yeast rna polymerase c and its subunits

Vol. 260, No. 28, Issue of December 5, pp. 15304-15310,1985 Printed in U.S.A.

THE JOURNAL OF BIOLOGICAL CHEMISTRY 0 1985 by The American Society of Biological Chemists, Inc.

Yeast RNA Polymerase C and Its Subunits SPECIFIC ANTIBODIES AS STRUCTURAL AND FUNCTIONAL PROBES*

(Received for publication, April 3, 1985)

Janine Huet, Michel Riva, Andre Sentenac, and Pierre Fromageot From the Departement de Biologie, Service de Biochimie, Centre d’Etudes Nucliaires de Saclay, 91291 Gif-sui-Yvette Cedex, France

Yeast RNA polymerase C purified by a simple large scale method was resolved into multiple components by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Specific antibodies directed against each polypeptide chain were prepared in rabbits and used as structural and functional probes. With minor excep- tions, each antibody recognized specifically the corresponding polypeptide by blot-immunodetection. Cross- reactions with purified RNA polymerases A and B confirmed our previous description of the subunits shared by the three nuclear RNA polymerases. Immu- noadsorption of RNA polymerase C at different stages of purification using antibodies to subunits Cleo and ClZ8 yielded the same collection of polypeptides as found in the purified enzyme: C~G,,, C128, CS2, c63, C,O, C37, Ca4, Cal, CZ,, c261 C23, C19, c14.6, C12.5, and G o .

Subunit-specific antibodies were used to probe the activity of RNA polymerase C in a specific, reconstituted transcription system as well as on a nonspecific template. Transcription of the tRNA2” gene in vitro was inhibited when RNA polymerase C was preincubated with antibodies directed to ClZs, cS2, c63, C34, C23, or CI9. Antibodies to c82, c63, and C34 were much less inhibitory in the nonspecific assay. Inhibition by anti-C128 or anti-C23 was relieved by preincubation of enzyme C with plasmid DNA prior to antibody addition. These results are discussed in terms of the participation of these polypeptides to the active enzyme molecule, and of their possible role in DNA binding or transcription factor recognition.

In eukaryotes, the three forms of nuclear RNA polymerase transcribe different classes of genes. RNA polymerase C(II1) synthesizes small RNA species: tRNAs, 5 S RNA, and small viral RNAs (1, 2). The regulation of class C genes is unique in that the control regions (promoter) are intragenic and recognized by regulatory proteins (transcription factors) (for a review see Ref. 1). Transcription factor IIIA specifically binds to the internal promoter of the 5 S RNA gene (3). Similarly, a factor in yeast binds to the split promoter of yeast tRNA genes as a preliminary step of gene recognition prior to transcription (4, 5 ) . RNA polymerase C has to recognize these factor-DNA complexes and additional factorb) required for gene activation.

Isolation of RNA polymerase C from a variety of organisms revealed an exceptional molecular complexity (6-14). The number and size range of most polypeptides, in all eukaryotic ~~ ~ ~

* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

cells, appear to be fairly conserved with two large subunits (MI around 160,000 and 130,000) and a complex collection of 10-12 smaller components (Mr between 90,000 and 10,000). One characteristic polypeptide around 80-90 kDa is found in all cases, except in insects (7, 10). The role of these polypeptide components in transcription, DNA binding, and RNA polymerase-factor interaction has to be established. There are reports on enzyme C variants having lost some polypeptides upon chromatography (14) or gel electrophoresis (8, 9), or having small components of slightly different size (6). These observations have not yet been exploited to compare the activity of these variants in the specific transcription of class C genes in uitro.

Antibodies to individual subunits of yeast RNA polymerases A(1) and B(I1) have proved to be valuable tools to clarify the molecular organization and structural relationship of the three forms of yeast enzyme (15-18). Remarkably, RNA polymerase C was found to share five subunits with enzyme A, three of which constitute a core of common subunits also present in enzyme B. The development of reconstituted systems for the specific transcription of yeast class C genes (19, 20), as well as the prospect of cloning the polymerase structural genes has encouraged us to extend our immunological approach and to prepare a collection of antibodies directed to the various polypeptide components of yeast RNA polymerase C. These subunit-specific antibodies have been used as structural and functional probes. The results presented here pro- vide a new insight on the structure of yeast RNA polymerase C and suggest a role for some enzyme components in DNA binding and RNA polymerase-factor interaction.

EXPERIMENTAL PROCEDURES

RNA Polymerases, Subunits, and Antibodies-RNA polymerase C was purified from Saccharomyces cereuisiae cells (pep4-3) by a procedure derived in part from that of Buhler et al. (21) and Wandzilak and Benson (22). Briefly, the cells (300 g) were broken in a Manton Gaulin homogenizer in a high salt buffer, and the high speed supernatant was mixed with phosphocellulose PI1 (Whatman) at reduced salt concentration (0.05 M ammonium sulfate). Optimal yield of enzyme C was obtained with 1 g of phosphocellulose, wet weight, for every 175 Azso units of supernatant. RNA polymerases A and C were eluted batchwise at 0.4 M ammonium sulfate, concentrated by ammonium sulfate precipitation and chromatographed on a DEAE- Sephadex column (25 X 2 cm) using a 400-ml gradient of ammonium sulfate from 0.1 to 0.6 M. Enzyme C activity eluted at around 0.3- 0.34 M salt was pooled and adsorbed immediately on a small column (2 X 0.8 cm) of heparin-Sephadex (or agarose) equilibrated in 0.2 M Tris-HC1, pH 8.3, 10 mM 2-mercaptoethanol, 0.5 mM EDTA, 25% glycerol, and 0.2 M ammonium sulfate. After a washing step at 0.28 M salt, enzyme C activity was eluted stepwise at 0.6 M salt, concentrated by dialysis against 70% glycerol and stored at -70 “C. Enzyme specific activity was 12,000 nmol of UMP incorporated per hour/mg of protein on a poly[d(A-T)] template (with 100 mM ammonium sulfate). Having used this procedure for several years we have noted a variation in the salt elution of enzyme C with different batches of

15304

Yeast RNA Polymerase C: Subunits and Specific Antibodies 15305

DEAE-Sephadex (from 0.28 to 0.35 M salt) or heparin-Sepharose 4B or -agarose, or with the same batch of heparin upon repeated cycling of the column. A DEAE-cellulose chromatography step may occasion- ally be required to complete the purification. Routinely, about 2 mg of enzyme C were obtained from 300 g of cells.

RNA polymerase A derived from the DEAE-Sephadex column was further purified by sedimentation and phosphocellulose chromatography as described (23). RNA polymerase B was purified by the standard procedure (24).

The polypeptide components of RNA polymerase C were isolated by preparative electrophoresis with SDS' and injected into rabbits as described for the subunits of enzyme A or B (15, 18). A total of 12 mg of enzyme C were used for antibody production. Immunoglobulins were purified and stored as described (15).

Immunoadsorption of RNA Polymerase C from Crude Fractions- RNA polymerase C was immunoadsorbed from three different fractions: 1) a yeast crude extract from about 25 g of cells was centrifuged at high speed. The soluble proteins were concentrated by ammonium sulfate precipitation and loaded on four 10-30% glycerol gradients containing 0.3 M ammonium sulfate. After 18 h of sedimentation a t 40,000 rpm in the Beckman SW 41 rotor, the fractions containing the peak of RNA polymerase activity (A, B, and C) were pooled, and the proteins were concentrated by ammonium sulfate precipitation. 2) Fifteen ml (A280 = 3) of the protein fraction eluted batchwise from phosphocellulose at 0.4 M salt (see enzyme C preparation), which contained RNA polymerases A and C, were concentrated by ammonium sulfate precipitation. 3) The protein fraction (4 ml, 100 pg of protein/ml) with maximum activity from the DEAE-Sephadex column (see enzyme C preparation) was concentrated as above. In each case the pellet was dissolved in 0.9 ml of 10 mM Tris-HC1, pH 8.2, 0.15 M NaCl, and dialyzed against the same buffer. Staphylococcus aureus cells (IgGsorb; The Enzyme Center, Inc.) were treated for 30 min at 95 "C in 10 mM Tris-HC1, pH 8, 10% 2-mercaptoethanol, 2% SDS to remove soluble proteins. In order to remove proteins that might bind nonspecifically to the S. aureus cells, 50 pl of a 10% cell suspension in 10 mM Tris-HC1, pH 8.2, 0.15 M NaCl, bovine serum albumin (1 mg/ml) were added to each protein sample and incubated for 20 min at 4°C. The cells were discarded by centrifugation, and the protein supernatant was supplemented with 20 pg each of anti- C160 and anti-CI28 IgG and 0.2 mM phenylmethylsulfonyl fluoride. The immune complexes formed after overnight incubation at 4 "C were adsorbed on S. aureus cells (50 pl of a 10% cell suspension) for 30 min at 4 "C. The cell pellet was washed five times with 1 ml of buffer containing 10 mM Tris-HC1, pH 8.2, 0.15 M NaC1, 0.01% SDS, and 2% Triton X-100 and once with 10 mM Tris-HC1, pH 8.2. Immunoadsorbed proteins were eluted by boiling the cell pellet for 5 min in 30 p1 of Laemmli's sample buffer with 2% SDS and subjected to electrophoresis on a 13% polyacrylamide gel with SDS (25). After staining with Coomassie Blue, the gel was scanned with an LKB Ultroscan laser densitometer.

Antibody Specificity and Effect on Enzyme Activity-The affinity spectrum of antibodies was analyzed by the immunoblotting technique (18) or spot-immunodetection (16) as previously described.

The effect of antibodies on enzyme activity was measured after preincubation of purified IgG with 1 pg of RNA polymerase C for 1 h a t 30 "C in 50 pl of buffer containing 10 mM Tris-HC1, pH 8, 2 mM NaC1,20% glycerol, and 0.1 mM EDTA. The final amount of IgG was kept constant (90 pg) by complementing with y-globulins from a normal rabbit. Residual enzyme activity was assayed for 20 min in a standard transcription mixture with poly[d(A-T)] (3 pg) as template (0.1-ml final volume) (23).

Effect of antibodies on specific transcription was measured after preincubation, as above, of enzyme C (0.05 pg) in 20 pl with purified IgG, as indicated in Fig. 7, for 1 h at 30 "C. The amount of IgG was adjusted to 8 pg with control y-globulins. Transcription of pY20 plasmid DNA which harbors the tRNA$'" gene was then run for 1 h in an RNA polymerase C-dependent reconstituted system (40 p1 final volume) (20). The synthesized pre-tRNA was isolated by gel electrophoresis and quantified by densitometry of the autoradiograph using a LKB Ultroscan laser densitometer.

RESULTS

Yeast RNA Polymerase C-Of the three forms of yeast nuclear RNA polymerase, enzyme C(II1) is the least well

' The abbreviation used is: SDS, sodium dodecyl sulfate.

A B C ", - .._- .

160 128 82 53

40 34 31 - - -27 23 19

14-5

FIG. 1. Polypeptide content of RNA polymerases as seen by SDS-polyacrylamide gel electrophoresis. RNA polymerase C (11 pg), purified as described under "Experimental Procedures," was analyzed on a 13% polyacrylamide slab gel with SDS according to Laemmli (25), together with a sample of RNA polymerase A or B (12 pg each). Polypeptide chains were revealed by Coomassie Blue staining. Molecular weights (X of enzyme C components are indicated.

characterized, structurally and functionally. Since the original report of Valenzuela et al. (8), alternative methods were developed, mostly in the same laboratory, to improve the preparation of this enzyme, which has been generally difficult to purify (13, 22, 26, 27). We have essentially followed, for the initial steps, the strategy of Buhler et al. (21) and Wand- zilak and Benson (22), with some modifications that increased the speed and yield of the enzyme preparation. Briefly, RNA polymerases A and C were purified from a whole cell extract by batchwise adsorption and elution from phosphocellulose and separated by chromatography on DEAE-Sephadex. At this stage, enzyme C is about 30% pure based on its polypeptide content, but fairly unstable. Final purification and concentration were achieved by a rapid salt step-elution from heparin-agarose. This procedure avoids all lengthy dialysis or sedimentation steps and is completed in 3 days. RNA polymerases A and B were conveniently purified from corresponding side fractions using our previously described procedures.

Fig. 1 shows the polypeptide composition of the three enzymes A, B, and C after electrophoresis on a polyacrylamide slab gel in the presence of sodium dodecyl sulfate. The subunit profile of the enzyme C preparation was: 160,000, 128,000, 82,000, 53,000, 40,000, 37,000, 34,000, 31,000, 27,000, 25,000, 23,000, 19,500, 19,000, and 14,500 daltons. Two additional polypeptides around 12,000 and 10,000 daltons are not seen in Fig. 1. The characteristic subunit C8*' (82,000 daltons) was present, as well as the polypeptides Css, C37, C34, and C31, which have sometimes been considered as contaminants (13, 26-28). Polypeptide C53, which often gives a double band (as in Fig. 1) or a fuzzy set of bands, is highly susceptible to proteolysis and for this reason was not found consistently by previous authors (see Fig. 1 in Ref. 8). Polypeptide CIS migrated slightly ahead of a more intense protein band of 19,500

For the nomenclature of RNA polymerase subunits, see Ref. 15.

15306 Yeast RNA Polymerase C: Subunits and Specific Antibodies

daltons, which we initially thought to be a contaminant. This point will be discussed later.

When analyzed by electrophoresis under nondenaturing conditions on a 5% polyacrylamide gel, enzyme C, assayed with poly[d(A-T)] as template, migrated as a single peak at the level of the major protein band representing more than 95% of the Coomassie Blue staining material (results not shown). RNA polymerase C had the slowest migration rate of the three nuclear enzymes.

Enzyme Subunits and Specific Antibodies-The polypeptide components of RNA polymerase C were isolated by preparative electrophoresis in a polyacrylamide gel with SDS. The components of RNA polymerase A and B prepared in this way retained antigenic determinants present in the native enzymes (15,16,18). Fig. 2 shows the electrophoretic analysis of the enzyme C components used to raise antibodies in rabbits. The double bands at the level of CS3 and C19 could not be resolved and were injected as such into the animal.

The affinity spectrum of the antibody preparations was determined by the protein blotting technique. RNA polymerase C polypeptides were separated by electrophoresis and transferred to a nitrocellulose membrane, and strips of membrane were incubated with the different purified immunoglobulins used at the same concentration (20 pg/ml). Immune complexes were revealed with '251-labeled protein A (Fig. 3). Antibodies directed to native enzyme C reacted with all the enzyme subunits (except C27, which is often transferred errat- ically (18)). This experiment served to locate the different enzyme components. When incubated with RNA polymerase A components, on another membrane strip, the same antibodies reacted with the subunits shared by the two enzymes: A ~ o , A27, A23, A19, and A14.5. Every antibody preparation directed at a different subunit gave a strong binding signal corresponding specifically to the injected antigen (Fig. 3). Minor side reactions were observed in some cases with unrelated subunits. These small cross-reactions were generally with the nearest neighbor subunit of higher molecular weight,

. -L l r

FIG. 2. Separation of RNA polymerase C subunits by polyacrylamide gel electrophoresis. Subunits of RNA polymerase C (2 mg) were isolated by preparative gel electrophoresis as described earlier (15, 18). Before each injection series, samples were subjected to analytical electrophoresis on SDS-polyacrylamide gel together with purified RNA polymerases A and C. The gel was then stained with Coomassie Brilliant Blue. In the figure, the subunit preparations are identified by the polypeptide molecular weight ( x ~ O - ~ ) .

FIG. 3. Specificity of antibodies analyzed by blot-immunodetection. Purified RNA polymerase C (100 pg) was subjected to electrophoresis in a SDS-polyacrylamide slab gel, and its subunits were transferred by diffusion to a nitrocellulose membrane. Individual strips of membrane were exposed to the different IgG preparations (20 pglml), and the bound antibodies were revealed with '251-labeled protein A (18). Antibodies to native enzymes C or A were used to locate the different enzyme C subunits (pol C strip) and to reveal the subunits common with enzyme A (pol A strip). The specific antibodies are identified by the name of the polymerase or subunit used as antigen.

while the reverse situation was never observed. Thus, these side reactions were probably due to a small contamination of the isolated subunit with proteolytic products from the neigh- boring higher molecular weight subunit. This was clearly the case with the very immunogenic subunit CS3, Antibodies to that subunit revealed a trailing of cross-reacting material which had remained undetected by Coomassie Blue staining. This analysis of the specificity of the antibodies makes very unlikely the possibility that some components of RNA polymerase C were related. A similar conclusion has been reached after immunological analysis of the components of RNA polymerase A and B (15, 18).

Co-immunoadsorption of RNA Polymerase C Components- The enormous complexity of RNA polymerase C with its 15 distinct polypeptides casts some doubt on the subunit status of all these components. As an additional step to confirm the subunit structure of this enzyme, we have used subunit- directed antibodies to analyze the polypeptide chains which remain firmly associated to the enzyme. Antibodies to the large subunits Cleo and cl26 were reacted with different enzyme C fractions at different stages of purification and adsorbed to s. aureus cells. After a thorough washing of the pellet, the immunoadsorbed polypeptides were solubilized with sodium dodecyl sulfate and analyzed by electrophoresis. As seen in Fig. 4A, the two large subunits c160 and cl26 were selectively adsorbed, together with the complete collection of polypeptides found in the purified enzyme. The pattern of immunoadsorbed polypeptides was the same whether the an-

Yeast RNA Polymerase C: Subunits and Specific Antibodies 15307

A

I 0 j j I! i

FIG. 4. Immunoadsorption of RNA polymerase C from various protein fractions. A mixture of antibodies to subunits Cleo and C128 was used to immunopurify RNA polymerase C from three different enzyme fractions at different stages of purity as described under "Experimental Procedures." Left, immunoadsorbed proteins were analyzed by electrophoresis with SDS and stained with Coomassie Blue. A, control purified RNA polymerase C; B, glycerol gradient fraction; C, phosphocellulose fraction; D, DEAE-Sephadex fraction. Right, densitometer tracings of control purified enzyme C (A) and of the glycerol gradient immunopurified proteins (B) .

tibodies were incubated with a crude enzyme fraction (a phosphocellulose batch adsorption step or a crude glycerol gradient fraction), or with the partially purified DEAE-Seph- adex fraction. The selectivity of the antibodies was attested to by the absence of the large polypeptides characteristic of RNA polymerase A which was present in the crude fractions. This experiment indicated that all the polypeptides found in purified enzyme C were indeed firmly bound to the enzyme molecule and were not independent contaminants. This in- formation was particularly important concerning the polypeptide chains whose subunit status was uncertain, like CE2, C53, C37, C34, c31, and c25. The presence of C53 could not be established as firmly because this polypeptide migrated close to the immunoglobulin heavy chains. (The light chains are heterogeneous in size and contribute to the background between c27 and Clg.)

The stoichiometry of the polypeptides in the purified enzyme and in the immunopurified fractions can be compared in the densitometer scans shown in Fig. 4B. Most polypeptide components of the purified enzyme were in nearly equimolar amount except for C53, which is extremely sensitive to proteolysis (unless one considers the whole heterogeneous peak as being C53), C37 (0.5), C27 (this common subunit has a stoichiometry close to 2 in all three RNA polymerases), and C25 (0.4). The protein band of 19,000 daltons contains the subunit C19, common to enzymes A and C, and an additional polypeptide with a stoichiometry of about 2 (see Fig. 1, lane c).

Remarkably, not only were the different polypeptides found in RNA polymerase C selectively retained on IgG S. aureus cells, but their relative stoichiometry was also maintained. Note that the low stoichiometry of Cs7 and c25 suggests some heterogeneity of RNA polymerase C.

Binding of Specific Antibodies to RNA Polymerases A, B, and C-A simple spot-immunodetection method was used to investigate the binding of subunit-directed antibodies to yeast RNA polymerases (16). As shown in Fig. 5, native enzyme C spotted on nitrocellulose filters gave a strong binding signal with the whole collection of antibodies, indicating that all the polypeptides of the filter-bound enzyme were accessible. The presence of subunits shared by yeast RNA polymerases was revealed using antibodies against the subunits of A and B enzymes (16). These results have now been extended using antibodies to enzyme C components. A strong binding signal was observed on the three enzymes with antibodies against C27, C23, and c14.5, which constitute the core of common subunits (29). Anti-Clo and anti-Clg recognized enzymes A and C to the same extent. In addition to these cross-reactions due to shared components, cross-reactions were also observed with antibodies against CS4 and Clo. The binding of anti-C34 to enzyme A was explored in more detail by the immunoblot technique (results not shown) and found in fact to be due to antibodies recognizing the common subunit AC40. This contamination is apparent in Fig. 3. The cross-reaction obtained with anti-Clo was not analyzed further because this component is not well characterized. The small but discrete cross- reaction on enzymes A and B with anti-Clso and anti-ClzE was again in favor of the conservation of a few determinants in the large polypeptides of the three enzymes (29).

Effect of Antibodies on Specific Versus Nonspecific Tran- scription-Antibodies to enzyme C components were used to probe the activity of enzyme C on a nonspecific template as well as on a cloned class C gene in a reconstituted transcription system. The effect of antibody binding on the general transcriptive ability of enzyme C was assayed with poly[d(A- T)] as template (Fig. 6). Based on their inhibition capacity, the antibody preparations ranked in the following order: C23, C19 > c128, Csz, C53, C40 > C34r c3l >> c160, CZ7, CI4.,. Anti-Clo were not tested. Antibodies to the common subunits were very inhibitory (c23, C19, and to a lesser extent c40) or not at all (C,,, c14.5). Similar levels of inhibition of enzyme C had been observed previously using antibodies to these common subunits originating from RNA polymerase A or B (18). Whatever the enzyme origin of these polypeptides, the antibodies were probably directed to the same set of antigenic sites. The inhibition of poly[r(A-U)] synthesis by antibodies to c128, c82, C53, c40, c23, or C19 strongly suggested that these polypeptides were part of the enzyme molecule. The conclusion was tess clear for C34 and c31 in this type of experiment since the

C

B

A FIG. 5. Binding of specific antibodies to yeast RNA polym-

erases A, B, and C. Purified RNA polymerase A, B, or C (about 0.5 pg) were spotted as indicated onto nitrocellulose filters and exposed to antibodies to the isolated subunits of enzyme C. Immune complexes were revealed by lZ5I-labeled protein A and the filters subjected to autoradiography. Antibodies are identified as in Fig. 3.

15308 Yeast RNA Polymerase C: Subunits and Specific Antibodies

O% 0 50 100

ANTI- RNA POLYMERASE C SUBUNITS

IgGWg) FIG. 6. Inhibition of poly[d(A-T)] transcription by subunit-

directed antibodies. RNA polymerase C was preincubated with varying amounts of purified IgG, as described under "Experimental Procedures," and residual enzyme activity was measured in a standard transcription mixture with poly[d(A-T)] as template. Results are given as per cent of control UMP incorporation (1.2 nmol) in the presence of control y-globulins. The different antibodies are identified as in Fig. 3.

corresponding antibodies were poorly inhibitory (20% inhibition at 300 pg/ml), even with a large excess of immunoglobulins (up to 1 mg/ml). Concerning the large subunits, whereas anti-C12a IgG was clearly inhibitory, anti-Cleo had practically no effect, even at very high concentrations. In previous work we also found that anti-B150 was poorly inhibitory (18).

In view of the molecular complexity of RNA polymerase C, it is likely that some enzyme components play a role in the recognition of regulatory components or transcription factors. Therefore, some antibodies that have little or no effect on the basic functions of the enzyme might be expected to interfere with the complex process of transcription of a specific gene. We have investigated the effect of the above collection of subunit-directed antibodies on the transcription of the yeast tRNbG'" gene in vitro. The antibodies were preincubated with purified RNA polymerase C prior to addition of the two protein fractions containing the required transcription factors (20) and of the cloned gene. After transcription, the synthesized 32P-labeled tRNA was isolated by electrophoresis and the gel was subjected to autoradiography. The efficiency of transcription was estimated by comparing densitometer scans of the tRNA signals with that of a control reaction comple- mented with antibodies from a nonimmune rabbit. Fig. 7 shows a summary of the results obtained with the different antibodies. Several antibodies were poorly or not inhibitory in the range of concentration explored (up to 200 pg/ml): anti-Cleo, -C27r -Cl4.5, and -Cl0. The same antibodies had little or no effect in the nonspecific assay. The small inhibition obtained with anti-Cro and anti-C31 was in good agreement also with that found with the poly[d(A-T)] assay at the same concentration (200 pglml). The other antibodies directed to Clzs, Cs2, CS3, C34, C23, and Clg were strongly inhibitory. Several observations indicated that the antibodies were acting through their interactions with RNA polymerase C. The same antibodies (at 200 pg/ml) had no effect on the signal intensity when added after 45 min of transcription and incubated for 30 more min (see Fig. 7, lower panel). Except for anti-CS3 and

-CS4 to a small extent, the antibodies had no significant effect if the preincubation step with RNA polymerase C was omitted (result not shown).

The profound inhibition of tRNA synthesis by antibodies to CS3 (90% inhibition at 50 pg/ml) or C34 (81% inhibition at 100 pg/ml) was particularly striking when compared to their almost zero effect in the nonspecific assay at the same concentration. The relative efficiency of the antibodies varied markedly when comparing the two assay systems. Hence, anti-Clo and anti-& were equally inhibitory in the poly[d(A- T)] assay (50% inhibition at 600 pg/ml) (see Fig. 6), whereas in the specific assay at 50 pg/ml anti-C5, IgG totally blocked tRNA transcription and anti-Clo had practically no effect (Fig. 7). Considering the couple CS4 and c31, anti-C34 was also much more inhibitory in the specific assay as compared to anti-C31 IgG. Similarly, anti-Csz was relatively more effective than anti-C40 in the specific system. In contrast, antibodies against c23 or C19 were equally inhibitory in the two transcription systems. These results suggested that some antibodies interfered predominantly with the specific transcription process.

DNA Protects the Enzyme from Inhibition-To determine the subunits involved in template binding in the specific transcription assay, we investigated whether DNA could pre- vent inhibition by some antibodies. RNA polymerase C was preincubated with pY20 plasmid DNA harboring the tRNA?' gene prior to antibody addition, then the other components, factors and substrates, were added t o initiate transcription. The pre-tRNAf'" transcript was estimated by gel electrophoresis and autoradiography. The results are shown in Fig. 8. The plasmid DNA protected enzyme C against two antibodies, anti-Clz8 and anti-CZ3, but it did not interfere with inhibition by anti-G3, -C34r or -C19. Densitom- eter scans of the RNA signal were used to estimate the extent of protection which was 60% with anti-ClZa IgG and reached 99% with anti-C23 IgG. In a template competition assay we found that preincubation of enzyme C with pY20 DNA inhibited 70% the subsequent transcription of poly[d(A-T)], which reflected a stable interaction of enzyme C with plasmid DNA. Nevertheless, the residual poly[r(A-U)] synthesis was still markedly sensitive to inhibition by anti-C23 antibodies (results not shown).

DISCUSSION

This is the last part of our immunological studies on the polypeptide components of yeast RNA polymerases. Antibod- ies to enzymes A and B have.been previously characterized and used for the structural analysis of eukaryotic RNA polymerases (15, 16, 18). With the present work on antibodies to enzyme C components, the complete set of antibodies against each of the three polymerases subunits is now available. This has been possible, until now, only with the yeast system.

Yeast RNA polymerase C is the most complex of the three nuclear enzymes. Variable polypeptide compositions have been reported. Some polypeptides were not found consistently, like C53, C37, and C3, (8, 27, 28) or were not taken into account, like C31 and c25 (26), and have sometimes been considered as contaminants. The polypeptide content of enzyme C derived from the present work is Cleo, ClzS, Cs2, C53, c40 (C37), C34, C31, CZ7, (c25), C23, (Clgs), C19, c14.5, (c12.51, and (Clo). Within parenthesis are the polypeptides with abnormal stoichiometry or that were poorly characterized. It is likely that the population of enzyme molecules is heterogeneous with respect to the components of low stoichiometry like C37 or CzS. That these many components are physically associated with the enzyme molecule is suggested by the fact that they were selectively immunoadsorbedusing antibodies to the large

L"~"IU""

- L C

0

C 0 V

C 160 C128 C 8 2 c 5 3 C 4 0 C 3 4 C 3 1 C 2 7 C 2 3 C 1 9 C14,5 C 1 0

FIG. 7. Effect of antibodies on the specific transcription of the tRNA!?" gene. RNA polymerase C was preincubated in 20 pl with varying amounts, as indicated, of subunit-directed IgG, and the mixture added to an RNA polymerase C-dependent reconstituted transcription system, as described under "Experimental Procedures." The synthesized tRNA was isolated by electrophoresis and visualized by autoradiography. Antibodies are identified by the name of the antigen. Densitometry of the tRNA signals indicated the following inhibition of transcription relative to control with 4 pg of antibodies: anti-ClW, 0%; anti-Clm, 70%; anti-Csz, 53%; anti-Csn (2 pg), 90%; anti- Cl0, 22%; anti-& 81%; anti-Cs1, 26%; anti-Cn, 0%; anti-Cm (2 pg), 79%; anti-C19 (2 pg), 76%; anti-C14.a or anti- Clo, 0%. Lower panel shows the tRNA bands obtained after 45 min transcription, followed by 30 min incubation a t 30 "C with 200 pg/ml of the various antibodies. From left to right, antibodies are as in the main panel (except for anti-Clo which was not tested).

SPECIFIC ANTlBODl ES

m u w I w ~ . w r " -

C 1 2 8 C 5 3 C 3 4 C 2 3 c 1 9

FIG. 8. DNA protects RNA polymerase C from inhibition by anti-Clle and anti-C IgG. RNA polymerase C (0.05 pg in 10 pl) was preincubated with, lr without (+ or -) 0.2 pg of plasmid pY20 DNA for 10 min at 30 before addition of purified I&: anti-Cls (4 pg), anti-Cs (0.8 pg), anti-CM, anti-Cm, or anti-C19 (2 pg each). The mixtures were further incubated for 1 h before reconstitution of the complete transcription system and the RNA transcripts were isolated, as described under "Experimental Procedures."

2 P r .

subunits C1, and C1, from crude RNA polymerase preparations and by the observation of a similar polypeptide pattern by Hager et al. (26) using a very different purification scheme. To conclude that all these polypeptides are enzyme subunits in the strictest sense, one should add the demonstration that they are required for the structural and/or functional integrity of the enzyme molecule. (Referring to eukaryotic RNA polymerases, the term of subunit has generally been taken in the broad sense of being physically associated with the enzyme.)

The specificity of the antibodies directed to the individual components of RNA polymerase C was remarkable since the majority of the polypeptides were specifically recognized by their respective antiserum. Minor cross-reactions can be ex- plained by a slight contamination of the injected polypeptide with a nearby subunit. This was difficult to avoid in view of the complexity of the polypeptide pattern. It can be reasona- bly concluded, therefore, that each component of the polymerase harbors a unique set of immunological determinants (Le. they are unrelated proteins). The same conclusion was reached previously for the components of RNA polymerases A and B (15, 18). The strong cross-reactions found between the three enzymes were again accounted for by their common subunits (ACdO, ABC2,, AB&, AC19, and ABC14.J (29). In addition, a strong cross-reaction was found with antibodies to component(s) Clo. This low molecular weight polypeptide(s1 being poorly characterized, the interpretation of this result remains uncertain. There is the interesting possibility that this component is an additional common subunit.

RNA polymerase C has been isolated from a great variety

15310 Yeast R N A Polymerase C: Subunits and Specific Antibodies

of organisms, from lower eukaryotes like yeast (8), amoeba (9), and fungi (14), to higher eukaryotes like plants ( l l ) , insects (7, lo), and vertebrates (6, 12). The present collection of antibodies will be useful to examine the immunological relationship of the various components of this class of enzyme to confirm their subunit status and study their evolution. The relatedness of several polypeptides of class B RNA polymerase from various organisms has previously been demonstrated using antibodies to yeast enzyme B subunits (16).

Antibodies can be used as functional probes to demonstrate the participation of a given polypeptide chain to the transcrip- tional machinery. Using a nonspecific assay with poly[d(A- T)] as template, one could show that antibodies to Clzs, Csz, Cb3, Ca, c23, and C19 interfered with transcription, albeit with different efficiencies. On this basis, these polypeptides were clearly part of the enzyme molecule. Whether the inhibition is due to the masking of important functional determinants, to a steric effect, or to an indirect effect of antibody binding on the enzyme structure is not known. From the low level of inhibition by antibodies to other components like Clm, CZ7, C14.s (whose participation to the enzyme structure is well established), or like C31, it can only be concluded that the dominant epitopes of these polypeptides (accessible in the enzyme) are not directly involved in transcription.

A reconstituted transcription system dependent upon the addition of purified RNA polymerase C for the accurate transcription of tRNA genes (20) was used in parallel to probe with antibodies the possible involvement of enzyme polypeptides in the recognition of specificity determinants. At least two transcription factors are required for tRNA synthesis (2, 19). One of them, a large protein in yeast called T factor, binds specifically to the internal control region of the tRNA genes (4, 5 ) . It was expected that some antibodies would interfere with interaction of RNA polymerase with the factor-DNA complex while having little effect in a nonspecific transcription assay. Indeed, antibodies directed to C53, C34, and Csz were more strongly inhibitory in the reconstituted system than in the poly[d(A-T)] assay. Some immunodominant epitopes are probably localized in regions of these subunits involved in enzyme-factor interaction (although an indirect effect of antibodies is also possible). On the other hand, antibodies to the common subunits c23 or CIS were very inhibitory in both transcription systems, suggesting a general role for these components in RNA synthesis. We have found a similar drastic inhibition of tRNA:'" gene transcription with antibodies to A23 to AI9?

Preincubation of RNA polymerase C with plasmid pY20 DNA efficiently relieved the inhibition by anti-C2, or anti- Clz, in the reconstituted transcription system. Since enzyme C does not appear to bind by itself specifically to the tRNA gene, the protection effect was due to its interaction with non-promoter DNA sequences. The complete protection observed in the case of anti-C,, antibodies suggests that the active enzymes were transferred to the gene-factor complex by a rapid unidimensional diffusion along the DNA, a mech- anism previously proposed for promoter recognition by the bacterial RNA polymerase (30). Subunits ClZs and c23 play probably an important role in DNA binding. The same conclusion was reached, using a nonspecific assay, concerning the same subunit ABC,, (18), subunit A1s5 (homologous to CIZS in size) (17), and the largest subunit of enzymes A and B (18).

The collection of antibodies to RNA polymerase A, B, and C subunits is being used for the further characterization of

S. Camier, J. Huet, and A. Ruet, unpublished results.

the transcription apparatus. Another important project is the cloning by the immunological approach of RNA polymerase structural genes (2)

Acknowledgments-We particularly appreciated the enthusiastic help of J.-M. Buhler at the initial stage of the project during RNA polymerase C and antibody preparation; A. Ruet and S. Camier helped with the reconstituted transcription system. We wish to thank them for their contribution and for fruitful discussions throughout this work. We are grateful to C. Mann for his correcting and improving the manuscript.

1.

2. 3.

4.

5.

6.

7.

8.

9.

10.

11. 12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25. 26.

27.

28.

29.

30.

REFERENCES Ciliberto, G., Castagnoli, L., and Cortese, R. (1983) Curr. Top.

Deu. Biol. 18, 59-88 Sentenac, A. (1985) Crit. Rev. Biochem. 18 , 31-90 Engelke, D. R., Ng, S.-Y., Shastry, B. S., and Roeder, R. G. (1980)

Cell 19,717-728 Stillman, D. J., and Geiduschek, E. P. (1984) EMBO J. 3 , 847-

Camier, S., Gabrielsen, O., Baker, R., and Sentenac, A. (1985)

Sklar, V. E. F., and Roeder, R. G. (1976) J. Biol. Chem. 2 5 1 ,

Sklar, V. E. F., Jaehning, J. A., Gage, L. P., and Roeder, R. G.

Valenzuela, P., Hager, G. L., Weinberg, F., and Rutter, W. J.

Spindler, S. R., D'Alessio, J. M., Duester, G. L., and Paule, M.

Gundelfinger, E., Saumweber, H., Dallendorfer, A., and Stein, H.

Jendrisak, J. (1981) Plant Physiol. 6 7 , 438-444 Engelke, D. R., Shastry, B. S., and Roeder, R. G. (1983) J. Biol.

Bell, G. I., Valenzuela, P., and Rutter, W. J. (1977) J. Biol. Chem.

Barreau, C., and Begueret, J. (1982) Eur. J. Biochem. 129,423-

Buhler, J.-M., Huet, J., Davies, K. E., Sentenac, A., and Froma-

Huet, J., Sentenac, A., and Fromageot, P. (1982) J. Biol. Chem.

Huet, J., Phallente, L., Buttin, G., Sentenac, A., and Fromageot,

BrBant, B., Huet, J., Sentenac, A., and Fromageot, P. (1983) J.

Klekamp, M. S., and Weil, P. A. (1982) J. Biol. Chem. 257 ,

Ruet, A., Camier, S., Smagowicz, W., Sentenac, A., and Froma-

Buhler, J.-M., Sentenac, A., and Fromageot, P. (1974) J. Biol.

Wandzilak, T. M., and Benson, R. W. (1978) Biochemistry 17 ,

Huet, J., Buhler, J.-M., Sentenac, A., and Fromageot, P. (1975)

Dezilke, S., Wyers, F., Sentenac, A., and Fromageot, P. (1976)

Laemmli, U. K. (1970) Nature 227 , 680-685 Hager, G. L., Holland, M. J., and Rutter, W. J. (1977) Bwchem-

istry 1 6 , l - 8 Valenzuela, P., Bell, G. I., Weinberg, F., and Rutter, W. J. (1978)

in Methods in Cell Biology (Prescott, D., ed) Vol. 19, pp. 1-26, Academic Press, New York

Valenzuela, P., Bell, G. I., Weinberg, F., and Rutter, W. J. (1976) Biochem. Biophys. Res. Commun. 71,1319-1325

Sentenac, A., and Hall, B. D. (1982) in The Molecuhr Biology of the Yeast Saccharomyces. Metabolism and Gene Expression (Strathern, J. N., Jones, E. W., and Broach, J. R., eds) pp. 561- 606, Cold Spring Harbor Laboratory, Cold Spring Harbor, New York

von Hippel, P. H., Bear, D. G., Winter, R. B., and Berg, 0. G. (1982) in Promoters: Structure and Function (Rodriguez, R. L., and Chamberlin, M. J., eds) Praeger Publishers, New York

853

EMBO J. 4,491-500

1064-1073

(1976) J. Biol. Chem. 251,3794-3800

(1976) Proc. Natl. Acad. Sci. U. S. A. 73, 1024-1028

R. (1978) J. Biol. Chem. 253,6242-6248

(1980) Eur. J. Biochem. 111,395-401

Chem. 258,1921-1931

252,3082-3091

428

geot, P. (1980) J. Biol. Chem. 255,9949-9954

257,2613-2618

P. (1982) EMBO J. 1, 1193-1198

Biol. Chem. 258,11968-11973

8432-8441

geot, P. (1984) EMBO J. 3,343-350

Chem. 249,5963-5970

426-431

Roc. Natl. Acad. Sci. U. S. A. 7 2 , 3034-3038

Eur. J. Biochem. 65,543-562

yeast rna polymerase c and its subunits

Documents