN3C1. PV557 also inhibited cleavage of Sub W byBACE1 (Fig. 5a).

When tested on the HEK293-APP cells, PV557 inhibitedproduction of Ab as well as processing of APP in the treatedcells. When HEK293-APP cells were treated with increasingconcentrations of PV557, the amount of Ab released into theculture medium was reduced in a concentration-dependentmanner. At 25 lmol/L of PV557, the amounts of Ab40 andAb42 released was reduced by about 64% and 35%,respectively (Fig. 5b). IC50 of PV557 for the inhibition ofAb40 secretion was about 15 lM. The inhibition of Ab42

production by PV557 was less effective than that Ab40.Similar pattern was observed with the cell-permeableBACE1 inhibitor (Stachel et al. 2004). As there have beenmany reports on the importance of intracellular Ab for thecytotoxicity (Echeverria and Cuello 2002), effect of PV557on the intracellular level of Abwas also investigated. PV557appears to be more effective for inhibition of intracellularAb40 production. About 50% inhibition of Ab40 productionwas observed at 3.12 lM PV557. On the other hand, only22% reduction was observed for Ab42 production at the sameconcentration (Fig. 5c). In contrast, b-secretase inhibitor IV

(b)

β-secretaseinhibitor IV (μmol/L)

PV557 (μmol/L)

2 Blank 0.78 1.56 3.12 6.25 12.5 25

CTFβ-PCTFβ

FL-APPMatureImmature

(d)

(c)

(e)

(a)



showed very poor inhibitory activity for the production ofintracellular Ab (about 8% for both Ab40 and Ab42). BothPV557 and b-secretase inhibitor IV also inhibited accumu-lation of the cleavage product CTFb, the fragment spanningfrom the b-cleavage site to the C-terminus of APP (Figs 5dand e). Of interest is that b-secretase inhibitor IV inhibitedaccumulation of only the non-phosphorylated form of CTFb.On the other hand, PV557 inhibited the accumulation of bothforms of CTFbs. Densitometric tracing of the gel pattern(Fig. 5e) showed that PV557 was somewhat more effectivein inhibiting the accumulation of phosphorylated CTFb thannon-phosphorylated CTFb.

PV557 showed higher inhibitory activity for Sub W and

Sub M than other substrates

We investigated whether the HC peptide would showspecificity for APP in terms of the inhibition of cleavage byBACE1. We investigated the inhibitory activity of PV557 forcleavage of various BACE1 substrates in vitro. Since differentsubstrates were cleaved by BACE1 at widely different rates,we chose conditions in which about 50% of substrates werecleaved by BACE1. The cleavage products were fractionatedby reversed phase HPLC and quantitated. The followingsubstrates were used: Sub W, Sub M, Sub-ST6Gal1, and Sub28. Sub-ST6Gal1 is the peptide composed of 19 amino acids(DYEALTL/QAKEFQMPKSQE) derived from the b-clea-vage site of ST6Gal1 (Kitazume et al. 2001, 2003). Sub 28 isthe peptide composed of 13 amino acids (SGISY/EVES-RHDW) that is known to be about a 420-fold better substratefor BACE1 than Sub W in vitro (Tomasselli et al. 2003). Theresult shows that PV577 blocked the cleavage of Sub W andSub Mmore efficiently than the other substrates (Fig. 6a). Wealso investigated the binding activity of PV557 for Sub WandSub-ST6Gal1. PV557 showed a 2.4-fold higher bindingactivity for bio-Sub W than bio-ST6Gal1. It was alsoimportant to investigate if PV557 binds to Ab since Ab is

cleaved by other types of secretase, especially a-secretase andother isoforms of b-secretase (Walter et al. 2001). The resultshows that PV557 does not bind to Ab42 (Fig. 6b).

Cytotoxicity and stability of PV557

Possible cytotoxocity of PV557 on HEK293-APP cells wasinvestigated by colorimetric MTT assay. MTT assay is basedon reduction of MTT by mitochondrial dehydrogenases(Shearman et al. 1995). The result shows that PV557 is non-toxic to the cells at up to 6.25 lmol/L. At 12.5 lmol/L and25 lmol/L, PV557 showed 7% and 14% of cytotoxicity,respectively (Fig. 7).

Stabilities of PV557 either in serum or during cell culturewere investigated. For the investigation of stability of PV557in animal serum, 25 lg of PV557 was incubated with ratserum for various time periods and remaining PV557 wasseparated by HPLC method. The half-life of PV557 in ratserum was about 2 h (Fig. 8).

Stability of PV557 during cell culture was also investi-gated. For this assay, HEK293-APP cells were treated withPV557 in serum free media and incubated for various timeperiods. After incubation, the PV557 in the culture mediumand the cells (combined) was analyzed by HPLC. The half-life of PV557 was about 16 h. (Fig. 8).

Discussion

In this report, we investigated the possibility of developingdrugs for Alzheimer’s disease by blocking the cleavage sitesof APP from the access of secretases. This was important inview of the potential side effects of secretase inhibitors due tothe non-specific inhibition of normal processing of essentialproteins by secretases. The results obtained with the HCpeptides targeted to the 10 amino acid region flanking theb-cleavage site of APP suggest that one can indeed developdrugs that specifically target the cleavage site of APP and

Fig. 5 Inhibitory activity of modified form of a HC peptide on pro-

duction of Ab. (a) Inhibition on the cleavage of Sub W by PV557. Either

PV557 (750 lmol/L) or c-Sub M DN3C1 (750 lmol/L) was preincu-

bated with Sub W (25 lmol/L) and the cleavage of Sub W by rhBACE1

(2 lmol/L) was carried out as described in Experimental procedures.

This experiment was repeated twice. Values shown are mean ± range.

(b) Effect of PV557 on the production of secreted Ab from the treated

cells. HEK293-APP cells were treated with different concentrations of

PV557 for 9 h, and the culture media were harvested. The amount of

Ab40 and Ab42 released was determined by the sandwich ELISA

system as described in Experimental procedures. As a control, the

effect of a known inhibitor for BACE1, b-secretase inhibitor IV, was

also investigated. The amount of Ab secreted from the non-treated

cells and peptide-treated cells was compared. The experiment shown

is representative of three independent experiments in triplicate. Values

shown are mean ± SEM. **, p < 0.01. (c) Effect of PV557 on the

production of intracellular Ab in the treated cells. HEK293-APP cells

were treated with different concentrations of PV557 for 9 h, and the

cells were harvested. The amount of intracellular Ab40 and Ab42 was

determined by the sandwich ELISA system as described in Experi-

mental procedures. As a control, the effect of a known inhibitor for

BACE1, b-secretase inhibitor IV, was also investigated. The amount of

Ab produced from the non-treated cells and peptide-treated cells was

compared. The experiment shown is representative of three inde-

pendent experiments in triplicate. Values shown are mean ± SEM. **,

p < 0.01. (d) Effect of PV557 on the processing of APP in the cells.

The membrane fraction was obtained from the HEK293-APP cells

after treatment with PV557 as in B. The membrane proteins were

fractionated by gel electrophoresis and transferred onto PVDF mem-

brane for immunoblotting. FL-APP (full length-APP) and CTFb were

detected with the 6E10 antibody. The experiment shown is repre-

sentative of two independent experiments. (e). Figure D was scanned,

and the density of each band was determined by Multi Gauge pro-

gram. The ratios of CTFb and CTFb-P to FL-APP (mature + immature)

were calculated based on the intensities.



such an approach may be able to avoid the issue of sideeffects of AD drugs.

We used theHC approach for designing peptides that bind tothe b-secretase cleavage site of APP. Bindings as well asinhibitory activities of these peptides were investigated insome detail with a substrate decapeptide derived fromSwedish

mutant APP (Sub M) since the substrate peptide derived fromwild-type APP was a very poor substrate for recombinanthuman BACE1 in vitro. Interestingly, only the HC peptide c-SubM (SEFCIQIHFR), derived from the non-coding strand ofmutant substrate (Sub M) DNA, bound to Sub M appreciablybut not the other HC peptides. c-Sub M, but not Sub M-c, alsoshowed inhibitory activity on the cleavage of Sub M byrhBACE1. Approximately 50% inhibition was obtained at aconcentration of c-SubM that was 100-fold higher than that ofthe substrate. c-Sub M apparently does not bind to rhBACE1.In an ELISA assay, rhBACE1 did not bind to the immobilizedc-Sub M (data not shown).

As most AD patients have wild-type APP, informationabout the activity of HC peptides on wild type substrate (SubW) was important. Therefore, the binding activities of HCpeptides either to Sub M or Sub W were compared.Interestingly, c-Sub M showed higher binding activity toSub W than c-Sub W did and also inhibited cleavage of SubW. The cleavage of Sub W was investigated by adding alarger amount of the enzyme and also by HPLC analysis ofthe cleavage products. Therefore, the HC peptide derivedfrom c-Sub M can also be an inhibitor for wild-type APP.

We found that the minimal core sequence of c-Sub M isCIQIHF and the N-terminal cysteine is important. CIQHIFbinds to Sub W and also inhibits cleavage of Sub W byBACE1. In order to demonstrate that CIQIHF can also

(a)

(b)

Fig. 6 Substrate preference of HC peptide. (a) Inhibition of the

cleavage of Sub W, Sub M, Sub-ST6Gal1, and Sub 28 by PV557.

Each BACE1 substrate (25 lmol/L) was pre-incubated with PV557

(750 lmol/L) for 2 h at 25�C and cleaved with rhBACE1. The amount

of enzyme was adjusted so that about 50% of the substrate was

cleaved. The cleaved peptide fragments were analyzed with reversed-

phase HPLC as described in Experimental procedures. This experi-

ment was repeated twice. Values shown are mean ± range. (b)

Binding of PV557 to various peptides. PV557 (200 lmol/L) was

chemically coupled on wells of a microtiter plate, and bio-Sub W

(50 lmol/L), bio-Sub-ST6Gal1 (50 lmol/L), or bio-Ab42 (50 lmol/L)

was added. The binding of substrate peptides was detected with Str-

HRP as described in Experimental procedures. This experiment was

repeated twice and results are means of duplicate determinations in

one of the two experiments.

Fig. 7 Cytotoxicity assay of PV557 on HEK293-APP cells. Various

concentrations of PV557 were added to HEK293-APP cells. After

incubation for 9 h, 1/10 volume of MTT was added to the cells and the

reduced MTT was determined by measuring absorbance at 570 nm as

described in Experimental procedures. The experiment shown is

representative of two independent experiments in triplicate. Values are

mean ± range.



inhibit cleavage of APP in the cells, the cysteine wasmodified in several ways. It was found that addition ofpolyarginine (9 residues) to the end of an HC peptideallowed the delivery of the peptide into the cells. However,the conjugated peptide did not inhibit processing of APP inthe cells (data not shown). However, addition of6-aminohexanoic acid to the N-terminus of CIQIHF(PV557) allowed the peptide to inhibit processing of APPas well as the production of Ab from the treated cells. Forexample, the IC50 value for inhibition of production ofAb40 was about 15 lmol/L. PV557 was less effective forthe inhibition of Ab42 production. Similar pattern wasobserved with the b-secretase inhibitor IV (Stachel et al.2004): the inhibitor is more effective for inhibition of Ab40production than that of Ab42. When the effect of PV557 onthe level of Ab in the cell was examined, it was found thatPV557 was more effective in preventing the production ofAb40 (IC50 of 3 lmol/L in contrast to IC50 of 15 lmol/Lfor the released Ab40). Again the effect of PV557 on theintra-cellular level of Ab42 was less pronounced. Thedifferential effect of PV557 on the production of Ab40 andAb42 may be related to the possible effect of PV557 on the

environment of membrane structure involving c-secretasecomplexes. It was reported that the choice of cleavage siteseither at Ab40 or Ab42 by c-secretase is very muchsusceptible to membrane environment (Fassbender et al.2001; Kim et al. 2005). Surprisingly, b-secretase inhibitorIV showed weak effect on the intracellular level of Ab40and Ab42 (less than 10% versus 78% inhibition ofproduction of the released Ab40). It is difficult to offer aplausible mechanism for this phenomenon. There has beenno report on the effect of b-secretase inhibitor IV on theintracellular level of Ab. Another interesting differencebetween PV557 and b-secretase inhibitor IV is that whilePV557 reduced the accumulation of both the phosphoryl-ated and non-phosphorylated forms of the C-terminalcleavage product, b-secretase inhibitor IV preferentiallyreduced the non-phosphorylated form of the C-terminalcleavage product. This may reflect possible difference in thesite of action of the two inhibitors in the cells, such as theendoplasmic reticulum/Golgi complex and endosome(Annaert and De Strooper 2000; De Strooper and Annaert2000). There are reports on the importance of intracellularAb for cytotoxicity (Echeverria and Cuello 2002). HCpeptide can be a more potent inhibitor for the production ofintracellular Ab than BACE1 inhibitors.

The HC peptide described in this study also showed apreference for APP for its inhibition of cleavage by BACE1over other BACE1 substrates such as ST6Gal1. The HCpeptide may not inhibit the processing of APP at positionsother than the b-cleavage site. PV557 did not show bindingactivity to Ab42 in vitro. Ab42 contains the cleavage sites fora-secretase and other isoforms of b-secretase.

PV557 shows half maximal inhibitory activity for pro-duction of secreted Ab at 10–20 lmol/L while the value forproduction of intra-cellular Ab is about 3 lmol/L. Theinhibitory activity of PV557 is certainly lower than some ofthe reported inhibitors of BACE1. For example, b-secretaseinhibitor IV (Stachel et al. 2004) shows about 10 fold higherinhibitory activity for production of secreted Ab than PV557.However, PV557 is quite comparable with JMV2764(Lefranc-Jullien et al. 2005) in terms of inhibition ofproduction of secreted Ab40. In this study, we only focusedon the modification of the N-terminal cysteine of CIQIHFpeptide. It is conceivable that highly effective inhibitors canbe derived from PV557 or CIQIHF by systematic modifica-tion of the structure.

In this report, we have provided proof of concept for anovel strategy aimed at development of AD drugs. Devel-opment of the molecules that block the secretase cleavagesites of APP is a viable approach for development of drugsfor AD that will reduce the potential risk of side effects bythe inhibitors of secretases. The approach described in thisreport will provide a new opportunity for the development ofdrugs that can be used to prevent and treat AD with minimalside effects.

Fig. 8 Stability assay of PV557 in serum and during cell culture. For

the study of PV557 stability in serum, PV557 (25 lg) was incubated

with rat serum at 37�C for the indicated time, and the incubated

sample was fractionated by C18 reversed phase HPLC and the

amount of the remaining PV557 was estimated as described in

Experimental procedures. For the study of metabolic stability of PV557

during cell culture, PV557 (25 lmol/L) in serum free DMEM was ad-

ded to HEK293-APP cell for the indicated time, and culture media and

cells were harvested together. The harvested sample was fractionated

by C18 reversed phase HPLC after sonication, and the amount of the

remaining PV557 was estimated as described in Experimental pro-

cedures. The experiment was repeated twice. Values are mean ±

range.



Acknowledgments

This work was supported by grants from Posco and Konkuk

University. We thank Dr T.W. Kim at Columbia University for the

HEK293-APP cells and advice on assays for APP cleavage. We also

thank Mi Rang Nam for technical assistance.

References

Annaert W. and De Strooper B. (2000) Neuronal models to studyamyloid precursor protein expression and processing in vitro.Biochim. Biophys. Acta 1502, 53–62.

Arbel M., Yacoby I. and Solomon B. (2005) Inhibition of amyloidprecursor protein processing by beta-secretase through site-directedantibodies. Proc. Natl Acad. Sci. USA 102, 7718–7723.

Blalock J. E. and Smith E. M. (1984) Hydropathic anti-complementarityof amino acids based on the genetic code. Biochem. Biophys. Res.Commun. 121, 203–207.

Bost K. L., Smith E. M. and Blalock J. E. (1985) Similarity between thecorticotropin (ACTH) receptor and a peptide encoded by an RNAthat is complementary to ACTH mRNA. Proc. Natl Acad. Sci. USA82, 1372–1375.

Brentani R. R., Ribeiro S. F., Potocnjak P., Pasqualini R., Lopes J. D.and Nakaie C. R. (1988) Characterization of the cellular receptorfor fibronectin through a hydropathic complementarity approach.Proc. Natl Acad. Sci. USA 85, 364–367.

Cumming J. N., Iserloh U. and Kennedy M. E. (2004) Design anddevelopment of BACE-1 inhibitors. Curr. Opin. Drug Discov. Dev.7, 536–556.

De Strooper B. and Annaert W. (2000) Proteolytic processing and cellbiological functions of the amyloid precursor protein. J. Cell Sci.113 (Pt 11), 1857–1870.

Dovey H. F., John V., Anderson J. P. et al. (2001) Functional gamma-secretase inhibitors reduce beta-amyloid peptide levels in brain.J. Neurochem. 76, 173–181.

Echeverria V. and Cuello A. C. (2002) Intracellular A-beta amyloid, asign for worse things to come?. Mol. Neurobiol. 26, 299–316.

Fassbender K., Simons M., Bergmann C. et al. (2001) Simvastatinstrongly reduces levels of Alzheimer’s disease beta-amyloid pep-tides Abeta 42 and Abeta 40 in vitro and in vivo. Proc. Natl Acad.Sci. USA 98, 5856–5861.

Fassina G., Roller P. P., Olson A. D., Thorgeirsson S. S. and OmichinskiJ. G. (1989) Recognition properties of peptides hydropathicallycomplementary to residues 356–375 of the c-raf protein. J. Biol.Chem. 264, 11 252–11 257.

Gho Y. S. and Chae C. B. (1997) Anti-angiogenin activity of the peptidescomplementary to the receptor-binding site of angiogenin. J. Biol.Chem. 272, 24 294–24 299.

Gruninger-Leitch F., Schlatter D., Kung E., Nelbock P. and Dobeli H.(2002) Substrate and inhibitor profile of BACE (beta-secretase)and comparison with other mammalian aspartic proteases. J. Biol.Chem. 277, 4687–4693.

Harrison T., Churcher I. and Beher D. (2004) gamma-Secretase as atarget for drug intervention in Alzheimer’s disease. Curr. Opin.Drug Discov. Dev. 7, 709–719.

Heal J. R., Bino S., Ray K. P., Christie G., Miller A. D. and Raynes J. G.(1999) A search within the IL-1 type I receptor reveals a peptidewith hydropathic complementarity to the IL-1beta trigger loopwhich binds to IL-1 and inhibits in vitro responses. Mol. Immunol.36, 1141–1148.

Heal J. R., Roberts G.W., Raynes J. G., BhakooA. andMiller A. D. (2002)Specific interactions between sense and complementary peptides: thebasis for the proteomic code. Chembiochem 3, 136–151.

Ida N., Hartmann T., Pantel J., Schroder J., Zerfass R., Forstl H.,Sandbrink R., Masters C. L. and Beyreuther K. (1996) Analysis ofheterogeneous A4 peptides in human cerebrospinal fluid and bloodby a newly developed sensitive Western blot assay. J. Biol. Chem.271, 22 908–22 914.

Ikeuchi T. and Sisodia S. S. (2003) The Notch ligands, Delta1 andJagged2, are substrates for presenilin-dependent ‘‘gamma-secret-ase’’ cleavage. J. Biol. Chem. 278, 7751–7754.

John V., Beck J. P., Bienkowski M. J., Sinha S. and Heinrikson R. L.(2003) Human beta-secretase (BACE) and BACE inhibitors.J. Med. Chem. 46, 4625–4630.

Kim S., Jeon T. J., Oberai A., Yang D., Schmidt J. J. and Bowie J. U.(2005) Transmembrane glycine zippers: physiological and patho-logical roles in membrane proteins. Proc. Natl Acad. Sci. USA 102,14 278–14 283.

Kitazume S., Tachida Y., Oka R., Shirotani K., Saido T. C. and Hashi-moto Y. (2001) Alzheimer’s beta-secretase, beta-site amyloid pre-cursor protein-cleaving enzyme, is responsible for cleavagesecretion of a Golgi-resident sialyltransferase. Proc. Natl Acad. Sci.USA 98, 13 554–13 559.

Kitazume S., Tachida Y., Oka R., Kotani N., Ogawa K., Suzuki M.,Dohmae N., Takio K., Saido T. C. and Hashimoto Y. (2003)Characterization of alpha 2,6-sialyltransferase cleavage byAlzheimer’s beta -secretase (BACE1). J. Biol. Chem. 278, 14 865–14 871.

Knutson V. P. (1988) Insulin-binding peptide. Design and characteriza-tion. J. Biol. Chem. 263, 14 146–14 151.

Lannfelt L., Viitanen M., Johansson K., Axelman K., Lilius L., AlmqvistE. and Winblad B. (1993) Low frequency of the APP 670/671mutation in familial Alzheimer’s disease in Sweden. Neurosci. Lett.153, 85–87.

Lefranc-Jullien S., Lisowski V., Hernandez J. F., Martinez J. and CheclerF. (2005) Design and characterization of a new cell-permeantinhibitor of the beta-secretase BACE1. Br. J. Pharmacol. 145,228–235.

Li Q. and Sudhof T. C. (2004) Cleavage of amyloid-beta precursorprotein and amyloid-beta precursor-like protein by BACE 1.J. Biol. Chem. 279, 10 542–10 550.

Lichtenthaler S. F., Dominguez D. I., Westmeyer G. G., Reiss K., HaassC., Saftig P., De Strooper B. and Seed B. (2003) The cell adhesionprotein P-selectin glycoprotein ligand-1 is a substrate for theaspartyl protease BACE1. J. Biol. Chem. 278, 48 713–48 719.

Mattson M. P. (2004) Pathways towards and away from Alzheimer’sdisease. Nature 430, 631–639.

Olson R. E., Copeland R. A. and Seiffert D. (2001) Progress towardstesting the amyloid hypothesis: inhibitors of APP processing. Curr.Opin. Drug Discov. Dev. 4, 390–401.

Periz G. and Fortini M. E. (2004) Functional reconstitution of gamma-secretase through coordinated expression of presenilin, nicastrin,Aph-1, and Pen-2. J. Neurosci. Res. 77, 309–322.

Ruiz-Opazo N., Kosik K. S., Lopez L. V., Bagamasbad P., Ponce L. R.and Herrera V. L. (2004) Attenuated hippocampus-dependentlearning and memory decline in transgenic TgAPPswe Fischer-344rats. Mol. Med. 10, 36–44.

Scherrmann J. M. (2002) Drug delivery to brain via the blood-brainbarrier. Vascul. Pharmacol. 38, 349–354.

Selkoe D. J. (1994) Alzheimer’s disease: a central role for amyloid.J. Neuropathol. Exp. Neurol. 53, 438–447.

Shai Y., Flashner M. and Chaiken I. M. (1987) Anti-sense peptiderecognition of sense peptides: direct quantitative characterizationwith the ribonuclease S-peptide system using analytical high-performance affinity chromatography. Biochemistry 26, 669–675.

Shearman M. S., Hawtin S. R. and Tailor V. J. (1995) The intracellularcomponent of cellular 3-(4,5-dimethylthiazol-2-yl)-2, 5-diphenyl-



tetrazolium bromide (MTT) reduction is specifically inhibited bybeta-amyloid peptides. J. Neurochem. 65, 218–227.

Shuto D., Kasai S., Kimura T. et al. (2003) KMI-008, a novel beta-secretase inhibitor containing a hydroxymethylcarbonyl isostere asa transition-state mimic: design and synthesis of substrate-basedoctapeptides. Bioorg. Med. Chem. Lett. 13, 4273–4276.

Sinha S. and Lieberburg I. (1999) Cellular mechanisms of beta-amyloidproduction and secretion. Proc. Natl Acad. Sci. USA 96, 11 049–11 053.

Sinha S., Anderson J. P., Barbour R. et al. (1999) Purification and clo-ning of amyloid precursor protein beta-secretase from human brain.Nature 402, 537–540.

Smith P. K., Krohn R. I., Hermanson G. T., Mallia A. K., Gartner F. H.,Provenzano M. D., Fujimoto E. K., Goeke N. M., Olson B. J. andKlenk D. C. (1985) Measurement of protein using bicinchoninicacid. Anal. Biochem. 150, 76–85.

Sommer B. (2002) Alzheimer’s disease and the amyloid cascade hypo-thesis: ten years on. Curr. Opin. Pharmacol. 2, 87–92.

Stachel S. J., Coburn C. A., Steele T. G. et al. (2004) Structure-baseddesign of potent and selective cell-permeable inhibitors of humanbeta-secretase (BACE-1). J. Med. Chem. 47, 6447–6450.

Tomasselli A. G., Qahwash I., Emmons T. L. et al. (2003) Employing asuperior BACE1 cleavage sequence to probe cellular APP pro-cessing. J. Neurochem. 84, 1006–1017.

Vassar R. (2002) Beta-secretase (BACE) as a drug target for Alzheimer’sdisease. Adv. Drug Deliv. Rev. 54, 1589–1602.

Vickers J. C., Dickson T. C., Adlard P. A., Saunders H. L., King C. E.and McCormack G. (2000) The cause of neuronal degeneration inAlzheimer’s disease. Prog. Neurobiol. 60, 139–165.

Walter J., Kaether C., Steiner H. and Haass C. (2001) The cell biology ofAlzheimer’s disease: uncovering the secrets of secretases. Curr.Opin. Neurobiol. 11, 585–590.

Willem M., Garratt A. N., Novak B., Citron M., Kaufmann S., RittgerA., DeStrooper B., Saftig P., Birchmeier C. and Haass C. (2006)Control of peripheral nerve myelination by the beta-secretaseBACE1. Science 314, 664–666.

Wong H. K., Sakurai T., Oyama F., Kaneko K., Wada K., Miyazaki H.,Kurosawa M., De Strooper B., Saftig P. and Nukina N. (2005) betaSubunits of voltage-gated sodium channels are novel substrates ofbeta-site amyloid precursor protein-cleaving enzyme (BACE1) andgamma-secretase. J. Biol. Chem. 280, 23 009–23 017.

Yoo S. A., Bae D. G., Ryoo J. W., Kim H. R., Park G. S., Cho C. S.,Chae C. B. and Kim W. U. (2005) Arginine-rich anti-vascularendothelial growth factor (anti-VEGF) hexapeptide inhibits col-lagen-induced arthritis and VEGF-stimulated productions of TNF-alpha and IL-6 by human monocytes. J. Immunol. 174, 5846–5855.



Inhibition of amyloid ?-peptide production by blockage of ?-secretase cleavage site of amyloid precursor protein

Documents