(Received for publication, May 17, 1996, and in revised form, September 23, 1996)
From the Department of Psychiatry and Behavioral Science, School of Medicine, State University of New York, Stony Brook, New York 11794 and the § Biology Department, Brookhaven National Laboratory, Upton, New York 11973
Neurodegenerative disorders of aging are
characterized by the intraneuronal accumulation of ubiquitin conjugates
into tangles and inclusions. Ubiquitin conjugates are degraded by
cellular particles known as proteasomes. We have previously shown that amyloid protein (A
) inhibits proteasomal activity and thereby blocks ubiquitin conjugate degradation. In the present studies, we
found that A
binds the 20 S proteasome and forms a proteasome-A
complex. The complex was detected by Western blot with anti-A
antibodies. Using a 1.4 nm Nanogold-labeled A
, we visualized proteasome-A
complexes by scanning transmission electron microscopy (STEM). Analysis of the side-on oriented proteasome-A
complexes revealed a single gold particle, corresponding to one gold-labeled A
, in the middle portion of the proteasome. On end-on views of proteasome-A
complexes, gold was detected within the area delimited by the proteasome circular projection. Both STEM views are consistent with A
localization inside the proteasome along the peptide channel. Direct interaction of A
with the inner catalytic compartment of the
proteasome may explain the generation of ubiquitin-containing lesions
in Alzheimer's disease and other neurodegenerative disorders. In
addition, detection of Nanogold-labeled peptide inside the 20 S
eukaryotic proteasome suggests that conformational constraints for
protein degradation in eukaryotic proteasomes are different from those
in archaebacteria proteasomes.
The ubiquitin/proteasome protein degradation pathway is one of the
two major cellular proteolytic systems. In this pathway, damaged and
abnormal proteins are targeted for degradation by the covalent
attachment of several ubiquitin moieties (ubiquitination) to form
ubiquitin conjugates. Ubiquitinated proteins are rapidly degraded by
the 26 S proteasome, the proteolytic component of the ubiquitin
degradation system (1, 2, 3). Several findings link the
ubiquitin/proteasome pathway to some pathological manifestations of
neurodegenerative disorders such as Alzheimer's disease
(AD),1 Lewy bodies disease, and to a lesser
extent normal aging of the brain, as well as pathologies of non-central
nervous system-related diseases such as inclusion body myositis (IBM).
In these disorders, high levels of ubiquitin and ubiquitin conjugates
are detected in abnormal intracellular inclusions and
cytoskeletal-derived fibrils (4, 5, 6). In AD, microtubule associated
protein is ubiquitinated but is not degraded. Ubiquitinated
is
found in paired helical filaments forming neurofibrillary tangles (NFT) (7). Development of NFT has been linked directly to neuronal degeneration in AD (8). Other, yet unidentified, proteins also accumulate into intracellular inclusions. These observations suggest that the ubiquitin/proteasome proteolytic pathway may be involved in
the formation of ubiquitinated intracellular lesions.
Proteasomes are abundant nonlysosomal multicatalytic proteinases
involved in a variety of cell functions (1, 9, 10, 11, 12). Proteasomes are
localized both in the cytoplasm and in the nucleus of cells (10). In
addition, proteasomes have been detected associated with plasma and
internal membranes (10). The 20 S proteasome is the catalytic core of
the 26 S proteasome (9, 10, 11). An increasing body of evidence points to a role of proteasomes in the accumulation of ubiquitinated proteins and
ubiquitin-containing lesions in neurodegenerative disorders. Antibodies
to proteasomes decorate some Lewy bodies and NFT (13). Cultured cells
exposed to proteasome inhibitors accumulate ubiquitin conjugates (14,
15) and in at least one neuronal cell line these ubiquitin conjugates
accumulated into intracellular AD-like inclusion
bodies.2 In addition, we found that
in vitro amyloid protein (A
) blocked the degradation
of ubiquitinated proteins by inhibiting the proteasome activity (16).
A
is a 39-42-amino acid peptide, derived from the proteolytic
cleavage of the membrane-bound amyloid
protein precursor, which has
been implicated in AD (17). Intracellular A
has been detected in
cultured cells (18, 19) and associated with NFT in AD (20, 21) and with
cytoplasmic tubulofilaments of vacuolated muscle fibers in IBM (5). We
showed that A
selectively inhibits the chymotrypsin-like activity in
the proteasome (16). Among the proteasome's different proteolytic
activities, the chymotrypsin-like activity has been linked directly to
the degradation of ubiquitin conjugates. Defects in the yeast
proteasome subunits bearing this catalytic specificity cause
accumulation of ubiquitin conjugates and reduced protein degradation
rates (22). Selective inhibition of the proteasome chymotrypsin-like
activity by A
is consistent with a critical function of the
proteasome-peptide interaction in the accumulation of ubiquitin
conjugates in AD. Furthermore, a recent finding that A
, used as the
bait in a yeast two-hybrid system, binds two distinct yeast proteasome
subunits, supports our results on the ability of A
to interact with
the proteasome (23).
In this work, we characterized the 20 S proteasome-A complex and
established the spatial localization of A
in the complex. The
results provide evidence for a molecular mechanism of A
interaction with the proteasome, which may be involved in the accumulation of
ubiquitin conjugates in neurodegenerative disorders and other age-related diseases.
Purified proteasomes were provided by Dr. M. Figueiredo-Pereira (Mount Sinai, CUNY, New York).
A1-39C40 was a gift of Dr. C. Gable (University of
California, Irvine). A
1-40 (A
) was purchased from
K-Biologicals (Rancho Cucamonga, CA). The peptides were dissolved in
water at the final concentration of 200 µM and stored in
aliquots at
80 °C. Before use, the peptide solutions were
microcentrifuged for 10 min. 6E10 is a monoclonal antibody generated
against the sequence 1-16 of A
(24). Nanogold, silver enhancement
kit, and methylamine vanadate were purchased from Nanoprobes (Stony
Brook, NY). All other proteins and reagents were obtained from
Sigma. Protein concentration was determined using BCA
(Pierce) with bovine serum albumin as the standard.
Purified 20 S proteasomes (2.3 × 107
M) in Tris-HCl, 20 mM, pH 7.4, were incubated
with A
at the indicated final peptide concentrations for 30 min at
room temperature. Samples were analyzed on 5% nondissociating polyacrylamide gel electrophoresis (PAGE) and transferred onto nitrocellulose membrane using Tris/glycine/SDS containing 20% methanol
as the transferring buffer. The membrane was probed with 6E10
antibodies, and the membrane-bound antibodies were detected by enhanced
chemiluminescence (Amersham). Quantitation of immunostained bands was
obtained from a phosphoanalyzer (Bio-Rad Laboratories) image using
screen-CH plate. Following immunodetection, proteins on the membrane
were stained with 0.1% Amido Black and destained with 30% methanol,
10% acetic acid.
In
order to covalently and selectively attach monomaleimido-Nanogold
particle to the peptide, we used a synthetic A1-40 variant in which the last amino acid was substituted with a cysteine residue (A
1-39C40). Ten-fold molar excess
A
1-39C40 over the gold particle was used as the
coupling conditions. The samples were incubated for 1 h at room
temperature according to the manufacturer's instructions. Samples were
chromatographed on Superose 12 (FPLC, Pharmacia) to separate the
product A
1-39C40-Nanogold (A
Au) from
unconjugated Nanogold and A
1-39C40. A
Au
eluted from gel filtration was analyzed on 14% Tris/Tricine PAGE (25)
and proteins were transferred onto PVDF membrane. Membrane-bound A
Au was detected by immunostaining with 6E10 antibodies
and by silver enhancement of gold. Silver enhancement staining does not
react with proteins.
The following procedures were performed at room
temperature. Purified 20 S proteasomes (2.3 × 107
M) were incubated with A
Au
(10
5 M) for 30 min to form
proteasome-A
Au complex. As the control, proteasomes were
incubated with unconjugated Nanogold. The complex was cross-linked
using 0.1% glutaraldehyde for 25 min, and the reaction was stopped by
the addition of 0.2 M glycine, pH 7.3, for 1 h. The
sample was concentrated to the original volume by ultrafiltration on
Microcon-100 (Amicon) and used for STEM analysis. Cross-linked
proteasome-A
Au complex was subjected to 5%
nondissociating PAGE. The proteasome was detected by Coomassie Blue
staining and in a parallel experiment, A
Au was detected
by the silver enhancement method.
Samples for STEM analysis were prepared by applying 5 µl of cross-linked proteasome-AAu complex to a
titanium grid with a holey film and 3 nm carbon film on it (26). After
1-min absorption time, the grids were washed, wicked, quick-frozen in
liquid nitrogen slush, and freeze-dried slowly overnight. Samples were
negatively stained in a low atomic number stain, methylamine vanadate,
which is denser than proteins but less dense than the gold
particle.
A and 20 S
eukaryotic proteasomes were incubated together to allow the formation
of a proteasome-A
complex. The complex was analyzed by
nondissociating PAGE (Fig. 1) and characterized by
Western blotting using 6E10 antibodies to detect A
(Fig.
1A) and by protein staining of the same membrane to detect
proteasomes (Fig. 1B). Upon incubation of A
with the
proteasome, anti-A
antibodies recognized a new band (Fig.
1A, lane 2) with a slower electrophoretic
mobility than free A
(Fig. 1A, lane 1). This immunoreactive band was not detected when A
was omitted from the
reaction (Fig. 1A, lane 3), indicating that
proteasomes alone did not form the new band and that there was no
cross-reactivity of anti-A
antibodies with proteasomes. Furthermore,
the same band was not formed when A
was incubated without the
proteasome (Fig. 1A, lane 1) excluding the
possibility that this complex is due to aggregated A
formed during
incubation. This new band was detected by Coomassie staining of the
membrane (Fig. 1B, lane 2) indicating that it
contained the proteasomes, thus confirming it was a proteasome-A
complex. Under these experimental conditions, A
was not degraded by
the proteasome since the intensity of the A
band as revealed by 6E10
antibody or by protein staining showed no difference between peptide
incubated in the presence (lanes 2, Fig. 1, A and
B) or in the absence (lanes 1, Fig. 1,
A and B) of proteasomes. This observation
confirms our previous finding that, under our experimental conditions,
the peptide is not degraded by the proteasome (16). However, our
results do not exclude that after a prolonged exposure to proteasomes a
certain level of A
degradation may occur.
We evaluated the formation of the proteasome-A complex at various
concentrations of the peptide (Fig. 1C). The intensity of
proteasome-A
complex bands was quantitated and plotted against peptide concentration (Fig. 1D). The curve in
Fig. 1D shows that even at the highest A
concentration
tested, no saturation of binding was achieved. This phenomenon could be
due to the tendency of A
to rapidly aggregate when used at
concentrations higher than 200 µM. Furthermore, even when
in a soluble state, A
is present as a mixture of monomer, dimer,
tetramer, and higher multimeric forms (27) which preclude any type of
standard kinetic analysis. An approximate extrapolation of the curve in
Fig. 1D for saturation conditions indicated that 50%
binding of A
to the proteasome was obtained at a micromolar
concentration of A
. This range is consistent with the reported
IC50 for A
inhibition of the proteasome chymotrypsin-like activity of 3-5 µM (16).
In
order to establish the spatial localization of A within the
proteasome-A
complex, we labeled an A
variant containing a
cysteine residue as the last amino acid (A
1-39C40) with 1.4 nm Nanogold which could be detected by electron microscopy. Formation of gold-labeled A
1-39C40-Nanogold
(A
Au) was confirmed by Western blotting using 6E10
antibodies to detect A
1-39C40 (Fig. 2A,
left panel) and by silver enhancement staining to visualize the gold particle (Fig. 2A,
right panel). The same band was detected by both staining
methods indicating that the 17-kDa band was gold-labeled
A
1-39C40 (Fig. 2A, lane 2). As
the controls, unconjugated A
1-39C40 and free Nanogold
were analyzed and their migration patterns are shown in Fig.
2A, lanes 1 and 3, respectively. In
both cases, the 17-kDa band was not detected.
In order to confirm that AAu maintained the ability to
bind the proteasomes, A
Au was incubated with the
proteasome to form a proteasome-A
Au complex. Formation
of the complex was analyzed by nondissociating PAGE. Proteasomes were
detected by Coomassie Blue staining (Fig. 2B, lane
3) while gold-labeled A
Au in the complex was
visualized using silver enhancement (Fig. 2C, lane
1). Specificity of labeling is demonstrated by the absence of the
proteasome-A
Au complex band as detected by silver
enhancement staining when proteasomes were cross-linked to unconjugated
Nanogold (Fig. 2C, lane 3). The upper band was
also not formed when A
Au was cross-linked in the absence
of proteasomes (Fig. 2C, lane 2). The results in
Figs. 1 and 2 indicate that A
, as well as gold-labeled A
, forms
complexes with proteasomes. Furthermore, A
Au was able to
inhibit ubiquitin-dependent degradation of lysozyme with
only a slight increase of the IC50 when compared with
A
1-40 (Fig. 3). As already discussed, no
further kinetic analysis could be performed due to the anomalous
behavior of A
in solution. Our result indicates that
A
Au, similar to unlabeled A
(16), interacted with the
26 S proteasome and thereby inhibited its proteolytic activity. Since
the 20 S proteasome is the catalytic component of the larger proteasome and contains the chymotrypsin-like activity previously shown to be
inhibited by A
, we concluded that at least one of the interactions of A
Au with the 26 S proteasome is with its 20 S
proteasome component. We analyzed this interaction and A
localization within the proteasome-A
Au complex by
electron microscopy. Our observations suggest that the A
conformation required for proteasome-peptide interaction was not
changed by the presence of the 1.4 nm Nanogold particle at the carboxyl
terminus of A
. However, based on these experiments, we cannot
exclude that coupling of Nanogold to A
may change the peptide
binding affinity for the proteasome.
STEM Analysis of Proteasome-A
Eukaryotic proteasomes are composed of 15-20 different
subunits, assembled to form four staked rings in a barrel-shaped
structure (10). The regulatory subunits (-type) are located in the
outer rings and the catalytic subunits (
-type) are in the two inner rings with the active sites oriented toward the proteasome internal cavity (9, 28, 29). Proteasomes on electron microscopy are seen in two
orientations, either end-on displaying ring-like projections or side-on
producing striated rectangular views (29). STEM analysis of
proteasome-A
Au complexes is shown in Fig.
4. The arrows in Fig. 4A indicate gold particles inside proteasomes. In end-on views (Fig.
4B), we found that the gold particle was in or close to the
center of the proteasome circular projection. In side-on views (Fig. 4C), we observed that A
Au was located in the
middle portion of the four-ring structure. Since each gold particle is
linked to a single peptide molecule, our results indicate a predominant
one-to-one ratio of proteasome and A
in the complex. More than 80%
of the labeled proteasomes had one A
Au cross-linked,
whereas the remaining carried two A
Au molecules (last
image in Fig. 4, B and C). No cross-linking of unconjugated Nanogold to proteasomes was observed in our control experiments (data not shown). The results from multiple samples of STEM
images are schematically summarized in Fig. 4D.
A
Au was predominantly detected inside the structure,
within the area along the peptide channel which runs through the
proteolytic structure and is hypothesized to be the route of substrate
access into the proteasome for degradation (9, 28, 29). As Fig.
4D indicates, A
did not randomly bind the proteasome, but
there appeared to be two well defined regions of preferential binding.
One between the second and the third ring and the other between the
first and the second ring. Although at present, we cannot determine which specific subunits were involved in the interactions, our findings
suggest that at least one interaction occurred with
-type subunits
in the inner rings, presumably the subunit with chymotrypsin-like specificity (16). A
-proteasome complex formation was not abolished by the addition of 100-fold excess of the peptide
Z-Leu-Leu-Leu-p-nitroanilide (data not shown) which is the
substrate used to measure the chymotrypsin-like activity (16). These
results suggest that A
inhibited the proteasome activity not through
a direct binding to the catalytic site(s), but via an indirect
modification of the same catalytic site(s).
The current mechanistic model of proteasome peptide degradation
proposes that peptides are directed into proteasomes and degraded in
the inner cavity, where the active sites of the catalytic subunits are
located (28, 29). Recent work by Wenzel and Baumeister (30) supported
this model and suggested that only unfolded peptides are able to slide
into the peptide channel which functions by size exclusion. By using
the same Nanogold reagent we used in our study, they found that the
gold particle attached to oxidized insulin B chain was stuck at the
entrance of the peptide channel and could not pass the narrow orifice.
Our results show that the same gold particle attached to A can enter
the proteasome. This difference could be due to the fact that we used
eukaryotic proteasomes whereas in Baumeister's work, archaebacteria
proteasomes were used. Although the two proteasome structures are very
similar, the dimension of the peptide channels might be different (12). In addition to a structural difference, archaebacteria proteasomes are
also functionally different since they possess only chymotrypsin-like activity (10, 28, 29), while the more complex eukaryotic proteasomes
comprise several distinct proteolytic specificities (9). Thus, our
findings suggest that the substrates' conformational constraints for
degradation by eukaryotic proteasome may be less stringent than
previously reported for archaebacteria proteasomes.
In the degradation of ubiquitin conjugates, the 20 S proteasome
requires the addition of other subunits which assemble into the 26 S
proteasome. However, within the 26 S proteasome, the 20 S component
remains the site with proteolytic activity. Thus, A interaction with
and inhibition of the 20 S proteasome bears direct connection with the
inhibitory effect of A
on the 26 S proteasome-dependent
degradation of ubiquitinated proteins. Consistent with this association
are the findings that cultured cells exposed to inhibitors of the 20 S
proteasome (14, 15) or mutations of its proteolytic subunits (22)
caused an accumulation of ubiquitin conjugates. However, our findings
do not exclude the possibility that A
may interact with the 26 S
proteasome at other sites in addition to those of the 20 S proteasome
reported here.
Understanding proteasome-A interaction is critical to establishing
its role in the accumulation of ubiquitin conjugates and the formation
of NFT as seen in AD and other neurodegenerative disorders. Our work
proposes a molecular mechanism of proteasome-A
interaction.
Moreover, the present findings provide evidence that the conformational
requirements in peptide degradation by eukaryotic proteasomes are
distinct from those of archaebacteria proteasomes.
We are particularly grateful to Dr. Maria Figueiredo-Pereira and Dr. Charles Glabe for their very generous gifts. We would also like to thank Dr. Chana Fuchs for her advice and suggestions during the preparation of this manuscript and Dr. Amit Banerjee and Dr. Roy MacCauley for reviewing this manuscript.