Optimizing separation efficiency of 2-DE procedures for visualization of different superoxide...

Cristina Di Poto1, 2

Paolo Iadarola1

Roberta Salvini1

Ileana Passadore1, 3

Cristina Cereda2

Mauro Ceroni2, 4, 5

Anna Maria Bardoni1

1Department of Biochemistry,University of Pavia,Pavia, Italy

2Laboratory of ExperimentalNeurobiology,IRCCS,Foundation “C. Mondino”Institute of Neurology,Pavia, Italy

3Department of Haematological,Pneumological andCardiovascular Sciences,University of Pavia,Pavia, Italy

4Department of Neurology,Policlinico of Monza,Monza, Italy

5Department of Neurosciences,University of Pavia,Pavia, Italy

Received March 2, 2007Revised May 17, 2007Accepted July 27, 2007

Research Article

Optimizing separation efficiency of 2-DEprocedures for visualization of differentsuperoxide dismutase forms in a cellularmodel of amyotrophic lateral sclerosis

Neurodegenerative diseases such as Alzheimer disease (AD) and Parkinson disease (PD)have been associated with increased production of reactive oxygen species. In AD and PDpatients, superoxide dismutase (SOD1) was also indicated as a major target of oxidativedamage. In particular, in brain tissue of these patients, different SOD1 isoforms have beenidentified, although their functional role still remains to be elucidated. In the light of thepossibility that different SOD1 entities could be expressed also in other neurodegenerativedisorders, as a sort of unifying event with AD and PD, we have investigated amyotrophiclateral sclerosis (ALS) using human neuroblastoma SH-SY5Y cells with mutated SOD1gene H46R as cellular model. 2-DE using a narrow-range IPG 4–7 strips in the first di-mension and linear 15% SDS-PAGE in the second allowed to separate different SOD1spots. MALDI-TOF MS and CapLC-MS/MS have been used for their complete identifica-tion. This is the first report in which the presence of SOD1 (iso)forms in a cellular model ofALS has been evidenced.

Keywords:

2-DE / MALDI-MS / Neurodegenerative disorders / Superoxide dismutaseDOI 10.1002/elps.200700162

4340 Electrophoresis 2007, 28, 4340–4347

1 Introduction

A common characteristic of neurodegenerative and psychia-tric diseases is that their molecular bases are unknown andindividual biomarkers are often not reliable. This is the caseof amyotrophic lateral sclerosis (ALS), Alzheimer disease(AD), and Parkinson disease (PD), chronic disorders causedby a combination of events that impair normal neuronalfunction and characterized by the accumulation of insolubleproteinaceous deposits [1–3]. Unifying events in these slowlyprogressive neurodegenerative disorders include accumula-tion of aberrant or misfolded proteins, protofibril formation,ubiquitin–proteasome system dysfunction, mitochondrialinjury, synaptic failure, altered metal homeostasis, and oxi-dative and nitrosative stress [1, 3]. Both AD and PD have

been associated with increased production of reactive oxygenspecies, which could result either from genetic predisposi-tion and/or environmental factors such as exposure to pesti-cides [1, 4]. Markers of oxidative damage have also beenidentified in the motor cortex and the spinal cord of patientswith familial ALS (FALS) and sporadic ALS (SALS) [5–7]. Inan effort to obtain a better understanding of the pathogenicmechanisms in these diseases, specific proteins that are tar-gets of oxidative stress have been largely investigated [1, 3, 6].As a result of these studies, ubiquitin carboxyl-terminalhydrolase L1 (UCH-L1) was identified as the causativegenetic defect for certain familial forms of PD [8, 9] andCu,Zn-superoxide dismutase (SOD1), an antioxidant en-zyme that plays an essential role in the cellular defenseagainst harmful superoxide radicals, was indicated as amajor target of oxidative damage in AD and PD [1]. In par-ticular, a total of four SOD1 isoforms differing in their pIhave been identified in brain of PD and AD subjects and aseries of mutations in the SOD1 gene encoding for the SOD1enzyme have been clearly evidenced in individuals withfamilial ALS [10, 11]. Among these, of particular interestappeared the H46R mutation which involves one of the resi-dues that binds catalytic copper at the active site [12] andoriginates a SOD with a dramatically reduced activity [11].Thus, in the light of the possibility that SOD1-modified

Correspondence: Professor Paolo Iadarola, Department of Bio-chemistry, University of Pavia, Via Taramelli 3/B, I-27100 Pavia,ItalyE-mail: [email protected]: 139-0382-423108

Abbreviations: AD, Alzheimer disease; ALS, amyotrophic lateralsclerosis; DTE, dithioerythritol; FALS, familial ALS; NL, nonlinear;PD, Parkinson disease; SOD1, superoxide dismutase

© 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.electrophoresis-journal.com

Electrophoresis 2007, 28, 4340–4347 Proteomics and 2-DE 4341

forms could be present not only in brain but also in othercellular compartments and for neurodegenerative disordersother than PD and AD, we have probed for these entities inALS using human neuroblastoma SH-SY5Ycells, transfectedwith mutated SOD1 gene H46R, as cellular model. Giventhat proteomics technologies have proved to be an ideal re-search platform for monitoring changes in the expressionlevel of complex protein mixtures, 2-DE profiling combinedwith Western blotting and MS were the methods of choicefor this study. Despite their well-recognized robustness andefficiency, the “conventional” procedures which employedbroad-range (pH 3–10) nonlinear (NL) IPG strips for IEF anda 9–16% T for SDS-PAGE in the second dimension, did notshow the power to resolve the spots of interest present in ourmixture and made assignment of specific protein identityunreliable.

We present here an accurate description of the analyticalprotocol developed for obtaining their separation and identi-fication. The data shown in this study demonstrate for thefirst time that different SOD1 entities are expressed also inthis cell model of ALS, although a great deal of research isstill needed to obtain their structural characterization and toestablish whether these forms may represent a sort of unify-ing event between this disorder and PD or AD.

Our investigation confirms that proteomics is an exqui-site tool for identifying target proteins that may be involvedat important time points in human (neurodegenerative) dis-orders.

2 Materials and methods

2.1 Reagents

Cell culture media and reagents, protease inhibitor cocktail,1,4-dithioerythritol (DTE), CHAPS, glycerol, iodoacetamide,agarose, HEPES, KOH, KCl, MgCl2, sucrose, TCA, urea,SDS, Tris, PBS, were purchased from Sigma–Aldrich (St.Louis, MO, USA). Bicinchoninic acid (BCA) was purchasedfrom Pierce (Rockford, IL, USA). Carrier ampholytes andIPG gel strips were from GE Healthcare (Uppsala, Sweden).Antibodies for Cu,Zn-SOD1 and polyclonal anti-Op18 forstathmin/oncoprotein 18 were purchased from Santa CruzBiotechnology (Santa Cruz, CA, USA).

2.2 Cell lines and cytosol preparation

Human neuroblastoma cells SH-SY5Y (SH) were main-tained in Dulbecco’s modified Eagle/F12 medium supple-mented with 10% fetal calf serum, 2 mM L-glutamine,100 U/mL penicillin, 10 mg/mL streptomycin, at 377C in aatmosphere humidified with 5% CO2. Monoclonal cell lineexpressing FALS-SOD1 H46R was obtained as previouslydescribed [13, 14] by transfection with expression plasmidpRc/CMV (Invitrogen, Carlsbad, CA, USA) containingcDNAs coding for the SOD1 protein. Cytosol from SH-SY5Y

cells was prepared using a modified version of method ofYang et al. [15]. Cells were centrifuged at 6006g for 5 min toobtain pellets that were washed twice with ice-cold PBS andresuspended in extraction buffer (20 mM HEPES/KOHpH 7.4, 10 mM KCl, 1.5 mM MgCl2, 1 mM DTE, 2 Msucrose, protease inhibitor cocktail). Cells were homoge-nized with 30 passes of a micropestle. Homogenates werecentrifuged at 8506g for 6 min at 47C originating a post-nuclear S1 supernatant that was harvested. The pellet wasresuspended in extraction buffer and homogenized as above.The centrifugation step was repeated: the two S1 super-natants were pooled and centrifuged at 13 0006g for 10 minat 47C originating a S2 supernatant whose protein con-centration was determined using the BCA assay. Ice-coldTCA (final concentration 12.5%) was added to precipitateproteins. Precipitates were centrifuged at 14 5006g for30 min and the pellet was washed twice with 1 mL of ace-tone. Precipitates were finally resuspended in electrophoreticbuffer and subjected to 2-DE as indicated in the followingparagraph.

2.3 2-DE

The IPG gel strips with a NL gradient range of pH 3–10 and alinear gradient of pH 4–7 (length 18 cm in both cases) fromGE Healthcare were rehydrated in a swelling solution(350 mL of a buffer containing 8 M urea, 4% w/v CHAPS,65 mM DTE, 0.8% v/v carrier ampholytes, and 5 mL of bro-mophenol blue), containing 0.8–2 mg of cytosolic proteins,for 8 h using a low voltage of 30 V at 167C. Each step of IEFwas carried out by using the same voltage regime both incase of NL IPG 3–10 or linear IPG 4–7 strips, according to aprogram driven by the Ettan IPGphor system (1 h at 120 V;30 min at 300 V; linear ramping from 300 to 3500 V in 3 h;10 min at 5000 V; and then 7950 V to reach at total of 62 KV/h). The focused IPG strips were incubated for 12 min atroom temperature in a first equilibration buffer containing6 M urea, 2% w/v SDS, 50 mM Tris (pH 6.8), glycerol 30%,and for 5 min in second equilibration buffer containing 2.5%w/v iodoacetamide and 2% w/v DTE [16]. The strips werehold in place with 0.4% low-melting temperature agaroseand loaded onto 20618 cm 9–16% or constant 15% SDS-polyacrylamide gels. Electrophoresis was carried out at aconstant current of 40 mA in a PROTEAN® II xi 2-D Cell(BioRad, Richmond, CA, USA) apparatus. The 2-D gels werestained with “Blue silver” according to the manufacturer’sinstructions [17]. Data were analyzed using the PDQuestVersion 7.2 software (BioRad).

2.4 Western blotting

The separated proteins were transferred onto NC mem-branes by using a Trans-Blot® Electrophoretic Transfer Cell(BioRad, Hercules, CA, USA) and applying a voltage of 100 Vfor 2 h. Individual membranes were incubated with mono-clonal and polyclonal anti-SOD1 specific antibodies for


4342 C. Di Poto et al. Electrophoresis 2007, 28, 4340–4347

Cu,Zn-SOD1 and polyclonal anti-Op18 for stathmin/onco-protein 18. Membranes were washed and proteins visualizedusing an ECL kit.

2.5 In situ digestion and MALDI-MS analysis

When 2-D runs were performed for preparative purposes,selected protein spots were excised from the gel and washedin 50 mM ammonium bicarbonate (pH 8.0) and 50% ACNuntil complete destaining. The proteins were dried undervacuum, resuspended in 50 mM ammonium bicarbonate(pH 8.0), and digested with sequencing grade trypsin (5 mg/mL, Promega, Madison, WI, USA) at 377C. After incubationovernight, peptides were extracted sequentially three timeswith 50% ACN, 5% TFA in water. Each extraction involved10 min of stirring followed by centrifugation and removal ofthe supernatant. The original supernatant and thoseobtained from sequential extractions were combined anddried. The peptide mixture was then solubilized with 0.5%TFA for MS analysis. MALDI-TOF analysis of the trypsindigest from the 2-D gel was performed using a Voyager DE-Pro Applied Biosystem mass spectrometer (Foster City, CA,USA) equipped with a pulsed N2 laser operating at 337 nm.Positive ions spectra were acquired in reflector mode over am/z range of 500–5000. The analysis was performed byspotting 1 mL of sample mixed with an equal volume of thematrix solution (a-cyano-4-hydroxycinnamic acid 10 mg/mLin 1:1 ACN/water containing 0.1% TFA) onto the targetplate. External mass calibration was performed using thepeptide mass standard kit provided by manufacturer. Data-base searches were performed against the NCBI non-redundant database using MASCOT (Matrix Science;www.matrixscience.com) search engine. MASCOT scoresgreater than 65 were significant (p ,0.05) and were subse-quently blasted against the Swiss-Prot database (http://www.expasy.ch/sprot).

2.6 LC-MS/MS analysis

When the identity of proteins could not be established byPMF, the peptide mixtures were further analyzed by LC-MS/MS using a CapLC capillary chromatography system (Waters,Massachusetts, USA) coupled online with a Q-TOF Ultimahybrid mass spectrometer (Micromass, Toronto, Canada)equipped with a Z-spray source. The peptide mixture (10 mL)was first loaded onto a RP trap column (Waters), using 0.2%formic acid as eluent, at a flow rate of 10 mL/min. The samplewas then transferred to a C18 RP capillary column(75 mm620 mm) at a flow rate of 280 nL/min and fraction-ated using a linear gradient from 7% eluent A (0.2% formicacid in 5% ACN) to 60% eluent B (0.2% formic acid in 95%ACN) in 50 min. The mass spectrometer was set up in a data-dependent MS/MS mode to alternatively acquire a full scan(m/z acquisition range from 400 to 1600 Da/e) and a tandemmass spectrum (m/z acquisition range from 100 to 2000 Da/e). The three most intense peaks in any full scan were

selected as precursor ions and fragmented by collisionenergy. For the peptide sequence searching, the mass spectrawere processed and analyzed using the MASCOT MS/MSion search program with the Swiss-Prot 49.1 database. Themass accuracy was within 50 ppm for the peptide mass tol-erance and within 0.25 Da for fragment mass tolerance.Protein identification was repeated at least once usingmatching spots from different gels.

2.7 N-Terminal sequence analysis

Sequence analyses were performed on the HP G1005A(Hewlett-Packard, Palo Alto, CA, USA) protein sequencingsystem using the routine 3.0 chemistry, according to themanufacturer’s protocol.

2.8 Reproducibility of the study

All gels were run in triplicate within the same day and mapspresented in this work are the best-representative 2-D gelsthat showed spots constantly present.

3 Results and discussion

3.1 Separation of cytosolic proteins on NL IPG 3–10

strips and immunoblotting

Total cytosolic proteins (750 and 800 mg) from SH-SY5Y hu-man neuroblastoma cells (used as model of reference), andfrom the same cells stably transfected with H46R humanSOD1, were separated “in parallel” for the first dimension on18 cm, pH 3–10 NL IPG gel strips. The focused IPG stripswere then loaded onto 20618 cm, 9–16% SDS-polyacryl-amide slabs for the second-dimensional run. Detailed mapsof proteins exhibited by these samples after staining withcolloidal Coomassie are shown in Fig. 1, panels a and b,respectively. Given the aim of our study, from among a totalof 176 and 289 spots observed in protein patterns a and b, wefocused our interest only on Coomassie-reactive proteinsdisplaying reactivity against anti-SOD1 antibody. Thus, todetermine the exact location of SOD1 spots on the gels, pro-teins separated by 2-DE were electroblotted onto NC mem-branes coated with a rabbit polyclonal anti-SOD1 antibody.The good matching between the immunosignals shown inpanels c and d of Fig. 1 and spots I–III or I–IV in the regionof the maps marked by dotted lines in panels a and b, allowedto assign tentatively SOD1 identity to these latter. To achievethe unambiguous identification of these proteins, reactivespots I–IV in panel b were excised, washed to a completedestaining and submitted to in-gel trypsin digestion follow-ing the procedure described in Section 2. PMF of spots anddatabase search with the peptide mass spectra obtainedallowed to produce the data summarized in Table 1. Spots IIIand IV were easily identified as human Cu,Zn-SOD1 (Swiss-Prot accession number P00441 and NCBI accession number



Figure 1. 2-DE gel images of total cytosol proteins from SH-SY5Y human neuroblastoma cells (“control”, panel a) and from the same cellsstably trasfected with H46R human SOD1 (panel b). Separation was performed on 18 cm long IPG strips with a wide-range NL pH gradient3–10, followed by SDS-PAGE in a 9–16% linear gradient gel. Protein detection was by Coomassie staining. Panels c and d: blot of the win-dows defined in panels a and b, respectively. Arrows indicate spots displaying SOD1 activity (annotated I–III in panel c and I–IV in panel d)that were located by matching the two images and that were further identified by MALDI-TOF MS. Proteins annotated V and VI will bediscussed below (see caption of Fig. 3).

Table 1. List of cytosolic proteins identified in spots indicated I–IV in Fig. 1, panel b

Spot Protein NCBI ObservedpI

TheoreticalpI

Observedmol. mass(kDa)

Theoreticalmol. mass(kDa)

Sequencecoverage(%)

MOWSEscore

I Stathmin1/oncoprotein 18 gi/61363614 5.50 5.76 19.75 17.35 66 69II Stathmin1/oncoprotein 18 gi/61363614 5.58 5.76 19.00 17.35 80 132III SOD1 gi/38489880 5.81 5.70 19.75 16.05 83 122IV Variant SOD1 gi/31615965 5.92 5.70 19.75 16.05 80 119

gi/38489880) and Apo-H46R Familial ALS Mutant HumanCu,Zn-SOD1 (Swiss-Prot accession number P00441 andNCBI accession number gi/31615965), respectively. In con-trast, the two spots I and II were identified as stathmin, alsoreferred to as oncoprotein 18 (Swiss-Prot accession numberP16949; NCBI accession number gi/61363614). Stathmin isa regulatory phosphoprotein known to play the role of a sec-

ond toxic messenger in the pathogenesis of neurodegenera-tion in FALS with SOD1 mutations [18]. Although the loca-tion of stathmin in the region of the gel mentioned abovewas in good agreement with its molecular size (Mr around17.225 Da and pI value around pH 5.89 according to ref.[19]), nevertheless the fact that this protein reacted with anti-SOD1 antibody generated ambiguity. Thus, to examine whe-



ther recognition of stathmin could be ascribed to nonspecificcrossreactivity exhibited by the antibody adopted, the blottingwas repeated using a mouse monoclonal anti-SOD1 anti-body. The results obtained, however, were rather disappoint-ing, the distribution of spots responsive to mAb being iden-tical to that shown in panel c. It means that MS spectra ofspots I–IV produced the same results previously reported inTable 1, i.e., confirmed the presence of SOD1 under spots IIIand IV and of stathmin under spots I an II (data not shown).It was apparent that spots I and II, rather than being homo-geneous spots comprising a single polypeptide chain, wereenvelopes of two or more components. The fact that theputative SOD1 entity was, most likely, the less abundantprotein in the bulk of other more abundant proteins, ren-dered its identification difficult. This hypothesis wasstrengthened by the insufficient quality of MS sequence datawhich made it difficult to determine exactly, from this analy-sis, which might be the proteins of interest within the spotsconsidered. Attempts to enrich the content of putative SOD1in spots I and II by applying on the gel high-protein load(runs with 1 and 2 mg of protein mixture have been per-formed) were not successful, the MS sequence data of stath-min being in any case the only good-quality MS data detectedin these spots. Thus, given that the current approach wasunlikely to precisely decipher the different proteins in spotsof interest, it became imperative to answer the question ofhow to resolve these spot overlaps.

3.2 Separation on a narrow-range IPG 4–7 strips

According to the numerous reports so far appeared in theliterature [16, 20, 21], the best option to improve resolution

and thereby minimize spot overlaps seemed to be the use ofnarrow-range (such as pH 4–7 or 4.5–5.5) IPG strips. In anattempt to maximize the number of detectable proteins inregion considered, we performed in the 1-D a series of runson an 18 cm long pH 4–7 gel. The results shown in Fig. 2,which represents the typical profiling achieved under theexperimental conditions mentioned above, were discourag-ing. Although an advantage could be expected from theimproved separation of spots of interest, their mass spectracompared well to those previously discussed and recon-firmed the presence of SOD1 under spots III and IV and ofstathmin under spots I and II. Despite supplementary runsusing 4.5–5.5 pH strips have been performed to circumventthese difficulties, no improvement was observed in thenumber and position of spots in comparison with profile ofFig. 2 (data not shown).

In a further effort to overcome these limitations and todefinitively establish or exclude the presence of SOD1 underspots I and II, we worked also on the optimization of gelporosity.

3.3 Combination of narrow-range IPG 4–7 strip and

linear 15% SDS-PAGE

A combination of a narrow pH range (pH 4–7) for the first-dimensional IEF and of a linear 15% porosity in the second,provided a powerful tool for obtaining a comprehensiveinventory of protein spots in region considered (Fig. 3, panel a).By visual inspection of the profiling shown in Fig. 3 (panel a),it was clear that this approach exhibited a performancehigher than that of procedures previously applied yielding a

Figure 2. 2-DE gel image of totalcytosolic proteins separated onan 18 cm long IPG strip with anarrow-range pH gradient 4–7,followed by SDS-PAGE in a 9–16% linear gradient gel. Proteindetection was by Coomassiestaining. Spots annotated I–IVare the same as those previouslyobserved in Fig. 1.



map that, at least in the region of interest, contained a greaternumber and more distinct spots. The eight protein spots I–VIII arrowed in Fig. 3, panel a were submitted to MS analysisand all of them, with the exception of spots V and VI, forwhich the quality of MS signal was insufficient, could beidentified. The results of these analyses may be summarizedas follows: spots I, II, and VII corresponded to stathmin(Swiss-Prot accession number P16949; NCBI accessionnumber gi/61363614); spots III and IV corresponded toSOD1 (Swiss-Prot accession number P00441; NCBI acces-sion number gi/38489880) and Apo-H46R Familial ALSMutant Human SOD1, respectively (Swiss-Prot accessionnumber P00441 and NCBI accession number gi/31615965);and spot VIII corresponded to dUTP pyrophosphatase(NCBI accession number gi/4503423). Once again the qual-ity of MS signal for spots V and VI was not sufficient to allowtheir unambiguous identification.

3.4 Identification of protein in spots V and VI by

means of LC-MS/MS

To improve the quality of MS signal for spots V and VI,separation was replicated under the experimental conditionsindicated above loading the gel with 2 mg of total cytosolicproteins from transfected cells. A good matching betweenspots obtained in this run and those of previous runs allowedto select with accuracy the candidate proteins for MS analy-ses. Spots of interest were excised from the gel, digested withtrypsin, and the resulting peptides analyzed by LC-MS/MS.For the peptide sequence searching, mass spectra were pro-cessed and analyzed using the Swiss-Prot 49.1 database. Upto four and six peptides with high-score sequence werematched from spots V and VI, respectively. The sequencecoverage of proteins was 60% for spot V and 66% for one ofthe two entities in spot VI. The results of MS analyses (sum-marized in Table 2) allowed to identify in spots V and VIpossible (iso)forms of SOD1 and mutant H46R SOD1,respectively, that had never been observed before. The latterentity in spot VI was also identified as SOD1 (NCBI: gi/223632), although with lower sequence coverage (43%) andMOWSE score (101) than the former. Four peptides (cover-ing the sequence regions: 10–23; 24–36; 116–128; and 144–153) out of a total of five identified by MS, were the same asthose shown in Table 2. The fifth peptide (covering thesequence 10–30) was peculiar of this entity. The findingsdescribed above were further validated by blotting the labeledregion of Fig. 3, panel a onto a NC membrane coated withanti-SOD1 mAb. As shown in panel b of the same figure, two(namely spots III and IV), out of the four spots responsive tomAb, were the SOD1 entities identified from the beginningof our experimental approach. The other two, namely spots Vand VI, were the newly separated SOD1 forms that over-lapped spots of stathmin (spots I and II) in former gels butwere well resolved in this latter. The dotted “empty” circleson the membrane indicate the former position (previouslyshown in Figs. 1 and 2) of stathmin plus SOD1 spots.

To prove definitively our assumption, supplementaryexperiments based on the determination of N-terminal pro-tein sequence were performed. An aliquot of protein recov-ered from gel spots V and VI was submitted to automatedEdman degradation providing the results shown in thescheme below:

10 20 30

SOD1 Sequence [22] atkavcvlkg dgpvqgiinf eqkesngpvKSpot V Sequence atkavavlkg dgpvqgiinf eqkesngpvKSpot VI Sequence atkavcvlkg dgpvqgiinf eqkesngpvK

Taken together, all these data are unambiguously con-sistent with the presence of SOD1/mutant SOD1 entitiesunder spots V and VI of protein pattern considered.

Figure 3. Panel a: 2-DE gel image of total cytosolic proteinsseparated on an 18 cm long IPG strip with a narrow-range pHgradient 4–7, followed by SDS-PAGE in a constant 15% gel. Pro-tein detection was by Coomassie staining. Selected spots (I–VIII)analyzed by MS or LC-MS/MS are shown in the region marked bydotted lines. Panel b: Blot of the gel region indicated above onto aNC membrane coated with mouse monoclonal anti-SOD1 anti-body. Selected spots (III–VI) analyzed by MS or LC-MS/MS corre-sponded to SOD1. The dotted empty circles on the membrane(annotated I and IV) indicate the former position of stathmin plusSOD1 previously shown in Figs. 1 and 2.



Table 2. List of tryptic peptides identified, by LC-MS/MS, in spots indicated V and VI in Fig. 3, panel a

Spot Protein NCBI Matched peptides Sequencecoverage(%)

MOWSEscore

ObservedpI

TheoreticalpI

Observedmol. mass(kDa)

Theoreticalmol. mass(kDa)

Start–end(No. ofpeptide)

Sequence (aa)

V SOD1 gi/349912 10–23 GDGPVQGIINFEQK

60 116 5.55 5.70 19.75 16.05

24–36 ESNGPVKVWGSIK

116–128 TLVVHEKADDLGK

144–153 LACGVIGIAQ

VI SOD1 gi/31615955 10–23 GDGPVQGIINFEQK

66 232 5.60 5.70 19.75 16.05

24–36 ESNGPVKVWGSIK

37–46 GLTEGLHGFR

47–69 VHEFGDNTAGCTSAGPHFNPLSR

116–128 TLVVHEKADDLGK

144–153 LACGVIGIAQ

Aim of the present report was to verify the presence ofdifferent SOD1 forms in a cellular model of ALS by applyingthe classical proteomics procedures. We have experiencedthat a series of drawbacks, including a large difference inabundance between proteins with very similar pI and Mr,prevent the separation and characterization of spots of inter-est. The limited sensitivity in the detection of less abundantproteins in fact made their identification difficult in the bulkof other more abundant entities. Nevertheless, the properchoice of pH gradient in the first dimension, of gel porosityin the second and of high-resolution mass spectrometricstrategies allowed to obtain their successful separation andunambiguous identification.

4 Concluding remarks

A combination of immunoblotting, LC-MS/MS, and N-ter-minal sequence analysis made identification of differentforms of SOD1 in human neuroblastoma SH-SY5Y plasmidpRc/CMV unambiguous. Since these newly identifiedSOD1-containing spots are located in the gel adjacent to oneanother and to spots of genuine SOD1, they are most likelycharacterized by subtle post-translational modifications suchas charge variations. In this context, it is tempting to spec-ulate that they may be referred to as “SOD1 isoforms” al-though further analysis is required to elucidate their struc-tural modifications and, possibly, their precise role and

function in ALS. To the best of our knowledge, this is the firstreport in which the presence of different forms of SOD1 inhuman neuroblastoma cell cultures is demonstrated.

The authors would like to thank Professor Maria TeresaCarrì, Università di Roma “Tor Vergata”, for invaluable supportin the construction of transfected SH-SY5Y cell line and Dr.Angela Flagiello, Università di Napoli Federico II, for performingMALDI-MS analyses. Fabio Ferrari, Marco Fumagalli, and Ser-ena Giuliano (Department of Biochemistry, University of Pavia)are also gratefully acknowledged for their technical support.

5 References

[1] Choi, J., Ress, H. D., Weintraub, S. T., Levey, A. I. et al., J. Biol.Chem. 2005, 280, 11648–11655.

[2] Ciechanover, A., Brundin, P., Neuron 2003, 40, 427–446.

[3] Bossy-Wetzel, E., Schwarzenbacher, R., Lipton, S., Nat. Med.2004, S2–S9.

[4] Beal, M. F., Free Radical Biol. Med. 2002, 32, 797–803.

[5] Cova, E., Cereda, C., Galli, A., Curti, D. et al., Neurosci. Lett.2006, 399, 186–190.

[6] Ferrante, R. J., Browne, S. E., Shinobu, L. A., Bowling, A. C. etal., J. Neurochem. 1997, 69, 2064–2074.

[7] Shibata, N., Nagai, R., Uchida, K., Horiuchi, S. et al., BrainRes. 2001, 917, 97–104.



[8] Choi, J., Levey, A. I., Weintraub, S. T., Rees, H. D. et al., J.Biol. Chem. 2004, 279, 13256–13264.

[9] Dawson, T. M., Dawson, V. L., Science 2003, 302, 819–822.[10] Valentine, J. S., Hart, P. J., Proc. Natl. Acad. Sci. USA 2003,

100, 3617–3622.[11] Deng, H. X., Hentati, A., Tainer, J. A., Iqbal, Z. et al., Science

1993, 261, 1047–1051.[12] Ogasawara, M., Matsubara, Y., Narisawa, K., Aoki, M. et al.,

Nat. Genet. 1993, 5, 323–324.[13] Polticelli, F., Battistoni, A., Bottaro, G., Carrì, M. T. et al.,

FEBS Lett. 1994, 352, 76–78.[14] Carrì, M. T., Ferri, A., Battistoni, A., Famhy, L. et al., FEBS

Lett. 1997, 414, 365–368.[15] Yang, J., Liu, X., Bhalla, K., Kim, C. N. et al., Science 1997,

275, 1129–1132.

[16] Görg, A., Weiss, W., Dunn, M. J., Proteomics 2004, 4, 3665–3685.

[17] Candiano, G., Bruschi, M., Musante, L., Santucci, L. et al.,Electrophoresis 2004, 25, 1327–1333.

[18] Strey, C. W., Spellman, D., Stieber, A., Gonatas J. O. et al.,Neurobiology 2004, 165, 1701–1718.

[19] Cheon, M. S., Fountoulakis, M., Cairns, N. J, Dierseen et al.,J. Neural Transm. Suppl. 2001, 61, 281–288.

[20] Wildgruber, R., Harder, A., Obermaier, C., Boguth, G. et al.,Electrophoresis 2000, 21, 2610–2616.

[21] Westbrokk, J. A., Yan, J. X., Wait, R., Welson, S. Y., Dunn, M.J., Electrophoresis 2001, 22, 2865–2871.

[22] Jabusch, J. R., Farb, D. L., Kerschensteiner, D. A., Deutsch, H.F., Biochemistry 1980, 27, 2310–2316.


Optimizing separation efficiency of 2-DE procedures for visualization of different superoxide...

Documents

Transcript of Optimizing separation efficiency of 2-DE procedures for visualization of different superoxide...