Small-molecule inhibitors of islet amyloid polypeptide fibril formation

Article abstract of DOI:10.1002/anie.200705372

(Chemical Equation Presented) Small and effective: The pathological aggregation of amylin (IAPP), which leads to type II diabetes mellitus, is effectively inhibited by small-molecule rhodanine-based inhibitors at nanomolar concentrations. The prevention o

Full text of DOI:10.1002/anie.200705372

Angewandte
Chemie
DOI: 10.1002/anie.200705372
Small-Molecule Inhibitors
Small-Molecule Inhibitors of Islet Amyloid Polypeptide Fibril
Formation**
Rajesh Mishra, Bruno Bulic, Daniel Sellin, Suman Jha, Herbert Waldmann, and Roland Winter*
Newly synthesized proteins in the cell adopt a functional
folded state resulting from a highly regulated process. Failure
to form this functional state leads to the degradation of
proteins inside the cell. However, under certain conditions,
some proteins can also adopt an alternative state by the
assembly of unfolded or partially folded monomers or protein
fragments into a b-sheet structure called amyloid fibril.[1] In
spite of arising from diverse amino acid sequences, they have
a similar fibrillar structure that binds the dye Congo red.[2]
These amyloids are involved in a number of devastating
diseases including Alzheimerꢀs disease, prion diseases, and
type-II diabetes mellitus.[3]
In the type-II diabetes, deposition of extracellular amyloid
plaques in pancreatic beta cells has been observed in humans.
Biochemical analysis of the plaques revealed the presence of
a 37-residue peptide called islet amyloid polypeptide (IAPP)
or amylin, which is co-secreted with insulin. It has a disulfide
bond between residues 2–7 and the C-terminus is ami-
dated.[4–6] IAPP is also known to interact with lipid mem-
branes which are able to induce and foster the fibril
formation.[7] Recently, a possible mechanism of IAPP fibril
formation at anionic lipid interfaces has been proposed in
which it has been shown that IAPP forms b-sheet-rich
amyloid fibrils via an intermediate a-helical state.[8] The
presence of IAPP amyloid finally leads to the apoptosis of
pancreatic beta cells.[9] However, it is still not clear whether
the fibrils themselves or their intermediate states are respon-
sible for the cell death.[10] In nature, IAPP amyloid fibril
formation can be prevented by altering the primary amino
acid sequence, such as in rat IAPP where three proline
residues, which are absent in the human IAPP, are thought to
prevent the amyloid fibril formation.[11] Recently, Kapurnio-
tuꢀs group succeeded in the synthesis of conformationally
constrained analogues of IAPP, which are methylated at
amide bonds and do not fibrillize.[12,13]
Inhibition of amyloid fibril formation is considered to be a
potentially key therapeutic approach towards diabetes and
other amyloid-related diseases.[14] Surprisingly, very little
attempt has been made to inhibit IAPP fibril formation by
small-molecule inhibitors.[15] Small-molecule inhibitors have
advantages over peptide inhibitors because they could more
easily cross the blood brain barrier, avoid immunological
response, and are more stable in biological fluids and
tissues.[16] In addition, the high flexibility of peptide inhibitors
may, for entropic reasons, prevent efficient binding. This
problem may be overcome by synthesis of conformationally
restricted peptides.[12,13] The bottleneck in the discovery of
small-molecule inhibitors of amyloid fibril formation is the
lack of structural information about amyloids. However, this
did not prevent the discovery of small-molecule inhibitors for
other amyloid fibrils such as, Ab[17,18] and tau,[19,20] which are
involved in Alzheimerꢀs disease.
In a recent study on a cellular model of tau aggregate
inhibition, two rhodanine-scaffold (2-thioxothiazolidin-4-
one) based inhibitors have been identified which have very
low cell toxicity.[21] These compounds were chosen because of
the presence of a rhodanine heterocyclic core, which is
biocompatible, non-mutagenic,[22] and has a drug-like profile.
Inspired by these results, we wanted to explore whether these
compounds will also inhibit amyloid fibril formation of IAPP
which has a completely different amino acid sequence but
shares a similar fibrillar morphology with the tau aggregate.
To our knowledge, this is the first study on such small-
molecule inhibitors of IAPP amyloid formation.
The compounds 1 and 2 (Figure 1) were synthesized as
described earlier.[21] Amyloid fibril formation was carried out
in 10 mm sodium phosphate buffer at pH 7.5 for 96 h. To
reveal the effect of the two potential inhibitors, different
concentrations of the compounds were added to the buffer
solution. Fibril formation was quantified by measuring the
fluorescence intensity of the amyloid-specific dye thioflavin T
(ThT) at a wavelength of 480 nm. The fluorescence intensity
of amyloid fibrils increases upon binding to ThT. The
efficiency of inhibition was monitored by measuring the
ThT fluorescence intensity with respect to that of pure IAPP
aggregate without inhibitor (100%). It is evident from
Figure 1b that both compounds have a marked inhibitory
effect. The concentration at which half of the fibril formation
is inhibited (IC50) is 1.23 mm for compound 1, and 0.45 mm for
compound 2. A similar trend has been observed for the
aggregation of tau, with IC50 values of 0.67 and 0.26 mm for
compounds 1 and 2, respectively.[21] From the results on tau
and IAPP aggregation it is clear that compound 2 is a more
[*] Dr. R. Mishra, D. Sellin, S. Jha, Prof. Dr. R. Winter
Faculty of Chemistry
Physical Chemistry I—Biophysical Chemistry
Technical University Dortmund
Otto-Hahn-Strasse 6, 44227 Dortmund (Germany)
Fax. (+49)231-755-3901
E-mail: roland.winter@uni-dortmund.de
Dr. B. Bulic, Prof. Dr. H. Waldmann
Max-Planck-Institute for Molecular Physiology
Department of Chemical Biology, and
Center for Applied Chemical Genomics
Otto-Hahn-Strasse 11, 44227 Dortmund (Germany)
[**] Financial support from the DFG, the Fonds der Chemischen
Industrie, the country NRWand the EU (Europäischer Fonds für
regionale Entwicklung) is gratefully acknowledged.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2008, 47, 4679 –4682
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4679

Communications
Figure 2. AFM images of the amyloid fibrils of a) pure IAPP and of
IAPP in the presence of a molar ratio 1:1 of b) compound 1 and
c) compound 2 (width of image: 4 mm; height scale on the right-hand
side). The solution was kept for 96 h at room temperature (258C)
before samples were taken, diluted, and measured. The IAPP concen-
tration was 50 mm.
Figure 1. a) Structures of compounds 1 and 2. b) Effect of different
~
*
concentrations of compound 1 ( ) and 2 ( ) on IAPP amyloid fibril
formation with a constant IAPP concentration of 10 mm. After 96 h,
samples were incubated for 1 h with 10 mm of ThT at room temper-
ature. The intensity of the ThT fluorescence at 480 nm is plotted
against a control IAPP sample (100%) with no inhibitor.
was incubated with compounds 1 and 2 in a 1:1 molar ratio.
No fibril formation and appearance of a few oligomeric
species only, was observed in the case of compound 1
(Figure 2b). For compound 2, abundance of small oligomeric
structures is visible (Figure 2c). Hence, the AFM data clearly
support the results obtained from the ThT fluorescence assay.
Morphological differences induced by the inhibition with
compounds 1 and 2 indicate a differential mechanism of
inhibition of IAPP fibril formation. What is clear from
Figure 2c is that inactive monomers and/or oligomers are
present as inhibition product, only, and compound 2 seems to
preferentially interact with the rate-limiting intermediate
oligomeric species.
Negatively charged lipid membranes are known to accel-
erate IAPP fibril formation.[7,8,26,27] It is believed that the
membrane interface provides sites for initiating the aggrega-
tion process. Therefore, we tested whether these inhibitors are
also able to inhibit fibril formation of IAPP in the presence of
lipid membranes. Lipid bilayer membranes were prepared
from large unilamellar vesicles (LUV) made up from dioleoyl
phosphatidylcholine (DOPC) and dioleoyl phosphatidylgly-
cerine (DOPG; which is negatively charged) in the molar
ratio 7:3. We applied—to our knowledge for the first time—
the surface-sensitive attenuated total reflection Fourier-trans-
form infrared (ATR-FTIR) spectroscopy technique[28] to be
able to follow the fibril formation at the lipid interface (top of
Figure 3). The buffer solution containing the lipid vesicles was
injected into the ATR cell leading to spontaneous formation
of supported lipid bilayers. ATR-FTIR absorbance spectra
are dominated by the signal from the supported membrane,
including the membrane-bound protein, whereas—because of
the low penetration depth (< 1 mm) of the evanescent IR
wave—the molecules far from the membrane do not contrib-
than two-fold more potent inhibitor than compound 1 (P-
value = 0.0085). The results obtained also indicate that it is
very likely that IAPP interacts with inhibitor 1 largely in the
ratio 1:1, which suggests that the inhibition occurs at a very
initial stage of the aggregation process. The non-equimolar
ratio inhibition observed for compound 2 indicates that this
compound is also active against small oligomers and prevents
them from forming mature fibrils. The interaction is probably
initialized by interaction of the heterocyclic ring system of the
inhibitors with hydrophobic patches on the surface of the
monomeric or oligomeric IAPP. The differences in the side
chains of the scaffold of compounds 1 and 2 may explain the
different extents of inhibition as was previously discussed in
the case of the tau protein.[21] Notably, the striking common
features shared between reported inhibitors of the aggrega-
tion of other amyloids[23,24] are the rigid planar hydrophobic
central core decorated with highly polar functionalities, such
as carboxylic acid, sulfonates, or phenols.
Owing to the fact that the fluorometric ThT assay may be
prone to errors arising from the inhibitor–ThT interaction,
additional techniques were employed to provide support of
these findings. Further evidence of inhibition of IAPP fibril
formation came from morphological studies using atomic
force microscopy (AFM). Fibrils grown for 96 h at room
temperature were diluted, applied on cleaved mica, and dried
before scanning at room temperature. IAPP forms long
unbranched fibrils as shown in Figure 2a, with heights of 3–
4 nm which is typical for amyloid fibrils.[25] To visualize the
effect of the two compounds on IAPP fibril formation, IAPP
4680 www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 4679 –4682

Angewandte
Chemie
Formation of b-sheet-rich amyloid fibrils is characterized by
their specific amide Isubband which appears at approxi-
mately 1623 cmÀ1.[29] As not all of the amino acids are
involved in the intermolecular b-sheet formation, a broad
shoulder remains around 1646 cmÀ1. The high absorbance
value indicates that IAPP molecules are in close contact to or
even inserted into the membrane. In agreement with other
recent spectroscopic data,[7,8] no such behavior is observed for
neutral lipids or in the absence of the anionic lipid bilayer in
the time-range covered in these experiments. When com-
pounds 1 and 2 were included in the buffer, no such
aggregation subband is visible (Figure 3b). Rather, broad
IR bands at about 1646–1648 cmÀ1 with low intensity appear,
which are characteristic of a large contribution of disordered
conformations arising from the presence of monomeric and
small oligomeric IAPP particles in solution close to the lipid
interface (in the evanescent wave region), only. Comparison
of Figures 2b and 2c reveals that in the presence of compound
2, IAPP forms small oligomeric structures of the size of 3–
4 nm, which are largely absent in the case of compound 1.
Such a scenario is in accord with the infrared data which
suggest that the IAPP-compound 2 complex corresponds to
the existence of a significant population of structurally
unordered oligomers. A slightly different picture emerges
for the scenario with compound 1, which exhibits a very low
IR absorbance only, typical for the presence of unfolded
monomeric IAPP in the solution. Again, this indicates that
the two inhibitors effectively interfere with the strong native
ability of IAPP to self-assemble, with differential specific
interactions, however, which lead to different populations of
the final inactive peptide species. This result also demon-
strates that even minor changes in the structure of these
small-molecule inhibitors may lead to several different
behaviors for modulating protein aggregation.
Generally, the lack of atomic resolution structures of
amyloid fibrils hampers the search for small-molecular
inhibitors. By screening of guiding structures, such inhibitors
may be found. Herein, we have reported the inhibition of
IAPP amyloid fibril formation by non-peptidic, small mole-
cules with a rhodanine scaffold at submicromolar concen-
trations. By using ThT fluorescence and ATR-FTIR spec-
troscopy as well as atomic force microscopy we were able to
show that the compounds 1 and 2 are effective inhibitors of
IAPP amyloid fibril formation. These inhibitors may be of
particular interest because of their biocompatibility and
negligible cytotoxicity at the concentrations investigated.[21]
As the cellular membrane is considered to be the primary
target for amyloid toxicity,[30] these compounds promise to be
also active against IAPP fibril formation in the cell, since they
already show inhibition in the presence of aggregation-
fostering lipid interfaces. Inhibition of fibrillation is an
important aspect in preventing (proto)fibril-induced toxicity.
However, an ideal inhibitor should be able to inhibit other
potential cytotoxic species, such as the oligomers as well.
Hence, compounds which act in the very early stage of the
fibrillation reaction, before cytotoxic species are formed,
would be ideal candidates for drug development. Finally, we
presume that the time-lapse ATR-FTIR spectroscopy tech-
nique introduced for studying aggregation processes of
Figure 3. a) Time-evolution of the ATR-FTIR spectra of 10 mm IAPP.
b) ATR-FTIR spectra of 10 mm IAPP (red), 10 mm IAPP + 1 mm com-
pound 1 (green) and 10 mm IAPP + 1 mm compound 2 (blue) after 20 h
incubation. All the spectra were collected in the presence of a lipid
bilayer made up from a 7:3 molar ratio of DOPC:DOPG at 258C and
deposited on the ATR crystal in buffer solution. Top: Scheme of the
ATR sample cell with an internal Ge reflection plate covered with the
DOPC–DOPG lipid bilayer. IAPP molecules adsorbing to the mem-
brane are detected while those distant from the membrane are barely
visible owing to the low penetration depth of the evanescent IR wave
(<1 mm).
ute significantly to the ATR-FTIR spectra. This situation
makes ATR-FTIR spectroscopy a surface-sensitive technique
that allows the study of protein binding to supported
membranes, and of the structural changes in both membrane
lipids and the protein that result from protein–membrane
interactions. The IAPP solution was then injected into the
ATR cell and allowed to adsorb at the membrane for up to
20 h. The ATR-FTIR spectra shown in Figure 3a show the
development of the aggregation band of IAPP with time.
Angew. Chem. Int. Ed. 2008, 47, 4679 –4682
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
4681

Communications
[14] F. E. Cohen, J. W. Kelly, Nature 2003, 426, 905 – 909.
[15] Y. Porat, A. Abramowitz, E. Gazit, Chem. Biol. Drug Des. 2006,
67, 27 – 37.
amyloidogenic proteins at lipid interfaces may prove very
valuable in forthcoming studies on related systems.
[16] C. Adessi, C. Soto, Curr. Med. Chem. 2002, 9, 963 – 978.
[17] M. Necula, R. Kayed, S. Milton, C. G. Glabe, J. Biol. Chem. 2007,
282, 10311 – 10324.
[18] M. Bartolini, C. Bertucci, M. L. Bolognesi, A. Cavalli, C.
Melchiorre, V. Andrisano, ChemBioChem 2007, 8, 2152 – 2161.
[19] M. Pickhardt, Z. Gazova, M. von Bergen, I. Khlistunova, Y.
Wang, A. Hascher, E. M. Mandelkow, J. Biernat, E. Mandelkow,
J. Biol. Chem. 2005, 280, 3628 – 3635.
Received: November 22, 2007
Revised: January 4, 2008
Published online: May 9, 2008
Keywords: amylin · amyloid · IAPP · small-molecule inhibitors ·
.
type-II diabetes
[20] M. Pickhardt, G. Larbig, I. Khlistunova, A. Coksezen, B. Meyer,
E. M. Mandelkow, B. Schmidt, E. Mandelkow, Biochemistry
2007, 46, 10016 – 10023.
[1] C. M. Dobson, Nature 2003, 426, 884 – 890.
[2] M. R. Nilsson, Methods 2004, 34, 151 – 160.
[3] F. Chiti, C. M. Dobson, Annu. Rev. Biochem. 2006, 75, 333 – 366.
[4] P. Westermark, C. Wernstedt, E. Wilander, D. W. Hayden, T. D.
OꢀBrien, K. H. Johnson, Proc. Natl. Acad. Sci. USA 1987, 84,
3881 – 3885.
[21] B. Bulic, M. Pickhardt, I. Khlistunova, J. Biernat, E. M.
Mandelkow, E. Mandelkow, H. Waldmann, Angew. Chem.
2007, 119, 9375 – 9379; Angew. Chem. Int. Ed. 2007, 46, 9215 –
9219.
[5] G. J. Cooper, A. C. Willis, A. Clark, R. C. Turner, R. B. Sim,
K. B. Reid, Proc. Natl. Acad. Sci. USA 1987, 84, 8628 – 8632.
[6] K. H. Johnson, T. D. OꢀBrien, C. Betsholtz, P. Westermark, Lab.
Invest. 1992, 66, 522 – 535.
[22] E. Zeiger, B. Anderson, S. Haworth, T. Lawlor, K. Mortelmans,
W. Speck, Environ. Mutagen. 1987, 9, 1 – 109.
[23] C. Chirita, M. Necula, J. Kuret, Biochemistry 2004, 43, 2879 –
2887.
[7] M. F. M. Engel, H. Yigittop, R. C. Elgersma, D. T. S Rijkers,
R. M. J. Liskamp, B. Kruijff, J. W. M. Höppener, J. A. Killian, J.
Mol. Biol. 2006, 356, 783 – 789.
[8] D. H. J. Lopes, A. Meister, A. Gohlke, A. Hauser, A. Blume, R.
Winter, Biophys. J. 2007, 93, 3132 – 3141.
[24] S. Taniguchi, N. Suzuki, M. Masuda, S. Hisanaga, T. Iwatsubo, M.
Goedert, M. Hasegawa, J. Biol. Chem. 2005, 280, 7614 – 7623.
[25] R. Jansen, W. Dzwolak, R. Winter, Biophys. J. 2005, 88, 1344 –
1353.
[26] J. D. Knight, A. D. Miranker, J. Mol. Biol. 2004, 341, 1175 – 1187.
[27] S. A. Jayasinghe, R. Langen, Biochim. Biophys. Acta 2007, 1768,
2002 – 2009.
[9] J. W. Höppener, B. AhrØn, C. J. Lips, N. Engl. J. Med. 2000, 343,
411 – 419.
[10] C. Y. Lin, T. Gurlo, R. Kayed, A. E. Butler, L. Haataja, C. G.
Glabe, P. C. Butler, Diabetes 2007, 56, 1324 – 1332.
[11] P. Westermark, U. Engström, K. H. Johnson, G. T. Westermark,
C. Betsholtz, Proc. Natl. Acad. Sci. USA 1990, 87, 5036 – 5040.
[12] L. M. Yan, M. Tatarek-Nossol, A. Velkova, A. Kazantzis, A.
Kapurniotu, Proc. Natl. Acad. Sci. USA 2006, 103, 2046 – 2051.
[13] L. M. Yan, A. Velkova, M. Tatarek-Nossol, E. Andreetto, A.
Kapurniotu, Angew. Chem. 2007, 119, 1268 – 1274; Angew.
Chem. Int. Ed. 2007, 46, 1246 – 1252.
[28] S. A. Tatulian, Biophys. J. 2001, 80, 789 – 800.
[29] W. Dzwolak, V. Smirnovas, R. Jansen, R. Winter, Protein Sci.
2004, 13, 1927 – 1932.
[30] R. Kayed, E. Head, J. L. Thompson, T. M. McIntire, S. C. Milton,
C. W. Cotman, C. G. Glabe, Science 2003, 300, 486 – 489.
4682 www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 4679 –4682