BetterScholar BetterScholar
Title Claps Level Year L/Y
RNA molecular weight determinations by gel electrophoresis under denaturing conditions, a critical reexamination.
H. Lehrach, D. Diamond, J. Wozney, H. Boedtker
RNA molecular weight measurements were carried out by gel electrophoresis under four different denaturing conditions including 99% formamide, 10 mM methyl mercury, 2.2 M formaldehyde, and 6 M urea at pH 3.8. Electrophoresis a t a series of gel conce…
RNA molecular weight measurements were carried out by gel electrophoresis under four different denaturing conditions including 99% formamide, 10 mM methyl mercury, 2.2 M formaldehyde, and 6 M urea at pH 3.8. Electrophoresis a t a series of gel concentrations and at least two different voltage gradients resulted in some RNA species exhibiting apparent molecular weights that vary with both gel concentration and voltage gradient. Three different deviations from the requirement for hydrodynamically equivalent conformations were observed: (1) deformation of the random coil structure of very large RNAs at moderately high gel concentrations and voltage gradients resulting, in extreme cases, in a molecular weight independent migration of RNA molecules; (2) incomplete denaturation of RNA molecules with very GC rich helical regions; and (3) varying charge/mass ratio due to G e l electrophoresis has become the major analytical procedure for characterizing charged macromolecules both because of the high resolution it provides and the relative simplicity of the technology it requires. The applicability of this procedure to molecular weight determination of nucleic acids and NaDodS04-protein complexes is based both on the fact that these macromolecules are polymers with a constant charge-mass ratio and hydrodynamically equivalent conformations, and on the linearity of the empirical log molecular weight-mobility relation used to determine molecular weights from mobilities. When these assumptions are fulfilled, molecular weights can be measured in a single experiment using standards of known molecular weight. Moreover, since RNA molecules do not have hydrodynamically equivalent conformations in aqueous solutions (Boedtker, 1968; Groot et al., 1970; MacLeod, 1975), several methods were developed in which R N A molecular weights could be determined by gel electrophoresis under denaturing conditions. Denaturing conditions first used included reaction with formaldehyde (Boedtker, 1971), 8 M urea at 60 'C (Rejinders et al., 1973) and 99% formamide at room temperature (Pinder et al., 1974), the latter being by far the most widely used. In addition to the denaturing gel systems first developed, three others have been reported more recently: 6 M urea at pH 3.5 run at 2 "C (Rosen et ai., 1975), 5 mM methyl mercury run at room temperature (Bailey and Davidson, 1976), and 99% formamide run at 45-55 'C (Spohr et al., 1976). This proliferation of denaturing gel systems arose because it was recognized that some of the denaturants first used were not able to denature very GC rich RNA and could not be used in the dilute agarose gels required t From the Department of Biochemistry and Molecular Biology, Harvard University, Cambridge, Massachusetts 02 138. Receiued March 30, 1977. H. Lehrach was the recipient of a postdoctoral fellowship from the Jane Coffin Childs Memorial Fund for Medical Research. John M. Wozney was awarded a Camille and Henry Dreyfus Foundation Summer Research Fellowship. This investigation was supported by National Institutes of Health Grant HD-01229. differential protonation at pH 3.8. Reliable molecular weight measurements of RNA molecules as large as 4.0 X lo6 containing G C rich helical regions could only be made on dilute (0.5-1.0%) agarose gels after reaction with either 2.2 M formaldehyde or 10 mM methyl mercury hydroxide. A theoretical justification for the use of the empirical log molecular weight-mobility relation is presented. It is also demonstrated that the gel electrophoretic behavior of a homologous series of random coils can be approximated by that of a series of spheres with radii proportional to the square root of radius of gyration of a random coil. Consequently, molecular weight determinations of denatured RNAs, especially those obtained by extrapolation, are more reliable if the square root of the molecular weight is plotted vs. log mobility. for molecular weight determinations of very large RNA molecules. We report here a comparison of the behavior of high molecular weight R N A molecules on four different denaturing gels: polyacrylamide in 99% formamide, agarose in 10 mM methyl mercury hydroxide, 2.2 M formaldehyde, and 6 M urea, pH 3.8. In addition, we offer a rationale for the use of the empirical log molecular weight-mobility relation and suggest an alternate method of obtaining RNA molecular weights from mobilities based on the finding that the migration of denatured RNA molecules depends on the square root of the radius of gyration, or the fourth root of the molecular weight. Materials and Methods RNA Samples. TMV and chicken ribosomal RNA were prepared as described previously (Boedtker, 1960; Boedtker et al., 1973). E. coli 23s and 16s rRNA was prepared by phenol-CHC13 extraction (Perry et al., 1972). followed by fractionation on linear 5-20% sucrose gradients. Mouse 28s rRNA was donated by Jesse F. Scott of the Harvard Medical School. Sindbis virus, a gift from Michelene McCarthy of the Department of Biochemistry and Molecular Biology, was extracted with phenol-CHC13 (Perry et al., 1972). Polyacrylamide Gel Electrophoresis in 99% Formamide. To achieve reproducible mobilities, pure reasonably dry deionized formamide is essential. 99% formamide (Eastman Chemical Co.) was deionized following the procedure of Pinder et al. (1 974) by stirring with 40 g /L mixed bed resin (Bio-Rad AG501-X8, 20-50 mesh). After about 5 h, the conductivity should decrease to about 70 pmho. We have found, however, that some batches of formamide are not deionized under these conditions even when more resin is used. Such formamide is unsuitable for gel electrophoresis because irreproducible polymerization of gels occurs. The deionized formamide was filtered through a sintered-glass filter and then distilled under B I O C H E M I S T R Y , V O L . 1 6 , N O . 2 1 , 1 9 7 7 4743 L E H R A C H E T A L electrophoresis buffer (E buffer): 0.05 M boric acid, 0.005 M NazB407.10H~0, 0.01 M sodium sulfate, and 0.001 M Na3EDTA, pH 8.2, without CH3HgOH. The suspension was heated for 5 min in an autoclave and diluted to the appropriate concentration with hot E buffer. CH3HgOH was added to the hot buffered agarose by syringe pipet and stirred rapidly at 60 “C. The gels were poured immediately using 2.5 mL per tube. Before samples were applied, the top of the gel was sliced off with a razor blade to provide a flat surface. RNA samples were prepared by dissolving the RNA in E buffer containing 5 or 10 mM methyl mercury hydroxide as indicated. One-half volume of a 1:l mixture of glycerol-HzO solution containing 0.004% bromophenol blue was added and the sample applied to the gel. Electrophoresis was carried out for various times as indicated at 2.5 or 3 mA per tube, at room temperature inside the closed hood. The latter was clearly designated as being hazardous both because of high voltage and CH3HgOH. All operations involving methyl mercury hydroxide were carried out in the hood, with the operator wearing gloves. The tops of the gels, leftover agarose, used gels, and any material contaminated with CH3HgOH (gloves, paper, disposable items) were placed in a disposable plastic bag and stored in the hood. CH3HgOH waste and contaminated items were disposed of every 3 months by the Department of Chemistry’s hazardous chemical disposal service, the Radiac Corp. The operator was monitored for CH3HgOH accumulation every 6 months by the Harvard Health Services. Agarose Gel Electrophoresis in 2.2 M Formaldehyde. RNA samples were heated in 2.2 M formaldehyde (prepared from 37% Mallinckrodt formaldehyde) in 50% formamide, 0.01 8 M NazHP04-0.002 M NaHzPO4, for 5 min at 60 “C. Agarose gels were prepared by heating a “3%” agarose-water suspension in the autoclave for 5 min and diluting wi th either 1 volume of 4.4 M formaldehyde in 0.036 M NazHP04-0.004 M NaH2P04 to make 1.5% gels or with 2 volumes of the latter plus 1 volume of water to make 0.75% gels. The gels were poured immediately using 2.5 mL per tube. Electrophoresis was carried out at 2 mA per tube at room temperature for the times indicated. The electrophoresis buffer was 2.2 M formaldehyde-0.018 M Na2HP04-0.002 M NaH2P04. Agarose Gel Electrophoresis in 6 M Urea at pH 3.8. Electrophoresis was performed following the procedure described by Rosen et al. ( I 975) with the following modifications. Three times agarose (usually 3 g/lOO mL of HzO) was heated in the autoclave and then diluted with 2 volumes of warm I .5X buffer to give the final concentration of agarose and 6.0 M urea, 0.025 M citric acid (pH 3.8), heated at 65 “ C for 20 s, and fast cooled. One-half volume of the glycerol dye was added as described above. Electrophoresis was carried out for 12 h at 0.38 mA per gel at 4 “C. Staining Gels with Ethidium Bromide. Agarose and polyacrylamide gels were stained overnight in 1 yg/mL ethidium bromide in 0.1 M ammonium acetate (Bailey and Davidson, 1976). After staining, the gels were photographed under short-wave U V light with a Polaroid M P 3 Land camera and high-speed type 57 film (Polaroid Corp.) using a yellow f i l ter.
Published in Biochemistry
111
 11 1977