Phosphorylative and functional modifications of nucleoplasmic RNA polymerase II by homologous adenosine 3':5'-monophosphate-dependent protein kinase from calf thymus and by heterologous phosphatase.

We have studied the effects of calf thymus nuclear CAMP-dependent protein kinase and of bacterial phosphatase on the phosphorylative and functional modification of calf thymus nucleoplasmic RNA polymerase II. Incubation of highly purified preparations of calf thymus nuclear CAMP-dependent protein kinase and of homologous RNA polymerase II in the presence of [yJ”PlATP and 1 PM CAMP led to a 3-fold stimulation of RNA polymerase II activity and to an average incorporation of 0.5 mol of [“‘Plphosphatelmol of RNA polymerase II. Analysis of the 32P-labeled RNA polymerase II by polyacrylamide gel electrophoresis under nondenaturing conditions showed that both forms of RNA polymerase II were phosphorylated. Polyacrylamide gel electrophoresis of the:‘“P-labeled RNA polymerase II in the presence of 0.1% sodium dodecyl sulfate revealed that the Z&000-dalton polypeptide subunit of RNA polymerase II (subunit B5) served as the principal [:“Plphosphate acceptor. In some but not all experiments a minor degree of :(2P incorporation into the 180,000-dalton subunit (subunit B2) was also observed. Ion exchange chromatography of acid-hydrolyzed”“P-labeled RNA polymerase II resulted in the identification of serine and threonine as the [“‘PIphosphate acceptor amino acids. Phosphorylation and activation of RNA polymerase II were dependent upon the presence of ATP and of the active catalytic subunit of the nuclear CAMP-dependent protein kinase. Substitution of ATP with adenylyl imidodiphosphate, an inhibitor of protein kinase, prevented protein kinase-mediated activation of RNA polymerase II. Furthermore, selective inhibition of the protein kinase catalytic subunit by a heat-stable protein kinase inhibitor from rabbit muscle led to both an inhibition of RNA polymerase II phosphorylation and to a proportional decrease of the degree of activation of RNA polymerase II. Incubation of :“P-labeled RNA polymerase II with alkaline phosphatase from E’scherichia coli resulted in a loss of :rzP label with a concomitant decrease of RNA polymerase

We have studied the effects of calf thymus nuclear CAMP-dependent protein kinase and of bacterial phosphatase on the phosphorylative and functional modification of calf thymus nucleoplasmic RNA polymerase II. Incubation of highly purified preparations of calf thymus nuclear CAMP-dependent protein kinase and of homologous RNA polymerase II in the presence of [yJ"PlATP and 1 PM CAMP led to a 3-fold stimulation of RNA polymerase II activity and to an average incorporation of 0.5 mol of ["'Plphosphatelmol of RNA polymerase II. Analysis of the 32P-labeled RNA polymerase II by polyacrylamide gel electrophoresis under nondenaturing conditions showed that both forms of RNA polymerase II were phosphorylated.
Polyacrylamide gel electrophoresis of the:'"P-labeled RNA polymerase II in the presence of 0.1% sodium dodecyl sulfate revealed that the Z&000-dalton polypeptide subunit of RNA polymerase II (subunit B5) served as the principal [:"Plphosphate acceptor. In some but not all experiments a minor degree of :(2P incorporation into the 180,000-dalton subunit (subunit B2) was also observed. Ion exchange chromatography of acid-hydrolyzed""P-labeled RNA polymerase II resulted in the identification of serine and threonine as the ["'PIphosphate acceptor amino acids. Phosphorylation and activation of RNA polymerase II were dependent upon the presence of ATP and of the active catalytic subunit of the nuclear CAMP-dependent protein kinase. Substitution of ATP with adenylyl imidodiphosphate, an inhibitor of protein kinase, prevented protein kinase-mediated activation of RNA polymerase II. Furthermore, selective inhibition of the protein kinase catalytic subunit by a heat-stable protein kinase inhibitor from rabbit muscle led to both an inhibition of RNA polymerase II phosphorylation and to a proportional decrease of the degree of activation of RNA polymerase II.
Incubation of :"P-labeled RNA polymerase II with alkaline phosphatase from E'scherichia coli resulted in a loss of :rzP label with a concomitant decrease of RNA polymerase * This investigation was supported by National Science Foundation Grant PCM-76-17213 and in part by the Research and Education Fund, Northwestern Memorial Hospital. The costs of publication of this article were defrayed i n part by the payment of page charges. This article must therefore be hereby "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. II activity. The results indicate that phosphorylative and functional modifications of RNA polymerase II by homologous nuclear protein kinases and by heterologous phosphatase may be achieved in vitro.
In recent years, CAMP-dependent and independent protein kinases from nuclei of a number of tissues have been identified and characterized (l-12). The biological role of nuclear protein kinases remains, however, unknown. It is conceivable that in combination with nuclear phosphoprotein phosphatase, nuclear protein kinases of the CAMP-dependent and independent variety act as converter enzymes achieving the phosphorylative and functional modification of proteins that partake in the diversity and complexity of nuclear function.
Alterations in nuclear RNA synthesis correlate closely with changes in the phosphorylation of nuclear nonhistone proteins (13)(14)(15)(16)(17)(18)(19), and there is indirect evidence suggesting that control of DNA transcription may be achieved through a phosphorylative and functional modification of RNA polymerases. We have previously reported the activation of RNA polymerases I and II by a CAMP-dependent protein kinase from calf ovary cytosol (20), and we demonstrated a functional modification of calf thymus RNA polymerase II by homologous nuclear CAMPdependent protein kinase and by phosphatase (21). Additionally, activation of rat liver RNA polymerase I (22) and of ascites tumor RNA polymerase II (23) was reported to occur through the action of homologous protein kinases.
In the light of these reports, it appears a distinct possibility that functional modification of polymerase activity may be achieved through phosphorylation and dephosphorylation of the core RNA polymerase. The following report presents evidence for the subunit phosphorylation and stimulation of the enzymatic activity of RNA polymerase II by a homologous nuclear CAMP-dependent protein kinase from calf thymus in uitro. Additionally, we demonstrate the in vitro dephosphorylation and deactivation of RNA polymerase II through the action of phosphatase.  Gissinger and Chambon (24). Tissue-Thymus glands from 4-to 5-month-old calves were obtained at the time of slaughter.
The tissues were immediately frozen in dry ice, stored at -8O", and used for experimentation within 4 weeks after their collection from the slaughter house.  Weber and Osborn (39) and scanned at 550 nm in a Gilford model 2000 spectrophotometer with a gel scanning device. After scanning the gels were sliced into l-or 2mm sections.
The sections were placed into counting vials and extracted for 24 h at 20" in 0.5 ml of Protosol (New England Nuclear Corp.).
After addition of 10 ml of phase-combining system, the samples were counted for the determination of radioactivity. protein kinase. When RNA polymerase II was incubated with lr-32PlATP in the absence of protein kinase, essentially no 32P radioactivity was recovered in the trichloroacetic acid-precipitated RNA polymerase II. However, some autophosphorylation of the CAMP-dependent protein kinase occurred during its incubation with [y-"'PIATP. Therefore, the degree of autophosphorylation of the CAMP-dependent protein kinase was determined under the experimental conditions of Fig. 1. The cpm 'YLP representing the autophosphorylation of the protein kinase were subtracted from the total cpm ?'P incorporated into RNA polymerase II in the presence of protein kinase. The data of Fig. 1, therefore, reflect the degree of [3ZP]phosphate incorporation into the RNA polymerase II only. The total amount of l:3YP]phosphate that was incorporated per mg of RNA polymerase II to achieve complete saturation of all polymerase phosphate acceptor sites varied with the different polymerase preparations.
The reason for this variation is unknown but may be related to the fact that RNA polymerase II may be partially phosphorylated in vivo or may have undergone phosphorylative modification by nuclear protein kinases and phosphatases during its isolation from the nuclear extracts. Assuming a molecular weight of 700,000 for calf thymus RNA polymerase II (261, an average value of 0.5 mol of [32P]phosphate incorporated/mol of RNA polymerase II at saturation levels was determined. Phosphorylation of RNA polymerase II under the conditions of Fig. 1 was neither stimulated by the addition of an ATPregenerating system, nor by the addition of native DNA (up to 200 pg/assay). The phosphorylation reaction was, however, progressively inhibited in reaction media of increasing ionic strength. All phosphorylation reactions were routinely carried out in the presence of 40 mM ammonium sulfate at which concentration the CAMP-dependent protein kinase exhibited 75% of its optimal catalytic activity observed in the absence of ammonium sulfate.
To determine that phosphorylation of RNA polymerase II was mediated by the catalytic protein kinase subunit, the effect of a heat-stable protein kinase inhibitor protein on the phosphorylation of RNA polymerase II was investigated. It has been shown that the hect-stable inhibitor protein inhibits CAMP-dependent protein kinase activity by directly interacting with the free catalytic subunit of CAMP-dependent protein kinase (27)(28)(29)(30). The data of Table I show that in the presence of CAMP-dependent protein kinase, CAMP, ATP, and increasing amounts of inhibitor protein incorporation of [R"P]phosphate into RNA polymerase II was progressively inhibited.

DEAE-cellulose
Chromatography and Polyacrylamide Gel Electrophoresis of ""P-labeled RNA Polymerase II -To obtain convincing evidence that ["'PIphosphate was specifically associated with RNA poiymerase II and to demonstrate that the binding of lq2P]phosphate to polymerase protein was covalent in nature, :?'P-labeled RNA polymerase II was subjected to chromatographic and electrophoretic analysis, and the presence of ["ZP]phosphoserine and ["'Plphosphothreonine was determined. To achieve this RNA polymerase II was incubated with CAMP-dependent protein kinase, CAMP, and lyJ2P]ATP under conditions which achieved maximum phosphorylation of the polymerase (for conditions see Fig. 2). To determine the amount of protein kinase autophosphorylation, CAMPdependent protein kinase was incubated under identical conditions in the absence of RNA polymerase II. Both preparations were subsequently subjected to chromatography on DEAE-cellulose to separate protein kinase and RNA polymerase II. Fig. 2, A and B, shows the '?!P elution profiles of the :"P-labeled mixture consisting of protein kinase and RNA polymerase II ( Fig. 2A) and of the :'ZP-labeled protein kinase without RNA polymerase II (Fig. 2B). The elution profile of Fig. 2A reveals the elution of 32P radioactivity in Fractions 56 to 66 at an ionic strength of 0.18 M ammonium sulfate, and in Fractions 72 to 87 at an ionic strength of 0.31 M ammonium sulfate. The latter peak of szP radioactivity coincided with the RNA polymerase II activity which eluted, as expected, at 0.31 M ammonium sulfate. Fig. 2E revealed the elution of R2P radioactivi'v in Fractions 56 to 66 at an ionic strength of 0.18 M ammonium sulfate but not at the higher concentrations of ammonium sulfate indicating that the radioactivity associated with RNA polymerase II (in Fig. 24) did not arise from the :'2P-labeled protein kinase or phosphorylated protein kinase breakdown products. The chemical nature of the R2P radioactivity eluted in Fractions 56 to 66 at 0.18 M ammonium sulfate is unknown. Some protein kinase activity was associated with that :IsP radioactivity but the majority of protein kinase activity applied onto the column eluted at low ionic strength in fractions which did not adsorb onto DEAE-cellulose in accordance with our previous findings (21). Analysis of the :'"P-labeled RNA polymerase II eluted from DEAE-cellulose (see Fig. 2A) by electrophoresis on polyacrylamide gels under nondenaturing conditions revealed the presence of two forms of RNA polymerase II in agreement with previous data (26) (Fig. 3). Both protein bands exhibited RNA polymerase activity (data not shown) and were associated with significant amounts of 'IzP radioactivity.
To determine the pattern of RNA polymerase II subunit phosphorylation, ""P-labeled RNA polymerase II eluted from DEAE-cellulose  Fig. 2). After dialysis and concentration fraction was used as the source of "*P-labeled RNA polymerase II.
(see legend of Fig. 31, 25 kg of "'P-labeled RNA polymerase II were applied onto a mixed polyacrylamide gel (consisting of 5% acrylam-"*P-labeled RNA nolvmerase II (20 ua) was aDDl i ed onto a 5% polyacrylamide gel and subjected to 'erectrophoresis as described ide in the upper half and 10% acrylamide in the lower half of the gel) containing 0.1% sodium dodecyl sulfate. Electrophoresis was under "Experimental Procedures." Electrophoresis was carried out carried out at 2.5 mA/gel for 9 h, at 25". After electrophoresis the for 4 h at 3 mA/eel at 4". The eel was stained. destained. scanned at 550 nm, and subsequently sliled into l-mm sections. The 82P radio-gel was stained, destained, scanned at 550 nm, and then sliced into 2-mm sections which were counted for the determination of s2P activity content of the slices was determined by liquid scintillation radioactivity. For experimental details see "Experimental Procecounting.
( Fig. 2A) was subjected to electrophoresis on polyacrylamide gels in the presence of 0.1% sodium dodecyl sulfate. The characteristic arrangement of six RNA polymerase II subunits was identified (Fig. 4). The subunits were designated Bl through B6 in the order of decreasing molecular weight and increasing mobility in the gels as defined by Kedinger and Chambon (26) (Fig. 4). Using both 5% acylamide gels (data not shown) and mixed gels (5% acrylamide in the upper half and 10% acrylamide in the lower half of the gel), the 25,000 molecular weight subunit (B5) of RNA polymerase II was identified as the principal 1""Plphosphate acceptor protein. In about 30% of the experiments low levels of :reP radioactivity (about 20% of the :V2P radioactivity seen in the B5 subunit) were also found associated with the 180,000 molecular weight subunit B2.
To identify the lVzPlphosphate acceptor amino acid, :r2Plabeled RNA polymerase II eluted from DEAE-cellulose (see Fig. 2A) was hydrolyzed and the hydrolysate was subjected to ion exchange chromatography on Dowex AG50W-5X. The amino acid elution profile shown in Fig. 5 reveals the presence of l:ViP]phosphoserine and of l"'P]phosphothreonine with free inorganic phosphate eluting in the early fractions. It was consistently observed that serine was the major l:'YPlphosphate acceptor amino acid. Additional confirmation that the phosphoryl groups attached to RNA polymerase II corresponded to those of phosphoserine and phosphothreonine was obtained when the acid and base stability of the phosphate-RNA II was subsequently precipitated by the addition of trichloroacetic acid, and a2P radioactivity was determined. The picomoles of 13*P]phosphate/sample plotted are corrected data and were obtained after subtraction of the cpm of "2P/sample contributed by the autophosphorylation of the CAMP-dependent protein kinase from the total cpm of "2P/sample obtained after incubation of RNA polymerase II with the various amounts of CAMP-dependent protein kinase.
The samples with nonradioactive ATP were cooled in ice. To measure the RNA polymerase activity of the samples, 6 nmol (4 @Zi) of [o-:mP]UTP and all other reactants for the RNA polymerase assay were added in 40 ~1. Incubation was carried out for 60 min at 37". At the end of the incubation period trichloroacetic acid-insoluble 32P-labeled product was determined as described under "Experimental Procedures." The values shown are the arithmetic means ofthree determinations. 0---0, R2P radioactivity; O-O, RNA polymerase II activity.   Hirsch and Martelo (60) reported the incorporation of ["?P]phosphate into serine and threonine residues of several subunits of the nucleolar RNA polymerase I from rat liver when phosphorylation was carried out with intact nuclear preparations. Under their experimental conditions no phosphorylation of RNA polymerase II was observed, but Rutter et al. (52) have previously reported the in vitro phosphorylation of purified rat liver RNA polymerase II with a CAMP-dependent protein kinase from both rat liver and rabbit muscle.
The results of our experiments demonstrate that a highly purified nuclear CAMP -dependent protein kinase carries out the in vitro phosphorylation of the 25,000-dalton polypeptide subunit of the homologous nucleoplasmic RNA polymerase II.
In several experiments a minor degree of :leP incorporation into the 180,000-dalton subunit was observed. The inconsistency of the 180,000-dalton subunit phosphorylation may indicate that the subunit underwent some phosphorylative modification in vivo or during its isolation. The identification of only two RNA polymerase subunits as phosphate acceptors does not exclude the possibility that other RNA polymerase subunits may become phosphorylated as well. The subunits may exist in an extensively phosphorylated state which makes the identification of additional [""PIphosphate incorporation impossible. Similarly, the observed average incorporation of only 0.5 mol of [:"P]phosphate/mol of RNA polymerase may reflect a partially phosphorylated state of the purified RNA polymerase.
A causal relationship between RNA polymerase II phosphorjlation and activation was at least partially confirmed by four experiments. Firstly, as shown in Fig. 6, there is a good correlation between the extent of phosphorylation and activation of RNA polymerase II. Secondly, selective inhibition of the protein kinase catalytic subunit by the heat-stable protein kinase inhibitor led to both an inhibition of RNA polymerase II phosphorylation and to a proportional decrease of the degree of activation of RNA polymerase II (Tables I and III). Thirdly, substitution of ATP and the ATP-analogue adenylyl imidodiphosphate, whose terminal phosphate group cannot be utilized and transferred to a protein substrate, prevented CAMP-dependent protein kinase-mediated activation of RNA polymerase II (see Table II). Finally, treatment of :32P-labeled RNA polymerase II with bacterial phosphatase led to both a dephosphorylation and concomitant deactivation of RNA polymerase II (Table IV).
The 25,000-dalton RNA polymerase II subunit (B5 subunit) was identified as the major phosphate acceptor polypeptide. Since phosphorylation of the B5 subunit was accompanied by a stimulation of RNA polymerase II activity, it may be conceivable that the B5 subunit plays a crucial role in determining the enzymatic activity of RNA polymerase II. Recently, Valenzuela et al. (61) showed that the molar ratio of the yeast RNA polymerase I 24,000-dalton subunit, a subunit which can become phosphorylated by yeast protein kinase (57, 581, is requisite for yeast RNA polymerase I activity. Although we have not demonstrated a relationship between the molar ratio of the B5 subunit and RNA polymerase II activity, we tentatively conclude from our data that phosphorylation of the B5 subunit polypeptide may conceivably participate in the regulation of polymerase activity. The major question of whether phosphorylation of RNA polymerase II subunits by nuclear protein kinases modulates RNA polymerase activity in viuo remains unsettled. The precise role of CAMP in modulating RNA polymerase activity in uiuo can only be subject to speculation. Although it appears clear from the present study that the polymerase activity is modulated in vitro by CAMP through its action on CAMPdependent protein kinases, CAMP-independent protein kinases are similarly implicated in the control of RNA polymerase activity (21, 23). Conceivably, both types of protein kinase participate in the regulation of RNA polymerase activity in uiuo but in response to different physiological stimuli.