852
R. Iida et al.
Reiko Iida Toshihiro Yasuda Daita Nadano Koichiro Kishi Department of Legal Medicine, Fukui Medical School, Fukui
Electrophoresis 1990,II, 852-855
Intensification of peroxidase-diaminobenzidine staining using gold-sulfide-silver:A rapid and highly sensitive method for visualization in immunoblotting A highly sensitive and rapid visualization method for protein detection by immunoblotting is described. Proteins blotted onto a Durapore membrane were visualized by the following procedure: after conventional peroxidase-based staining with 3,3'-diaminobenzidine (DAB), the produced DAB precipitates were intensified by treating with (i) gold trichloride (acid), (ii) sodium sulfide, and (iii) a developer containing silver nitrate. This postintensification method was employed for the detection of the genetic polymorphism of human proteins, such as deoxyribonuclease I in urine, and group specific component, transferrin and a,-antitrypsin in serum after polyacrylamide gel-isoelectric focusing, followed by immunoblotting. This postintensification technique was found to be simple, giving up to 16- to 64-fold amplification of the conventional peroxidase-DAB staining.
1 Introduction
2.2 Sample treatment
Since the first application of the blotting technique for specific detection of proteins in 1979 [ 11, many visualization systems have been developed including enzymatic, fluorescence, autoradiographic, and colloidal metal/silver development methods. The enzyme-based systems are preferred, since they are fast, easy, and safe, and therefore useful for routine work in the fields of biochemistry, forensic science and anthropology. Several workers have reported methods for postintensification of peroxidase-based staining with 3,3 '-diaminobenzidine (DAB) as immunocytochemical techniques 12-51. Especially the intensification by a gold-sulfide-silver (GOSS) method [41 has been found to be highly sensitive in histochemical use [61. We have employed the GOSS-intensification method for immunoblotting detection of some genetic markers in serum and urine by electrophoretic analysis, and succeeded in devising a rapid, sensitive and simplified version of this method. In this report, we describe the application of this new method for phenotyping analysis of deoxyribonuclease I (DNase I) in urine and group specific component (GC), transferrin (TF) and a,-antitrypsin (PI) in serum using polyacrylamide gelisoelectric focusing (PAG-IEF).
For T F and DNase I phenotyping, plasma or urine samples were treated with sialidase from Clostridiurn perfringens according to our previous reports [8,91. For PI phenotyping, 10 pL plasma was mixed with 10 pL 20 mM dithiothreitol and left for 30 min at room temperature, and then 10 pL 15 mM iodoacetamide was added and left for a further 30 min prior to focusing.
2 Materials and methods 2.1 Sample preparation
2.3 Antibodies and chemicals Rabbit antibodies against human GC, T F and PI were purchased from Dakopatts (Glostrup, Denmark); peroxidase-conjugated goat anti-rabbit immunoglobulin was from Bio-Rad (Richmond, CA). An antibody specific for human DNase I was produced in a rabbit as described previously [ 101. Carrier ampholytes Pharmalyte 2.5-5, 4.2-4.9 and 4.5-5.4 were purchased from Pharmacia (Uppsala, Sweden), Ampholine 4-6.5 from LKB (Bromma, Sweden), Durapore membrane from Millipore (Bedford, MA), gold trichloride (acid) from Nakarai (Kyoto, Japan), N-(2-acetamido)-2aminoethanesulfonic acid (ACES) from Sigma (St. Louis, MO), and 3,3'-diaminobenzidine.4HCI (DAB) from Dojin (Kumamoto, Japan). Other chemicals wereofreagent grade. 2.4 PAG-IEF
Plasma samples were collected from heparinized blood and stored at -80 OC until use. Urine samples were concentrated, dialyzed, and lyophilized according to our previous method 171.
PAG-IEF was performed mainly according to our previous papers [8, 10-121. Gels measuring 0.5 x 90 x 120 mm were prepared using the following components: 1.4 mL acrylamide-N,N'-methylenebisacrylamide(19.4 % w/v, 0.6 % w/v), 2.3 mL sucrose-glycerol (20 % w/v, 10 % vlv), 280 pL carrier ampholytes (Pharmalyte 4.5-5.4 for GC typing, Ampholine 4-6.5 for TF, Pharmalyte 4.2-4.9 for PI, and Pharmalyte 2.5-5 for DNase I), 1 mL distilled water, 10 pL Correspondence: K. Kishi, M. D., Department of Legal Medicine, Fukui N,N,N',N'-tetramethylethylenediamine,and 40 pL 1.2 % w/v Medical School, Matsuoka-cho, Fukui 910-1 1, Japan ammonium persulfate. In the case of PI phenotyping, 1 mL 6 % w/v ACES was added instead ofdistilled water. Plasma or Abbreviations: AB, avidin-biotin; ACES, N-(2-acetamido)-2-aminolyophilized urine samples were diluted with 0.3 % w/v bovine ethanesulfonic acid; DAB, 3,3'-diaminobenzidine; DNase I, deoxyribonuclease I; GC, group specific component; GOSS, gold-sulfide-silver ; serum albumin in order to protect them from pattern distorPAG-IEF, polyacrylamide gel-isoelectric focusing; PAP, peroxidase-anti- tion caused by a lowering of protein concentration. Electrophoresis was performed at a constant 2.5 W for 30 min, and peroxidase; PBS, phosphate-buffered saline; PBS-T, PBS containing then continued at 1000 V for 4 h under cooling at 10 "C. 0.05 96 Tween 20; PI, a,-antitrypsin; TF, transferrin 0VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1990
0173-0835/90/0909-08.52%3.50+.25/0
Electrophoresis 1990,11,852-855
2.5 Immunoblotting Transfer of proteins onto Durapore membrane strips (4 x 8 cm) was carried out by a capillary blotting method according to our previous papers [13,141. After transfer, the strips were briefly washed with 20 mM phosphate-buffered saline, pH 6.7, containing 0.05 % Tween 20 (PBS-T), and immersed in PBST containing 0.1 % w/v gelatin for 30 min. They were then washed with three changes of PBS-T and incubated for 1 h with an appropriate primary rabbit antibody diluted 1:400 in PBS-T. After washing three times in PBS-T, the strips wereincubated for 1 h with a peroxidase-conjugated secondary antibody (goat anti-rabbit immunoglobulin) diluted 1:400 in PBS-T. After washing three times with PBS-T, they were incubated for 2 min in PBS containing 0.01 % wlv DAB and 0.005 % v/v H,O, until the background was stained faintly pink, followed by washing with distilled water to stop the reaction. They were then kept in distilled water until further intensification.
2.6 Postintensificationprocedure (GOSS method) Following peroxidase-DAB staining, the protein patterns on the strips were detected using the improved immunoblotting version of the GOSS method for histochemical staining [41, as outlined in Table 1. The conditions shown are optimal, and the concentrations of the reagents and other particulars are indicated in the footnotes to Table 1.Durapore membranes were manipulated with clean Teflon forceps.
3 Results and discussion The most commonly nonradioactive protein detection methods employ intense immunoblotting stains, such as peroxidase-antiperoxidase (PAP) [ 151, avidin-biotin (AB) 161, and immunogold/silver [ 171. These stains lack the appropriate sensitivity to detect proteins present at low or trace concentrations (unpublished observation), although we have applied them to genetic studies of urinary proteins [7-10, 181. Body fluids, such as cerebrospinal and amniotic fluids, are often difficult to obtain in quantity, and frequently contain many proteins. This sensitivity problem was overcome by the application of a histologically derived GOSS postintensification procedure [41 for proteins on immunoblotting transfer membranes. Although the principle of the intensification is unknown, it
Rapid and highly sensitive visualization in irnrnunoblotting
853
Table 1. Postintensification procedure for peroxidase-DAB staining by the GOSS method") 1 Incubate the strips (4 x 8 cm) for 5 min in 0.1 % w/v gold trichloride (acid)b) and wash in three changes of distilled water (20 mL). 2 Incubate for 5 min in neutralized 0.3 % w/v sodium sulfide)and wash in three changes of distilled water (20 mL).
3 Incubate in an ice-cold silver developeld) until the desired intensification of the protein bands is obtained (about 10-20 min). 4 Rinsequickly intwochangesof 1 %v/vaceticacid tostopthereaction.
5 Incubate in a destaining solutione)until the background becomes clear (- 1-5 min). 6 Wash in a large volume of distilled water and then dry.
a) Unless otherwise indicated, all steps were performed with gentle shaking at room temperature (18-20 "C) in a glass tray (2 x 6 x 10 cm) with a reagent volume of I0 mL. b) Hydrogen tetrachloroaurate (1-), tetrabydrate (HAuC1,.4H2O) is of the purest grade and should be kept tightly closed and protected from light. c) Adjust the pH to 7.0 with 3 N HC1. d) Stock solution A: dissolve 2.5 g of sodium carbonate, anhydrous in 300 mL of distilled water. Stock solution B: dissolve 0.1 g of ammonium nitrate, 0.1 g of silver nitrate, 5 g of tungstosilicic acid and 0.25 mL of 37 %formaldehyde, inthatorder,in300mLofdistilledwater.Thestock solutions are kept at 4 "C. The silver developer is prepared fresh just before use by slow addition of 5 mLof stock solution A to 5 mL of stock solution B with stirring, and then chilledon ice. The chemicals used here are of the purest grade commercia!ly~available. e) Dissolve 2.5 g of sodium thiosulfate (pentahydrate) and 0.15 g sodium sulfite (anhydrous) in 40 mL of 1 % v/v acetic acid. Although not essential, it is often convenient to obtain avery clear background. It is possible to substitute 5 % w/v Fujifix, an acid hardening fixerreagent(Fuji Photo Film, Tokyo, Japan), for the solution.
may be based on the high afinity of DAB for metal salts and catalyzing acitivity ofmetal sulfides for silverdeposition under reducing conditions 141. The postintensification procedures described in this study are subject to the following three major variables: the temperature of the reagents, reagent concentration, and time of application of each reagent. In order to achieve a low background and slow development, the incubation temperature of the developer (step 3 inTable 1) should be kept as low as possible. The final intensity of the stain is also under the investigator's control by deciding when to stop the reduction reaction (steps 4 and 5 in Table 1).As theconcentration of stock solutions A and B for developer were diluted to one sixth of the original [41, no precipitation appeared when they were mixed without vigorous stirring. Moreover, the time required for the intensification was shortened to about 30 min.
Figure 1. (a) Pattern and (b) schematic representation of PAG-IEF of DNase I visualized by immunoblotting using peroxidase-DAB staining with intensification by the GOSS method. Anode is at the top. (a) Three common phenotypes of DNase I detected in 0.05 % w/v lyophilized urine samples (5 yL), corresponding to an approximately 5-fold concentration of the original urine with Pharmalyte pH 2.5-5. Lanes (1) and (4) phenotype 1; (2) and (5) phenotype 1-2; (3) and ( 6 ) phenotype 2. (b) Lane (1) phenotype I ; (2) phenotype 1-2; (3) phenotype 2.
854
R.Iida et al.
Electrophoresis 199O,Il, 852-855
Durapore membranes were substituted for Immobilon or nitrocellulose membranes; nitrocellulose gave high background staining. Heavy metals, CoCl, or NiC12,were adopted for the intensification of DAB-peroxidase reaction products for immunocytochemistry [21, but this technique did not produce good intensification for the immunoblotting version in the present study. Electrophoretic analysis of urine with commonly used immunostains requires a 100-fold to 1000-fold concentration in order to detect the trace amounts of protein that are normally present(l-2mg/100mL)~19,201.Unconcentratedor10-fold concentrated urine (10 pL) stained by the ordinary peroxidase-DAB method revealed no DNase I bands, whereas GOSS postintensification of the same strip revealed clear DNase I bands, comparable to the patterns obtained by conventional peroxidase-DAB staining of urine concentrated 100-fold (Fig. la). GOSS intensification reduces the effort required in analyzing large numbers of urine samples by replacing the need for exhaustive 100- to 1000-fold concentrations with relatively simple 10- to 20-fold concentrations. This should facilitate the development of urine electrophoresis as a diagnostic tool. GC, T F and PI patterns of diluted (ca. 1:1000-8000) human serum in PAG-IEF from different individuals are also shown in Fig. 2. Sensitivity comparison of the peroxidase-DAB/GOSS intensification with the conventional peroxidase-DAB staining is shown in Fig. 3. For the detection of G C phenotype 2, only 2.0 x pL of a sample of serum was used with the GOSS intensification, while 0.3 x lW3 pL,or more, of the serum was required with the conventional staining. When postintensification by the GOSS method was used to detect specific proteins immobilized on Durapore strips, protein bands appeared dark brown on avirtually unstained white background. Since excess deposits of DAB-polymer on the strips produce a high background, the DAB concentration and the incubation time for prestaining should be less than 0.01 % and 2 min, respectively. The patterns of DNase I, GC, TF and PI showed high contrast and were similar to the patterns obtained by conventional peroxidase-DAB staining [ 8-1 21. A 16-64-fold increase in sensitivity over our previous technique was thus achieved (Table 2).
Figure 2. PAG-IEF patterns of GC,'TF, and PI visualized by immunoblotting using peroxidase-DAB staining with intensification by the GOSS method. Anode is at the top. (a) Sixcommon phenotypes of G C detected in 5120-fold diluted serum samples (I0 pL) with Pharmalyte pH 4.5-5.4. Lane (1) phenotype 2-1s; (2) phenotype 1s; (3) phenotype 1F-IS; (4) phenotype 1 F; ( 5 ) phenotype 2-1F; (6) phenotype 2. (b) Three common phenotypes of T F detected in 7680-fold diluted serum samples ( 5 FL) with Ampholine pH 4-6.5. Lanes (1) and (4) phenotype C 1 ; (2) and (5) phenotype C2- 1; (3) and (6) phenotype C2. (c) Three common phenotypes of PI detected in 1200-fold diluted serum samples (5 wL) with Pharmalyte pH4.2-4.9 containing 1.2 o/o w/vACES. Lanes (1) and(4)phenotypeMl; (2) phenotype M l M 2 ; (3) phenotype M2.
Figure 3. Sensitivity comparison between (a) peroxidase-DAB/GOSS intensification, and (b) conventional peroxidase-DAB staining. Anode is at the top. G C protein was examined in a 10 FL sample of serially diluted serum of phenotype 2 by two methods. Conditions of electrophoresis were the same as in Fig. 2a. Lane( 1)serumdilution 1:100;lane(2) 1:200;lane(3) 1:400;lane(4) 1:800;lane(5) 1:1600;lane(6) 1 :3200;lane(7) 1:6400;lane (8) 1:12 800;lane(Y) 1:25 600;lane(10) 1:51 200.
Rapid and highly sensitive visualization in immunoblotting
Elecirophoresis 1990,11, 852-855
855
Table 2. Detection limits of four genetic markers (GC, TF, PI and DNase I) for immunoblotting using peroxidase-DAB/GOSS staining after PAG-IEF Genetic markers
GC TF PI DNase I
Before intensification Sample dilution")
Detection limit (ng)h)
1:160 1:120 1:120 1:l
19-38 83-125 66-146 N.D.c)
After intensification Sample dilutiona) 15120 1:7680 1:960-1920 12-16
Intensification rate
Detection limit (ng)h)
Before: After
0.3- 0.6 1.3- 2.0 4.1-18.2 NDC)
1:32 1:64 1:16-32 1: 8-16
a) These figures indicate degrees of serum dilution (GC, TF, PI) and urine dilution of 0.5 % solution of lyophilized urine sample, corresponding to a 50-fold concentration of the original urine [ 191. b) Detection limits were calculated from the following serum concentrations (GC: 0.3-0.6 mg/mL 1211, TF: 2-3 mg/mL [221, PI: 1.6-3.5 mg/mL[231), andfrorn thetransfereficiencyofabove90 %,sincetheresidual gels gave no band using protein staining after blotting. c) Not determined 19,241.
4 Concluding remark The GOSS postintensification technique described here is a convenient and rapid method that does not require expensive Or darkroom conditions for genetic and should prove useful for a variety of applications.
This work was supported in part by grantsfrom Japan Brain Foundation and Tokyo Immunopharmacology Institute, and by a Grant-in-Aidfor Scientific Research from The Ministry of Education, Science and Culture of Japan. We wish to express our appreciation to Miss Y. Ikehara and Mrs. F. Nakamura for their excellent technical and secretarial assistance. Received April 24, 1990
5 References 111 Renart, J., Reiser, J. and Stark, G. R., Proc. Natl. Acad. Sci. USA 1979, 76,3116-3120. [21 Adams, J. C., J. Histochem. Cytochem. 198 I, 29, 775. 131 Gallyas, F., Gorcs, T. and Merchenthaler, I., J. Histochem. Cytochem. 1982,30,183-184. [41 Newman, G. R., Jasani, B. and Williams, E. D.,J.Microscopy 1983, 132, RP1-RP2. [51 Merchenthaler, I., Gallyas, F. and Liposits, Z., Techniques in Immunochemistry 1989,4, 217-252. 161 Thomas, N., Bennett, R. andJones, C. N.,J.Immunol.Methods 1987, 104,201-207.
171 Kishi, K. and Yasuda, T., Hum. Genet. 1987, 75,209-212. 181 Kishi, K., Ikehara, Y., Yasuda, T., Mizuta, K. and Sato, W., Forensic Sci. Znt. 1990,45, 225-230. [91 Kishi, K., Yasuda, T., Awazu, S. and Mizuta, K., Hum. Genet. 1989, 81,295-297. 1101 Yasuda, T., Mizuta, K., Ikehara, Y. and Kishi, K., Anal. Biochem. 1989,183,84-88. I1 11 Yasuda, T., Ikehara, Y., Takagi, S., Mizuta, K. and Kishi, K., Hum. Genet. 1989,82, 89-91. [121 Sawazaki, K., Yamaba, T., Yasuda, T., Mizuta, K. and Kishi, K., Hum. Herd. 1990,40, 187-189. 1131 Yasuda, T., Sato, W. and Kishi, K., Biochim. Biophys. Acta 1988, 965, 185-194. 1141 Yasuda, T., Sato, W., Mizuta, K. and Kishi, K., Arn.J. Hum. Genet. 1988,42,608-614. [I51 De Blas, A. L. and De Cherwinski, H. M.,Anal. Biochem. 1983,133, 214-2 19. [161 Lanzillo, J. J., Stevens, J., Tumas, J . and Fanburg, B. L., Electrophoresis 1983,4, 313-316. I171 Surek, B. and Latzko, E., Biochem. Biophys. Res. Commun. 1984, 121,284-289. 1181 Iida, R., Sawazaki, K.,Ikehara,Y.,Yasuda,T.,Mizuta,K. and Kishi, K., Forensic Sci. Znt. 1990,47, 71-77. [ 191 Kishi, K. and Iseki, S., Proc. Japan Acad. 1982,58B, 229-23 1. [201 Iida, R.,Yasuda,T. and Kishi, K.,J.Biochem. 1987,101,357-363. L21 I Kawakami, M. and Goodman, De W. S., Biochemistry 1981,20, 5881-5887. [221 Goya, N., Miyazaki, S., Kodate, S. and Usio, B., Blood 1972, 40, 239-245. 1231 Dietz, A. A., Rubinstein, H. M. and Hodges, L. V., Clin. Ckem. 1974, 20,396-399. 1241 Kishi, K., Yasuda, T., Ikehara, Y., Sawazaki, K., Sato, W. and Iida, R., Am. J. Hum. Genet. 1990,47, 121-126.
R. Iida et al.
Reiko Iida Toshihiro Yasuda Daita Nadano Koichiro Kishi Department of Legal Medicine, Fukui Medical School, Fukui
Electrophoresis 1990,II, 852-855
Intensification of peroxidase-diaminobenzidine staining using gold-sulfide-silver:A rapid and highly sensitive method for visualization in immunoblotting A highly sensitive and rapid visualization method for protein detection by immunoblotting is described. Proteins blotted onto a Durapore membrane were visualized by the following procedure: after conventional peroxidase-based staining with 3,3'-diaminobenzidine (DAB), the produced DAB precipitates were intensified by treating with (i) gold trichloride (acid), (ii) sodium sulfide, and (iii) a developer containing silver nitrate. This postintensification method was employed for the detection of the genetic polymorphism of human proteins, such as deoxyribonuclease I in urine, and group specific component, transferrin and a,-antitrypsin in serum after polyacrylamide gel-isoelectric focusing, followed by immunoblotting. This postintensification technique was found to be simple, giving up to 16- to 64-fold amplification of the conventional peroxidase-DAB staining.
1 Introduction
2.2 Sample treatment
Since the first application of the blotting technique for specific detection of proteins in 1979 [ 11, many visualization systems have been developed including enzymatic, fluorescence, autoradiographic, and colloidal metal/silver development methods. The enzyme-based systems are preferred, since they are fast, easy, and safe, and therefore useful for routine work in the fields of biochemistry, forensic science and anthropology. Several workers have reported methods for postintensification of peroxidase-based staining with 3,3 '-diaminobenzidine (DAB) as immunocytochemical techniques 12-51. Especially the intensification by a gold-sulfide-silver (GOSS) method [41 has been found to be highly sensitive in histochemical use [61. We have employed the GOSS-intensification method for immunoblotting detection of some genetic markers in serum and urine by electrophoretic analysis, and succeeded in devising a rapid, sensitive and simplified version of this method. In this report, we describe the application of this new method for phenotyping analysis of deoxyribonuclease I (DNase I) in urine and group specific component (GC), transferrin (TF) and a,-antitrypsin (PI) in serum using polyacrylamide gelisoelectric focusing (PAG-IEF).
For T F and DNase I phenotyping, plasma or urine samples were treated with sialidase from Clostridiurn perfringens according to our previous reports [8,91. For PI phenotyping, 10 pL plasma was mixed with 10 pL 20 mM dithiothreitol and left for 30 min at room temperature, and then 10 pL 15 mM iodoacetamide was added and left for a further 30 min prior to focusing.
2 Materials and methods 2.1 Sample preparation
2.3 Antibodies and chemicals Rabbit antibodies against human GC, T F and PI were purchased from Dakopatts (Glostrup, Denmark); peroxidase-conjugated goat anti-rabbit immunoglobulin was from Bio-Rad (Richmond, CA). An antibody specific for human DNase I was produced in a rabbit as described previously [ 101. Carrier ampholytes Pharmalyte 2.5-5, 4.2-4.9 and 4.5-5.4 were purchased from Pharmacia (Uppsala, Sweden), Ampholine 4-6.5 from LKB (Bromma, Sweden), Durapore membrane from Millipore (Bedford, MA), gold trichloride (acid) from Nakarai (Kyoto, Japan), N-(2-acetamido)-2aminoethanesulfonic acid (ACES) from Sigma (St. Louis, MO), and 3,3'-diaminobenzidine.4HCI (DAB) from Dojin (Kumamoto, Japan). Other chemicals wereofreagent grade. 2.4 PAG-IEF
Plasma samples were collected from heparinized blood and stored at -80 OC until use. Urine samples were concentrated, dialyzed, and lyophilized according to our previous method 171.
PAG-IEF was performed mainly according to our previous papers [8, 10-121. Gels measuring 0.5 x 90 x 120 mm were prepared using the following components: 1.4 mL acrylamide-N,N'-methylenebisacrylamide(19.4 % w/v, 0.6 % w/v), 2.3 mL sucrose-glycerol (20 % w/v, 10 % vlv), 280 pL carrier ampholytes (Pharmalyte 4.5-5.4 for GC typing, Ampholine 4-6.5 for TF, Pharmalyte 4.2-4.9 for PI, and Pharmalyte 2.5-5 for DNase I), 1 mL distilled water, 10 pL Correspondence: K. Kishi, M. D., Department of Legal Medicine, Fukui N,N,N',N'-tetramethylethylenediamine,and 40 pL 1.2 % w/v Medical School, Matsuoka-cho, Fukui 910-1 1, Japan ammonium persulfate. In the case of PI phenotyping, 1 mL 6 % w/v ACES was added instead ofdistilled water. Plasma or Abbreviations: AB, avidin-biotin; ACES, N-(2-acetamido)-2-aminolyophilized urine samples were diluted with 0.3 % w/v bovine ethanesulfonic acid; DAB, 3,3'-diaminobenzidine; DNase I, deoxyribonuclease I; GC, group specific component; GOSS, gold-sulfide-silver ; serum albumin in order to protect them from pattern distorPAG-IEF, polyacrylamide gel-isoelectric focusing; PAP, peroxidase-anti- tion caused by a lowering of protein concentration. Electrophoresis was performed at a constant 2.5 W for 30 min, and peroxidase; PBS, phosphate-buffered saline; PBS-T, PBS containing then continued at 1000 V for 4 h under cooling at 10 "C. 0.05 96 Tween 20; PI, a,-antitrypsin; TF, transferrin 0VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1990
0173-0835/90/0909-08.52%3.50+.25/0
Electrophoresis 1990,11,852-855
2.5 Immunoblotting Transfer of proteins onto Durapore membrane strips (4 x 8 cm) was carried out by a capillary blotting method according to our previous papers [13,141. After transfer, the strips were briefly washed with 20 mM phosphate-buffered saline, pH 6.7, containing 0.05 % Tween 20 (PBS-T), and immersed in PBST containing 0.1 % w/v gelatin for 30 min. They were then washed with three changes of PBS-T and incubated for 1 h with an appropriate primary rabbit antibody diluted 1:400 in PBS-T. After washing three times in PBS-T, the strips wereincubated for 1 h with a peroxidase-conjugated secondary antibody (goat anti-rabbit immunoglobulin) diluted 1:400 in PBS-T. After washing three times with PBS-T, they were incubated for 2 min in PBS containing 0.01 % wlv DAB and 0.005 % v/v H,O, until the background was stained faintly pink, followed by washing with distilled water to stop the reaction. They were then kept in distilled water until further intensification.
2.6 Postintensificationprocedure (GOSS method) Following peroxidase-DAB staining, the protein patterns on the strips were detected using the improved immunoblotting version of the GOSS method for histochemical staining [41, as outlined in Table 1. The conditions shown are optimal, and the concentrations of the reagents and other particulars are indicated in the footnotes to Table 1.Durapore membranes were manipulated with clean Teflon forceps.
3 Results and discussion The most commonly nonradioactive protein detection methods employ intense immunoblotting stains, such as peroxidase-antiperoxidase (PAP) [ 151, avidin-biotin (AB) 161, and immunogold/silver [ 171. These stains lack the appropriate sensitivity to detect proteins present at low or trace concentrations (unpublished observation), although we have applied them to genetic studies of urinary proteins [7-10, 181. Body fluids, such as cerebrospinal and amniotic fluids, are often difficult to obtain in quantity, and frequently contain many proteins. This sensitivity problem was overcome by the application of a histologically derived GOSS postintensification procedure [41 for proteins on immunoblotting transfer membranes. Although the principle of the intensification is unknown, it
Rapid and highly sensitive visualization in irnrnunoblotting
853
Table 1. Postintensification procedure for peroxidase-DAB staining by the GOSS method") 1 Incubate the strips (4 x 8 cm) for 5 min in 0.1 % w/v gold trichloride (acid)b) and wash in three changes of distilled water (20 mL). 2 Incubate for 5 min in neutralized 0.3 % w/v sodium sulfide)and wash in three changes of distilled water (20 mL).
3 Incubate in an ice-cold silver developeld) until the desired intensification of the protein bands is obtained (about 10-20 min). 4 Rinsequickly intwochangesof 1 %v/vaceticacid tostopthereaction.
5 Incubate in a destaining solutione)until the background becomes clear (- 1-5 min). 6 Wash in a large volume of distilled water and then dry.
a) Unless otherwise indicated, all steps were performed with gentle shaking at room temperature (18-20 "C) in a glass tray (2 x 6 x 10 cm) with a reagent volume of I0 mL. b) Hydrogen tetrachloroaurate (1-), tetrabydrate (HAuC1,.4H2O) is of the purest grade and should be kept tightly closed and protected from light. c) Adjust the pH to 7.0 with 3 N HC1. d) Stock solution A: dissolve 2.5 g of sodium carbonate, anhydrous in 300 mL of distilled water. Stock solution B: dissolve 0.1 g of ammonium nitrate, 0.1 g of silver nitrate, 5 g of tungstosilicic acid and 0.25 mL of 37 %formaldehyde, inthatorder,in300mLofdistilledwater.Thestock solutions are kept at 4 "C. The silver developer is prepared fresh just before use by slow addition of 5 mLof stock solution A to 5 mL of stock solution B with stirring, and then chilledon ice. The chemicals used here are of the purest grade commercia!ly~available. e) Dissolve 2.5 g of sodium thiosulfate (pentahydrate) and 0.15 g sodium sulfite (anhydrous) in 40 mL of 1 % v/v acetic acid. Although not essential, it is often convenient to obtain avery clear background. It is possible to substitute 5 % w/v Fujifix, an acid hardening fixerreagent(Fuji Photo Film, Tokyo, Japan), for the solution.
may be based on the high afinity of DAB for metal salts and catalyzing acitivity ofmetal sulfides for silverdeposition under reducing conditions 141. The postintensification procedures described in this study are subject to the following three major variables: the temperature of the reagents, reagent concentration, and time of application of each reagent. In order to achieve a low background and slow development, the incubation temperature of the developer (step 3 inTable 1) should be kept as low as possible. The final intensity of the stain is also under the investigator's control by deciding when to stop the reduction reaction (steps 4 and 5 in Table 1).As theconcentration of stock solutions A and B for developer were diluted to one sixth of the original [41, no precipitation appeared when they were mixed without vigorous stirring. Moreover, the time required for the intensification was shortened to about 30 min.
Figure 1. (a) Pattern and (b) schematic representation of PAG-IEF of DNase I visualized by immunoblotting using peroxidase-DAB staining with intensification by the GOSS method. Anode is at the top. (a) Three common phenotypes of DNase I detected in 0.05 % w/v lyophilized urine samples (5 yL), corresponding to an approximately 5-fold concentration of the original urine with Pharmalyte pH 2.5-5. Lanes (1) and (4) phenotype 1; (2) and (5) phenotype 1-2; (3) and ( 6 ) phenotype 2. (b) Lane (1) phenotype I ; (2) phenotype 1-2; (3) phenotype 2.
854
R.Iida et al.
Electrophoresis 199O,Il, 852-855
Durapore membranes were substituted for Immobilon or nitrocellulose membranes; nitrocellulose gave high background staining. Heavy metals, CoCl, or NiC12,were adopted for the intensification of DAB-peroxidase reaction products for immunocytochemistry [21, but this technique did not produce good intensification for the immunoblotting version in the present study. Electrophoretic analysis of urine with commonly used immunostains requires a 100-fold to 1000-fold concentration in order to detect the trace amounts of protein that are normally present(l-2mg/100mL)~19,201.Unconcentratedor10-fold concentrated urine (10 pL) stained by the ordinary peroxidase-DAB method revealed no DNase I bands, whereas GOSS postintensification of the same strip revealed clear DNase I bands, comparable to the patterns obtained by conventional peroxidase-DAB staining of urine concentrated 100-fold (Fig. la). GOSS intensification reduces the effort required in analyzing large numbers of urine samples by replacing the need for exhaustive 100- to 1000-fold concentrations with relatively simple 10- to 20-fold concentrations. This should facilitate the development of urine electrophoresis as a diagnostic tool. GC, T F and PI patterns of diluted (ca. 1:1000-8000) human serum in PAG-IEF from different individuals are also shown in Fig. 2. Sensitivity comparison of the peroxidase-DAB/GOSS intensification with the conventional peroxidase-DAB staining is shown in Fig. 3. For the detection of G C phenotype 2, only 2.0 x pL of a sample of serum was used with the GOSS intensification, while 0.3 x lW3 pL,or more, of the serum was required with the conventional staining. When postintensification by the GOSS method was used to detect specific proteins immobilized on Durapore strips, protein bands appeared dark brown on avirtually unstained white background. Since excess deposits of DAB-polymer on the strips produce a high background, the DAB concentration and the incubation time for prestaining should be less than 0.01 % and 2 min, respectively. The patterns of DNase I, GC, TF and PI showed high contrast and were similar to the patterns obtained by conventional peroxidase-DAB staining [ 8-1 21. A 16-64-fold increase in sensitivity over our previous technique was thus achieved (Table 2).
Figure 2. PAG-IEF patterns of GC,'TF, and PI visualized by immunoblotting using peroxidase-DAB staining with intensification by the GOSS method. Anode is at the top. (a) Sixcommon phenotypes of G C detected in 5120-fold diluted serum samples (I0 pL) with Pharmalyte pH 4.5-5.4. Lane (1) phenotype 2-1s; (2) phenotype 1s; (3) phenotype 1F-IS; (4) phenotype 1 F; ( 5 ) phenotype 2-1F; (6) phenotype 2. (b) Three common phenotypes of T F detected in 7680-fold diluted serum samples ( 5 FL) with Ampholine pH 4-6.5. Lanes (1) and (4) phenotype C 1 ; (2) and (5) phenotype C2- 1; (3) and (6) phenotype C2. (c) Three common phenotypes of PI detected in 1200-fold diluted serum samples (5 wL) with Pharmalyte pH4.2-4.9 containing 1.2 o/o w/vACES. Lanes (1) and(4)phenotypeMl; (2) phenotype M l M 2 ; (3) phenotype M2.
Figure 3. Sensitivity comparison between (a) peroxidase-DAB/GOSS intensification, and (b) conventional peroxidase-DAB staining. Anode is at the top. G C protein was examined in a 10 FL sample of serially diluted serum of phenotype 2 by two methods. Conditions of electrophoresis were the same as in Fig. 2a. Lane( 1)serumdilution 1:100;lane(2) 1:200;lane(3) 1:400;lane(4) 1:800;lane(5) 1:1600;lane(6) 1 :3200;lane(7) 1:6400;lane (8) 1:12 800;lane(Y) 1:25 600;lane(10) 1:51 200.
Rapid and highly sensitive visualization in immunoblotting
Elecirophoresis 1990,11, 852-855
855
Table 2. Detection limits of four genetic markers (GC, TF, PI and DNase I) for immunoblotting using peroxidase-DAB/GOSS staining after PAG-IEF Genetic markers
GC TF PI DNase I
Before intensification Sample dilution")
Detection limit (ng)h)
1:160 1:120 1:120 1:l
19-38 83-125 66-146 N.D.c)
After intensification Sample dilutiona) 15120 1:7680 1:960-1920 12-16
Intensification rate
Detection limit (ng)h)
Before: After
0.3- 0.6 1.3- 2.0 4.1-18.2 NDC)
1:32 1:64 1:16-32 1: 8-16
a) These figures indicate degrees of serum dilution (GC, TF, PI) and urine dilution of 0.5 % solution of lyophilized urine sample, corresponding to a 50-fold concentration of the original urine [ 191. b) Detection limits were calculated from the following serum concentrations (GC: 0.3-0.6 mg/mL 1211, TF: 2-3 mg/mL [221, PI: 1.6-3.5 mg/mL[231), andfrorn thetransfereficiencyofabove90 %,sincetheresidual gels gave no band using protein staining after blotting. c) Not determined 19,241.
4 Concluding remark The GOSS postintensification technique described here is a convenient and rapid method that does not require expensive Or darkroom conditions for genetic and should prove useful for a variety of applications.
This work was supported in part by grantsfrom Japan Brain Foundation and Tokyo Immunopharmacology Institute, and by a Grant-in-Aidfor Scientific Research from The Ministry of Education, Science and Culture of Japan. We wish to express our appreciation to Miss Y. Ikehara and Mrs. F. Nakamura for their excellent technical and secretarial assistance. Received April 24, 1990
5 References 111 Renart, J., Reiser, J. and Stark, G. R., Proc. Natl. Acad. Sci. USA 1979, 76,3116-3120. [21 Adams, J. C., J. Histochem. Cytochem. 198 I, 29, 775. 131 Gallyas, F., Gorcs, T. and Merchenthaler, I., J. Histochem. Cytochem. 1982,30,183-184. [41 Newman, G. R., Jasani, B. and Williams, E. D.,J.Microscopy 1983, 132, RP1-RP2. [51 Merchenthaler, I., Gallyas, F. and Liposits, Z., Techniques in Immunochemistry 1989,4, 217-252. 161 Thomas, N., Bennett, R. andJones, C. N.,J.Immunol.Methods 1987, 104,201-207.
171 Kishi, K. and Yasuda, T., Hum. Genet. 1987, 75,209-212. 181 Kishi, K., Ikehara, Y., Yasuda, T., Mizuta, K. and Sato, W., Forensic Sci. Znt. 1990,45, 225-230. [91 Kishi, K., Yasuda, T., Awazu, S. and Mizuta, K., Hum. Genet. 1989, 81,295-297. 1101 Yasuda, T., Mizuta, K., Ikehara, Y. and Kishi, K., Anal. Biochem. 1989,183,84-88. I1 11 Yasuda, T., Ikehara, Y., Takagi, S., Mizuta, K. and Kishi, K., Hum. Genet. 1989,82, 89-91. [121 Sawazaki, K., Yamaba, T., Yasuda, T., Mizuta, K. and Kishi, K., Hum. Herd. 1990,40, 187-189. 1131 Yasuda, T., Sato, W. and Kishi, K., Biochim. Biophys. Acta 1988, 965, 185-194. 1141 Yasuda, T., Sato, W., Mizuta, K. and Kishi, K., Arn.J. Hum. Genet. 1988,42,608-614. [I51 De Blas, A. L. and De Cherwinski, H. M.,Anal. Biochem. 1983,133, 214-2 19. [161 Lanzillo, J. J., Stevens, J., Tumas, J . and Fanburg, B. L., Electrophoresis 1983,4, 313-316. I171 Surek, B. and Latzko, E., Biochem. Biophys. Res. Commun. 1984, 121,284-289. 1181 Iida, R., Sawazaki, K.,Ikehara,Y.,Yasuda,T.,Mizuta,K. and Kishi, K., Forensic Sci. Znt. 1990,47, 71-77. [ 191 Kishi, K. and Iseki, S., Proc. Japan Acad. 1982,58B, 229-23 1. [201 Iida, R.,Yasuda,T. and Kishi, K.,J.Biochem. 1987,101,357-363. L21 I Kawakami, M. and Goodman, De W. S., Biochemistry 1981,20, 5881-5887. [221 Goya, N., Miyazaki, S., Kodate, S. and Usio, B., Blood 1972, 40, 239-245. 1231 Dietz, A. A., Rubinstein, H. M. and Hodges, L. V., Clin. Ckem. 1974, 20,396-399. 1241 Kishi, K., Yasuda, T., Ikehara, Y., Sawazaki, K., Sato, W. and Iida, R., Am. J. Hum. Genet. 1990,47, 121-126.
