We describe here a fluorometric method of detection of proteins fractionated by electrophoresis in poly- acrylamide-SDS gels. This method, using ethidium.
Molec. BioLRep. Vol. 5, 4: 209-214, 1979
A RAPID AND SENSITIVE METHOD FOR DETECTION OF PROTEINS IN POLYACRYLAMIDE SDS GELS: STAINING WITH ETHIDIUM BROMIDE
A. VINCENT & K. SCHERRER
Laboratoire de Biochimie de la Diff~renciation, Institut de Recherche en Biologie Mol~culaire du C.N.R.S., Paris, France (Received April 15, 1979)
Abstract
Materials and methods
We describe here a fluorometric method of detection of proteins fractionated by electrophoresis in polyacrylamide-SDS gels. This method, using ethidium bromide as fluorescent dye, is performed within 40 minutes after the end of the electrophoretic run. It does not require treatment of proteins prior to electrophoresis, and entails neither fixation of proteins in the gel, nor destaining. It is sufficiently sensitive to detect 0.5 - 1.0 gg of protein per band. Furthermore, the ~nultaneous electrophoretic resolution and detection of protein and RNA on a single SDS-polyacrylamide gradient gel is reported.
Chemicals Ethidium bromide (2,7 diamino-10 ethyl-9 phenyl phenantidrium bromide) was purchased from Sigma Chemical Co. (St-Louis, USA). Acrylamide and N,N'Methylen -bis- acrylamide were obtained from Fluka and recrystallized according to Loening (5). All other chemicals used were reagent grade. Before use, buffers were f'dtered through 0.22/a Millipore nitrocellulose membrane f'dters.
Introduction
SDS sample buffer contained 0.08 M Tris-HC1, pH 6.8, 2% SDS, 2% /~-mercaptoethanol; 10% sacharose and bromophenol blue as tracing dye.
The characterization of polypeptides by electrophoresis in supporting gels of agarose, starch, and particularly of polyacrylamide is one of the most widely applied analytical techniques in biochemistry. Traditionally, the developed .gels are stained with amido black (1), Fast green FCF (2) or Coomassie Brilliant Blue (3, 4). Yet, these staining procedures are rather time consuming since they involve destaining of the supporting gel and have the further drawback that, in general, they lead to irreversible denaturation and fixation of the proteins in the gel. It is, however, often necessary to recover a polypeptide from the gel to pursue its further characterization. We report here a staining method which is both rapid and allows recovery of unfixed polypeptides from SDS gels. Correspondence to: Klaus Scherrer, Institut de Recherche en Biologie Mol6culaire, Universit6 de Paris VII, 2, Place Jussieu (Tour 43), 75221 ParisCedex 05, France
Solutions
Preparation of RNA and protein samples Ribosomal RNA was phenol extracted from duck erythroblast polyribosomes and subsequently separated on SDS sucrose gradients (6). 9 S globin messenger RNA was prepared using the same procedure as for ribosomal RNA and was further purified by aff'mity chromatography on oligo-dT cellulose columns. Commercial 4 S yeast tRNA was from Sigma Chemical Co. (USA). EDTA dissociated ribosomal subunits were prepared from duck erythroblasts according to published methods (7). Proteins were recovered from the phenolic phase by precipitation with nine volumes of icecold acetone after phenol extraction of purified ribosomal subunits. Protein standards were: BSA (68,000 Mr), ovalbumin (45,000 Mr), aldolase (39,000 Mr) Chymotrypsinogen (25,000 Mr) and Cytochrome C (12,500 Mr), from Boehringer (Marmheim). 209
Polyacrflamide-SDS gels Exponential gradient (4% to 15%) polyacrylamide (acrylamide]bis-acrylamide: 39/1) 1.5 n u n thick slab gels, prepared as described in detail elsewhere (8), were ovedayered by a 4% polyacrylamide stacking gel (9). Two-dimensional electrophoresis was according to O'Farrell (13).
Electrophoresis Just prior tO electrophoresis, RNA and protein samples were incubated for 3 minutes at 65 ~ C in SDS sample buffer. Electrophoresis was conducted at 18~ for 6 hours. The current was kept at 20 mA while the voltage increased from initial 100 V to 220 V at the end of the run.
Staining with ethidium bromide Following electrophoresis, the gels were soaked under gentle agitation in 20 volumes of bidistilled water for 10 minutes at room temperature, rinsed once with water and incubated for 20 min. in 200 ml of an ethidium bromide solution at 1/zg/ml.
Photography For photography, the gels were placed on a black support and illuminated with two short wave lamps (CS 215 U.V. Products, San Gabriel, California) (10). Black and white photography was with polaroid type 665 positive f'dm using a red falter (Wratten 24). Exposure time was for 40 see. for detection of nucleic acids and 1.5 min for detection of proteins.
Staining with Coomassie Brilliant Blue If necessary, after ethidium bromide treatment, the gels were stained overnight with 0.05% Coomassie Brilliant Blue G 250 (Serva) in 12.5% TCA and destained by repeated washing in 7.5% acetic acid.
mination (1, 3). The main effect of SDS is to give an uniform charge density to polypeptides prealably incubated with a reducing agent to disrupt disulfide bridges. In our investigations of messenger ribonucleoprotein complexes (11, 12), we were led to develop an convenient method for the simultaneous analysis of ribonudeic acid and proteins present in these complexes. Exponential gradient polyacrylamide gels were adopted in the course of this investigation, thus allowing estimation of the MW of proteins (13) and RNA (14, 8) over a much wider range than possible on constant concentration gels. We show here that electrophoresis of a mixture of RNA and proteins on 4% to 15% polyacrylamide SDS gels allows the simultaneous resolution of proteins with MW greater than 10,000 D (fig. 1A) and ribonucleic acid in the MW range of 104 to 106 D(fig. 1Band 1C). Figure 2 shows a comparison of migration of proteins and RNA present in small EDTA derived ribosomal subunits before and after phenol extraction, and in the presence or absence of a large amount of added exogeneous RNA (slots 3-5, fig. 2A, and fig. 2C). Since the patterns obtained are identical, we may conclude that the presence of both RNA and RNA binding proteins (ribosomal proteins) in a sample subjected to electrophoresis in these conditions does not influence the migration of either type of molecule. This result eliminates the possible occurrence of artifactual bands due to the formation of RNA-protein complexes during electrophoresis under our experimental conditions. This conclusion is in agreement with results recently obtained on the detection of nucleic acid binding proteins and nucleases in polyacrylamide SDS gels (15, 16). Indeed, in the experiments cited above, SDS had to be removed from the gels in order to allow a protein to recover its nucleic acid binding capacity, or its enzymatic (e.g. nuclease) activity. Although simultaneous analysis of RNA and proteins is possible, it should be noted that, under the conditions used, no correlation exists between the respective apparent molecular weights of RNA and proteins (fig. 1C).
Results and discussion
Detection of proteins with ethidium bromide Simultaneous electrophoretic analysis of RNA and proteins in polyacrylamide SDS gels SDS polyacrylamide gel electrophoresis, first introduced by Shapiro et al, to analyse proteins has proven to be a very powerful method for MW deter210
In the course of our investigations, we used ethidium bromide to stain RNA fractionated by. electrophoresis on polyacrylamide SDS gels. Ethidium bromide is generally used to stain RNA or DNA bands in polyacrylamide or agarose gel electrophoresis (17, 18).
Fig. 1. Electrophoresis of proteins and RNA on exponential 4% to 15% polyerylamide SDS gels. (A) Protein standards (1 /~g of each protein): bovine serum albumin (BSA), 68,000 Mr, Ovalbumin 45,000 Mr, Aldolase 39,000 Mr, Chymotrypsinogen A (CTG A) 25,000 Mr, Cytoehrome C (Cyt C) 12,500 Mr. Coomassie Blue staining. (B) RNA: (1) 18 S ribosomal RNA (2 zg) (2) 5 S ribosomal RNA (2 ~g) (3) 9 S globin messenger RNA (2 ~tg) (4) 4 S yeast tRNA (2 tLg) Ethidium Bromide staining. (C) Molecular weight calibration curves for proteins and RNA based on migration shown in (A) and (B) respectively. Molecular weight for RNA was taken from ref. 14.
According to our observations, protein contamination represents a source of error in the detection of nucleic acids in gels containing SDS. Indeed, to our surprise, we observed on overexposed films of ethidium bromide stained gels, that protein bands could be detected as well as RNA (fig. 2B compared with fig. 2A and 2C). To improve the detection of protein, the background of fluorescence emitted by the supporting gel itself and obscuring the fluorescence of protein bands was abolished by merely washing the gels with water before staining. Ethidium bromide staining allows the detection of less than 1 /ag of protein per band (fig. 2B and 3A). To verify that this staining procedure was
not restricted to some particular polypeptides, we further used a second dye, the sensitive and protein specific Coomassie Brilliant Blue (4). The protein patterns obtained with either ethidium bromide or Coomassie Brilliant Blue (fig. 3A compared to 3B), appeared to be identical. Hence, we may conclude that ethidium bromide staining of proteins is not a selective process and may be applied generally for the detection ofprotein-SDS complexes. This method, not requiring acid fixation of the proteins in the gel, allows detection of about 0.5/gg of protein per band. Different procedures are currently used for the detection of protein in polyacrylamide SDS gels. Several dyes (1-4) are used as well as a phosphor211
Fig. 2. Staining of proteins with ethidium bromide following electrophoresis on exp. 5% to 16% polyacrylamide SDS cels. (1) total polyribosomal RNA (4 t~g) (2) 18 S ribosomal RNA (2 ttg) (3) 30 S EDTA derived ribosomal subunits (15 ~g) (4) same as (3) and addition of yeast tRNA (30 ~g) (5) 30 S n'bosomal subunits proteins (10 t~g) (6, 7) protein standards (same as in fig. 1A); (6) 1.5 ag, (7) 0.75 ~g of each protein (A, B) staining with ethidium bromide. Film exposure: (A) 40 see., (B) 1.5 min., (C) staining with Coomassie Brilliant Blue.
escence (19), and fluorescence (20). This latter procedure, using OPT (O-phthalaldehyde) as the fluorescent marker, although very sensitive, has the disadvantage that it necessitates the covalent linkage of the reagent, and hence, an irreversible modification of polypeptides. The simple and efficient method proposed here is based on the fluorescence emitted under UV illumination by the ethidium bromide bound to the protein-SDS complex in the electrophoretic bands. The reason for the fluorescence of the protein-SDS-ethidium bromide complex in the gels is not known. Addition of 0.1% SDS to an ethidium bromide solution resulted in decreased fluorescence emitted by ethidium bromide. Moreover, in control experiments (data not shown), detection by ethidium bromide of 10 tag 212
of protein fractionated by gel electrophoresis with the same buffer system but lacking SDS was totally unsuccessful. We may therefore conclude tentatively that the complexing of protein with SDS is necessary to enhance ethidium bromide fluorescence. It might be interesting to follow further experimentally and theoretically this phenomenon. The main objective of this communication resides, however, in the practical applications of this observation.
Conclusion The principal interest of the staining procedure reported here resides in its quickness and its sire-
plicity, without loosing much sensitivity in comparison v~ith more widely used procedures. Although non-fluorescent dyes such as Coomassie Brilliant Blue, provide very sensitive means for the detection of electrophoretically separated proteins, it requires acid fixation o f the proteins in the gel and timeconsuming staining and destaining. The ethidium bromide staining o f proteins, performed within 40 min., is particularly suitable to locate proteins that must be recovered from the gel to be further analyzed in reconstitution experiments, in studies of enzymatic activity, or used as antigens in immunological experiments. Fig. 3 shows our method to be very useful to detect proteins resolved by bidimension_al gel electrophoresis (13). Indeed, the identification o f a protein band for further processing, easy to achieve on one-dimensional gels by the
staining of an adjacent lane, is problematic in the case of two-dimensional gels which are never exactly superposable. The ethidium bromide technique allows processing of the stained spot itself since dye and SDS can easily be eliminated and the unaltered polypeptide recovered. Furthermore, since ethidium bromide staining o f proteins does not prevent subsequent staining with Comassie Brilliant Blue (fig. 2), the combination of these two procedures, as described he~e, allows the simultaneous electrophoretic resolution on a unique polyacrylamide SDS gel, of RNA or DNA-protein complexes.
Acknowledgements We would like to thank Drs. K. Maundrell for critical reading of the manuscript, J.F. Buri and S. Goldenberg for their interest in this work, and Mrs O. Champion and Mr R. Schwartzmann for the photographs. This work was supported by the French C.N.R.S. (contrat ATP No 13557) and the D616gation G~n~rale ~ la Recherche Scientifique et Technique (contrat no 7670710). A.V. was supported by a fellowship from the Fondation pour la Recherche M6dicale Fran~aise.
References
Fig. 3. Analysis of proteins in two dimensions according to O'FarreU (13). Separation in the first dimension is by isoelectric focusing and in the second dimension by SDS electrophoresis in 13% polyaerylamide gels. (A, B) - Analysis of proteins from "19 S" particles isolated from the post polysomal supernatant of duck erythroblasts (12). (A) staining with ethidium bromide. Film exposure : 1 rain 30 see. (B) staining with Coomassie Brilliant Blue. Molecular weight standards were BSA, ovalbumin, aldolase and ehymotripsinogen (not shown).
1. Shapiro, A.L., Vinuela, E. and Maizel, J.V., Biochem. Biophys. Res. Commun. 28, 815 (1967). 2. Gorovsky, K., Carlson, K. and Rosenbaum, J.L., Anal. Biochem. 35,359-370 (1970). 3. Weber, K. and Osborn, M., J. Biol. Chem. 244, 44064412 (1969). 4. Diezel, W., Kopperschlager, G. and Hofmann, E., Anal. Biochem. 48, 617-620 (1972). 5. Loening, U.E., Biochem. J. 102, 251-257 (1967). 6. Scherrer, K. in Fundamental methods in Virology, Eds. Habel, K. and Salzmann, N.P., (Academic Press, New York), 413-422 (1969). 7. Gander, E.S., Luppis, B., Stewart, A.G. and Scherrer, K., Eur. J. Biochem. 29,369-376 (1972). 8. Mirault, M.E. and Scherrer, K. and Hansen, L., Eur. J. Biochem. 23,372-386 (1971). 9. Laemmli, U.K., Nature (London) 227,680-686 (1970). 10. Prunell, A., Strauss, F. and Leblanc, B., Anal. Biochem. 78, 57-65 (1977). 11. Vincent, A., Civelli, O.C., Buri, J.F. and Scherrer, K. FEBS Lett. 77,281-286 (1977). 12. Maundrell, K.G., Maxwell, S., Civelli, O.C., Vincent, A., Goldenberg, S., Buri, J.F., Imalzumi-Scherrer, M.T., and Scherrer, K., Mol. Biol. Reports, 5, 43-51 (1979). 13. O'FarreU, P.H., J. Biol. Chem. 250, 4007-4021 (1975). 14. Spohr, G., Mirault, M.E., Imaizumi, M.T. and Scherrer, K. Eur. J. Biochem.62, 313-322 (1976). 213
15. Huet, J., Sentenac, A. and Fromageot, P., FEBS Letters 94, 28-32 (1978). 16. Rosenthal, A.L., and Lacks, S.A.J. Biol. Chem. 253, 8674-8676 (1978). 17. Burgoyne, L.A., Hewish, D.R., and Mobbs, J., Biochem. J. 143, 67-72 (1974). 18. Sharp, P.A., Sugden, B. and Sambrook, J. Biochemistry 12, 3055-3063 (1973). 19. Mardian, J.K.W. and lsenberg, I., Anal. Biochem. 91, 1-12 (1978). 20. Weidekamm, E., Wallach, D.F.H. and Fliickiger, R., Anal. Biochem. 54, 102-114 (1973).
214
