Interconvertible Forms of Escherichia Coli RNA Polymerase

Proceedings of the National Academy of Sciences Vol. 68, No. 1, pp. 152-154, January 1971

Interconvertible Forms of Escherichia coli RNA Polymerase

CESAR A. CHELALA, LUISA HIRSCHBEIN*, AND HECTOR N. TORRESt§ Instituto de Investigaciones Bioquimicas "Fundacion Campomar"; and Facultad de Ciencias Exactas y Naturales, Buenos Aires, Argentina Communicated by Luis F. Leloir, October 26, 1970

ABSTRACT Escherichia coli RNA polymerase exists in liquid) against the same solution. The enzyme fractions thus two different forms, one less active than the other under standard assay conditions. Conversion of the active to the obtained, called 0.25 and 0.40 were stored at 00C. The last inactive form requires ATP-Mg++. The reverse, inactive to fraction was used as source of polymerase activity. active enzyme conversion, requires Mg++. Covalently Incubation for polymerase conversion was carried out as bound adenylic acid residue(s) are contained in the in- follows: about 0.140 mg of protein of the 0.40 enzyme fraction, active form. containing the RNA polymerase activity, was incubated at Transcription of phage and bacterial genes by RNA polymer- 370C for different periods, with the indicated additions, in a ase seems to be regulated by two different control mechanisms. total volume of 0.03 ml. The reaction was stopped by the One of them stimulates the ability of the enzyme to transcribe addition of 1.5 ml of ice-cold buffer solution C. RNA poly- certain DNA templates, and requires a specific protein factor, merase activity was assayed on 0.15-ml aliquots of these di- called 6 (1). The other control mechanism prevents the trans- luted samples plus the following additions: 0.2 mM CTP, cription by an active polymerase and requires the presence of a UTP, GTP, and (H] ATP (2000 cpm/nmol), 1.6 mM MnCl2, specific protein factor-the repressor (2). These two different 0.01 mg calf-thymus DNA, 40 mM Tris* HCl buffer, pH 7.4, control mechanisms appear to be quite specific. The first and 4 mM mercaptoethanol. The total volume was 0.25 ml. operates as a positive control in the sequential expression of Incubations were at 370C for 10 min; the reactions were certain genomes; the second exerts a negative control on the stopped by the addition of 3 ml of cold 5% trichloroacetic genome transcription coordinated with the input of environ- acid. The precipitates were collected on nitrocellulose (Milli- mental signals. pore) filters, washed several times with 5% cold trichloro- This paper reports that Escherichia coli RNA-polymerase acetic acid, and counted for radioactivity in a scintillation exists in two interconvertible forms of different activity. The spectrometer using a toluene-2,5 diphenyloxazole-1, 4 bis[2- results obtained in this work provide evidence for the existence (4-methyl-5-phenyl) oxazolyl]-benzene mixture. of another mechanism of control of bacterial gene expression. In some experiments, the enzyme samples were passed through Sephadex G-25 columns prior to the assay for poly- MATERIALS AND METHODS merase activity. The volumes of the 0.40 fraction and the addi- The following buffer solutions were used: A, 10 mM Tris* HCl tions were doubled and the reaction was stopped with 0.07 (pH 8.4)-i mM EDTA-5 mM ,3-mercaptoethanol; B, 50 mM ml of ice-cold buffer solution B. After this, 0.01 ml of a 3% Tris HCl (pH 7.5)-20 mM EDTA-50 mM NaF-10 mM,- hemoglobin solution was added to the mixture. The samples mercaptoethanol; C, 50 mM Tris HCl (pH 7.5)-i mM EDTA- were then passed through Sephadex G-25 columns (0.8 X 10 5 mM NaF-5 mM (3-mercaptoethanol. cm) equilibrated with buffer solution C. RNA polymerase RNA polymerase was purified from E. coli A-19 by the activity was assayed on the colored effluent as indicated above. method of Babinet (3) up to step Si. Column chromatography The labeling of the enzyme fraction was carried out as of this fraction was as follows: about 13 ml of the Si fraction, follows: 0.14 mg of protein of the 0.40 fraction was incubated containing 26 mg protein/ml, was loaded on a DEAE-Sepha- with a or -y labeled [32P]ATP (10 mCi/,Mmol) or [U-14C]ATP dex A-50 column (3 X 25 cm) equilibrated with buffer solution (0.5 mCi/ymol) with the indicated concentrations of MgCl2. A. The column was washed with the same buffer solution until The total volume was 0.03 ml. Incubations were performed the absorbance of the eluate, measured at 280 nm, was below at 370C. The reaction was stopped by the addition of 3 ml of 0.1. Then it was eluted successively with 250 ml of buffer solu- cold 5% trichloroacetic acid plus 10 Mmol of ATP. The pre- tion A containing 0.25 M KCl and 250 ml of this buffer solu- cipitates were collected and counted for radioactivity as in- tion containing 0.4 M KCl. The fractions from each elution dicated for the polymerase assay. In some experiments, the step were pooled and precipitated with solid ammonium sul- volumes of the incubation mixtures were increased 60 times fate (60 g/100 ml). After centrifugation for 10 min at 10,000 and the concentration of MgCl2 was adjusted to 0.74 mM. X g, the precipitates were taken up in a minimal volume of In these circumstances, the reaction was stopped by the ad- buffer solution A, and were dialyzed for 4 hr (2 changes of the dition of an equal volume of the buffer solution B. After the solution was mixed, the protein was precipitated with 3 volumes of cold saturated ammonium sulfate solution. The *Permanent address: Institut de Microbiologie, Facult6 des precipitate was collected by centrifugation at 10,000 X g for Sciences, Orsay, France. 10 min, then it was taken in a small volume of the same buffer $ Career investigator of the Consejo Nacional de Investigaciones solution. This labeled enzyme fraction was dialyzed (3 changes Cientificas y Tecnicas (Argentina). § Reprint requests may be addressed to Dr. H. N. Torres, of liquid) against buffer solution A. Instituto de Investigaciones Bioqufmicas "Fundaci6n Campo- The fraction obtained after incubation with [a32P]ATP was mar," Obligado 2490, Buenos Aires (28), Argentina. chromatographed on DEAE-Sephadex as indicated above, 152 Downloaded by guest on September 28, 2021 Vol. 68, 1971 RNA Polymerase of E. coli 153

except that the elution steps were performed with 3 changes TABLE 1. Reaction of labeled ATP with enzyme in the concentration of KCl: 0.25, 0.33, and 0.40 respectively. Fractions of 17 ml were collected. On the other hand, the Radioactivity fraction obtained after incubation with [14C ]ATP was loaded precipitated on a DEAE-Sephadex column (4.7 X 11 cm) equilibrated with trichloro- with buffer solution A. After the column was washed with 50 Conditions acetic acid ml of the same buffer solution containing 0.25 M KCl, the Incu- (cpm) enzyme was eluted with a linear gradient from 0.25 M to 0.50 Total MgCl2 bation Incu- Con- M KCl in the same buffer. Fractions of 10 ml were collected; Expt. ATP (cpm) (mM) (min) bated trol the radioactivity in protein was measured after precipitation 1 [la82p] 1.6 X 107 5.5 10 750,000 1200 of an aliquot of each fraction with 5% trichloroacetic acid and [y_32P] 1.4 X 107 5.5 10 2,700 900 filtration on a nitrocellulose filter. 2 [l4C]aden- 5 X 105 0.74 6.5 66,210 1778 Protein content of the enzyme samples was measured by osine the method of Lowry et al. (4) Electrophoresis on polyacryl- 3 [a-82p] 5.7 X 106 0.74 6.5 5,000,000 5077 amide gels containing sodium dodecylsulfate was as described elsewhere (5). Controls received trichloroacetic acid before the incubations [a32P]ATP was prepared by enzymatic phosphorylation of began. labeled 5'-AMP. 5'-['2P]AMP was synthesized by a chemical reaction between isopropylidene adenosine and 32P-labeled changes were observed after a 50-fold dilution of the incuba- inorganic phosphate in the presence of trichloroacetonitrile tion mixture. Moreover, the same results were obtained when and triethylamine by a modification of the method of Green- the enzyme samples were passed through Sephadex G-25 lees and Symons (6). [9y32P]ATP was prepared as previously columns prior to the assay. described (7). In order to determine if an adenylylation or a phosphorylation of the enzyme lead to polymerase inactivation, the 0.40 RESULTS AND DISCUSSION fraction was incubated with Mg++ and either [a32P]-, [-y- Fig. 1 shows that incubation of RNA polymerase preparation 32p]_, or [14C ]ATP, precipitated with 5% trichloroacetic acid, in the presence of ATP-Mg++ leads to a time-dependent in- filtered, and counted. As shown in Table 1, the incorporation activation of the enzyme. This effect is reversible since the of radioactivity into the trichloroacetic acid-insoluble fraction activity was restored by addition of the 0.25 fraction. No proceeded only when [a32P]- or [14C JATP was used as sub- important changes in the enzyme activity were observed strate. The incorporation from [9y82P]ATP was negligible. when the polymerase was incubated without any addition or Therefore, it appears that inactivation of the polymerase is the in the presence of ATP or magnesium ions alone. result of an adenylylation of the enzyme molecule. In order Some facts indicated that the changes observed in the to verify this presumptive conclusion, the 0.40 fraction was polymerase activity were the result of interconversions be- incubated either with [a-32P]- or [14C ]ATP. After that, the tween different molecular entities of the enzyme. These enzyme was precipitated with ammonium sulfate, dialyzed, and chromatographed on a DEAE-Sephadex column. As can be seen in Fig. 2 the remaining enzyme activity was eluted in a peak, with almost all of the radioactivity insoluble in 5% trichloroacetic acid. This peak was further precipitated with ammonium sulfate and dialyzed. As shown in Fig. 3, when the labeled enzyme preparation was incubated in the presence of Mg++, it lost radioactivity in a time-dependent reaction. Under these conditions, a 14C-labeled product appeared; it was soluble in 5% trichloroacetic acid. This compound be- haved during paper chromatography (ethanol-1 M ammonium acetate, pH 7.4,70:35) as 5'-AMP. Some experimental results indicate that the radioactivity was covalently bound to the protein. The radioactivity copre- cipitated with the protein in 5% trichloroacetic acid and, after exhaustive dialysis of the labeled protein against 40% urea and water, the radioactivity remained in the sack. In addition, after tryptic hydrolysis of the labeled enzyme, a radioactive product soluble in 5% trichloroacetic acid was obtained. M I N U T E S The evidence thus obtained might indicate that E. coli RNA polymerase interconverts according to the following FIG. 1. Inactivation and reactivation of E. coli RNA-poly- scheme: merase. The 0.40 enzyme fraction (0.02 ml containing 0.24 mg ATP-Mg + +(1) protein) was incubated in the presence (0,O) or absence (0) of ACTIVE ENZYMEI 4.1 mM ATP and 4.1 mM MgCl2. At the time indicated by the Mg+ (2) arrow, the 0.25 enzyme fraction (5 pl, containing 0.07 mg protein) INACTIVE ENZYME (adenylylated) diluted in 10 mM MgC12 was added to the incubations, performed in the presence of ATP-Mg++ (0). (U), Control reactions Some properties of the reactions (1) and (2) responsible for containing only the 0.25 enzyme fraction plus ATP-Mg++. these interconversions were studied. Downloaded by guest on September 28, 2021 154 Biochemistry: Chelala et al. Proc. Nat. Acad. Sci. USA

I 0.016 Jo 0 0-

T C- *X @

0.1 3: 0 @ UQ008

0 2 4 20 0 2 4 6 6 8 10 12 FRAC TIONS MINU TE S M9gC12 (mM)

(Left) FIG. 2. DEAE-Sephadex column chromatography of the E. coli RNA polymerase labeled with 32P (A) or 14C (B). (0) Absorb- ance measured at 280 nm; (0) enzyme activity assayed on 0.15-ml aliquots of the indicated fractions for 20 min at 370C; (A) 32P or 14C, total radioactivity of the indicated fractions precipitable with 5% trichloroacetic acid. Broken line in B indicates the KCl concentration in the elution buffer. (Center) FIG. 3. '4C-labeled enzyme incubated in the presence (-) or absence (0) of 2 mM MgCI2. The labeled enzyme (20 AI, containing 0.018 mg protein) was incubated at 370C for the indicated times. Reactions were stopped with 3 ml of 5% trichloroacetic acid. (Right) FIG. 4. Inactivation of E. coli RNA-polymerase as a function of Mg++ concentration. The 0.40 enzyme fraction was incubated for 5 min (0) in the presence of 4.1 mM ATP; the concentration of MgCl2 was varied. (0), Control, not incubated. The vertical dotted line indicates the concentration of EDTA in the reaction mixture. Other conditions were as in Fig. 1.

Purification of the polymerase on DEAE-Sephadex accord- diminution in the rate of reaction (1) leads to an increase in ing to Babinet's method (3), eluting the column in three steps: the polymerase activity. No polymerase activity was detected 0.25, 0.33, and 0.40 M KC1, respectively, led to a polymerase in the 0.25 fraction. preparation that did not have any ATP-Mg++ effect. On the In conclusion, our results show that E. coli RNA poly- contrary, when the elution step with 0.33 M KCI was omitted, merase can exist in two different forms, one more active than the enzyme could be inactivated by ATP-Mg++. This result the other under the conditions of the assay. The less-active indicates that a factor required for reaction (1) is eluted with form appears to be the adenylylated form of the active en- 0.33 M KCl. Also, Mn++ could not substitute for Mg++, and zyme. Interconversion between active and inactive forms of GTP, UTP, or CTP could not replace ATP in the inactivation E. coli RNA polymerase may play an important role in the reaction. The extent of the inactivation was studied at a fixed transcription of bacterial genes during the cell cycle of growth. incubation time and constant concentration of ATP, varying This kind of control mechanism differs from those previously the concentration of Mg++. As can be seen in Fig. 4, the described (1, 2). maximum rate of inactivation was obtained when the ratio We express our special gratitude to Dr. Jos6 M. Olavarria for of ATP to Mg++ was about 1. The increase in activity of the his invaluable assistance, to Dr. Luis F. Leloir for his helpful polymerase at higher concentrations of Mg++ might indicate criticism, and to other members of the Instituto de Investi- an increase in the rate of reaction (2). This assumption is gaciones Bioqufmicas for their collaboration. favored by the fact that the liberation of radioactivity from a This work was supported in part by grants from the U.S. '4C-labeled enzyme was strongly stimulated in the presence of Public Health Service (GM 03442) and the Consejo Nacional de Investigaciones Cientfficas y Tecnicas (Argentina). Mg++ (Fig. 3). This enzyme preparation was submitted to a polyacrylamide gel electrophoresis in the presence of sodium 1. Travers, A. A., and R. R. Burgess, Nature, 222, 537 (1969). 2. Jacob, F., and J. Monod, J. Mol. Biol., 3, 318 (1961). dodecylsulfate. Five protein bands were obtained, with the 3. Babinet, Ch., Biochem. Biophys. Res. Commun., 26, 639 mobilities of ##r', 6, 1A a, and w as were previously described (1967). (5, 8). Probably one of these bands corresponds to the enzyme 4. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J. responsible for reaction (2). Randall, J. Biol. Chem., 193, 265 (1951). 5. J. M. and In the experiment shown in Fig. 1, the addition of a Hirschbein, L., Dubert, Ch. Babinet, FEBS protein Lett., 3, 260 (1969). fraction eluted from the DEAE-Sephadex column with 0.25 6. Greenlees, A. W., and R. H. Symons, Biochim. Biophys. M KCI (0.25 fraction) almost reverses the effect of ATP-Mg++ Acta, 119, 241 (1966). in a time-dependent reaction. As this fraction contains an 7. Torres, H. N., and C. A. Chelala, Biochim. Biophys. Acta, active adenylate kinase and 5'-nucleotidase, its effect could be 198, 495 (1970). 8. Zillig, W., E. Fuchs, P. Palm, D. Rabussay, and K. due to the removal of ATP. Since the observed effects would Zechel, in RNA Polymerase and Transcription, ed. L. G. Silvestry be a balance between reactions (1) and (2), it is obvious that a (North Holland, Amsterdam, 1969), p. 151. Downloaded by guest on September 28, 2021