485

Biochimica et Biophysica Acta, 537 (1978) 485---494 © Elsevier/North-Holland Biomedical Press

BBA 38065

CYCLIC AMP-BINDING PROTEINS AND PROTAMINE KINASES IN PORCINE THYROID CYTOSOL

ALAIN TIRARD and MARIE ROQUES

Laboratoire de Biochimie associd au CNRS 040178 et Unitd 38 de I 'INSERM 27, Bd Jean Moulin, 13385 MarseiUe Cedex 4 (France)

(Received May 10th, 1978)

Summary

Partial purification of cyclic AMP-binding proteins from porcine thyroid cytosol was performed by gel filtration on Bio Gel 1.5 m followed by ion exchange chromatography on DEAE Sephadex A25. Three fractions presenting cyclic AMP-binding activities were resolved by gel filtration (I, II, III). Approxi- mate molecular weights were respectively 280 000, 145 000 and 65 000. Frac- tion I was further resolved into two peaks (Is and I~) on DEAE-Sephadex A25. Fractions I, Is, I~ comigrated with protein kinase activity whereas peaks II and III did not. These fractions differed with respect to the following characteris- tics: rate and stability of cyclic AMP binding to isolated fractions were differ- ently affected by pH (4.0 or 7.5). Electrophoretic mobility on polyacrylamide gels (5%) of fractions preincubated with cyclic [3H]AMP showed similar mobil- ities for Is, I~ or II (Rf 0.37) whereas fraction III displayed a much greater mo- bility (R F 0.73); Scatchard plots were linear for fractions Is, II and III with an apparent Kd in the same range (2 to 5 riM) whereas fraction I~ generated a biphasic plot with Ka 0.4 nM and 20 nM; cyclic [3H]AMP added to fraction I, Is or I~ generated a cyclic [3H]AMP-binding protein complex of lower molecu- lar weight as shown by Sephadex G 150 filtration; on the basis of the elution volume, this complex was not distinguished from fraction II. In the course of this work, we separated at the first step of purification (Bio Gel 1.5 m) a pro- tein kinase not associated with cyclic AMP binding activity which exhibited marked specificity for protamine as compared to histone II A.

Introduction

Most of the numerous effects produced by thyroid-stimuiating hormone (TSH) on thyroid function appear to be mediated by cyclic AMP [1,2]. That

Abbreviation: TSH, thyrold-stimulating hormone.

486

changes in cyclic AMP concentrat ion in the intact tissue should act through activation of cyclic AMP-dependent protein kinases was shown by Spaulding and Burrow [3] and Field et al. [4]. Cyclic AMP-binding proteins and cyclic AMP-dependent protein kinases have been described in subfractions of thyroid cells [5,6]. Soluble cyclic AMP-dependent protein kinases from thyroid cytosol have been purified by several authors [ 7--11]. However, these authors reported only on protein kinase activities and not on cyclic AMP-binding activities. Nevertheless, bet ter knowledge on the transmission of the hormonal signal through the cell requires information on proteins able to bind cyclic AMP and on the characteristics of the binding to these proteins. Accordingly, this report describes the partial purification and some properties of soluble cyclic AMP binding proteins and their relationship to protein kinases identified during the purification procedure.

Materials and Methods

Chemicals: Calf thymus histone II A, protamine sulfate, dithiothreitol, EDTA disodium salt and bovine serum albumin were purchased from Sigma. Phosphorylase a was from Boehringer. Cyclic [all]AMP (32 Ci/mmol) and [~/.32p] ATP (2--3 Ci/mol) were from Radiochemical Centre Amersham England. Bio Gel 1.5 m was obtained from Bio Rad Laboratories, DEAE-Sephadex A25 and Sephadex G150 from Pharmacia Fine Chemicals. Acrylamide, N,N'-metyl- ene bisacrylamide were from BDH. Other chemicals were obtained from Merck. Cellulose ester filters (HAWP 02500) were from Millipore Corporation and columns from Pharmacia.

Assay for cyclic AMP binding activity. Standard cyclic AMP binding assays were performed by a method derived from Gilman's procedure [12]. The reac- tion mixture contained in a final volume of 0.2 ml, 50 mM acetate buffer, pH 4.0, 0.2 uM cyclic [3H]AMP and 0.16 ml fraction. Incubation was carried out for 1 h at 2°C and then s topped by filtering 0.16 ml incubation mixture on a miUipore filter premoistened with 1.5 ml cold phosphate buffer (20 mM, pH 6.0). The filter was rinsed with 10 ml of cold phosphate buffer and dried at 150°C. Radioactivity retained by the filter was estimated by counting in 10 ml of toluene permafluor I mixture.

Data for Scatchard binding plots were obtained at pH 7.5 in the homogeniza- tion buffer with longer incubation times and cyclic [3H]AMP concentrations ranging from 0.1 or 0.33 to 100 or 330 nM. Blank values were measured by omitting the protein fraction ('filter blanks'). Using thyroid cytosol, it was found that 'dilution blanks' {i.e. in the presence of 1 mM unlabelled cyclic AMP) were never larger than filter blanks. Thus, results were only corrected for filter blanks. Assays used to obtain Scatchard plots were done in triplicate.

Assay for protein kinase activity. The protein kinase activity was assayed using protamine as substrate as described earlier [5] with some modifications: [~/-a2P]ATP was 0.1 mM and cyclic AMP when added was 1 gM. Incubation was carried out at 30°C and the radioactivity incorporated into proteins was measured following precipitation on paper with a mixture of 10% trichloro- acetic and 0.25% phosphotungstic acid.

When histone II A was used as substrate, the incubation medium contained in

487

a final volume of 70 ~1, 15 mM phosphate buffer pH 6.5, 4 mM Mg 2÷, 0.4 mM [7-32P]ATP and 300 ~g histone. Cyclic AMP, when present, was 2 pM and sample was I added as 28 pl. Incubation was carried out at 33°C for 25 min.

Preparation of cytosol. Porcine thyroid glands obtained from a local abbatoir were kept on :ice. Glands were trimmed free of fat and connective tissues and sliced with a S~ie-Riggs microtome. The slices were suspended in 3 volumes (v/w) of 10 mM Tri$-HC1, 4 mM EGTA Na2, 1 mM DTT pH 7.0 (buffer H) and homogenized using an Ultra-Turrax homogenizer at half maximum speed for 10 s (3 times). ~ The homogenate was filtered through three layers of gauze centrifuged at 1500 ×~g for 15 min and the resulting supernatant was clarified by centrifugation at 105 000 × g for 1 h.

Polyacrylamide gel ~ei~eVtrophoresis. Polyacrylamide gel electrophoresis was performed according t o Davis [13] in Tris/glycine buffer pH 8.5 at 2°C. The fraction was loaded in 20% sucrose directly on the gel. After electrophoresis (1 mA per gel for 10 min, then 3 mA per gel) gels were cut into slices (2 mm), dissolved in H202 at 60°C overnight and radioactivity counted in Instagel.

Results

I. Separation and purification of cyclic AMP binding proteins (a) Bio Gel 1.5 m: Gel filtration of thyroid cytosol on Bio Gel 1.5 m (Fig. 1)

constantly resolved three cyclic AMP binding protein fractions designated I, II and III, which were well separated from major protein components, especially thyroglobulin. Fraction I was associated with protein kinase activity. This pro- tein kinase phosphorylated protamine and histone and its activity was stimu- lated by cyclic AMP. An apparent Mr ~ 280 000 was estimated after standard- ization of the column with protein markers. Fraction II (apparent Mr-- 145 000) was eluted slightly after a second cyclic AMP-dependent histone and protamine kinase activity (apparent Mr = 170 000). Fraction III devoid of kinase activity exhibited an apparent Mr = 65 000. Finally, between fractions

I

0J 4 ~ ~ 17 x 2

~ r ~

// " ~ Q I .~e 2 CL •

° .., / i

40 810 80 I~10 fraction|

Fig. 1. S e p a r a t i o n by Bio Gel 1 .5 m f i l t r a t i on of cycl ic AMP binding and prote in kinase activities f r o m p o r c i n e t h y r o i d cy toso l . 40 ml of c y t o s o l were app l i ed to a 5 × 80 c m e o i n m n of Bio Gel 1.5 m p rev ious ly

equ i l i b r a t ed w i t h b u f f e r H. T h e c o l u m n w a s e i n t e d w i t h the s a m e b u f f e r ( f low ra t e 24 m l / h ) . F r a c t i o n s o f

a b o u t 11 m l were co l lected and assayed for cycl ic AMP binding and p r o t e i n k inase activities wi th pro t - a m i n e or h i s t o n e II A as subs t r a t e . Ph = p h o s p h o r y l a s e a. SA = b o v i n e s e r u m a l b u m i n .

488

II and III, a peak of protein kinase activity devoid of cyclic AMP-binding activ- ity (apparent Mr = 81 000) was eluted. This protein kinase differed from the preceding enzymatic fractions by its high specificity for protamine, histone being a very poor substrate. Specific cyclic AMP binding activities from most active pooled fractions were 7, 11 and 2 fold higher for fractions I, II, III, respectively than that of the starting cytosol (3030 counts • min -1 • A - Z 2 8 0 n m .

More than 85% of the initial activity was recovered in the total three peaks. (b) DEAE-Sephadex A25 chromatography: The pooled most active fractions

of peaks I to III were further purified by ion-exchange chromatography on DEAE-Sephadex A25. Fig. 2 represents typical profiles obtained from one of at least three similar experiments for each fraction. The major part of protein was excluded. Fractions II and III were eluted as a single peak at about 0.05 M NaC1 (peak II) and 0.08 M NaC1 (peak III). Peak I (associated with protein kinase activity) was resolved into two peaks. The first (Ia) was eluted at <0.1 M NaC1 (about 0.07 M ) and the second (I~) at >0.1 M NaC1 (about 0.12 M). Both peaks were associated with protamine and histone kinase activities. Their specific activities were 50 to 200-fold higher than in the initial cytosol and more than 50% of the initial activity were recovered in the purified fractions.

H. Some properties of purified cyclic AMP-binding fractions. (a). Effect of pH on cyclic AMP binding: Binding of cyclic AMP as a function

of incubation time was studied for each fraction either in buffer H (pH 7.5) or acetate buffer (pH 4.0) at low cyclic AMP concentrat ion (1 nM). The following criteria were examined: rate of binding, maximum binding and stability of binding after long incubation times. Cyclic AMP binding to fraction Ia increased up to 15 h at either pH 4.0 or pH 7.5 and was stable up to 30 h. More cyclic AMP was bound at pH 4.0 than at pH 7.5. Maximum cyclic AMP binding to fraction I~ was obtained after 25 min and was stable up to 30 h at pH 7.5, whereas a small decrease was observed at 30 h at pH 4.0. Binding to fraction II was greatly affected by low pH: at pH 4.0 cyclic AMP binding reached a maxi- mum at 10 min, then decreased to 0 from 10 min to 3 h. At pH 7.5, maximum binding was achieved at 80 min and was stable for 18 h. Maximum binding was

0, i

• o!

to 20 30 40 10 20 30 f rect ~ons I r ac t ~ons

10 2o 3o 4o t r a c t J o n l

F i g . 2 . P u r i f i c a t i o n b y D E A E - S e p h a d e x A 2 5 c h r o m a t o g r a p h y o f p o o l e d f r a c t i o n s I, I f , I I I f r o m B i o G e l 1 . 5 m . 1 5 m l ( f r a c t i o n s I o r I I ) o r 3 0 m l ( f r a c t i o n I I I ) o f p o o l e d f r a c t i o n s f r o m B i o G e l 1 . 5 m w e r e a p p l i e d o n a 0 . 9 X 1 3 c m c o l u m n e q u i l i b r a t e d w i t h b u f f e r H. C y c l i c A M P - b l n d i n g a c t i v i t i e s w e r e e l u t e d w i t h a l i nea~ g r a d i e n t f r o m 0 t o 4 0 0 m M N a C I i n b u f f e r H . F l o w r a t e w a s 8 m l / h a n d v o l u m e o f c o l l e c t e d f r a c t i o n s w a s 4 m l . I n s e t s h o w s d i s c o n t i n u o u s e l u t i o n s ( I : N a C I 1 5 0 m M a n d 3 5 0 m M ; I I I 1 1 5 m M ) . F r a c - t i o n s w e r e a s s a y e d f o r c y c l i c A M P - b i n d l n g a c t i v i t y as d e s c r i b e d i n M e t h o d s .

4 8 9

at any time higher at pH 7.5 than at pH 4.0. In the case of fraction HI, maxi- mum binding was achieved at pH 4.0 after 60 rain, then decreased for 18 h. At pH 7.5 longer incubation times (6 h) were needed to achieve maximum binding and the latter was higher than the maximum obtained at pH 4.0. When same experiments were conducted at high cyclic AMP concentrations (100 riM), the instability of binding at pH 4.0 in fractions II and III was partially corrected indicating a protective effect of cyclic AMP. Under these conditions, initial rates of binding were higher at pH 4.0 than at pH 7.5 for fractions II and III. These results were used to achieve maximum and stable binding in further experiments.

(b). Polyacrylamide gel electrophoresis: Polyacrylamide gel electrophoresis of fractions obtained from the DEAE-Sephadex purification step was per- formed after incubation with cyclic [3H]AMP. Distribution of radioactivity in the gels (Fig. 3) showed that bound cyclic AMP remained associated with the binding proteins during electrophoresis while free cyclic AMP migrated to the front of the gel. Cyclic AMP bound to fraction Ia or I~ migrated as a major

31 Io~

o

1

o

J I I ? 10 2 0 tractions

~frel

3- I I

b

x 2

(J

1

,

10 2 0 fractions

? o

x

E o. L)

T o~

o

x

E cL o "1-

o~

1

I J

10 2 0

4 III

3

o

? fractions

tree c A M P

c o

T 10 2 0 f r ° "

Fig. 3. Polyacry lamide gel e lectrophores is o f purif ied fract ions f rom D E A E - S e p h a d e x A25 . Act ive frac- tinr.s obta ined f rom D E A E - S e p h a d e x wexe incubated w i t h cycl ic [ ~ H ] A M P 0.1 ~M fox 18 h a t 2°C in buf fer H (final v o l u m e 200 /~l). Suczose (20% ~) was added to an a l iquot of 100 ~I, which was then, loaded on the po lyacry lamide gel (5%) and e lectxophoresis per formed as described in Methods .

490

peak (RF 0.37) with sometimes a shoulder of RF 0.30. Cyclic AMP bound to fraction II migrated as a single peak in the same position as the major peak from fractions Is and Ifi (RF 0.37). Fraction III had a much greater mobil i ty and migrated as a single peak (RF 0.73).

(c). Affinity for cyclic AMP: The apparent affinities of the different frac- tions for cyclic AMP were determined after incubation with cyclic. [3H]AMP at concentrations ranging from 0.3 to 300 nM under conditions (Fig. 4) in which maximum and stable binding were achieved. Fractions Is, II and III generated linear Scatchard plots with an apparent K d ranging from 2 to 5 nM. Fraction Ifi, however, generated a biphasic Scatchard plot with apparent Kd 0.4 nM and 20 nM. In the same conditions as described for fractions II and III (i.e. incubations in buffer H for 18 h) Is generated a linear plot and I/~ a biphasic plot.

(d). Dissociating effect of cyclic AMP on fraction I: Fraction I from Bio Gel 1.5 m, initially exhibiting cyclic AMP binding activity and kinase activity was dissolved by cyclic AMP. Both activities were separted as shown in Fig. 5. Frac- tion I from Bio Gel 1.5 m was applied to a column of DEAE-Sephadex A25. Washing with 1 pM cyclic [3H]AMP resulted in elution of protein kinase activ- ity no t associated with cyclic AMP binding activity {catalytic subunit(s)). The purified dissociated catalytic subunit phosphorylated both histone II A and protamine. Free cyclic [3H]AMP was eluted immediately after the catalytic subunit (not shown). The binding pro te in , cyclic [3H]AMP complex was eluted with a linear NaC1 gradient at 170 mM. This fraction, devoid of protein kinase activity (regulatory subunit) was eluted in this experiment as a single peak. However, in some experiments, two peaks of cyclic AMP binding activ- ities were eluted with the NaC1 gradient at 150 and 350 mM. In all experi- ments, no well characterized protamine or histone kinase peak was eluted with the NaC1 gradient (0--0.6 M).

(e). Further characterization of fractions Is, I{J and H: Similar to fraction I from Bio Gel 1.5 m, subfractions Is and I~ from DEAE-Sephadex were both associated with kinase activity and dissociated by cyclic AMP. The cyclic AMP- binding moieties were compared to fraction II by Sephadex G 150 filtration. Fractions were or were not incubated with 0.1 pM cyclic [3H]AMP at 2°C. 1 ml of the fraction or of the incubation medium was loaded on a Sephadex G150

Id

06 Kd : 24 x ~0 "e M

10

O8

01

~0 "~° M tO "e M B

Kd : 1.8x~O4M

a

I0 "e M 2x tO'e M 8

° Kd ; 2.7 x 104 M

o

~ , o M ~r s M B

Fig. 4. Cyclic AMP binding Scatchazd p lo t s o f puri f ied frac t ions f r o m the O E A E - S e p h a d e x s tep. Frac t ions w e r e i n c u b a t e d w i t h c y c l i c [3H]AMP 0.3 nM t o 3 0 0 nM at 2°C. I n c u b a t i o n s w e r e carried o u t in the elu- tlon buf fer at pH 7.5 for f rac t ion I~ (14 h) II (18 h) and III (24 h) or at pH 4.0 for fraction I s (20 h).

491

i |

. .

c

3 N ~ o. -- ~

40 HlO " ~ 0 ~ 0 | r ac t *on8

Fig. 5. Dissociat ion of f r ac t ion I by cycl ic AMP on D E A E - S e p h a d e x . 50 m l of poo led p e a k I f r o m Bio Gel 1.5 m were appl ied to a c o l u m n (1 .5 × 25 c m) of D E A E - A 25 equ i l ib ra ted w i t h b u f f e r H. Th e c o l u m n was washed wi th the s ame b u f f e r un t i l A 2 8 0 n m was nea r 0; 52 m l of cycl ic [ 3 H ] A M P 1 ~M were t hen appl ied and the c o l u m n was washed wi th b u f f e r H. Bound cyclic [ 3 H ] A M P was e lu ted wi th a l inear g rad ien t f o r m 0 to 0.3 M NaCI in the b u f f e r H : F low ra te was 8 m l / h and v o l u m e of the f rac t ions 4 m L B o u n d cycl ic [ 3 H ] A M P ( o - - o) e lu ted in the NaC1 grad ien t was e s t i m a t e d as r ad ioac t iv i ty r e t a ined on a mi l l ipore f i l ter a f te r f i l t ra t ion of an a l iquo t of each f rac t ion . Cyclic AMP binding ac t iv i ty (o o) was meastvred in f rac t ions p r io r to the NaC1 grad ien t s tep, as desc r ibed in Methods . Pro te in kinase ac t iv i ty (e e) was assayed wi th p r o t a m i n e as subs t ra te as descr ibed in Methods .

column (1 × 74 cm) and eluted with buffer H. Fractions of I ml were collected and either assayed for cyclic AMP binding or counted in Instagel for measure- ment of radioactivity. When fraction Is or I~ was chromatographed on this column, cyclic AMP binding activity appeared with 1.12 void volume. Preincu- bation with cyclic [3H]AMP resulted in elution of bound cyclic [3H]AMP with 1.33 void volume. For comparison the cyclic [3H]AMP regulatory subunit com- plex obtained as described above also eluted at 1.33 void volume and chroma- tography of fraction II resulted in elution of cyclic AMP binding activity with 1.33 void volume.

Discussion

Separation of proteins from thyroid cytosol on Bio Gel 1.5 m result in three fractions exhibiting cyclic AMP binding activity (I, II, III). The heaviest one (Fraction I) is associated with cyclic AMP dependent protein kinase activity and has an apparent molecular we~ht of 280 000. This result agrees with the report of Rubin et al. [13] on Cyclic AMP dependent protein kinase from bovine heart muscle (280 000) and with that of Leonard et al. [11] on cyclic AMP dependent protein kinase from thyroid cytosol (230 000). However, this size is probably overestimated as was suggested for the protein kinase from bovine heart muscle [14]. Fraction I phosphorylates both histone and prot- amine. Dissociation by cyclic AMP after adsorption on a DEAE-Sephadex column resulted in the release of a purified protein kinase activity not associ- ated with binding activity (catalytic subunit that phosphorylates both histone and protamine) as previously shown with protein kinase from bovine heart muscle [14]. Further elution with a NaC1 gradient did not result in further elu- tion of protein kinase activity. Thus, protein kinase from peak I was fully dis- sociated and eluted by cyclic AMP.

However, fraction I is heterogeneous since it can be resolved into two peaks of cyclic AMP binding activity (Ia and I~) on DEAE-Sephadex A 25 after elu-

492

tion with a NaC1 concentration gradient. Both peaks exhibit histone and prot- amine kinase activities and are dissociated by cyclic AMP as shown by chroma- tography on Sephadex G150. Considering their elution profile on DEAE-Sepha- dex (one at low and the other at high ionic strength), these two cyclic AMP- binding protein activity peaks could represent the protein kinases designated as 'type I' and 'type II', present in different tissues [16]. To ascertain this propo- sition, it remains to verify that peaks Is and I~ also show the other known prop- erties of protein kinase 'I' and 'II' [16], especially autophosphorylation of the regulatory subunit of type II and absence of autophosphorylation of the type I enzyme [17].

An important difference between Is and I~ is revealed by cyclic AMP bind- ing saturation analyses. Despite the fact that, according to the reaction models proposed for cyclic AMP binding to cyclic AMP dependent protein kinases, Scatchard plots should not be linear [18--21], fraction Is generates a straight line whereas fraction I~ generates a biphasic curve with a high affinity compo- nent {apparent Kd 0.4 nM). This could reflect different mechanisms for cyclic AMP binding on fraction Is and I~. Values for apparent Kd are much lower (2 orders of magnitude) than the physiological concentration of cyclic AMP in the thyroid. However, it was demonstrated that small changes in cyclic AMP con- centration cause a significant intracellular activation of thyroid protein kinase activity [3,4]. In intact tissue, the apparent Kd may be very different, due to interactions with other factors. Such differences between in vitro and in vivo values, ranging from 1 to 10 were noticed for the binding of cyclic AMP to receptors of rat myometrium [22] or to receptors of rat diaphragm [23]. These differences were explained in the case of cyclic AMP dependent protein kinases by the inhibitory effect of MgATP [25--27] and the effect of protein kinase concentration in the test [17,18]. With respect to these possibilities, it may be noticed that in our experiments, dilution of cyclic AMP binding activity in purified fractions as compared to initial tissue was in the range of 1150 and that previous results [28] indicated that ATP at very low concentrations inhibited cyclic AMP binding in whole thyroid cytosol.

We also characterized two other binding proteins. Fraction III of lower molec- ular weight differs form the others by its high electrophoretic mobility. We do not know whether this fraction represents a special kind of cyclic AMP binding protein or a preferential degradation product of fraction I or II.

Fraction II and fractions Is or I~ previously treated with cyclic [3H]AMP could not be discriminated from each other by their electrophoretic migration nor by molecular weight elution. Therefore, fraction II very probably repre- sents a regulatory subunit of cyclic AMP-dependent protein kinases. Its appar- ent molecular weight was 145 000; however, as for fraction I, this value may be overestimated [ 15].

In the course of this study, we also noticed the presence of two other pro- tein kinase activities in the thyroid cytosol. The first eluted slightly ahead of fraction II on Bio Gel 1.5 m and, on the basis of its molecular weight should be the protein kinase named K II as described by Leonard et al. [11]. The authors suggested that this fraction was an interconvertible form of the heavier protein kinase comprising different states of aggregation of the regulatory and catalytic subunits. Similar conclusions were reported by Rubin et al. [14] for cyclic

493

AMP-dependent protein kinase from bovine heart muscle. We also report on another protein kinase which eluted from Bio Gel 1.5 m between fractions II and III. This fraction is not associated with a cyclic AMP binding activity peak. An important characteristic of this enzyme is its high specificity for protamine as compared to histone II A. Since catalytic subunits obtained after dissocia- tion of peak I phosphorylated both histone and protamine and since catalytic subunits from type I and type II protein kinases are thought to be identical [29], it seems improbable that the specific protamine kinase would be the free catalytic subunit of fraction Is or I~. Consequently, this kinase should be a spe- cial kind of enzyme or the catalytic subunit linked to some 'modulator' modi- fying its specificity. This question should be of special interest with respect to the following reports: (1) Spaulding et al. [9] reported on a protein kinase from thyroid exhibiting some specificity for protamine and which was cyclic AMP dependent when histone was used as substrate and cyclic AMP independent with protamine as substrate: (2) protein kinases with different specificities for histone and protamine were reported to be independently regulated by differ- ent hormones in rat liver [3]; (3) recently Inoue et al. [31] reported on the existence in the cytosol from 7 different tissues of a proenzyme of molecular weight 77 000 which very specifically phosphorylated protamine; (4) cyclic AMP independent protein kinases have been described to be related to contrac- tile proteins such as light chain myosin [32,33] or parvalbumin from dog-fish skeletal muscle which is phosphorylated by a cyclic AMP-independent protein kinase exhibiting much greater specificity for protamine than for histones [34].

Acknowledgements

The authors are grateful to Jeannine Rapaud for expert assistance in some experiments and to Maryse Guantini and Edith Giraud for their secretarial assistance. This work was supported by the Centre National de la Recherche Scientifique (LA 178) and the Institut National de la Sant~ et de la Recherche Mddicale (U 38).

References

1 Dumont, J.E. (1971) Vitam. Horm. 29, 297--412 2 Field, J.B. (1975) MetaboH~n 24, 381--391 3 Spaulding, S.W. and Burrow, G.N. (1974) Biochem. Biophys. Res. Commun. 59, 386--391 4 Field, J.B., Bloom, G., Kerlns, M.E., Chayoth, It. and Zor, U. (1975) J. Biol. Chem. 250, 4903--4910 5 Roques, M., Tirsxd, A. and Torresani, J. (1973) Biochimie 55, 1421--1430 6 Roques, M., Tirard, A. and Limitzky, S. (1976) MoL Cell. Endocrlnol. 2, 303--316 7 Rapoport , B. and DeGroot, L.J. (1972) Endocrinology 91, 1259--1266 8 Yamashita, K. and Field, J.B. (1972) MetabolL~m 21, 150--156 9 Spaulding, S.W. and Burrow, G.N. (1972) Endocrinology 91, 1343--1349

10 Rappaport, L., Leterrier, J.F. and Nunez, J. (1971) Blochimie 53, 721--726 11 Leonard, J.L. and Rosenberg, L.L. (1977) Bioehim. Btophys. Acta 484, 336--347 12 Gilman, A.G. (1970) Proc. Natl. Acad. SeA. U.S. 67, 305--312 13 Davis, B.J. (1964) Ann. N.Y. Acad. Sei. 121,404--427 14 Rubin, C.S., Erllchman, J. and Rosen, O.M. (1972) J. Biol. Chem. 247, 36--44 15 Erlichman, J., Rubin, C.S. and Rosen, O.M. (1973) J. Biol. Chem. 248, 7607--7609 16 Corbin, J.D., Keely, S.L. and Park, C.S. (1974) J. Biol. Chem. 250, 218--226 17 Walter, U., Uno, I., Liu, A.Y-C and Greengaxd, P. (1977) J. Biol. Chem. 252, 6588--6590 18 SwiBens, S., VanCauter, E. and Dumont , J.E. (1974) Biochim. Biophys. Acta 364, 260--259

4 9 4

19 Swillens, S. and Dumont , J.E. (1975) Biochem. J. 149, 779--782 20 Ogez, J.R. and Segel, I.H. (1976) J. Biol. Chem. 251, 4451--4456 21 MacKenzie, C.W. and Stenwagen, R.H. (1974) J. Biol. Chem. 249, 5763--5771 22 Harbon, S., Dokhac, K. and Vesin, M-F (1976) Mol. Cell. Endocrinol. 6, 17--34 23 Dokhac, L., Harbon, S. and Clauser, H.J. (1973) Eur. J. Biochem. 40, 177--185 24 Walsh, D.A. and Ashby, C.D. (1973) Recent. Prog. Horm. Res. 29, 329--359 25 Wilcheck, M., Salomon, Y., Lowe, M. and Selinger, Z. (1971) Bioehem. Biophys. Res. Commun. 45,

1177--1184 26 Haddox, M.K., Newton, N.E., Hattie, D.K. and Goldberg, N.D. (1972) Biochem. Biophys. Res. Com-

mun. 47,653---661 27 Beavo, J.A., Bechtel, P.J. and Krebs, E.G. (1974) Proc. Natl. Acad. Sci U.S. 71, 3580--3583 28 Tirard, A. (1975) Thesis Universit~ Aix Marseille II, 25--27 29 Bechtel, P.J., Beavo, J.A. and Krebs, E.G. (1977) J. Biol. Chem. 252, 2691--2697 30 Gorin, E., Honeyman, T.W. and Goodman, H.M. (1976) Biochim. Biophys. Acta 451 ,491- -498 31 Inoue, M., Kishimito, A., Takai, Y. and Nishizuka, Y. (1977) J. Biol. Chem. 252, 7610--7616 32 Dabrowska, R., Aromatoria , D., Sherry, J.M.F. and Hartshorne, D.J. (1977) Biochem. Biophys. Res.

Commun. 78, 1263--1272 33 Daniel, J.L. and Adelstein, R.S. (1976) Biochemistry 15, 2370--2377 34 Blum, H.E., Pocin Wong, S. and Fischer, E.H. (1974) Proc. Natl. Acad. Sci. U.S. 71, 2198--2202