Purification and characterization of the cytoplasmic casein kinase I from rat liver

Biochimica et Biophysica Acta, 700 (1982) 221-228 Elsevier Biomedical Press

221

BBA 31028

PURIFICATION AND CHARACTERIZATION OF THE CYTOPLASMIC CASEIN KINASE I FROM RAT LIVER M. PIERRE * and J.E. LOEB

lnstitut de Recherches Scientifiques sur le Cancer, B.P. No. 8, 94802 Villejuif (France) (Received July 2nd, 1981) (Revised manuscript received October 2nd, 1981)

K ~ words: Casein kinase I; Protein phosphorylation; (Rat liver)

Two cyclic nucleotide-independent protein kinases, which preferentially utilize casein and phosvitin as substrates, exist in rat liver. In contrast to cytosol the 'light' form of these enzymes was predominant in the 'microsomal extract'. This form (30000-40000 daltons, casein kinase I) was separated from the 'heavy' form (130000 daltons, casein kinase II) by gel filtration. This enzyme was then purified by successive chromatography on carboxymethyl-Sephadex, phosvitin-Sepharose and hydroxyapatite. The activity of the purified enzyme was 2000-3000-fold the casein kinase activity of the cytosol. It had a S~w of approx. 3 S as determined on sucrose density gradient. After iodination or incubation with IT-3z P]ATP, it was analyzed on polyacrylamide gel in the presence of sodium dodecyl sulfate (SDS) and appeared to be composed of a single polypeptide (36000---1000 daltons) which self-phosphorylated. In contrast to casein kinase II, casein kinase I preferentially utilized ATP over GTP. The K m value for ATP was determined to be 14 p M. The K m value for phosvitin was 0.17 mg/ml. Casein kinase I phosphorylated sites different from those of casein kinase II (as shown with ribosomes or SV40 T antigen). Casein kinase I was further characterized by studying its thermal stability. The half-life at 37°C was 6 min and 1 min 30 s at 54°C. In the presence of two substrates (ATP and phosvitin), the half-life at 54°C increased from 1 min 30 s to 4 min. Hemin strongly increased the rate of inactivation of the casein kinase I at 37°C in the absence or presence of the substrates. Nethylmaleimide also inactivated casein kinase I. Phosvitin, though not ATP, protected the enzyme. This observation may indicate that thiols are involved at the binding site of the enzyme for the protein substrate.

Introduction

Evidence is now accumulating that phosphorylation of proteins plays a fundamental role in the regulation of many cellular functions in eukaryotes. Various protein kinases have been identified: cyclic AMP-dependent protein kinases, cyclic GMPdependent protein kinase, phosphorylase kinase, myosin light chain kinase, casein-phosvitin kinases (for review see Ref. 1). * Present address: U 96 INSERM, Htpital du Kremlin-Bic~tre, 78, rue du Gtntral Leclerc, 94270 Kremlin-Bic~tre, France. 0167-4838/82/0000-0000/$02.75 © 1982 Elsevier Biomedical Press

The role of many of them remains speculative. This is the case for casein kinases. This class of protein kinases is defined by its ability to catalyse the incorporation of phosphate into acidic proteins such as casein or phosvitin without phosphorylate basic proteins like histones. These protein kinases probably exist in all eukaryotic cells, since they have been found in a variety of tissues including rat liver [2-5], calf brain [6], human lymphocytes [7], rabbit reticulocytes [8], mouse plasmocytoma [9], Novikoff ascites tumor cells [10] and yeast [11]. In some mammalian cells, two distinct molecular species have been found [3,4,12-15] and named I

222 and II according to the nomenclature of Hathaway and Traugh [15] or S and TS according to that of Clari et al. [13]. Casein kinase II has been characterized from some tissues, but casein kinase I has been purified only from rabbit reticulocytes [15]. However, a casein kinase which presents similar properties has been isolated from rat liver nuclei [16]. As a prerequisite to the research of the biological function of casein kinase I in rat liver, we have undertaken to purify the cytoplasmic casein kinase I in order to study its properties. Materials and Methods

Materials Chemicals were purchased from the following sources: phosvitin, bovine serum albumin, deoxyribonucleaseI, hemin and N-ethylmaleimide from Sigma; Ultrogel AcA34 and HA-Ultrogel from I.B.F.; CM-Sephadex (A50), Sephacryl $200 and CNBr-activated Sepharose 4B from Pharmacia; ATP, GTP, catalase and cytochrome c from Boehringer; [,/-32p]ATP and [y-32p]GTP from Amersham; 1251from New England Nuclear; X-ray film from Kodak; Quinacrine from Specia. Preparation of phosvitin Sepharose. CNBractivated Sepharose was coupled to phosvitin as follows. Phosvitin (10 mg/ml Sepharose) was dissolved at 5 mg/ml in 0.1 M NaHCO 3, pH 8.3/0.5 M NaC1, and then mixed with Sepharose previously washed with 1 mM HC1. The suspension was gently stirred for 16h at 4°C. After washing with the above buffer, the gel was stirred for 3 h at 4°C in 0.1 M NaHCO 3, pH 8.3/0.5 M NaC1/1 M glycine. Then, Sepharose was washed successively in 0.1 M NaHCO 3, pH 8.3/0.5M NaC1/0.1 M NaCH3COO, pH 4.0/0.5 M NaC1. The washing was repeated twice. Methods Protein lanase assay. Protein kinase assays were performed in a final volume of 50 #1. Standard reaction mixture contained 50 mM Tris-HC1, pH 7.5/5 mM MgC12/1 mg/ml phosvitin/50 #M ATP or GTP (0.5-1 #Ci)/enzyme (0-5 #g) and, when indicated, 5 mM 2-mercaptoethanol. Reaction was initiated by the addition of ATP (or GTP) and stopped after 10 min incubation at

37°C by spotting 40 #1 on a Whatman 3 MM paper disc. Discs were then washed as previously described [17]. Isolation of microsomes and preparation of the "microsomal extract" Male Sprague-Dawley rats were killed and their livers were quickly removed and homogenized in 50 mM Tris-HC1, pH 7 . 5 / 2 m M MgC12 (bufferA) containing 0.25M sucrose (for a typical preparation 25 animals were killed). Further procedures were performed at 4°C. Cell debris, nuclei and mitochondria were removed by 15 min centrifugation at 20000Xg. Microsomes were obtained by 90 rain centrifugation at 105000 X g. Then microsomes were washed with buffer A containing 0.15 M KC1 and, finally, with buffer A containing 0.85 M KCI. The last supernatant obtained after 90 min at 105000Xg is called the 'microsomal extract'. Separation of casein kinase I and casein kinase H on Ultrogel AcA34. Microsomal extract was concentrated with poly(ethylene glycol) 10000. After rapid dialysis against buffer A containing 0.5 M NaC1, microsomal extract (approx. 10 ml) was chromatographed on a large column of Ultrogel AcA34 (4.4 X 50 cm). The flow rate was 20 ml/h. Fractions of 6 ml were collected and assayed for enzyme activity. Active fractions corresponding to casein kinase I were pooled and dialysed overnight against buffer A containing 0.15 M NaC1 (buffer B). Chromatography on CM-Sephadex. CMSephadex column (2.5 X 12 cm) was equilibrated in buffer B. Entire crude casein kinase I obtained by Ultrogel chromatography was loaded on this column, eluted with buffer A containing increasing concentrations of NaCI: 0.25, 0.5 and 0.75 M. Each time, the volume of buffer was 100 ml. Fractions of 5 ml were collected and assayed for activity. Active fractions were pooled and dialyzed overnight against buffer B. Affinity chromatography. A small column (0.4 X 6 cm) of phosvitin-Sepharose was equilibrated in buffer B. After loading the whole enzyme preparation, the column was eluted with buffer A containing increasing concentrations of NaCI: 0.25, 0.4 and 0.6 M (each time with 25 ml). Fractions of 2 ml were collected and assayed for enzyme activity. Active fractions were pooled. Bovine serum albumin (200 /xg/ml) was added. The casein

223

kinase I was dialyzed for 3 h against 25 mM potassium phosphate, pH 7.0. Chromatography on HA-Ultrogel (hydroxyapatite). A small column (0.4 × 7 cm) of HAUltrogel was equilibrated in 25 mM potassium phosphate, pH 7.0. After loading of the whole enzyme preparation, the column was eluted with a 150 ml linear gradient of 0.025-0.5 M potassium phosphate, pH 7.0. Fractions of 2 ml were collected and assayed for activity. Active fractions were pooled and dialysed overnight against buffer A + 0.35 M NaC1 and 50% glycerol. Casein kinase II was partially purified by the same procedure (unpublished data). Protein concentration was determined according to McKnight [18]. Polyacrylamide gel electrophoresis. Gel electrophoresis (10-15% acrylamide) was carried out in the presence of sodium dodecyl sulfate according to Laemmli and Favre [19]. Gels were stained with Coomassie brilliant blue R-250 in 30% methanol/7% acetic acid. Prior to electrophoresis, 20 #1 of the preparation of casein kinase I was: (1) iodinated according to Gibson [20] in the presence of chloramineT with 1 mCi '25I in 50 mM TrisHC1, pH 7.5/4 M urea/0.1% SDS. Bound and free 12sI were separated on a small column of Sephadex G25 (3 ml) equilibrated in 50 mM Tris-HC1, pH 7.5/4 M urea/0.1% sodium dodecyl sulfate (sodium dodecyl sulfate was then raised to 2%); or (2) incubated at 37°C with [-r-a2p]ATP in 50 mM Tris-HC1, pH 7.5/5 mM MgC12/7 mM NaC1/10% glycerol. Reaction was stopped by adding sodium dodecyl sulfate (2% final). Finally the samples were boiled for 3 min in 2% 2-mercaptoethanol. Autoradiography was performed by applying X-ray film (Kodirex or RPXomat) at -70°C. Isolation of 40S ribosomal subunits and SV40 Tantigen. Ribosomal subunits from rat liver were prepared according to Martin and Wool [22]. Polysomes were dissociated in 50 mM Tris-HC1, pH 7.5/12.5 mM MgC12/20 mM 2mercaptoethanol/0.88 M KC1/0.1 mM puromycine. Ribosomal subunits were separated by centrifugation on a 10-30% sucrose density gradient. After dialysis against 50 mM Tris-HC1, pH 7.5/12.5 mM MgC12/20 mM 2-mercaptoethanol/ 80 mM KC1, 40 S subunits were precipitated with 0.2 vol. ethanol.

SV40 T antigen containing immune complexes bound to protein A Sepharose, prepared according to Schwyzer [23], was kindly provided by Dr. M. Kress (IRSC--Villejuif, France). Results

Clari et al. [13] have observed that casein kinase II, the heavy form of casein kinase, also named TS by these workers, is predominant in the cytosol. In contrast, we found, as shown in Fig. 1, that casein kinase I, the light form, is the major form in the 'microsomal extract' (0.85 M KC1 wash). The apparent molecular weight of casein kinase I was 30000-40000. Furthermore, since the casein kinase activity of the microsomal extract was about 10fold that of the cytosol, we performed subsequent purification from this fraction. Purification. Casein kinases I and II of the microsomal extract were separated by gel filtration according to Clad et al. [13] but on Ultrogel in place of Sepharose 6B. In subsequent chromatog-

200K 160K

68K

12K

I

u a w r~

2 8 n,.

i"% 1

z

n

lOO

50

VOLUME (ml)

Fig. 1. Chromatography of casein kinase activity from microsomal extract on Sephacryl $200. Concentrated microsomal extract (1 ml) was dialyzed against b u f f e r A + 0 . 5 M NaCI and was then chromatographed on a Sephacryl S200 column (86 × 1.5 cm) previously equilibrated in the same buffer. Fractions of 2 ml were collected and assayed for enzyme activity (vertical axis). The column was previously calibrated with immunoglobulin (160K), bovine serum albumin (68K) and cyto. chrome c (11.7 K). • . . . . . , A~sonm; O O, 32p incorporated•

224

raphy on CM-Sephadex casein kinase I was bound and then eluted with 0.5 M NaC1 after washing with 0.25 M NaCI. Then, the enzyme was chromatographed on phosvitin-Sepharose and eluted with 0.4 M NaC1 after washing with 0.25 M NaC1. Casein kinase I adhered to HA-Ultrogel and eluted as a single peak between 0.12 and 0.18 M potassium phosphate. Although considerable loss of activity was observed in the last two steps, the casein kinase activity of the pure enzyme was 2000-3000-fold that of the cytosol (Table I). However, the enzyme could be stored at - 7 0 ° C after dialysis against 50% glycerol/0.35 M NaC1 containing buffer A.

g-. x

o

OU0.5 .q

Determination of the sedimentation coefficient. An aliquot of the pure enzyme, not dialyzed against buffer containing glycerol, was sedimented on 520% sucrose density gradient in buffer A containing 0.2 M NaCI (for details see legend to Fig. 2). Casein kinase I had a sedimentation coefficient of nearly 3 S determined with bovine serum albumin and deoxyribonuclease I used as markers (Fig. 2). SDS-polyacrylamide gel electrophoresis. The enzyme was analyzed on polyacrylamide gel. It was iodinated or self-phosphorylated (see Materials and Methods) prior to electrophoresis. After iodination one single band (36000 - 1000 daltons) and one doublet (68000-62000 daltons) were observed (Fig. 3A). The doublet might correspond to albumin and a degradation product of the albumin used since no component having a molecular weight above 50000 daltons was observed when the enzyme preparation was

4.3

2~

i 1.£

5

~O A

VOLUME

L

Fig. 2. Coefficient of sedimentation of casein kinase I. An aliquot of 0.4 ml casein kinase I (last stage of purification without dialysis against glycerol) was loaded on the top of a linear gradient (13.2 ml) of 5-20% sucrose in 50 mM Tris-HCl, p H 7 . 5 / 2 m M MgCI2/0.2 M NaC1. The reference proteins were deoxyribonuclease I, 2.8 S, bovine serum albumin 4.3 S. Centrifugation was performed in a SW41 rotor at 40000 rev./min for 15 h at 4°C. Fractions of 0.3 ml were collected and assayed for enzyme activity.

analyzed just after phosvitin-Sepharose chromatography (before addition of albumin). Then, it seems likely that the 36000-dalton component corresponds to the kinase. After self-phosphorylation (Fig. 3B), only one band (of this molecular weight) was seen on the autoradiograph. These results, taken together with Ultrogel

TABLE I PURIFICATION OF CASEIN KINASE I

Cytosol Microsomal extract Ultrogel CM-Sephadex Phosvitin-Sepharose HA-Ultrogel a

~

(ml)

Specific activity nmol 32p incorporated/ min per mg enzyme

Casein kinase activity relative to cytosol

0.075 0.93 2.1 19.8 80 200-300

12.4 28 264 1066 2 600-3 800

Yield (%)

I

a Protein concentration was very low and could not be precisely determined.

I00 82.4 48.2 17.4 4.5

225

A

68K

40K

30K

7

E c

0

5

10

68K 43K

94~

30K

55K

,oK

17K

5

Dlstonce of rnigrotion (cm)

Fig. 3. Densitometric tracing of the polyacrylamidegel electrophoresis in the presence of SDS. Casein kinase I (last stage of purification) was iodinated (A) or incubated with [-y-32p]ATP (B), Then, samples were submitted to electrophoresisaccording to Laemmli and Favre [18] in 15% polyacrylamide gels. Gels were dried and an autoradiography was performed. Autoradiograms were scanned with a Joyce Loebl densitometer. Standards proteins used for calibration were ovalbumin (40000 daltons), bovine serum albumin (68000 daltons) and carbonic anhydrase (30000 daltons).

chromatography and sucrose density gradient analysis, suggested that the casein kinase I consists of a single polypeptide of 36000 daltons which is able to self-phosphorylate. This interpretation accords well with the observation of Hathaway and Traugh for the reticulocyte enzyme [15].

Specificity for phosphate donors and acceptors. The K m value for ATP was determined to be 14 /~M and for phosvitin 0.17 m g / m l . The enzyme preferentially utilized A T P over G T P (3.2% incorporation with GTP, at 5 • 10 -5 M, relative to ATP) in contrast to casein kinase II (72% incorporation with G T P at 5 • 10 -5 M, relative to ATP). We have checked, as previously observed [13], that casein ldnase I phosphorylated only serine residues while casein kinase II phosphorylated serine and threonine residues. Phospho-amino acids were separated according to Heald [21]. Incubation of 40 S ribosomal subunits with ATP and casein kinases I and II in standard conditions (see details in legend Fig. 4) showed that the two enzymes recognized different sites on ribosomal proteins (Fig. 4A). In the same conditions, two (94000 and 55000) of the three polypeptides iden-

a

b A

a

b

c

B

Fig. 4. Specificityfor phosphate acceptors. 40 S subunits of rat liver ribosomes (A) and antigen T containing immune complexes obtained from 40 #1 extract performed according to Schwyzer [23] from mouse cells transformed by SV40 (B) were incubated for 15 min at 37°C with [7-32p]ATP and casein kinase I or II (5 ,ttl of the preparation). Incubation was performed in 50 #1 of standard reaction mixture for the protein kinase assay. Reaction was stopped by adding SDS (2% final concentration). Samples were submitted to electrophoresis in 15% polyacrylamideaccording to Laemmli and Favre [ 18]. Gels were dried and autoradiography was performed, a, + caseine kinase II; b, +caseine kinase I; c, autophosphorylation of the antigen T containing immune complex. tiffed as SV40 T antigen in immune complexes [23] are phosphorylated by casein kinase I while none are phosphorylated by casein kinase I while none is phosphorylated by casein kinase II (Fig. 4B). All these observations clearly demonstrated the difference of specificity of casein kinase I and II for protein substrates. Heat inactivation. Casein kinase I was further characterized by studying its heat sensitivity. The half-life in the absence of substrates was found to be 6 rain at 37°C and 1 rnin 30 s at 54°C (Fig. 5). Phosphate donors and phosphate acceptors protected the enzyme. In the presence of A T P and phosvitin, the half-life was determined to be 4 rain at 54°C. Both substrates participated in the protection of the enzyme, since, in the presence of A T P or phosvitin, the measured half-life was 3 min at 54°C. Inactivation by heroin. As we had previously seen [24] that heroin inhibits the incorporation of phosphate into phosvitin when catalysed by the

226

lOC

F-

soi

~ so

i W o

'". "'.....

W hi

f8

W 11

2-

i

Fig. 5. Heat inactivation. Purified casein kinase I was preincubated at 54°C in 50 mM Tris-HC1, pH 7.5/4.5 mM MgC12/5 mM NaC1/7.5% glycerol in the absence or presence of substrates: 50 # M ATP and 1 m g / m l phosvitin. At various times, an aliquot was removed and activity was measured in standard conditions (ATP and phosvitin were adjusted in each case respectively to 50 # M and 1 mg/ml). @, control; O, preincubation in the presence of phosvitin; II, preincubation in the presence of phosvitin and ATP.

microsomal extract, we examined the effect of hemin on the rate of inactivation of the pure enzyme. We observed that heroin increased the rate of inactivation (Fig. 6). At 37°C, in the presence of hemin (24 #M), the half-life was determined to be 2 rain 15 s instead of 6 min in the absence of heroin. When Mg 2+ concentration was decreased from 4.5 to 1 mM, the half-life in the presence of hemin decreased to approx. 1 rain although the rate of inactivation of the control was not strongly modified. Inactivation by N-ethylmaleimide. We also examined the effect on the activity of casein kinase I of N-ethylmaleimide, which is known to react with thiols (Fig. 6). At 37°C and in the presence of 2 m M N-ethylmaleimide, the half-life was measured to be 2min 30s instead of 6rain in its absence. Thus, it appeared that thiol groups play an important role in the function of the enzyme. The presence of ATP did not give an efficient protection. In contrast, phosvitin (1 m g / m l ) completely prevented the effect of N-ethylmaleimide. This observation suggested that thiol groups are localized on the enzyme at the binding site of protein substrates.

S

~8

PREINCUBATION TIME (minutes)

HEATING TIME (minutes)

Fig. 6. Inactivation by hemin and N-ethylmaleimide. Purified casein kinaseI was preincubated at 37°C with 2 mM Nethylmaleimide or 24 # M hemin in 50 mM Tris-HC1, pH 7.5,/4.5 mM MgC12/5 mM NaCI/7.5% glycerol in the absence or presence of substrates: 50 # M ATP and I m g / m l phosvitin. At various times, an aliquot was removed and the activity was determined. 5 mM 2-mercaptoethanol were added in the reaction mixture to destroy excess N-ethylmaleimide. When hemin was present in the preincubation mixture, Quinacrine was added (80 /~M final) in the reaction mixture to prevent the action of hemin present in the aliquots. (We have shown previously [24] that tilorone or quinacrine prevent the inactivation of casein kinase I by hemin.) O, control; V, +hemin: [], + N-ethylmaleimide; I10 + N-ethylmaleimide+ ATP; O, + Nethylmaleimide+phosvitin; . . . . . . , +hemin but Mg 2+ was decreased to 1 mM (control curves at 4.5 or 1 mM Mg 2+ were not significantly different).

Discussion This paper reports the purification and the characterization of a rat liver casein kinase isolated fom microsomal 0.85 M KC1 extract. The significance of the association of this kinase with the microsomes is not clear. This association might be due to the presence of substrates of this enzyme in the microsomes. This enzyme may be clearly distinguished by its substrate specificity from the protamine kinase activity which is tightly bound to the microsomal membrane [25]. Although our preparations still contained serum albumin added before the last stage of the purification scheme to protect the enzyme, its specific activity was 20003000-fold that of the cytosol in which both casein kinases I and II were present. Hathaway and Traugh [15] have purified, from rabbit reticulocytes, a protein kinase with properties similar to those of our enzyme (size, K m for substrate, specificity).

227

In other tissues, some workers [4,12-15] have observed a casein kinase of this size. Furthermore, Clari et al. [13] have also shown that the fight form of casein kinase from rat liver cytosol phosphorylated only serine residues in contrast to the heavy form. Some years ago, Takeda et al. [3] made the same observation for chromatin casein kinases. It is possible then that casein kinase I exists in most mammalian cells and in various compartments. In our laboratory, the observation that the light form of casein kinase from cytosol and chromatin has several analogous properties supports this view [26]. However, Thornburg and Lindell [16] have isolated a light form of casein kinase from chromatin and have proposed that it is a dimer (2 × 22000 daltons). At this time, it is not clear whether this enzyme is different from our casein kinase. It is also possible that distinct enzymes of this type exist in chromatin. If the same enzymes were present in the cytosol and in the nucleus, some controls might act on the eventual translocation between these compartments. For example, triiodothyronine, which increases casein kinase activity in rat liver chromatin [28] might act in this process. Otherwise, casein kinase I clearly differs from casein kinase II isolated from various tissues [6,12 -15] by its size and its specificity for phosphate donor and phosphate acceptor. We also further characterized casein kinase I by its heat sensitivity and its sensitivity to some chemicals. Hemin which inactivates other protein kinases like elF2 kinase [29] and cyclic AMPdependent protein kinase [30], also inactivates casein kinase I. The biological significance of this effect is not evident because free heme concentration in cells seems to be very low and also because the biological role of casein kinase I remains unknown. Experiments with N-ethylmaleimide demonstrate that thiol groups are important in enzyme functions and suggest that these groups are localized at the binding site for the protein substrate. The presence of thiols raises the possibility that they play a role in the regulation of the enzyme activity. Self-phosphorylation might also be the basis of a control mechanism. Other protein kinases are able to self-phosphorylate. In the case of cyclic AMP-dependent protein kinases, the rate of subunit reassociation is altered by this modification

[31]. But at this time, no regulatory mechanism is known for casein kinase I. The biological role of this enzyme is still speculative. Hathaway et al. [32] have shown that in vitro the reticulocyte enzyme phosphorylates certain initiation factors (elF4B and elF5). Itarte and Huang [33] have isolated a protein kinase from rabbit muscle with properties similar to those of casein kinaseI, which inactivates glycogen synthetase in vitro. On the other hand [34], the Phihp Cohen group has isolated a glycogen synthetase kinase (GSK3), independent of cyclic AMP and from Ca 2+ and calmodulin, and almost completely devoid of casein kinase activity. Thus, the question arises whether casein kinase I is able to effectively inactivate glycogen synthetase, or whether the preparation of casein kinase I contains GSK3. Research in our laboratory is currently in progress to identify substrates of the casein kinase I and to determine its structural requirements to recognize a site on a protein substrate. Meggio et al. [35] have observed that, with casein as substrate, casein kinase I recognized a serine residue near a block of acidic residues and phosphoserines placed on the N terminal side. We feel that it is now possible to study structural requirements with a protein other than casein, since we report here that SV40 T antigen may be phosphorylated in vitro by casein kinase I (the primary sequence of this protein is known and it contains blocks of acidic residues and serine).

Acknowledgments This work was supported by a grant No. 78-70340 from the D616gation G6n6rale/~ la Recherche Scientifique et Technique.

References 1 Krebs, E.G. and Beavo, J.A. (1979) Annu. Rev. Biochem. 48, 923-959 2 Baggio, B. and Moret, V. (1971) Biochim. Biophys. Acta 250, 346-350 3 Takeda, M., Yamamura, H. and Ohga, Y. (1971) Biochem. Biophys. Res. Commun. 42, 103-110 4 Desjardins, P.R., Lue, P.E., Liew, C.C. and Gornall, A.G. (1972) Can. J. Biochem. 50, 1249-1259 5 Dastugue, B., Tichonicky, L. and Kruh, J. (1974) Biochimie 56, 491-500

228 6 Walinder, O. (1973) Biochim. Biophys. Acta 293, 140-149 7 Kemp, B.E., Froscio, M., Rogers, A. and Murray, A.W. (1975) Biochem. J. 145, 241-249 8 Traugh, T.A. and Traut, R.R. (1974) J. Biol. Chem. 249, 1207-1212 9 Schmitt, M., Kempf, J. and Quirin-Stricker, C. (1977) Biochim. Biophys. Acta 481,438-449 10 Dahmus, M.E. and Natzle, J. (1977) Biochemistry 16, 1901 -1908 11 Lerch, K., Muir, LW. and Fischer, E.H. (1975) Biochemistry 14, 2015-2023 12 Keller, R.K., Chandra, T., Sehroder, W.T. and O'Malley, B.W. (1976) Biochemistry 15, 1958-1967 13 Clari, G., Pinna, L.A. and Moret, T.V. (1976) Biochim. Biophys. Acta 451,484-490 14 Cochet, C., Job, D. and Chambaz, E.M. (1977) FEBS Lett. 83, 53-58 15 Hathaway, G.M. and Traugh, J.A. (1979) J. Biol. Chem. 254, 762-768 16 Thornburg, W. and Lindell, T.J. (1978) J. Biol. Chem. 253, 4638-4641 17 Martelo, O.J., Woo, S.L.C., Reimann, E.M. and Davie, E.V. (1970) Biochemistry 9, 4807-4813 18 McKnight, S.G. (1977) Anal. Biochem. 78, 86-91 19 Laemmli, U.K. and Favre, M. (1973) J. Mol. Biol. 80, 575-599 20 Gibson, W. (1974) Virology 62, 319-336 21 Heald, P.J. (1958) Biochem. J. 68, 580-584

22 Martin, T.E. and Wool, I.G. (1969) J. Mol. Biol. 43, 151 - 16 I 23 Schwyzer, M. (1977) EMBO INSERM Workshop, Early Proteins, Oncogenic DNA viruses 69, 63-68 24 Pierre, M. and Loeb, J.E. (1977) Biochem. Biophys. Res. Commun. 77, 481-488 25 Sommarin, M. and Jergil, R. (1978) Eur. J. Biochem. 88, 49-60 26 Dabauvalle, M.C., Meggio, F., Creuzet, C. and Loeb, J.E. (1979) FEBS Lett. 107, 193-197 27 Thornburg, W., Gamo, S., O'Malley, A. and Lindell, T. (1979) Biochim. Biophys. Acta 571, 35-44 28 Kruh, J. and Tichonicky, L. (1976) Eur. J. Biochem. 62, 109-115 29 Balkow, K., Hunt, T. and Jackson, R.L. (1975) Biochem. Biophys. Res. Commun. 67, 366-375 30 Datta, A., De Haro, C., Sierra, J.M. and Ochoa, S. (1977) Proc. Natl. Acad. Sci. U.S.A. 74, 3326-3329 31 Rangel-Aldoa, R. and Rozen, E.M. (1976) J. Biol. Chem. 251, 3375-3380 32 Hathaway, G.M., Lundak, T.S., Tahara, S.M. and Traugh, J.A. (1979) Methods Enzymol. 60, 495-511 33 Itarte, E. and Huang, K.P. (1979) J. Biol. Chem. 254, 4052-4057 34 Embi, N., Rylatt, D.B. and Cohen, P. (1980) Eur. J. Biochem. 107, 519-527 35 Meggio, F., Donella Deana, A. and Pinna, L.A. (1979) FEBS Lett. 106, 76-80