Supporting Information for Proteomics DOI 10.1002/pmic.200500596

Li-Tai Jin, Sun-Young Hwang, Gyurng-Soo Yoo and Jung-Kap Choi

A mass spectrometry compatible silver staining method for protein incorporating a new silver sensitizer in sodium dodecyl sulfate-polyacrylamide electrophoresis gels

ª 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.proteomics-journal.com Supplementary Information

A mass spectrometry compatible silver staining method for protein

incorporating a new silver sensitizer in sodium dodecyl sulfate-

polyacrylamide electrophoresis gels

Li-Tai Jin, Sun-Young Hwang, Gyurng-Soo Yoo and Jung-Kap Choi

College of Pharmacy, Chonnam National University, Kwangju 500-757, South

Korea

Correspondence: Prof. Jung-Kap Choi, Lab. of Analytical Biochemistry,

College of Pharmacy, Chonnam National University, Yongbong-Dong 300, Buk-

Ku, Kwangju 500-757, South Korea (Phone: +8262-530-2930; Fax: +8262-530-

2911; E-mail: [email protected])

Abbreviations used: EBT, eriochrome black T

Keywords: Electrophoresis / Eriochrome black T / Mass spectrometry / Protein silver staining

1 1 Supplementary Introduction

In life sciences, sodium dodecyl sulfate-polyacrylamide gel electrophoresis

(SDS-PAGE) is a reliable and widely used analytical technique for the separation and identification of proteins. Given the development of proteomics, the ability to analyze and identify protein in a gel matrix at high sensitivity is becoming more important [1,2]. Various protein staining methods have been developed based on visible organic dyes, fluorescent and silver based systems, and radiolabeling. Of these, organic dyes are easily used but have low sensitivities, with detection limits in the range of 1-50 ng for proteins [2,3]. Fluorescent staining methods are a little more sensitive than dye-based methods but for data reading, special instruments are necessary that involve UV. These methods have detection limits that fall in the range of 0.1-10 ng for proteins [2,4]. Radiolabeling of proteins with radioactive isotopes is likely to remain the most sensitive method available. However, they are hazardous and require complicated handling procedures. With the possible exception of fluorescent staining and radiolabeling, silver staining offers greatest sensitivity for protein detection. Numerous silver staining protocols with different merits have been described [2,5,6,7,8]. However, relatively few are compatible with MS analysis. In the present study, we developed an eriochrome black T

(EBT)-silver staining method that can detect 0.05-0.2 ng protein within 60 min while glutaraldehyde-silver method can only detect 0.2-0.6 ng protein within 100 min. The EBT dye (Fig. 1), which acts as a silver sensitizer, easily binds to

2 proteins and silver ions, and contains a diazo group, which reduces silver ions by cleavage in alkaline solution. We found that these properties of EBT significantly improve traditional silver staining in terms of sensitivity, speed, and MS compatibility. And more, the dye is cheap and widely used in other fields. So we believe that the developed EBT-silver method will be found useful for routine proteomics research.

2 Supplementary Materials and methods

2.1 Materials

Acrylamide, Bis, TEMED, ammonium persulfate (APS), Tris base, SDS, eriochrome black T (EBT, dye content approx. 60%), iodoacetamide, glycerol, bromophenol blue, silver nitrate, sodium thiosulfate, potassium ferricyanide, formaldehyde, trypsin, potassium carbonate and molecular weight marker proteins

(SDS-6H) including myosin heavy chain, β-galactosidase, phosphorylase b, bovine serum albumin (BSA), ovalbumin (OVA), carbonic anhydrase (CA) were purchased from Sigma-Aldrich Chemical Co. (St. Louis, MO, USA). EDTA,

CHAPS, dithiothreitol (DTT), phenylmethylsulfonyl fluoride (PMSF), urea, IPG strip, cover oil and IPG buffer were from Amersham Biosciences (Uppsala,

Sweden). All other chemicals used were of analytical grade and were obtained

3 from various commercial sources.

2.2 Electrophoresis

2.2.1 Preparation for 1-DE

Escherichia coli BL 21 (E. coli) was chosen for one-dimensional electrophoresis

(1-DE) analysis. Total cell protein samples were obtained from E. coli using the following method. After cell culture, cells were harvested by centrifugation at

3,000 rpm for 10 min, sonicated in buffer containing 50 mM Tris, pH 7.5, 1 mM

EDTA, and 0.4 mM PMSF for 5×1 min (1 g cell/10 ml buffer) and then centrifuged at 15,000 rpm for 20 min at 4 . Protein concentrations in supernatant were ca. 16 mg/ml as determined by Bradford's method using a Bio-Rad protein assay kit. Molecular weight marker proteins (SDS-6H) and E. coli extract were dissolved in buffer containing 60 mM Tris, pH 6.8, 25% glycerol, 2% SDS, 1%

DTT, and 0.002% bromophenol blue for 1-DE. Prior to electrophoresis, protein samples were heated at 100 for 5 min in a boiling water bath and then cooled to room temperature. Twofold serial dilutions of marker proteins and E. coli samples were loaded onto gels at loadings of 100 ng to 0.2 ng/band and 4,000 ng to 8 ng/band, respectively.

4 Electrophoresis was carried out using polyacrylamide slab gels (60×80×0.75 mm) and a discontinuous buffer system [9]. 4.5% of stacking gel was overlaid on a

11.5% polyacrylamide separating gel with an acrylamide:Bis ratio of 30:0.8. The running buffer consisted of 0.025 M Tris, 0.2 M glycine, and 0.1% SDS. Gels were run in a Mini-protein III dual slab cell (Bio-Rad, Hercules, CA, USA) at a constant current of 22 mA per slab gel using a POWER PAC 300 supply (Bio-

Rad).

2.2.2 Preparation for 2-DE

The total cell protein samples of E. coli were separated using IPG gel strips

(linear 4-7 pH gradient, 13 cm) for 2-DE analysis. Each strip loaded with 48 µg of sample was rehydrated in an Immobiline DryStrip Reswelling tray (Amersham

Pharmacia Biotech, USA). Electrophoresis in the first dimension (isoelectric focusing) was performed using a Multiphor Electrophoresis unit (Amersham

Pharmacia Biotech, USA) according to the manufacturer’s instructions. IPG strips containing the samples were transferred to the focusing tray and covered with mineral oil. Isoelectric focusing was performed using an EPS 3501 XL power supply (Amersham Pharmacia Biotech, USA) with a 5-step program (0 to 150 V for 1 h with gradient, 150 V to 300 V for 3 h, 300 V to 1,500 V for 1.5 h, 1,500 V to 3,500 V for 1.5 h, and hold at 3,500 V for 5 h). After isoelectric focusing, strips were removed and excess mineral oil was allowed drain off. Subsequently, strips

5 were incubated in 16 ml of first equilibration buffer (1% DTT, 50 mM Tris-HCl, pH 8.8, 6 M urea, 30% glycerol, 2% SDS, 0.002% bromophenol blue) for 15 min.

For the second equilibration, the strips were incubated in 16 ml of second equilibration buffer (4% iodoacetamide, 50 mM Tris-HCl, pH 8.8, 6 M urea, 30% glycerol, 2% SDS, 0.002% bromophenol blue) for 15 min. The strips were placed in individual tubes with the support film toward the tube wall and the tubes were shaken on an adjustable tilt rocker (Labnet International Inc., Woodbridge, NJ,

USA) during equilibration. The strips were transferred for electrophoresis in the second dimension. A separating gel of 11.5% polyacrylamide with an acrylamide:Bis ratio of 30:0.8 was cast using Hoefer SE 600 electrophoresis units

(Amersham Pharmacia Biotech, USA). Gels were run at a constant current of 25 mA per slab in running buffer consisting of 0.025 M Tris, 0.2 M glycine, and

0.1% SDS.

2.3 Protein staining methods

For all stainings, working solutions were prepared fresh with deionized water and clean glassware or plastic ware. All steps were carried out at room temperature with shaking.

2.3.1 EBT-silver staining

6

The EBT-silver staining procedure is summarized in Table 1. After electrophoresis, gels (0.75 or 1 mm thickness, 8×10 cm) were fixed in 200 ml of fixing solution (40 % v/v ethanol, 10% v/v acetic acid solution) for 2×20 min (or more to overnight). Gels were sensitized (stained) with 100 ml of staining solution containing 0.006% w/v EBT, 30% v/v ethanol solution by shaking for 2 min, and then destained using 200 ml of 30% ethanol solution for 2 min, and washed in 200 ml of deionized water for 2×2 min. Gels were then impregnated in 100 ml of

0.25% w/v silver nitrate containing 0.037% w/v formaldehyde for 5 min, washed in 200 ml of deionized water for 2×20 sec, and immersed in 100 ml of 2% w/v potassium carbonate, 0.04% w/v sodium hydroxide, 0.007% w/v formaldehyde, and 0.002% w/v sodium thiosulfate solution to develop image. After silver ion reduction, gels were immersed in 100 ml of 1.5% w/v EDTA for 10 min (or more to overnight) to stop the development. Stained gels were stored in deionized water or dried on filter paper under vacuum at 65 for 40 min. The EBT stock solution can be prepared 0.5% w/v with deionized water and stored for 1 year in dark.

The volumes of the working solutions used in each step were proportional to the gel area, e.g., 400 ml of fixing solution for a 15×15 cm regular gel.

2.3.2 Glutaraldehyde-silver staining

7 Silver nitrate staining using glutaraldehyde as a sensitizer was performed using a modification of the method described by Heukeshoven and Dernick [8]. Briefly, after electrophoresis, gels (0.75 or 1 mm thickness, 8×10 cm) were fixed in 125 ml of 40% v/v ethanol, 10% v/v acetic acid solution for 30 min, and then reacted in 125 ml of 6.8% sodium acetate, 0.125% glutaraldehyde and 0.2% sodium thiosulfate solution for 30 min, and finally washed in 125 ml deionized water for

3×5 min. Gels were then impregnated in 125 ml of 0.015% formaldehyde, 0.25% silver nitrate solution for 20 min, washed in 125 ml deionized water for 2×1 min, and immersed in 125 ml of 3% sodium carbonate, 0.007% formaldehyde solution.

After silver ion reduction, gels were immersed in 125 ml of 1.5% EDTA for 10 min to stop development. Stained gels were stored in deionized water or dried on a filter paper under vacuum at 65 for 40 min. The volumes of the working solutions used in each step were proportional to the gel area.

2.4 MALDI-TOF MS analysis

For MALDI-TOF MS analysis, five maker proteins (β-galactosidase, phosphorylase b, BSA, OVA and CA) were separated by 1-DE using a 0.75 mm gel. The electrophoresed gels were stained using the EBT-silver method, and 3, 6,

12, 25, 50, and 100 ng/band protein spots were excised from stained gels. Protein digestion was performed using a protocol modified by EMBL and Russell et al.

8 (Russell et al., 2001). Briefly, gel pieces were soaked in 0.1 ml of destaining solution (stock solution of 100 mM sodium thiosulfate and 30 mM potassium ferricyanide were mixed in a 1:1 ratio immediately prior to use) for 10 min to convert the silver metal to a water-soluble form. After removing the destaining solution, the gel pieces were each washed for 3×15 min in 1 ml of deionized water and then incubated in 0.04 ml 50% acetonitrile for 15 min, and 0.04 ml 100% acetonitrile for 10 min. The opaque gel pieces were then equilibrated with 0.04 ml

100 mM ammonium bicarbonate (pH 8.0) for 5 min and an equal volume of 100% acetonitrile was added for 15 min. After removing the solution, gel pieces were dried in a SpeedVac for 15 min, and the dried gel pieces were reduced with 0.04 ml of 10 mM DTT containing 100 mM ammonium bicarbonate at 56 for 45 min and alkylated with 0.04 ml of 55 mM iodoacetamide containing 100 mM ammonium bicarbonate for 30 min in the dark. After removing the solution, the pieces were washed with 0.04 ml 100 mM ammonium bicarbonate for 5 min and an equal volume of 100 % acetonitrile was added for 10 min. The pieces were again dried in a SpeedVac for 15 min and digested with 0.04 ml 0.01% trypsin solution at 4 for 45 min. To remove salt or contaminant from the peptide mixture, they were purified using a pipette tip (ZipTipC18; Millipore Corporation

Bedford, USA). MS analysis was performed on a PerSeptive Biosystems MALDI-

TOF Voyager DE-STR MS in the delayed extraction/ACTH reflector mode

(PerSeptive Biosystems, Framingham, MA, USA). The ACTH reflector mode involved the following; 20 kV accelerating voltage, 65 % grid voltage, 1.12 mirror voltage ratio, and 150 nsec extraction delay time. The detailed MS

9 parameters were shown in Table 2. Masses were internally calibrated using standard peptides: angiotensin ([M+H]+=1296.6853), ACTH 1-17

([M+H]+=2093.0867), ACTH 18-39 ([M+H]+=2465.1989). Proteins were identified by database searching with MS-Fit (http://prospector.ucsf.edu/) by comparing peptide masses with those in the NCBInr protein database.

2.5 Image analysis

After gel drying, the quantitation of the protein bands was performed with a scanner (SIS 3800, Samsung Co., South Korea) interfaced to a Samsung computer

(DV 25, Samsung Co., South Korea). Linear dynamic range analysis was performed using a band analysis software program (TINA 2.09, Raytest Co.,

Straubenhardt, Germany); for counting the number of spots, it was performed using a 2-D spot analysis software program (UVIspot 10.01, UVItyec Limited,

UK).

3 Supplementary Results

3.1 Comparison of EBT-silver and glutaraldehyde-silver methods in terms

of 1-DE and 2-DE results

10

In order to compare the sensitivities of the EBT-silver and glutaraldehyde-silver methods, marker proteins (SDS-6H) were electrophoresed and stained using both methods (Fig. 2). For the detection of serial dilutions of marker proteins from 100 to 0.2 ng (twofold dilution), the detection limit of the EBT-silver method was ca. 0.05-0.2 ng, 2-4 folds higher than that of the glutaraldehyde- silver method. In order to confirm sensitivities and to compare protein staining patterns, total E. coli cell protein samples were separated by 1-DE or 2-DE, and stained using both methods. 1-DE stained patterns are shown in Figure 3. In addition to its higher sensitivity, EBT-silver was also found to have a better resolution than glutaraldehyde-silver. Some bands were unresolved by the glutaraldehyde-silver but resolved by EBT-silver. The 2-DE stained patterns are shown in Figure 4. The 2-DE protein map of the EBT-silver method produced more spots than the glutaraldehyde-silver method with better contrast. The numbers of detected spots using EBT-silver and glutaraldehyde-silver stainings were 577 and 482, respectively. The above results demonstrate that the EBT- silver method has better sensitivity and resolution than the glutaraldehyde-silver method.

3.2 Compatibility with MALDI-TOF MS

11 In order to determine the compatibility of the EBT-silver method with MALDI-

TOF MS, five maker proteins from 3 to 100 ng were used. MALDI-TOF MS spectra were acquired from each of the protein bands, and evaluated: β- galactosidase (E. coli, 116.5 KDa / pI 5.3), phosphorylase (Rabbit, 97.3 KDa / pI

6.8), BSA (Bovine, 69.3 KDa / pI 5.8), OVA (Chick, 42.8 KDa / pI 5.2) and CA

(Bovine, 28.9 KDa / pI 6.4). A summarized MALDI-TOF MS data of these proteins including the number and percent of mass matched peptides, protein sequence coverage, and total mass tolerance (ppm) were provided in Table 3.

Generally, the valid identification of protein requires the sequence coverage over

25%. Therefore, the detection limits of the EBT-silver method for MALDI-TOF

MS were found to be β-galactosidase, 6 ng; phosphorylase, 6 ng; BSA, 25 ng;

OVA, 6 ng, and CA, 3 ng.

3.3 Image analysis

In terms of linear dynamic range of the EBT-silver staining, stained density and band areas were determined by using the scanned data from standard protein

(Fig. 5). The linear dynamic ranges of the amount of proteins were myosin (0.4-

50 ng, correlation coefficient, 0.942), phosphorylase b (0.4-50 ng, 0.972), β- galactosidase (0.4-50ng, 0.958), ovalbumin (0.4-50 ng, 0.924) and carbonic anhydrase (0.4-50 ng, 0.939), respectively.

12

4 Supplementary Discussion

The mechanism of silver staining of proteins is not clearly understood [7]. But one of the most efficient ways of suppressing gel background staining and of increasing sensitivity is to introduce an enhancing step, by treating gel with special reagents before and during silver impregnation. The step represents a

‘seeding’ process to form initial nucleation centers, the so-called pre-nucleation process in impregantion. The creation of such centers is followed by the rapid buildup of silver deposits from free silver ions during the developing step.

Therefore, reductants, sulfiding agents were widely used as silver ion sensitizer in the enhancing step. Reductants produce minute silver metals particles that act as nucleation centers during impregnation step, and which can accelerate and optimize the build-up of silver deposits at protein zones during further silver ion reductions; sulfiding agents produce minute silver sulfide deposits, and these deposits of sulfide in protein zones also act as silver nucleation centers.

Sometimes formaldehyde or glutaraldehyde is added to silver nitrate solution directly to facilitate the formation of nucleation centers in silver impregnation.

In the present study, a new silver sensitizer for protein silver staining, namely

EBT, was developed. Its possible staining mechanism is as follows. The EBT has several functional groups, e.g., sulfonate, hydroxyl, nitrogen dioxide, and diazo, and thus it is classified as an anionic acidic dye. Initially, the EBT binds to

13 proteins via electrostatic interaction between its sulfonate group and proteinaceous amine cation at acidic pH’s. Hydrophobic interaction, Van der

Waals forces, and hydrogen bonding also contribute to binding between dye and protein, as described for CBB-based dyes [10,11]. Thus, one reason for the compatibility of the EBT-silver method with MS may be due to the lack of direct covalent bonding between EBT and protein. Second, besides silver ions that directly bind to protein, they also bind to EBT to form silver-dye complexes, like π-complexes (Figure 6, A) or silver-dye coordinated complexes in silver impregnation [12]. The formation of these complexes may improve the selectivity of silver at protein zones. In addition, the diazo bond of EBT has the ability to reduce silver ions in alkaline solution [12], and it has been suggested that this ability has a beneficial effect on silver nucleation and silver reduction.

At a high pH, the diazo bond is under strain, due to the repulsion between the negatively charged groups on the two naphthalenic moieties. Therefore, the dye molecule becomes more susceptible to nucleophilic attack. This would facilitate breakage of the diazo bond with the release of nitrogen. Moreover, silver ions can further be reduced by dihydroxynaphthalenes produced from EBT cleavage; the possible redox reactions of EBT and silver are shown in Figure 6, [12].

Finally, EBT wholly loses its color by diazo bond cleavage after silver reduction. We believe that these properties of EBT increase the sensitivity, speed, and MS compatibility of the silver staining.

Glutaraldehyde is widely used in silver staining protocols because it is an excellent sensitizing agent. These protocols are based on cross-linking between

14 glutaraldehyde and proteinaceous lysine by Schiff’s base formation. However, proteins so treated are of limited use for subsequent MS analysis because of the presence of these covalent modifications [13]. In the present study, therefore the glutaraldehyde-based staining method was compared with EBT-silver staining method only in sensitivity.

In sensitive silver staining, gel buffer should be degassed before polymerization, otherwise point streaking will be caused.

As to sample preparation for silver staining, many protocols are documented in the literature. These articles comprehensively discuss the uses of 2- mecaptoethanol, DTT, DTE, and TBP [14,15,16,17]. Reductants present during sample preparation reduce disulfide bridges and prevent protein aggregation in sample solutions. However, high concentrations of reductants can cause point streaking by the charge of reductant in silver staining. About 5% of 2- mercaptoethanol is used in dye-based staining protocols, but for silver staining this concentration is unnecessary, because silver staining methods are 100-1,000 folds more sensitive than dye-based methods. Therefore, in general, about 2% of

2-mercaptoethanol or 1% of DTT is sufficient for silver staining.

In proteomics research, protein detecting methods should be compatible with MS analysis. However silver staining methods inherently modify proteins to different levels. Thus, substantial protein modifications induced by protein staining make them unsuitable for MS analysis. i.g., 1) Methylation (or ethylation) takes place preferentially at glutamic acid residues due to either TCA or methanol (or ethanol)

15 in staining solutions. 2) Cystein residue alkylation often occurs after iodoacetamide treatment in the 2-D equilibration step. And, 3) Lysine can form protein cross-links in the presence of glutaraldehyde or formaldehyde [15,18]. In general, MS compatible silver staining protocols can identify ca. 50 ng of protein

[19,20]. But we found that the EBT-silver staining offers detection limits as low as

3-25 ng method for MS analysis, possibly because of the gentle treatment and short staining times involved.

5 References

16 [1] Hames, B. D., and Rickweed, D., Gel electrophoresis of proteins, A practical

approach, Oxford University Press, Oxford 1994.

[2] Patton, W. F., J. Chromatography B 2002, 771, 3-31.

[3] Candiano, G., Bruschi, M., Musante, L., Santucci, L., et al., Electrophoresis

2004, 25, 1327-1333.

[4] Kang, C., Kim, H.J., Kang, D., Jung, D.Y., Suh, M., Electrophoresis 2003, 24,

3297-3304.

[5] Kondo, H., Ikeda, K., Miyazaki, N., J. Neuroscience Methods 1996, 68(2),

275-280.

[6] Nesterenko, M. V., Tilley, M., Upton, S. J., J. Biochem. Biophys. Mehods 1994,

28, 239-242.

[7] Syrovy, I., Hodny, Z., J. Chromatogr. 1991, 569, 175-196.

[8] Heukeshoven, J., Dernick, R., Electrophoresis 1985, 6, 103-112.

[9] Laemmli, U. K., Nature 1970, 227, 680-685.

[10] Jin, L. T., Hwang, S. Y., Yoo, G. S., Choi, J. K., Electrophoresis 2004, 25,

2494-2500.

[11] Jung, D. W., Yoo, G. S., Choi, J. K., Electrophoresis 1998, 19, 2412-2415.

[12] Zhai, X., Efrima, S., J. Phys. Chem. 1996, 100, 1779-1785.

[13] Swain, M., Ross, N. W., Electrophoresis 1995, 16, 948-951.

17 [14] Gorg, A., Obermaler, C., Boguth, G., Harder A., et al., Electrophoresis, 2000,

21, 1037-1053.

[15] Shaw, M. M., Riederer, B. M., Proteomics 2003, 3, 1408-1417.

[16] Jacobs, D. I., Heijden, R., Verpoorte, R., Phytochem. Anal. 2000, 11, 277-287.

[17] Herbert, B., Galvani, M., Hamdan, M., Olivieri, E., et al., Electrophoresis

2001, 22, 2046-2057.

[18] Haebel, S., Albrecht, T., Sparbier, K., Walden, P., et al., Electrophoresis 1998,

19, 679-686.

[19] Richert, S., Luche, S., Chevallet, M., Dorsselaer, A. V., et al., Proteomics 2004,

4, 909-916.

[20] Mortz, E., Krogh, T. N., Vorum, H., Gorg, A., Proteomics 2001, 1, 1359-1363.

18

Supplementary Figure 1. The structure of eriochrome black T (EBT) dye.

19

Supplementary Figure 2. Comparison of the sensitivities of the EBT–silver and the glutaraldehyde-silver methods in 1-DE mini-gels using marker proteins.

The staining procedures (A) EBT-silver and (B) glutaraldehyde-silver were performed as described in Materials and methods. The twofold serial dilutions of marker proteins (SDS-6H) loaded onto gels (from left to right) were; lane (1)

100; (2) 50; (3) 25; (4)12.5; (5) 6.3; (6) 3.2; (7) 1.6; (8) 0.8; (9) 0.4 and (10) 0.2 ng per band, respectively.

20

Supplementary Figure 3. Comparison of the sensitivities of the EBT-silver and the glutaraldehyde-silver methods in 1-DE mini-gels using total cell proteins of E. coli.

The staining procedures (A) EBT-silver and (B) glutaraldehyde-silver were performed as described in Materials and methods. The twofold serial dilutions of E. coli total cell protein samples were loaded onto gels (from left to right) were; lane (1) 4,000; (2) 2,000; (3) 1,000; (4) 500; (5) 250; (6) 125; (7) 62.5; (8)

31.3; (9) 16 and (10) 8 ng, respectively.

21

Supplementary Figure 4. Comparison of the sensitivities of the EBT-silver and the glutaraldehyde-silver methods in 2-DE regular gels (14 × 16 cm) using the total cell proteins of E. coli.

The staining procedures (A) EBT-silver and (B) glutaraldehyde-silver were performed as described in Materials and methods. 48 µg of E. coli total cell protein sample per gel was loaded into 13 cm, pH 4-7 linear IPG strips.

22

Supplementary Figure 5. Linear dynamic ranges of the EBT-silver staining with standard marker proteins.

The EBT-silver stain and image analysis were performed as described in

Materials and methods. Five protein molecular mass standards, myosin, β- galactosidase, phosphorylase b, ovalbumin (OVA) and carbonic anhydrase (CA) were separated in 4.5% and 10 % discontinual polyacrylamide gel. The image analysis was performed using a TINA 2.09 software program.

23

Supplementary Figure 6. (A) An outline of the redox reaction between silver ions and EBT. (B) Reduction of silver ions by dihydroxynaphthalene.

24 Supplementary Table 1. EBT-silver protein staining protocol in SDS-PAGE gels (0.75 or 1.0 mm, 8×10 cm).

Steps Reagents Volume (ml) Time (min)

20 × 2 (or to Fix 40% ethanol, 10% acetic acid 200 overnight)

Sensitize 0.006% EBT, 30% ethanol 100 2

Destain 30% ethanol 200 2

Rinse Deionized water 200 2× 2

Silver 0.25% silver nitrate, 0.03% formaldehyde 100 5

Rinse Deionized water 200 20 sec× 2

2% potassium carbonate, 0.04% sodium Develop hydroxide, 0.007% formaldehyde, 100 2-8 0.002% sodium thiosulfate

10 (or to Stop 1.5% EDTA 100 overnight)

Comments: a) A fixing time of 2 hours is recommended for maximal sensitivity. b) The EBT stock solution can be stored for 1 year and should be shaken before use. c) The working solutions should be mixed immediately before use. d) The volumes of the working solutions used in each step should be proportional to the gel area, e.g., 400 mL of fixing solution for a 15×15 cm regular gel. e) Gel buffer should be degassed before polymerization.

25 Supplementary Table 2. Parameters of MALDI-TOF MS in determination.

Mode of operation: Reflector Extraction mode: Delayed Polarity: Positive Acquisition control: Manual Accelerating voltage: 20000 V Grid voltage: 65% Mirror voltage ratio: 1.12 Guide wire 0: 0% Extraction delay time: 135 nsec Acquisition mass range: 800 -- 3500 Da Number of laser shots: 100/spectrum Laser intensity: 1700 Laser Rep Rate: 20.0 Hz Calibration type: Default Calibration matrix: a-Cyano-4-hydroxycinnamic Low mass gate: 800 Da acid Timed ion selector: Off Digitizer start time: 41.5095 Bin size: 0.5 nsec Number of data points: 90383 Vertical scale 0: 500 mV Vertical offset: 0.75% Input bandwidth 0: 500 MHz Sample well: 13 Plate ID: PLATE1 Serial number: 4239 Instrument name: Voyager-DE STR Plate type filename: C:\VOYAGER\100 well plate.plt Lab name: PE Biosystems Absolute x-position: 13141.3 Absolute y-position: 44383.3 Relative x-position: 1393.82 Relative y-posititon: 2155.82 Shots in spectrum: 100 Source pressure: 1.063e-007 Mirror pressure: 2.388e-008 TC2 pressure: 0.03566 TIS gate width: 15 TIS flight length: 1183

Supplementary Table 3. MALDI-TOF-MS data of five proteins obtained

26 from EBT-silver stained gels.

ng/band 100 50 25 12 6 3

Protein species fmol/band 1027 514 257 123 61.6 30.8

Beta-galactosidase Mass matched 34/82 30/75 30/78 24/76 21/93 10/99

(41%) (40%) (38%) (31%) (23%) (10%)

Sequence coverage (%) 43 44 45 35 32 15 Total ppm 24.8 24.8 29.1 36 30.4 35.5

Phosphorylase b Mass matched 37/81 33/85 27/86 12/29 16/92 -

(45%) (38%) (31%) (12%) (17%) -

Sequence coverage (%) 54 50 44 23 30 - Total ppm 24.9 24.4 28.9 41.6 26.5 -

Bovine albumin Mass matched 21/78 15/101 20/77 14/114 10/97 8/84

(BSA) (26%) (14%) (25%) (12%) (10%) (9%)

Sequence coverage (%) 40 28 32 22 14 18 Total ppm 21.0 22.4 27.9 22.0 30.5 37.7

Ovalbumin Mass matched 14/66 14/68 12/69 12/75 10/83 6/86

(OVA) (21%) (20%) (17%) (16%) (12%) (6%)

Sequence coverage (%) 51 51 45 44 37 28 Total ppm 26.3 10.4 6.7 22 16.9 18.1

Carbonic Mass matched 13/67 11/55 12/61 10/68 11/69 6/65

anhydrase (CA) (19%) (20%) (19%) (14%) (15%) (9%)

Sequence coverage (%) 60 60 66 60 61 41 Total ppm 28.5 15.6 22.3 21.8 23.4 35.4

27