Cleavage agents for soluble oligomers of amyloid β peptides

Article abstract of DOI:10.1002/anie.200702399

(Chemical Equation Presented) Clearing the mind: Peptide-cleaving agents for amyloid β-42 (Aβ₄₂) oligomers (see scheme), the intermediate neurotoxic species in the pathology of Alzheimer's disease, have been obtained from a combinatorial librar

Full text of DOI:10.1002/anie.200702399

Communications
DOI: 10.1002/anie.200702399
Peptide Cleavage
Cleavage Agents for Soluble Oligomers of Amyloid b Peptides**
Junghun Suh,* Sang Ho Yoo, Min Gyum Kim, Keunhong Jeong, Jae Young Ahn, Myoung-
soon Kim, Pil Seok Chae, Tae Yeon Lee, Jaehwa Lee, Jeongkuk Lee, Yun Ah Jang, and
Eun Hwa Ko
Alzheimerꢀs disease (AD) is characterized by neuronal loss
and the presence of senile plaques containing amyloid b (Ab)
peptides in the brain.[1] Ab peptides are primarily composed
of amyloid b-40 (Ab40) and amyloid b-42 (Ab42) peptides,
which contain 40 and 42 amino acid residues, resepctively.[2]
Even when the concentration of Ab42 peptides is much lower
than that of Ab40, Ab42 peptides are the predominant
components in the plaques.[3] An increase in the Ab42/Ab40
ratio is associated with familial forms of early onset AD.[4]
The association process of Ab40 or Ab42 peptides involves
the formation of several oligomers, protofibrils, and fibrils as
summarized in Scheme 1.[5,6] Here, the species placed in the
forms oligomers and fibrils considerably faster than does the
Ab40 peptide.[8,9] A more recent study indicated that the Ab42
peptide exists in solution as a binary mixture of the monomer
and high-molecular-mass oligomers, whereas a monomer–
dimer equilibrium exists for the Ab40 peptide.[10]
Soluble oligomers of the Ab42 peptide, instead of the
monomer or insoluble amyloid fibrils, are proposed to be
responsible for synaptic dysfunction in the brains of patients
with AD.[11] Several lines of evidence have been reported in
support of the role of soluble oligomers, such as the
dodecamer, of the Ab42 peptide as the intermediate neuro-
toxic species in the pathology of AD.[5,12–14]
As a therapeutic option to alleviate Ab42-induced neuro-
toxicity, various attempts have been made to lower the level
of the Ab42 assembly in the brain.[15–19] One way to do this is to
inhibit the action of either b-secretase or g-secretase which
produces Ab peptides from b-amyloid precursor proteins.[11]
One could attempt to prevent oligomerization of the Ab42
peptide by the use of Ab42 immunization.[15,16] Small mole-
cules with high affinity toward the Ab42 peptide may suppress
its oligomerization.[17] Enhancement of the activity of pro-
teases known to degrade the Ab42 peptide is an alternative
way to suppress its oligomerization in the brain.[18]
A novel method discovered in this study to reduce the
level of Ab42 oligomers is to cleave the Ab42 peptide included
in an oligomer: As shown in Scheme 1, where the cleavage
agent is indicated as (R)-(LCoIII), cleavage of the Ab42
peptide included in a target oligomer reduces the concen-
tration of the target oligomer, thereby leading to a decrease in
the concentrations of other oligomers which are readily
transformed into the target oligomer. Reduction of the
concentration of the oligomers slows down the formation of
protofibrils and fibrils. Herein we present the design, syn-
thesis, and activity of four agents that effectively cleave Ab42
oligomers.
Scheme 1. Formation of various assemblies of the Ab40 or Ab42
peptides and cleavage of the assemblies.
rectangle are the soluble oligomers that possess unique[7]
structures. The conversion of large assemblies such as
protofibrils and fibrils into smaller ones is slow, and the
formation of the large assemblies is irreversible or partially
irreversible.[6] Immediately after peptide production, the Ab40
peptide produces monomers, dimers, trimers, and tetramers in
rapid equilibrium, whereas the Ab42 peptide preferentially
forms pentamers or hexamers.[6] In addition, the Ab42 peptide
We discovered that target-selective artificial proteases can
be designed for proteins using myoglobin or peptide defor-
mylase as the target proteins.[20–23] As the catalytic center for
the target-selective artificial proteases, the CoIII complex of
cyclen ([CoIIIcyclen]) is one of the most preferred in view of
its ability to hydrolyze[21–25] peptide bonds and the inertness of
CoIII complexes to exchange.
[*] Prof. Dr. J. Suh, S. H. Yoo, M. G. Kim, K. Jeong, J. Y. Ahn, M.-s. Kim,
P. S. Chae, T. Y. Lee, J. Lee, J. Lee, Y. A. Jang, E. H. Ko
Department of Chemistry
Seoul National University
Seoul 151-747 (Korea)
Fax: (+82)2-874-3704
E-mail: jhsuh@snu.ac.kr
The combinatorial library of candidates of agents for the
cleavage of soluble oligomers of the Ab42 peptide was
constructed by exploiting [CoIIIcyclen] as the cleavage
center in this study. Several aromatic moieties are known to
possess affinity for b-amyloid plaques when appropriately
derivatized, and have been used as essential parts of imaging
agents for b-amyloid plaques.[26] To facilitate the selective
Homepage: http://chem3.snu.ac.kr/~jhsuh/
[**] This work was supportedby the Korea Science andEngineering
Foundation through the National Research Laboratory Program
(no. M10500000001-06J0000-00110) funded by the Ministry of
Science andTechnology.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
7064
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 7064 –7067

Angewandte
Chemie
recognition of soluble oligomers of the Ab42 peptide, those
aromatic moieties were employed as auxiliary binding
components in the combinatorial library. A chemical library
containing 888 compounds was constructed (see the Support-
ing Information).
The activity of the candidates as cleavage agents was
checked by MALDI-TOF MS after incubating each library
member with 4.0 mm of the Ab42 peptide at pH 7.50 (0.050m
phosphate) and 378C for 24h. The activity for cleavage of the
Ab42 assembly was judged by the appearance of peptide
fragments. After repeated screening, A–D were selected and
NaOH before introducing it into pH 7.50 medium.[28] The
results (see the Supporting Information) of filtration through
a membrane with a cut-off molecular weight (Mw) of 10000
indicate that most (> 80%) of the Ab42 peptide (Mw 4514)
passes through the filter immediately after transfer from a
1 mm NaOH environment to a pH 7.50 medium. Since the
amount of dimer or trimer is not significant compared to that
of the monomer,[6,10] the amount of Ab42 peptide that passes
through the filter represents the amount of monomer. During
the filtration step, which takes about 10 minutes, small
amounts of large oligomers may form, which accounts for
the incomplete passage through the filter. Thus, the Ab42
peptide appears to exist mostly as the monomer immediately
after exposure to the reaction buffer. Within 3 hours, about
two-thirds of the initially added Ab42 peptide forms aggre-
gates that cannot pass through the filter. About 10% of the
initially added Ab42 peptide remains as the monomer even
after 36 hours, in agreement with literature results.[10] On the
other hand, more than 90% and about 50% of the initially
added Ab40 peptide (Mw 4330) passes through the membrane
immediately and after 24h, respectively, after addition of the
stock solution to the pH 7.50 buffer. The monomer, the dimer,
and, to some extent, the trimer of the Ab40 peptide are
expected to pass through the membrane. The results agree
with the previous observations that the Ab40 peptide produces
small oligomers in equilibrium with the monomer and that the
Ab40 peptide is converted into protofibrils and fibrils consid-
erably more slowly than is the Ab42 peptide.[8–10]
MALDI-TOF mass spectra obtained after incubation of
the Ab40 or Ab42 peptides (4.0 mm) with A–D (see the
Supporting Information) showed the presence of Ab1–20 and
Ab1–21 peptides as some of the cleavage products. Here, the
Ab fragments are named according to the amino acid
sequence of the Ab42 peptide. The structures of the cleavage
products were confirmed by MALDI LIFT-TOF/TOF MS.[29]
Since the intensity of a MALDI-TOF MS signal does not
indicate its relative concentration, some oligopeptide frag-
ments may be present in significant concentrations without
showing strong MALDI-TOF MS signals.
Unless noted otherwise, the cleavage reaction was ini-
tiated by adding the stock solution of the Ab40 or Ab42
peptides to the buffer solution containing one of the cleaving
agents A–D. The product solution obtained by the reaction of
the Ab40 or Ab42 peptides (4.0 mm) with A–D was filtered
through a membrane with a cut-off molecular weight of 10000
to remove aggregates of the two peptides, and the resulting
solution was subjected to HPLC separation to isolate the
oligopeptide fragments. The oligopeptide fragments were
hydrolyzed under alkaline conditions, and the resulting amino
acids were quantified with fluorescamine (see the Supporting
Information). The amounts of oligopeptide fragments
obtained by the action of A–D were estimated as mol% of
the initially added amount of the Ab40 or Ab42 peptides and
are defined as the cleavage yields. The cleavage yields
measured after reaction with various initial concentrations
(Co) of A–D after 36 h at 378C and pH 7.50 are plotted
against logCo MÀ1 in Figure 1. Plateau values of the cleavage
yields are 10–30%, and significant yields were observed at a
Co value of 100 nm. As analyzed in the Supporting Informa-
synthesized on larger scales as described in the Supporting
Information. Whether A–D could cleave the Ab40 peptide was
also examined by incubating A–D with 4.0 mm of the Ab40
peptide at pH 7.50 and 378C: A and B were found to effect
cleavage, whereas C and D were inactive.[27]
The aggregation behavior of the Ab peptides is known to
be sensitive to the experimental conditions.[10] We have,
therefore, carried out the following filtration experiments
under the experimental conditions to obtain quantitative
information on the rate of decrease in the amount of
monomer and/or small oligomers (see the Supporting Infor-
mation).
To ensure generation of the monomeric form of the Ab40
or Ab42 peptides, the synthetic Ab peptide was treated with
Angew. Chem. Int. Ed. 2007, 46, 7064 –7067
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 7065

Communications
pH 7.50. The results (Supporting Information) reveal that the
yield for cleavage of the peptides by A does not increase
appreciably upon reaction for more than 24h; this result is
likely due to consumption of the target oligomer through both
the cleavage by A and formation of protofibrils and fibrils.
At present, the identity of the oligomers of the Ab40 or
Ab42 peptides cleaved by A–D is unknown. Nevertheless,
reduction of the concentration of the target oligomer would
decrease the concentrations of other oligomers, since the
oligomers are in equilibria with one another.
Agents A and B cleaved both the Ab42 and Ab40 peptides,
even though they were selected from the combinatorial
library by screening against the Ab42 peptide. The b-amyloid
precursor protein is cleaved by secretases mainly to produce
the Ab40 peptide throughout life, and this peptide carries out
physiological functions.[30] Excessive cleavage of the Ab40
peptide may, therefore, interfere with its normal functions.
In the brains of patients with AD, however, the level of
soluble Ab40 peptide is 30–40 times higher than those of
nondemented elderly controls.[31] Partial cleavage of soluble
oligomers of Ab40 during cleavage of the Ab42 peptide may
not cause considerable side effects for patients with AD.
Many more cleavage agents for the Ab42 oligomers can be
synthesized by combining a CoIII–ligand complex and a
binding auxiliary with affinity for the Ab42 oligomers. After
performing proper in vivo tests, some of the synthetic
cleavage agents may be found suitable for therapeutic
treatment of patients with AD.
Figure 1. Plot of the cleavage yield(CY, mol%) against log Co/m for the
*
*
cleavage of Ab40 ( ) or Ab42 ( ) (4.0 mm) by A (a), B (b), C (c), and D
(d) measured after reaction for 36 h at 378C andpH 7.50. In the
present study, the estimated values of the cleavage yields are the
average of results of 4 or 6 measurements carriedout with different
reaction mixtures, respectively, and their relative standard deviations
are 5–15%.
tion, the efficiency for cleavage of the Ab42 peptide by A–D is
expected to be much higher in patients with AD.
To examine the effect of aggregation of the Ab40 or Ab42
peptides prior to exposure to the cleavage agent, either the
Ab40 or Ab42 peptide was incubated in the buffer solution for
various periods of time before treating the solution with A
(see the Supporting Information). Little cleavage was
observed when A was added to the reaction mixture after
preincubation of the Ab42 peptide for 24h, probably because
of extensive polymerization of the peptide leading to the
formation of protofibrils or fibrils. When A was added to the
reaction mixture after preincubation of the Ab42 peptide for
3–6 h, the amounts of products formed by cleavage of the
peptide were not much smaller than that obtained with A
added without preincubation of the peptide. This result stands
in contrast with the considerable reduction in the amount of
Ab42 monomer during the initial 3–6 hours. Thus, the frag-
ments are formed mainly by cleavage of oligomers instead of
monomer, protofibrils, or fibrils.
The addition of A after preincubation of the Ab40 peptide
for 24h leads to considerable cleavage, while preincubation
for longer periods reduces the cleavage yield. This observa-
tion is consistent with the slower formation of protofibrils and
fibrils by the Ab40 peptide compared with that from the Ab42
peptide. In addition, it reveals that the protofibrils or fibrils of
the Ab40 peptide are not the main source of the fragments.
Since the yield for cleavage of the Ab40 peptide by A does not
decrease considerably by preincubation for 3–18 h, the
monomer of Ab40 is not the main source of the fragments,
in view of the results of the filtration experiment.
Received: June 2, 2007
Published online: August 20, 2007
Keywords: Alzheimer’s disease · amyloid b peptides ·
.
combinatorial chemistry · drug design · peptide cleavage
[1] D. J. Selkoe, Physiol. Rev. 2001, 81, 741 – 766.
[2] K. Sambamurti, N. H. Greig, D. K. Lahiri, J. Neuromol. Med.
2002, 1, 1 – 31.
[3] S. A. Gravina, L. B. Ho, C. B. Eckman, K. E. Long, L. Otvos, Jr.,
L. H. Younkin, N. Suzuki, S. G. Younkin, J. Biol. Chem. 1995,
270, 7013 – 7016.
[4] D. Scheuner et al., Nat. Med. 1996, 2, 864– 870.
[5] S. LesnØ, M. T. Koh, L. Kotilinek, R. Kayed, C. G. Glabe, A.
Yang, M. Gallagher, K. H. Ashe, Nature 2006, 440, 352 – 357.
[6] G. Bitan, M. D. Kirkitadze, A. Lomakin, S. S. Vollers, G. B.
Benedek, D. B. Teplow, Proc. Natl. Acad. Sci. USA 2003, 100,
330 – 335.
[7] B. Urbanic, L. Cruz, S. Yun, S. V. Buldyrev, G. Bitan, D. B.
Teplow, H. E. Stanley, Proc. Natl. Acad. Sci. USA 2004, 101,
17345 – 17350.
[8] R. Kayed, E. Head, J. L. Thompson, T. M. McIntire, S. C. Milton,
C. W. Cotman, C. G. Glabe, Science 2003, 300, 486 – 489.
[9] J. T. Jarrett, E. P. Berger, P. T. Lansbury, Jr., Biochemistry 1993,
32, 4693 – 4697.
[10] R. W. Hepler et al., Biochemistry 2006, 45, 15157 – 15167.
[11] J. Hardy, J. Selkoe, Science 2002, 297, 353 – 356.
[12] R. E. Tanzi, Nat. Neurosci. 2005, 8, 977 – 979.
[13] E. M. Snyder et al., Nat. Neurosci. 2005, 8, 1051 – 1058.
[14] S. Barghorn et al., Neurochemistry 2005, 95, 834– 847.
[15] D. Schenk et al., Nature 1999, 400, 173 – 177.
To examine the progress of the cleavage reaction, the
cleavage yield was measured by treating the Ab40 or Ab42
peptides with A for various periods of time at 378C and
7066 www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 7064 –7067

Angewandte
Chemie
[16] R. B. DeMattos, K. R. Bales, D. J. Cummins, J. C. Dodart, S. M.
Paul, D. M. Holtzman, Proc. Natl. Acad. Sci. USA 2001, 98,
8850 – 8855.
[17] T. Cohen, A. Frydman-Marom, M. Rechter, E. Gazit, Biochem-
istry 2006, 45, 4727 – 4735.
[18] D. S. Choi, D. Wang, G. Q. Yu, G. Zhu, V. N. Kharazia, J. P.
Paredes, W. S. Chang, J. K. Deitchman, L. Mucke, R. O. Messing,
Proc. Natl. Acad. Sci. USA 2006, 103, 8215 – 8220.
[19] C. M. Dobson, Science 2004, 304, 1259 – 1262.
[20] J. Suh, Acc. Chem. Res. 2003, 36, 562 – 570.
[21] J. W. Jeon, S. J. Son, C. E. Yoo, I. S. Hong, J. B. Song, J. Suh, Org.
Lett. 2002, 4, 4155 – 4158.
[22] J. W. Jeon, S. J. Son, C. E. Yoo, I. S. Hong, J. Suh, Bioorg. Med.
Chem. 2003, 11, 2901 – 2910.
[23] P. S. Chae, M. Kim, C. Jeung, S. D. Lee, H. Park, S.; Lee, J. Suh, J.
Am. Chem. Soc. 2005, 127, 2396 – 2397.
[24] D. A. Buckingham, A. M. Sargeson, F. R. Keene, J. Am. Chem.
Soc. 1974, 96, 4981 – 4983.
[25] M. K. Takasaki, J. H. Kim, E. Rubin, J. Chin, J. Am. Chem. Soc.
1993, 115, 1157 – 1159.
[26] H. I. Sair, P. M. Doraiswamy, J. R. Petrella, Neuroradiol. 2004,
46, 93 – 104.
[27] To confirm the selectivity of A–D for the Ab42 and Ab40 peptides,
several control experiments were performed, as described in the
Supporting Information.
[28] Y. Fezoui, D. M. Hartley, J. D. Harper, R. Khurana, D. M. Walsh,
M. M. Condron, D. J. Selkoe, P. T. Lansbury, Jr., A. L. Fink,
D. B. Teplow, Amyloid 2000, 7,166 – 178.
[29] D. Suckau, A. Resemann, M. Schuerenberg, P. Hufnagel, J.
Franzen, A. Holle, Anal. Bioanal. Chem. 2003, 376, 952 – 965.
[30] L. E. Plant, N. J. Webster, J. P. Boyle, M. Ramsden, D. B. Freir,
C. Peers, H. A. Pearson, Neurobiol. Aging 2006, 27, 1673 – 1683.
[31] L. F. Lue, Y. M. Kuo, A. E. Roher, L. Brachova, Y. Shen, L. Sue,
T. Beach, J. H. Kurth, R. E. Rydel, J. Rogers, Am. J. Pathol. 1999,
155, 853 – 862.
Angew. Chem. Int. Ed. 2007, 46, 7064 –7067
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 7067