Peroxynitrite, the product of the reaction between nitricoxide (•NO) and superoxide (O2
•�), is a very reactive andshort-lived molecule. In biological systems its concen-tration is contingent on the production and decomposi-tion of its precursors. Since it decomposes rapidly underphysiological conditions, it is not possible to measure itdirectly. However, the concentrations of nitric oxide andsuperoxide are 10–100 nM and 0.1–1 nM, respectively,under normal processing conditions and can go up to�Munder pathological circumstances [1,2].
Peroxynitrite is responsible for the oxidation of me-thionine, cysteine, and tryptophan residues, the increaseof protein carbonyls content, the formation of dityrosine,and protein fragmentation [3–5]. These modifications
induce changes to the protein surface hydrophobicity,which render the molecule more susceptible to proteol-ysis.
It is now well established that the proteolytic systeminvolved in the removal of oxidatively modified proteinsis the proteasome [6–14]. Its catalytic core is the 20Sproteasome, a 670 kDa barrel-shaped particle, composedof four heptameric rings: the two outer rings formed by�-subunits and the two inner ones by�-subunits. The�-subunits harbor the catalytic sites of the complex[15,16]; in particular three of them, X, Y, and Z, expressthe well-characterized catalytic activities: the chymot-rypsin-like (ChT-L) activity, the peptidyl-glutamil pep-tide hydrolase (PGPH) activity, and the trypsin-like(T-L) activity, respectively [10]. The three catalytic sub-units (X, Y, and Z) are replaced by the interferon-�inducible subunits, LMP2, LMP7, and MECL1(LMP10), in the so-called immunoproteasome. Theyconfer to the proteasome the ability to degrade proteinsubstrates to class I antigenic peptides.
Address correspondence to: Dr. Anna Maria Eleuteri, University ofCamerino, Department of Molecular, Cellular, and Animal Biology,Via Scalzino, 7, 62032 Camerino (MC), Italy; Tel:�39 0737 403267;Fax: �39 0737 403350; E-Mail: [email protected].
The two types of proteasomes show different sub-strates specificity; in fact, the assembly of the LMPsubunits causes an almost complete loss of the PGPHactivity, a decrease in the ChT-L activity, and the ex-pression of a very active component that prefers tocleave after two amino acids with branched chain oraromatic residues at the carboxyl side: the branched-chain amino acid-preferring (BrAAP) activity. This latteractivity is also present in the constitutive enzyme, even ifin a latent state that can be activated by various com-pounds such as low concentrations of SDS and DCI [10].
In most organs, the two proteasomes are coexpressed,even if at different degrees; however, in some tissues onetype prevails over the other. For example, the thymuscontains almost exclusively the immunoproteasome [17],whereas the brain proteasome has been demostrated asbeing an XYZ complex [18].
Taking into account that the proteasome is a protein,it undergoes oxidative stress that can alter the structureand the functions of the enzyme, unbalancing the equi-librium between oxidized proteins and their degradation.This leads to damaged proteins accumulation, which isdangerous for the cells.
Being that the two proteasomes are different in struc-ture, they could be differently influenced by oxidation sothat they could contribute to the antioxidant defense atvarying degrees. We undertook this study to examine theeffects of peroxynitrite-induced oxidation on the consti-tutive and immunoproteasomes isolated from brain andthymus, respectively. We looked at the changes in struc-tures and proteolytic activities of the two peroxynitrite-treated proteasomes to explore the influence of the oxi-dative stress on the two proteasomal systems and,consequently, to evaluate the susceptibility of differentorgans to oxidation.
MATERIALS AND METHODS
Materials
Bovine brain and thymus were obtained from the localslaughterhouse. They were rapidly frozen in liquid nitro-gen and maintained at �70°C. The anti-subunit X, Y, Z,LMP7, LMP2, and MECL1 antibodies were purchasedfrom Affiniti Research Products Ltd. (Mamhead Castle,Exeter, England). The anti-nitrotyrosine antibody wasfrom Upstate Biotechnology (Lake Placid, NY, USA).Substrates for assaying the ChT-L, PGPH, and T-L pro-teolytic activities were purchased from Sigma-AldrichCorp. (St. Louis, MO, USA; Z-GGL-pNA, Z-LLE-2NA,Z-GGR-2NA). The Z-GPALG-pAB substrate was thekind gift of Professor M. Orlowski. Other reagents wereobtained from Sigma-Aldrich Corp.
Isolation and purification of proteasomes
Isolation and purification of 20S proteasomes werecarried out following experimental protocols very similarto those used previously for the MPC isolation fromother bovine organs [17,19] and essentially based on afractionation from 40% to 60% in ammonium sulfate, anionic exchange chromatography, and two gel filtrationcolumns that favor the removal of lower molecularweight contaminants. A higher degree of purification wasobtained by adding a hydrophobic interaction chroma-tography step that seems to improve the separation ofproteasome from the copurifying chaperonine Hsp90.
Determination of proteolytic activities
The ChT-L, T-L, PGPH, and BrAAP activities of 20Sproteasomes were determined as reported previously[17,20,21], using Z-GGL-pNA, Z-GGR-2NA, Z-LLE-2NA, and Z-GPALG-pAB, respectively, as substrate.Aminopeptidase N (EC 3.4.11.2), used for coupled assayto detect the BrAAP activity [21], was purified from pigkidney, as reported elsewhere [22,23].
Exposure of proteasomes to peroxynitrite
Peroxynitrite was synthesized according to the proto-col reported by Uppu et al. and stocked at �80°C [24].The concentration of peroxynitrite was determined usingUV-visible spectroscopy at � � 302 nm (� � 1670 M�1
cm�1). The peroxynitrite stock solution was diluted inKOH 0.1 M before use. Preliminary assays were per-formed to establish the inorganic buffer in which eitherproteasome maintains its activities or peroxynitrite-buffer interactions never occur. Tris-HCl (50 mM, pH8.0), Hepes (20 mM, pH 7.0), Mops (25 mM, pH 7.4),and phosphate buffer (20 mM, pH 7.0) were tested (datanot shown). Moreover, influences of KOH 0.1 M werechecked.
20S proteasomes (1.4 � 10�11 M) were treated with2 �l of peroxynitrite (final concentration from 0 to 100�M), diluted with KOH 0.1 M at different concentra-tions; after 5 min of incubation at room temperature,specific substrates were added and then incubated in afinal volume of 250 �l of phosphate buffer (20 mM, pH7.0) for 1 h at 37°C.
Peroxynitrite decomposition in phosphate buffer wasfollowed spectrophotometrically (� � 302 nm); within 1min the absorbance went down to the baseline. Forcontrol experiments, regarding the byproducts’ effects,peroxynitrite was added to the buffer and allowed todecompose for 5 min before adding the enzyme. Eachexperimental set was repeated four times and relativemean values and standard errors were calculated. TheChT-L and BrAAP activities were stopped and measuredcolorimetrically as described above; the T-L and PGPH
activities were evaluated spectrofluorimetrically (�exc �335 nm, �em � 410 nm) by a Hitachi F-4500 spectroflu-orometer (Hitachi Ltd., Tokyo, Japan).
Determination of casein degradation by 20Sproteasomes
20S proteasomes from bovine brain and thymus werepreincubated at room temperature with peroxynitrite (40and 80 �M) in 20 mM phosphate buffer, pH 7.0. There-fore, 22.5 �g of treated proteasome, dialyzed againstphosphate buffer to get rid of contaminating and decom-position products such as nitrite and nitrate, were incu-bated with 100 �g of �-casein and 20 mM phosphatebuffer, pH 7.0, in a final volume of 100 �l.
The mixtures were incubated at 37°C, and 20 �laliquots were withdrawn at different times (from 0 to 90min) and subjected to HPLC on a Hamilton PRP-3column (reversed phase 300Å HPLC column, 4.1 � 150mm) by a Perkin Helmer HPLC system. Elution wascarried out with a linear gradient established between10% and 50% acetonitrile, containing 0.1% trifluoroace-tic acid, at a flow rate of 1 ml/min. The rate of caseindegradation was determined measuring the peak heightof the casein in relation to controls in which the enzymewas not treated with peroxynitrite [25].
Each experimental set was repeated three times andrelative mean values and standard errors were calculated.
Polyacrylamide gel electrophoresis and Westernblotting
Sodium dodecyl sulfate-polyacrylamide gel electro-phoresis (SDS-PAGE) [26] was done in 12% gels. Im-munoblot experiments using anti-X, anti-Y, and anti-Zsubunits or anti-LMP2, anti-LMP7, and anti-MECL1subunits antibodies (Affiniti Research Products Ltd.)were performed by electroblotting 10 �g of the enzymes,previously separated on a 12% SDS gel, onto PVDFmembrane (Millipore Corp., Bedford, MA, USA) ac-cording to Towbin and Burnette [27,28].
The detection of 3-nitrotyrosine groups was per-formed by exposing the enzymes to increasing concen-trations of peroxynitrite (0, 10, 20, 40, and 80 �M). Afterdialyses against 20 mM phosphate buffer, pH 7.0, 10 �gof treated enzymes per lane were loaded. Immunoblotdetection of the 3-nitrotyrosine groups was performedusing anti-nitrotyrosine antibodies (Upstate Biotechnol-ogy). To check the specificity of the anti-nitrotyrosineantibodies binding, control experiments were done re-ducing the 3-nitrotyrosine to aminotyrosine with dithio-nite. The membrane was washed with 100 mM dithionitein 100 mM sodium borate, pH 9.0, extensively washedwith distilled water, and then incubated with the anti-nitrotyrosine antibodies [29,30].
The immunoblot detection was carried out with ECL(Enhanced Chemiluminescence) Western blotting analy-sis system (Amersham, Braunschweig, Germany) usingperoxidase-conjugated anti-rabbit or anti-mouse second-ary antibodies.
Dye-binding study
A Hitachi model F-4500 spectrofluorometer (HitachiLtd.) was used for fluorescence measurements. Fluores-cence emission spectra data (�em � 450–600 nm, �exc �370 nm) obtained from titration with ANS (from 0.0248to 1.0 M) were collected for 150 �g of 20S proteasomestreated or not with 100 �M peroxynitrite [31,32]. Thebuffer effect on ANS fluorescence was subtracted forevery dye concentration. Data from the binding studywere analyzed according to a fitting equation based on ageneral binding model:
P � nANS � P(ANSn) (1)
where the equilibrium constant is:
Ka �[P�ANS�n]
[P] � [ANS]n (2)
Assuming F� as the fluorescence of [P(ANS)n] at protein-saturating conditions, the fitting equation is:
F �F� � (Ka � [ANS]n)
�n � Ka � [ANS]n)(3)
Quenching of fluorescence
Quenching studies were performed using selectivefluorescence quenchers such as potassium iodide (KI)and 2,2,2-trichloroethanol (TCE) [33]. KI was dissolvedin 20 mM phosphate buffer, pH 7.0, whereas TCE wasdissolved in ethylene glycol. The titrations were per-formed for 20S proteasomes either treated or not with100 �M peroxynitrite. The enzyme concentration was1.61 � 10�7 M in all experiments.
RESULTS
20S proteasomes characterization
From immunoblot analyses using anti-X, anti-Y, an-ti-Z, anti-LMP2, anti-LMP7, and anti-MECL1 antibod-ies, the constitutive 20S proteasome was predominantlyexpressed in the bovine brain, whereas the immunopro-teasome was detectable in the bovine thymus (Fig. 1).
Immunoblot detection of 3-nitrotyrosine groups
The tyrosine nitration is a specific biomarker of theattack of reactive nitrogen species such as peroxynitrite
upon proteins [34,35]. Immunoblot detections of 3-nitroty-rosine groups on 20S proteasomes subunits purified frombovine brain and thymus were performed to evaluate thedifferent susceptibilities of the enzymes to oxidation.
The results obtained (Fig. 2) clearly indicated thatincreasing amounts of peroxynitrite (0, 10, 20, 40, and 80�M) induced a gradual increase of 3-nitrotyrosinegroups on the proteasome subunits. More specifically, 40�M peroxynitrite concentration (Fig. 2, lane 9) caused arelevant nitration on the thymus proteasome, whereas thebrain proteasome showed a significant signal only at 80�M peroxynitrite (Fig. 2, lane 5). The control experi-ments revealed a high specificity of the anti-nitrotyrosineantibodies; in fact, no bands were detected after reduc-tion with dithionite (data not shown).
Peroxynitrite effect on the catalytic activities ofproteasomes
The effect of increasing amounts of peroxynitrite(from 0 to 100 �M) on the ChT-L, T-L, BrAAP, andPGPH activities of 20S proteasomes from bovine brainand thymus is shown in Figs. 3A–3F. All assayed pro-teolytic activities of the enzymatic complexes were in-fluenced by peroxynitrite, even if in different ways. TheChT-L (Fig. 3A) activity of XYZ proteasome was acti-vated (1.75 times compared with control) at 50 �Mperoxynitrite, whereas that of LMP proteasome was 30%inhibited at 30 �M peroxynitrite.
The T-L component was the least influenced (Fig.3B): the XYZ proteasome was gradually and weaklyinhibited, whereas the immunoproteasome was not af-fected.
Fig. 1. Immunoblot detection of proteasomes �-subunits. 20S protea-somes purified from bovine thymus (lanes 1) and brain (lanes 2) wereseparated on a 12% SDS-PAGE and then electroblotted onto PVDFmembranes. Autoradiographic films were obtained from the immuno-blot analyses performed using anti-X, anti-Y, anti-Z, anti-LMP7, anti-LMP2, and anti-MECL1 antibodies.
Fig. 2. Immunoblot detection of 3-nitrotyrosine groups. 20S proteasomes purified from bovine brain (lanes 1–5) and thymus (lanes6–10) were treated with increasing amounts of peroxynitrite (0 �M: lanes 1 and 6; 10 �M: lanes 2 and 7; 20 �M: lanes 3 and 8; 40�M: lanes 4 and 9; 80 �M: lanes 5 and 10), separated on a 12% SDS-PAGE, and then electroblotted onto a PVDF membrane. (A)The Comassie brilliant blue-stained PVDF membrane; (B) autoradiographic film obtained from the immunoblot analyses performedusing anti-nitrotyrosine antibodies.
The brain 20S proteasome showed a significant acti-vation of BrAAP activity (3.5 times compared with thecontrol) at peroxynitrite concentration lower than 50 �M(Fig. 3C), while the LMP complex was inhibited to 65%
(Fig. 3D). The PGPH activity was four times increased inthe XYZ proteasome at 40 �M peroxynitrite (Fig. 3E),whereas it was 15% inhibited in the LMP proteasome(Fig. 3E).
Fig. 3. Effect of increasing concentrations of peroxynitrite on the proteasomes peptidase activities. 20S proteasomes from bovinebrain (�) and thymus (�) (1.4 � 10�11 M) were treated with 2 �l of peroxynitrite (final concentration from 0 to 100 �M) dilutedwith KOH 0.1 M and then incubated in a final volume of 250 �l of phosphate buffer (20 mM, pH 7.0), with synthetic substrates((A) Z-GGL-pNA for ChT-L activity; (B) Z-GGR-2NA for T-L activity; (C and D) Z-GPALG-pAB for BrAAP activity; (E andF) Z-LLE-2NA for PGPH activity), at 37°C for 1 h. The ChT-L and BrAAP activities were stopped and measured colorimetri-cally, as described previously [20]; the T-L and PGPH activities were evaluated spectrofluorimetrically (�exc � 335 nm, �em �410 nm) by a Hitachi F-4500 spectrofluorometer (Hitachi Ltd.). For control experiments, peroxynitrite was added to the bufferand allowed to decompose for 5 min before adding the enzyme. Each data point is subtracted of its control, is the mean value� 4% standard error, and comes from four separate determinations. Error bars are not shown because they correspond to the sizeof the symbol.
The proteolytic activity of 20S proteasomes on mac-romolecular substrates was assayed using �-casein assubstrate. Figures 4A and 4B show the effect of per-oxynitrite increasing concentrations (0, 40, and 80 �M)on the caseinolytic activity of brain and thymus 20Sproteasomes, respectively. Both enzymes degraded �-ca-sein in the absence of a nitrating agent, even if at differ-ent rates. Upon peroxynitrite exposure, the XYZ com-plex was not influenced and the immunoproteasome wasinhibited.
Conformational changes induced by peroxynitrite:spectrofluorimetric studies
To check structural changes induced by peroxynitriteon the 20S complexes, spectrofluorimetric studies were
performed testing intrinsic tryptophan fluorescence forthe two enzymes treated and not with 100 �M peroxyni-trite. Figures 5A and 5B show emission spectra (�em �310–400 nm, �exc � 295 nm) of brain and thymusproteasomes, respectively.
Before treatment with the nitrating agent, the XYZproteasome showed an emission spectrum centered at337 nm (F � 1817 a.u.); upon oxidation, a decrease of32.66% fluorescence intensity without a significant shiftof the maximal emission wavelength was observed. Dif-ferently, the native LMP proteasome exhibited an emis-sion spectrum centered at 335 nm (F � 1429 a.u.); afterperoxynitrite exposure, the fluorescence spectrumshowed a relevant decrease of fluorescence intensity(84.24%) with a blue shift of the maximal emissionwavelength (from 335 to 331 nm).
Binding of 1-anilinonaphthalene-8-sulfonic acid to 20Sproteasomes
Binding assays with the fluorescence dye ANS wereperformed to test conformational changes induced byperoxynitrite on the complex structure. Fluorescenceemission spectra (�em � 450–600 nm, �exc � 370 nm)
Fig. 4. Effect of peroxynitrite (control (�), 40 �M (�), and 80 �M(E)) on the casein degradation by the 20S proteasome from (A) bovinebrain and (B) thymus. Each data point is the mean value � 4% standarderror and comes from three separate determinations. Error bars are notshown because they correspond to the size of the symbol.
Fig. 5. Emission spectra (�em � 310–400 nm, �exc � 295 nm) for (A)constitutive and (B) immunoproteasome treated (�) or not (E) with100 �M peroxynitrite.
992 M. AMICI et al.
for the brain and thymus 20S proteasomes (150 �g),treated or not with 100 �M peroxynitrite at each ANSaddition (final concentration from 0.0248 to 1.0 M), werecollected. Results of ANS fluorescence were correctedtaking into account the buffer interferences. All spectrashowed an emission maximum of ANS centered at 482nm.
A quantitative analysis of the experimental data wasperformed according to Eqns. 2 and 3.
Equation 3 allows us to obtain the total binding sitesfor ANS on the proteasomal molecule and the equilib-rium constant related to ANS-saturating conditions. Fit-ting results are reported in Table 1.
Quenching of tryptophan groups
To evaluate if microenvironment changes of trypto-phan residues occurred in the two native or peroxynitrite-treated 20S proteasomes, quenching studies were per-
formed using hydrophobic and polar probes such as TCEand KI. Emission spectra from 310 to 400 nm (�exc �295 nm) were registered at any addition of the probe.
The concentration range for TCE was from 0 to 0.16M and for KI from 0 to 0.6 M. The final quencherconcentration was different for the two enzymes and alsovaried before and after peroxynitrite exposure. Thebuffer contribution was considered for every spectrum.
The Stern-Volmer plots for the quenching experi-ments are shown in Fig. 6. The accessible fraction (fa)and the Stern-Volmer quenching constant (KSV) wereobtained using a modified Stern-Volmer equation; dataare summarized in Table 2.
DISCUSSION
The characterization of isolated 20S proteasomes, us-ing specific antibodies against the catalytic subunits,
Table 1. Fitting Results for the Binding of 1-Anilinonaphthalene-8–Sulfonic Acid to 20S Proteasomes
F� (arbitrary units) �max (nm) Ka (mM�1) n. sitesANS
F� � fluorescence of [P(ANS)n] at protein-saturating conditions; �max � emission maximum wavelength, Ka
� equilibrium constant; n. sitesANS � number of sites for ANS.
Fig. 6. Stern-Volmer plots for the quenching of tryptophan residues in (A and B) XYZ and (C and D) LMP proteasomes with (A andC) potassium iodide and (B and D) 2,2,2-trichloroethanol. Experiments were executed for the two enzymes treated (Œ) or not (�) with100 �M peroxynitrite.
993Peroxynitrite and proteasomes
showed a predominant expression of the XYZ complexin bovine brain, whereas the LMP proteasome waspresent in thymus (Fig. 1).
Since peroxynitrite half-life is estimated to be lessthan 100 ms, it is almost impossible to detect it directly.Therefore, the nitration of tyrosine residues is considereda useful marker for detecting the peroxynitrite effect onprotein substrates [34–36]. Data in Fig. 2 confirmed thatexposure to the nitrating agent resulted in the nitration,peroxynitrite concentration dependent, of the two protea-somes. From the blotting analyses, it was revealed thatthe LMP proteasome showed a higher susceptibility withrespect to the constitutive proteasome. Looking at thesequences (obtained from the “Expert Protein AnalysisSystem Molecular Biology Server” database, Swiss In-stitute of Bioinformatics, Geneva, Switzerland) of con-stitutive and interferon-� inducible �-subunits, it turnsout that the number of tyrosine residues is similar in thetwo complexes. Under identical experimental conditions,the differences in the amounts of 3-nitrotyrosine groupscould derive from a different accessibility of the hydro-phobic regions of the two enzyme structures. Regardingthe proteolytic activities, a clear effect of peroxynitrite ata concentration less than 10 �M is evident: the brain 20Sproteasome showed an overall activation, with the PGPHand BrAAP activities being the most stimulated uponperoxynitrite treatment and with the exception of the T-Lcomponent, which was 20% inhibited. On the other hand,the LMP proteasome was inhibited, especially theBrAAP activity, and the T-L activity was not affected.
Recently, other papers regarding the peroxynitrite ef-fect on the proteasomal system have been published.Grune et al. [8] and Gieche et al. [37] reported studies oncell cultures, measuring the ChT-L activity with thesubstrate suc-LLVY-MCA; both presented an inhibitionof that activity. To our knowledge, Reinheckel et al. [38]observed an inhibitory effect of the suc-LLVY-MCAhydrolyzing activity on the 20S proteasome isolatedfrom human erythrocytes. In our case, we detected adecrease of the ChT-L activity measured with the sub-strate Z-GGL-pNA only for the LMP 20S proteasome;on the contrary, the XYZ 20S proteasome was signifi-cantly activated. Therefore, our data are not comparable
with the above-mentioned studies because the proteaso-mal systems and the synthetic substrates were different.Furthermore, to gain insights into the effects of thenitrating agent, we assayed other peptidase activities,like the PGPH, BrAAP, and T-L activities; for the twoenzymes, we obtained two distinct trends already evidentat low peroxynitrite concentrations, namely the immu-noproteasome being inhibited and the constitutive com-plex being activated. The only exception was the T-Lcomponent, which was slightly inhibited in the XYZproteasome and not affected in the immunoproteasome.
The investigation of the influence of peroxynitrite onthe proteolytic activity was carried out using �-casein assubstrate for 20S proteasomes exposed or not to thenitrating agent. Increasing amounts of peroxynitrite in-duced a gradual inhibition of �-casein degrading rate bythe LMP proteasome, whereas peroxynitrite did not in-fluence the constitutive complex.
Proteins exposure to peroxynitrite results in modifi-cations that include tyrosine nitration as well as trypto-phan oxidation. The fluorescence of tryptophan dependson the integrity of the indole ring; therefore, its disrup-tion by oxidation or N-nitration may be responsible forthe loss of fluorescence. Analysis of intrinsic fluores-cence spectra of two 20S proteasomes, treated or notwith 100 �M peroxynitrite, clearly demostrated thattryptophan residues on the LMP proteasome were moresusceptible to the nitrating agent compared to the con-stitutive complex.
Upon ANS binding, the two native enzymes behaveddissimilarly; in fact, F� and Ka values for the LMP andXYZ complexes, reported in Table 1, showed statisti-cally significant differences. This is related to a higheraffinity and a larger hydrophobic surface for ANS in theimmunoproteasome than in the constitutive one.
After treatment with peroxynitrite, spectra obtainedfrom the ANS-binding experiments showed a decrease ofemission maxima without a significant shift of the �max,indicating that both enzymes underwent an increase inthe polarity of the ANS-binding sites. However, theimmunoproteasome showed a more sensitive changewith respect to the constitutive complex: a 75% decreaseof its equilibrium constant and a reduction of its number
of available sites (n) for ANS was detected, whereas onlya 30% decrease characterized the XYZ complex withoutany variation of n.
Furthermore, the quenching studies of tryptophan res-idues with TCE displayed a linear Stern-Volmer plot andfull accessibility to a quencher for the constitutive pro-teasome, treated and not with peroxynitrite, with a sig-nificant decrease of KSV; a linear Stern-Volmer plot anda full accessibility to an ionic quencher (KI) was ob-tained for the native XYZ complex, whereas oxidationproduced a down curvature of the Stern-Volmer plotwith a decrease of fa and an increase of KSV.
On the contrary, only the native LMP proteasomeshowed a curvature of the Stern-Volmer plot for TCE,whereas nitration produced a decrease of KSV and anincrease of fa; a linear Stern-Volmer plot for KI wasobtained for the enzyme treated and not with peroxyni-trite, whereas oxidation induced a slight increase of KSV
and a decrease of accessibility of the polar quencher.The results obtained from these experiments con-
firmed what was found with ANS: the native immuno-proteasome has a larger hydrophobic surface than theXYZ complex and the peroxynitrite exposure causes adecrease in hydrophobicity, with this effect being morerelevant in the immunoproteasome than in the constitu-tive one. On the other hand, a decrease of the number ofbinding sites for KI was evident after peroxynitrite treat-ment in both enzymes, most likely due to an increase ofcharged residues, but only the XYZ complex doubled theaffinity for the ionic quencher.
CONCLUSION
Increasing evidence suggests that oxidative stress isimplicated in aging and numerous diseases, many ofwhich are associated with defects in proteolytic systems[39,40]. Previous studies have indicated that alterationsof proteasome activity may be a cause of the increase inprotein oxidation and protein aggregation, typical of theaging process and neurodegeneration [41–44].
It has been demostrated that oxidative modification ofthe proteasome and its interaction with oxidized or 4-hy-droxy-2-nonenal- (HNE-) modified proteins occur, todifferent extents, in regions of the central nervous sys-tem, leading to a significant inhibition of the proteasome[45–49].
On the contrary, proteasome complexes isolated fromerythrocytes and exposed to mild oxidative conditionsare stimulated, reinforcing the theory of the proteasomekey role in the cellular antioxidant defense [38,50].
From our study, it is evident that the two 20S protea-somes each have a different susceptibility to peroxyni-trite. The modifications induced on the XYZ 20S protea-some cause a slight structural rearrangement that leads to
an overall activation of the hydrolysis of the short pep-tides used as substrates, without affecting the caseino-lytic activity. Meanwhile, the immunoproteasome oxida-tion produces a more significant conformational changethat damages the enzyme functionality. Therefore, sincethe constitutive 20S proteasome is relatively resistant tothe peroxynitrite-induced oxidative stress while the 20Simmunoproteasome is inhibited by the nitrating agent, itis reasonable to suppose that the two proteasomes pro-vide antioxidant defense for the various organs to differ-ent degrees. It is tempting to speculate a preferentialassembly of constitutive proteasomal systems with re-spect to their immunocounterparts in tissues physiolog-ically exposed to oxidative conditions. In fact, this couldcome either from an a priori downexpression of LMPsubunits or from an a posteriori removal of oxidizedimmunoproteasomes.
REFERENCES
[1] Groves, J. T. Peroxynitrite: reactive, invasive, and enigmatic.Curr. Opin. Chem. Biol. 3:226–235; 1999.
[2] Patel, R. P.; McAndrew, J.; Sellak, H.; White, C. R.; Jo, H.;Freeman, B. A.; Darley-Usmar, V. M. Biological aspects ofreactive nitrogen species. Biochim. Biophys. Acta 1411:385–400;1999.
[3] Pryor, W. A.; Squadrito, G. L. The chemistry of peroxynitrite: aproduct from the reaction of nitric oxide with superoxide. Am. J.Physiol. 268:L699–L722; 1995.
[4] Koppenol, W. H. The basic chemistry of nitrogen monoxide andperoxynitrite. Free Radic. Biol. Med. 25:385–391; 1998.
[5] Ischiropoulos, H.; Al-Mehdi, A. B. Peroxynitrite-mediated oxi-dative protein modification. FEBS Lett. 364:279–282; 1995.
[6] Grune, T.; Reinheckel, T.; Davies, K. J. A. Degradation of oxi-dized proteins in mammalian cells. FASEB J. 11:526–534; 1997.
[7] Suoza, J. M.; Choi, I.; Chen, Q.; Weisse, M.; Daikhin, E.; Yud-koff, M.; Obin, M.; Ara, J.; Horwitz, J.; Ischiropoulos, H. Pro-teolytic degradation of tyrosine nitrated proteins. Arch. Biochem.Biophys. 380:360–366; 2000.
[8] Grune, T.; Klotz, L. O.; Gieche, J.; Rudeck, M.; Sies, H. Proteinoxidation and proteolysis by the nonradical oxidants singlet ox-ygen or peroxynitrite. Free Radic. Biol. Med. 30:1243–1253;2001.
[9] Baumeister, W.; Walz, J.; Zuhl, F.; Seemuller, E. The protea-some: paradigm of a self-compartmentalizing protease. Cell 92:367–380; 1998.
[10] DeMartino, G. N.; Slaughter, C. A. The proteasome. A novelprotease regulated by multiple mechanisms. J. Biol. Chem. 274:22123–22126; 1999.
[11] Orlowski, M.; Wilk, S. Catalytic activities of the 20S proteasome,a multicatalytic proteinase complex. Arch. Biochem. Biophys.383:1–16; 2000.
[12] Merker, K.; Grune, T. Proteolysis of oxidized proteins and cel-lular senescence. Exp. Gerontol. 35:779–786; 2000.
[13] Grune, T. Oxidative stress, aging, and the proteasomal system.Biogerontology 1:31–40; 2000.
[14] Davies, K. J. A. Degradation of oxidized proteins by the 20Sproteasome. Biochimie 83:301–310; 2001.
[15] Lowe, J.; Stock, D.; Jap, B.; Zwickl, P.; Baumeister, W.; Huber,R. Crystal structure of the 20S proteasome from the archaeon T.acidophilum at 3.4 A resolution. Science 268:533–539; 1995.
[16] Groll, M.; Ditzel, L.; Lowe, J.; Stock, D.; Bochtler, M.; Bartunik,H. D.; Huber, R. Structure of 20S proteasome from yeast at 2.4 Aresolution. Nature 386:463–471; 1997.
[17] Eleuteri, A. M.; Angeletti, M.; Lupidi, G.; Tacconi, R.; Bini, L.;Fioretti, E. Isolation and characterization of bovine thymus mul-ticatalytic proteinase complex. Protein Express. Purif. 18:160–168; 2000.
[18] Noda, C.; Tanahashi, N.; Shimbara, N.; Hendil, K. B.; Tanaka, K.Tissue distribution of constitutive proteasomes, immunoprotea-somes, and PA28 in rats. Biochem. Biophys. Res. Commun.277(2):348–354; 2000.
[19] Orlowski, M.; Michaud, C. Pituitary multicatalytic proteinasecomplex. Specificity of components and aspects of proteolyticactivity. Biochemistry 28:9270–9278; 1989.
[20] Wilk, S.; Orlowski, M. Evidence that pituitary cation-sensitiveneutral endopeptidase is a multicatalytic protease complex.J. Neurochem. 40:842–849; 1983.
[21] Orlowski, M.; Cardozo, C.; Michaud, C. Evidence for the pres-ence of five distinct proteolytic components in the pituitary mul-ticatalytic proteinase complex. Properties of two componentscleaving bonds on the carboxyl side of branched chain and smallneutral amino acids. J. Biochem. 32:1563–1572; 1993.
[22] Pfleiderer, G. Particle-bound aminopeptidase from pig kidney.Methods Enzymol. 19:514–521; 1970.
[23] Almenoff, J.; Orlowski, M. Membrane-bound kidney neutral met-alloendopeptidase: interaction with synthetic substrates, naturalpeptides, and inhibitors. Biochemistry 22:590–599; 1983.
[24] Uppu, R. M.; Squadrito, G. L.; Cueto, R.; Pryor, W. A. Synthesisof peroxynitrite by azide-ozone reaction. Methods Enzymol. 269:311–321; 1996.
[25] Eleuteri, A. M.; Kohanski, R. A.; Cardozo, C.; Orlowski, M.Bovine spleen multicatalytic proteinase complex (proteasome).Replacement of X, Y, and Z subunits by LMP7, LMP2, andMECL1 and changes in properties and specificity. J. Biol. Chem.272:11824–11831; 1997.
[26] Laemmli, U. K. Cleavage of structural proteins during the assem-bly of the head of bacteriophage T4. Nature 277:680–685; 1970.
[27] Towbin, H.; Staehelin, T.; Gordon, J. Electrophoretic transfer ofproteins from polyacrylamide gels to nitrocellulose sheets: pro-cedures and some applications. Proc. Natl. Acad. Sci. USA 76:4350–4354; 1979.
[28] Burnette, W. N. Western blotting: electrophoretic transfer ofproteins from sodium dodecyl sulfate-polyacrylamide gels to un-modified nitrocellulose and radiographic detection with antibodyand radioiodinated protein A. Anal. Biochem. 112:195–203; 1991.
[29] Ye, Y. Z.; Strong, M.; Huang, Z.; Beckman, J. S. Antibodies thatrecognize nitrotyrosine. Methods Enzymol. 269:201–209; 1996.
[30] Viera, L.; Ye, Y. Z.; Estevez, A. G.; Beckman, J. S. Immunohis-tochemical methods to detect nitrotyrosine. Methods Enzymol.301:373–381; 1999.
[31] Johnson, J. D.; El-Bayoumi, M. A.; Weber, L. D.; Tulinsky, A.Interaction of �-chymotrypsin with the fluorescent probe 1-anili-nonaphtalene-8-sulfonate in solution. Biochemistry 18:1292–1296; 1979.
[32] Conconi, M.; Petropoulos, I.; Emond, I.; Turlin, E.; Biville, F.;Friguet, B. Protection from oxidative inactivation of the 20Sproteasome by heat-shock protein 90. Biochem. J. 333:407–415;1998.
[33] Eftink, M. N.; Ghiron, C. A. Exposure of tryptophanyl residues inproteins. Quantitative determination by fluorescence quenchingstudies. Biochemistry 15:672–680; 1976.
[34] Beckman, J. S.; Ischiropoulos, H.; Zhu, L.; van der Woerd, M.;Smith, C. D.; Harrison, J.; Martin, J. C.; Tsai, J.-H. M. Peroxyni-trite-mediated tyrosine nitration catalyzed by superoxide dis-mutase. Arch. Biochem. Biophys. 298:438–445; 1992.
[35] Ischiropoulos, H. Biological tyrosine nitration: a pathophysiolog-ical function of nitric oxide. Arch. Biochem. Biophys. 356:1–11;1998.
[36] Radi, R.; Peluffo, G.; Alvarez, M. N.; Naviliat, M. Cayota, A.Unraveling peroxynitrite formation in biological systems. FreeRadic. Biol. Med. 30:463–488; 2001.
[37] Gieche, J.; Mehlhase, J.; Licht, A.; Zacke, T.; Sitte, N.; Grune, N.Protein oxidation and proteolysis in RAW264.7 macrophages:
effects of PMA activation. Biochim. Biophys. Acta 1538:321–328; 2001.
[38] Reinheckel, T.; Sitte, N.; Ullrich, O.; Kuckelkorn, U.; Davies,K. J. A.; Grune, T. Comparative resistance of the 20S and 26Sproteasome to oxidative stress. Biochem. J. 335:637–642; 1998.
[39] Torreilles, F.; Salman-Tabcheh, S.; Guerin, M. C.; Torreilles, J.Neurodegenerative disorders: the role of peroxynitrite. Brain Res.Rev. 30:153–163; 1999.
[40] Szweda, P. A.; Friguet, B.; Szweda, L. I. Proteolysis, free radi-cals, and aging. Free Radic. Biol. Med. 33:29–36; 2002.
[41] Ding, Q.; Keller, J. N. Proteasomes and proteasome inhibition inthe central nervous system. Free Radic. Biol. Med. 31:574–584;2001.
[42] Keller, J. N.; Gee, J.; Ding, Q. The proteasome in brain aging.Ageing Res. Rev. 1:279–293; 2002.
[43] Bulteau, A. L.; Szweda, L. I.; Friguet, B. Age-dependent declinesin proteasome activity in the heart. Arch. Biochem. Biophys.397:298–304; 2002.
[44] Carrard, G.; Bulteau, A. L.; Petropoulos, I.; Friguet, B. Impair-ment of proteasome structure and function in aging. Int. J. Bio-chem. Cell Biol. 34:1461–1474; 2002.
[45] Friguet, B.; Szweda, L. I. Inhibition of the multicatalytic protein-ase (proteasome) by 4-hydroxy-2-nonenal cross-linked protein.FEBS Lett. 405:21–25; 1997.
[46] Okada, K.; Wangpoengtrakul, C.; Osawa, T.; Toyokuni, S.;Tanaka, K.; Uchida, K. 4-Hydroxy-2-nonenal-mediated impair-ment of intracellular proteolysis during oxidative stress. J. Biol.Chem. 274(34):23787–23793; 1999.
[47] Keller, J. N.; Hanni, K. B.; Markesbery, W. R. Possible involve-ment of proteasome inhibition in aging: implications for oxidativestress. Mech. Ageing Dev. 113:61–70; 2000.
[48] Keller, J. N.; Huang, F. F.; Markesbery, W. R. Decreased levelsof proteasome activity and proteasome expression in aging spinalcord. Neuroscience 98:149–156; 2000.
[49] Shringarpure, R.; Grune, T.; Sitte, N.; Davies, K. J. 4-Hydroxy-2-nonenal-modified amyloid-� peptide inhibits the proteasome:possible importance in Alzheimer’s disease. Cell. Mol. Life Sci.57:1802–1809; 2000.
[50] Strack, P. R.; Waxman, L.; Fagan, J. M. Activation of the mul-ticatalytic endopeptidase by oxidants. Effects on enzyme struc-ture. Biochemistry 35:7142–7149; 1996.
