Cells from embryonic rat striatum cocultured with mesencephalic glia express dopaminergic phenotypes

Neuroscience Letters, 128 (1991) 1-3

© 1991 ElsevierScientificPublishers Ireland Ltd. 0304-3940/91/$03.50 ADONIS 030439409100310S NSL 07827

Cells from embryonic rat striatum cocultured with mesencephalic glia express dopaminergic phenotypes C o r d i a n B e y e r 1, C h r i s t o p h Pilgrim 1, Ingrid Reisert I a n d U l r i c h Misgeld 2 1Ab teilung Anatomie und Zellbiologie, Universitiit Ulm, Ulm ( F.R.G. ) and 2Physiologisches Institut der Universitiit Heidelberg, Heidelberg ( F.R.G. )

(Received30 January 1991;Revisedversionreceived25 March 1991;Accepted27 March 1991) Key words: Coculture;Dopamine; Immunocytochemistry;Neostriatum; Substantianigra; Tyrosinehydroxylase;Rat

To study region-specifictransmitter phenotype expression, cells of embryonicday 14 (El4) rat neostriatum (ganglionic eminenceplus cortical plate) or of the substantia nigra (ventral mesencephalon)were cultured on glial cells either from substantia nigra or neostriatum (E21). By antityrosine hydroxylaseimmunocytochemistry,immunoblottingof tyrosine hydroxylaseprotein and quantitation of dopamine and its metabolites by HPLC, dopaminergiccells were revealed in nigral and neostriatal cultures plated on nigral glial cells. No dopaminergiccells were found among neostriatal neuronsplated on neostriatal glialcells. It is concludedthat glia from substantia nigra but not glia from neostriatumis capable of inducing developmentor promoting survivalof dopaminergiccells.

Interactions between nerve- and glial cells in the developing central nervous system are well documented. To ensure proper development of the brain, these two cell types must interact in specific ways [8]. For example, glial cells guide the migration of immature neurons [13] and secrete molecules needed for neuronal growth and survival [10]. Recently interest has focused on the regional heterogeneity of glial properties and on region-specific interactions between glial cells and distinct nerve cell populations in the developing brain [5]. However, up to the present, little information is available regarding the influence of glial cells on the development of identified neuronal phenotypes [9]. The goal of the present study was to compare specific morphological and biochemical correlates of neurons developing either on homo- or heterotypical glial cells. For this purpose, we cultured rat neostriatum cells (embryonic day 14 = El4) on glia (E21) derived from the neostriatum (S/Sg coculture) or on glia from the substantia nigra (S/Ng coculture). For comparison, cells from substantia nigra (El4) were grown on glial cells from substantia nigra E21 (N/Ng coculture). In the mature brain, the neostriatum and the substantia nigra differ in various respects, one being the presence in substantia nigra or absence in neostriatum of dopaminergic neurons. Culturing techniques were described previously [4, 12]. Correspondence. C. Beyer, Abteilung Anatomic und Zellbiologie, Universit/it Ulm, 7900Ulm, F.R.G. Fax: (49) (731) 176-2038

Primary cultures of neostriatal or nigral glial cells were obtained by plating neostriatal or nigral tissues of E21 rat fetuses on poly-o-lysine-coated plastic dishes. After 2-4 weeks, cultures of nigral tissue were entirely composed.of glial cells and no more phase-bright cells indicative of the presence of neurons. In neostriatal cultures, only a few such cells were evident. These glial cultures were then used as a substrate for plating dissociated cells of the forebrain or of the ventral mesencephalon from E l 4 fetuses. The forebrain tissue was chosen in such a way that it contained ganglionic eminence plus cortical plate and is referred to as neostriatal culture (S). For the same reason, cultures from ventral mesencephalic tissue are called nigral cultures (N) in this paper. The cells developed an extensive neuronal network within 3 weeks and remained viable for up to 4 months. Anti-tyrosine hydroxylase (TH) immunocytochemistry was performed using a monoclonal antibody (l:1000, Boehringer, Mannheim) and ABC complex (Camon, Wiesbaden). Immunoreaction was visualized by DAB in the presence of 0.01% H202 [for details see ref. 3]. Numbers of immunolabeled perikarya were counted b y screening entire cultures. In addition, some cultures of each experimental set containing only the glial monolayer from neostriatum or substantia nigra were immunostained for TH. No TH-immunoreactive (TH-IR) cells were found in these cultures. For Western blots [2], cultures were pooled in groups of two and the proteins were separated into an insoluble and a soluble protein fraction by lysis in a buffer con,

TABLE I ABSOLUTE NUMBERS OF TYROSINEHYDROXYLASEIMMUNOREACTIVE CELLS IN DISSOCIATED CELL CULTURES FROM NEOSTRIATUM (S) AND SUBSTANTIANIGRA (N) COCULTURED WITH GLIAL CELLS FROM NEOSTRIATUM (Sg) OR SUBSTANTIANIGRA (big) Numbers represent means+S.E.M, from 3 independent experiments (eachconsisting of 2-4 cultures), n.d., not detectable. Time in culture

N/Ng

S/Ng

S/Sg

1 week 2 weeks 4 weeks

482 + 78 437__+71 417 +45

218 _ 49 207__+35 183+ 34

n.d. n.d. n.d.

taining 10 m M NaH2PO4, 110 m M NaC1, 50 m M E D T A and 0.5% (v/v) Triton X-100 (pH 6.5). After centrifugation (15 min, 12,000 gay, 4°C), the supernatant (soluble) and pellet (insoluble) were dissolved in 62.5 m M TrisHCI, 2 m M EDTA, 2% (w/v) SDS and 5% mercaptoethanol (pH 6.8), heated to 100°C for 5 min and stored at - 2 0 ° C . SDS-PAGE was carried out under reducing conditions on 10% polyacrylamide gels. After electrophoresis, proteins were transferred to nitrocellulose and incubated overnight with anti-TH (1:1500 in TBSTween, pH 7.6, 4°C). Immunoreaction was visualized as described for immunocytochemistry. TH-protein was identified by comparison with an adrenal medulla standard (1 /~g/ml) treated in the same way. Blanks were obtained by replacing T H antiserum with normal swine serum. Intracellular dopamine (DA), norepinephrine (NE), epinephrine (E), dihydroxyphenylalanine (DOPA) and homovanillic acid (HVA) were assayed using perchloric acid (50 mM) extractions of cell cultures and reversed phase HPLC (Cl8 column) with electrochemical detection [3]. Extraction and analysis was done in the presence of dihydroxybenzylamine (2 pmol) as an internal standard. Detection limit was 1 pg cate~holamine per injection (20 gl). Two cultures each were pooled to obtain sufficient material for HPLC analysis. Medium supplemented with horse serum which had been used for feeding the cultures was also analyzed for catecholamines. As expected, T H - I R cells were found in N/Ng cultures, but surprisingly also in S/Ng cultures, while they were absent in S/Sg cultures (Table I). Numbers of THIR cells in relation to all cells (neurons plus glia) quantified after Methylene blue staining were about 5% in N/ Ng- and 2.5% in S/Ng cultures. The number of T H - I R cells remained constant through the various culturing times tested (Table I). However, the morphological appearance of T H - I R cells showed distinct differences between S/Ng- and N/Ng cultures (Fig. 1). Whereas N/

Fig. 1. Tyrosine hydroxylase-immunoreactivecells in dissociated cell cultures from neostriatumcoculturedwith substantia nigra glia (a) and substantia nigra cocultured with substantia nigra glia (b). Magnification x 400 (a), x 440 (b).

Ng cultures contained T H - I R cells with large perikarya and several long varicose processes (Fig. lb), most THIR cells in S/Ng cultures had small round soma and no visible processes (data not shown). A fraction of the latter cells exhibited nuclear in addition to cytoplasmatic staining. Only a few scattered T H - I R cells issued long thin T H - I R processes (Fig. la). Because immunohistochemistry to T H may stain other cellular components such as histone proteins [11], the immunoreactivity was further characterized by Western blots (Fig. 2). A heavily stained protein band with a Mr of about 60 kDa with an electrical mobility identical to the TH-subunit observed in a preparation from adrenal medulla was found in N/Ng cultures. All other bands seen in Fig. 2 were also present when the primary antibody was omitted and are therefore unspecific. TH-IR bands were not found in blots from S/Ng- and S/Sg cultures, which may be due to the low numbers of IR-cells and an insufficient sensitivity of this technique. Therefore, cultures were further analyzed for the presence of DA and its metabolites by H P L C (Table II). The culturing medium contained no DA but N E (13 ng/ml) and E (1 ng/ml). In N/Ng cultures, we measured similar intracellular amounts of DA, NE and E, but low levels

S/Ng

N/Ng a

b

a

b

S/Sg a

AM

b -- 8 4 kD 5 8 kD 4 8 kD

Fig. 2. Immunoblot of TH in the insoluble (a) and soluble (b) fraction prepared from 3-week-old cell cultures from neostriatum and substantia nigra eocultured with glia from neostriatum or substantia nigra. Right-hand blot shows immunoblot of a preparation from adrenal medulla (AM). Note the absence of TH-immunoreactivity in blots of S/Ng and S/Sg.

of HVA. Likewise in S/Ng cultures, we found endogenous DA, NE and E but no metabolites. The catecholamines extracted from S/Sg cultures were NE and E but not DA. The presence of catecholamines (NE, E) in S/Sg cultures could be the result of an uptake of these neurotransmitters from cultivation medium (see above) by astroglial cells [7]. Our data demonstrate the presence of dopaminergic neurons in cultures of a brain region in which this transmitter phenotype is normally not expressed in the adult even though TH-IR cells have occasionally been observed in this area [15]. The requirement was that striatal neurons were cultured on glial cells from the ventral mesencephalon and not from the forebrain. It can presently not be excluded that the nigral glial cultures contain some neuronal precursors which are induced to express TH upon the addition of striatal cells. However, it is more likely that the glial cells of the nigral feeder layer are responsible for the induction of dopaminergic properties in some striatal neurons or neuroblasts. This TABLE II INTRA.CELLULAR CONCENTRATIONS OF CATECHOLAMINES IN DISSOCIATED CELL CULTURES FROM NEOSTRIATUM (S) AND SUBSTANTIA NIGRA (N) COCULTURED WITH GLIA FROM NEOSTRIATUM (Sg) OR SUBSTANTIA NIGRA (Ng) AFTER 3 WEEKS IN VITRO Concentrations are expressed as means_+ S.E.M. (ng/well) from two independent experiments (each consisting of 3-5 cultures), n.d., not detectable. Catecholamine

N/Ng

S/Ng

S/Sg

medium

DA NE E HVA DOPAC

0.54 + 0.05 0.74+0.12 0.44_+0.04 0.09 _+0.02 n.d.

0.29 ___0.03 0.33-t-0.05 0.47_+0.05 n.d n.d.

n.d 0.66_+0,07 0.61_+0.08 n.d n.d.

n.d 13.2_+1.4 0.94_+0.95 n.d n.d.

would conform to the idea that nerve-glia interactions direct the differentiation of dopaminergic cells [5, 9] and that glia contributes to the region-specific development of the mammalian brain [1, 6]. Alternatively, nigral glia could promote the survival of dopaminergic cells which may otherwise occur only transiently during development [14]. The authors thank Dr. M. Bijak for technical support and A. Lewen for excellent technical assistance. The study was supported by the Deutsche Forschungsgemeinschaft (SFB 317, B 113 and RE 413/2-2/3). 1 Autillo-Touati, A., Chamak, B., Araud, D., Vuillet, J., Seite, R. and Prochiantz, A., Region-specific neuro-astroghal interactions: ultrastructural study of the in vitro expression of neuronal polarity, J. Neurosci. Res., 19 (1988) 326-342. 2 Beyer, C., Epp, B., Fassberg, J., Reisert, I. and Pilgrim, C., Regionand sex-related differences in maturation of astroeytes in dissociated cell cultures of embryonic rat brain, Glia, 3 (1990) 55-64. 3 Beyer, C., Pilgrim, C. and Reisert, I., Dopamine content and metabolism in mesencephalic and diencephalic cell cultures: sex difference and effects of sex steroids, J. Neurosci., 11 (1991) 1325-1333. 4 Bijak, M., Jarolimek, W. and Misgeld, U., Effects of antagonists on quisqualate and nicotinic receptor-mediated currents of midbrain neurones in culture, Br. J. Pharmacol., 102 (1991) 699-705. 5 Denis-Donini, S., Glowinski, J. and Prochiantz, A., Glial heterogeneity may define the three-dimensional shape of mouse meseneephalic dopaminergic neurons, Nature, 307 (1984) 641--643. 6 Denis-Donini, S. and Estenoz, M., Interneurons versus efferent neurons: heterogeneity in their neurite outgrowth response to glia from several brain regions, Dev. Biol., 130 (1988) 237-249. 7 Kimelberg, H.K., Catecholamine and serotonin uptake in astrocytes. In S. Fedoroff and A. Vernadakis (Eds.), Astrocytes Vol. 2, Academic Press, New York, 1988, pp. 107-131. 8 Lauder, J.M. and McCarthy, K.D., Neuronal-glial interactions. In S. Fedoroff and A. Vernadakis (Eds.), Astrocytes Vol. 1, Academic Press, New York, 1986, pp. 295-314. 9 Lieth, E., Towle, A.C. and Lauder, J., Neuronal-glial interactions: quantitation of astrocytic influence on development of catecholaminergic neurons, Neurochem. Res., 14 (1989) 979-985. 10 Manthorpe, M., Rudge, J.S. and Varon, S., Astroghal cell contributions to neuronal survival and neurite growth. In S. Fedoroff and A. Vernadakis (Eds.), Astrocytes Vol. 1, Academic Press, New York, 1986, pp. 314-376. 11 Mayerhofer, A. and Russel, L.D., Characterization of a transiently expressed nuclear protein localized in elongating spermatids, Am. Soc. Cell Biol. (Suppl.), San Diego, 1990. 12 Misgeld, U. and Dietzel, I., Synaptic potentials in the rat neostriatum in dissociated embryonic cell culture, Brain Res. 492 (1989) 149-157. 13 Rakic, P., Guidance of neurons migrating to the fetal monkey neocortex, Brain Res., 33 (1971) 471-476. 14 Satoh, J. and Suzuki, K., Tyrosine hydroxylase-immunoreactive neurons in the mouse cerebral cortex during postnatal period, Dev. Brain Res., 53 (1990) 1-5. 15 Tashiro, Y., Sugimoto, T., Hattori, T., Uemura, Y., Nagatsu, I., Kikuchi, H. and Mizuno, N., Tyrosine hydroxylase-like immunoreactive neurons in the striatum of the rat, Neurosci. Lett., 97 (1989) 6-10.