ARTICLE continued from page 9 pensive method to maintain very low basal may be observed in the absence of glucose. tionary phase, we recommend the addition expression levels of T7 RNA polymerase in Overall, the optimal combination of strin- of 0.5 to 1.0% glucose to the medium, so the λDE3 lysogenic expression hosts used gent uninduced repression and high in- that the glucose effect can be exploited to in the pET System. This is especially true duced expression may be achieved by initial reduce basal expression. when λDE3 hosts carrying pET plasmids growth in the presence of glucose, followed are grown to stationary phase. A disadvan- by switching to medium without glucose tage with glucose addition is that after an for induction. REFERENCES initial rapid growth phase the metabolic Novagen’s recommendations for growth 1. Grossman, T. H., Kawasaki, E. S., breakdown products of glucose will lead to and induction of pET constructs in expres- Punreddy, S. R., and Osburne, M. S. acidic culture conditions and lower cell sion hosts are based on the information pre- (1998) Gene 209, 95–103 density at stationary phase. The data pre- sented above. For innocuous proteins, any 2. Pan, S. and Malcolm, B. A. (2000) sented in Figure 2 demonstrate that strong pET vector and λDE3 lysogen are suitable BioTechniques 29, 1234–1238. induction can be achieved from λDE3 lyso- in a variety of media. But for proteins that 3. Le Grice, S. F. J., Matzura, H., Marcoli, gens in the presence of glucose for some are potentially toxic to the bacterial cell, we R., Iida, S., and Bickle, T. A. (1982) J. Bacteriol. 150, 312–318. target proteins. Note, however, that theoret- recommend using either a pET vector with 4. Kelley, K. C., Huestis, K. J., Austen, D. ically the strongest induction of T7 RNA a T7lac promoter or expression hosts that A., Sanderson, C. T., Donoghue, M. polymerase would be expected when glu- carry the pLysS plasmid. In addition, our A., Stickel, S. K., Kawasaki, E. S., and cose is absent and cAMP levels are elevated. general advice is to avoid growing a λDE3 Osburne, M. S. (1995) Gene 15, Accordingly, in some cases (see preceding lysogen carrying a pET plasmid to station- 33–36. article), higher target protein expression ary phase. If the cells must be grown to sta- Preparation of protein samples for SDS-polyacrylamide gel electrophoresis: procedures and tips Anthony C. Grabski1 and Richard R. Burgess2—1Novagen, Inc. and 2McArdle Laboratory for Cancer Research, University of Wisconsin-Madison, Madison, WI 53706

odium dodecyl sulfate polyacry- components because the proteins are fo- dious gel casting, buffer preparation and ap- lamide gel electrophoresis (SDS- cused, or “stacked,” as thin bands prior to paratus set-up, but on careful sample prepa- PAGE) is the most widely used ana- entering the resolving gel. Second was the ration and treatment prior to loading the lytical method to resolve separate use of the detergent sodium dodecyl sulfate gel. This article describes techniques and Scomponents of a protein mixture. It is (SDS) and reducing agents to denature pro- procedures as a guide for preparation of pro- almost obligatory to assess the purity of a teins (7). SDS binds strongly to proteins at tein samples for SDS-PAGE analysis. protein through an electrophoretic method. an approximate ratio of 1 dodecyl sulfate Sample buffer preparation SDS-PAGE simultaneously exploits differ- molecule per 2 amino acid residues (8). ences in molecular size to resolve proteins Therefore, the negative charge/unit mass To ensure consistent and successful differing by as little as 1% in their elec- ratio when SDS is bound to the polypeptide PAGE analysis, the highest purity reagents trophoretic mobility through the gel matrix chain is similar for all proteins. Third was should be used to prepare sample buffer (1). The technique is also a powerful tool for the combination of the first two discoveries stock solutions. After a reliable source of estimating the molecular weights of proteins employing a simple Tris-glycine buffer electrophoresis reagents has been identified, (2, 3). The success of SDS-PAGE as an in- system (9). More recently, buffer combina- the vendor and buffer component chemicals dispensable tool in protein analysis has been tions such as Tris-borate (10) and Tris- should be maintained. High purity elec- attributed to three innovations that permit- tricine (11) have improved the resolving trophoresis, Ultrol® grade, and molecular bi- ted the correlation of electrophoretic mobil- power of the original methods. Modern ology grade reagents are available through ity with a protein’s molecular mass (4). First SDS-PAGE has evolved to use microslab Novagen’s partner brand, Calbiochem. was the introduction of discontinuous buffer precast gels (12). Precast and packaged gels Solutions must be carefully and safely pre- systems where the sample and gel running in a wide variety of gel formulations, acry- pared, dated, and chemical lot numbers buffers differ in both composition, Tris- lamide percentages, thicknesses, well for- recorded. Concentrated stock solutions HCl/Tris-glycine, and pH, 6.8/8.3, respec- mats, and buffer systems are now commer- should not be stored for long periods of tively (5, 6). Discontinuous buffer systems cially available from several manufacturers. time. Tris base, rather than Tris-Cl, should allow larger sample volumes to be loaded Therefore, successful SDS-PAGE analysis of be used for buffer preparation and pH ad- while maintaining good resolution of sample protein samples no longer depends on te- justment made with HCl. Use of Tris-Cl

inNovations 13 10 will result in a higher ionic strength, poor Table 1. SDS-PAGE sample buffer recipes 1.4 µg SDS per 1.0 µg polypeptide, but a migration and diffuse protein bands (13). A ratio of 3:1 is recommended (15). The 2X Component Concentration respirator or dust mask should be worn sample buffer prepared as shown in Table 1 2X 4X when handling powdered SDS. Sulfhydryl Tris-HCl, pH 6.81 0.125 M 0.25 M contains 40 µg/µl SDS. Maintained reduc- reagents, dithiothreitol (DTT) and 2-mer- SDS 4% 8% tion of protein sulfhydryls is essential in captoethanol (2-ME) can be unstable in so- 2-ME2 5% 10% order to prevent intramolecular disulfide lution and are toxic. These chemicals should DTT3 0.15 M 0.3 M bond formation through oxidized cysteines. be measured in a fume hood while wearing Glycerol 20% 30% If artifactual band heterogeneity or unusual gloves and safety glasses. Although 2-ME is Bromphenol blue .01% .02% doublets are noted in SDS-PAGE results historically the chemical of choice for reduc- 1. Prepared using Tris base, pH adjusted with HCl. from samples containing sulfhydryls, insuf- tion of protein disulfide bonds in SDS- 2. If 2-ME is used, omit DTT. ficient reducing agent was present during PAGE (7, 9), DTT is also a very effective al- 3. If DTT is used, omit 2-ME. sample treatment or the oxidation of cys- ternative (14). Glycerol is added to increase (BCA) assay prior to sample buffer addition. teines may have occurred during the stack- the sample density, facilitating gel loading Loading too much protein will result in dis- ing phase of electrophoresis. These artifacts and preventing convective migration out of torted, poorly resolved bands in the over- may be prevented if samples are treated with the sample wells. A small amount of loaded lane and distorted electrophoretic iodoacetamide (IAA) after heating in the ap- bromphenol blue is added as a visual aid patterns in adjacent lanes. Underloading propriate concentration of sample buffer. during sample loading and as a tracking dye, simply prevents detection of minor compo- The IAA treatment irreversibly blocks allowing easy monitoring of electrophoretic nents while even major bands will be too sulfhydryls and destroys excess reducing progress. faint for photographic reproduction of the agent (16). Therefore, the sample buffer The sample buffer recipes listed in Table recipes in Table 1 do not necessarily indicate 1 are commonly used for Tris-glycine SDS- functional dilution factors, but rather they PAGE analysis of protein samples under de- are convenient stock concentrations permit- naturing, reduced conditions (7, 9, 13). ting correct addition of reagents to samples When preparing these buffers, wear gloves of high and low protein concentration. to avoid keratin (skin protein) contamina- Delayed heating of samples after sample tion. A heterogeneous cluster of bands buffer addition or excessive heating can around 55 kDa can be seen when a keratin- cause electrophoretic artifacts due to protein contaminated sample or sample buffer is degradation and peptide bond cleavage, re- used. This is particularly obvious with high spectively. Upon addition of SDS sample sensitivity silver staining methods. The buffer, samples should be immediately sample buffer should be divided into 1-ml Sample preparation is critical mixed and heated to 85°C for three min- aliquots and can be stored frozen (–70°C) for clear and accurate resolution utes. This treatment is usually sufficient to for several months. Prior to use, warm of protein bands. reduce disulfides, solubilize and dissociate (37°C) and mix the solution briefly to com- proteins without peptide bond cleavage. pletely dissolve the SDS. gel. Depending on the well size and gel Addition of SDS sample buffer will begin to thickness, the amount of protein loaded denature most proteins. However, proteases Protein sample preparation should range from 0.5–4.0 µg for purified are known to be resistant to SDS denatura- Sample preparation is critical for clear samples and from 40–60 µg for crude sam- tion alone (15, 17). Partially denatured sam- and accurate resolution of protein bands. ples if a Coomassie blue stain (e.g., ples (particularly crude extracts) are there- Photographic quality results are routinely RAPIDstain™) is used. Silver staining fore extremely sensitive to proteolytic possible if samples are carefully prepared. methods (such as the FASTsilver™ Kit) are degradation as protease active sites within Common mistakes during sample prepara- approximately 100-fold more sensitive, and the polypeptides become exposed by SDS tion include using an incorrect protein-to- therefore require less protein per sample. treatment. Immediate heating limits degra- sample buffer ratio, delayed heating, over- SDS-PAGE sample buffer treatment is dation by completely denaturing all proteins heating, failure to remove insoluble designed to completely dissociate all pro- including resistant proteases through the material, and overloading and underloading teins into their subunit polypeptides. combination of heat, SDS, and reductant. of protein. To prevent inadequate sample Proteins heated in the presence of SDS are Protease inhibitors may also be used during buffer-to-protein ratios, overloading, and denatured and imparted with a strong nega- sample preparation to limit proteolysis. underloading of samples, the protein con- tive charge. Thiol reagents in the sample Excessive heating, e.g., 100°C for prolonged centration of the sample should be deter- buffer reduce disulfide bonds. It is impor- periods, may break peptide bonds or cause mined using a standard protein assay such as tant to use enough sample buffer in order to selective aggregation and band smearing the CB-Protein Assay™, Non-Interfering maintain an excess of SDS. Most polypep- (18). Asp-Pro bonds have been demon- Protein Assay™, or bicinchoninic acid tides bind SDS in a constant mass ratio of strated to be sensitive to thermal cleavage. continued on page 12 inNovations 13 11 ARTICLE continued from page 11 In some cases more extreme heating may be precipitation method is also described in 14. Cleland, W. W. (1964) Biochemistry 3, necessary to completely denature the protein Novagen Technical Bulletin 012, available at 480–482. (19, 20). Therefore, if prolonged heating at www.novagen.com. Crude cell extracts are 15. Hames, B. D. (1990) in “Gel 100°C is necessary for complete dissociation often extremely viscous due to the high con- Electrophoresis of Proteins,” Hames, B. D. and Rickwood, D., eds., pp. 1–147, of a thermally stable protein, the effects of centration of unsheared nucleic acids. The Oxford University Press, New York. such treatment upon peptide bond cleavage high viscosity is problematic during gel 16. Crow, M. K., Karasavvas, N., and Sarris, must be considered (21). Some proteins loading, because samples are difficult to A. H. (2001) BioTechniques 30, such as histones and membrane proteins pipet and will not be evenly distributed in 311–316. may not completely dissolve by heating in the sample well. Viscosity can be eliminated 17. Gallagher, S. R. (2000) in “Current SDS sample buffer alone and may require by treatment of samples with Benzonase® Protocols in Protein Science,” Coligan, addition of 6–8 M urea or a nonionic deter- Nuclease prior to addition of sample buffer. J. E., Dunn, B. M., Ploegh, H. L., Speicher, D. W., and Wingfield, P. T., gent such as Triton X-100 (22, 23). After This recombinant endonuclease completely eds., pp. 10.1.29–10.1.34, John Wiley heat treatment in SDS sample buffer, insolu- degrades all forms of DNA and RNA and is & Sons, New York. ble material must be removed by brief cen- free from proteolytic activity. Viscosity can 18. Gallagher, S. R. and Leonard, R. T. trifugation. This is easily accomplished by a also be reduced by physical shearing of the (1987) Plant Physiol. 87, 120–125. two-minute spin in a microcentrifuge at nucleic acids through sonication or by vigor- 19. Bensadoun, A., Ehnholm, C., Steinberg, 17,000 × g. Failure to remove precipitated ous vortex mixing of the heated sample. D., and Brown, W. V. (1974) J. Biol. insoluble material from the sample will Employing these pre-treatment protocols Chem. 249, 2220–2227. cause streaking within the gel. The super- will allow successful SDS-PAGE analysis of 20. Slutzky, G. M. and Ji, T. H. (1974) Biochim. Biophys. Acta 373, 337–346. natant of the treated sample is now ready to samples that are very dilute, viscous or cont- 21. Deutsch, D. G. (1976) Anal. Biochem. load. The sample may be stored at 4°C aminated with interfering compounds. 71, 300-303. overnight or frozen at –20°C for longer peri- 22. Franklin, S. G. and Zweidler, A. (1977) ods. Warm stored samples briefly at 37°C to Nature 266, 273–275. redissolve the SDS and recentrifuge to REFERENCES 23. Siu, C. H., Lerner, R. A., and Loomis, remove insoluble material prior to loading. 1. Scopes, R. K. (1994) “Protein W. F. (1977) J. Mol. Biol. 116, 469–488. Preparation of difficult samples Purification: Principles and Practice,” 3rd ed., Springer Verlag, New York. 24. Hager , D. A. and Burgess, R. R. (1980) Samples that are dilute, acidic, very vis- 2. Weber, K., Pringle, J. R., and Osborn, Anal. Biochem. 109, 76–86. cous, or that contain interfering compounds M. (1971) Meth. Enzymol. 26, 3–27. pose unique challenges to the SDS-PAGE 3. Chambach, A. and Rodbard, D. (1971) Product Size Cat. No. analysis method. However, these difficult Science 172, 440–450. 4. Marshak, D. R., Kadonaga, J. T., Benzonase® Nuclease, samples can be analyzed by SDS-PAGE Purity > 90% 10,000 U 70746-3 through the application of one or more of Burgess, R. R., Knuth, M. W., Brennan, W. A., and Lin, S. -H. (1996) Aquacide I 1 kg 1785 the following pre-treatment techniques. “Strategies for Protein Purification and Aquacide II 1 kg 17851 Samples too dilute for analysis can be con- Characterization,” Cold Spring Harbor Aquacide III 1 kg 17852 centrated by several methods including Laboratory Press, New York. CB-Protein Assay™ 1 kit 219468 lyophilization, spin concentrators, dialysis 5. Ornstein, L. (1964) Ann. NY Acad. Sci. Non-Interfering against concentrated polyethylene glycol 121, 321–349. Protein Assay™ 1 kit 488250 6. Davis, B. J. (1964) Ann. NY Acad. Sci. (PEG), and absorption of excess solvent by Tris Base, Ultrol® 100 g 648311 exposure of the dialysis bag containing 121, 404–427. Grade 1 kg sample to dry PEG, Aquacide or gel filtra- 7. Shapiro, A. L. and Maizel, J. V. (1969) Cleland’s Reagent, 1 g 233155 Anal. Biochem. 29, 505–514. tion media such as Sephadex®. Samples con- Reduced (DTT) 5 g 8. Reynolds, J. A. and Tanford, C. (1970) centrated through these methods may be di- Glycerol, Molecular 100 ml 356352 Proc. Natl. Acad. Sci. USA 66, Biology Grade 1 liter alyzed against 50 mM Tris-HCl, pH 6.8 to 1002–1007. Sodium n-Dodecyl Sulfate, remove low molecular weight impurities 9. Laemmli, U. K. (1970) Nature 227, High Purity 25 g 428016 prior to addition of SDS sample buffer. 680–685. 4X SDS Sample Buffer 2 ml 70607-3 Dilute samples, acidic samples and samples 10. Neville, D. M. and Glossmann, H. Perfect Protein™ containing interfering compounds such as (1974) Meth. Enzymol. 56, 58–66. Markers, 15–150 kDa 100 lanes 69149-3 potassium, guanidine hydrochloride, or 11. Schagger, H. and Von Jagow, G. (1987) Perfect Protein™ ionic detergents can be precipitated by Anal. Biochem. 166, 368–379. Markers, 10–225 kDa 100 lanes 69079-3 trichloroacetic acid or acetone to concen- 12. Matsudaira, P. T. and Burgess, R. R. Trail Mix™ Protein (1978) Anal. Biochem. 87, 386–396. Markers 100 lanes 70980-3 trate the proteins and remove contaminants. 13. Patel, D. (1994) in “Gel RAPIDstain™ 1000 ml553215 Protocols for TCA, acetone/methanol, and Electrophoresis,” Rickwood, D. and FASTsilver™ Kit 1 kit 341298 ethanol precipitation are described in refer- Hames, B. D., eds, pp. 3–5, John Wiley ences 4, 15, 17 and 24. A modified acetone & Sons, New York. = Calbiochem brand product inNovations 13 12