Green tea (2)-epigallocatechin-gallate modulates
early events in huntingtin misfolding and reduces
toxicity in Huntington’s disease models
Dagmar E. Ehrnhoefer1, Martin Duennwald2, Phoebe Markovic1, Jennifer L. Wacker3,
Sabine Engemann1, Margaret Roark5, Justin Legleiter3,4, J. Lawrence Marsh5,
Leslie M. Thompson6, Susan Lindquist2, Paul J. Muchowski3,4and Erich E. Wanker1,*
1Max Delbrueck Center for Molecular Medicine (MDC), Department of Neuroproteomics, Robert-Roessle-Straße 10,
13092 Berlin, Germany,2Whitehead Institute for Biomedical Research, Nine Cambridge Center, Cambridge, MA
02142, UK,3Department of Pharmacology, University of Washington, Seattle, WA 98195-7280, USA,4Gladstone
Institute of Neurological Disease, and Departments of Biochemistry and Biophysics, and Neurology, University of
California, San Francisco, CA 94158, USA,5Department of Developmental and Cell Biology and6Departments of
Psychiatry and Human Behavior and Biological Chemistry, University of California, Irvine, CA 92697, USA
Received May 16, 2006; Revised and Accepted July 31, 2006
Huntington’s disease (HD) is a progressive neurodegenerative disorder for which only symptomatic treat-
ments of limited effectiveness are available. Preventing early misfolding steps and thereby aggregation of
the polyglutamine (polyQ)-containing protein huntingtin (htt) in neurons of patients may represent an attrac-
tive therapeutic strategy to postpone the onset and progression of HD. Here, we demonstrate that the green
tea polyphenol (2)-epigallocatechin-3-gallate (EGCG) potently inhibits the aggregation of mutant htt exon 1
protein in a dose-dependent manner. Dot-blot assays and atomic force microscopy studies revealed that
EGCG modulates misfolding and oligomerization of mutant htt exon 1 protein in vitro, indicating that it inter-
feres with very early events in the aggregation process. Also, EGCG significantly reduced polyQ-mediated htt
protein aggregation and cytotoxicity in an yeast model of HD. When EGCG was fed to transgenic HD flies
overexpressing a pathogenic htt exon 1 protein, photoreceptor degeneration and motor function improved.
These results indicate that modulators of htt exon 1 misfolding and oligomerization like EGCG are likely to
reduce polyQ-mediated toxicity in vivo. Our studies may provide the basis for the development of a novel
pharmacotherapy for HD and related polyQ disorders.
Huntington’s disease (HD) is caused by an unstable CAG
repeat expansion in the first exon of the IT-15 gene which
encodes huntingtin (htt), a ?350 kDa protein, functionally
involved in clathrin-mediated endocytosis, vesicle transport
processes and transcriptional regulation (1,2). The trinucleo-
tide expansion translates into an elongated polyglutamine
(polyQ) stretch, and the disease appears when the pathological
threshold of 37 glutamine residues is exceeded (3). The dis-
order is characterized by a progressive loss of cortical and
striatal neurons and the formation of neuronal inclusions
containing aggregated htt protein (4,5). There is evidence
that mutant htt aggregate formation is causally linked to the
progressive neuropathology of the disease (6), though it is
not clear whether large insoluble, fibrillar structures or
smaller assemblies of htt are the toxic agents responsible for
neuronal damage and loss [reviewed by (7)]. Furthermore,
polyQ-containing proteins, i.e. transcription factors or wild-
type htt, into the neuronal inclusions, which would result in
a loss of their normal cellular functions [see review by (8)].
Several studies have focused on the identification of mole-
cules that are able to interfere with aggregation of mutant htt.
# The Author 2006. Published by Oxford University Press. All rights reserved.
For Permissions, please email: email@example.com
*To whom correspondence should be addressed. Tel: þ49 30 9406 2157; Fax: þ49 30 9406 2552; Email: firstname.lastname@example.org
Human Molecular Genetics, 2006, Vol. 15, No. 18
Advance Access published on August 7, 2006
by guest on September 13, 2015
Antibodies directed against the elongated polyQ tract or the N-
terminal region of htt efficiently inhibit aggregation in vitro or
in cell culture models of HD (9,10). Similarly, bivalent
polyQ-containing peptides were found to suppress htt aggrega-
tion and to inhibit photoreceptor degeneration in fly models of
HD (11,12). However, peptides have a high probability of
inducing allergic reactions and are barely able to pass the
These problems do not necessarily occur with chemical
compounds. In recent years, several research groups have
been searching for small molecule inhibitors of polyQ aggre-
gation using cell-freeand
Compounds like Congo red, PGL-34, thioflavine S, gossypol
or trehalose have been found to prevent htt aggregation (15).
However, it is not known so far how or at which stage of
the aggregation process these compounds exert their inhibitory
influence. Also, for most of these substances it is unclear
whether they have a therapeutic effect in animal models
A crucial issue in the development of novel therapeutics is
tolerance of the drug by the human organism, especially con-
sidering that in diseases like HD, drugs are taken chronically
for long time periods. Well tolerated are e.g. the flavonoids,
a large group of naturally occurring polyphenolic substances
that are found in many plants (16). The everyday intake of fla-
vonoids in food is in the range of 50–800 mg, depending on
the fruit, vegetables and beverages consumed (16). Previous
studies have demonstrated that they are able to modulate
various cellular processes; their pharmacological effects
have mostly been linked to their antioxidant properties (17).
Furthermore, their ability to reduce oxidative stress leads to
neuroprotection in models of Alzheimer’s and Parkinson’s
disease (17), suggesting that they could be developed into
therapies for neurodegenerative disorders.
In this study, we have screened a library of natural com-
pounds and identified (2)-epigallocatechin-3-gallate (EGCG)
cell-based assays (13,14).
and related polyphenols as potent inhibitors of mutant htt
exon 1 protein aggregation in vitro. We found that EGCG
modulates misfolding as well as the assembly of oligomers
in cell-free assays and reduces both toxicity and aggregate for-
mation in yeast and fly models of HD. The relevance of our
findings with regard to the HD pathomechanism and therapy
development is discussed.
Identification of green tea polyphenols as inhibitors of
htt exon 1 aggregation
In order to identify inhibitors of htt aggregation, we screened a
library of ?5000 natural substances using a membrane filter
retardation assay (14). Mutant GST-tagged htt exon 1 fusion
protein with 51 glutamines (GST-HDQ51) was incubated for
16 h at 378C in the presence of both the chemical compounds
to be tested and elastase, a protease that very efficiently trig-
gers the aggregation process by completely removing the
GST-tag from the fusion protein. The samples were denatured
by boiling in SDS/DTT, and aggregates were trapped on a cel-
lulose acetate filter followed by immunodetection with the
CAG53b antibody (18). With this approach we detected six
plant extracts, which we did not analyze further, and one puri-
fied natural compound, EGCG. EGCG is a polyphenol present
in green tea (for structure see Fig. 1) that has antioxidant prop-
erties and has been shown to cross the blood–brain barrier
(19). It suppressed the assembly of HDQ51 aggregates in a
concentration-dependent manner with an IC50 value of
?1 mM, corresponding to a molar ratio of drug to protein of
2:1 (Fig. 2A and B).
We also tested the stereoisomer (2)-gallocatechin 3-gallate
(GCG) andtwo otherrelated
(2)-gallocatechin (GC) and (2)-epigallocatechin (EGC)
that have known antioxidant activities in vitro (Fig. 1). GCG
Figure 1. Chemical structures of the aggregation inhibitors tested.
2744Human Molecular Genetics, 2006, Vol. 15, No. 18
by guest on September 13, 2015
inhibited aggregation to a lesser extent than EGCG (IC50of
?2 mM), whereas GC and EGC, which unlike EGCG and
GCG do not possess a gallate moiety, did not show a signifi-
cant effect on HDQ51 aggregation (Fig. 2A and B). This indi-
cates that the presence of the gallate esters rather than the
antioxidative properties of polyphenols in general are crucial
for the inhibitory effect in vitro. In good agreement with
this,also the strong antioxidants
a-tocopherol did not significantly influence the assembly of
SDS-stable htt exon 1 aggregates in the cell-free assays
(Figs 1, 2A and B).
EGCG modulates the formation of htt exon 1
oligomers in vitro
Previous studies have demonstrated that htt exon 1 fibrillogen-
esis is a complex process involving the formation of oligomers
that can be observed by atomic force microscopy (AFM) (20).
Here, we examined whether the addition of EGCG to aggrega-
tion reactions modulates the assembly of such structures in
vitro. GST-taggedfusion protein
(GST-HDQ53) was incubated with the site-specific PreScis-
sion protease to remove the GST tag and to initiate the aggre-
gation of HDQ53. At 5 h after GST cleavage, aliquots were
taken from the aggregation reactions and the assembly of oli-
gomers was analyzed by AFM. As shown in Figure 3A and B,
the dominant structure observed was a heterologous popu-
lation of spherical oligomers with a diameter of 20–80 nm,
which is in agreement with previous studies (20). Strikingly,
in EGCG-treated samples, the density (particles/field) of
these structures was significantly reduced, whereas the
number of spherical oligomers with a larger diameter
(?120–200 nm) was increased (Fig. 3A and B). A similar
result was obtained with the stereoisomer GCG (data not
shown), whereas the compounds EGC and GC lacking the
gallate moiety (Fig. 1) did not significantly modulate oligo-
merization in vitro (data shown exemplarily for GC, Fig. 3A
and B). This indicates that EGCG suppresses the formation
of small HDQ53 oligomers by stimulating the formation of
EGCG interferes with a conformational change in
mutant htt exon 1 protein
Schaffar et al. (21) demonstrated that proteolytic cleavage of
GST-htt exon 1 fusion protein induces a rapid conformational
change in the polyQ-containing htt fragment, which can be
monitored by FRET assay. We investigated whether EGCG
can influence this intramolecular rearrangement upon cleavage
of the htt fragment from the GST tag. To monitor the struc-
tural change, we used a dot-blot assay and the monoclonal
antibody MW1, which specifically registers the expanded
polyQ tract (22) and has been used previously to observe the
occlusion of the polyQ epitope after cleavage of a GST-htt
exon 1 fusion protein with PreScission protease (20).
Figure 4A shows that the MW1 antibody very efficiently
recognized the uncleaved fusion protein. When the GST tag
was removed with elastase, however, 60% of the immunoreac-
tivity was lost immediately after addition and no signal was
Figure 2. The polyphenols EGCG and GCG inhibit the formation of HDQ51
aggregates in vitro. (A) Effect of the indicated compounds on HDQ51 aggre-
gation as monitored by the filter retardation assay and immunodetection with
the CAG53b antibody. (B) Quantification of the filter assay results shown in
(A). The amount of aggregates retained on the filter in the solvent control
was arbitrarily set at 100%. The data reported are representative for three inde-
pendent experiments +SE.
Figure 3. EGCG stimulates the formation of large spherical oligomers. (A)
GST-HDQ53 was cleaved with PreScission protease and incubated for 5 h
with a 2-fold molar excess of EGCG, GC or the solvent alone. Aliquots
were analyzed by AFM (image size 3 mm2each). (B) The particles were
grouped with respect to their size (20 nm diameter intervals). Five hundred
and forty-one particles were counted for the solvent, 312 for GC and 340
for EGCG. The correlation coefficients between all data sets were 0.9765,
0.9736 and 0.9331 for solvent, GC and EGCG, respectively. The asterisk indi-
cates particles that were only observed after EGCG treatment.
Human Molecular Genetics, 2006, Vol. 15, No. 182745
by guest on September 13, 2015
detected after an incubation of 2 min (Fig. 4A). Such a dra-
matic effect was neither seen with the antibody MW7,
which specifically recognizes the poly-proline region in the
htt exon1 protein (22), nor with the CAG53b antibody,
which detects both the intact GST fusion protein and its frag-
ments after proteolytic cleavage (18). Thus, our findings
strongly support previous results (21) indicating that the htt
fragment undergoes a very rapid structural rearrangement
upon cleavage from GST that can be monitored by the
polyQ-specific antibody MW1.
MW1-immunoreactivity was slowed down significantly (but
notprevented) ina concentration-dependent
(Fig. 4B and C). When a 5-fold molar excess of EGCG was
added to the aggregation reaction, MW1 immunoreactivity
was still clearly detectable after an incubation time of
30 min. This suggests that the compound binds to the unstruc-
tured polyQ sequence and interferes with the conformational
rearrangement assumed to occur immediately after cleavage
EGCG reduces toxicity and aggregate load in a
yeast model of HD
In order to test whether EGCG influences htt exon 1 toxicity in
vivo, an yeast model of HD was used. The htt exon 1 GFP
fusion proteins GFP-HDQ25 and GFP-HDQ72 were overex-
pressed in yeast and cell growth was measured by monitoring
the optical density at 600 nm. As shown in Figure 5A, overex-
pression of GFP-HDQ25 did not significantly influence yeast
cell growth.When GFP-HDQ72
however, cell proliferation was completely blocked, indicating
that the GFP fusion protein with the expanded polyQ tract is
toxic in yeast. Strikingly, when these cells were treated with
(Fig. 5A), indicating that the compound is able to reduce the
toxicity of the GFP-HDQ72 protein in vivo.
A recent study demonstrated polyQ aggregation and toxicity
in yeast to be dependent on the prion state of the protein Rnq1
(23). To exclude the possibility that yeast growth in EGCG-
containing media is due to a spontaneous conversion of
Rnq1 from the aggregated to the soluble state, centrifugation
assays were performed. Supernatant and pellet fractions
were analyzed by western blotting. We found that EGCG-
treatment did not influence the aggregation state of Rnq1, indi-
cating that solubilization of Rnq1 cannot be the cause for the
EGCG-mediated reduction of polyQ toxicity (Supplementary
Material, Fig. S1).
Furthermore, we tested whether EGCG-treated cells expres-
sing the toxic GFP-HDQ72 protein can regrow on SD þ Gal
medium lacking the compound. We found that the yeast
cells, after removal of the compound, do not grow on
SD þ Gal plates (galactose is required for the induction of
SD þ Glu medium. This strongly indicates that the EGCG
treatment did not induce spontaneous mutations in the yeast
genome responsible for the improved growth phenotype
(data not shown).
Next, we examined whether EGCG can influence the for-
mation of insoluble protein aggregates in yeast. Fluorescence
Figure 4. EGCG interferes with a conformational change in HDQ51. (A) 0.6 mM
GST-HDQ51 was incubated with elastase for the indicated times and samples
were spotted onto nitrocellulose for immunodetection with the antibodies
MW1, MW7 and CAG53b. MW1 immunoreactivity disappeared rapidly after
cleavage, indicating a fast intramolecular conformational change in the PolyQ
tract. (B) 0.6 mM GST-HDQ51 was incubated with elastase and 0.3, 0.6 or
3 mM EGCGorthesolventat378Candspottedontonitrocelluloseattheindicated
time points. The disappearance of MW1 immunoreactivity was significantly
slowed down by the addition of increasing concentrations of EGCG, indicating
that the compound modulates the misfolding of HDQ51 in vitro. No such effect
was observed with the polyclonal CAG53b antibody, which detects both the
GST-tag and the htt exon 1 protein. (C) Quantification of the results obtained
with the MW1 antibody after the addition of 3 mM EGCG or solvent as shown
in (B). The signal obtained for uncleaved reactions was arbitrarily set at 100%.
The data reported represent three independent experiments +SE.
2746Human Molecular Genetics, 2006, Vol. 15, No. 18
by guest on September 13, 2015
microscopy showed that EGCG treatment reduces the number
of yeast cells containing GFP-HDQ72 protein aggregates
(?40% reduction compared with controls, Fig. 5B). Similarly,
when cell extracts were analyzed with the membrane filter
retardation assay (24), a lower amount of SDS-insoluble
GFP-HDQ72 aggregates was observed in EGCG-treated
cells (Fig. 5C), indicating that EGCG suppresses not only toxi-
city but also protein aggregation in an yeast model of HD.
EGCG reduces photoreceptor degeneration and
motor impairment in HD transgenic flies
Previous studies in transgenic flies have demonstrated that
overexpression of htt exon 1 protein with an expanded
polyQ sequence of 93 glutamines (Htt93Q) causes progressive
photoreceptor neuron degeneration that can be monitored by
light microscopy (25). The disruption of the regular trapezoidal
arrangement of seven visible photoreceptor neurons (rhabdo-
meres) can be examined as a marker for more widespread
improves photoreceptor degeneration, transgenic flies expres-
sing Htt93Q in all neurons from embryogenesis onward
were fed with the compound and the number of rhabdomeres
per ommatidium was analyzed using the pseudo-pupil tech-
nique (25). Freshly eclosed Htt93Q adults were transferred
to vials containing increasing doses of compound or solvent
alone (DMSO). After 7 days, the number of rhabdomeres
was determined. Because of the fact that the transgenic
protein is already expressed during development, eyes have
begun to degenerate when the flies emerge. Compound effi-
cacy is evaluated by comparing treated flies at day 7 with
day 0 control flies. In the absence of EGCG, htt-expressing
neurons deteriorate until on average ?3.5 photoreceptor
neurons/ommatidium remain after 7 days. When increasing
amounts of compound were present, a dose-dependent
slowing down of the degenerative process could be observed.
With the highest EGCG concentration tested (100 mM), an
average of ?4.2 neurons/ommatidium was found after 7
days, corresponding to a rescue of ?29% (Fig. 6B). A
visual representation of rescue of the number of rhabdo-
meres/ommatidium is shown in Figure 6A.
Finally, we tested whether EGCG treatment can improve
motor function in the HD transgenic flies using a climbing
assay (25). Flies overexpressing HttQ93 show a progressively
abnormal movement and climbing behavior with age because
of impaired motor and neuronal function (Fig. 6C). This pheno-
type, however, was dramatically improved when the HD flies
were mated and raised on sugar supplemented with 500 mM
EGCG. While untreated transgenic flies, at an age of 18 days,
couldclimbonanaverage4–5 cm/60 sinaverticalglasscylin-
der, the EGCG-treated flies climbed about 18 cm in the same
time, indicating that the compound not only reduces photo-
mal motor ability in the transgenic flies. Together, these studies
indicate that EGCG is not only a potent inhibitor of
polyQ-mediated htt exon 1 protein aggregation in vitro but is
also able to reduce toxicity in different in vivo models of HD.
Growing evidence suggests that the misfolding and aggrega-
tion of htt is central to HD pathogenesis. Using a well-
established filter assay, we screened a library of natural
substances and identified EGCG and related compounds
from green tea as potent inhibitors of htt exon 1 aggregation.
Green tea derivatives are very attractive drug candidates
because they are naturally occurring phytopharmaca with
known neuroprotective effects (16). EGCG e.g. has been
shown to reduce oxidative stress and neurotoxicity in different
model systems of Alzheimer’s and Parkinson’s disease and to
modulate the expression of cell survival and cell death genes
(17). Furthermore, EGCG was reported to pass the blood–
brain barrier in mammals (19) and to be safe for humans
when tested in clinical studies (26).
Figure 5. EGCG diminishes toxicity and aggregation of mutant htt fragments
in yeast. (A) Growth of yeast expressing GFP-HDQ25 or GFP-HDQ72 in the
presence or absence of 500 mM EGCG was monitored by measuring the OD600
every 30 min over an interval of 144 h. (B) Representative fluorescence
microscopy images of yeast cells expressing GFP-HDQ72 after cultivation
for 4 h in the presence of 500 mM EGCG or the solvent. The percentage of
cells containing visible aggregates was determined (a total of 1500 cells
was counted), the average results from three independent experiments +SE
are shown. (C) Yeast cells expressing GFP-HDQ72 were grown in the pre-
sence of 500 mM EGCG or solvent for 6 h. Cell lysates were spotted onto
nitrocellulose (left) or subjected to the filter retardation test (right). The
total amounts of GFP-htt spotted or the amount of aggregates retained on
the filter were detected with a GFP antibody.
Human Molecular Genetics, 2006, Vol. 15, No. 18 2747
by guest on September 13, 2015
Our data obtained from in vitro models of HD demonstrate
for the first time that green tea polyphenols are able to
modulate early steps in the aggregation process of an amyloi-
dogenic polyQ-containing protein. The inhibitory effect on
the assembly of mutant htt exon 1 fragments in the cell-free
assays is concentration-dependent and does not require the
antioxidant properties of the polyphenols. Nevertheless,
the known beneficial activities of EGCG and its derivatives
such as radical scavenging, reduction of reactive oxidative
species or chelating of metal ions may contribute to the
decrease of htt aggregation and toxicity in the in vivo
models of HD (17).
The aggregation of htt exon 1 fragments in vitro is a multi-
step and, potentially, multipathway process that involves an
initial conformational change in the polyQ tract (21), the for-
mation of annular or spherical oligomers and protofibrils and
the self-assembly of mature fibrils (27). Two different
models for the conversion of polyQ proteins into amyloid-like
fibrils have been proposed. On the one hand, it has been
suggested that multiple conformations of misfolded htt mono-
mers coexist and give rise to several distinct oligomeric or
amorphous assemblies as well as fibrillar structures (20). On
the other hand, misfolded htt monomers are believed to
proceed via transient oligomers or protofibrils into amyloid-
like structures (28). In this model, oligomers are metastable
aggregation intermediates that form during the nucleation
phase and act as seeds for the assembly of large fibrillar struc-
tures. However, the precise assembly pathways for the differ-
ent types of htt aggregates observed in vitro are currently
unclear. Moreover, it is unknown which type of protein assem-
bly causes dysfunction and toxicity in mammalian cells.
Using an in vitro aggregation assay, we found that the epi-
topes of expanded polyQ tracts very rapidly become inaccess-
ible for the MW1 antibody after proteolytic cleavage. This
observation is in good agreement with the data presented pre-
viously (21) and suggests that fragments of mutant htt undergo
a conformational change very early in the aggregation process.
In our cell-free assay, EGCG significantly prolonged the
accessibility of the polyQ epitope for the MW1 antibody in
a concentration-dependent manner. This indicates that the
compound modulates the intramolecular rearrangement of
mutant htt exon1 prior to aggregate formation, influencing
the initial misfolding step in the aggregation cascade.
We propose that in untreated reactions, protease cleavage of
the GST fusion protein triggers a very rapid intramolecular
change in the polyQ-containing htt exon 1 fragment that
leads to a compact b-sheet-rich structure. This compaction
is most likely caused by the formation of hydrogen bonds
between the main chain and side chain amides in the polyQ
tract. In EGCG-treated samples, however, either the formation
of stable hydrogen bonds is slowed down, or the binding of
EGCG molecules to htt might induce the formation of a
stable protein-drug adduct that has a novel structure and
cannot be recognized by the polyQ-specific MW1 antibody.
Previous studies have shown that gallate esters are import-
ant for the association of polyphenols with proteins, probably
because they mediate weak hydrophobic interactions with
proline-rich regions in target proteins, which is a prerequisite
for subsequent hydrogen bond formation (29). Our findings
that only EGCG and GCG, but not the related compounds
GC and EGC are able to modulate htt exon 1 aggregation in
vitro are in agreement with the observations that the gallate
moiety is critical for the binding of polyphenols to proteins.
We suggest that the proline-rich region in the htt exon 1
protein might function as an anchoring point for EGCG
binding and the adjacent polyQ sequence causes the formation
of stable hydrogen bonds between the compound and the poly-
peptide chain. However, more detailed studies will be necess-
ary to characterize the structure of the polyphenol-htt complex
formed in vitro.
Figure 6. EGCG diminishes the toxicity of mutant htt fragments in fly models
of HD. (A) Rescue of degeneration of photoreceptor neurons (PR) in Droso-
phila expressing mutant htt fragments with 93 glutamines. Representative
deep pseudopupil images of eyes from 7-day-old EGCG-treated and solvent-
treated flies. Left panel: Solvent-treated flies show extensive degeneration
with many ommatidia containing either three (white arrowhead) or four
(yellow arrowhead) rhabdomeres. Right panel: Flies fed 100 mM EGCG
show less degeneration, most ommatidia contain four to five rhabdomeres
with some retaining seven (green arrowhead). (B) Newly eclosed flies were
kept on food supplemented with different amounts of EGCG (0.01, 0.1, 1,
10 and 100 mM) or solvent only. The average number of photoreceptor cells
per ommatidium was scored 7 days post-eclosion. Error bars represent
+SE. Day 0 shows the neuronal loss that has occurred during pupal develop-
ment prior to drug treatment. Values ?0.1 mM are all significantly different
(P , 0.002) from the no drug control, 0.01 mM EGCG had no significant
effect. Maximal rescue in the representative experiment shown was 29%.
(C) Age-dependent deterioration of climbing abilities in flies expressing
mutant htt fragments with 93 glutamines (UAS-httQ93) in neurons was moni-
tored by placing the flies on the bottom of a vertical glass tube and measuring
the distance covered in 60 s. Flies mated and raised on food with 500 mM
EGCG performed significantly better than solvent-treated controls.
2748Human Molecular Genetics, 2006, Vol. 15, No. 18
by guest on September 13, 2015
AFM analysis also revealed that the addition of EGCG
to aggregation reactions changed the formation of htt exon
1 oligomers in vitro. EGCG significantly reduced the density
of small oligomers with a diameter of 20–80 nm while
larger, spherical oligomers with a diameter of 120–200 nm
had appeared. This is a clear indication that the compound
not only alters the rapid conformational change of the htt
exon 1 fragment after proteolytic cleavage of the fusion
protein, but also has a modulating impact on later steps in
the aggregation process. We propose that the larger spherical
oligomers, which are formed after EGCG treatment are off-
pathway for fibril formation and are more biologically inert
than the small oligomeric structures that form in the absence
of the chemical compound. Thus, EGCG may reduce polyQ
toxicity in a similar way as the molecular chaperones Hsp70
and Hsp40 do because it attenuates the assembly of small
spherical oligomers that appear very early in the aggregation
The effect of EGCG on polyQ-mediated htt exon 1 aggrega-
tion and toxicity was examined in an yeast model of HD. We
found that the compound significantly reduced both the for-
mation of SDS-stable htt aggregates as well as toxicity in
this in vivo model system, supporting the hypothesis that com-
pounds that function as chemical chaperones and modulate
early steps in the aggregation pathway (misfolding and oligo-
merization) have a high potential for therapy development
(30). This assumption is further strengthened by our finding
that EGCG ameliorates the pathological effects of mutant htt
in a fly model of HD. Currently, the molecular mechanisms
by which htt fragments with pathogenic polyQ sequences
exert their toxic effects in a cellular environment are
unclear. Misfolded b-sheet-rich htt monomers or small
soluble oligomers form abnormal protein–protein interactions
with other polyQ proteins such as the transcription factors
TBP and CBP in vitro, suggesting that such interactions
might also occur in vivo in neuronal cells and contribute to
polyQ-mediated htt toxicity in mammalian cells by changing
the conformation of the disease protein and by preventing
the formation of destructive protein–protein interactions.
We conclude that EGCG is a potent inhibitor of polyQ
aggregation that has beneficial effects in vivo. It may protect
neuronal cells expressing a mutant htt protein from its
noxious properties. Finally, our study reveals EGCG to have
considerable potential as drug candidate for the development
of treatments of HD and of protein misfolding and amyloid
diseases in general.
might reduce the
MATERIALS AND METHODS
Proteins, antibodies and chemical compounds
bacteria and purified as described (14,31). The recombinant
proteins were stored at 2808C after shock-freezing in liquid
nitrogen. Before each experiment, the proteins were either
centrifuged at 230.000g or 18.000g for 20–30 min at 48C.
The polyclonal anti-htt antibody CAG53b has been
described (18). Themonoclonal
polyproline antibodies MW1 and MW7 (22), respectively,
andGST-HDQ53 were overexpressedin
were obtained from the Developmental Studies Hybridoma
Bank (University of Iowa). The anti-GFP antibody was pur-
chased from Roche (Indianapolis, USA).
The compounds EGCG, GCG, (2)-epigallocatechin (EGC)
and (2)-gallocatechin (GC), ascorbic acid and a-tocopherol
were purchased from Sigma–Aldrich (Hamburg, Germany).
Stock solutions of the compounds of 10 mM were prepared
in DMSO and stored at 48C.
Compound screening and filter retardation assay
A compound library containing ?5000 natural substances was
screened for inhibitors of htt aggregation as described (14).
Total removal of the GST-tag from the fusion protein was
achieved by elastase treatment (3 min at 378C in 150 mM
NaCl, 20 mM Tris–HCl pH 8.0 and 2 mM CaCl2) before the
addition of chemical compounds. HDQ51 of 0.6 mM was
then incubated with compound (0.6, 3, 6 mM) or the solvent
DMSO alone for 16 h at 378C to allow aggregate formation.
For the filter retardation assay, aliquots were mixed with 2%
SDS and 50 mM DTT (final concentrations) and denatured at
988C for 7 min. Aliquots corresponding to 125 ng of
GST-HDQ51 protein were filtered through a cellulose
acetate membrane (0.2 mm pore size, Schleicher & Schuell,
Dassel, Germany). Aggregates captured on the membrane
were detected with the CAG53b antibody, an alkaline
phosphatase-conjugated secondary antibody and the fluor-
escent substrate AttoPhos. Signals were quantified using the
AIDA image analysis software (Raytest, Straubenhardt,
Analysis of epitope accessibility
Aggregation reactions with GST-HDQ51 were set up as for
the filter retardation assay, however, elastase and the chemical
compounds were added to the protein at the same time to
monitor early compound effects. At different time points, ali-
quots corresponding to 2 mg fusion protein were spotted onto
ProtranTMnitrocellulose membranes (Schleicher & Schuell).
Membranes were then immunodetected with primary antibody
(MW1, MW7 or CAG53b), an alkaline phosphatase coupled
secondary antibody and the fluorescent substrate AttoPhos.
Signals were quantified using the AIDA image analysis
GST-HD53Q protein was prepared at a concentration of 6 mM
in buffer A (50 mM Tris–HCl pH 7, 150 mM NaCl, 1 mM
DTT, 1 mM PMSF, 0.5 mM leupeptin, 0.5 mM pepstatin A).
At time zero, 12 mM compound and PreScission protease
(4 units/100 mg fusion protein) (Amersham Biosciences,
Piscataway, USA) were added to initiate removal of GST
and HDQ53 aggregation. Aliquots of HDQ53 protein were
removed from the reactions and diluted into 50 mM Tris–
HCl pH 7. The protein was spotted on freshly cleaved mica,
allowed to adhere for 2 min and then washed with 200 ml dis-
tilled water. The samples were partially dried with compressed
air, dried to completion at room temperature and imaged in
air with a digital multimode NanoscopeIII scanning probe
Human Molecular Genetics, 2006, Vol. 15, No. 182749
by guest on September 13, 2015
microscope operating in tapping mode. Representative 3 mm2
images obtained in three separate experiments were used for
size/density analysis with custom written software (32).
Yeast strains employed in this study were in the W303 (MATa
can1-100 ade2-1 his3-11, 15 trp1-1 ura3-1 leu23,112) genetic
background and contained integrated plasmids (pRS303 back-
bone) for the expression of green fluorescent protein- (GFP)
HDQ25 or HDQ72 fusion proteins under the control of the
inducible GAL1 promotor.
Growth of yeast cultures in selective media containing raf-
finose (2%) as the sole carbon source to mid-log phase fol-
lowed by growth in galactose (2%) containing selective
media induced GFP-HDQ25/72 expression.
media with glucose (2%). Cells were washed and diluted to an
OD600of 0.02 in media containing 2% galactose. Growth curves
were determined with the Bioscreen instrument by taking OD600
measurements at 30 min intervals over a period of 72 h.
Filter retardation assays of aggregated material were per-
formed as described (24) with total lysates prepared from
yeast cells expressing GFP-HDQ72. Dot-blots were performed
like filter retardation assays except that PVDF membranes
were used. Aggregates and total htt protein on the membranes
were immunodetected with the GFP-antibody.
The percentage of cells containing GFP-HDQ72 aggregates
was determined with a Zeiss Axioplan II microscope and the
Openlab (ImproVision, Lexington, USA) software. In total,
1500 cells of three independent cultures were counted.
Drosophila melanogaster methods
Expression of the htt protein in transgenic flies is driven by the
bipartite expression system upstream activator sequence
(UAS)-GAL4 (yeast transcriptional activator). Stocks w;
w þ ;
UAS-Q93httexon1)P463were mated in order to obtain flies
expressing mutant htt fragments in all neurons from embryo-
genesis on. Flies were mated at 258C on standard food sup-
plemented with the compound to be tested, transferred to
fresh food daily and assayed for neurodegeneration at day 7
post-eclosion (25). Concentrations used in the feedings were:
0.01, 0.1, 1, 10 and 100 mM, using a 20 mM stock in DMSO.
Vials were normalized to the same concentration of DMSO,
always ?0.5% to control for DMSO effects (25). The
plotted values are an average of 10 eyes from 10 flies
assayed at each concentration with a minimum of 100 omma-
tidia (omm) counted and averaged per eye. The significance of
the resulting differences in average number of PRs was
analyzed using the Wilcoxon rank-sum test. Error bars in
Figure 5A represent +SE. Percent rescue was calculated
as [(ave no. PR/omm in drug-fed flies) 2 (ave no. PR/omm
in no-drug control flies)] 4 [(ave no. PR/omm at day
zero) 2 (ave no. PR/omm in no-drug control flies)] ? 100.
The average number of PR/omm present at day 0 in a repre-
sentative sample of flies is measured prior to drug feeding.
For the climbing assay, flies were mated and raised on
media containing 500 mM EGCG or the DMSO solvent. At
P(w þ mC ¼
different timepoints after eclosion, 20 flies were placed in a
vial and the height half of them were able to climb from the
bottom within 60 s was determined. The values shown are
the average of five determinations +SE.
Supplementary Material is available at HMG Online.
We thank S. Schnoegl for critical reading of the manuscript
and editorial support. The project was funded by the NGFN
LSHM-CT-2003-503330, Deutsche Forschungsgemeinschaft
grants Wa-1151/5-1 and He-1928/8-2, the Huntington’s
Disease Society of America (HDSA), the HDF, NINDS
(NS54753) and the High Q foundation.
Conflict of Interest statement. The authors do not declare any
conflict of interest.
1. The Huntington’s Disease Collaborative Research Group (1993) A novel
gene containing a trinucleotide repeat that is expanded and unstable on
Huntington’s disease chromosomes. Cell, 72, 971–983.
2. Harjes, P. and Wanker, E.E. (2003) The hunt for huntingtin function:
interaction partners tell many different stories. Trends Biochem. Sci., 28,
3. Rubinsztein, D.C., Leggo, J., Coles, R., Almqvist, E., Biancalana, V.,
Cassiman, J.J., Chotai, K., Connarty, M., Crauford, D., Curtis, A. et al.
(1996) Phenotypic characterization of individuals with 30-40 CAG repeats
in the Huntington disease (HD) gene reveals HD cases with 36 repeats and
apparently normal elderly individuals with 36–39 repeats. Am. J. Hum.
Genet., 59, 16–22.
4. Vonsattel, J.P., Myers, R.H., Stevens, T.J., Ferrante, R.J., Bird, E.D. and
Richardson, E.P., Jr. (1985) Neuropathological classification of
Huntington’s disease. J. Neuropathol. Exp. Neurol., 44, 559–577.
5. DiFiglia, M., Sapp, E., Chase, K.O., Davies, S.W., Bates, G.P., Vonsattel,
J.P. and Aronin, N. (1997) Aggregation of huntingtin in neuronal
intranuclear inclusions and dystrophic neurites in brain. Science, 277,
6. Yamamoto, A., Lucas, J.J. and Hen, R. (2000) Reversal of neuropathology
and motor dysfunction in a conditional model of Huntington’s disease.
Cell, 101, 57–66.
7. Glabe, C.G. and Kayed, R. (2006) Common structure and toxic function
of amyloid oligomers implies a common mechanism of pathogenesis.
Neurology, 66, S74–S78.
8. Cattaneo, E., Zuccato, C. and Tartari, M. (2005) Normal huntingtin
function: an alternative approach to Huntington’s disease. Nat. Rev.
Neurosci., 6, 919–930.
9. Colby, D.W., Garg, P., Holden, T., Chao, G., Webster, J.M., Messer, A.,
Ingram, V.M. and Wittrup, K.D. (2004) Development of a human light
chain variable domain (V(L)) intracellular antibody specific for the amino
terminus of huntingtin via yeast surface display. J. Mol. Biol., 342,
10. Heiser, V., Scherzinger, E., Boeddrich, A., Nordhoff, E., Lurz, R.,
Schugardt, N., Lehrach, H. and Wanker, E.E. (2000) Inhibition of
huntingtin fibrillogenesis by specific antibodies and small molecules:
implications for Huntington’s disease therapy. Proc. Natl Acad. Sci. USA,
11. Kazantsev, A., Walker, H.A., Slepko, N., Bear, J.E., Preisinger, E.,
Steffan, J.S., Zhu, Y.Z., Gertler, F.B., Housman, D.E., Marsh, J.L. et al.
(2002) A bivalent Huntingtin binding peptide suppresses polyglutamine
aggregation and pathogenesis in Drosophila. Nat. Genet., 30, 367–376.
2750Human Molecular Genetics, 2006, Vol. 15, No. 18
by guest on September 13, 2015
12. Nagai, Y., Tucker, T., Ren, H., Kenan, D.J., Henderson, B.S., Keene, J.D., Download full-text
Strittmatter, W.J. and Burke, J.R. (2000) Inhibition of polyglutamine
protein aggregation and cell death by novel peptides identified by phage
display screening. J. Biol. Chem., 275, 10437–10442.
Washington, S.L., Maxwell, M.M., Marsh, J.L., Thompson, L.M. et al.
(2005) A potent small molecule inhibits polyglutamine aggregation in
Huntington’s disease neurons and suppresses neurodegeneration in vivo.
Proc. Natl Acad. Sci. USA, 102, 892–897.
14. Heiser, V., Engemann, S., Brocker, W., Dunkel, I., Boeddrich, A.,
Waelter, S., Nordhoff, E., Lurz, R., Schugardt, N., Rautenberg, S. et al.
(2002) Identification of benzothiazoles as potential polyglutamine
aggregation inhibitors of Huntington’s disease by using an automated
filter retardation assay. Proc. Natl Acad. Sci. USA, 99(Suppl. 4),
15. Herbst, M. and Wanker, E. (2006) Therapeutic approaches to
polyglutamine diseases: combating protein misfolding and aggregation.
Curr. Pharm. Des., 12, 2543–2555.
16. Commenges, D., Scotet, V., Renaud, S., Jacqmin-Gadda, H.,
Barberger-Gateau, P. and Dartigues, J.F. (2000) Intake of flavonoids and
risk of dementia. Eur. J. Epidemiol., 16, 357–363.
17. Mandel, S.A., Avramovich-Tirosh, Y., Reznichenko, L., Zheng, H.,
Weinreb, O., Amit, T. and Youdim, M.B. (2005) Multifunctional
activities of green tea catechins in neuroprotection. Modulation of cell
survival genes, iron-dependent oxidative stress and PKC signaling
pathway. Neurosignals, 14, 46–60.
18. Davies, S.W., Turmaine, M., Cozens, B.A., DiFiglia, M., Sharp, A.H., Ross,
C.A., Scherzinger, E., Wanker, E.E., Mangiarini, L. and Bates, G.P. (1997)
Formation of neuronal intranuclear inclusions underlies the neurological
dysfunction in mice transgenic for the HD mutation. Cell, 90, 537–548.
19. Suganuma, M., Okabe, S., Oniyama, M., Tada, Y., Ito, H. and Fujiki, H.
(1998) Wide distribution of [3H](-)-epigallocatechin gallate, a cancer
preventive tea polyphenol, in mouse tissue. Carcinogenesis, 19,
20. Wacker, J.L., Zareie, M.H., Fong, H., Sarikaya, M. and Muchowski, P.J.
(2004) Hsp70 and Hsp40 attenuate formation of spherical and annular
polyglutamine oligomers by partitioning monomer. Nat. Struct. Mol.
Biol., 11, 1215–1222.
Sakahira, H., Siegers, K., Hayer-Hartl, M. and Hartl, F.U. (2004) Cellular
toxicity of polyglutamine expansion proteins: mechanism of transcription
factor deactivation. Mol. Cell., 15, 95–105.
22. Ko, J., Ou, S. and Patterson, P.H. (2001) New anti-huntingtin monoclonal
antibodies: implications for huntingtin conformation and its binding
proteins. Brain Res. Bull., 56, 319–329.
23. Meriin, A.B., Zhang, X., He, X., Newnam, G.P., Chernoff, Y.O. and
Sherman, M.Y. (2002) Huntington toxicity in yeast model depends on
polyglutamine aggregation mediated by a prion-like protein Rnq1. J. Cell
Biol., 157, 997–1004.
24. Wanker, E.E., Scherzinger, E., Heiser, V., Sittler, A., Eickhoff, H. and
Lehrach, H. (1999) Membrane filter assay for detection of amyloid-like
polyglutamine-containing protein aggregates. Meth. Enzymol., 309,
25. Agrawal, N., Pallos, J., Slepko, N., Apostol, B.L., Bodai, L., Chang, L.W.,
Chiang, A.S., Thompson, L.M. and Marsh, J.L. (2005) Identification of
combinatorial drug regimens for treatment of Huntington’s disease using
Drosophila. Proc. Natl Acad. Sci. USA, 102, 3777–3781.
26. Henning, S.M., Niu, Y., Lee, N.H., Thames, G.D., Minutti, R.R.,
Wang, H., Go, V.L. and Heber, D. (2004) Bioavailability and
antioxidant activity of tea flavanols after consumption of green tea,
black tea, or a green tea extract supplement. Am. J. Clin. Nutr., 80,
27. Poirier, M.A., Li, H., Macosko, J., Cai, S., Amzel, M. and Ross, C.A.
(2002) Huntingtin spheroids and protofibrils as precursors in
polyglutamine fibrillization. J. Biol. Chem., 277, 41032–41037.
28. Wanker, E.E. (2000) Protein aggregation and pathogenesis of
Huntington’s disease: mechanisms and correlations. Biol. Chem., 381,
29. Murray, N.J., Williamson, M.P., Lilley, T.H. and Haslam, E. (1994) Study
of the interaction between salivary proline-rich proteins and a polyphenol
by1H-NMR spectroscopy. Eur. J. Biochem., 219, 923–935.
30. Sacchettini, J.C. and Kelly, J.W. (2002) Therapeutic strategies for human
amyloid diseases. Nat. Rev. Drug Discov., 1, 267–275.
31. Muchowski,P.J.,Schaffar,G., Sittler,A.,Wanker,E.E.,Hayer-Hartl,M.K.
and Hartl, F.U. (2000) Hsp70 and hsp40 chaperones can inhibit
self-assembly of polyglutamine proteins into amyloid-like fibrils. Proc.
Natl Acad. Sci. USA, 97, 7841–7846.
32. Legleiter, J., DeMattos, R.B., Holtzman, D.M. and Kowalewski,
T. (2004) In situ AFM studies of astrocyte-secreted apolipoprotein
E- and J-containing lipoproteins. J. Colloid Interface Sci., 278,
Human Molecular Genetics, 2006, Vol. 15, No. 18 2751
by guest on September 13, 2015