654
J tleukeshoven and R. Dcrnicli
Jochen Heukeshoven Rudolf Dernick Heinrich-Pette-lnstitut fur Experimentelle Virologie und lmmunologie an der Universitiit Hamburg
Native horizontal ultrathin polyacrylamide gel electrophoresis of proteins under basic and acidic conditions The preparation of homogeneous ultrathin native polyacrylamide gels, using a basic as well as an acidic buffer system is described. The basic buffer system consists of Tris-HCl/Tris-glycine, the same buffer as in sodium dodecyl sulfate (SDS)-gel electrophoresis but without SDS. The acidic system uses potassium acetate, pH 4.3, as gel buffer and 0-alanine, pH 4.6, acetic acid as electrolytes. The gels are covalently bound on glass plates. Binding of acidic gels requires a special pretreatment of glass plates. The whole procedure is simple and extraordinarily fast: 100-120 min from the start of gel preparation to the end of electrophoresis. Coomassie staining is done in 40 min and silver staining in 90 min.The native gels are excellently suited for diffusion blotting. Further attractive properties of these gels are easy handling, simple drying and dimensional stability.
result from two parameters, gel thickness and fixation on a support.
1 Introduction Electrophoresis *separationtechniques with a horizontally arranged separation matrix have been known for many years. Paper electrophoresis and electrophoresis on cellulose acetate sheets in clinical diagnostics are the classical methods of horizontal electrophoresis. Later, peptide separations on thin layers of' cellulose or silica gel and several applications of electrophoresis in agarose, including isoelectric focusing (IEF) in agarose, were introduced. With the use of polyacrylamide gels as separation matrix, horizontal I E F in thin layers was developed [l-31. Although gels are mainly used in a vertical configuration at present, some techniques of horizontal sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) were described recently [4-71. Native electrophoresis in acidic or basic media is performed relatively seldom and then exclusively in vertical gels [8-111. Quite recently horizontal native electrophoresis was demonstrated using the PhastSystem, a semi-automatic electrophoresis unit [12]. We have developed a horizontal gel system for SDS-PAGE which uses ultrathin, homogeneous gels (without a stacking gel) bound on glass plates.The preparation of these gels is simple and fast, the separation power is excellent, and staining and drying is faster in comparison to conventional 1 m m slab gels [13, 141. Here we describe the adaption of this gel system to native electrophoresis in both basic and acidic media. Preliminary results were presented previously [ 151. We prcfer horizontal gel systems because they possess a number of advantages in comparison to vertical systems. Horizontal gels are usually thin gels bound to rigid supports [ I , 3,5, 7) and therefore their favorable properties .-
Correspondence: Dr. Jochen Heukeshoven, Heinrich-Pette-Institut fiir ExperiinentelleVirologie und lmmunologie an deruniversitiit Hamburg, Mariinistr. 52. DW-2000 klamburg 20, Germany Abbreviations: IEF, isoclectric focusing; SDS-PAGE, sodium dodecyl sulfate-polyacrylaniide gt.1 electrophoresis; TEMED, N,iV,N', N'-tetramethplethyienedlanlinc. T, total gel concentration = acrylamjde plus crosslinking agent W / C ; 0% C, gcl crosslinking =percent crosslinking agent/ percent total gel concentralion
The specific characteristics of horizontal gels are summarized briefly. Horizontal gels are open systems and consequently it is feasible to use nearly all gel dimensions or parts of a gel in a single apparatus; the geometry and number of sample slots are freely selectable. Fitting problems with gel cassettes and buffer tanks will not arise. Sample application is simplified. The use of thin gels and buffer strips results in a remarkable reduction in the consumption of chemicals (acrylamide, buffers). Gels bound to supports are dimensionally stable during washing, staining, drying, blotting [ 161and, if necessary, rehydration. Especially in the case of gels < 0.3 mm, all processes of electrophoresis, fixation, washing, staining and drying are significantly faster than with 1 m m thick gels. Furthermore, the sharpness of bands and detection sensitivity are improved. In our gel system we use exclusively homogeneous gels, developed in the first instance for SDS-PAGE 1141. This gel type has additional advantages over those with stacking and separation gels. The preparation is extraordinarily simple: we need only one gel solution and after a single pouring step the gel is ready for use. We do not need further equipment, e.g. gradient mixers. The polymerization conditions can be optimally standardized, resulting in a more reproducible preparation of gels. Staining and drying is less critical than with gradient gels.The risk of breaking and nonuniform staining is not a problem. All advantages acquired in our previous work with SDS-PAGE have now been incorporrated into the preparation and use of native gels.
2 Preparation of thin homogeneous gels 2.1 Glass plates versus polyester sheets We have bound the homogeneous thin gels only on glass plates as support because (i) glass is cheaper than pretreated polyester sheets, (ii) it can be used repeatedly, and (iii) it is easier to handle during casting, staining, blotting [16] and, above all, during drying. The anchorage of poly-
N a t i v e polyacrylamide gel electrophorcsi?
EIi~rirophori~.\is1992. 13, 654-659
655
acrylamide gels on glass is normally better (exceptions see Section 4.1) than on polyester. On glass, sharper zones are usually deserved. The gel layer is always 0.25 mm thick. 2.2 Gel preparation The preparation of the casting mold was described earlier [14].A thick glass plate of 4-5 mm is provided with spacers and slot formers and is then hydrophobized with trimethylsilyl chloride. A 1.5-2 mm glass plate is used as support to bind the gel covalently by reaction with the 3-methacryloxypropyl groups of Polyfix 1000 [3,7,17,18].The special activation of glass plates for use in acidic polymerized gels containing detergents will be described in Section 4.1. Gels are cast horizontally using a combination of capillary attraction and sliding [7], as described earlier for SDS gels [14]. The slot plate (slots at the top), laying on a table or a bigger glass plate, is fixed with the aid of a water film for better adherence, with 80% of its length being covered by the gel plate. The mixed and degassed gel solution is pipetted slowly on the open tenth of the slot plate (Fig. I). Entry of the polymerization solution between the two glass plates should be promoted by slightly knocking on the gel plate near the front of the solution. Finally the gel plate is slid over the remaining 20% to cover the slot plate completely. With this technique the polymerization solution can be introduced without air bubbles within 20--30 s. The casting device is weighted by about 1 kg to press the two glass plates together tightly. The basic gel is polymerized within 3045 min at room temperature or after 15 rnin at 50-60°C and is then ready for use. The acidic gels must be kept at 5060°C for 15 min to start and complete the polymerization, otherwise the gel does not polymerize. The procedure is always the same for SDS, basic and acidic native gels; only the polymerization solutions differ. Gel preparation is simple and requires only 30 min from mixing the solutions to the finished polymerized gel, if all stock solutions and the glass plates were prepared before. Rehydration of a dried gel is possible but requires the same or more time than the preparation of a fresh gel.
Figuic I. Horizontal casting procedure of ultrathin polyacrylamide gels. The gel solution is pipetted slowly on the open slot plate (about 20% of total length).
Table 1 Composition ofgel solutions for preparation o f a native standard uolvacrylamide nel 210 X 110 X 0.25 mm Solution Acrylamide, 40% TJ2% C Acrylamide, 40% T/10% C KAc, IM, pH 4.3 Tris/WCl, 2 ~pH , 8.9 Prosolv IT Tetrabutylammonium . H S0 4 , 10% H2O
2.40
1.80 0.60 0.80
mL
4.00
-
-
0.032 0.04 4.61 0.02 0.10 8.00
1.54 0.01 0.05 8.00
Persulfate 10% Total
5
Acidic gel
-
TEMED
0
Basic gel mL
10
15
%T
3 Basic gels
Figure 2. Electrophoresis in four homogeneous ultrathin native polyacrylamide gels in Tris/HCI buffer, pH 8.9, with 2% C and 6, 8, 10 and
3.1 Gel buffer
12% T. Electrophoresis of marker proteins until the dye Bromothymol Blue reached the end of the gel. Silver staining. The log Rf values (mobility related to dye front) were calculated and plotted against %T.
Three buffer system previously used in native gradient gels [ 19-21], were adapted to our homogeneous gels. Neither the Tris-borate-EDTA system nor the Tris-acetate system produced sufficiently sharp protein bands. The best choice, and also the simplest, was the Tris-HCl/Tris-glycine system, already applied by us in ultrathin SDS-PAGE [14]. We employ the same buffer at a high concentration (1 M TrisHCl, pH 8.9) as in SDS-PAGE but without SDS [15]. 3.2 Gel compositon The electrophoresis mobility of protein depends on their size and their charge at a given pH as well as on their retardation by the gel matrix. We recommend starting with an intermediate acrylamide concentration of 10-12 O/o T and 2 % C (Table 1). To optimize the separation and to estimate molecular weights using a Ferguson plot (log mobility vs.
%T) [22], 3-4 additional gels with higher or lower acrylamide contents are necessary. The useful range may be between 5 and 17% T. 3.3 Ferguson plot
Due to the simple and fast preparation ofhomogeneous native gels (4-5 gels with different acrylamide content can be produced and run per day), a graph of a Ferguson plot is easy to draw [22]. Data from four gels with 6,8,10 and 12%T were evaluated. The relative mobility, R, (related to the dye front) of some marker proteins was calculated and their log values were plotted against %T (Fig. 2).We obtained straight lines, from which we constructed a calibration
656
J Heukeahoven m d R Dernick
E l r o r o p h o r a i s 1992, J3, 654-689
curve to estimate the molecular weights. Only a few of the obtained values, mainly from aggregated proteins, do not fit in the linear (calibration curve (Fig. 3). 3.4 Electrophoresis buffer and paper buffer strips The electrophoresis buffer (0.4 M Tris - 0.1 M glycine, pH 8.6), the same as in SDS-PAGE but without SDS, is identical for cathode and anode. As electrode bridges we use filter paper strips, 3-4 cm wide and 20-21 cm long. They should be able to take up 15-20 m L of electrophoresis buffer. We use 7-10 layers of 4 cm wide filter paperfrom Schlei-
'50
I----
chert&Schiill (AMGL2043). This paper is sufficiently clean so as not to impair silver staining. The quality of the filter paper is not as critical in native electrophoresis as it is in SDS-PAGE, and therefore it is not necessary to use polyacrylamide gel strips. The paper strips overlap the gel about 5 m m and their full length is tightly pressed against the gel layer by a glass plate. The platinum electrodes are placed on the strips at the greatest distance from each other.
3.5 Electrophoretic conditions Electrophoresis is performed for 75-90 min at limited settings of 400 V,,,,, 12 W,,,, and 20 mA,,,, and the temperature of the cooling plate is set to 12"C.These values are optimal for our standard gels (12% T; 210 mrn wide, 110 m m long and 0.25 m m thick). For gels with a higher acrylamide concentration the power should be slightly reduced, to 10 Wrn',,.
BSA
Y 0
I 50
150
100
TIM OV The CA TI MVO *La
200
Mot -Gew in kD
f i g u r e 3. Fcrguson p l o ~based on data obtained from Fig. 2 and a similar set of log R, - '%T.The slope of the curves was estimated and plotted against molecular weight. Abbreviations: Pho. phosphorylase: BSA, bovine serum albumin; TIM, triosephosphate isomerase; OV, ovalbumin; The, thermolysin; CA, carbonic anhydrase; TI, trypsin inhibitor; Myo, myoglobin horse; a-La, a-lactalbumin; P-Gal. 8-galactosidase; ALD, aldolase;(XX),. aggrcgates of protein XX.
3.6 Sample preparation Extracts or protein samples should be dissolved in Tris/ HC1 buffer, pH 8.5-9.0; small amounts (0.1-0.2 O/o) of mild detergents like Triton or Tween may be added. The buffer concentration is not critical and up to 0.5 M will be tolerated, because the applied sample volume is only 3-4 pL and the gel buffer also has a high molarity. The amount of protein in this volume should be in the range of 10-100 ng/ band (3-30 p/mL) if the gel is silver stained, or 100-1000 ng/band when utilizing an effective Coomassie staining. As tracking dye we recommend Bromothymol Blue. Figure 4 shows separations of a mixture of marker proteins, selected single marker proteins and extracts of bovine and fish muscle.
Flgure 4. Native polyacrylamide gel electrophoresis in ultrathin, homogeneous gel with 12S0/uT and 2%C under basic conditions using 1 M Tris/HCI gel buffer, pH 8.9, and 0.1 M Trid0.4 M glycine electrode buffer, pH 8.6. Original size: 210 X 110 X 0.25 mm; electrophoresis condi-
tions: 400 Vn,:,,> 20 mA,,,,, and 12W,,,
for 90 min. Silver staining.
Elccrrophoreris 1992, 13, 654-659
4 Acidic gels 4.1 Pretreatment of glass plates and gel buffer As mentioned in Section 2.2, acrylamide can be polymerized at acidic pH value, 4.3, initiated by N,N,N',N'-tetraniethylethylenediamine (TEMED)-persulfate, if the concentration of persulfate is doubled and the polymerization is performed at higher temperature (50-60°C). Gels prepared under these conditions adhere well to 3-methacryloxypropyltrimethoxy silane pretreated glass plates. None of the buffer systems tested produce sufficiently sharp bands, the modified potassium acetate@-alanin system according to Reisfeld et al. [23] being most promising. Zones with extensive tailing indicated an interaction between glass plate and proteins. Therefore, low concentrations (0.1-0.2% Triton X-100, Nonidet P-40,Tween) of nonionic detergents were added to overcome the adsorption. Indeed, the detergent had the desired effect, but in the presence of neutral detergents the gel is no longer bound covalently to the glass plate, although glass plates pretreated in the same way are able to bind gels prepared without neutral detergents. The detergent-containing gel detached at the latest during the fixation or staining step. The protein band in such gels were relatively sharp.
Native polyacrylamide gel
deCtIODhoreSlS
657
The gel buffer was an adaption of the buffer system of Reisfeld et al. [23].The pH was varied between 4 and 5, the concentration between 0.05-0.4 M potassium acetate. The buffer concentration could not be as high as in basic native gels. The best results were obtained with 0. I M potassium acetate, pH 4.3. Sodium acetate and ammonium acetate, according to Bonner et al. 1241,were also tested but the separations were not as good as with potassium acetate. Prosolv I1 (0.1 Yo from Elektrophorese-Technik, Leonberg, Germany) was the best choice as nonionic detergent, with cleaner separations than with Triton X-100. A small amount (0.05%) of tetrabutylammonium hydrogensulfate promotes a more uniform dispersion of the proteins in the sample slots.
4.2 Gel composition
Acidic gels have larger pores than basic gels at the same acrylamide concentrations.This could be shown with a washed and dried basic gel, rehydrated in an acidic buffer. The electrophoretic mobility of the proteins was remarkably reduced in such a rehydrated geLTherefore, using acidic polymerization, we recommend a gel composition of 12O/o T and 4% C as standard gel (Table 1). Surprisingly, the migration properties do not change with a variation of gel concentration. Anumber of proteins were run in acidic gels with 8,10, For improved gel binding, the glass plates were intensively 12, 14, 16, and 18% T, and their migration distances were washed with 20% potassium hydroxide in 50% ethanol, compared. The same patterns were observed. With increaswith 10% chromosulfuric acid, and with water. These pre- ing gel concentration the protein bands were sharper, but treated plates were reacted with 10% 3-methacryloxypro- gels with a high acrylamide content tend to dry out during pyltrimethoxy silane in chloroform or dimethylformamide electrophoresis. at 60°C. However, none of these preventive measures improved gel binding in the presence of neutral detergents. 4.3 Electrophoresis buffer and paper buffer strips It was obvious that the number of reactive sites on the glass surface was too low to firmlybind the gel. We can onlyspec- Reisfeld et al. [23] used (3-alanine acetate, pH 4-5, while in d a t e that the acryl-silane residues on the surface were their acidic system, Bonner et al. [24] employed glycine acepartly hidden by a layer of detergent molecules, preventing tate, approximately pH 3, as running buffer. In our experithe reaction with acrylamide during polymerization. To ments the amino acids glycine, alanine and 6-alanine were overcome this problem we tried to enlarge the reactive sur- chosen as cationic buffer substances because they are relatiface by a mechanical or chemical procedure. The simplest vely inexpensive.The buffer capacity of glycine and alanine and most efficient method is to frost the glass surface with at pH > 4 is low and therefore, with them, an acetate buffer waterproof emery paper of finest grain size (400-600) for could only be used at pH < 4. Only P-alanine can be used 0.5-1 min. A more homogeneous buffing can be achieved with an acetate buffer in the pH range 4-5. The results of a by chemical corroding with 10Vo hydrofluoric acid or a mix- few electrophoretic separations demonstrated that the moture of 10% KF/lO% H,SO, for 20 h at room temperature. bility of a number of marker proteins depends only slightly Because of the risk in working with these chemicals we pre- on the pH in the range ofpH 4.3-4.8.To obtain an adequate fer the simpler emery paper treatment, which makes the buffer capacity, moderate conductivity, and also a tolerable glass surface highlyreactive,so that the binding of gels after electrophoresis time, we have to adjust both the pH and activation with 3-methacryloxypropyltrimethoxy silane is buffer concentration accordingly. The acceptable concenalso adequate in the presence of detergents. Not all rsilanol tration range lies between 0.1-0.3 M 0-alanine at a pH of groups react with the acryl silane, resulting in a residual ad- 4.3-4.8. The optimum anode buffer contains 0.2 M (3-alasorption partially inhibiting the formation of sharp protein nine, adjusted with acetic acid to pH 4.6. As cathode solubands, especially in work with basic proteins. An additional tion we used 1'Yo acetic acid. As with basic gels, the same coating with trimethylsilyl chloride was essential, as in the kind of buffer strips (7-10 layers, 4 cm wide filter paper endcapping procedure known from chromatographic silica strip) are applied and soaked with 15-20 mL of the elecmatrices. Summarizing, the pretreatment of glass plates trode solutions. suitable for binding acidic gel prepared with neutral detergent requires the following steps: (i) cleaning in alkaline solution (10% NaOH, also suitable for removing gels, over- 4.4 Electrophoretic conditions night), (ii) washing in acid (5% sulfuric acid, minimum 1 h), Electrophoresis of acidic gels is performed under similar (iii) frosting with wet emery paper (0.5-1 min), (iv) coating with 3-methacryloxypropyltrimethoxy-silane, 2% chloro- conditions as with thin homogeneous SDS-gels. The standard gels were run for 60-80 min at limited values of 600 form, (v) second coating with trimethylsilyl chloride, 5 O/o in chloroform, and (vi) rinsing with ethanol and drying. V, 15 mA,,, and 30 W,,, at a temperature of 12°C.Power
658
J H e u k a h o v e n a n d R. Dei-nich
and current can be adjusted to highervalues than with basic gels because the electrosmotic transport of water is reduced under acidic conditions.
Electroiihor?~t,\1992. 13. 654-630
background is unnecessary. The gel i s rinsed shortly with water and dried in air. 5.2 Silver staining
4.5 Sample preparation Proteins must be soluble in acidic buffer at pH 4-5. Small amounts of mild tletergents may improve the solubility.We obtained the best results with 0.1 M P-alanine-acetate buffer, pH 5.0, containing 0.2 %Tween 20. Methyl Green or Malachite Green are suitable as tracking dye. The amount of protein which is applicable to the gel is in the same range as in basic gels. Figure 5 shows a separation of marker proteins, some individual proteins und some extracts in an acidic native polyacrylamide gel.
5 Staining Coomassie and silver staining methods described for ultrathin SDS-gels [141 were worked out for the staining of native gels, whereby it was not necessary to differentiate between basic and acidic gels.
5.1 Coomassie staining Fast staining is accomplished in two steps: (i) fixation with 10% trichloroacetic acid for 10 min; (ii) the gels are stained with 0.002% Serva Blue R in 10% acetic acid. The staining time is usually 1-2 h at room temperature,but it can be considerably accelerated by heating the solution to 50-60°C. Agitation ofthe gel for 10-20 rnin with a hot solution is adequate. At the low dye concentration, destaining to clear the
Silver staining is performed according to our method as published previously for the PhastSystem 1251and ultrathin SDS-gels [14,26]. After 10 rnin fixation in 10% trichloroacetic acid, a second fixation step is applied for 30-60 min in a solution containing 0.5% glutaraldehyde, 0.25 O h sodium thiosulfate and 0.4 M sodium acetate in 30% ethanol. The gel is thoroughly washed with water before it is incubated in 0.2% silver nitrate for 20 min. Finally, the usual proceTable 2 Staining protocol for ultrathin basic or acidic native polyacrylamide gels Coomassie staining 1 Fixation in 10% trichloroacetic acid at room temperature 2 Staining with 0.002% Serva Blue R at 50-60°C in 100/0 acetic acid 3 Destaining not necessary Total
Silver staining 1 Fixation i n 10% trichloroacetic acid at room temperature 2 Fixation in 30% ethanol with 0.4 M / L sodium acetate 0.25% Na2S203,0.5"/0 glutaraldehyde 3 Wash with water, 3 X 5 min 4 Silver nitrate, 0.2% 5 Development in 2.5% Na2C03, pH 11.7 and 0.01 Vo formaldehyde 6 Stop bath, 1% glycine 7 Wash with water, 2 X 3 rnin Total
10 min 30 min 0 min 40 rnin
10 min
30 rnin 15 min 20 min
3-5 min 1-2 min 6 min 87-90 min
Figure 5. Native polyacrylamide gel electrophoresis in ultrathin, homogeneous gel with 12%T and 4%0Cunder acidic conditions using 0.1 M potassium acetate gel buffer, pH 4.3, with additives given in Table 1, and 0.2 M 8-alanine/acetate, pH 4.6, as electrode buffer. Original size: 210 X 110 X 0.25 mm; electrophoresis conditions: 600V,,,, 15 mA,,, and 30 W,,, for 75 min. Samples dissolved in 0.1 M 8-alanine buffer, pl-1 5 , with 0.2% Tween to decrease the absorption of proteins in the sample vial. Silver staining.
Elccrrophotrsis 1992, 13, 654-659
dure of development with 2.5 010 sodium carbonate/O.Ol O/o formaldehyde follows. The staining protocols are given in Table 2. The indicated times are minimum values and can be extended (except for the fixation with trichloroacetic acid and development with carbonate solution).
Native polyacrylamide gel electrophoresis
659
The Heinrich-PtJtte-lnstitueis financially supported by Freie und Hansestudt Hamburg and by Bundesministerium f u r Gesundheit, Bonn. The authors thank Anja Kirschtier for excellent technical assistance. Received August 11,1992
6 Blotting
7 References
The transfer from polyacrylamide gels to nitrocellulose or Immobilon membranes is an important technique, especially for proteins separated in the native state. For immunological studies the conservation of native structures is often essential. Using ultrathin gels bound to a support, the transfer can only be achieved by diffusion b1otting.A~demonstrated previously with homogeneous SDS-gels, these thin gels are excellently suited for this type of blotting because of the short diffusion pathway of only 0.2-0.25 mm [ 161. As previously mentioned [14], the transfer efficiency depends on a weak fluid stream, generated by suction with one layer of filter paper. The gel layer should not be desiccated too fast. If membranes with excessive moisture are used, the liquid stream is reversed and will be directed against the diffusion. The technique established for SDSgels works just as well as with native gels. We recommend the following procedure to build up the gel-membranepaper sandwich. (i) The nitrocellulose membrane should be washed in water for 5-10 min and slightly dried for 5 min. (ii) Immediately after the end of electrophoresis the membrane is placed on the gel without entrapping air bubbles. (iii) A sheet of filter paper (Whatman# 17, 0.92 mm thick) and a glass plate are placed on the membrane. All parts should have the same size. (iv) The stack is weighted by about 20 g/cm2.The bulk of the protein is transfkrred within the first 1-2 h, but the yield can be improved by extending the blotting time. The membrane and the gel should be dried to stabilize the protein-membrane binding before the membrane is carefully taken from the gel; slight moistening may be necessary. The transfer yield can be as high as 70-80%. It is not reduced by Prosolv I1 or Triton X-100 in the gel. The incomplete transfer may also have a positive aspect: silver staining of the remaining proteins in the dimensionally stable gel results in a print of the overall protein pattern, which helps to localize the immunostained bands on the membrane.
[ I ] Gorg,A.,Postel,W.andWestermeier,R.,Anal.Biorhem. 1978,89,6070. [2] Gorg, A., Postel, W. and Westermeier, R., in: Radola,B. J, (Ed.),Electrophoresis '7Y de Gruyter, Berlin 1980, pp. 67-78. [3] Radola, B. L., Electrophoresis 1980, I , 43-56. [4] Gorg, A,. Postel, W. and Westermeier, R., 2. Lebensmirtelunters.ForSch. 1979, 168,25-28. [S] Gorg, A,, Postel, W., Westermeier, R., Gianazza, E. and Righetti, P. G., J. Biochem. Biophys. Methods 1980, 3, 273-284. [6] Gorg, A,, Postel,W.,Weser, J., Boesken,W. and Schiwara, H. W., Sci. TOOIS1985, 32, 5-9. [7] Ansorge, W. and De Maeyer, L., J. Chromarogr. 1980,202, 45-53. IS] Lasky, M., in: Catsimpoolas, N,(Ed.) Elecfrophoresis '78, Elsevier, North Holland Amsterdam 1978, pp. 195-210. [9] Lambin, P. and Fine, J. M., Anal. Biochem. 1979,98, 160-168. [lo] Every, D., Anal. Biochem. 1981, 115, 7-10. [ l l ] Campbell, W. P., Wrigley, C. W. and Margolis, J., Anal. Biochem. 1983, 129,31-36. [I21 PhastSysrein Owner's Manual, Separation Technique Files 121 and 130, Pharmacia, Uppsala, Sweden, 1987. [13] Heukeshoven, J. and Dernick, R., Electrophoresis 1985, 6, 103-112. [14] Heukeshoven,J.and Dernick,R.,in: Radola,B. J.(Ed.), Electrophoresis Forum '89, Technische Universitat, Munchen 1989, pp. 283-292. 1151 Heukeshoven, J. and Dernick, R., in: Radola, B. J. (Ed.) Electrophorese Forum '91, Technische Universitat, Munchen 1991, pp. 415-420. [16] Heukeshoven, 3. and Dernick, R., in: Radola, B. J. (Ed.), Elekrrophorese Forum '86,Technische Universitat, Miinchen 1986, pp. 247-252. [17] Neuhoff, V., Electrophoresis 1984, 5, 251. 1181 Dernick, R. and Heukeshoven, J.. in: Mayr, A., Bachmann, P. A,, Mayr-Bilbrack, B. and Wittmann, G. (Eds.), Virologische Arbeitsmethoden, Gustav Fisher-Verlag, Stuttgart 1989, Vol. 3, pp. 188-347. [19] Lampin, P., Rochu, D. and Fine, J. M., Anal. Biochem. 1976, 74,567575. [20] Lampin, P., Anal. Biochem. 1978, 85, 114-125. [2 11 PhastSysrem Owner's Manual, Separation Technique Files 120, Pharmacia, Uppsala, Sweden, 1987. [22] Ferguson, K. A., Metabolism 1964, 13,985-995. [23] Reisfeld, R. A . , Lewis, J. J. and Williams, E. E., Nature 1962, 4838, 281-283. [24] Bonner, W. M., West, M. H. P. and Stedman, J. D., Eur. J. Biochem. 1980, IOY, 17-23. [25] Heukeshoven, J. and Dernick, R., Electrophoresis 1988, 9, 28-32. [26] Heukeshoven, J. and Dernick, R., in: Radola, B. J. (Ed.), Elektrophorese Forum '86, Technische Universitat, Munchen 1986, pp. 22-27.
J tleukeshoven and R. Dcrnicli
Jochen Heukeshoven Rudolf Dernick Heinrich-Pette-lnstitut fur Experimentelle Virologie und lmmunologie an der Universitiit Hamburg
Native horizontal ultrathin polyacrylamide gel electrophoresis of proteins under basic and acidic conditions The preparation of homogeneous ultrathin native polyacrylamide gels, using a basic as well as an acidic buffer system is described. The basic buffer system consists of Tris-HCl/Tris-glycine, the same buffer as in sodium dodecyl sulfate (SDS)-gel electrophoresis but without SDS. The acidic system uses potassium acetate, pH 4.3, as gel buffer and 0-alanine, pH 4.6, acetic acid as electrolytes. The gels are covalently bound on glass plates. Binding of acidic gels requires a special pretreatment of glass plates. The whole procedure is simple and extraordinarily fast: 100-120 min from the start of gel preparation to the end of electrophoresis. Coomassie staining is done in 40 min and silver staining in 90 min.The native gels are excellently suited for diffusion blotting. Further attractive properties of these gels are easy handling, simple drying and dimensional stability.
result from two parameters, gel thickness and fixation on a support.
1 Introduction Electrophoresis *separationtechniques with a horizontally arranged separation matrix have been known for many years. Paper electrophoresis and electrophoresis on cellulose acetate sheets in clinical diagnostics are the classical methods of horizontal electrophoresis. Later, peptide separations on thin layers of' cellulose or silica gel and several applications of electrophoresis in agarose, including isoelectric focusing (IEF) in agarose, were introduced. With the use of polyacrylamide gels as separation matrix, horizontal I E F in thin layers was developed [l-31. Although gels are mainly used in a vertical configuration at present, some techniques of horizontal sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) were described recently [4-71. Native electrophoresis in acidic or basic media is performed relatively seldom and then exclusively in vertical gels [8-111. Quite recently horizontal native electrophoresis was demonstrated using the PhastSystem, a semi-automatic electrophoresis unit [12]. We have developed a horizontal gel system for SDS-PAGE which uses ultrathin, homogeneous gels (without a stacking gel) bound on glass plates.The preparation of these gels is simple and fast, the separation power is excellent, and staining and drying is faster in comparison to conventional 1 m m slab gels [13, 141. Here we describe the adaption of this gel system to native electrophoresis in both basic and acidic media. Preliminary results were presented previously [ 151. We prcfer horizontal gel systems because they possess a number of advantages in comparison to vertical systems. Horizontal gels are usually thin gels bound to rigid supports [ I , 3,5, 7) and therefore their favorable properties .-
Correspondence: Dr. Jochen Heukeshoven, Heinrich-Pette-Institut fiir ExperiinentelleVirologie und lmmunologie an deruniversitiit Hamburg, Mariinistr. 52. DW-2000 klamburg 20, Germany Abbreviations: IEF, isoclectric focusing; SDS-PAGE, sodium dodecyl sulfate-polyacrylaniide gt.1 electrophoresis; TEMED, N,iV,N', N'-tetramethplethyienedlanlinc. T, total gel concentration = acrylamjde plus crosslinking agent W / C ; 0% C, gcl crosslinking =percent crosslinking agent/ percent total gel concentralion
The specific characteristics of horizontal gels are summarized briefly. Horizontal gels are open systems and consequently it is feasible to use nearly all gel dimensions or parts of a gel in a single apparatus; the geometry and number of sample slots are freely selectable. Fitting problems with gel cassettes and buffer tanks will not arise. Sample application is simplified. The use of thin gels and buffer strips results in a remarkable reduction in the consumption of chemicals (acrylamide, buffers). Gels bound to supports are dimensionally stable during washing, staining, drying, blotting [ 161and, if necessary, rehydration. Especially in the case of gels < 0.3 mm, all processes of electrophoresis, fixation, washing, staining and drying are significantly faster than with 1 m m thick gels. Furthermore, the sharpness of bands and detection sensitivity are improved. In our gel system we use exclusively homogeneous gels, developed in the first instance for SDS-PAGE 1141. This gel type has additional advantages over those with stacking and separation gels. The preparation is extraordinarily simple: we need only one gel solution and after a single pouring step the gel is ready for use. We do not need further equipment, e.g. gradient mixers. The polymerization conditions can be optimally standardized, resulting in a more reproducible preparation of gels. Staining and drying is less critical than with gradient gels.The risk of breaking and nonuniform staining is not a problem. All advantages acquired in our previous work with SDS-PAGE have now been incorporrated into the preparation and use of native gels.
2 Preparation of thin homogeneous gels 2.1 Glass plates versus polyester sheets We have bound the homogeneous thin gels only on glass plates as support because (i) glass is cheaper than pretreated polyester sheets, (ii) it can be used repeatedly, and (iii) it is easier to handle during casting, staining, blotting [16] and, above all, during drying. The anchorage of poly-
N a t i v e polyacrylamide gel electrophorcsi?
EIi~rirophori~.\is1992. 13, 654-659
655
acrylamide gels on glass is normally better (exceptions see Section 4.1) than on polyester. On glass, sharper zones are usually deserved. The gel layer is always 0.25 mm thick. 2.2 Gel preparation The preparation of the casting mold was described earlier [14].A thick glass plate of 4-5 mm is provided with spacers and slot formers and is then hydrophobized with trimethylsilyl chloride. A 1.5-2 mm glass plate is used as support to bind the gel covalently by reaction with the 3-methacryloxypropyl groups of Polyfix 1000 [3,7,17,18].The special activation of glass plates for use in acidic polymerized gels containing detergents will be described in Section 4.1. Gels are cast horizontally using a combination of capillary attraction and sliding [7], as described earlier for SDS gels [14]. The slot plate (slots at the top), laying on a table or a bigger glass plate, is fixed with the aid of a water film for better adherence, with 80% of its length being covered by the gel plate. The mixed and degassed gel solution is pipetted slowly on the open tenth of the slot plate (Fig. I). Entry of the polymerization solution between the two glass plates should be promoted by slightly knocking on the gel plate near the front of the solution. Finally the gel plate is slid over the remaining 20% to cover the slot plate completely. With this technique the polymerization solution can be introduced without air bubbles within 20--30 s. The casting device is weighted by about 1 kg to press the two glass plates together tightly. The basic gel is polymerized within 3045 min at room temperature or after 15 rnin at 50-60°C and is then ready for use. The acidic gels must be kept at 5060°C for 15 min to start and complete the polymerization, otherwise the gel does not polymerize. The procedure is always the same for SDS, basic and acidic native gels; only the polymerization solutions differ. Gel preparation is simple and requires only 30 min from mixing the solutions to the finished polymerized gel, if all stock solutions and the glass plates were prepared before. Rehydration of a dried gel is possible but requires the same or more time than the preparation of a fresh gel.
Figuic I. Horizontal casting procedure of ultrathin polyacrylamide gels. The gel solution is pipetted slowly on the open slot plate (about 20% of total length).
Table 1 Composition ofgel solutions for preparation o f a native standard uolvacrylamide nel 210 X 110 X 0.25 mm Solution Acrylamide, 40% TJ2% C Acrylamide, 40% T/10% C KAc, IM, pH 4.3 Tris/WCl, 2 ~pH , 8.9 Prosolv IT Tetrabutylammonium . H S0 4 , 10% H2O
2.40
1.80 0.60 0.80
mL
4.00
-
-
0.032 0.04 4.61 0.02 0.10 8.00
1.54 0.01 0.05 8.00
Persulfate 10% Total
5
Acidic gel
-
TEMED
0
Basic gel mL
10
15
%T
3 Basic gels
Figure 2. Electrophoresis in four homogeneous ultrathin native polyacrylamide gels in Tris/HCI buffer, pH 8.9, with 2% C and 6, 8, 10 and
3.1 Gel buffer
12% T. Electrophoresis of marker proteins until the dye Bromothymol Blue reached the end of the gel. Silver staining. The log Rf values (mobility related to dye front) were calculated and plotted against %T.
Three buffer system previously used in native gradient gels [ 19-21], were adapted to our homogeneous gels. Neither the Tris-borate-EDTA system nor the Tris-acetate system produced sufficiently sharp protein bands. The best choice, and also the simplest, was the Tris-HCl/Tris-glycine system, already applied by us in ultrathin SDS-PAGE [14]. We employ the same buffer at a high concentration (1 M TrisHCl, pH 8.9) as in SDS-PAGE but without SDS [15]. 3.2 Gel compositon The electrophoresis mobility of protein depends on their size and their charge at a given pH as well as on their retardation by the gel matrix. We recommend starting with an intermediate acrylamide concentration of 10-12 O/o T and 2 % C (Table 1). To optimize the separation and to estimate molecular weights using a Ferguson plot (log mobility vs.
%T) [22], 3-4 additional gels with higher or lower acrylamide contents are necessary. The useful range may be between 5 and 17% T. 3.3 Ferguson plot
Due to the simple and fast preparation ofhomogeneous native gels (4-5 gels with different acrylamide content can be produced and run per day), a graph of a Ferguson plot is easy to draw [22]. Data from four gels with 6,8,10 and 12%T were evaluated. The relative mobility, R, (related to the dye front) of some marker proteins was calculated and their log values were plotted against %T (Fig. 2).We obtained straight lines, from which we constructed a calibration
656
J Heukeahoven m d R Dernick
E l r o r o p h o r a i s 1992, J3, 654-689
curve to estimate the molecular weights. Only a few of the obtained values, mainly from aggregated proteins, do not fit in the linear (calibration curve (Fig. 3). 3.4 Electrophoresis buffer and paper buffer strips The electrophoresis buffer (0.4 M Tris - 0.1 M glycine, pH 8.6), the same as in SDS-PAGE but without SDS, is identical for cathode and anode. As electrode bridges we use filter paper strips, 3-4 cm wide and 20-21 cm long. They should be able to take up 15-20 m L of electrophoresis buffer. We use 7-10 layers of 4 cm wide filter paperfrom Schlei-
'50
I----
chert&Schiill (AMGL2043). This paper is sufficiently clean so as not to impair silver staining. The quality of the filter paper is not as critical in native electrophoresis as it is in SDS-PAGE, and therefore it is not necessary to use polyacrylamide gel strips. The paper strips overlap the gel about 5 m m and their full length is tightly pressed against the gel layer by a glass plate. The platinum electrodes are placed on the strips at the greatest distance from each other.
3.5 Electrophoretic conditions Electrophoresis is performed for 75-90 min at limited settings of 400 V,,,,, 12 W,,,, and 20 mA,,,, and the temperature of the cooling plate is set to 12"C.These values are optimal for our standard gels (12% T; 210 mrn wide, 110 m m long and 0.25 m m thick). For gels with a higher acrylamide concentration the power should be slightly reduced, to 10 Wrn',,.
BSA
Y 0
I 50
150
100
TIM OV The CA TI MVO *La
200
Mot -Gew in kD
f i g u r e 3. Fcrguson p l o ~based on data obtained from Fig. 2 and a similar set of log R, - '%T.The slope of the curves was estimated and plotted against molecular weight. Abbreviations: Pho. phosphorylase: BSA, bovine serum albumin; TIM, triosephosphate isomerase; OV, ovalbumin; The, thermolysin; CA, carbonic anhydrase; TI, trypsin inhibitor; Myo, myoglobin horse; a-La, a-lactalbumin; P-Gal. 8-galactosidase; ALD, aldolase;(XX),. aggrcgates of protein XX.
3.6 Sample preparation Extracts or protein samples should be dissolved in Tris/ HC1 buffer, pH 8.5-9.0; small amounts (0.1-0.2 O/o) of mild detergents like Triton or Tween may be added. The buffer concentration is not critical and up to 0.5 M will be tolerated, because the applied sample volume is only 3-4 pL and the gel buffer also has a high molarity. The amount of protein in this volume should be in the range of 10-100 ng/ band (3-30 p/mL) if the gel is silver stained, or 100-1000 ng/band when utilizing an effective Coomassie staining. As tracking dye we recommend Bromothymol Blue. Figure 4 shows separations of a mixture of marker proteins, selected single marker proteins and extracts of bovine and fish muscle.
Flgure 4. Native polyacrylamide gel electrophoresis in ultrathin, homogeneous gel with 12S0/uT and 2%C under basic conditions using 1 M Tris/HCI gel buffer, pH 8.9, and 0.1 M Trid0.4 M glycine electrode buffer, pH 8.6. Original size: 210 X 110 X 0.25 mm; electrophoresis condi-
tions: 400 Vn,:,,> 20 mA,,,,, and 12W,,,
for 90 min. Silver staining.
Elccrrophoreris 1992, 13, 654-659
4 Acidic gels 4.1 Pretreatment of glass plates and gel buffer As mentioned in Section 2.2, acrylamide can be polymerized at acidic pH value, 4.3, initiated by N,N,N',N'-tetraniethylethylenediamine (TEMED)-persulfate, if the concentration of persulfate is doubled and the polymerization is performed at higher temperature (50-60°C). Gels prepared under these conditions adhere well to 3-methacryloxypropyltrimethoxy silane pretreated glass plates. None of the buffer systems tested produce sufficiently sharp bands, the modified potassium acetate@-alanin system according to Reisfeld et al. [23] being most promising. Zones with extensive tailing indicated an interaction between glass plate and proteins. Therefore, low concentrations (0.1-0.2% Triton X-100, Nonidet P-40,Tween) of nonionic detergents were added to overcome the adsorption. Indeed, the detergent had the desired effect, but in the presence of neutral detergents the gel is no longer bound covalently to the glass plate, although glass plates pretreated in the same way are able to bind gels prepared without neutral detergents. The detergent-containing gel detached at the latest during the fixation or staining step. The protein band in such gels were relatively sharp.
Native polyacrylamide gel
deCtIODhoreSlS
657
The gel buffer was an adaption of the buffer system of Reisfeld et al. [23].The pH was varied between 4 and 5, the concentration between 0.05-0.4 M potassium acetate. The buffer concentration could not be as high as in basic native gels. The best results were obtained with 0. I M potassium acetate, pH 4.3. Sodium acetate and ammonium acetate, according to Bonner et al. 1241,were also tested but the separations were not as good as with potassium acetate. Prosolv I1 (0.1 Yo from Elektrophorese-Technik, Leonberg, Germany) was the best choice as nonionic detergent, with cleaner separations than with Triton X-100. A small amount (0.05%) of tetrabutylammonium hydrogensulfate promotes a more uniform dispersion of the proteins in the sample slots.
4.2 Gel composition
Acidic gels have larger pores than basic gels at the same acrylamide concentrations.This could be shown with a washed and dried basic gel, rehydrated in an acidic buffer. The electrophoretic mobility of the proteins was remarkably reduced in such a rehydrated geLTherefore, using acidic polymerization, we recommend a gel composition of 12O/o T and 4% C as standard gel (Table 1). Surprisingly, the migration properties do not change with a variation of gel concentration. Anumber of proteins were run in acidic gels with 8,10, For improved gel binding, the glass plates were intensively 12, 14, 16, and 18% T, and their migration distances were washed with 20% potassium hydroxide in 50% ethanol, compared. The same patterns were observed. With increaswith 10% chromosulfuric acid, and with water. These pre- ing gel concentration the protein bands were sharper, but treated plates were reacted with 10% 3-methacryloxypro- gels with a high acrylamide content tend to dry out during pyltrimethoxy silane in chloroform or dimethylformamide electrophoresis. at 60°C. However, none of these preventive measures improved gel binding in the presence of neutral detergents. 4.3 Electrophoresis buffer and paper buffer strips It was obvious that the number of reactive sites on the glass surface was too low to firmlybind the gel. We can onlyspec- Reisfeld et al. [23] used (3-alanine acetate, pH 4-5, while in d a t e that the acryl-silane residues on the surface were their acidic system, Bonner et al. [24] employed glycine acepartly hidden by a layer of detergent molecules, preventing tate, approximately pH 3, as running buffer. In our experithe reaction with acrylamide during polymerization. To ments the amino acids glycine, alanine and 6-alanine were overcome this problem we tried to enlarge the reactive sur- chosen as cationic buffer substances because they are relatiface by a mechanical or chemical procedure. The simplest vely inexpensive.The buffer capacity of glycine and alanine and most efficient method is to frost the glass surface with at pH > 4 is low and therefore, with them, an acetate buffer waterproof emery paper of finest grain size (400-600) for could only be used at pH < 4. Only P-alanine can be used 0.5-1 min. A more homogeneous buffing can be achieved with an acetate buffer in the pH range 4-5. The results of a by chemical corroding with 10Vo hydrofluoric acid or a mix- few electrophoretic separations demonstrated that the moture of 10% KF/lO% H,SO, for 20 h at room temperature. bility of a number of marker proteins depends only slightly Because of the risk in working with these chemicals we pre- on the pH in the range ofpH 4.3-4.8.To obtain an adequate fer the simpler emery paper treatment, which makes the buffer capacity, moderate conductivity, and also a tolerable glass surface highlyreactive,so that the binding of gels after electrophoresis time, we have to adjust both the pH and activation with 3-methacryloxypropyltrimethoxy silane is buffer concentration accordingly. The acceptable concenalso adequate in the presence of detergents. Not all rsilanol tration range lies between 0.1-0.3 M 0-alanine at a pH of groups react with the acryl silane, resulting in a residual ad- 4.3-4.8. The optimum anode buffer contains 0.2 M (3-alasorption partially inhibiting the formation of sharp protein nine, adjusted with acetic acid to pH 4.6. As cathode solubands, especially in work with basic proteins. An additional tion we used 1'Yo acetic acid. As with basic gels, the same coating with trimethylsilyl chloride was essential, as in the kind of buffer strips (7-10 layers, 4 cm wide filter paper endcapping procedure known from chromatographic silica strip) are applied and soaked with 15-20 mL of the elecmatrices. Summarizing, the pretreatment of glass plates trode solutions. suitable for binding acidic gel prepared with neutral detergent requires the following steps: (i) cleaning in alkaline solution (10% NaOH, also suitable for removing gels, over- 4.4 Electrophoretic conditions night), (ii) washing in acid (5% sulfuric acid, minimum 1 h), Electrophoresis of acidic gels is performed under similar (iii) frosting with wet emery paper (0.5-1 min), (iv) coating with 3-methacryloxypropyltrimethoxy-silane, 2% chloro- conditions as with thin homogeneous SDS-gels. The standard gels were run for 60-80 min at limited values of 600 form, (v) second coating with trimethylsilyl chloride, 5 O/o in chloroform, and (vi) rinsing with ethanol and drying. V, 15 mA,,, and 30 W,,, at a temperature of 12°C.Power
658
J H e u k a h o v e n a n d R. Dei-nich
and current can be adjusted to highervalues than with basic gels because the electrosmotic transport of water is reduced under acidic conditions.
Electroiihor?~t,\1992. 13. 654-630
background is unnecessary. The gel i s rinsed shortly with water and dried in air. 5.2 Silver staining
4.5 Sample preparation Proteins must be soluble in acidic buffer at pH 4-5. Small amounts of mild tletergents may improve the solubility.We obtained the best results with 0.1 M P-alanine-acetate buffer, pH 5.0, containing 0.2 %Tween 20. Methyl Green or Malachite Green are suitable as tracking dye. The amount of protein which is applicable to the gel is in the same range as in basic gels. Figure 5 shows a separation of marker proteins, some individual proteins und some extracts in an acidic native polyacrylamide gel.
5 Staining Coomassie and silver staining methods described for ultrathin SDS-gels [141 were worked out for the staining of native gels, whereby it was not necessary to differentiate between basic and acidic gels.
5.1 Coomassie staining Fast staining is accomplished in two steps: (i) fixation with 10% trichloroacetic acid for 10 min; (ii) the gels are stained with 0.002% Serva Blue R in 10% acetic acid. The staining time is usually 1-2 h at room temperature,but it can be considerably accelerated by heating the solution to 50-60°C. Agitation ofthe gel for 10-20 rnin with a hot solution is adequate. At the low dye concentration, destaining to clear the
Silver staining is performed according to our method as published previously for the PhastSystem 1251and ultrathin SDS-gels [14,26]. After 10 rnin fixation in 10% trichloroacetic acid, a second fixation step is applied for 30-60 min in a solution containing 0.5% glutaraldehyde, 0.25 O h sodium thiosulfate and 0.4 M sodium acetate in 30% ethanol. The gel is thoroughly washed with water before it is incubated in 0.2% silver nitrate for 20 min. Finally, the usual proceTable 2 Staining protocol for ultrathin basic or acidic native polyacrylamide gels Coomassie staining 1 Fixation in 10% trichloroacetic acid at room temperature 2 Staining with 0.002% Serva Blue R at 50-60°C in 100/0 acetic acid 3 Destaining not necessary Total
Silver staining 1 Fixation i n 10% trichloroacetic acid at room temperature 2 Fixation in 30% ethanol with 0.4 M / L sodium acetate 0.25% Na2S203,0.5"/0 glutaraldehyde 3 Wash with water, 3 X 5 min 4 Silver nitrate, 0.2% 5 Development in 2.5% Na2C03, pH 11.7 and 0.01 Vo formaldehyde 6 Stop bath, 1% glycine 7 Wash with water, 2 X 3 rnin Total
10 min 30 min 0 min 40 rnin
10 min
30 rnin 15 min 20 min
3-5 min 1-2 min 6 min 87-90 min
Figure 5. Native polyacrylamide gel electrophoresis in ultrathin, homogeneous gel with 12%T and 4%0Cunder acidic conditions using 0.1 M potassium acetate gel buffer, pH 4.3, with additives given in Table 1, and 0.2 M 8-alanine/acetate, pH 4.6, as electrode buffer. Original size: 210 X 110 X 0.25 mm; electrophoresis conditions: 600V,,,, 15 mA,,, and 30 W,,, for 75 min. Samples dissolved in 0.1 M 8-alanine buffer, pl-1 5 , with 0.2% Tween to decrease the absorption of proteins in the sample vial. Silver staining.
Elccrrophotrsis 1992, 13, 654-659
dure of development with 2.5 010 sodium carbonate/O.Ol O/o formaldehyde follows. The staining protocols are given in Table 2. The indicated times are minimum values and can be extended (except for the fixation with trichloroacetic acid and development with carbonate solution).
Native polyacrylamide gel electrophoresis
659
The Heinrich-PtJtte-lnstitueis financially supported by Freie und Hansestudt Hamburg and by Bundesministerium f u r Gesundheit, Bonn. The authors thank Anja Kirschtier for excellent technical assistance. Received August 11,1992
6 Blotting
7 References
The transfer from polyacrylamide gels to nitrocellulose or Immobilon membranes is an important technique, especially for proteins separated in the native state. For immunological studies the conservation of native structures is often essential. Using ultrathin gels bound to a support, the transfer can only be achieved by diffusion b1otting.A~demonstrated previously with homogeneous SDS-gels, these thin gels are excellently suited for this type of blotting because of the short diffusion pathway of only 0.2-0.25 mm [ 161. As previously mentioned [14], the transfer efficiency depends on a weak fluid stream, generated by suction with one layer of filter paper. The gel layer should not be desiccated too fast. If membranes with excessive moisture are used, the liquid stream is reversed and will be directed against the diffusion. The technique established for SDSgels works just as well as with native gels. We recommend the following procedure to build up the gel-membranepaper sandwich. (i) The nitrocellulose membrane should be washed in water for 5-10 min and slightly dried for 5 min. (ii) Immediately after the end of electrophoresis the membrane is placed on the gel without entrapping air bubbles. (iii) A sheet of filter paper (Whatman# 17, 0.92 mm thick) and a glass plate are placed on the membrane. All parts should have the same size. (iv) The stack is weighted by about 20 g/cm2.The bulk of the protein is transfkrred within the first 1-2 h, but the yield can be improved by extending the blotting time. The membrane and the gel should be dried to stabilize the protein-membrane binding before the membrane is carefully taken from the gel; slight moistening may be necessary. The transfer yield can be as high as 70-80%. It is not reduced by Prosolv I1 or Triton X-100 in the gel. The incomplete transfer may also have a positive aspect: silver staining of the remaining proteins in the dimensionally stable gel results in a print of the overall protein pattern, which helps to localize the immunostained bands on the membrane.
[ I ] Gorg,A.,Postel,W.andWestermeier,R.,Anal.Biorhem. 1978,89,6070. [2] Gorg, A., Postel, W. and Westermeier, R., in: Radola,B. J, (Ed.),Electrophoresis '7Y de Gruyter, Berlin 1980, pp. 67-78. [3] Radola, B. L., Electrophoresis 1980, I , 43-56. [4] Gorg, A,. Postel, W. and Westermeier, R., 2. Lebensmirtelunters.ForSch. 1979, 168,25-28. [S] Gorg, A,, Postel, W., Westermeier, R., Gianazza, E. and Righetti, P. G., J. Biochem. Biophys. Methods 1980, 3, 273-284. [6] Gorg, A,, Postel,W.,Weser, J., Boesken,W. and Schiwara, H. W., Sci. TOOIS1985, 32, 5-9. [7] Ansorge, W. and De Maeyer, L., J. Chromarogr. 1980,202, 45-53. IS] Lasky, M., in: Catsimpoolas, N,(Ed.) Elecfrophoresis '78, Elsevier, North Holland Amsterdam 1978, pp. 195-210. [9] Lambin, P. and Fine, J. M., Anal. Biochem. 1979,98, 160-168. [lo] Every, D., Anal. Biochem. 1981, 115, 7-10. [ l l ] Campbell, W. P., Wrigley, C. W. and Margolis, J., Anal. Biochem. 1983, 129,31-36. [I21 PhastSysrein Owner's Manual, Separation Technique Files 121 and 130, Pharmacia, Uppsala, Sweden, 1987. [13] Heukeshoven, J. and Dernick, R., Electrophoresis 1985, 6, 103-112. [14] Heukeshoven,J.and Dernick,R.,in: Radola,B. J.(Ed.), Electrophoresis Forum '89, Technische Universitat, Munchen 1989, pp. 283-292. 1151 Heukeshoven, J. and Dernick, R., in: Radola, B. J. (Ed.) Electrophorese Forum '91, Technische Universitat, Munchen 1991, pp. 415-420. [16] Heukeshoven, 3. and Dernick, R., in: Radola, B. J. (Ed.), Elekrrophorese Forum '86,Technische Universitat, Miinchen 1986, pp. 247-252. [17] Neuhoff, V., Electrophoresis 1984, 5, 251. 1181 Dernick, R. and Heukeshoven, J.. in: Mayr, A., Bachmann, P. A,, Mayr-Bilbrack, B. and Wittmann, G. (Eds.), Virologische Arbeitsmethoden, Gustav Fisher-Verlag, Stuttgart 1989, Vol. 3, pp. 188-347. [19] Lampin, P., Rochu, D. and Fine, J. M., Anal. Biochem. 1976, 74,567575. [20] Lampin, P., Anal. Biochem. 1978, 85, 114-125. [2 11 PhastSysrem Owner's Manual, Separation Technique Files 120, Pharmacia, Uppsala, Sweden, 1987. [22] Ferguson, K. A., Metabolism 1964, 13,985-995. [23] Reisfeld, R. A . , Lewis, J. J. and Williams, E. E., Nature 1962, 4838, 281-283. [24] Bonner, W. M., West, M. H. P. and Stedman, J. D., Eur. J. Biochem. 1980, IOY, 17-23. [25] Heukeshoven, J. and Dernick, R., Electrophoresis 1988, 9, 28-32. [26] Heukeshoven, J. and Dernick, R., in: Radola, B. J. (Ed.), Elektrophorese Forum '86, Technische Universitat, Munchen 1986, pp. 22-27.
