[36]
GEL-STAININGTECHNIQUES
477
tein mixture, water, and ampholytes. However, pI precipitation may require that 3 M urea be included for solubility. When higher urea concentrations are needed, the Rotofor cell is run at 12°. Detergents (1-2%, w/v) may also be added to samples. Zwitterionic detergents such as CHAPS, CHAPSO, and nonionic octylglucoside are satisfactory. Triton X-100 and NP-40 may be less satisfactory due to their slight charge content.
Removal of Ampholytes from Proteins There are a number of ways to separate ampholytes from proteins.l-4 Electrophoresis, ammonium sulfate precipitation, and gel filtration, ionexchange, and hydroxylapatite chromatographies have all been used. Dialysis is a simple and effective method for removing ampholytes from solutions of proteins. First, adjust the pooled fractions to 1 M NaCi to disrupt weak electrostatic complexes between ampholytes and proteins, then dialyze the solutions into appropriate buffers. Extensive dialysis is required for thorough removal of ampholytes. There is no good way to demonstrate complete absence of ampholytes in a protein solution, but for many applications they need not be removed.
[36] G e l - S t a i n i n g T e c h n i q u e s
By CARL R. MERRIL Protein Stains Naturally colored proteins such as myoglobin, hemoglobin, ferritin, and cytochrome ¢ may be directly observed in gels illuminated with light in the visual spectrum, providing that their chromophores are not damaged during electrophoresis. 1However, the visualization of most proteins requires the use of dyes or stains. Organic stains were first utilized for the detection of proteins on gels. Recently metal-based stains, such as the silver stains, have achieved widespread use because of their increased sensitivity. A number of organic stains have been adapted for the detection of electrophoretically separated proteins, including Bromphenol Blue, 2 Fast Green (Food Green 3) and Amido Black (Acid Black 1). 3 Some of these 1 B. D, Davis and E. J. Cohn, Ann. N.Y. Acad. Sci. 39, 209 (1939). 2 E. L, Durrum, J. Am. Chem. Soc. 72, 2943 (1950). W. Grassman and K. Hannig, Z. Physiol. Chem. 290, 1 (1952).
METHODS IN ENZYMOLOGY, VOL. 182
Copyright © 1990 by Academic Press, Inc. All fights of reproduction in any form reserved.
478
PURIFICATION PROCEDURES: ELECTROPHORETIC METHODS
[36]
stains preferentially stain certain classes of proteins: Lipoproteins may be stained by Oil Red O, 4 while glycoproteins can be detected by a red color that is produced by their oxidation with periodic acid and subsequent reaction with fuchsin sulfurous acid (Schiff's reagent)/ Of the organic stains, Coomassie Blue has proved to be one of the most sensitive. Proteins may also be detected by the use of fluorescent stains. These stains can detect proteins in the nanogram range. 6 However, fluorescent stains usually require reaction conditions that are best provided prior to electrophoresis. Furthermore, they may alter the charge of the protein. 7 However, such charge alteration do not generally present problems for electrophoretic techniques that separate proteins on the basis of molecular weight, such as with sodium dodecyl sulfate (SDS) electrophoresis. Silver staining currently offers the highest sensitivity. These stains generally provide more than a 100-fold increase in sensitivity over that attained by the most commonly used organic protein stain, Coomassie Blue.8, 9 Coomassie Blue Staining If one is primarily interested in detection of fairly abundant proteins, and not concerned with the determination of purity or the detection of trace proteins, the Coomassie Blue stains may be useful. They were originally developed as acid wool dyes and they were named "Coomassie dyes" to commemorate the 1896 British occupation of the Ashanti capital, Kumasi or "Coomassie," now in Ghana. Coomassie Blue R-250 (the letter " R " stands for a reddish hue while the number "250" is a dye strength indicator) was the first of these triphenylmethane stains to be introduced.l°,11 Other Coomassie stains, such as Coomassie Blue G-250 ( " G " indicates that this stain has a greenish hue), have augmented the original Coomassie stain. Coomassie Blue G-250 has a diminished solubility in 12% TCA, permitting its use as a colloidal dispersion which does not 4 E. L. Durrum, M. H. Paul, and E. R. B. Smith, Science 116, 428 (1952). 5 E. Koiw and A. Gronwell, Scand. J. Clin. Lab Invest. 4, 244 (1952). 6 B. O. Barger, F. C. White, J. L. Pace, D. L. Kemper, and W. L. Ragland, Anal. Biochem. 70, 327 (1976). 7 H. F. Bosshard and A. Datyner, Anal. Biochem. 82, 327 (1977). 8 C. R. Merril, R. C. Switzer, and M. L. Van Keuren, Proc. Natl. Acad. Sci. U.S.A. 76, 4335 (1979). 9 R. C. Switzer, C. R. Merril, and S. Shifrin, Anal. Biochem. 98, 231 (1979). ~0S. Fazekas de St. Groth, R. G. Webster, and A. Datyner, Biochim. Biophys. Acta 71, 377 (1963). n T. S. Meyer and B. L. Lamberts, Biochim. Biophys. Acta 107, 144 (1965).
[36]
GEL-STAINING TECHNIQUES
479
penetrate gels. This property permits rapid staining of proteins without an undesired background. ~2 Another Coomassie stain, Coomassie Violet R-150, has gained some favor by virtue of its ability to rapidly stain proteins on polyacrylamide gels while not staining carrier ampholytes, and for its ease in destaining) TM Coomassie dyes are no longer made by Imperial Chemical Industries and they are now often sold under a number of different trade names. General Coomassie Staining Method Gels are stained immediately after electrophoresis in a solution containing 50% (v/v) methanol, 10% (v/v) acetic acid, and 0.25% (w/v) Coomassie Blue for 3 hr. This solution should be filtered (Whatman No. 1) prior to use. Gels are destained overnight in a solution containing 5% (v/v) acetic acid and 10% methanol. The destaining solution must be changed repeatedly, or alternatively it may be pumped continuously through a felt filter. The felt clarifies the destaining solution by binding the Coomassie stain as it diffuses out of the gels. Rapid Coomassie Stain for Isoelectric Focusing Gels A simple method for staining gels which do not contain sodium dodecyl sulfate utilizes a 6% (w/v) perchloric acid solution containing 0.04% (w/v) of Coomassie G-250. Dense protein bands or spots stain an intense blue and can often be observed within less than a minute. The background stains a pale orange. Less dense proteins can usually be visualized within 90 min. A 3-fold increase in sensitivity can be achieved by placing the gel in 5% (v/v) acetic acid. The background changes to pale blue in the acetic acid. 15 Properties of Coomassie Blue Stains Coomassie Blue Stain Binding Coomassie Blue staining requires an acidic medium for the generation of an electrostatic attraction between the dye molecules and the amino groups of the proteins. This ionic attraction, together with van der Waals forces, binds the dye-protein complex together. The binding is fully reversible by dilution under appropriate conditions.l° Polypeptides rich in 12w. Diezel, G. Kopperschlager, and E. Hofman, Anal. Biochem. 48, 617 (1972). 13R. Frater, J. Chromatogr. 50, 469 (1970). ~4B. J. Radola, Electrophoresis 1, 43 (1980). 15A. H. Reisner, this series, Vol. 104, p. 439.
480
PURIFICATION PROCEDURES" ELECTROPHORETIC METHODS
[36]
lysine and arginine are aggregated by Coomassie G dye molecules, suggesting that the dye interacts with the basic groups in the polypeptides. 16 Studies of proteins with known sequences have confirmed these observations and demonstrated a significant correlation between the intensity of Coomassie Blue staining and the number of lysine, histidine, and arginine residues in the protein.17 Coomassie Blue stains exhibit three times the staining intensity of Fast Green and six times the intensity of Amido Black.18 The staining intensities of these dyes is approximately proportional to their relative molar absorption coefficients. One milligram of protein will bind 0.17 mg of Amido Black, 0.23 mg of Fast Green, 1.2 mg of Coomassie Blue R-250, and 1.4 mg of Coomassie Blue G-250. ~9 Since the molecular weights of these dyes vary by only 1.4-fold these 5- to 8-fold variations in dye binding are most likely due to differences in the number of dye molecules bound per protein molecule. The higher staining intensity of Coomassie Blue may also be due to its higher efficiency at forming dye-dye interactions. Secondary binding mechanisms may also occur with Amido Black and Fast Green dyes, although perhaps not at the levels observed with Coomassie Blue R-250, as these dyes display metachromatic effects with certain proteins similar to the metachromasy observed with Coomassie Blue R-250. Amido Black produces blue-green bands with certain histones rather than its characteristic blue-black color, while Fast Green produces a difference in the ratio of blue to green hues. 2° Collagen and histones often produce redstaining bands or spots with Coomassie Blue. 2~ These metachromatic effects are dependent on temperature, concentrations, and the solvents in the gel. Coomassie stains give a linear response up to 20/~g/cm. 10,, However, the relationship between stain density and protein concentration varies for each protein. ~0 Preelectrophoretic Fluorescent Stains Fluorescent stains usually involve the covalent binding of a fluorescent residue to the protein prior to electrophoresis. The advantages of this type of stain include the possibility of performing stoichiometric reactions 16 p. G. Righetti and F. Chillemi, J. Chromatogr. 157, 243 (1978). 17 M. Tal, A. Silberstein, and E. Nusser, J. Biol. Chem. 260, 9976 (1985). 18 C. M. Wilson, this series, Vol. 91, p. 236. z9 C. M. Wilson, Anal. Biochem. 96, 236 (1979). 2o R. McMaster-Kaye and J. S. Kaye, Anal. Biochem. 61, 120 (1974). 2z R. C. Duhamel, E. Meezan, and K. Brendel, Biochim. Biophys. Acta 626, 432 (1980).
[36]
GEL-STAININGTECHNIQUES
481
with proteins without the diffusion limitations imposed by staining within a gel matrix, the feasibility of following the process of electrophoresis visually with "stained" proteins, and the absence of background problems due to dye trapping or reaction of the dye with the gel. These advantages may be offset in some applications by the alterations in the charge of the proteins, unless an amphoteric stain is employed. 7 This objection is not of consequence for sodium dodecyl sulfate (SDS) electrophoresis, as the mobility of the protein depends solely on molecular weight with this method, and the dye molecules are usually too small to produce an appreciable size effect. However, the fluorescent stains require ultraviolet light for visualization, and direct quantitation requires fairly sophisticated equipment. Currently fluorescent stains, such as fluorescamine, are the most sensitive preelectrophoretic stains. At room temperature and alkaline pH, fluorescamine can react with the primary amines of the protein to yield a fluorescent derivative. This stain has proved capable of detecting as little as 6 ng of myoglobin. 22,23 A related compound, 2-methoxy-2,4-diphenyl3(2H)-furanone (MDPF), has the same speed and simplicity of reaction as fluorescamine, while its protein derivative is 2.5 times as fluorescent as a fluorescamine-labeled protein. Furthermore, its fluorescent derivative does not fade as rapidly. As little as 1 ng of protein has been detected with MDPF. This stain has a linear response from 1 to 500 ng. As with most other protein stains, a plot of relative fluorescence versus protein concentration reveals a different slope for each protein. 6
General Fluorescence Staining Method To label proteins with fluorescamine or MDPF, first add 50/zl of 0.2 M borate (pH 9.0) buffer to a protein solution containing 50 to 100/zg of protein. Then add, with vortex mixing, 30/.d of MDPF or fluorescamine stock solution. (The fluorescamine or MDPF stock solution contains 2 mg of stain dissolved in 1 ml of acetone.) Continue the mixing for about 1 min. The proteins will be labeled within this time. There is no need to remove the remaining unreacted stain reagent as it is not fluorescent and it degrades rapidly in water. It also does not interfere with the electrophoretic separation of the labeled proteins. Although maximal fluorescence of the labeled proteins is obtained at pH 8-8.5, fluorescence can be detected over a wider pH range. 22 W. L. Ragland, J. L. Pace, and D. L. Kemper, Anal. Biochem. 59, 24 (1974). 23 j. L. Pace, D. L. Kemper, and W. L. Ragland, Biochem. Biophys. Res. Commun. 57, 482 (1974),
482
PURIFICATION PROCEDURES: ELECTROPHORETIC METHODS
[36]
Silver Staining Studies in which the purity of a protein is critical, or in which there is a need to monitor trace proteins, should employ the highly sensitive silver stains. Silver stain protocols can be divided into three basic categories: the diamine or ammoniacal silver stains, the nondiamine chemical development silver stains, and the photoreduction silver stains. The diamine or ammoniacal silver stains have proved to be particularly good for the staining of proteins separated in gels thicker than 1 mm. The nondiamine chemical development stains are generally more rapid than the diamine stains and they work best with l-mm or thinner gels. The photoreduction silver stains are the most rapid, but they currently lack the sensitivity of the other silver stain methods. Diamine Silver Stains These stains rely on the stabilization of the silver ions by the formation of silver diamine complexes with ammonium hydroxide. Silver ion concentrations are usually very low in these stains, as most of the silver is bound in the diamine complexes, z4 In these diamine stains, the ammoniacal silver solution must be acidified, usually with citric acid, for image production to occur. The addition of citric acid lowers the concentration of free ammonium ions, thereby liberating silver ions to a level where their reduction by formaldehyde to metallic silver is possible. The optimal concentration of citric acid results in a controlled rate of silver ion reduction, preventing the nonselective deposition of silver. Diamine Staining Method Gels are washed for 5 min in deionized water and then placed in a solution containing 5% (v/v) ethanol, 5% (v/v) acetic acid, and deionized water for 3 hr. The deionized water used to make these solutions should have a conductivity of less than 1 mho/cm. The gels may be stored in this solution overnight prior to staining. The gels are then washed with deionized water for 5 min and then soaked for 30 min in a 10% (v/v) glutaraldehyde solution. The unreacted glutaraldehyde is removed by five 30-min washes with deionized water. These glutaraldehyde-treated gels are then soaked in an ammoniacal silver nitrate solution for 10 min. The ammoniacal silver nitrate solution is prepared by slowly adding, with stirring, 30 ml of a 1.2 M silver nitrate solution to a solution containing 10 ml of concentrated ammonium hydroxide and 1.5 ml of 10 N sodium hydroxide in 160 ml of deionized water. After the silver nitrate is dissolved the final volume is adjusted to 750 ml. 24 W. J. H. N a u t a a n d P. A. Gygax, Stain Technol. 26, 5 (1951).
[36]
GEL-STAININGTECHNIQUES
483
The gels are removed from the ammoniacal silver nitrate and treated with three 5-min deionized water washes. The image is developed with a solution containing 0.1 g of citric acid and 1 ml of formaldehyde (37% commercial formaldehyde) per liter of deionized water. When the image is sufficiently developed, usually about 3 min, the reaction is stopped by placing the gels in solution containing 5% (v/v) acetic acid. The gels are then washed in a 10% (v/v) ethanol solution. If the gels are left in a solution containing acetic acid some of the trace bands or spots will be lost. Gels are stored in a 7% (v/v) glycerol, 10% (v/v) ethanol solution. Diamine stains tend to become selectively sensitive for glycoproteins if the concentration of silver ions is maintained at a low level during image development. This specificity can be minimized by maintaining a sufficient sodium-to-ammonium ion ratio in the diamine solution. 25 However, in some applications, an emphasis on the specificity of the diamine stain has proved useful, as in the adaptation of a diamine histological silver stain to visualize neurofilament polypeptides in electrophoretic analyses of spinal cord homogenates. 26 Nondiamine Chemical Development Silver Stains These stains are relatively simple and rapid. They rely on the reaction of silver nitrate with protein sites under acidic conditions, followed by the selective reduction of silver ion to metallic silver by formaldehyde under alkaline conditions. Sodium carbonate and/or hydroxide and other bases are used to maintain an alkaline pH during development. Formic acid, produced by the oxidation of formaldehyde, is buffered by the sodium carbonate. Nondiamine Staining Method Gels are fixed for 20 min in a solution containing 50% (v/v) methanol, 10% (v/v) acetic acid, and deionized water. The gels may be stored in this solution overnight prior to staining. The gels are then washed for 30 min in a solution containing 10% (v/v) methanol, 5% (v/v) acetic acid, and deionized water. These gels are then soaked in a 3.4 mM potassium dichromate solution containing 3.2 m M nitric acid for 5 min. The gels are then rinsed with deionized water and placed in 12 mM silver nitrate for 20 min. Image development is achieved by rinsing the gels with agitation in 0.28 M sodium carbonate containing 0.5 ml formaldehyde (37% commercial formaldehyde) per liter of deionized water. This step requires at least two changes of the solution to prevent precipitated silver salts from adsorbing 55 R. C. Allen, Electrophoresis 1, 32 (1980). 26 p. Gambetti, L. Autilio-Gambetti, and S. C. H. Papasozomenos, Science 213, 1521 (1981).
484
PURIFICATION PROCEDURES: ELECTROPHORETIC METHODS
[36]
to the surface of the gel. The pH of the gel is made alkaline so that the formaldehyde can reduce ionic silver to metallic silver. Image development is stopped when a slightly yellowish background appears by placing the gel in a 3% (v/v) acetic acid solution for 5 min. The gels are then washed in a 10% (v/v) ethanol solution. If the gels are left in acetic acid some of the trace bands or spots will be lost. Gels are stored in a solution containing 7% (v/v) glycerol and 10% (v/v) ethanol.
Photodevelopment Silver Stains The use of photoreduction provides for a rapid, simple, staining method for the detection of proteins. It permits the visualization of protein patterns within 10 min after an electrophoretic separation. However, the method currently lacks the sensitivity of the other silver staining methods and it should be reserved for studies of dense protein bands or spots. 27'28Photodevelopment stains utilize energy from photons of light to reduce ionic to metallic silver.
Photodevelopment Silver Staining Method Gels are fixed for 5 min in a solution containing 50% (v/v) methanol, 10% (v/v) acetic acid, 2% (w/v) citric acid, and 2% (w/v) sodium chloride in deionized water. The gels are rinsed briefly with deionized water to remove surface chloride and placed in a solution containing 50% (v/v) methanol, 10% (v/v) acetic acid, and 2% (w/v) silver nitrate. The gels immersed in this solution are then transilluminated by placing them 2.5 cm above a uniform fluorescent light source (a 160-W fluorescent grid lamp with a clear Lucite diffusion screen) until an image appears. Image development may be stopped at any time by placing the gel i n t h e dark. Image preservation, which is very good with the other silver-staining methods, is difficult with this photodevelopment silver stain. Archival storage can only be achieved by photographing these gels. Properties of Silver Stains
Silver Stain Reactive Groups Amino acid homopolymers, individual amino acids, and peptides of known sequence have been studied to gain information about reactive groups that may be involved in the silver-staining reactions, z9The consen27 C. R. Merril and M. G. Harrington~ Clin. Chem. 30, 1938 (1984). 28 C. R. Merril, M. Harrington, and V. Alley, Electrophoresis 5, 289 (1984). 29 C. R. Merrii and M. E. Pratt, Anal. Biochem. 117, 307 (1986).
[36]
GEL-STAININGTECHNIQUES
485
sus findings of these studies indicate that the principal reactive groups are the free amines and the sulfur groups contained on the proteins. 3°-32 The importance of the basic and the sulfur-containing amino acids has been corroborated by observations with purified peptides and proteins of known amino acid sequence. 29 The importance of the basic amino acids has been further substantiated by evaluations of the relationship between the amino acid mole percentages of proteins and their ability to stain with silver. The best correlations are achieved when comparisons are made between the slope of the staining curve of a denatured protein and that protein's mole percentage of basic amino acids. 29 Color Effects with Silver Stains Most proteins stain with monochromatic brown or black colors. However, silver stains can produce other colors. Lipoproteins tend to stain blue while some glycoproteins appear yellow, brown or red. 33 This color effect has been demonstrated to be due to the diffractive scattering of light by the microscopic silver grains. A pronounced and reproducible dependence of color and silver-grain size has been observed. 34 Modifications of the silver-staining procedures, such as lowering the concentration of reducing agent in the image development solution, prolonging the development time, adding alkali, or elevating the temperature during staining, often enhance color formation. Some silver stain protocols have been developed to produce colors that may aid in the identification of certain p r o t e i n s . 32,35,36 Combinations of stains may also be employed for protein identification. In a study of erythrocyte membrane proteins, sialoglycoproteins and lipids were stained yellow with a silver stain, while other membrane proteins counterstained with Coomassie Blue. 37 30 H. C. Freeman, in "Inorganic Biochemistry" (G. L. Eichhorn, ed.), Vol. 1, p. 121. Elsevier, Amsterdam, 1973. 31 j. Heukeshoven and R. Demick, Electrophoresis 6, 103 (1985). 32 B. L. Nielsen and L. R. Brown, Anal. Biochem. 144, 311 (1984). 33 D. Goldman, C. R. Merril, and M. H. Ebert, Clin. Chem. 2,6, 1317 (1980). 34 C. R. Merril, M. E. Bisher, M. Harrington, and A. C. Steven, Proc. Natl. Acad. Sci. U.S.A. 85, 453 (1988). 35 D. W. Sammons, L. D. Adams, and E. E. Nishizawa, Electrophoresis 2, 135 (1981). 36 D. W. Sammons, L. D. Adams, T. J. Vidmar, A. Hatfield, D. H. Jones, P. J. Chuba, and S. W. Crooks, in "Two-Dimensional Gel Electrophoresis of Proteins" (J. E. Celis and R. Bravo, eds.), p. 112. Academic Press, New York, 1984. 37 j. K. Dzandu, M. H. Deh, D. L. Barratt, and G. E. Wise, Proc. Natl. Acad. Sci. U.S.A. 81, 1733 (1984).
486
PURIFICATION PROCEDURES: ELECTROPHORETIC METHODS
[36]
Protein-Specific Silver Stains Silver stains can demonstrate considerable specificity. Stains specific for nucleolar proteins 38 and neurofilament polypeptides 26 have been described. Many silver stain protocols detect not only proteins but also DNA, 3~I lipopolysaccharides, 42 and polysaccharides. 43 All silver stains do not detect all proteins. It is difficult to stain calmodulin and troponin C with some silver stain protocols. However, pretreatment of these proteins with glutaraldehyde appears to enhance their ability to stain. 44 Some histones may also fail to stain with silver. Fixation with formaldehyde coupled with the simultaneous prestaining of these proteins with Coomassie Blue has been reported to partially alleviate this problem. However, even with this procedure the sensitivity for histones is reported to be decreased 10-fold when compared with the detection of neutral proteins. 45 Another example of differential sensitivity has been demonstrated in a study utilizing four different silver stain protocols to stain salivary proteins. Different protein bands were visualized with each of the stains. 46
Silver Stain Quenching of Autoradiography Quenching of ~4C-labeled proteins is minimal with most nondiamine silver stains. Even the most intense diamine-stained radioactive proteins can be detected by autoradiography with only a 50% decrease in image density. This loss of autoradiographic sensitivity can generally be compensated for by longer film exposures. However, detection of 3H-labeled proteins is severely quenched by all silver stains. Destaining of silver-stained gel with photographic reducing agents can often permit detection of as much as half of the fluorographic density of 3H-labeled proteins, providing that the initial staining was performed with a nondiamine silver stain. Many diamine stains continue to quench, even after treatment with photographic reducing agents, so that fluorographic detection of 3H-labeled proteins is not feasible with the diamine stains. This impediment to 3H detection with diamine stains is likely to be due to a greater amount of residual silver deposited throughout the gels by these stains, which block the weak/3 emissions from 3H. 38 H. R. Hubbell, L. I. Rothblum, and T. C. Hsu, Cell Biol. Int. Rep. 3, 615 (1979). 39 L. L. Somerville and K. Wang, Biochem. Biophys. Res. Commun. 10, 53 (1981). 4o T. Boulikas and R. J. Hancock, Biochem. Biophys. Methods 5, 219 (1981). 41 D. Goldman and C. R. Merril, Electrophoresis 3, 24 (1982). 42 C. M. Tsai and C. E. Frasch, Anal. Biochem. 119~ 115 (1982). 43 G. Dubray and G. Bezard, Anal. Biochem. 119, 325 (1982). M. Schleicher and D. M. Watterson, Anal. Biochem. 131, 312 (1983). 45 S. lrie and M. Sezaki, Anal. Biochem. 134, 471 (1983). R. D. Friedman, Anal. Biochem. 126, 346 (1982).
[36]
GEL-STAININGTECHNIQUES
487
Residual silver has been demonstrated in gels that have been "cleared" by photographic reducing agents, by the reappearance of a faint silver image of the proteins in "cleared" gels which are dried with heat. Silver has also been demonstrated in these "cleared" gels by electron beam analysis .47
Common Staining Artifact and Background Staining Artifactual bands with molecular weights ranging from 50K to 68K have been commonly observed in silver-stained gels. Evidence has been presented indicating that these contaminating bands are due to keratin skin proteins. 48 Background staining has been demonstrated to be due in part to the chemistry of the polyacrylamide gels. Preliminary experiments indicate that alterations in the chemistry of the polyacrylamide gels may result in reduced background staining. 49
Q~antitation with Protein Stains
Quantitation Using Organic Stains An accuracy of plus or minus 10% in measuring the concentrations of proteins in the range of 0.5-20/~g was reported with the introduction of the Coomassie Blue R-250 stain. ~° However, while individual proteins displayed linear relationships between absorbance and concentration within this protein range, the slopes differed for each protein. This variation in Coomassie Blue staining now appears to be related to the mole percent of the basic amino acids in the protein.17 Therefore, a standard curve must be produced for each protein assayed and quantitative comparisons limited to equivalent protein spots or bands.
Quantitation Using Silver Stains Most silver stain protocols provide a reproducible relationship between silver stain density and protein concentration. The linear portion of this relationship generally extends over a 40-fold range in concentration, beginning at 0.02 ng/mm2.16,27,5°,51 Protein concentrations greater than 47 M. L. Van Keuren, D. Goldman, and C. R. Merril, Anal. Biochem. 116, 248 (1981). 4s D. Ochs, Anal. Biochem. 135, 470 (1983). 49 D. F. Hochstrasser, A. Patchornik, and C, R. Merril, Anal. Biochem. 173, 412 0988). 50 C. R. Merril, D, Goldman, and M. L. Van Keuren, Electrophoresis 3, 17 (1982). ~1 C. R. Merril and D. Goldman, in "Two-Dimensional Gel Electrophoresis of Proteins" (J. E. Celis and R. Bravo, eds.), p. 93. Academic Press, New York, 1984.
488
PURIFICATION PROCEDURES: ELECTROPHORETIC METHODS
[37]
2 ng/mm 2 generally cause saturation of silver images, resulting in nonlinearity above that concentration. In a manner analogous to that observed with the organic stains, the relation between the silver stain densities and the protein concentrations varies for each protein. 8,9,29,5°,51 Protein-specific staining curves have also been observed with most protein assays (see [6] in this volume). The observation that each protein produces a unique density versus concentration curve in these studies illustrates the dependence of the staining reaction on a specific reactive group(s) contained in each protein. The importance of the basic amino acids, particularly lysine and histidine, for both the silver stains and the Coomassie Blue stains indicates the need for a careful choice of "standard protein(s)." If a protein containing an abnormally large number of stainreactive groups is chosen as a standard it will produce a curve which would tend to underestimate the concentration of proteins containing normal numbers of reactive groups.17 Alternatively, in intergel comparisons only equivalent spots offer valid quantitative comparisons. [Editor's note: See [33] and [35] in this volume for additional information on gel staining.]
[37] E l u t i o n o f P r o t e i n f r o m Gels
By MICHAEL G. HARRINGTON Electrophoretic separation of proteins in various types of polyacrylamide gels is employed from the analytical to the preparative scale. After separation, it is frequently necessary to extract, or elute, a specific protein from the gel for further study: this might include amino acid composition or sequence analysis, or partial enzyme or chemical digestion. For optimal efficiency of elution, it is desirable to have a simple technique that successfully extracts all protein from the gel and avoids any additional chemical modification to the protein. The diffusion method of elution involves agitation of the gel fragments in a sodium dodecyl sulfate (SDS) solution. This approach is simple, takes 3-12 hr, but is less efficient than electroelution, and will not be discussed further. Electroelution is more controlled than diffusive elution, and can be performed either during or after electrophoresis. The author's limited experience with elution during electrophoresis precludes further description of what is a less flexible method. The following detailed laboratory procedure pertains to the identification in the gel of a specific protein and the subsequent elution of that METHODS IN ENZYMOLOGY, VOL. 182
Copyright © 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.