Se

PNa

b

c

d

a

ARRAA

KSTLsS

1

aam[sD[amcd

0h

Journal of Pharmaceutical and Biomedical Analysis 90 (2014) 72– 77

Contents lists available at ScienceDirect

Journal of Pharmaceutical and Biomedical Analysis

jou rn al hom epage: www.elsev ier .com/ locate / jpba

uperoxide dismutase isozyme detection using two-dimensional gellectrophoresis zymograms

loypat Niyomploya, Chantragan Srisomsapb, Daranee Chokchaichamnankitb,awaporn Vinayavekhinc, Aphichart Karnchanatatd, Polkit Sangvanichc,∗

Program in Biotechnology, Faculty of Science, Chulalongkorn University, Bangkok 10330, ThailandLaboratory of Biochemistry, Chulabhorn Research Institute, Bangkok 10210, ThailandDepartment of Chemistry, Faculty of Science, Chulalongkorn University, Bangkok 10330, ThailandInstitute of Biotechnology and Genetic Engineering, Chulalongkorn University, Bangkok 10330, Thailand

r t i c l e i n f o

rticle history:eceived 22 July 2013eceived in revised form 18 October 2013ccepted 23 October 2013vailable online 20 November 2013

eywords:uperoxide dismutasewo-dimensional gel electrophoresisiquid chromatography–masspectrometry (LC–MS)temona tuberosa

a b s t r a c t

Superoxide dismutases (SODs) are ubiquitous antioxidant enzymes involved in cell protection from reac-tive oxygen species. Their antioxidant activities make them of interest to applied biotechnology industriesand are usually sourced from plants. SODs are also involved in stress signaling responses in plants, andcan be used as indicators of these responses. In this article, a suitable method for the separation of dif-ferent SOD isoforms using two-dimensional-gel electrophoresis (2D-GE) zymograms is reported. Themethod was developed with a SOD standard from bovine erythrocytes and later applied to extractsfrom Stemona tuberosa. The first (non-denaturing isoelectric focusing) and second (denaturing sodiumdodecylsulphate-polyacrylamide gel electrophoresis) dimensions of duplicate 2D-GE gels were stainedwith either Coomassie brilliant blue G-250 for total protein visualization, or SOD activity (zymogram)using riboflavin/nitroblue tetrazolium. For confirmation, putative SOD activity positive spots were subjectto trypsin digestion and nano-liquid chromatography tandem mass spectrometry, followed by searchingthe MASCOT database for potential identification. The method could separate different SOD isoforms from

a plant extract and at least partially maintain or allow renaturation to the native forms of the enzyme.Peptide sequencing of the 2D-GE suggested that the SODs were resolved correctly, identifying the con-trol CuZn-SOD from bovine erythrocytes. The two SODs from S. tuberosa tubers were found to be likelyhomologous of a CuZn-SOD. SOD detection and isoform separation by 2D-GE zymograms was efficientand reliable. The method is likely applicable to SOD detection from plants or other organisms. Moreover,a similar approach could be developed for detection of other important enzymes in the future.

. Introduction

Superoxide dismutase (SOD, EC 1.15.1.1) is an antioxidant met-lloenzyme [1]. The different isozymes are currently classifiedccording to the metal cofactor in their active sites into the fourain and broad types of CuZn-SOD, Mn-SOD, Fe-SOD and Ni-SOD

2–5]. SODs have an important role in catalyzing the destruction ofuperoxide radicals, which are cytotoxic agents to cell membranes,NA and other biomolecules, to hydrogen peroxide and oxygen

6–8]. In general, SODs have been found in microorganisms, animalsnd plants. The SODs used as antioxidant agents for applications in

edicine, and in the cosmetic, chemical and food industries are

urrently sourced and extracted (or cloned and recombinant pro-uced) from plants [9]. Furthermore, at least some SOD isoforms

∗ Corresponding author. Tel.: +66 22187637.E-mail address: polkit@gmail.com (P. Sangvanich).

731-7085/$ – see front matter © 2013 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.jpba.2013.10.035

© 2013 Elsevier B.V. All rights reserved.

in plants are involved in the defense against pathogens and insignaling responses for various stresses. Since the different SODisoforms in plants show different responses to infection and stress[10–12], the level of expression of specific SOD isozymes in key tis-sues can be used as an indicator to determine the growth stage orstress/infection circumstances of that plant [13].

Currently several methods have been reported for SOD detectionwith one of the popular methods being detection by zymograms.This method detects active enzymes across the spectrum of thefour broad types of SODs, rather than, for example, immunologicaldetection which often cannot discriminate between enzymicallyactive and inactive forms, and can only detect those isoformswith conserved epitopes. The original zymogram detection methodfor SODs was created by Beauchamp and co-workers and used

native polyacrylamide gel electrophoresis (Native-PAGE) followedby enzymatic staining for SOD activity [14]. There are two combinedreactions in the SOD activity assay. The first is the autooxi-dation from riboflavin (oxidizing agent) and the second is the

utical

raaNt(iasnb

iwri(fstsorc

(td

tgtdctu(eiu

2

2

fsH

apwgAumpc1ttaea

P. Niyomploy et al. / Journal of Pharmace

iboflavin/nitroblue tetrazolium (NBT) reduction, which uses NBTs a chromogenic substrate. The SOD enzyme can be determined asn achromatic zone on Native-PAGE because the enzyme inhibitsBT reduction [15], but Native-PAGE does not allow determina-

ion of valuable information, such as the subunit molecular weightMW), nor discriminate between isozymes with a different MW orsoelectric point (pI). However, the simultaneous detection of SODctivity and evaluation of their MW on one dimensional denaturingodium dodecyl sulphate-PAGE (1D-SDS-PAGE) zymograms can-ot separate the different SOD isoforms that share a similar MW

ut differ in their pI.Two-dimensional gel electrophoresis (2D-GE) is one of the

mportant techniques for protein detection and is often coupledith tryptic digestion and mass spectrometry (MS) to identify the

esolved protein(s) [16]. Proteins are separated by 2D-GE accord-ng to their pI in the first dimension and their apparent MW

three dimensional size) in the second dimension, using isoelectricocusing (IEF) and denaturing SDS-PAGE, respectively [17]. Con-equently, proteins with a different pI but similar MW, as well ashose that differ in apparent MW, can be resolved even in complexamples, such as different SOD isozymes from a whole organismr tissue extracts [18]. According to 2D-GE theory, the capacity ofesolution for the method is up to 10,000 different proteins, and itan detect less than 2 ng per spot [17].

However, due to the denaturing running conditions of 2D-GESDS, dithiothreitol and urea plus sample heating) this method isypically unsuitable for SOD (and many other enzymes) zymogramsue to the inactivation of the enzyme [16].

Accordingly, in this research, a 2D-GE approach was developedhat appears to maintain SOD activity and so it is suitable for zymo-ram based detection of enzymically active SOD isoforms fromissue extracts. The method combined a non-denaturing IEF firstimension and SDS-PAGE second dimension, before staining dupli-ate gels with either Coomassie blue or SOD activity. Finally, in-gelrypsin digestion and peptide extraction was coupled with liq-id chromatography tandem MS (LC–MSMS) of a known standardCuZn-SOD from bovine erythrocytes) and an experimental tissuextract (crude total protein extract from the root of the medic-nal plant Stemona tuberosa Lour.) to evaluate the reliability andsability of the method.

. Methods and materials

.1. Isolation and extraction of SOD from S. tuberosa

Fresh tubers of S. tuberosa (∼3 kg, fresh weight) were purchasedrom Chatujak market, Bangkok, Thailand, in June 2009. A voucherpecimen (BK 244965) is deposited at the Sirindhorn Bangkokerbarium, Bangkok, Thailand.

The fresh roots of S. tuberosa were peeled, cut into small cubesnd homogenized in 5 L of extraction buffer (0.1 M NaCl, 20 mMhosphate buffer pH 7.2) at 4 ◦C. The suspension-solution mixtureas stirred at 4 ◦C overnight before being clarified by centrifu-

ation (10,178 × g, 30 min, 4 ◦C) and harvesting the supernatant.mmonium sulfate was then added to the supernatant to 90% sat-ration and left overnight at 4 ◦C, before harvesting the precipitatedaterial by centrifugation (15,904 × g, 30 min, 4 ◦C), dissolving the

recipitate in 400 mL de-ionized water, and dialyzing against 4–5hanges of 1000 mL each of 20 mM phosphate buffer pH 7.2 over8 h at 4 ◦C. The dialyzed extract was then freeze-dried to yieldhe dark brown crude protein extract, from which the IC50 value of

he SOD preparation was determined as reported [19]. In addition,

commercial preparation of the CuZn-SOD isozyme from bovinerythrocytes (Merck, USA), a homodimer enzyme of 32.50 kDa MW

nd a pI of 4.95, was used as a positive control.

and Biomedical Analysis 90 (2014) 72– 77 73

2.2. Resolution and detection of SOD isozymes

2.2.1. One dimensional reducing sodium dodecylsulphate-polyacrylamide gel electrophoresis (1D-SDS-PAGE)

Reducing 1D-SDS-PAGE was run according to the modifiedmethod as previously reported [19,20]. Either 12.5 �g (per trackof the gel) of the reference pure CuZn-SOD homodimer enzymefrom bovine erythrocytes, as a positive control, or 25 �g (per trackof the gel) crude protein extract from S. tuberosa tubers was mixedwith reducing sample buffer (62.5 mM Tris–HCl pH 6.8, 14.4 mM2-mercaptoethanol, 10%, v/v glycerol, 2%, w/v SDS) and 1% (w/v)bromophenol blue at room temperature and then subjected toduplicate 1D-SDS-PAGE resolution using a 10% or 12.5% (w/v) acryl-amide resolving gel for the bovine erythrocyte CuZn-SOD or thecrude protein extract from S. tuberosa, respectively. After elec-trophoresis, one of each pair of duplicate gels was stained withCoomassie brilliant blue G-250 to visualize the protein bands,whilst the other, as a SOD zymogram, was washed and stained forSOD enzyme activity (see Section 2.3) to determine the presence ofenzymically active SOD.

Crude protein extracts were also mixed with reducing samplebuffer or non-reducing sample buffer (62.5 mM Tris–HCl pH 6.8,10%, v/v glycerol and 1%, w/v bromophenol blue) and applied (25 �gper track of the gel) to a 1D-SDS-PAGE and 1D-Native-PAGE (both12.5%, w/w gel resolving gel), respectively, for comparison of theSOD activity level.

2.2.2. Non-denaturing two-dimensional polyacrylamide gelelectrophoresis (2D-GE)

The reference CuZn-SOD isozyme from bovine erythro-cytes (40 �g per gel), and the crude protein extract fromS. tuberosa (150 �g per gel) were dissolved in lysis solution(40 mM Tris, 4%, w/v 3-(3-cholamidopropyl) dimethylammonio-1-propanesulfonate (CHAPS), 1 mM ethylenediaminetetraaceticacid(EDTA), 2%, v/v immobilized pH gradient (IPG) buffer, 10%, v/v glyc-erol and 1%, w/v bromophenol blue), vortexed every 30 min for1–2 h and left on ice as previously reported [21]. Each protein sam-ple was loaded onto duplicate 7-cm, pH 3–6 IPG gel strips (Bio-RADLaboratories, CA, USA) and left overnight at room temperature (RT).The first-dimensional IEF electrophoresis was performed at 4 ◦C ona Pharmacia LKB Multiphore II system at 300 V for 1200 Vh and thenincreasing the voltage step wise to 1000 V for 300 Vh, 5000 V for4500 Vh and 5000 V for 1000 Vh. After IEF, the IPG strips were equil-ibrated in equilibration buffer (50 mM Tris–HCl buffer pH 6.8, 6 Murea, 1%, w/v SDS, 30%, v/v glycerol, 1%, w/v dithiothreitol (DTT))and were alkylated with equilibration buffer containing 2.5% (w/v)iodoacetamide (IAA). Finally, the equilibration IPG strips were sep-arated in the second dimension using the reducing SDS-PAGE andthen the duplicate gels stained for protein and SOD activity, respec-tively, as described in Sections 2.2.1 and 2.3, respectively.

2.3. Gel washing procedures and SOD staining activity assay

After electrophoresis, the 1D-SDS-PAGE or 2D-GE gel waswashed as reported previously [22,23]. In brief, the gel was firstsoaked twice in 100 mL of 25% (v/v) isopropanol in 0.01 M Tris pH 7for 10 min each at RT to remove the SDS. The gel was then washedtwice with 100 mL of 2 �M ZnCl2/0.01 M Tris pH 7 for 10 min eachat RT to remove the isopropanol, then twice with 100 mL of 0.1 MTris pH 7 for 20 min each, and once with 100 mL of 0.01 M Tris pH7 for 10 min, all at RT.

2.4. SOD isozyme identification by in-gel trypsin digestion and

tryptic peptide mass spectrometry analysis

Protein spots on the Coomasie stained 2D-GE that matched thespots on the SOD activity zymogram in the paired duplicate gel

7 utical and Biomedical Analysis 90 (2014) 72– 77

wbw2o4afgpdtaootttp15Vi

2m

t(cflcs(S±

3

3e

weS

oetp(Htam

aspp(ta

Table 1Comparison of the reported SOD activity (as IC50 values) from 10 plants.

Plants IC50 value (�g/mL) Reference

Wheat germ 250 [25]Alfalfa leaf 700 [26]Sphenostylis stenocarpa >1000 [27]Pea (F2) 1000 [28]Gnetum gnemon (Gg-AOPI) 40 [29]Melon 800 [30]Hedychium coronarium 0.729 ± 0.008 [31]Kaempferia galanga Linn. 0.681 ± 0.003 [31]Zingiber officinale Roscoe. 0.432 ± 0.003 [31]

4 P. Niyomploy et al. / Journal of Pharmace

ere excised and washed three times in 100 �L distilled waterefore cutting into small pieces. The gel pieces were then washedith 0.1 M NH4HCO3 in 50% (v/v) acetonitrile (ACN) at 30 ◦C for

0 min to de-stain, dried in a Speed Vacuum and incubated in 50 �Lf buffer solution (0.1 M NH4HCO3/10 mM DTT/1 mM EDTA) for5 min at 60 ◦C to reduce the gels. The supernatant was discardednd the gels incubated in 50 �L of 100 mM IAA in 0.1 M NH4HCO3or 30 min at RT in the dark. The supernatant was removed and theels were washed three times with 50 �L each of 0.05 M Tris–HClH 8.5 in 50% (v/v) ACN and then dried in a Speed Vacuum. Trypticigestion was performed by the addition of 30 �L of trypsin diges-ion buffer (0.05 M Tris–HCl pH 8.5, 0.1 �g/�L trypsin in 1%, v/vcetic acid, 10%, v/v ACN and 1 mM CaCl2) to the excised regionf the gel containing the desired spot and incubating at 37 ◦Cvernight. The reaction was stopped by adding 20 �L of 2% (v/v)rifluoroacetic acid and incubating for 30 min at 60 ◦C, and thenhe supernatant was harvested. The gels were then extracted threeimes with 50 �L 2% (v/v) trifluoroacetic acid and 0.05 M Tris–HClH 8.5 containing 1 mM CaCl2 and 2% (v/v) formic acid, and for0 min at 30 ◦C on a shaker (1500 rpm), and then sonicated for

min. The harvested solutions were pooled and dried in a Speedacuum for further characterization by nano-LC–MSMS as detailed

n Section 2.5.

.5. SOD characterization by liquid chromatography tandemass spectrometry (LC–MSMS)

Each tryptic digested zymogen spot protein (puta-ive SOD enzyme, see Section 2.4) was mixed with 0.1%v/v) formic acid before being analyzed by nano liquidhromatography-electrospray ionization quadrupole-time ofight MS (nano-LC–ESI–MSMS) using an EASY-nLCII spectrometeroupled with a MicroTOF QII (Bruker, Germany). The tandem masspectra of the tryptic peptides were searched from Mascot databasehttp://www.matrixscience.com/cgi/search formpl?FORMVER=2&EARCH=MIS). The precursor and MSMS tolerances were set to1.2 Da and ±0.6 Da, respectively.

. Results and discussion

.1. Evaluation of the 2D-GE method using the bovinerythrocyte CuZn-SOD isozyme as a known standard

The CuZn-SOD from bovine erythrocytes, a known homodimerith a pI of 4.95 and MW of 32.5 kDa, was used to determine the

fficiency of the developed 2D-GE method for the detection of theOD enzyme.

Fig. 1 shows a representative 1D-SDS-PAGE and 2D-GE res-lution gel zymogram of the CuZn-SOD isozyme from bovinerythrocytes after staining for SOD activity (Fig. 1A and B) or for pro-ein (Fig. 1C and D). The 1D-SDS-PAGE resolution revealed a strongroteinband at ∼15–16 kDa (Fig. 1C) with a weak SOD activityFig. 1A), which presumably represents the monomer (15–16 kDa).owever, the intensity of this protein spot is markedly reduced in

he 2D-GE (Fig. 1D, spot 6), and showed no detectable SOD stainingctivity (Fig. 1B). This is potentially due to the renaturation of theonomer to homodimeric units.The 1D-SDS-PAGE gel also revealed three protein bands at

round 55–58 and >66 kDa (Fig. 1C), at least two of which had atrong SOD activity in the zymogram (Fig. 1A). These were alsootentially found in the 2D-GE as two spots at 55–58 kDa with a

I of 4.95 (Fig. 1D, spots 1 and 2) that also displayed SOD activityFig. 1B). These may be the homodimer but in native forms, wherehe abnormal (low) levels of bound SDS reduce the charge densitynd so accounts for the apparently higher MW in terms of the gel

Curcuma zedoaria Roscoe. 0.259 ± 0.006 [31]Stemona tuberosa 0.039 ± 0.006 Fig. S3

mobility [24]. If so, although these results show an advantage of themethod in that it can maintain the native form of the SOD enzyme(or allow its partial renaturation) even though the second dimen-sion was run under a denaturing condition, it also suggests that thediscrimination of isoforms by MW, and the assignment of MW fromthe 2D-GE are both unreliable.

Accordingly, to ascertain if the two large Coomassie blue stainedspots on the 2D-GE gel (spots 1 and 2 in Fig. 1D) were the CuZn-SOD,the spots were cut out, digested by trypsin and the resultant peptidefragment sequences identified by nano-LC–EIS–MSMS were usedto search the Mascot database to identify their likely source. Bothspots 1 and 2 matched the Cu-Zn SOD from Bos taurus with an aminoacid match of 59/59, which over the 152 amino acids gives a 38.8%sequence coverage (Fig. 2A and Supplementary Table S1).

3.2. SOD isoform separation from a crude protein extract of S.tuberosa tubers

To evaluate the efficiency of this 2D-GE method in real situa-tions, the crude protein extract from S. tuberosa tubers was used asa model system. The initial choice of S. tuberosa was based upon itsSOD activity, since the derived SOD activity (Supplementary Figs. S1and S2) had a higher (lower IC50 value in the NBT reduction assay)than that reported in nine other plants (Table 1).

The crude protein extract (150 �g) from S. tuberosa was ini-tially resolved by 2D-GE using a pH 3–10 linear IEF strip (Fig. 3).However, most of the proteins, including those that displayed SODactivity, were found to be in the acidic zone resulting in poor res-olution. To solve this problem, a narrow range linear IEF strip ofpH 3–6 was used in the first dimension IEF to improve the resolu-tion of the individual SOD proteins from each other and from theother non-SOD active proteins. After comparing the spots on thegel stained with Coomassie blue with those tested for SOD activity,three spots that potentially represent different (active) SOD iso-forms were revealed. Two of these had an apparently similar MW

of 53 kDa but slightly different pI values at 4.25 and 4.95 (spots 3and 4 in Fig. 4D, respectively), while the third was smaller at 38 kDaand more acidic with a pI of 4.0 (spot 5 in Fig. 4D).

These three SOD-active spots (spots 3–5) were excised fromthe corresponding Coomassie blue stained 2D-GE, digested withtrypsin and then analyzed via nanoLC–ESI–MSMS. However, onlyspots 3 and 5 could be identified, presumably because the limiteddata in the plant protein sequence databases prevented matches ofthe tryptic peptides from spot 4. Those from spot 3 matched to theCuZn-SOD from Ananas comosus at 25/25 amino acids (a coverage of16.4% over the total 152 amino acids), whilst spot 5 matched to theCuZn-SOD from Solanum lycopersicum and Zantedeschia aethiopica

at 10/10 amino acids (a coverage of 6.6% over the total 152 aminoacids) (Fig. 2C and D, respectively). The results from the 2D-GE cou-pled with SOD staining activity, therefore, suggest the presence ofat least three different active SODs in S. tuberosa. More importantly,

P. Niyomploy et al. / Journal of Pharmaceutical and Biomedical Analysis 90 (2014) 72– 77 75

Fig. 1. Representative (A and C) 1D-SDS-PAGE and (B and D) 2D-GE of the CuZn-SOD isozyme (12.5 and 40 �g, respectively) from bovine erythrocytes after staining for (Aand B) SOD activity or (C and D) Coomasie blue for protein. Note that the spot appearing at ∼30 kDa with pI ∼ 4.0 in C and D corresponds to the carbonic anhydrase enzymewithin the commercial SOD standard.

(A)

(B)

(C)

(D)

1 MATKAVCVLK GDGPVQGTIH FEAKGDTVVV TGSITGLTEG DHGFHVHQFG51 DNTQGCTSAG PHFNPLSKKH GGPKDEERHV GDLGNVTADK NGVAIVDIVD

101 PLISLSGEYS IIGRTMVVHE KPDDLGRGGN EESTKTGNAG SRLACGVIGI151 AK

1 MVKAVAVLGS SEGVKGTIYF TQEGDGPTTV TGSISGLKPG LHGFHVHALG51 DTTNGCMSTG PHFNPAGNEH GAPEDETRHA GDLGNVTVGE DGTVNVNIVD101 SQIPLSGSNS IIGRAVVVHA DPDDLGKGGH ELSKTTGNAG GRVACGIIGL151 QG

1 MVKAVAVLNS SEGVSGTYLF TQVGVAPTTV NGNISGLKPG LHGFHVHALG51 DTTNGCMSTG PHYNPAGKEH GAPEDEVRHA GDLGNITVGE DGTASFTITD101 KQIPLTGPQS IIGRAVVVHA DPDDLGKGGH ELSKSTGNAG GRIACGIIGL151 QG

1 MVKAVAVLTG SEGVQGTVFF AQEGEGPTTI TGSLSGLKPG LHGFHVHALG51 DTTNGCMSTG PHFNPAGKEH GAPEDGNRHA GDLGNVTVGE DGTVNFTVTD101 SQIPLTGLNS VVGRAVVVHA DSDDLGKGGH ELSKTTGNAG GRLACGVIGL151 QA

Fig. 2. Deduced amino acid sequences of (A) the CuZn-SOD isoforms and (B–D) the matches to the tryptic peptide sequences of the SOD isozymes from a crude proteinextract of S. tuberosa tubers. Matching amino acid sequences (bold) of the tryptic peptides for (A) B. taurus CuZn-SOD spots 1 and 2 (see Fig. 1D) with the sequence from thebovine (B. taurus) erythrocyte CuZn-SOD. For peptides from S. tuberosa, the putative SOD isoforms matched with the CuZn-SOD sequence from (B) Ananas comosus (Fig. 4D,spot 3), and from (C) Solanum lycopersicum and (D) Zantedeschia aethiopica (Fig. 4D, spot 5).

76 P. Niyomploy et al. / Journal of Pharmaceutical and Biomedical Analysis 90 (2014) 72– 77

Fig. 3. Active SOD isozymes resolved and detected in the crude protein extract from S. tuberosa. Crude protein extract (150 �g) after 2D-GE resolution with a broad rangepH (3–10) IEF strip in the first dimension and stained for (A) SOD activity or (B) with Coomassie blue for protein.

F n extra

tra

3r

idSEomEmttIdfi

ig. 4. (A and B) 1D-SDS-PAGE and (C and D) 2D-GE resolution of the crude proteictivity and (B and D) total proteins by Coomassie blue.

he LC–MSMS results reveal the capability of this method in sepa-ating two different SOD isozymes from each other (Fig. 4, spots 3nd 4 with a similar MW but different pI).

.3. The key feature for improving the developed 2D-GE methodesolution

The 2D-GE method developed for the detection and discrim-nation of active SOD isozymes used a non-denaturing IEF firstimension (IEF-focusing) and an apparently non-fully denaturingDS-PAGE second dimension. For the IEF-focusing step, Tris andDTA were used in the lysis solution and rehydration buffer insteadf urea, thiourea and DTT. Since the cell membrane structure isaintained by calcium and magnesium ions, their chelation by

DTA will destabilize the cell membrane, whilst the Tris bufferaintains a stable pH for SOD activity after cell lysis. In addition,

he salts obtained from the SOD extraction process can disturb

he electrophoresis process giving a poor protein focusing in theEF. Accordingly, the crude protein extract from S. tuberosa wasialyzed against distilled water, but this was still found to be insuf-cient (data not shown). Increasing the IEF time (Vh) moves the

act (40 and 150 �g, respectively) from S. tuberosa after staining for (A and C) SOD

salt ions to the end of the strip and was found to improve theresultant resolution (Supplementary Fig. S2). Thus, the appropri-ate buffer and prolonged IEF electrophoresis time are importantsteps to improve the resolution of the non-denaturing IEF-focusingstep.

Another step to increase the zymogram resolution was the gelwashing steps after the SDS-PAGE to remove the bound SDS beforestaining for SOD activity. The presence of SOD interferes with theSOD staining activity and produces an irregular purple backgroundon the gel. The reason for this observation remains unknown,although it is possible that SDS might inhibit SOD from binding toand reacting with superoxide radicals, thereby allowing the reduc-tion of NBT to a deep blue-colored Formazan dye, even when SODis present [15].

The removal of SDS by isopropanol, as previously reported[22,23], was used in the first wash, followed by two ZnCl2 solu-tion washes to remove the isopropanol and then soaking the gel

sequentially with 0.1 M and 0.01 M Tris–HCl (pH 7). This may alsolead to renaturation of (some of) the gel-bound SOD monomers tofunctional homodimers and so restore (some of the) SOD activity[24].

utical

4

tStouau

Stotpei

A

RtRpAth

A

t

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

P. Niyomploy et al. / Journal of Pharmace

. Conclusion

Here, the first developed 2D-GE method for SOD isozyme detec-ion was reported. The method was successfully applied to separateOD isozymes that differ in pI and/or MW from a complex pro-ein mixture, maintaining at least some of the different isoformsf SOD in the native form of the enzyme and/or allowing renat-ration of some non-permanently denatured monomers to formctive homodimers, even though the second dimension was runnder denaturing conditions.

So far, there has been no report on using 2D-GE coupled withOD staining activity. However, this developed method could proveo be useful to many SOD-related research areas, including the usef specific isozymes as markers of plant stress responses, the iden-ification of SOD from novel sources and the evaluation of theirI for determining appropriate resins and buffer for the large scalextraction and enrichment in order to obtain purified SOD enzymesn a timely manner.

cknowledgements

The authors thank the Thailand Research Fund through theoyal Golden Jubilee Ph.D. Program (Grant No. PHD/0013/2552),he 90th Anniversary of Chulalongkorn University, and the Nationalesearch University Project of CHE; and the Ratchadaphiseksom-hot Endowment Fund (Aging Society Cluster CU-56-AS05 anddvanced Material Cluster CU56-AM09) for financial support of

his research. We also gratefully thank Dr. Apaporn Boonmee forer kind advice and guidance.

ppendix A. Supplementary data

Supplementary data associated with this article can be found, inhe online version, at http://dx.doi.org/10.1016/j.jpba.2013.10.035.

eferences

[1] Y. Wang, K. Osatomi, Y. Nagatomo, A. Yoshida, K. Hara, Purification, molecularcloning, and some properties of a manganese-containing superoxide dismutasefrom Japanese flounder (Paralichthysolivaceus), Comp. Biochem. Physiol. B 158(2011) 289–296.

[2] M.P. Babitha, H.S. Prakash, H. ShekarShetty, Purification and partial characteri-zation of manganese superoxide dismutase from downy mildew resistant pearlmillet seedlings inoculated with Sclerospora graminicola, Plant Sci. 163 (2002)917–924.

[3] D. Vyas, S. Kumar, Purification and partial characterization of a low tempera-ture responsive Mn-SOD from tea (Camellia sinensis (L) O. Kuntze), Biochem.Biophys. Res. Commun. 329 (2005) 831–838.

[4] H.D. Youn, E.J. Kim, J.H. Roe, Y.C. Hah, S.O. Kang, A novel nickel-containingsuperoxide dismutase from Streptomyces spp., Biochem. J. 318 (1996)889–896.

[5] A.F. Miller, Superoxide dismutase: active sites that save, but a protein that kills,Curr. Opin. Chem. Biol. 8 (2004) 162–168.

[6] C.L. Fattman, J.J. Enghild, J.D. Crapo, L.M. Schaefer, Z. Valnickova, T.D. Oury,Purification and characterization of extracellular superoxide dismutase inmouse lung, Biochem. Biophys. Res. Commun. 275 (2000) 542–548.

[

and Biomedical Analysis 90 (2014) 72– 77 77

[7] N.I.A. Haddad, Q. Yuan, Purification and some properties of Cu, Zn superoxidedismutase from Radix lethospermi seed, kind of Chinese traditional medicine, J.Chromatogr. B 818 (2005) 123–131.

[8] R. Ozturk-Urek, L. Tarhan, Purification and characterization of superoxide dis-mutase from chicken liver, Comp. Biochem. Physiol. B 128 (2001) 205–212.

[9] N. He, Q. Li, D. Sun, X. Ling, Isolation, purification and characterization of super-oxide dismutase from garlic, Biochem. Eng. J. 38 (2008) 33–38.

10] C. Zhao, J. Wang, M. Cao, C. Zhao, J. Shao, T. Lei, J. Yin, G.G. Hill, N. Xu, S. Liu,Proteomic changes in rice leaves during development of field-grown rice plants,Proteomics 5 (2005) 961–972.

11] H.Y. Wu, Y.X. Duan, Defense response of soybean (Glycine max) to soybean cystnematode (Heterodera glycines) race 3 infection, J. Anim. Plant Sci. 21 (2001)165–170.

12] S. Kanazawa, S. Sano, T. Koshiba, T. Ushimaru, Changes in antioxidative enzymesin cucumber cotyledons during natural senescence: comparison with thoseduring dark-induced senescence, Physiol. Plant. 109 (2000) 211–216.

13] Z. Zhu, G. Wei, J. Li, Q. Qian, J. Yu, Silicon alleviates salt stress and increasesantioxidant enzymes activity in leaves of salt-stressed cucumber (Cucumissativus L.), Plant Sci. 167 (2004) 527–533.

14] C. Beauchamp, I. Fridovich, Superoxide dismutase: improved assays an assayapplicable to acrylamide gel, Anal. Biochem. 44 (1971) 276–287.

15] J.R. Chen, C.W. Liao, S.J.T. Mao, T.H. Chen, C.N. Weng, A simple technique for thesimultaneous determination of molecular weight and activity of superoxidedismutase using SDS-PAGE, J. Biochem. Biophys. Methods 47 (2001) 233–237.

16] L. Palomares, G. Prestamo, Detection of peroxidase activity in two-dimensionalgel electrophoresis, Eur. Food Res. Technol. 220 (2005) 644–647.

17] J.L. Lopez, Two-dimensional electrophoresis in proteome expression analysis,J. Chromatogr. B 849 (2007) 190–202.

18] A. Boonmee, C. Srisomsap, D. Chokchaichamnankit, A. Karnchanatat, P. Sang-vanich, A proteomic analysis of Curcuma comosa Roxb. rhizomes, Proteome Sci.9 (2001) 43.

19] A. Boonmee, C. Srisomsap, A. Karnchanatat, P. Sangvanich, An antioxidant pro-tein in Curcuma comosa Roxb. rhizomes, Food Chem. 124 (2011) 476–480.

20] U.K. Laemmli, Cleavage of structural proteins during the assembly of the headof bacteriophage T4, Nature 227 (1970) 680–685.

21] H. Fukamachi, Y. Nakano, S. Okano, Y. Shibata, Y. Abiko, Y. Yamashita,High production of methyl mercaptan by l-methionine-�-deamino-�-mercaptomethanelyase from Treponema denticola, Biochem. Biophys. Res.Commun. 331 (2005) 127–131.

22] W.C. Hou, Y.L. Lu, S.Y. Liu, Y.H. Lin, Activities of superoxide dismutase and glu-tathione peroxidase in leaves of different cultivars of Lirio pespicata L. on 10%SDS-PAGE gels, Bot. Bull. Acad. Sin. 44 (2003) 37–41.

23] The Green Lab, A detailed procedure for RNase activity staining on SDS-PAGE,http://ag.udel.edu/plsc/faculty/pamgreen/rnase.htm

24] J.L. Abernethy, H.M. Steinman, R.L. Hill, Bovine erythrocyte superoxide dismu-tase subunit structure and sequence location of the intrasubunit disulfide bond,J. Biol. Chem. 249 (1974) 7339–7347.

25] Y.H. Cheng, Z. Wang, S.Y. Xu, Antioxidant properties of wheat germ proteinhydrolysates evaluated in vitro, J. Cent. S. Univ. Technol. 13 (2006) 160–165.

26] Z. Xie, J. Huang, X. Xu, Z. Jin, Antioxidant activity of peptides isolated from alfalfaleaf protein hydrolysate, Food Chem. 111 (2008) 370–376.

27] C.F. Ajibola, J.B. Fashakin, T.N. Fagbemi, R.E. Aluko, Effect of peptide size onantioxidant properties of African Yam bean seed (Sphenostylis stenocarpa) pro-tein hydrolysate fractions, Int. J. Mol. Sci. 12 (2011) 6685–6702.

28] T.L. Pownall, C.C. Udenigwe, R.E. Aluko, Effect of cation property on the in vitroantioxidant activities of pea protein hydrolysate fractions, Food Res. Int. 44(2011) 1069–1074.

29] T.A. Siswoyo, E. Mardiana, K.O. Lee, K. Hoshokawa, Isolation and characteriza-tion of antioxidant protein fractions from Melinjo (Gnetum gnemon) seeds, J.Agric. Food Chem. 59 (2011) 5648–5656.

30] J. Carillon, D. Del Rio, P.L. Teissedre, J.P. Cristol, D. Lacan, Antioxidant capacityand angiotensin I converting enzyme inhibitory activity of a melon concentrate

rich in superoxide dismutase, Food Chem. 135 (2012) 1298–1302.

31] W. Moon-ai, P. Niyomploy, R. Boonsombat, P. Sangvanich, A. Karnchanatat, Asuperoxide dismutase purified from the rhizome of Curcuma aeruginosa Roxb.as inhibitor of nitric oxide production in the macrophage-like RAW 264.7 cellline, Appl. Biochem. Biotechnol. 166 (2012) 2138–2155.