BBRC
Biochemical and Biophysical Research Communications 305 (2003) 709 718
www.elsevier.com/locate/ybbrc
Selective degradation of oxidatively modified protein substrates
by the proteasome
Tilman Grune,a Katrin Merker,a Grit Sandig,a and Kelvin J.A. Daviesb,*
a
Neuroscience Research Center, Medical Faculty (Charite) Humboldt University Berlin, Schumannstr. 20/21, 10117 Berlin, Germany
e
b
Ethel Percy Andrus Gerontology Center, and Division of Molecular and Computational Biology, University of Southern California,
Los Angeles, CA 90089-0191, USA
Received 9 March 2003
Abstract
Oxidative stress in mammalian cells is an inevitable consequence of their aerobic metabolism. Oxidants produce modifications to
proteins leading to loss of function (or gain of undesirable function) and very often to an enhanced degradation of the oxidized
proteins. For several years it has been known that the proteasome is involved in the degradation of oxidized proteins. This review
summarizes our knowledge about the recognition of oxidized protein substrates by the proteasome in in vitro systems and its
applicability to living cells. The majority of studies in the field agree that the degradation of mildly oxidized proteins is an important
function of the proteasomal system. The major recognition motif of the substrates seems to be hydrophobic surface patches that are
recognized by the 20S ÔcoreÕ proteasome. Such hydrophobic surface patches are formed by partial unfolding and exposure of hy-
drophobic amino acid residues during oxidation. Oxidized proteins appear to be relatively poor substrates for ubiquitination, and
the ubiquitination system does not seem to be involved in the recognition or targeting of oxidized proteins. Heavily oxidized proteins
appear to first aggregate (new hydrophobic and ionic bonds) and then to form covalent cross-links that make them highly resistant
to proteolysis. The inability to degrade extensively oxidized proteins may contribute to the accumulation of protein aggregates
during diseases and the aging process.
Ó 2003 Elsevier Science (USA). All rights reserved.
Keywords: Protein oxidation; Proteolysis; Proteasome; Lysosomes; Calpains; Free radicals
Life in an oxygen environment inevitably involves the tained with purified components of the proteasomal
production of free radicals and other oxidants. One of system, with cell lysates, and with intact cells of various
the consequences of this process is the continuous for- lineages [1 31].
mation of oxidatively modified proteins, both within It has been concluded that mild oxidation of globu-
cells and in extracellular fluids. Since modified proteins lar, soluble proteins enhances their proteolytic suscep-
often experience significant loss of function (or gain of tibility and makes them targets of the proteasomal
undesirable function) they have to be replaced by the system [23,32,33]. An increasing body of literature in-
cellular protein synthesis machinery. To avoid excessive dicates that mildly oxidized proteins are readily de-
accumulation of damaged proteins, such non-functional graded, whereas severe oxidation stabilizes proteins due
oxidized proteins have to be removed by proteolytic to aggregation, cross-linking, and/or decreased solubil-
systems. Substantial evidence, accumulated over the ity, thus increasing their half-lives [21 24,27 31]. The
past 20 years, strongly suggests that the proteasomal inability to degrade extensively oxidized proteins may
system is responsible for degrading oxidized proteins in contribute to certain disease states, including various
the cytoplasm, nucleus, and endoplasmic reticulum of neurodegenerative diseases such as AlzheimerÕs disease
eucaryotic cells [1 31]. These findings have been ob- [34], or ParkinsonÕs disease [35], and could also play a
significant role in the aging process [36 38].
*
The aim of this review is to summarize the current
Corresponding author. Fax: [phone].
E-mail address: [e-mail] (K.J.A. Davies). knowledge about the protein substrates that have been
0006-291X/03/$ - see front matter Ó 2003 Elsevier Science (USA). All rights reserved.
doi:10.1016/S0006-291X(03)00809-X
710 T. Grune et al. / Biochemical and Biophysical Research Communications 305 (2003) 709 718
used in studies of protein oxidation and proteolysis. In thoroughly investigated cross-links is the formation of a
particular, we have focused our attention on those 2,20-biphenyl bond between two tyrosyl radicals, to form
substrates that have been used to test the selectivity of dityrosine or bityrosine [16]. During the process of co-
the proteasome for the oxidized forms of intracellular valent cross-linking of protein aggregates, non-protein
proteins. components such as carbohydrates and oxidized lipids
can also react and add to the growing oxidized mass of
material [29,42,43]. It is now clear that these protein
Oxidative protein modification aggregates are poor substrates for proteases, which re-
sults in their accumulation within cells [15 17,23,29,32,
The degree of protein oxidation caused by a given 38,41].
oxidant depends on many factors, including the nature,
relative location, and flux rate of the oxidant, and the
presence (or absence) of antioxidants. Both the oxida- Accumulation of cross-linked proteins
tion of free amino acids and the oxidation of peptides
and proteins have been studied by many laboratories, Several diseases, and aging processes, are accompa-
and numerous amino acids are known to be susceptible nied by the accumulation of cross-linked proteins. This
to oxidation [4 6,11,16,18,39,40]. Although chemical accumulation of oxidized protein aggregates can occur
reactions occur during amino acid side chain oxidation, both extracellularly, and within various cellular com-
the products of these processes differ only minimally in partments. Differences in the effects of protein aggre-
molecular weight from those of the original amino acid gates on various cellular or organismal functions may be
residues. On the other hand fragmentation of polypep- expected, depending on the rate of formation and the
tide backbones can occur, leading to the formation of exact location of such aggregates. In several cases ag-
protein fragments which are not true peptides (these are gregated/cross-linked material will be autophagocyto-
formed during proteolytic degradation of polypeptides), sed, resulting in a major accumulation of the material in
but protein fragments with derivatized terminal amino lysosomes [44,45].
acids [4 6,11,41]. Depending on the conditions of the The accumulation of oxidized proteins can result
oxidation and the oxidant itself, these fragments will from several kinds of malfunctions of cellular metabo-
vary in length and in rate of formation. lism. As demonstrated in Fig. 1 the accumulation of a
Protein aggregates can also form during free radical protein can be the result of a genetically determined
reactions [5,6,23,38,41]. It is suggested that the formation lower proteolytic susceptibility, or an innate proneness
of protein aggregates occurs initially on a non-covalent to aggregation (Fig. 1, pathway 1). The oxidation of
basis, largely involving such forces as hydrophobic protein aggregates in this case is most likely the result of
bonds and electrostatic interactions. Subsequently, these the prolonged half-life of such proteins. A prominent
aggregates tend to form covalent cross-links due to re- example of such a mutation is the Huntington disease,
actions between carbon-, oxygen-, and nitrogen-centered where the mutated Huntigtin protein has a prolonged
radicals of amino acid side chains. One of the most poly-Glu-strech that causes aggregation. Under other
Fig. 1. Possible mechanisms for the accumulation of oxidized proteins. If a given protein accumulates, the corresponding gene might be mutated
(pathway 1), the protein might undergo excessive or artificial posttranslational modification (pathway 2) or the balance between synthesis and
degradation might be disturbed due to poor metabolic regulation (pathway 3). The accumulating (non-degraded) protein might in all cases undergo
secondary oxidation.
T. Grune et al. / Biochemical and Biophysical Research Communications 305 (2003) 709 718 711
conditions, an increase in posttranslational modification
can be the reason for the formation of protein aggre-
gates (Fig. 1, pathway 2). Since posttranslational mod-
ifications change the proteolytic susceptibility of
substrates, some proteins will experience decreased
proteolysis. Alternatively, a high formation rate of
posttranslationally modified proteins might simply
overwhelm the capacity of the proteolytic systems. Ex-
amples for posttranslational modifications changing the
proteolytic susceptibility of substrates include oxidative
modifications [1 31] and phosphorylation of the tau
protein [46]. Furthermore an imbalance between protein
synthesis and proteolytic capacity might be the reason
for an accumulation of cross-linked proteins (Fig. 1,
pathway 3). Examples are the ceroid lipofuscinoses with
their decline in proteolytic capacity [47,48].
With time such aggregated proteins are transferred
into a highly cross-linked and reactive material, referred
to as lipofuscin, ceroid or AGE-pigment-like fluoro-
phores by several authors [49,50]. The involvement of
Fig. 2. Proteolytic susceptibility of oxidized proteins. Due to increasing
free radicals as one of the initial steps in the formation
oxidation, the proteolytic susceptibility of the substrate increases to a
of fluorescent oxidized/cross-linked aggregates has been
certain point (Ôoptimal oxidant concentrationÕ) but at higher oxidant
postulated [51,52].
concentrations begins to decline. Initially, oxidation causes charge
rearrangements within the protein substrate, with random refolding
and partial exposure of internal hydrophobic amino acid Ôpatches.Õ
These hydrophobic patches bind to the core 20S proteasome and en-
Degradation of oxidized proteins
able the protein substrate to be fully degraded. With further oxidation,
however, the surface hydrophobicity of the substrate protein continues
Following exposure to oxidants one can detect
to increase, due to further oxidation-induced unfolding, and these
changes in the proteolytic susceptibility of a number of hydrophobic patches attract each other (or are repelled by the aqueous
environment) so quickly that proteasome has difficulty competing, and
protein substrates (Fig. 2). This change in proteolytic
aggregation becomes a significant problem. Over time, new covalent
susceptibility has a biphasic response. At moderate
bonds and ionic bonds form between oxidized amino acid residues in
oxidant concentrations proteolytic susceptibility in-
the aggregates, making them highly resistant to any form of proteo-
creases, whereas at higher oxidant concentrations a de-
lysis. For further information see text.
crease (sometimes even below the Ôbasal degradationÕ
level) in proteolytic susceptibility occurs (see Fig. 2).
Between these extremes, the oxidant reaches an Ôoptimal tration or dosage of the oxidants reported in Tables 1
concentrationÕ characterized by a Ômaximal degradationÕ and 2 varies widely, due to the different reactivity and
rate for the given conditions. The increase in degrada- stability of each oxidizing agent, an increase in proteo-
tion or the Ôproteolytic stimulationÕ is the ratio of the lytic susceptibility could always be achieved at an Ôop-
Ômaximal degradationÕ to the Ôbasal degradation.Õ timal oxidant exposure.Õ Naturally, the increase in
As one can see in Table 1, a large number of varied proteolytic susceptibility depends on the oxidizing
proteins have been used as substrates to test for oxidant- agent, the protein substrate, and the exact experimental
induced proteolytic stimulation. The phenomenon seems conditions (Table 2). However, the increase in proteo-
to be a common feature of all globular, soluble proteins lytic stimulation differs also as a function of the nature
with defined secondary and tertiary structures. The or- and the source of the proteolytic enzyme(s) employed
igin of the protein, however, whether extracellular, cy- (Tables 3 and 4). Furthermore, several reports have
demonstrated that increased proteolytic susceptibility is
tosolic, nuclear, native or recombinant, seems not to
have any importance. Essentially ÔstructurelessÕ proteins, limited to a certain oxidant concentration range, which
such as casein, are inherently good substrates for pro- is often very tight [53], implicating the possibility of
teolysis and their susceptibility is not increased by mild missing this Ôoptimal concentrationÕ range in any ex-
oxidation; although it can be decreased by heavy oxi- perimental setup. This reveals one of the potential dif-
dation. ficulties in measuring the in vitro effects of oxidants on
In addition to numerous proteolytic substrates, vari- the proteolytic susceptibility of a given protein sub-
ous oxidizing systems, employing either bolus treat- strate.
ments or fluxes of oxidants, have been used to oxidize On the other hand a number of different sources for
many proteins (Tables 1 and 2). Although the concen- the proteasomal system have been employed by various
Table 1
Overview of the most common oxidized protein substrates used for in vitro degradation assays
Protein Oxidant Proteasome source Reference
Hemoglobin H2O2, phenylhydrazine Rabbit erythrocytes and reticulocytes, human erythrocytes, isolated [5,13,15 18,21,22,71 74]
proteasome (20S and 26S), and C9 and K562 cells
Superoxide dismutase O 2 , H2O2, OH, SIN-1 Rabbit reticulocytes and erythrocytes, human erythrocytes and [5,9,13,21,24,71,75]
reticulocytes, bovine erythrocytes, isolated proteasome,
and C9 and RAW264 cells
Laminin H2O2 Rabbit erythrocytes and reticulocytes, human erythrocytes, and [76]
primary microglia
Bovine serum albumin OH, phenylhydrazine O 2 , H2O2 Rabbit reticulocytes and erythrocytes, human erythrocytes, isolated [3,5,13,14,74,77 79]
proteasome, and mice kidney cells
a-Casein OH, AGE, CML, H2O2 Rabbit erythrocytes and reticulocytes, human erythrocytes, isolated [3,5,13,80]
proteasome, and K562 cells
Aconitase H2O2, SNAP, ONOO , SIN-1 Isolated proteasome [24]
Myosin, ovalbumin, collagenase, SIN-1 Isolated proteasome [24]
and carbonic anhydrase
Myoglobin SIN-1, H2O2 Isolated 20S proteasome [24,56]
Ferritin SIN-1, ONOO , NDPO2, rose Rabbit erythrocytes and reticulocytes, human erythrocytes, isolated [24,53,65,81 83]
bengal + light, DEA-NO, X/XO, proteasome, K562, CH E36, and ts20 E1 mutant cells
H2O2
Catalase H2O2, SIN-1 Isolated proteasome [5,24]
Histones 1, 2A, 2B, 3, and 4 H2O2 K562, C9 cells, isolated proteasome [27,28]
Protein disulfide isomerase H2O2 C9 cells [84]
Ezrin H2O2 Primary mice liver cells [85]
Lysozyme FeCl3/Ascorbate H2O2 CH E36 and ts20 E1 mutant cells, isolated proteasome [65,86]
Glutamine synthetase O 2 , FeCl3/ascorbate, HNE, Rat liver [1,2,42,43,58,87,88]
FeSO4/citrate
G6PDH HNE, FeSO4/Citrat, AGE, CML, Isolated 20S proteasome [42,43,53,89,90]
H2O2
Calmodulin H2O2 HeLa cells, isolated 26S proteasome [90,91]
60
a, bL, bH, c-Crystallin Co-irradiation, OH, H2O2, UV Isolated proteasome, bovine lens epithelial cells ź BLEC, [20,92 95]
rabbit erythrocytes, and mice macrophages
Apolipoprotein B (human) CuCl2 B lymphoblastoid cell line ź LCL [96]
Tetanus toxin OH RAW264 cells [97]
Myelin basic protein ź MBP H2O2 Primary microglia [76]
712
T. Grune et al. / Biochemical and Biophysical Research Communications 305 (2003) 709 718
T. Grune et al. / Biochemical and Biophysical Research Communications 305 (2003) 709 718 713
Table 2
Increased proteolytic susceptibility of ferritin after treatment with various oxidants
Oxidant Optimal oxidant concentration Proteolytic stimulation References
SIN-1 2 mM 12.6-fold [24]
SIN-1 5 lmol mg protein 1 4-fold [82]
ONOO 100 lM 1.9-fold [81]
NDPO2 50 lM 2.9-fold [81]
DEA-NO 1 lmol mg protein 1 2.5-fold [82]
X/XO 0.75 nmol mg protein 1 4.5-fold [81]
H2O2 10 lmol mg protein 1 5.5-fold [83]
H2O2 15 mM 3.7-fold [65]
H2O2 10 lmol mg protein 1 3-fold [81]
investigators, using diverse sources for crude cell lysates, phages, tumor cells, and Escherichia coli cells [1,2,9,14
purified cell lysates or isolated proteasomes (see Tables 16,21,22,24,30,31,54 57] are able to selectively degrade
1 4). Since in a large number of these experiments the oxidatively modified proteins. What forms the recogni-
species of the substrate protein in question and the tion motif of oxidized proteins for the proteasome is one
source of the proteasome do not match, one can assume of the key research questions surrounding the fate of
a general, species overlapping mechanism for the rec- oxidized proteins. The selective oxidation of several
ognition of oxidized proteins. Therefore, it can be con- amino acids, and their resulting products, has to be
cluded that the removal of   minimally  oxidized taken into account as possible recognition markers.
proteins is an essential function for maintaining cellular Levine et al. [58], for example, could clearly demon-
homeostasis by preventing the accumulation of highly strate an increase in the proteolysis of glutamine syn-
oxidized and cross-linked proteins, which are no longer thetase after oxidizing a threshold level of methionine
degradable and which may threaten cellular/organismal residues. Lasch et al. [59] found a clear correlation be-
viability. tween tyrosine oxidation and proteasomal degradation
of RNase A. Numerous other examples of single amino
acid changes that correlate with altered proteolytic
Recognition of the oxidized protein substrates by the susceptibility can be found in the literature. Although
proteasome such findings are important, one has to be aware that
many protein oxidation processes correlate with the
It has been shown that erythrocytes and reticulocytes oxidation of several amino acids and with changes in
from rabbits, cows, and human beings, as well as rat the secondary, tertiary, and even quaternary structures
muscles in vitro, rat hepatocytes, fibroblasts, macro- of substrate proteins. Furthermore, as one can see in
Table 3
Increased proteolytic susceptibility of hemoglobin after treatment with the hydroxyl radical under various experimental conditions
Proteasome source Optimal OH concentration Proteolytic stimulation References
Rabbit erythrocytes 20nmol OH/nmol protein 9-fold [13]
Rabbit reticulocytes 20nmol OH/nmol protein 10-fold [13]
Isolated proteasome 20nmol OH/nmol protein 5.5-fold [15]
Human erythrocytes 25 nmol OH/nmol protein 7.5-fold [15]
Isolated proteasome 20nmol OH/nmol protein 18-fold [71]
Rabbit erythrocytes 150nmol OH/0.33 mg protein 2.6-fold [5]
Rabbit reticulocytes 150nmol OH/0.33 mg protein 12.9-fold [5]
Table 4
Increase in proteolytic susceptibility of superoxide dismutase after treatment with hydrogen peroxide under various experimental conditions
Proteasome source Optimal H2O2 concentration (mM) Proteolytic stimulation References
Isolated proteasome 1028-fold [71]
C9 cells 15 5.4-fold [22]
Rabbit erythrocytes 304-fold [75]
Bovine erythrocytes 302.5-fold [75]
Bovine erythrocytes 303.4-fold [9]
Rabbit erythrocytes 3013-fold [13]
Rabbit reticulocytes 3012-fold [13]
714 T. Grune et al. / Biochemical and Biophysical Research Communications 305 (2003) 709 718
Table 5
of oxidized proteins in vitro it is not yet clear whether
Comparison of the amino acid composition of commonly used sub-
this is also the main mechanism responsible for the re-
strates in oxidation-degradation assays and their maximal proteolytic
moval of oxidized proteins from living cells.
stimulation
Amino acid Hemoglobin Ferritin SOD1
L-chain
L
Role of further components of the proteasomal system in
Total number 147 136 154
the recognition of oxidized proteins
Ala 10.2 13.2 6.5
Cys 1.4 0.7 2.6
Numerous studies have been performed using the
Asp 4.8 2.9 7.1
Glu 5.4 5.1 6.5 isolated 20S ÔcoreÕ proteasome to degrade oxidized
Phe 5.4 2.9 2.6
proteins. However, since our knowledge about the
Gly 8.8 5.9 16.2
proteasomal system, its components, and its coordi-
His 6.1 1.5 5.2
nated action with various ubiquitination systems is quite
Ile 05.9 5.8
extensive, the question arises as to which form of the
Lys 7.5 9.6 7.1
Leu 11.6 8.8 5.8 proteasome is involved in the degradation of oxidized
Met 1.4 1.5 0.6
proteins. Today it is accepted that the proteasome is just
Asn 4.1 0.7 4.5
the core proteolytic particle of a whole system of regu-
Pro 4.8 4.4 3.2
latory factors, many of which interact with the ubiqui-
Gln 2.7 5.9 1.9
tination system and several heat shock and chaperone
Arg 2.013.2 2.6
Ser 3.4 4.4 6.5 proteins [62 64]. Therefore, the question was raised
Thr 4.8 7.4 5.2
whether any of these factors is involved in the recogni-
Val 12.2 3.7 9.1
tion of oxidized proteins. Clearly in studies using the
Trp 1.4 00.6
isolated 20S ÔcoreÕ proteasome it was demonstrated that
Tyr 2.02.2 0
this protease is able to recognize oxidized proteins [1
Maximal proteolytic 28-fold [71] 12.6-fold [24] 18-fold [71]
31]. However, whether this is a process with physiolog-
stimulation
ical relevance in living cells was, at first, unclear.
[Reference]
Therefore, more complex systems such as cell lysates
were used to test the degradation of oxidized proteins.
Table 5, the substrate proteins that have been studied A number of early studies demonstrated that ATP
differ widely in amino acid content, and therefore also in has no stimulating effect on the degradation of oxidized
the oxidizable amino acid side chains, yet all show an proteins in cell lysates, thus denying involvement of the
oxidation-induced increase in proteolytic susceptibility. 19S/PA700 proteasome activator [22]. In fact ATP ac-
Several years ago we proposed that oxidized proteins tually inhibits the degradation of all oxidized proteins
are partially unfolded due to the loss of regular sec- investigated so far in cell free lysates, by some 10 20%.
ondary and tertiary structures within the domain of the Furthermore, no evidence has been reported that oxi-
oxidative impact [15,17,23]. Lasch et al. [59] were able to dized proteins are specifically recognized by the ubiq-
demonstrate an up to 50% unfolding of RNase A under uitination system. In contrast, we were recently able to
conditions allowing an increase in proteolytic suscepti- demonstrate that there is no ubiquitination of oxidized
bility. Such unfolding is clearly accompanied by an ex- and highly degradable ferritin and lysozyme in an in
posure of hydrophobic patches from the interior of the vitro system, whereas the heat denatured forms of these
protein globule to the outside. This is a possible reason proteins were ubiquitinated [65]. The lack in ubiquiti-
for the aggregation of oxidized proteins through hy- nation of oxidized proteins may be due to oxidative side
drophobic interactions. On the other hand, it was also chain modification of lysine residues, which are the
proposed that these sites might serve as recognition binding site for ubiquitin. The accumulation of ubiqui-
motifs for proteolysis. The strong correlation between tinated proteins and ubiquitinated oxidized proteins due
the increase in proteolytic susceptibility and the increase to oxidative stress which was reported by Shang and
in hydrophobicity of the protein substrate was demon- Taylor [20] and Shang et al. [66] is probably, therefore, a
strated by separation of substrates, according to their non-specific effect.
hydrophobicity, by hydrophobic interaction chroma-
tography [15,17] or by using fluorescence labels detect-
ing the surface hydrophobicity of proteins [58,60]. Since Recognition of oxidized proteins in cells
it was shown that the proteasome has a preference to
bind hydrophobic and aromatic amino acids [61] the Although it is generally accepted that oxidized, un-
recognition of these hydrophobic   unfolded  patches by folded proteins can be degraded by the isolated 20S
the proteasome seems likely. However, although it is ÔcoreÕ proteasome in vitro, it has been rather less clear if
assumed that this is the main mechanism for recognition this same form of the proteasome actually has physio-
T. Grune et al. / Biochemical and Biophysical Research Communications 305 (2003) 709 718 715
logical relevance in living cells. Rivett [1,2] demonstrated olysis, and proteins whose structure allows rapid ubiq-
the selective degradation of oxidatively modified gluta- uitination will undergo rapid turnover in vivo. Similarly,
mine synthetase in a non-lysosomal pathway by a cy- energy from ATP hydrolysis is only required to remove
tosolic protease. Subsequently, numerous studies have polyubiquitin chains from substrate proteins and to un-
demonstrated that this key enzyme is the proteasome, fold globular proteins. Our studies of protein oxidation
although the literature is confusing because at least 20 and proteolysis have consistently shown a lack of ATP
different names have been used for the same proteinase. involvement in the degradation of oxidized proteins in
Since the proteasome is responsible for the degradation mammalian cells and cell extracts [5 7,9,13 18,21 24,
of most soluble intracellular proteins it appeared quite 27 29,71,77,78,83 85]. The possible involvement of the
reasonable that proteasome would also be responsible ubiquitination system in the recognition of oxidized
for the degradation of oxidized intracellular proteins. cellular proteins was recently overruled by the recent
Perhaps the clearest answer about the degradation of work of Shringarpure et al. [65]. In this work we were
oxidized proteins by the proteasomal system was pro- able to show that cells deficient in the ubiquitination
vided by means of anti-sense oligodeoxynucleotides di- system, and therefore unable to ubiquitinate proteins, do
rected against the (essential) C2 proteasomal subunit. not lose the ability to degrade oxidized proteins. There-
Anti-sense treated cells are essentially depleted of the fore, it seems clear that the proteasome is responsible for
proteasome and are no longer able to increase protein the degradation of oxidized non-ubiquitinated proteins.
turnover after oxidative stress, to degrade oxidatively We have proposed that partial denaturation, partial
modified proteins [21,22,24]. Additionally it was shown unfolding, and exposure of internal Ôhydrophobic pat-
that the proteasome-specific or selective inhibitors such chesÕ of amino acids are the key to selective recognition
as lactacystin, b-lactone, NLVS, and epoxomicin all and degradation of oxidized proteins by the 20S pro-
prevent the removal of oxidized groups from the protein teasome [4 6,13,15 17,21 24,28,29,56,57,65,71,77,78].
pool [55,71,84]. In our model, the hydrophobic patches of oxidized
One particular difficulty with deciding which form of proteins can bind to the a-subunits at the entrance to the
the proteasome is important in degrading oxidized 20S proteasome core cylinder. This step would be im-
proteins in vivo was an assumption, by several in the portant because the entrance to the 20S core proteasome
field, that 20S core proteasome would not exist without cylinder is normally closed. Binding of such partially
bound regulatory proteins in living cells. Thus, many unfolded, oxidized substrates is then proposed to acti-
researchers felt that proteasome would always be found vate further substrate unfolding and complete proteol-
with bound 19S or 11S regulators, to form the 26S ysis. Recent work by other groups [69,70] has provided
proteasome or the Ôimmunoproteasome,Õ respectively. independent evidence to support the idea that substrate
Recently, the actual distribution of all the proteasome binding can result in proteasome opening and activa-
forms has been carefully assessed, and the presence of tion, without ATP or ubiquitin.
free 20S ÔcoreÕ proteasome particles in living cells has In closing, it now seems clear that the 20S core pro-
been demonstrated [67,68]. In fact, these important teasome plays a major role in the degradation of oxi-
studies have actually revealed that the concentration of dized proteins in the cytoplasm, nucleus, and
free 20S ÔcoreÕ proteasome exceeds all other proteasome endoplasmic reticulum of eucaryotic cells. Diminished
forms in living cells by 2- to 3-fold. proteasome activity and consequent diminished clear-
Many, perhaps most, intracellular proteins are de- ance of oxidized proteins in aging [41,56,57,71,89] and
graded by the 26S proteasome system (the 20S core various diseases [29,43,71,98 100] underlies the impor-
proteasome, with a 19S regulatory subunit at each end) tance of this physiological function. The possibility that
in a process that requires initial ubiquitination of the either, or both, the 26S proteasome and the immuno-
protein substrate. The primary and secondary structures proteasome may also degrade oxidized proteins by
of proteins make their lysine residues either more or less mechanisms that do not require ATP or ubiquitin still
prone to ubiquitination by a series of ubiquitin activat- exists, however [70]. Furthermore, it is also still possible
ing, ubiquitin conjugating, and ubiquitin ligating en- that other proteasome regulators and/or chaperone
zymes. The result is that proteins eventually obtain a proteins may be involved.
long polyubiquitin Ôtail.Õ This hydrophobic polyubiquitin
chain is recognized and bound by the 19S regulators of
the 26S proteasome. Energy from ATP hydrolysis is then
References
used by the 19S regulators to remove the polyubiquitin
chain and to unfold the substrate protein. The unfolded,
[1] A.J. Rivett, Preferential degradation of the oxidatively modified
de-ubiquitinated substrate protein is then ÔfedÕ into the
form of glutamine synthetase by intracellular mammalian pro-
cylinder of the 20S ÔcoreÕ proteasome, where it is pro-
teases, J. Biol. Chem. 260 (1985) 300 305.
teolytically degraded in a process that requires no ATP.
[2] A.J. Rivett, Purification of a liver alkaline protease which
Thus, ubiquitin serves only to ÔtagÕ a protein for prote- degrades oxidatively modified glutamine synthetase. Character-
716 T. Grune et al. / Biochemical and Biophysical Research Communications 305 (2003) 709 718
ization as a high molecular weight cysteine proteinase, J. Biol. [23] T. Grune, T. Reinheckel, K.J.A. Davies, Degradation of
Chem. 260 (1985) 12600 12606. oxidized proteins in mammalian cells, FASEB J. 11 (1997)
[3] K.J.A. Davies, A.L. Goldberg, Oxygen radicals stimulate prote- 526 534.
olysis and lipid peroxidation by independent mechanisms in [24] T. Grune, I.E. Blasig, N. Sitte, B. Roloff, R. Haseloff, K.J.A.
erythrocytes, J. Biol. Chem. 262 (1987) 8220 8226. Davies, Peroxynitrite increases the degradation of aconitase and
[4] K.J.A. Davies, M.E. Delsignore, S.W. Lin, Protein damage and other cellular proteins by proteasome, J. Biol. Chem. 273 (1998)
degradation by oxygen radicals: II. Modification of amino acids, 10857 10862.
J. Biol. Chem. 262 (1987) 9902 9907. [25] J. Jahngen-Hodge, M.S. Obin, X. Gong, F. Shang, T.R. Nowell
[5] K.J.A. Davies, Protein degradation and damage by oxygen Jr., J. Gong, H. Abasi, J. Blumberg, A. Taylor, Regulation of
radicals: I. General aspects, J. Biol. Chem. 262 (1987) 9895 9901. ubiquitin-conjugating enzymes by glutathione following oxida-
[6] K.J.A. Davies, M.E. Delsignore, Protein damage and degrada- tive stress, J. Biol. Chem. 272 (1997) 28218 28226.
tion by oxygen radicals: III. Modification of secondary and [26] M.S. Obin, F. Shang, X. Gong, G. Handelman, J. Blumberg, A.
tertiary structure, J. Biol. Chem. 262 (1987) 9908 9913. Taylor, Redox regulation of ubiquitin-conjugating enzymes:
[7] K.J.A Davies, A.L. Goldberg, Proteins damaged by oxygen mechanistic insights using the thiol-specific oxidant diamide,
radicals are rapidly degraded in extracts of red blood cells, J. FASEB J. 12 (1998) 561 569.
Biol. Chem. 262 (1987) 8227 8234. [27] O. Ullrich, N. Sitte, O. Sommerburg, V. Sandig, K.J.A. Davies,
[8] C.C. Chao, Y.S. Ma, E.R. Stadtman, Modification of protein T. Grune, Influence of DNA binding on the degradation of
surface hydrophobicity and methionine oxidation by oxidative oxidized histones by the 20S proteasome, Arch. Biochem.
systems, Proc. Natl. Acad. Sci. USA 94 (1997) 2969 2974. Biophys. 362 (1999) 211 216.
[9] D.C. Salo, R.E. Pacifici, S.W. Lin, C. Giulivi, K.J.A. Davies, [28] O. Ullrich, T. Reinheckel, N. Sitte, R. Hass, T. Grune, K.J.A.
Superoxide dismutase undergoes proteolysis and fragmentation Davies, Poly-ADP ribose polymerase activates nuclear protea-
following oxidative modification and inactivation, J. Biol. Chem. some to degrade oxidatively damaged histones, Proc. Natl. Acad.
265 (1990) 11919 11927. Sci. USA 96 (1999) 6223 6228.
[10] E.R. Stadtman, Oxidation of proteins by mixed-function oxida- [29] R. Shringarpure, T. Grune, N. Sitte, K.J.A. Davies, 4-Hydroxy-
tion systems: implication in protein turnover, aging and neutro- nonenal-modified amyloid-b peptide inhibits the proteasome:
phil function, TIBS 11 (1986) 11 12. possible importance for AlzheimerÔs disease, Cell. Mol. Life Sci.
[11] E.R. Stadtman, Oxidation of free amino acids and amino acid 57 (2000) 1802 1809.
residues in proteins by radiolysis and by metal-catalyzed reac- [30] J. Mehlhase, J. Gieche, O. Ullrich, N. Sitte, T. Grune, LPS-
tions, Annu. Rev. Biochem. 62 (1993) 797 821. induced protein oxidation and proteolysis in BV-2 microglial
[12] E.R. Stadtman, C.N. Oliver, Metal-catalyzed oxidation of cells, IUBMB Life 50 (2000) 331 335.
proteins: physiological consequences, J. Biol. Chem. 266 (1991) [31] J. Gieche, J. Mehlhase, A. Licht, T. Zacke, N. Sitte, T. Grune,
2005 2008. Protein oxidation and proteolysis in RAW264.7 macrophages:
[13] R.E. Pacifici, D.C. Salo, K.J.A. Davies, Macroxyproteinase effects of PMA activation, Biochim. Biophys. Acta 1538 (2001)
(M.O.P.): a 670kDa proteinase complex that degrades oxida- 321 328.
tively denaturated proteins in red blood cells, Free Radic. Biol. [32] E.R. Stadtman, Protein oxidation in aging and age-related
Med. 7 (1989) 521 536. diseases, Ann. N. Y. Acad. Sci. 928 (2001) 22 38.
[14] R.E. Pacifici, K.J.A. Davies, Protein degradation as an index of [33] J. Mehlhase, T. Grune, Proteolytic response to oxidative stress in
oxidative stress, Methods Enzymol. 186 (1990) 485 502. mammalian cells, Biol. Chem. 383 (2002) 559 567.
[15] R.E. Pacifici, Y. Kono, K.J.A. Davies, Hydrophobicity as the [34] A. Nunomura, G. Perry, G. Aliev, K. Hirai, A. Takeda, E.K.
signal for selective degradation of hydroxyl radical-modified Balraj, P.K. Jones, H. Ghanbari, T. Wataya, S. Shiohama, S.
hemoglobin by the multicatalytic proteinase complex, protea- Chiba, C.S. Atwood, R.B. Petersen, M.A. Smith, Oxidative
some, J. Biol. Chem. 268 (21) (1993) 15405 15411. damage is the earliest event in AlzheimerÕs disease, J. Neuropa-
[16] C. Giulivi, K.J.A. Davies, Dityrosine and tyrosine oxidation thol. Exp. Neurol. 60 (2001) 759 767.
products are endogenous markers for the selective proteolysis of [35] K.R. Sidell, S.J. Olson, J.J. Ou, Y. Zhang, V. Amarnath, T.J.
oxidatively modified red blood cell hemoglobin by (the 19S) Montine, Cysteine and mercapturate conjugates of oxidized
proteasome, J. Biol. Chem. 268 (1993) 8752 8759. dopamine are in human striatum but only the cysteine conjugate
[17] C. Giulivi, R.E. Pacifici, K.J.A. Davies, Exposure of hydrophobic impedes dopamine trafficking in vitro and in vivo, J. Neurochem.
moieties promotes the selective degradation of hydrogen perox- 79 (2001) 510 521.
ide-modified hemoglobin by the multicatalytic proteinase com- [36] L.H. Butterfield, A. Merino, S.H. Golub, H. Shau, From
plex, proteasome, Arch. Biochem. Biophys. 311 (1994) 329 341. cytoprotection to tumor suppression: the multifactorial role of
[18] C. Giulivi, K.J.A. Davies, Dityrosine: a marker for oxidatively peroxiredoxins, Antioxid. Redox Signal 1 (1999) 385 402.
modified proteins and selective proteolysis, Methods Enzymol. [37] Y. Christen, Oxidative stress and Alzheimer disease, Am. J. Clin.
233 (1994) 363 371. Nutr. 71 (2000) 621s 629s.
[19] R.L. Levine, J.A. Williams, E.R. Stadtman, E. Shacter, Carbonyl [38] K. Merker, T. Grune, Proteolysis of oxidized proteins and
assays for determination of oxidatively modified proteins, cellular senescence, Exp. Gerontol. 35 (2000) 779 786.
Methods Enzymol. 233 (1994) 346 357. [39] R.T. Dean, S. Fu, R. Stocker, M.J. Davies, Biochemistry and
[20] F. Shang, A. Taylor, Oxidative stress and recovery from pathology of radical-mediated protein oxidation, Biochem. J. 324
oxidative stress are associated with altered ubiquitin conjugating (1997) 1 18.
and proteolytic activities in bovine lens epithelial cells, Biochem. [40] R.A. Dunlop, K.J. Rogers, R.T. Dean, Recent developments in
J. 307 (1995) 297 303. the intracellular degradation of oxidized proteins, Free Radic.
[21] T. Grune, T. Reinheckel, M. Joshi, K.J.A. Davies, Proteolysis in Biol. Med. 33 (2002) 894 906.
cultered liver epithelial cells during oxidative stress: role of the [41] T. Grune, Oxidative stress, aging and the proteasomal system,
multicatalytic proteinase complex, proteasome, J. Biol. Chem. Biogerontology 1 (2000) 31 40.
270 (1995) 2344 2351. [42] B. Friguet, E.R. Stadtman, L.I. Szweda, Modification of glucose-
[22] T. Grune, T. Reinheckel, K.J.A. Davies, Degradation of oxidized 6-phosphate dehydrogenase by 4-hydroxy-2-nonenal. Formation
proteins in K562 human hematopoietic cells by proteasome, J. of cross-linked protein that inhibits the multicatalytic protease, J.
Biol. Chem. 271 (1996) 15504 15509. Biol. Chem. 269 (1994) 21639 21643.
T. Grune et al. / Biochemical and Biophysical Research Communications 305 (2003) 709 718 717
[43] B. Friguet, L.I. Szweda, Inhibition of the multicatalytic protein- [63] O. Coux, K. Tanaka, A.L. Goldberg, Structure and functions of
ase (proteasome) by 4-hydroxy-2-nonenal cross-linked protein, the 20S and 26S proteasomes, Annu. Rev. Biochem. 65 (1996)
FEBS Lett. 405 (1997) 21 25. 801 847.
[44] A. Terman, U.T. Brunk, Ceroid/lipofuscin formation in cultured [64] J.M. Peters, Proteasomes: protein degradation machines of the
human fibroblasts: the role of oxidative stress and lysosomal cell, TIBS 19 (1994) 377 382.
proteolysis, Mech. Ageing Dev. 104 (1998) 277 291. [65] R. Shringarpure, T. Grune, J. Mehlhase, K.J.A. Davies,
[45] U.T. Brunk, A. Terman, The mitochondrial lysosomal axis Ubiquitin-conjugation is not required for the degradation of
theory of cellular aging, in: E. Cadenas, L. Packer (Eds.), oxidized proteins by the proteasome, J. Biol. Chem. 278 (2003)
Understanding the Process of Aging, Marcel Dekker, Basel, 311 318.
1998, pp. 229 250. [66] F. Shang, X. Gong, A. Taylor, Activity of ubiquitin-dependent
[46] T.G. Mack, R. Dayanandan, M. van Siegtenhorst, A. Whone, pathway in response to oxidative stress, J. Biol. Chem. 272 (1997)
M. Hutton, S. Lovestone, B.H. Anderton, Tau levels with 23086 23093.
frontotemporal dementia-17 mutations have both altered expres- [67] N. Tanahshi, Y. Murakami, Y. Minami, N. Shimbara, K.B.
sion levels and phosphorylation profiles in differentiated neuro- Hendil, K. Tanaka, Hybrid proteasomes. Induction by interfer-
blastoma cells, Neuroscience 108 (2001) 701 712. on-gamma and contribution to ATP-dependent proteolysis, J.
[47] J.M. Weimer, E. Kriscenski-Perry, Y. Elshatory, D.A. Pearce, Biol. Chem. 275 (2000) 14336 14345.
The neuronal lipofuscinoses: mutations in different proteins [68] A.J. Rivett, Intracellular distribution of proteasomes, Curr.
result in similar disease, Neuromol. Med. 1 (2002) 111 124. Opin. Immunol. 10 (1998) 110 114.
[48] J.M. Holopainen, J. Saarikoski, P.K. Kinnunen, I. Jarvela, [69] M. Groll, L. Ditzel, J. Lowe, D. Stock, M. Bochtler, H.D.
Elevated lysosomal pH in neuronal lipofuscinoses, Eur. J. Bartunik, R. Huber, Structure of 20S proteasomes from yeast at
Biochem. 268 (2001) 5851 5856. 2.4 A resulution, Nature 286 (1997) 463 471.
A
[49] D. Yin, Studies on age pigments evolving into a new theory of [70] A.F. Kisselev, D. Karganovich, A.L. Goldberg, Binding of
biological aging, in: K. Kitani, G.O. Ivy, H. Shimasaki (Eds.), hydrophobic peptides to several non-catalytic sites promotes
Lipofuscin and ceroid pigments, S. Karger, Basel, 1995, pp. 159 peptide hydrolysis by all active sites of the 20S proteasomes.
170. Evidence for peptide-induced channel opening in the alpha rings,
[50] D. Yin, Biochemical basis of lipofuscin, ceroid, and age pigment- J. Biol. Chem. 277 (2002) 22260 22270.
like fluorophores, Free Radic. Biol. Med. 21 (1996) 871 888. [71] K.J.A. Davies, Degradation of oxidized proteins by 20S protea-
[51] M. Nakano, F. Oenzil, T. Mizuno, S. Gotoh, Age-related some, Biochimie 83 (2001) 1 10.
changes in the lipofuscin accumulation of brain and heart, in: K. [72] J.M. Fagan, L. Waxman, The ATP-independent pathway in red
Kitani, G.O. Ivy, H. Shimasaki (Eds.), Lipofuscin and Ceroid blood cells that degrades oxidant-damaged hemoglobin, J. Biol.
Pigments, Karger, Basel, 1995, pp. 69 77. Chem. 267 (1992) 23015 23022.
[52] D. Yin, Lipofuscin-like fluorophores can result from reactions [73] J.M. Fagan, L. Waxman, A.L. Goldberg, Red blood cells
between oxidized ascorbic acid and glutamine. Carbonyl protein contain a pathway for the degradation of oxidant-damaged
cross-linking may represent a common reaction in oxygen radical hemoglobin that does not require ATP or ubiquitin, J. Biol.
and glycosylation-related ageing processes, Mech. Ageing Dev. Chem. 261 (1986) 5705 5713.
62 (1992) 35 46. [74] W. Matthews, J. Driscoll, K. Tanaka, A. Ichihara, A.L.
[53] O. Ullrich, T. Reinheckel, N. Sitte, T. Grune, Degradation of Goldberg, Involvement of the proteasome in various degradative
hypochlorite-damaged glucose-6-phosphate dehydrogenase by processes in mammalian cells, Proc. Natl. Acad. Sci. USA 86
the 20S proteasome, Free Radic. Biol. Med. 27 (1999) 487 492. (1989) 2597 2601.
[54] K.J.A. Davies, S.W. Lin, Oxidatively denatured proteins are [75] D.C. Salo, S.W. Lin, R.E. Pacifici, K.J.A. Davies, Superoxide
degraded by an ATP-independent proteolytic pathway in Esche- dismutase is preferentially degraded by a proteolytic system
richia coli, Free Radic. Biol. Med. 5 (1988) 225 236. from red blood cells following oxidative modification by
[55] N. Sitte, K. Merker, T. Grune, Proteasome-dependent degrada- hydrogen peroxide, Free Radic. Biol. Med. 5 (1988) 335
tion of oxidized proteins in MRC-5 fibroblasts, FEBS Lett. 440 339.
(1998) 399 402. [76] A. Stolzing, A. Wengner, T. Grune, Degradation of oxidized
[56] N. Sitte, K. Merker, T. von Zglinicki, T. Grune, K.J.A. Davies, extracellular proteins by microglia, Arch. Biochem. Biophys. 400
Protein oxidation and degradation during cellular senescence of (2002) 171 179.
human BJ fibroblasts: part I effects of proliferative senescence, [77] K.J.A. Davies, S.W. Lin, R.E. Pacifici, Protein damage and
FASEB J. 14 (2000) 2495 2502. degradation by oxygen radicals. IV. Degradation of denatured
[57] N. Sitte, K. Merker, T. von Zglinicki, K.J.A. Davies, T. Grune, protein, J. Biol. Chem. 262 (1987) 9914 9920.
Protein oxidation and degradation during cellular senescence of [78] K.J.A. Davies, Intracellular proteolytic systems may function as
human BJ fibroblasts: part II aging of non-dividing fibroblasts, secondary antioxidant defenses: an hypothesis, Free Radic. Biol.
FASEB J. 14 (2000) 2502 2509. Med. 2 (1986) 155 173.
[58] R.L. Levine, L. Mosoni, B.S. Berlett, E.R. Stadtman, Methio- [79] K. Okada, C. Wangpoengtrakul, T. Osawa, S. Toyokuni, K.
nine residues as endogenous antioxidants in proteins, Proc. Natl. Tanaka, K. Uchida, 4-Hydroxy-2-nonenal-mediated impairment
Acad. Sci. USA 93 (1996) 15036 15040. of intracellular proteolysis during oxidative stress. identification
[59] P. Lasch, T. Petras, O. Ullrich, J. Backmann, D. Naumann, T. of proteasomes as target molecules, J. Biol. Chem. 274 (1999)
Grune, Hydrogen peroxide-induced structural alterations of 23787 23793.
RnaseA, J. Biol. Chem. 276 (2001) 9492 9502. [80] A.L. Bulteau, P. Verbeke, I. Petropoulos, A.F. Chaffotte, B.
[60] R.L. Levine, J. Moskovitz, E.R. Stadtman, Oxidation of Friguet, Proteasome inhibition in glyoxal-treated fibroblasts and
methionine in proteins: roles in antioxidant defenses and cellular resistance of glycated glucose-6-phosphate dehydrogenase to 20S
regulation, IUBMB Life 50 (2000) 301 307. proteasome degradation in vitro, J. Biol. Chem. 276 (2001)
[61] R. Hough, G. Pratt, M. Rechsteiner, Purification of two high 45662 45668.
molecular weight proteases from rabbit reticulocyte lysate, J. [81] T. Grune, L.O. Klotz, J. Gieche, M. Rudeck, H. Sies, Protein
Biol. Chem. 262 (1987) 8303 8313. oxidation and proteolysis by the nonradical oxidants singlet
[62] A. Ciechanover, The ubiquitin proteasome proteolytic pathway, oxygen or peroxynitrite, Free Radic. Biol. Med. 30 (2001) 1243
Cell 79 (1994) 13 21. 1253.
718 T. Grune et al. / Biochemical and Biophysical Research Communications 305 (2003) 709 718
[82] M. Rudeck, T. Volk, N. Sitte, T. Grune, Ferritin oxidation in vitro aging are selectively degraded by 26S proteasomes without
vitro: implication of iron release and degradation by the 20S ubiquitination, J. Biol. Chem. 275 (2000) 20295 20301.
proteasome, IUBMB Life 49 (2000) 451 456. [92] K. Murakami, J.H. Jahngen, S.W. Lin, K.J.A. Davies, A.
[83] T. Reinheckel, N. Sitte, O. Ullrich, U. Kuckelkorn, K.J.A. Taylor, Lens proteasome shows enhanced rates of degradation of
Davies, T. Grune, Comparative resistance of the 20S and 26S hydroxyl radical modified alpha-crystallin, Free Radic. Biol.
proteasome to oxidative stress, Biochem. J. 335 (1998) 637 642. Med. 8 (1990) 217 222.
[84] T. Grune, T. Reinheckel, R. Li, J.A. North, K.J.A. Davies, [93] K. Murakami, J.H. Jahngen, S.W. Lin, K.J.A. Davies, A.
Proteasome-dependent turnover of protein disulfide isomerase in Taylor, Lens proteasome shows enhanced rates of degradation of
oxidatively stressed cells, Arch. Biochem. Biophys. 397 (2002) hydroxyl radical modified alpha-crystallin, Free Radic. Biol.
407 413. Med. 8 (1990) 217 222.
[85] T. Grune, T. Reinheckel, J.A. North, R. Li, P.B. Bescos, R. [94] O. Sommerburg, O. Ullrich, N. Sitte, T. von Zglinicki, W. Siems,
Shringarpure, K.J.A. Davies, Ezrin turnover and cell shape T. Grune, Dose- and wavelength-dependent oxidation of crys-
changes catalyzed by proteasome in oxidatively stressed cells, tallins by UV light selective recognition and degradation
FASEB J. 16 (2002) 1602 1610. by the 20S proteasome, Free Radic. Biol. Med. 24 (1998)
[86] S. Goto, A. Hasegawa, H. Nakamoto, A. Nakamura, R. 1369 1374.
Takahashi, I.V. Kurochkin, Age-associated changes of oxidative [95] F. Shang, L. Huang, A. Taylor, Degradation of native and
modification and turnover of proteins, in: R.G. Cutler, L. Packer oxidized beta- and gamma-crystallin using bovine lens epithelial
(Eds.), Oxidative Stress and Aging, J. Bertram & A. Mori, 1995, cell and rabbit reticulocyte extracts, Curr. Eye Res. 13 (1994)
pp. 151 158. 423 431.
[87] R.L. Levine, Oxidative modification of glutamine synthetase. I. [96] W. Jessup, E.L. Mander, R.T. Dean, The intracellular storage
Inactivation is due to loss of one histidine residue, J. Biol. Chem. and turnover of apolipoprotein B of oxidized LDL in macro-
258 (1983) 11823 11827. phages, Biochim. Biophys. Acta 1126 (1992) 167 177.
[88] R.L. Levine, Oxidative modification of glutamine synthetase. II. [97] M. Pernollet, C. Villiers, F. Gabert, C. Drouet, M. Colomb, OH
Characterization of the ascorbate model system, J. Biol. Chem. treatment of tetanus toxin reduces its susceptibility to limited
258 (1983) 11828 11833. proteolysis with more efficient presentation to specific T cells,
[89] B. Friguet, L.I. Szweda, E.R. Stadtman, Susceptibility of Mol. Immunol. 30 (1993) 1639 1646.
glucose-6-phosphate dehydrogenase modified by 4-hydroxy-2- [98] J.N. Keller, W.R. Markesbery, Inhibition of the proteasome
nonenal and metal-catalyzed oxidation to proteolysis by the increases poly-DP ribosylation: implications for neuron death, J.
multicatalytic protease, Arch. Biochem. Biophys. 311 (1994) 168 Neurosci. Res. 61 (2000) 436 442.
173. [99] J.N. Keller, F.F. Huang, H. Zhu, J. Yu, Y. Ho, M.S. Kindy,
[90] D.A. Ferrington, H. Sun, K.K. Murray, J. Costa, T.D. Williams, Oxidative stress-associated impairment of proteasome activity
D.J. Bigelow, T.C. Squier, Selective degradation of oxidized during ischemia-reperfusion injury, J. Cerebral Blood Flow
calmodulin by the 20S proteasome, J. Biol. Chem. 276 (2001) Metab. 20 (2000) 1467 1473.
937 943. [100] Q. Ding, J.N. Keller, Proteasome inhibition following oxidative
[91] E. Tarcsa, G. Szymanska, S. Lecker, C.M. OÕConnor, A.L. stress: implications for heat shock proteins, J. Neurochem. 77
Goldberg, Ca2þ-free calmodulin and calmodulin damaged by in (2001) 1010 1017.