Purification and subunit structure of deoxyribonucleic acid-dependent ribonucleic acid polymerase III from the mouse plasmacytoma, MOPC 315.

Class III DNA-dependent RNA polymerases were purified from the mouse plasmacytoma, MOPC 315. RNA polymerases IIIA and IIIB were solubilized from a whole cell extract and resolved by chromatography on DEAE-Sephadex. Chromatography on DEAE-cellulose, DEAE-Sephadex, CM-Sephadex, and phosphocellulose ion exchange resins and sedimentation in sucrose density gradients yielded chromatographically homogeneous Enzymes IIIA and IIIB which were purified approximately 22,000 and 53,000-fold respectively, relative to whole cell extracts. The specific activity of these enzymes was comparable to that reported for other purified eukaryotic RNA polymerases. Sucrose gradient sedimentation analysis suggested a molecular weight of approximately 650,000 for each of the class III enzymes.

In the mouse plasmacytoma, class III RNA polymerases have been shown to synthesize tRNA and 5 S RNA species (14). The cellular levels of solubilized RNA polymerase III activity vary among different cell types and in the same cell type under different physiological conditions (2-4) and may therefore regulate directly the cellular rate of tRNA and -5 S RNA synthesis. Determination of the specific mechanism(s) accounting for these variations in RNA polymerase III activity might provide insights into the regulation of tRNA and 5 S RNA synthesis and cell growth rates.
To investigate these problems, we have chosen the mouse plasmacytoma, MOPC 315,2 a rapidly growing malignant cell. These cells contain high levels of RNA polymerase III (2), probably reflecting a high level of tRNA and 5 S RNA synthesis characteristic of a rapid rate of cellular proliferation (15). This paper reports the purification and subunit structures of the class III enzymes from MOPC 315 cells, which have permitted a structural comparison of homologous class I, II, and III RNA polymerases (16). Evidence is presented that the heterogeneous class III enzymes, designated III, and III., have minor differences in their physical properties and subunit compositions. These studies also provide evidence that fluctuations in the levels of RNA polymerase III activity may, in part, be mediated via changes in enzyme concentration.   (04) or presence (0---0) of 0.5 rig/ml of cY-amanitin; -, ammonium sulfate concentration; ., absorbance at 280 nm. Procedures" (see "First Phosphocellulose Chromatography").
All of the activity was adsorbed to the column and eluted in a single sharp peak with a maximal enzyme concentration (peak tube) of 120,000 units/ ml. A total of 105,000 units of activity were recovered. Those fractions which contained enzyme concentrations in excess of 30,000 units/ml were subjected individually to sucrose gradient sedimentation at an ammonium sulfate concentration of 0.08 M, as described under "Experimental Procedures." A total of 95,000 units were loaded onto the sucrose gradient and the apparent yield of activity in this step was 103%. The sedimentation profile was similar to that observed in experiments described below. The final purification step was adsorption of the enzyme from the peak sucrose gradient fractions to a second phosphocellulose column and elution with a linear salt gradient (Fig. 3).
Overall Purification and Recovery-The purification of the individual class III enzymes after DEAE-Sephadex chromatography is summarized in Table II The apparent specific activities of RNA polymerases IIIA and IIIB in the crude cellular extract are, respectively, 0.014 and 0.0058 units/pg of protein with calf thymus DNA as template (see legend to Table I). Thus RNA polymerases III, and III, are purified approximately 22,000 and 53,000-fold, respectively, relative to whole cell extracts.
As summarized in Table II  and degree of dilutions of enzyme solutions, by the inclusion of bovine serum albumin in the buffers used for CM-Sephadex and the first phosphocellulose chromatography, and by the addition of bovine serum albumin and Nonidet P-40 to sucrose density gradients (17).
Properties-RNA polymerases III, and III, have many similar properties which distinguish them from the corresponding class I and class II enzymes. These include (a) biphasic salt activation profiles with native DNA templates (2); (b) distinct chromatographic behavior on DEAE-cellulose (elution at low ionic strength) uersus DEAE-Sephadex (elution at high ionic strength) (Figs. 1 and 2); (c) sensitivity to high concentrations of Lu-amanitin (50% inhibition at 20 pg/ml) (2); and (d) increased activity (ll-to 1Bfold) with [d(A-T)], as template, relative to native DNA (Figs. 2 and 3).
Thus far only minor differences in the properties of RNA polymerase III, and III, have been detected. As shown above, the enzymes show distinct chromatographic properties on DEAE-Sephadex (Fig. 2). In addition, RNA polymerase III, can be distinguished from RNA polymerase III. by sucrose gradient sedimentation at intermediate ionic strengths (Fig.  4). In the presence of 0.125 M ammonium sulfate, Enzyme III, sediments as a single peak of activity, while Enzyme III, sediments as a double peak of activity (Fig. 4,  both enzymes sediment as single peaks of activity as shown in Panels C and F (Fig. 4). However, at concentrations of ammonium sulfate lower than 0.1 M (Fig. 4, Panels A and D), both enzymes display heterogeneous peaks of activity. These data are consistent with the idea that Enzyme III, aggregates at both low and intermediate ionic strengths, whereas Enzyme III, does so only at low ionic strengths. Although these enzymes have been purified by chromatography on two cation exchange columns and on two strong anion exchange columns prior to sedimentation on sucrose gradients, the possibility remains that a contaminating substance (e.g. a nucleic acid) may be responsible for the characteristic sedimentation properties of RNA polymerases III ,, and III B.
Polyacrylamide Gel Electrophoresis under Denaturing Conditions-Individual phosphocellulose gradient fractions, containing RNA polymerase III, activity, were subjected to electrophoresis in the presence of sodium dodecyl sulfate (Fig.  5). The 10 polypeptides designated IIIa,b,c,d,el,e2,f,g,h, and i, in order of decreasing molecular weight, are regarded as putative subunits based on the following observations. First, the mass of each of these polypeptides is directly proportional to the enzyme activity present in each phosphocellulose gradient fraction (Fig. 5). In contrast, this relationship does not hold for the additional polypeptides which are apparent in the various gradient fractions, with the possible exception of the 24,000-and 22,000-dalton polypeptides which migrate between IIIh and IIIi (see also below). Second, the unlabeled polypeptides in Fig. 8  IIIa and IIIb and by less than 106 in the case of subunits 111~ to i. Molar ratios were normalized to subunit IIIb. Molar ratios for the indicated subunits, measured across phosphocellulose gradient fractions, were fairly constant, varying less than 20% from the average values shown (Fig. 5). Molar ratios for electrophoretic forms III,-1 and III,-2 were obtained from Fig. 7. In some instances (see Fig.  7 (Table III). However, heterogeneity in polypeptide IIIf has also been observed in the present studies, when the enzyme is subjected to electrophoresis in a high resolution polyacrylamide gel slab (Fig. 8). The molar ratios of the individual heterogeneous IIIf polypeptides vary in different enzyme preparations, but their sum is approximately unity. The basis for the heterogeneity in subunit IIIf is not known.

Polyacrylamide
Gel Electrophoresis under Nondenaturing Conditions-Highly purified RNA polymerase III, (phosphocellulose gradient fraction) was analyzed by polyacrylamide gel electrophoresis under nondenaturing conditions. When these gels were stained for protein, two major bands (designated III,-1 and III,-2) were routinely observed, and one minor diffuse band was occasionally detected (Fig. 6). Greater than 95% of the protein stain was associated with these bands. The migration of these bands and their relative intensities were somewhat variable in different experiments, yielding two patterns as shown in Fig. 6 (compare Gels 1 and 2). The cause of this variability is not clear, but samples containing high concentrations of enzyme seemed to yield the pattern shown in Gel 1 (Fig. 6), while less concentrated samples yielded the pattern shown in Gel 2 (Fig. 6). However, other factors such as minor differences in the salt concentration of the samples may contribute to this variability. No activity measurements (see Ref. 17) were attempted on the electrophoretically separated protein bands.

Polyacrylamide
Gel Electrophoresis under Denaturing Conditions Following Electrophoresis under Nondenaturing Conditions-The subunit compositions of electrophoretic forms III,-1 and III,-2 have been determined.
An unstained polyacrylamide gel, containing 6 times more sample protein than Gel 1 (Fig. 6), was divided into l-mm wide slices and the protein in the individual slices was subjected to electrophoresis under denaturing conditions as described previously (17). Panel A in Fig. 7 shows the subunit composition of the phosphocellulose enzyme prior to electrophoresis under nondenaturing conditions. Panels B and C in Fig. 7 show the polypeptide compositions of electrophoretic forms III,-1 and III A-2, respectively. The subunit compositions of electrophoretic forms III,-1 and III,-2 were identical and, except for a shoulder of staining material migrating slightly faster than subunit IIIf, they were the same as that of the phosphocellulose enzyme. Subunit molecular weights and molar ratios are summarized in Table III The enzyme preparation analyzed in this experiment (Fig. 7) also appeared to contain two polypeptide components between polypeptides IIIh and IIIi (cf. Fig. 5). Although the polypeptide bands were diffuse (Panels B and C), these two components appeared to remain associated with electrophoretic forms III,-1 and III,-2. As noted above, however, these components are not readily detected in all RNA polymerase III preparations.
A small amount of protein was recovered from slices corresponding to the minor diffuse band in Gel 1 of Fig. 6 Fig. 6. This gel was run in parallel with Gel 1 in Fig. 6. After electrophoresis, the gel was sliced and each slice was subsequently subjected to electrophoresis on 10% polyacrylamide 'gels containing sodium dodecyl sulfate as described previously (17). Due to the large amount of protein used, bands III,-1 and III,-2 were somewhat broadened.
Nevertheless, two peaks of protein, corresponding in migration position to the two electrophoretic forms of III,,, were clearly separated by a gel slice containing negligible amounts of protein.' Panel A portrays the characteristic subunit pattern for RNA polymerase III,, prior to electrophoresis under nondenaturing conditions (see Fig. 5). Panels B and C portray characteristic subunit patterns for the two major bands, designated III,-1 and III,-2, respectively, in Fig. 6. These results were obtained from three polyacrylamide gels which were run in parallel.
Enzyme subunits, the molecular weights and molar ratios of which are summarized in Table III (Fig. 6). The subunit compositions of electrophoretic forms IIIR-1 and III,-2 were not investigated.
Electrophoresis of individual phosphocellulose gradient fractions containing RNA polymerase III, activity revealed a subunit pattern similar to the one obtained with Enzyme III,.3 To resolve minor differences in the subunit compositions between RNA polymerases III, and IIIs, these enzymes were subjected to electrophoresis individually and in combination on a 25-cm polyacrylamide gel slab under denaturing conditions (Fig. 8). These data clearly illustrate that RNA polymerases III, and III, differ only in one subunit. Except for a 32,000 dalton subunit IIIgA which is unique to Enzyme III, and a 33,000 dalton subunit IIIgH which is unique to Enzyme IIIB, the FIG subunit molecular weights in Enzyme IIIH appear identical with those in Enzyme III, (Table III). The molar ratios of the subunits in Enzyme III" are similar to those for the analogous subunits in Enzyme III,.3 The electrophoretic system used here (Fig. 8) resolves polypeptide IIIf (Fig. 5) into two components as observed previously (Ref. 16, see also Footnote 4). This electrophoretic system also resolves subunit IIIi into several polypeptides. Although subunit IIIi appears to be present in stoichiometric excess after electrophoresis under nondenaturing conditions (Fig. 7), it is not clear which of the low molecular weight polypeptides observed in Fig. 8 remain associated with the enzyme under these conditions. DISCUSSION Purification, Structure, and Heterogeneity of RNA Polymerases ZZZ* and III,-Two chromatographic forms of RNA polymerase III are present in the mouse plasmacytoma MOPC 315. Previous studies have shown the presence of RNA polymerase III activity in both cytoplasmic and nuclear fractions following cellular disruption and fractionation (2,5,8,9). We have, therefore, purified the MOPC class III enzymes from whole cells in order to study the total cellular population of these molecules. RNA polymerases III, and III, were resolved by chromatography on DEAE-Sephadex and purified by ion exchange chromatography and sucrose gradient sedimentation. Relative to whole cell extracts the overall purifications were 22,000-and 53,000-fold, respectively, for enzymes III, and III,.
Chromatographically homogeneous RNA polymerases III, and IIIR each contains at least 10 putative subunits designated IIIa,b,c,d,el,e2,f,g,h, and i. That these RNA polymerase III-associated polypeptides represent enzyme subunits is suggested by the following observations: (a) the ratio of the amount of each polypeptide to the amount of enzyme activity is approximately constant for individual phosphocellulose gradient fractions; (b) the molar ratios of these polypeptides are approximately unity, with the exception of polypeptide IIIi which is present in a higher but constant molar ratio; (c) the molecular weight of RNA polymerase III calculated from the molecular weights and molar ratios of the individual polypeptides (695,000) is compatible with that estimated from sucrose gradient sedimentation (650,000); (d) polypeptides IIIa to i co-sediment with RNA polymerase III activity upon sucrose density gradient sedimentation and they remain associated with the major protein bands when RNA polymerase III is subjected to electrophoresis under nondenaturing conditions; and (e) the murine and amphibian class III enzymes contain analogous polypeptides of the same or similar size (16), even though these enzymes are from grossly different cell types.
The subunit compositions of RNA polymerases III, and III, are very similar, differing only in subunit IIIg which is slightly smaller in Enzyme III, (IIIg,-32,000 daltons) than in Enzyme III, (IIIgR-33,000 daltons). As reported previously (2), no evidence for interconversion of these enzyme forms could be found since they maintain their distinctive properties upon rechromatography on DEAE-Sephadex. The presence of serine protease inhibitors (i.e. PMSF and iPr,P-F) during enzyme isolation and purification did not alter the subunit patterns of the purified enzymes.3 The general similarity between the subunit structures of Enzymes III, and III, correlates with the similarities in their catalytic properties and a-amanitin sensitivities. However, the minor structural difference between subunits IIIg, and IIIg, may be responsible for differences in