Maturation-related gene expression of rat astroblasts in vitro studied by two-dimensional polyacrylamide gel electrophoresis

Cell Differeniiation and Development, Elsevier Scientific Publishers Ireland,

25 (1988) 37-46 Ltd.

31

CDF 00524

Maturation-related gene expression of rat astroblasts in vitro studied by two-dimensional polyacrylamide gel electrophoresis Camille

Loret

r, Monique

1 and Gerard

Sensenbrenner

’ Centre de Neurochimie du CNRS and ’ INSERM

(Accepted

Labourdette

2

lJ 44, 5, rue Blaise Pascal, 67084 Strasbourg Cedex, France

6 April 1988)

Maturation of rat astroblasts was induced in vitro by acidic fibroblast growth factor (aFGF), hydrocortisone and dibutyryl cyclic AMP. Cells grown for 20 days were treated for 48 h and labelled with [35S]methionine only during the last 18 h of treatment. Cell proteins solubilized in lithium dodecyl sulfate were submitted to two-dimensional polyacrylamide gel electrophoresis. About 300 radioactive proteins could be analysed visually and compared. All treatments induced visible quantitative and sometimes qualitative changes. A total of 81 proteins had their rate of biosynthesis modified. For some proteins, this rate was changed by only one treatment, while for others it was changed by two or even by the three treatments, mostly in the same way. These results are consistent with the hypothesis that the proteins involved in the maturation process are organized in sets, proteins belonging to one set always being regulated together under a common control. Some sets would be regulated by only one effector while others would be regulated by several effecters. Astrocytes;

Fibroblast growth factor; Maturation in culture; Two-dimensional

Introduction Homogeneous primary monolayer cultures of rat astroblasts can be grown easily starting from brain hemispheres of newborn animals (Booher and Sensenbrenner, 1972). These flat irregularshaped cells are immature when seeded, but some age-dependent maturation develops in vitro. The maturation stage of the cells is usually defined by the relative level of expression of one or two specific proteins. Some proteins, specific or localized predominantly in astroglial cells, in the central

Correspondence address: G. Labourdette, INSERM rue Blake Pascal, 67084 Strasbourg Cedex, France. 0922-3371/88/$03.50

0 1988 Elsevier Scientific

U 44, 5,

Publishers

Ireland,

electrophoresis

nervous system have been characterized. These include the glial fibrillary acidic protein (GFAP) which is in the gliofilaments (Moonen et al., 1976; Bock et al., 1977) S-100 protein (Labourdette and Mandel, 1978) glutamine synthetase (GS) (Martinez-Hemandez et al., 1977) and acu-enolase (Ghandour et al., 1981). The spontaneous maturation of the rat astroblasts is limited in culture (Pettmann et al., 1981) but it can be greatly enhanced by treatment of the cells with several maturation effecters, mainly hormones and growth factors. For instance, the hormone norepinephrine (NE) induces a modification of the astroglial cell morphology (Narumi et al., 1978) the cells then resembling the fibrous astrocytes found in vivo (Shapiro, 1973). Furthermore, dibutyryl cyclic Ltd.

38

AMP (db-CAMP) also induces an increase in the level of GFAP (Sensenbrenner et al., 1980) and in GS activity, and a morphological change. This compound is an active derivative of CAMP which is itself the second messenger mediating the effects of the hormone NE (Narumi et al., 1978). Glucocorticoids do not modify the cell morphology but enhance the GS activity (Juurlink et al., 1981). A growth factor, the astroglial growth factor 1 (AGF l), purified in our laboratory from bovine brain (Pettmann et al., 1985) and probably identical to the acidic fibroblast growth factor (aFGF) (Pettmann et al., 1987) isolated in 1984 by Gospodarowicz et al. (1984), induces a drastic modification of the cell morphology which is different from that induced by db-CAMP (Pettmann et al., 1987), and an increase of GS activity (Weibel et al., 1985) as well as of the level of S-100 protein (Sensenbrenner et al., 1985). As it is difficult to define cell maturation by the determination of the variation of levels of only a small number of proteins, we began to investigate this cellular process by using the technique of two-dimensional polyacrylamide gel electrophoresis (2D-PAGE). This technique allows an overview of the most abundant proteins synthesized in a cell type and an evaluation of the changes in the synthesis which can occur under various circumstances, particularly after treatments with hormones, growth factors or other extracellular signals (we will call them maturation inducers or effecters). The 2D-PAGE has been used in preliminary studies on the maturation of other cell types like hepatomas (Ivarie and O’Farrell, 1978) or mouse 3T3 cells (Thomas and Luther, 1981; Levenson et al., 1985). Its use for analysing the modulation of the protein biosynthesis of astroglial cells has been very scarce (Bridoux et al., 1986). The present study was undertaken as a first approach to gain more insight into the phenomenon of astroglial maturation and to investigate its mechanisms by comparing the effects of various maturation effecters. We used a growth factor (aFGF), a hormone, hydrocortisone, and db-CAMP which we will sometimes refer to as a hormone since, as mentioned before, CAMP is the second messenger of the hormone NE.

Materials and Methods Materials and reagents Falcon tissue culture Petri dishes were purchased from Becton and Dickinson, Waymouth’s MD 705/l medium from Flow Laboratories and fetal calf serum (FCS) from Gibco. Bovine insulin (I-5500), fatty. acid-free bovine serum albumin (BSA) (A-6003), db-CAMP (D-0627), hydrocortisone (H-0888) and Nonidet P-40 (NP-40) (N-6507) were from Sigma. aFGF was purified from bovine brain according to the procedure of Pettmann et al., (1985). [35S]Methionine was supplied by CEA, France. Lithium dodecyl sulfate (LDS) and 3-[(3-cholamidopropyl)dimethyl-ammoniol-l-propane sulfonate (CHAPS) were from Serva. Dithiothreitol (DTT), N, N, N ‘, N ‘-tetramethylenediamine (TEMED), bis-acrylamide and ammonium persulfate were from BioRad. Acrylamide and 2-mercaptoethanol were from Merck. Sephadex IEF and agarose IEF were from Pharmacia. Rat astroblast cultures Cultures were prepared by a modification of the method of Booher and Sensenbrenner (1972). Brain hemispheres of newborn rats were dissected, cleaned of their meningeal membranes, and dissociated by passages through a l-mm diameter needle in a small volume of nutrient medium. This medium was Waymouth’s MD 705/l medium supplemented with sodium pyruvate (110 mg/l), 50 units/ml penicillin, 50 pgg/ml streptomycin, and 10% FCS. The cell suspension was added to a suitable volume of complete nutrient medium to obtain a final concentration of one brain for 60 ml. Two ml of the cell suspension were dispensed per Petri dish (35-mm diameter). Cultures were incubated at 37’C in a 5% CO, humidified atmosphere. Culture medium was changed after 5 days and subsequently twice a week. After 20 days of culture, rat astroglial cells at confluence stage were switched to a serum-free chemically defined medium consisting of Waymouth’s medium supplemented with 5 pg/ml insulin and 0.5 mg/ml BSA, antibiotics and sodium pyruvate (Weibel et al., 1984). aFGF (5 ng/ml), db-CAMP (1 mM) or

39

hydrocortisone (10 PM) were added to the culture just after the culture medium change (final concentrations in the culture medium are given). Labelling of cells with [3sSJmethionine Thirty hours after the addition of the effecters, the cells were labelled for 18 h with 30 PCi of [ 35S]methionine (1200 Ci/mM : 41.4 TBq/mM) in 1 ml of fresh chemically defined medium containing a lower amount of methionine (373 pg/l instead of 50 mg/l) and the effector. The amount of cold methionine in the medium during labelling was adjusted in such a way that methionine was still in excess at the end of the incorporation, while incorporation of [35S]methionine was maximum. After labelling, the radioactive medium was removed and dishes were washed three times with cold 0.15 M NaCl. Cells were scraped off with a rubber policeman and pelleted by low speed centrifugation. The supernatants were removed and the pellets were stored at - 80’ C. Sample preparation Sample preparation was performed as described by Garrels (1979) with some modifications. A solution of 0.3% LDS and 1% 2-mercaptoethanol (solution A) was heated to 100 o C in a water bath. The hot solution A was added to the frozen cell pellets. Samples were immediately vortexed and placed in a boiling water bath for 2 min. Then the samples were quickly frozen and lyophilized. Immediately after release of the vacuum, the samples were dissolved in lysis buffer containing 9.5 M urea, 2% (w/v) CHAPS, 20 mM DTT and 2% (w/v) ampholytes consisting of a mixture of 0.5% (v/v) LKB ampholine pH 3.5-9.5, 2% (v/v) LKB ampholine pH 5-8, 0.5% (v/v) servalyte pH 3-10 and 2% (v/v) pharmalyte pH 5-8. The samples were then either run immediately or frozen at - 80 “C. Aliquots of a volume ranging from 10 to 25 ~1, containing lo6 cpm (about 25 pg protein) were loaded on the gel. Protein concentration of the samples was determined by the method of Lowry et al. (1951). Two-dimensional electrophoresis Two-dimensional electrophoresis was performed essentially as described by O’Farrell(l975).

First dimension. Isoelectric focusing gels were made in glass tubing (180 mm x 1.5 mm i.d.) siliconized with Repel-Silane (LKB 1850-252) and sealed at the bottom with parafilm. The acrylamide gel consisted of 9.2 M urea, 3.78% (w/v) acrylamide, 0.22% (w/v) bis-acrylamide, 2% (w/v) NP-40, 2% (w/v) ampholytes (same mixture as for the lysis buffer). To 10 ml of gel mixture, 7 ~1 of TEMED and 10 ~1 of 10% ammonium persulfate were added, and the solution was immediately put into tubes up to 140 mm height. The gels were overlaid with 10 ~1 of water. After polymerisation (about 1 h), the water overlay was removed and replaced by 10 ~1 of lysis buffer. The tubes were then filled with 0.02 M NaOH and placed in the gel electrophoresis chamber. The lower reservoir was filled with 0.01 M H,PO, and the upper reservoir with degassed 0.02 M NaOH. The gels were then prerun as described by Garrels (1979): initial voltage was 300 V until current stabilization and 1000 V thereafter. NaOH and lysis buffer were removed from the gel tubes and 10 ~1 of Sephadex IEF in lysis buffer (10 mg/ml) added. The excess of liquid was taken off after gel sedimentation. Samples were then loaded and overlaid with 10 ~1 of overlay buffer containing 9 M urea and 1% (w/v) ampholytes (same mixture as for the lysis buffer). The tubes and the upper reservoir were then filled with 0.02 M NaOH. Isoelectric focusing was performed at 1000 V for 19 h. When isoelectric focusing was complete, the gels were placed at once in LDS sample buffer: 10% (w/v) glycerol, 2.3% (w/v) LDS and 0.0625 M Tris-HCl, pH 6.8. They were then frozen at -80°C. Second dimension (slab gel electrophoresis). The second dimension was the LDS-polyacrylamide gel electrophoresis described by Laemmli (1970). Gels (200 x 200 x 1 mm) consisted only of a separation gel containing 15% (w/v) acrylamide (bis-acrylamide/ acrylamide = 0.006). The isoelectric focusing gels were applied to the slab gels and covered with a 1% agarose IEF solution (1 g agarose/lOO ml of LDS sample buffer and containing bromophenol blue). The running buffer (0.025 M Tris, 0.19 M glycine and 0.1% LDS) was poured into the two reservoirs. Gels were run at 20 mA/gel constant current

40

overnight until the front dye reached the bottom of the gel. Drying and autoradiography. The gels were incubated in 50% methanol overnight, dried and exposed to Kodak Diagnostic film X-Omat AR5 at 4°C for 24 h, 4 days or 15 days. -69

Results Treatments of astroblasts with aFGF, hydrocortisone and db-CAMP modulate the biosynthesis of 25% of the most abundant proteins The detergent-soluble radioactive proteins of confluent rat astroblasts in primary culture were analysed by 2D gel electrophoresis. An autoradiogram of a control untreated culture, exposed for 15 days, is shown in Fig. 1. The positions of actin (A), vimentin (V) and GFAP (G) are indicated. They were determined by immunoblotting with antisera kindly provided by J. Cieselski-Treska (not shown). Astroblasts were treated with three inducers: aFGF, hydrocortisone and db-CAMP. Two independent experiments were run in duplicate, giving four different samples per treatment. Two aliquots of the four samples were submitted to 2D gel electrophoresis. Thus, for each treatment, eight different gels were run, giving a total number of 32 gels run. For each gel, three autoradiograms were exposed for 24 h, 4 days or 15 days. The different durations of exposure allow the comparison of spots of different densities. Proteins with a very high rate of synthesis can be compared only on the 24-h exposed autoradiograms. At longer time of exposure, the saturation of the film does not permit any more comparison of these proteins. All the autoradiograms exposed for identical time were compared. A total of 320 spots were considered. After analysis we found that the biosynthesis of 81 proteins was reproducibly modified, compared with the untreated cells. Only these proteins were numbered and circled on Fig. 1. Biosynthesis was increased for 44 proteins, decreased for 29 proteins and increased or decreased depending on the treatments for eight proteins (from results shown in Table I). For most proteins, the variations were seen in all the eight

-46

Fig. 1. Two-dimensional gel electrophoretic pattern of the proteins of cultured rat astroblasts. Cells dissociated from brain hemispheres of newborn rats were grown as monolayer for 20 days. They were allowed to incorporate, for 18 h, [35S]methionine (30 pCi/ml culture medium) in a medium containing a low level of unlabelled methionine. An amount of cell proteins, detergent solubilized, containing lo6 cpm was submitted to 2D gel electrophoresis as described in Materials and Methods. The autoradiogram shown was exposed to the gel for 15 days. Some proteins have been localized. A, actin; V, vimentin; G, GFAP. All the proteins whose rate of biosynthesis is modulated under the effect of aFGF, hydrocortisone or db-CAMP have been circled and numbered (except in the delimited areas) according to the classification shown in Table I. The delimited areas will be shown at higher magnification and after treatments of the cells in the other figures. They are numbered in Roman numerals.

autoradiograms exposed for the same duration. For some proteins, variation was taken into consideration, if it was seen in at least five autoradiograms. We have chosen six areas, indicated in Fig. 1, in which changes of the rate of the biosynthesis of

41 TABLE

I

Classification and numbering of the proteins of rat astroblasts effects of aFGF, hydrocortisone and db-CAMP Protein no.

according

to the modulation

of their rate of biosynthesis

aFGF

Hydrocortisone

db-CAMP

l-4 5-11

+ -

+ -

+

4 7

12-16

-

+

+

5

17-19 20 21.22

+ + -

+

3 1 2

3

+

23,24 25-30 31-33

+

2 6 3

2

1

1

19 8

27

8 3

11

3 6

9

+

34

Number proteins

+ -

+

35-53 54-61

+ -

62-69 70-72

+ -

73-75 76-81

+

of ’

under

the

Number of proteins per set b I1

3

9

The sign+indicates that the biosynthesis is increased and the sign- that it is decreased by the matching treatment. Proteins have been classified and numbered according to their modulation. The first numbers are for the proteins modified by the three treatments and the last for those modulated by only one inducer. The location of the numbered proteins appears on Figs. 1-7. This Table has been established from the comparison of results from two independent experiments run in duplicate, each sample being submitted to two different 2D gel electrophoreses. A total of 96 autoradiograms have been compared. ’ This is the total number of proteins whose biosynthesis is modulated by the various inducers as shown on the same line. b A set is composed of proteins, the syntheses of which are modulated in the same way (+ or -) by the treatments shown.

IEF-

IEF-

IEF-

IEF-

cna c3 v)

Fig. 2. Effect of aFGF, hydrocortisone and db-CAMP on the rate of synthesis of the proteins localized in Area I. Cells were treated for 48 h and labelled with [ 35S]methionine for the last 18 h of treatment. Other specifications as for Fig. 1. (a) Control untreated cells; (b) cells treated with aFGF; (c) treated with hydrocortisone; (d) treated with db-CAMP. Autoradiograms were exposed for 15 days. Arrows show the proteins of interest.

,“av, Fig. 3. Effect of the various treatments on the synthesis of the proteins in Area II. Other specifications as for Fig. 2. (a) Control; (b) aFGF; (c) hydrocortisone; (d) db-CAMP. Proteins shown behave very differently depending on treatment.

42

IEF-

IEF-

V)a

Fig. 4. Comparison of the protein biosynthesis in Area III. Specifications as for Fig. 2, but autoradiograms were exposed for 4 days. (a) Control; (b) aFGF; (c) hydrocortisone; (d) db-CAMP. Area around actin which is the most predominant protein of rat astroblasts. The biosynthesis of the four proteins shown is modified differently by the three treatments.

some proteins are very important or well visible and in which various patterns, concerning the modulations of the biosynthesis by the inducers, are represented. In Fig. 2, which shows area I, the proteins numbered 5, 6 and 11 almost disappear with the three treatments, while protein no. 2 is increased by these treatments. Treatments with hydrocortisone and db-CAMP increased the synthesis of protein no. 28 but decreased that of no. 32. The biosynthesis of the protein no. 78 was decreased after treatment with db-CAMP. In Fig. 3, where area II is enlarged, protein no. 16 is depressed by aFGF but enhanced by the two hormones, while protein no. 43 is induced only by aFGF. Spot no. 49 is seen only after aFGF treatment and spot no. 66 only after hydrocortisone treatment. In area III (Fig. 4), three proteins, nos. 25, 27 and 34, are hardly seen in the control cells (Fig. 4a). Proteins 25 and 27 are strongly labelled after hydrocortisone and db-CAMP treatments. In

IEF-

ma

IEF-

IEFa

b

%,?

Fig. 6. Effect of aFGF on the rate of synthesis of proteins nos. 7 and 8 in Area V. (a) Control cells; (b) cells treated with aFGF. G, GFAP. Autoradiograms were exposed for 15 days.

this area and at the exposure of 4 days, no difference is seen in aFGF-treated cells compared with untreated cells. In area IV (Fig. 5) the biosynthesis of protein no. 10 was almost stopped after aFGF treatment (Fig. 5b). Similar behaviour was seen with the other two treatments (not shown). In area V (Fig. 6) the biosynthesis of two proteins, nos. 7 and 8, was completely inhibited after aFGF treatment (Fig. 6b) and after the two

IEFb

u tn

46 B

-60

X

t

Fig. 5. Effect of aFGF on the rate of synthesis of protein no. 10 in Area IV. (a) Control cells; (b) cells treated with aFGF. Autoradiograms were exposed for 15 days.

Fig. 7. GFAP aFGF; tin;

Effect of the various treatments on the synthesis of and of other proteins in Area VI. (a) Control; (b) (c) hydrocortisone; (d) db-CAMP. A, actin; V, vimenG, GFAP. Autoradiograms were exposed for 24 h.

43

other treatments (not shown). In area VI (Fig. 7) proteins 12 (GFAP), 13, 14 and 15 are seen. Their biosynthesis decreased after aFGF treatment and increased after hydrocortisone and db-CAMP treatments.

12- 16) which were modulated differently by aFGF and the two other effecters (Table I). Table II summarizes the results. It shows that aFGF modified the rate of biosynthesis of 51 proteins, corresponding to about 17% of the proteins analysed while hydrocortisone or db-CAMP modified 40 proteins each and a total of 54 proteins together. The biosynthesis of 47 proteins was altered by only one of the inducers tested, while that of 34 proteins was altered by at least two of the three inducers.

Classification of the proteins whose biosynthesis is modulated Proteins were classified and numbered (Table I) according to the modulation of their biosynthesis: first, proteins whose synthesis changed after the three treatments, then those changed by two treatments and finally those changed by only one treatment. For most proteins whose changes were elicited by two or more effecters, these changes were in the same direction (26 proteins out of 34). For spots nos 12-16, the effect of aFGF was opposite to that of the hormones. A difference was also seen for proteins nos. 23 and 24 between aFGF- and db-CAMP-treated cells and for protein no. 34 between hydrocortisoneand db-CAMPtreated cells. Three sets were defined by the proteins which were changed only by one treatment, 27, 11 and 9 proteins for aFGF, hydrocortisone and db-CAMP respectively. A minimum number of 7 sets of proteins whose biosynthesis was modulated in the same way by one or more treatments was defined. Another set would contain the proteins (nos.

TABLE Counting

Discussion In this report, we have shown that different inducers of the maturation of astroblasts in culture modulate differently the biosynthesis of a number of proteins. Together, they modify the expression of about one fourth of the 300 most abundant proteins. This high ratio justifies a more extensive study of the maturation process which is usually defined by the variation of levels of only a few specific proteins or enzymes. The number of proteins involved in the maturation process seems to be much higher than expected. Comparison of the effects of various effecters indicates possible mechanisms of cell maturation at the level of regulation of protein biosynthesis. The results show that, in our experimental condi-

II of all the proteins

Number of proteins involved a

whose biosynthesis

is modulated

by the various db-CAMP

treatments

aFGF

Hydrocortisone

Hydrocortisone

or db-CAMP

b

Total ’

Only modulated by the treatment(s) shown

27

11

9

20

47

Also modulated other treatments

24

29

31

34

34

51

40

40

54

81

by

Total (modulated by one or more treatment(s))

a For one treatment, one column, the number of proteins modulated by that treatment is shown on the first line, on the second line is the number of proteins modulated by that treatment and also by at least another treatment. Sum of these two numbers is shown on the third line. b In this column proteins modulated by hydrocortisone and db-CAMP have been put together to allow a comparison of the hormones with the growth factor. ’ In this column, on the first line is the total number of proteins modulated by only one treatment (27 + 11+ 9). On the second line is the total number of proteins modulated by two or three treatments, calculated from data of Table I.

44

tions, the synthesis of some proteins is modulated by only one effector, but that the synthesis of many other proteins is regulated by two or more effecters. Some correlations can be proposed between some sets of proteins and some known effects of the compounds tested. For instance, on astroblast proliferation, aFGF is stimulatory, hydrocortisone without effect and db-CAMP inhibitory. Since only db-CAMP inhibits the proliferation, some of the proteins modulated only by db-CAMP (nos. 73-81) could be involved in this inhibition. In contrast, proteins involved in the stimulation of proliferation could be found in the proteins modulated only by aFGF (nos. 35-61) or among the proteins modulated differently by aFGF and the hormones (nos. 12-16 and 23, 24). On the other hand, there are proteins whose biosynthesis is modulated in the same direction by aFGF and at least by another treatment, hydrocortisone or db-CAMP or both (proteins nos. 1-11 and 17-22). These proteins should be unambiguously involved in the cell maturation process, like the nine proteins (nos. 25-34) modulated by the two hormones only and the 20 proteins (nos. 62-81) modulated by one hormone. The proteins in the first set (nos. l-11), the rate of biosynthesis of which is modified in the same way by the three treatments, could have a special function. They could participate in a common mechanism involved after the action of any maturation inducer. The effect of hydrocortisone on cell morphology is very weak, while aFGF and db-CAMP modify it drastically. Protein no. 20, the only protein induced specifically by aFGF and dbCAMP, could be related to this effect. Our results suggest a high level of complexity for the maturation process. As a first approximation, the proteins involved in the processes of proliferation and maturation could be organized in sets, all proteins in a set always being regulated together. Each effector would regulate various sets. Some of them would be regulated by only one inducer, while others would be controlled by several inducers. Some sets could be related to the proliferative effects and some others to the maturation effects.

Several other effecters will be tested and quantitative analysis will be done in order to elaborate a more precise model.

Acknowledgements We thank Marc Weibel, Genevieve Daune and Brigitte Pettmann who participated in the preliminary experiments, Robert Wahlen from the Institut Pasteur (Paris), Pierre Vincens and Philippe Tarroux from the Ecole Normale Superieure (Paris), Hervt Thiellement from the LGSV (Gif sur Yvette) and Yarka Cieselski-Treska from INSERM U44 (Strasbourg) for discussions and suggestions concerning the 2D electrophoresis technique. We thank Marie-France Knoetgen for astroblast cultures. We are indebted to Serge Stein for the making of the electrophoresis apparatus. This work was supported in part by a grant from the ARC (Association pour la Recherche sur le Cancer, Villejuif, France).

References Bock, E., M. Moller, C. Nissen and M. Sensenbrenner: Glial fibrillary acidic protein in a primary astroglial cell culture derived from newborn rat brain. FEBS Lett. 83, 207-211 (1977). Booher, J. and M. Sensenbrenner: Growth and cultivation of dissociated neurons and glial cells from embryonic chick, rat and human brain in flask cultures. Neurobiology 2, 97-105 (1972). Bridoux, A.M., C. Fages, D. Couchie, J. Nunez and M. Tardy: Protein synthesis in astrocytes: “spontaneous” and cyclic AMP-induced differentiation. Dev. Neurosci. 8, 31-43 (1986). Garrels, J.I.: Two-dimensional gel electrophoresis and computer analysis of proteins synthesized by clonal cell lines. J. Biol. Chem. 254, 7961-7977 (1979). Ghandour, MS., O.K. Langley and A. Keller: A comparative immunohistological study of cerebellar enolases: double labelling technique and immunoelectronmicroscopy. Exp. Brain Res. 41, 271-279 (1981). Gospodarowicz, D., J. Cheng, G.-M. Lui, A. Baird and P. Bohlen: Isolation of brain fibroblast growth factor by heparin-Sepbarose affinity chromatography: identity with pituitary fibroblast growth factor. Proc. Natl. Acad. Sci. USA 81,6963-6967 (1984). Ivarie, R.D. and P.H. O’Farrell: The glucocorticoid domain: steroid-mediated changes in the rate of synthesis of rat hepatoma proteins. Cell 13, 41-55 (1978).

45 Juurlink, B.H.J., A Schousboe, OS. Jorgensen and L. Hertz: Induction by hydrocortisone of glutamine synthetase in mouse primary astrocyte cultures. J. Neurochem. 36, 136-142 (1981). Labourdette, G. and P. Mandel: SlOO protein in monolayer cultures of glial cells: basal level in primary and secondary cultures. B&hem. Biophys. Res. Commun. 85, 1307-1313 (1978). Laemmli, U.K.: Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685 (1970). Levenson, R., K. Iwata, M. Klagsbum and D.A. Young: Growth factor- and dexamethasone-induced proteins in Swiss 3T3 cells. Relationship to DNA synthesis. J. Biol. Chem. 260, 8056-8063 (1985). Lowry, O.H., N.J. Rosebrough, A.L. Farr and R.J. Randall: Protein measurements with the Folin phenol reagent. J. Biol. Chem. 193, 265-275 (1951). Martinez-Hemandez, A., K.P. Bell and M.D. Norenberg: Glutamine synthetase: glial localization in brain. Science 195, 1356-1358 (1977). Moonen, G., E. Heinen and G. Goessens: Comparative ultrastructural study of the effects of serum-free medium and dibutyryl cyclic AMP on newborn rat astroblasts. Cell Tissue Res. 167, 221-227 (1976). Narumi, S., H.K. Kimelberg and R.S. Bourke: Effects of norepinephrine on the morphology and some enzyme activities of primary monolayer cultures from rat brain. J. Neurochem. 31, 1479-1490 (1978). O’Farrell, P.H.: High resolution two-dimensional electrophoresis of proteins. J. Biol. Chem. 250, 4007-4021 (1975). Pettmann, B., G. Labourdette, G. Devilliers and M. Sensenbrenner: Effects of brain extracts from chick embryo on the development of astroblasts in culture. Dev. Neurosci. 4, 37-45 (1981). Pettmann, B., M. Weibel, M. Sensenbrenner and G. Labour-

dette: Purification of two astroglial growth factors from bovine brain. FEBS Lett. 189, 102-108 (1985). Pettmann, B., C. Gensburger, M. Weibel, F. Perraud, M. Sensenbrenner and G. Labourdette: Isolation of two astroglial growth factors from bovine brain, comparison with other growth factors; cellular iocalization. In: Glial-Neuronal Communication in Development and Regeneration, eds. H.H. Althaus and W. Seifert (Springer-Verlag, Berlin) pp. 451-478 (1987). Sensenbrenner, M., G. Devilhers, E. Bock and A. Porte: Biochemical and ultrastructural studies of cultured rat astroglial cells. Effect of brain extract and dibutyryl cyclic AMP on glial fibrillary acidic protein and glial filaments. Differentiation 17, 51-61 (1980). Sensenbrenner, M., B. Pettmann, G. Labourdette and M. Weibel: Properties of a brain growth factor promoting proliferation and maturation of rat astroglial cells in culture. In: Hormones and Cell Regulation, eds. J.E. Dumont, B. Hamprecht and J. Nunez (Elsevier, Amsterdam). pp. 345-360 (1985). Shapiro, D.L.: Morphological and biochemical alterations in foetal rat brain cells cultured in the presence of monobutyryl cyclic AMP. Nature 241, 203-204 (1973). Thomas, G., and H. Luther: Transcriptional and translational control of cytoplasmic proteins after serum stimulation of quiescent Swiss 3T3 cells. Proc. Natl. Acad. Sci. USA 78, 5712-5716 (1981). Weibel, M., B. Pettmann, G. Daune, G. Labourdette and M. Sensenbrenner: Chemically defined medium for rat astroglial cells in primary culture. Int. J. Dev. Neurosci. 2, 355-366 (1984). Weibel, M., B. Pettmann, G. Labourdette, M. Miehe, E. Bock and M. Sensenbrenner: Morphological and biochemical maturation of rat astroglial cells grown in a chemically defined medium: influence of an astroglial growth factor. Int. J. Dev. Neurosci. 3, 617-630 (1985).