Cyclic-amp dependent protein kinase in mouse striatal neurones and astrocytes in primary culture: development, subcellular distribution and stimulation of endogenous phosphorylation

Neurochem. Int. Vol. 14, No. 1, pp. 25-34, 1989 Printed in Great Britain. All rights reserved

0197-0186/89 $3.00+ 0.00 Copyright © 1989PergamonPress plc

CYCLIC-AMP D E P E N D E N T PROTEIN KINASE IN MOUSE STRIATAL N E U R O N E S A N D ASTROCYTES IN PRIMARY CULTURE: DEVELOPMENT, SUBCELLULAR DISTRIBUTION A N D STIMULATION OF E N D O G E N O U S PHOSPHORYLATION S. BIRMAN,* J. CORDIER, J. GLOWlNSKIand H.CHNEIWE1SSt I N S E R M U. 114, Chaire de Neuropharmacologie, Coll~ge de France, 11 place Marcelin Berthelot,

75231 Paris Cedex 5, France (Received 11 July 1988; accepted 18 July 1988)

Al~raet--The cAMP-dependent protein kinase (cAMPdPK) activity and the endogenous cAMPdependent phosphorylation of protein have been studied on pure populations of striatal neurones or astrocytes in primary culture originating from embryonic mouse brain. The appearance of cAMPdPK in cultured striatal neurones was rapid and paralleled neuritic outgrowth and cell maturation. Its highest value, reached between day 6 and 9 in culture, was comparable to that found in adult tissue. In cultured neurones as in adult striatum, cAMPdPK was found both in membrane and in cytoplasmic fractions. In astrocytes, cAMPdPK activity was low and only detectable in the cytoplasm. The presence of several cAMP-regulated phosphoproteins could be demonstrated in cultured cells, some of which being neuronespecific or astrocyte-specific.The phosphorylation of these striatal proteins was either enhanced, or, surprisingly, inhibited in the presence of cAMP. This study indicates that primary cultures of nerve cells provide valuable preparations for analysing protein phosphorylation processes induced by neurotransmitters.

Biological effects of cAMP are thought to be mediated mainly through activation of cAMP-dependent protein kinase (cAMPdPK), which can phosphorylate several cellular soluble or membrane-bound proteins (Krebs and Beavo, 1979; Cohen, 1982; Nestler and Greengard, 1984; Edelman et al., 1987). Inactive cAMPdPK is a tetramer containing two regulatory (R) and two catalytic (C) subunits. Binding of four molecules of cAMP to cAMPdPK promotes dissociation of the hoioenzyme into a R-subunit dimer and two monomeric fully active C subunits. Recently, we have characterized several monoamine and neuropeptide receptors coupled to an adenylate cyclase on neurones and astrocytes from the striatum of mouse embryos grown in primary culture (Chneiweiss et al. 1985a, b, 1986). The present study

was thus undertaken in order to estimate cAMPdPK activity in pure populations of striatal neurones and astrocytes grown in primary culture. The development of this activity was then followed during neuronal maturation in culture and the data obtained in 7-9 day-old neurones were compared to those found in 28 day-old astrocytes and adult striatal tissue. The subcellular distribution of cAMPdPK activity was examined in the three preparations (neurones, astrocytes, adult tissue). Finally, using corresponding subcellular fractions, the pattern of major proteins phosphorylated by endogenous cAMPdPK activity were compared. Most of the adult stfiatal phosphoproteins could be found in cultured cells. Neurone-specific or astrocyte-specific cAMP-regulated phosphoproteins could be demonstrated. In addition, some phosphoproteins were shown to be dephosphorylated in the presence of cAMP.

*Seconded for military service from the D616gation ~i la Recherche et ~ila Technologie (DRET). Present address: D6partement de Neurochimie, Laboratoire de Neurobiologie Cellulaire et Mol6culaire, CNRS, 91198 Gif sur Yvette Cedex, France. tTo whom correspondence should be addressed,

EXPERIMENTAL PROCEDURES

Neuronal cultures

Neuronal cultures were prepared using a procedure originally described by Di Porzio et al. (1980) and slightly 25

26

S. BIRMANet al.

modified according to Weiss et al. (1986). Briefly, culture dishes (60 mm dia; Falcon) were coated successively with poly(L-ornithine) (l.5/~g/ml: MW 40,000; Sigma) and a culture medium containing 10% fetal calf serum which was withdrawn after 2 h of incubation at 37°C. Striatal cells from 16 day/old Swiss mouse embryos (Iffa Credo) were dissociated and plated on rinsed culture dishes (5 × 104 each). The culture medium consisted of a mixture (1:1) of minimal essential medium (MEM) and F-12 nutrient (Gibco) supplemented with 33 mM glucose, 2 mM glutamine, 3 mM sodium bicarbonate, 5 mM Hepes (N-2-hydroxyethylpiperazine.N'2-ethanesulphonic acid), 25 gg/ml insulin, 100 #g/ml transferrin, 20 nM progesterone, 60#M putrescine, and 30 nM selenium salt NaSeO 3. Under these conditions according to different morphological and immuno-histochemical (number of cells stained with an antibody against the 70 K neurofilament protein) criteria, the number of non-neuronal cells represented < 5% of the total cell population until 15 days in culture. Glial primary cultures The protocol used for neurones was also used to start glial cultures. However, the medium included 10% Nu-serum (Collaborative Research) and was changed every 3 days for 4 weeks until glial elements had formed a confluent monolayer, devoid of neuronal cells. As previously described, the absence of surviving neurones was checked by indirect immunofluorescence using an antibody against the 70 K neurofdament protein (Chneiweiss et al. 1985b). More than 95% of the cells could be stained significantly with the immunofluorescent technique using a rabbit antibody directed against glial fibrillary acidic protein (GFAP), indicating that cells were mainly mature astrocytes. A rat monoclonal antibody against galactocerebroside and a rabbit antiserum against fibronectin were used to reveal the possible presence of oligodendrocytes and fibroblasts respectively. No staining could be observed, indicating that these cell types were not present in the cultures. The cultures were also checked to confirm the absence of macrophages using a non-specific fluorescent immunoglobulin. Homogenate preparation After the indicated times for neurones, and 4 weeks for astrocytes, the culture medium was removed and the cells washed 3 times at room temperature in phosphate buffered saline (PBS). Ice-cold lysis buffer [10 mM Tris-HCl (pH 7.4), 2 mM EDTA, I mM dithiothreitol, 0. I mM phenylmethylsulphonylfluoride (PMSF, Sigma), 100U/ml aprotinin (Boehringer), 0.3 ml per dish] was then added and the cells were detached using a rubber policeman. The resulting suspension was homogenized at 4°C by 10 strokes of a Teflon pestle in a Potter-Elvehjem glass homogenizer. Adult striata were dissected rapidly from decapitated female mouse brain and homogenized as above in ice-cold lysis buffer (0.5% w/v). Subcellular fractionation The fractionation procedure was that described by Sobel and Tashjian (1983) with a few modifications. The cellular lysate was centrifuged for 10 s in an Eppendorf (model 3200) table-top centrifuge to remove nuclei and cell debris. The clarified homogenate obtained was then centrifuged for 5 rain at 30 p.s.i. (139,000g) in a Beckman Airfuge. The resulting supernatant was used as the "cytoplasmic" fraction. The pellet was resuspended and recentrifuged for 5 min in the

Airfuge. The final pellet was resuspended in lysis buffer (one-third of the initial volume of the homogenate) and referred to as the "membrane" fraction. All fractions were frozen at the same time in liquid nitrogen, 30-45 rain after collection. Biochemical determinations Protein was determined by the colorimetric method of Bradford (1976) using bovine serum albumin as a standard. The DNA content was estimated in cellular homogenates with the microfluorometric method of Cesarone et al. (1979) using 33258 Hoechst fluorochrome. cAMP-dependent protein kinase assay Several substrates including bovine serum albumin (fatty acid free grade V, Pentex), protamine sulphate, casein sulphate, histones type II S, II AS and III (Sigma) and Synapsin I (a generous gift from P. Greengard) were tested. The final assay selected for the estimation of cAMPdPK activity was based on the phosphorylation of histone II AS. Cellular fractions (10 # g protein) were mixed with 25 ttg of histone II AS in a reaction medium containing 20 mM 3-(N-morpholino) propane sulphonic acid (MOPS), 10 mM Mg acetate, 1 mM EDTA, 1 mM dithiothreitol, 0.05% (v/v) Triton X100, (pH 6.6) After 1 min preincubation at 21°C with (or without) 5/zM cAMP and l mM isobutylmethylxanthine (IBMX, Sigma) to prevent cAMP degradation, the reaction was started by further addition of 50/z M [~,-32p]ATP (New England Nuclear; 0.5 #Ci per assay tube). The incubation was carried out at 21°C for 10 min in a final volume ofS0/zl. The reaction was stopped by blotting 25 pl of the mixture onto Whatman Phosphocellulose/P81 filter paper (1 x 2 cm) which was then rinsed with tap water as described by Witt and Roskoski (1975). In these conditions, the reaction was found to be linear with time up to 15 min (data not shown). Covalent incorporation of radioactive phosphate into the histone II AS substrate was estimated by liquid scintillation counting. Endogenous phosphorylation o f striatal proteins Incorporation of [32p]phosphate into cytoplasmic or membrane proteins was demonstrated by incubating aliquots (35 ~tg proteins) of the subcellular fractions in a reaction mixture containing 50mM Hepes (pH7.4), 10raM Mg acetate, 2 mM EDTA, 1 mM EGTA and l mM dithiothreitol in a final volume of 0.1 rrd. The cAMP-dependent phosphorylation was stimulated by adding 2 #M cAMP and 1 mM IBMX (Walaas et al. 1983c). After I min preincubation at 21°C, the reaction was initiated by adding 20/zM [~, - ~2p]. ATP (10#Ci) and stopped after 30 (membranes) or 90s (cytoplasm) by the addition of 25/~1 of electrophoresis buffer containing 12.5% (w/v) SDS and 25% (v/v) fl-mercaptoethanol, and boiling for 5 min. Gel electrophoresis of the samples was performed in 7.5 or 12% polyacrylamide gels according to the method of Laemmli (1970). After fixing and drying, gels were exposed for autoradiography for 1-5 days using Fuji RX films and Lanex Regular Intensifying screens (Kodak). RESULTS (1) In vitro maturation o f mouse striatal neurones Morphological differentiation o f striatal neurones grown in primary culture can easily be observed 24 h

cAMP-dependent protein kinase in striatal neurones and astrocytes after plating. This complex process is characterized by classical modifications of the cell bodies and increased growth and intricacy of the neuritic arborization (data not shown). The biochemical parameters used to follow cell maturation were D N A content, which reflects the number of surviving cells at each time, and protein content. A multiphasic decrease in D N A content was observed in the course of the culture (Fig. 1). Following a rapid exponential decrease (tm/2= 1 day, phase I), about one-third of initial D N A content was lost during the first 3 days of the culture. From day 4 to day 9, a plateau was reached (phase II), indicating that cell death was reduced markedly. A second rapid fall (phase III) in D N A content was then observed around day 10. Thereafter (phase IV), the D N A levels, which represented 25 % of initial D N A content, remained stable until complete death of the culture, which occurred between day 17 and day 21. As shown in Fig. 1 (left panel), the evolution of protein content followed a bell-shaped curve, maximal values being reached between day 4 and day 9, a period corresponding to phase II of the culture. The protein/DNA ratio, which provides an estimation of the protein content per cell, appears to be a reliable biochemical index of neuronal maturation (Fig. 1, right panel). A 3-fold increase in the protein/DNA ratio, which paralleled the progressive neuritic development, was observed from the onset of culture to day 7. This ratio then reached a value of 3.5 to 4 which hereafter did not change significantly, indicating that the

protein content per cell had stabilized in the surviving neurones.

(2) cAMP-dependent protein kinase activity in striatal neurones Homogenates from 7 day-old neuronal primary cultures and adult mouse striata were first compared for their cAMP-dependent protein kinase activity. Phosphorylation of several exogenous basic protein substrates was enhanced significantly when cAMP was added to the reaction medium (Table 1). Furthermore, this effect was slightly more pronounced with homogenates of neuronal cultures than with those of adult striatal tissue. The best protein substrates were synapsin I and histone II AS, whose phosphorylation was stimulated more than 2-fold. In subsequent experiments, histone II AS was chosen to detect cAMPdPK activity. The maturation of cAMPdPK activity was estimated during in vitro growth of striatal neurones. As shown in Fig. 2, cAMPdPK activity was undetectable in striatal cells from 16 day-old mouse embryos before plating and after 1 day in culture. After 3 days in culture the specific activity of cAMPdPK (expressed per mg protein) approached its maximal value. This rapid appearance of cAMPdPK activity was even more impressive when data were expressed per mg of D N A (index of cAMPdPK activity per neurone): the increase in activity after 3 days in culture was some three times higher than when expressed per mg of protein [Fig. 2, right panel]. Protein kinase activity

NEURONES

:! 6OO

200

~

100

3OO

o a-

o a.

o ~ DAYS

27

i

•

~ ~,o

i

IN CULTURE

Fig. 1. In vitro maturation of mouse striatal neurones. Left panel: cells from 8 striata from 16 day-old mouse embryos were dissociated and either homogenized in lysis buffer (day 0), or plated in dishes and grown in a neuronal culture medium as described in Experimental Procedures (day 1-14). After various times in primary culture, cells were homogenized and their total DNA and protein contents were determined. Right panel: Variations in the protein/DNA ratio as a function of time in culture, given as an index of neuronal maturation (see text). Each point is the mean + SEM of values obtained in three diffe~nt culture experiments. Each individual value was the result of a triplicate determination.

2~

S. BIRMAN et al. Table I. cAMP-dependent protein kinase activity in homogenates from mouse adult striatum and pure striatal neurones in primary culture Adult striatum Exogenous substrates

cAMP

BSA

-+ + + + + ÷

Protamine Histone I1 S Histone II AS Histone 1II Synapsin 1

3:P incorporated (counts/min) 1110 887 25,559 24,151 1863 1999 2674 5280 3680 5086 529 1470

Neurones

Ratio +_

32p incorporated (counts/rain) 1571 1826 45,405 42,444 2302 3829 5424 12,307 6582 10,265 595 1506

0.8 1.0 1A 2.0* 1.4 2.8

Ratio + 1.2 1.0 1.8" 2.3* 1.6" 2.5*

Pure striatal neurones grown for 7 days in primary culture or adult striatum were homogenized in lysis buffer. After elimination of nuclei and cell debris by a short (10 s) centrifugation in an Eppendorf microfuge, an aliquot of the resulting homogenates (10 #g protein) was added to 25/zg of each exogenous protein substratc in the reaction medium (see Experimental Procedures). After 1 rain preincubation at 21"C with ( + ) or without ( - ) 5/~M cAMP plus 1 mM [BMX, the reaction was started by the addition of 50/~M [yJ2P]ATP (0.5/~Ci), and terminated after 10 rain by blotting onto Whatman PSI paper. The data are the mean of results from two independent experiments, each individual value being determined in triplicate. Ratios (mean results of two experiments, SEM always < 5%) indicate the fold-stimulation factor of phosphorylation in the presence of cAMP. *Statistically significantly different from 1 for P < 0.01 using the Student's t-test.

STRIATAL Per mg ~-~ 1501

NEURONES Per mg

of protein

///~'~.C

Of D N A

A MP

e~

~F i ©300 ~.-..

i~oo

-4 '°I//

n.

2

4

6

8

n

10 12 14

DAYS

2

IN

4

6

8

10 ll2 14

CULTURE

Fig. 2. Time course of appearance of cAMP-dependent protein kinase activity during development of striatal neurones in primary culture. Left panel: cells from 8 striata from 16 day-old mouse embryos were dissociated, grown in primary culture, and homogenized at various times as described in Fig. 1. After elimination of nuclei and cell debris, homogenates were frozen quickly in liquid nitrogen. Samples from one culture experiment (day 0-14) were assayed simultaneously, and their protein kinase activity was determined using histone II AS as the exogenous substrate. The reaction was performed in the presence (O O ) or absence ( 0 O ) of 5 # M cAMP plus 1 rnM IBMX. cAMP-dependent protein kinase activity ( A A ) was determined by subtracting values obtained with and without cAMP. Data presented are those from one representative experiment out of three giving similar results. Each point is the mean of triplicate determinations (SEM always <5%). Right panel: data obtained in the same experiment were expressed per mg DNA. The hatched zone indicates the stage of neuronal development period used for further studies on cAMP-dependent phosphorylation of endogenous neuronal proteins.

cAMP-dependent protein kinase in striatal neurones and astrocytes

29

Striatot neurones

7-8

doys in eutture

Fig. 3. Micrograph of striatal neurones in culture. Cells were grown in serum-free medium as described in Experimental Procedures. This mierograph ( x 400) illustrates the neurones used for protein phosphorylation experiments.

hardly changed during phase II and III of the culture, however its highest value always was observed at day 7. A slight but significant decrease in cAMPdPK activity occurred after day 13 (Fig. 2, right and left panel). Further studies on cAMPdPK activity and endogenous physphorylation were made between day 6 and 9 in culture [hatched region in Fig. 2 (right panel), see Fig. 3].

(3) Subcellular distribution of cAMP-dependent protein kinase activity in striatal neurones and astrocytes in primary culture. Comparison with adult striatal tissue As found in the adult mouse striatum, both cytoplasmic and membrane fractions from pure striatal neurones in primary culture (Phase II) contained cAMPdPK activity (Table 2). Specific activities were comparable in cultured neurones and adult tissue and regularly were found to be higher in the membrane fraction. In striatal astrocytes, however, cAMPdPK activity was much weaker and only detectable in the cytoplasmic fraction where it represented only 12% of that found in neuronal homogenates (Table 2).

(4) cAMP-dependent phosphorylation of membrane and cytoplasmic proteins in neurones, astrocytes and strial adult tissue When membrane or cytoplasmic fractions of striatal neurones were incubated in the presence of [7-3zP]ATP, but without known activators of protein kinases, such as Ca 2+ ions or cAMP, several cellular proteins nonetheless were rapidly phosphorylated. These could be detected by autoradiography after polyacrylamide gel electrophoresis. In neurones, activation of cAMPdPK by cAMP (5/~ M) in the presence of IBMX stimulated the incorporation of radioactive phosphate into several membranes or cytoplasmic proteins, which might be considered endogenous protein substrates for the kinase. Representative autoradiograms are shown in Fig. 4, lanes N. Similar experiments performed with cytoplasmic fractions from cultured astrocytes led to comparable results [Fig. 4(c), lane G]. In contrast, when membrane fractions from astrocytes were used, cAMP had virtually no effect on the phosphorylation pattern [Fig. 4(a), lane G], as would be suggested by the absence of detectable cAMPdPK activity. Experi-

30

S. BIP,MANet

al.

Table 2. cAMP-dependent protein kinase activity in subcellular fractions from adult striatal tissue and striatal neurones or astrocytes grown in primary culture Protein kinase activity (pmol 32P/min/mg protein) Subcellular fraction Adult striatum

C M C M C M

Neurones Astrocytes

N

Proteins (%)

- cAMP

+ cAMP

A

3 3 8 6 3 3

43 27 56 22 64 9

145 _ 20 175 + 18 52 + 13 167 + 23 42 ± 5 81+24

239 + 41 448 ± 51 152 + 23 351 + 47 59 ± 7 78+25

94 + 20 274 _+61 100 ± 17 184 ± 38 17 + 4 ND

Striatal neurones (7-9 days in culture), astrocytes (4 weeks in culture) and adult mouse striatum were homogenized and subfractionated as described in Experimental Procedures. Protein kinase activity was determined in membrane (M) and cytoplasmic (C) fractions using histone II AS as the exogenous protein substrate, with or without cAMP (5/~M) plus IBMX (1 raM) added to the reaction medium. The difference in the results obtained with and without cAMP (A) was considered to be the cAMP-dependent protein kinas¢ specific activity. Experiments made with the Protein Kinase Inhibitor (with or without cAMP) provided data identical to those found without cAMP (not shown). Data are the mean (+ SEM) results of N independent experiments. Proteins are expressed as a percentage of the initial protein content in each homogenates (protein recovery was about 85%). ND, non detectable. (see Fig. 4). A s i m i l a r result w a s o b t a i n e d w i t h cytoplasmic fractions from astrocytes. F u r t h e r c o m p a r i s o n s o f r e s u l t s o b t a i n e d with astrocytes and neurones indicated that some cAMPr e g u l a t e d p h o s p h o p r o t e i n s were a s t r o c y t e - s p e c i f i c o r n e u r o n e - s p e c i f i c . A few c A M P - r e g u l a t e d p h o s p h o p r o t e i n s f o u n d in t h e a d u l t were n o t o b s e r v e d in c u l t u r e d cells ( T a b l e 3, a d u l t tissue specific).

m e n t s were also p e r f o r m e d w i t h t o t a l h o m o g e n a t e s o f a s t r o c y t e s . H o w e v e r , since t h e p r o t e i n s in glial m e m b r a n e s c o n t r i b u t e o n l y 1 0 % to t h e p r o t e i n c o n t e n t o f t h e w h o l e h o m o g e n a t e ( T a b l e 2), t h e p r o t e i n s p h o s p h o r y l a t e d in t h e p r e s e n c e o f c A M P were s i m i l a r to t h o s e o b s e r v e d for t h e c y t o p l a s m i c f r a c t i o n o f a s t r o cytes, e x c e p t for o n e b a n d w i t h a n a p p a r e n t m o l e c u l a r w e i g h t o f 58 k D a ( d a t a n o t s h o w n ) . N e a r l y 50 a u t o r a d i o g r a m s were a n a l y s e d to identify h i g h a n d low m o l e c u l a r w e i g h t p r o t e i n s p h o s p h o r y l a t e d in t h e presence o f c A M P . T h e results s u m m a r i z e d in T a b l e 3, p e r t a i n to t h o s e m a j o r p h o s p h o p r o t e i n s r e p r o d u c i b l y o b s e r v e d in t h r e e i n d e p e n d e n t cell culture and fractionation experiments or more. Major c A M P - r e g u l a t e d p h o s p h o p r o t e i n s f o u n d in t h e c y t o plasmic and membranes fractions of cultured neurones were also p r e s e n t in a d u l t t i s s u e a l t h o u g h t h e y f r e q u e n t l y e x h i b i t e d different labelling i n t e n s i t i e s

(5) Inhibition by cAMP of phosphorylation of certain proteins The phosphorylation of certain proteins was i n h i b i t e d in t h e p r e s e n c e o f c A M P , p a r t i c u l a r l y in n e u r o n a l m e m b r a n e s [see for i n s t a n c e a 45 k D a p r o t e i n in Fig. 4(b)]. T h e cellular l o c a l i z a t i o n a n d t h e apparent molecular weight of these phosphoproteins a r e i n d i c a t e d in T a b l e 3.

Table 3. Cellular localization of striatal phosphoproteins regulated by endogenous cAMP-dependent protein kinase Neurones Membrane Phosphorylation stimulated by cAMP

290 275 130

Phosphorylation inhibited by cAMP

80 74 55 170 52 45

Astrocytes Cytoplasmic

34 27 24.5 21.5 17.5 33 28

270* 185" 160*

48* 45 33 60* 43

Cytoplasmic 30* 27.5 26* 21.5* 17.5

Adult tissue specific Membrane Cytoplasmic

70* 45

84 66

37,5*

58.5

33 27.5 17.5 43

62.5

19

None

Autoradiograms similar to those presented in Fig. 4 were analyzed and the pattern of cAMP-regulated phosphoproteins was compared in striatal neurones or astrocytes in primary culture and in adult striatum. Apparent molecular weights of major proteins whose phosphorylation was clearly enhanced or inhibited in the presence of cAMP in three independent experiments or more are listed. Apparent molecular weights are expressed in kDa. Asterisks (*) indicate neurone- or astrocyte-specific cytoplasmic phosphoproteins. Adult tissue specific refers to phosphoproteins not observed in neurones or astrocytes in primary culture.This table summarized results obtained after analysis of 50 autoradiograms, obtained from 10 (neurones), 5 (astrocytes), and 5 (adult tissue) independent experiments.

cAMP-dependent protein kinase in striatal neurones and astrocytes

~

O ~o .~

~

,

O .~

~.~~ ~ o

÷ I

o

+ t

oo

oo

•

•

~.

•

o u~

.~ ,',

O

o

~

Y

VY

O~

~

,~

~

V

Y

V

Y

z c o

,x E

' ~ 0 '~"

~

N

!

g

!

!

o ;~~ ~ o ~

o

~Z •

~ o I~ ,-~ ~

~ ~.~

~

~ N _ ~ . ~

~'~

~-~..~

-~

.,, •

~

~

~,,~

31

32

S. BIRMAN et ai.

DISCUSSION In homogenates from 7 day-old neurones, the phosphorylation of exogenous proteins such as synapsin I and histone (II S, II AS or III) was increased in the presence of cAMP to an extent and with a protein substrate specificity very similar to that observed in adult striatal tissue. The development of cAMPdPK activity in fact was very rapid: although undetectable in 1 day-old neuronal culture, it had reached adult levels at day 5. Neuritic outgrowth seems, therefore, to be accompanied by the appearance and fast synthesis of cAMPdPK which may play a role in early stages of differentiation. Such a rapid synthesis was also observed for protein kinase C in neuronal primary cultures from rat brain (Burgess et al., 1986). Subcellular estimations of neuronal cAMPdPK activity were made in 6-9 day-old cultures since at this stage the maturation index (protein/DNA ratio) was high, cell death was virtually absent and the specific activity of cAMPdPK was close to its maximal value. Neuronal cAMPdPK activity was found both in the membrane and cytoplasmic fractions, as was also observed in the adult striatum; this is in agreement with previous studies made with rat (Maeno et al., 1971) or bovine (Rubin et al., 1979) cerebral tissue. Although we have recently found cAMPdPK activity in growth cones isolated from postnatal 5 day-old rat brain (Lockerbie and Chneiweiss, unpublished observations), little is yet known concerning the precise intracellular localization of the enzyme in growing striatal neurones. It might be largely present in neuritic processes, since one-third of the total soluble form of cAMPdPK has been reported to be tightly linked with microtubule-associated protein (MAP 2) in the calf cerebral cortex (Theurkauf and Vallee, 1982). After 28 days in culture, striatal astrocytes were at their maximal development, forming a confluent monolayer devoid of neuronal contamination. In contrast to the results obtained with neurones, little cAMPdPK activity was found in these striatal astrocytes, the enzyme being present in the cytoplasm but undetectable in membranes. This suggests that the membrane-bound form of cAMPdPK is predominantly, if not totally, neuronal in the mouse striatum. Besides cAMPdPK activity, cAMP-dependent phosphorylation of proteins could be demonstrated in neuronal or glial fractions. Examination of the autoradiographs obtained after gel electrophoresis of labelled phosphoproteins formed in the absence and presence of cAMP allowed several conclusions to be made. (i) On the basis of their apparent molecular

weight, nearly all cAMP-regulated phosphoproteins found in the subcellular fractions of the adult mouse striatum were present in the corresponding fraction of embryonic cells (neurones plus astrocytes) in primary culture. (2) Neurone-specific and astrocyte-specific cAMP-regulated phosphoproteins could be observed. (3) Several neuronal cAMP-dependent highly labelled phosphoproteins were observed both in cytoplasmic and membrane fractions. This was the case particularly for the still unidentified 17.5 and 21.5 kDa phosphoproteins. However, results obtained with one-dimensional gel electrophoresis do not allow one to conclude whether these proteins with similar apparent molecular weight are indeed identical. Comigration with purified proteins and 2D-PAGE analysis are now in progress to clarify these questions. Our results on soluble cAMP-regulated phosphoproteins can be compared to those of Walaas et al. (1983c) who have reported that several rat brain proteins which are phosphorylated in the presence of cAMP are localized specifically in the basal ganglia. For instance the rat 21, 32, 34 kDa phosphoproteins might correspond respectively to the mouse 21.5 and 30. kDa neurone-specific phosphoproteins and to the 33 kDa phosphoprotein found both in the cytoplasmic fractions of neurones and astrocytes from the mouse striatum. In particular, the 30 kDa neuronespecific phosphoprotein which appeared to be the most prominent soluble cAMP-regulated phosphoprotein in the adult mouse striatum, might be the mouse form of DARPP 32~ the dopamine and cAMPregulated phosphoprotein found in large amounts in the rat striatum (Walaas et al.. 1983a). Several studies have looked for endogenous phosphorylation in brain m e m b r a n e fractions, and certain phosphoproteins have been well characterized, particularly in synaptic membranes (De Bias et al., 1979; Kelly et al., 1979; Lohmann et al., 1980; MartinezMillan and Rodnight, 1982; Walaas et al., 1983b). One of these phosphoproteins, synapsin I, which corresponds to the 86-80 kDa doublet (Lohman et al.. 1980) heavily phosphorylated in the presence of cAMP in the membrane fraction from the adult mouse striatum, was represented only by a faint 80 kDa phosphoprotein in mouse striatal neurones in primary culture. Similarly the 66kDa cAMPregulated phosphoprotein, another major membranebound phosphoprotein found both in the rat (Walaas et al., 1983b) and the adult mouse (this study) was not detected in the membrane fraction of cultured striatal neurones. This may be related to an incomplete differentiation of 7-9 day-old striatal neurones in culture. Weiss et al. (1986) indeed have reported that

cAMP-dependent protein kinase in striatal neurones and astrocytes synapsin I is detectable by immunofluorescent staining in neuronal varicosities only after 12 days in culture. It should be added that synapsin I appears also as a single protein band in PC 12 cells not treated with N G F (Romano et al., 1987). The abundance of synapsin I and of the 66 kDa protein in the adult tissue could also be related to their localization in striatal afferences. In contrast with the above observations, the 74 and 55 kDa cAMP-regulated phosphoproteins were found in membranes from both adult striatal tissue and cultured striatal neurones. They might correspond to proteins Ilia and IIIb which were found to be highly concentrated in nerve terminals (Huang et al., 1982; Nestler and Greengard, 1984). Surprisingly, our experiments revealed the existence of a class of phosphoproteins in which the incorporation of labelled phosphate was strongly decreased in the presence of cAMP. These phosphoproteins, less numerous than the cAMP-stimulated ones, were either neurone-specific (the 60 kDa soluble phosphoprotein or the 52 and 28 kDa membrane-bound phosphoproteins) or common to striatal neurones and astrocytes (i.e. the cytoplasmic 43 kDa phosphoproteins. Recently, it has been also shown that cAMP can reduce the state of phosphorylation of myelin basic proteins in mixed rat brain cells cultures containing oligodendrocytes (Ulmer et al., 1987). Moreover, the activation ofcAMPdPK was found to inhibit phorbol ester-dependent phosphorylation of several proteins in $49 mouse lymphoma cells (Kiss and Steinberg, 1985). Both a cAMP-dependent inactivation of protein kinases other than cAMPdPK, and cAMPdependent activation of specific phosphatases might be involved in the cAMP induced reduction in 32p_ incorporation into several proteins observed in our experiments. In conclusion, the present results demonstrate that both striatal neurones and astrocytes from the mouse grown in primary culture possess a cAMPdPK activity and the machinery required for the endogenous phosphorylation of proteins. Interestingly, the pattern of phosphorylated proteins in the presence of cyclic AMP in neurones and/or astrocytes was similar to that found in adult striatal tissue but neurone-specific and astrocyte-specific phosphoproteins could be demonstrated. It remains now to be determined whether or not any of these cAMP-regulated pbosphoproteins are phosphorylated under the action of neurotransmitters acting on receptors coupled to adenylate cyclase in intact neurones or astrocytes in primary culture. Acknowledgements--We wish to thank Dr J. Pr6mont for his support and advises, and Dr A. Sobel for stimulating N.C.I. 1 4 / 1 ~

33

discussions and helpful suggestions on the manuscript. This work was supported by Grants from INSERM, DRET (85.078) and Rh6ne Poulenc SA. REFERENCES

Bradford M. M. (1976) A rapid and sensitivemethod for the quantification of microgram quantities of protein utilizing the principle of protein-dye binding Analyt. Biochem. 72, 248-254. Burgess S. K., Sahyoun N., Blanchard S. G., Le Vine H. III, Chang K. J. and Cuatrecasas P. (1986) Phorbol ester receptors and protein kinase C in primary neuronal cultures: development and stimulation of endogenous phosphorylation. J. Cell Biol. 102, 312-319. Cesarone C. F., Bolognesi C. and Santi L. (1979) Improved microfluorimetric DNA determination in biological material using 33258 Hoechst. Analyt. Biochem. 100, 188-198, Chneiweiss H., Glowinski J. and Pr6mont J. (1985a) VIP receptors linked to an adenylate cyclaseand their relationship with biogenic amines- and somatostatin-sensitive adenylate cyclases on central neuronal and glial cells in primary cultured cells. J. Neurochem. 44, 779-786, Chneiweiss H., Glowinski J. and Pr~mont J. (1985b) Modulation by monoamines of somatostatin-sensitiveadenylate cyclase on neuronal and glial cells from the mouse brain in primary cultures. J. Neurochem. 44, 1825-1831. Chneiweiss H., Glowinski J. and Pr~mont 3. (1986) Do secretin and vasoactive intestinal peptide have independent receptors on striatal neurons and glial cellsin primary cultures? J. Neurochem. 47, 608-613. Cohen O. (1982) The role of protein phosphorylation in neural and hormonal control of cellular activity. Nature 296, 613-620. De Blas A. C., Wang Y. J., Sorensen R. and Mahler H. R. (1979) Protein phosphorylation in synaptic membranes regulated by adenosine 3'-5'-monophosphate: regional and subcellular distribution of the endogenous substrates. J. Neurochem. 33, 647-659. Di Porzio U., Daguet M. C., Glowinski J. and Prochiantz A. (1980) Effect of striatal cells on /n rive maturation of mesencephalic dopaminergic neurones grown in serum free conditions. Nature 288, 370-373. Edelman A. M., Blumentahl D. K. and Krebs E. G. (1987) Protein serine/threonine kinases. A. Rev. Biochem. 56, 567-613. Huang C. K., Browning M. D. and Greengard P. (1982) Purification and characteriation of protein IIIb, a mammalian brain phosphoprotein. J. biol. Chem. 257, 6524-6528. Kelly P. T., Cottman C. W. and Largen M. (1979) Cyclic AMP-stimulated protein kinases at brain synaptic junctions. J. biol. Chem. 254, 1564--1575. Kiss Z. and Steinberg R. A. (1985) Interaction betweencyclic AMP- and phorbol ester-dependent phosphorylation system in $49 mouse lymphoma cells. J. Cell Physiol. 125, 200-206. Krebs E. G. and Beavo J. A. (1979) Phosphorylation and dephosphorylation of enzymes. A. Rev. Biochem. 48, 923959. Laemmli U. K. (1970)Cleavage of structural proteins during the assembly of the head of the bacteriophage T4. Nature 227, 680-685.

34

S. BIRMANet al.

Lohmann S. M., Walter U. and Greengard P. (1980) Identification of endogenous substrate proteins for cAMP-dependent protein kinase in bovine brain. J. biol. Chem. 255, 9985 9992. Maeno H.. Johnson E. M. and Greengard P. (1971) Subcellular distribution of adenosine-Y-5'-monophosphate dependent protein kinase in brain. J. biol. Chem. 246, 134- 142. Martinez-Millan L. and Rodnight R. (1982) Intrinsic protein phosphorylation in synaptosomal plasma membrane fragments. A comparison of cerebral cortex tissue from several species, including human biopsy specimen. J. Neurochem. 39, 1302-1307. NestLer E. J. and Greengard P. (1984) Protein Phosphorvlation in the Nervous System. Wiley, New York. Romano C., Nichols R. A., Greengard P. and Greene L. A. (I 987) Synapsin I in PC 12 cells. I. Characterization of the phosphoprotein and effect of chronic NGF treatment. J. Neurosci. 7, 1294~1299. Rubin C. S., Rangel-Aldao R., Sakar D., Erlichman J. and Flescher N. (1979) Characterization and comparison of membrane-associated and cytosolic cAMP-dependent protein kinases. J. biol. Chem. 254, 3797 3805. Sobel A. and Tashjian A. H. Jr (1983) Distinct patterns of cytoplasmic protein phosphory-iation related to regulation of synthesis and release of prolactin by GH cells. J. biol. Chem. 258, 10312-10324. Theurkauf W. E. and Vallee R. B. (1982) Molecular charac-

terization of the cAMP-dependent protein kinase bound to microtubule-associated protein 2. J. biol. Chem. 257, 3284-3290. Ulmer J. B., Edwards A. M., McMorris F. A. and Braun P. E. (1987) Cyclic AMP decreases the phosphorylation state of myelin basic proteins in rat brain cell cultures. J. biol. Chem. 262, 1748-1755. Walaas I., Aswad D. W. and Greengard P. (1983a) A dopamine- and cyclic AMP-regulated phosphoprotein enriched in dopamine-innervated brain regions. Nature 301, 6~71. Walaas 1., Nairn A. C. and Greengard P. (1983b) Regional distribution of calcium- and cyclic adenosine-Y-5'-monophosphate-regulated protein phosphorylation systems in mammalian brain. 1. Particulate system. J. Neurosci. 3, 291 301. Walaas I., Nairn A. C. and Greengard P. (1983c) Regional distribution of calcium- and cyclic adenosine-Y-5'-monophosphate-regulated protein phosphorylation systems in mammalian brain. I1. Soluble systems. J. Neurosci. 3, 302 311. Weiss S., Pin J. P., Sebben M., Kemp D. E., Sladeczec F., Gabrion J. and Bockaert J. (1986) Synaptogenesis of cultured striatat neurons in serum-free medium: a morphological and biochemical study. Proc. natn. Acad. Sci. U.S.A. 85, 2238-2242. Wilt J. J. and Roskoski R. Jr (1975) Rapid protein kinase assay using phosphocellulose-paper adsorption. Analyt. Biochem. 66, 253 258.