Region-specific regulation of preproenkephalin mRNA in cultured astrocytes

Molecular Brain Research, 11 (1991) 65-69 ¢~ 1991 Elsevier Science Publishers B.V. 0169-328X/91/$03.50 ADONIS 0169328X91703187

65

BRESM 70318

Region-specific regulation of preproenkephalin mRNA in cultured astrocytes David K. Batter and John A. Kessler Departments of Neurology and Neuroscience, Albert Einstein College of Medicine, Bronx, NY 10461 (U.S.A.)

(Accepted 12 March 1991) Key words: Glial heterogeneity; Astrocyte culture; Preproenkephalin mRNA; CyclicAMP; Opiate peptide precursor; mRNA regulation

Regulation of preproenkephalin (PPE) mRNA was examinedin astrocytescultured from several regionsof the. neonatal rat brain. Astrocytes from these regions expressed differing levelsof PPE mRNA, with higher levelsin astrocytes from the hypothalamusfollowed by frontal cortex and striatum. Further, PPE mRNA was regulated differently in hypothalamic than in striatal gila. Treatment of striatal astrc4.vtes with the /I-adrenergic agonist, isoproterenol, or with agents which directly increased intracellular cAMP (forskolin or 8-bromo-cAMP) elevated levels of PPE mRNA. By contrast, none of these treatments altered levelsof PPE mRNA in hypothalamicastrocytes despite increasingcAMP levels 60-fold. These observations ind,.'catethat there is st_.~._k.ingregional heterogeneity in the expression and regulation of PPE mRNA by astrocytes, suggesting that proenkephafin or its derived peptides help to mediate region-specificbrain functions. INTRODUCTION Astrocytes are the most numerous cell type in the mammalian central nervous system. Although all astrocytes share some molecular characteristics, such as the expression of unique intermediate filaments containing gliai fibrillary acidic protein (GFAP), it has become increasingly clear that they are heterogeneous in many respects. For example, astrocytes classically have been divided on morphologic grounds into fibrous and protoplasmic subtypes ~4. Recently, they have been subdivided into two populations on the basis of immunoreactivity to the monoclonal antibody A2Bsts,t6. Astrocytes are also heterogeneous with regard to the expression of various phenotypic traits such as receptors and reuptake systemss'23. Furthermore, astroglial cells from different regions of the brain interact differently with cultured neurons s, and they differ with respect to the protein composition of their cellular membranes 1. These observations suggest that such heterogeneity underlies regionspecific functions. Until recently, expression of neuropeptides has been thought to be restricted to neurons. However, observations from this laboratory22 and others n,~s have shown that primary cultures of astrocytes derived from a number of different brain regions express high levels of preproenkephalin (PPE) mRNA. In addition, PPE mRNA is expressed in C6 glioma cells24 (a transformed

glial cell line), and in neuroblastoma-glioma hybrids 17'25. Astrocytes can also express mRNAs encoding other peptides such as angiotensinogen21, somatostatin and cholecystokinin is. The role of these peptides in astrocyte function is not known. A number of laboratories have shown that the PPE gene can be regulated by various mechanisms including depolarization 9.t°, glucocorticoids 13, phorbol esters 3 and cAMP (for review, see ref. 7). Specifically with regard to gliai cells, treatments that increase intracellular cAMP elevate preproenkephalin mRNA in primary cultures of cortical and striatai astrocytes n'm, as well as in C6 glioma cells24 and glioma-neuroblastoma hybrids 25, In this report we show that isoproterenol (a /~adrenergic agonist), forskolin (a direct stimulator of adenylate cyclase), and 8-bromo-cAMP (a membrane permeant cAMP analog), three agents which can increase intraceilular cAMP, all increase PPE mRNA in primary cultures of striatal astrocytes. In dramatic contrast, PPE mRNA levels in cultures of hypothalamic astrocytes do not change in response to any of these agents, suggesting a region-specific difference in astrocyte function. MATERIALS AND METHODS Cell culture

Gfial cultures were prepared from neonatal rat striatum, frontal cortex, or hypothalamus as previously described" with minor modifications. Briefly, meninges were removed and brain regions

Correspondence: D.K. Batter, Department of Neurology, Albert Einstein Collegeof Medicine, 1300Morris P~k Avenue, Bronx, NY 10461,

U.S.A.

66 were rapidly dissected in ice-cold calcium/magnesium-free Puck's saline G. Tissue was minced and incubated in puck's saline G containing 0.1% trypsin for 30 min at 37 °C, switched to culture media and dissociated by trituration through a Pasteur pipet. Viable cells, as assessed by Trypan blue exclusion, were counted in a hemocytometer and plated at a density of 500 cells/mm2 in culture dishes coated with poly-D-lysine containing EMEM/F12/FBS (45%/ 45%/10%). Cultures were maintained at 37 °C in an atmosphere of 5% CO2-95% air at nearly 100% relative humidity. Cells were fed after 3 days with ice-cold media to minimize neuronal contamination, and every 3 days thereafter until confluence was achieved ( - 7 - 1 0 days). The cultures were characterized by morphology and immunocytochemistry. Cultures grown to confluence contained 90-95% GFAPimmunoreactive cells with no detectable neurons (see rc~. 22 for details).

Extraction of total cellular RNA Total RNA was isolated by the method of Chomczyinski and Sacchi2 with minor modifications. In short, cultures were rinsed with Puck's saline G, then lysed in solution D (4 M guanidinium isothiocyanate, 25% sarcosyi, 25 mM sodium citrate, 0.1 M 2-mercaptoethanol). Following addition of 2 M sodium acetate (0.1 vol), H20-equilibrated phenol (1 vol), and CHCI 3 (0.4 vol), extracts were incubated on ice for 15 min and the phases separated by centrifugation. RNA in the aqueous phase (free of contaminating

DNA) was precipitated 2 times, washed with 70% ethanol, dried, and resuspended in H20. RNA concentration was determined spectrophotometrically.

Northern blot analysis Denatured RNA samples were electrophoresed in 1.2% agarose gels containing 0.66 M formaldehyde. RNAs were transferred to Gene Screen Plus (NEN) by overnight capillary blotting in 10× SSC (Ix SSC -- 0.15 M sodium chloride, 0.015 M sodium citrate) and baked in vacuo at 80 °C for 2 h. 32p-Labelled RNA probes synthesized using SP6 RNA polymerase were used to detect preproenkephalin mRNA (1.5 kb) or cyclophilin mRNA (0.9 kb). The rat preproenkephalin eDNA (pENK, a.k.a, pRPE2) was kindly provided by Dr. Steven Saboi26 and the rat cyciophilin eDNA (1B15) was generously supplied by Dr. James Douglass4. Northern blots were prehybridized for 30 min at 65 °C in 50% formamide, 0.6 M NaCI, 1% SDS, and 0.1 mg/ml denatured herring sperm DNA. Blots were hybridized at 65 °C overnight in prehybridization solution to which 1 × 106 cpm/ml of one of the riboprobes had been added. Blots were washed to a final stringency of 0.2× SSC at 65 °C and exposed to Kodak XAR-5 film at -70 °C. The resulting hybridization signals were volume integrated using a Molecular Dynamics scanning densitometer. All blot~ were integrated within the linear range of the film (0.2 to 2 0 D . ) and the resulting optical density measurements were used in all subsequent calculations.

cAMP assay Levels of total cellular cyclic AMP were determined using a commercially available cAMP assay kit (Amersham). Total protein was determined by the Bradford micro-assay according to manufacturers instructions (Bio-Rad). RESULTS

ppE

p

q m

N

lels,.- U m

Primary cultures of neonatal rat astrocytes were established and characterized as previously described22, and preproenkephalin mRNA from confluent dishes was examined by Northern blot analysis (Fig. 1). Astrocytes from all brain regions tested expressed the 1450 nudeotide PPE mRNA. However, on a per microgram RNA basis, astrocytes derived from different brain regions varied in their basal levels of preproenkephalin mRNA. Hypothalamic astrocytes consistently expressed the highest level of PPE mRNA, followed by frontal cortex and striatum (Table I). Hybridization to the constitutively expressed cyclophilin mRNA (1B15) indicated that equal amounts of total RNA were loaded per lane. TABLE I

Regional differences in astrocyticpreproenkephalin mRNA levels

COR-

HYP-

STR

Fig. 1. Preproenkephalin mRNA levels in astrocytes cultured from three brain regions. Primary glial cultures were prepared from neonatal rat brain and grown to confluence. Cells were harvested and 5 ~g o[ total RNA from each dish was used in northern analysis. Blots were probed with a 32p-labelled riboprobe for preproenkephalin (PPE) or reprobed for cyciophilin (1B15) and autoradiographed. COR, frontal cortex; HYP, hypothalamus; STR, striatum.

Total RNA from confluent cultures of rat neonatal striatal or hypothalamic astrocytes was subjected to Northern blot analysis as in Fig. 1. The data are expressed as percent of striatal nigral 4- S.E.M. (n)

Region

PPE mRNA/1 B15 mRNA

Striatum Cortex Hypothalamus

100(4) 137 -+ 18 (4) 266 4. 53* (3)

*P < 0.05 by ANOVA as compared to striatum.

67 The preproenkephalin gene contains a cAMP-responsive element 3"7 and levels of PPE m R N A in cultured cortical and striatal astrocytes are stimulated by increased cAMP sl'ss. To determine if there is regional heterogeneity in this regulation, astrocytes from different brain regions were treated with agents that elevate intracellular cAMP, and PPE m R N A levels were measured. Astrocytes derived from the striatum showed approximately 3-fold increases in PPE m R N A following t r ~ n e n t with isoproterenol, forskolin, or 8-bromo-cAMP (Fig. 2). In contrast, PPE m R N A from hypothalamic astrocytes showed no significant increase in response to any of these treatments. Levels of cyclophilin m R N A were not changed by any of these treatments. To characterize these differences in more detail, dose response curves for treatment with isoproterenol or forskolin were examined (Fig. 3). Doses of isoproterenol as low as 10- ? M significantly elevated striatal PPE m R N A more than 4-fold. By contrast, hypothalamic astrocyte PPE m R N A was not altered by any dose of isoproterenol from 10-9 to 10- 4 M. Similarly, forskolin treatment failed to alter hypothalamic astrocyte PPE m R N A but treatment with 10-6 to 10- s M elevated striatal PPE m R N A 3-fold. Interestingly, higher doses of

CI

FB

forskolin were not effective in stimulating striatal PPE mRNA. In toto, these data clearly show that, while striatal astrocytes respond robustly over a wide concentration range, hypothalamic PPE m R N A levels remain virtually unchanged over all dosages tested. One possible explanation for these findings was a

400 350

$

t

A

300 Z

250 tO 0 m Z 0 z ee v

200 150

M O.

100 50 0 0.0001

I

Ir

|

I

I

0.001

0.01

0.1

1

10

-" I 100

iSOPROTERENOL (uM)

C I FB

400

B 350

,i

Z

PPE

•

m

lit

300 250

m~

200 ....

zu

150 eL_

100 ~ ~iii:i ~ i

:

50-

1815)m--

0

o.01 i

.

I

I

I

I

1

lO

lOO

10oo

FORSKOHN (uM)

l

• ~:! ¸iii¸~i ~i:ii

STR

I

0.1

HYP

Fig. 2. Regional heterogeneity of preproenkephalin mRNA regulation in response to cAMP-stimulating agents. Confluent astrocyte cultures prepared from neonatal rat striatum (STR) or hypothalamus (HYIP) were treated for 4 h with (1) media only (C), (2) 0.! pM isoproterenol (I), (3) 5 juM forskolin (F), or (4) 8-bromo-cAMP (B). Northern analysis was performed on 5/~g of total RNA as in Fig. L PPE, preproenkephalin mRNA; IBI5, cyclog;nilin mRNA.

Fig. 3. Dose response curves of preproenkephalin mRNA expression in striatal and hypothalamic astrocytes following treatment with cAMP-stimulating agents. Confluent cultures of astrocytes prepared from neonatal striatum (0) or hypothalamns ( a ) were treated for 4 h with the indicated doses of (A) isoproterenol or (B) forskolin. Cells were harvested and 5/tg of total RNA was subjected to northern analysis as in Fig. 1. Preproenkephalin mRNA hybridization signals were quanfitated with a Mo!e~lar Dynarmc~ scanning densitometer and normalized to those of the constitutively expressed cydophilin mRNA. The control treatments were: media only for the isoproterenol experiments, and 0.1% EtOH in media for the forskolin experiments. Each value is the result of three dishes + S.D. (n = 3). *P < 0.05 by ANOVA.

68

TABLE II cAMP levels in striatum t~nd hypothalamus after treatment with isoproterenol or forskolin

Confluent culturesof rat neonatal striatal or hy~othalamicastrocytes were treated with 1 pM isoproterenol or 5/~M forskolin for 30 rain and assayed for cAMP levels. Sister dishes were assayed for total protein. The data are expressed as mean _+S.D. (n = 3). Treatment

Control Isoproterenol Forskolin

pmol cAMP/mg protein Striatum

Hypothalamus

90.91 + 23.23 6753 + 595.1" 2121 + 74.98*

53.52_+13.50 3042 + 298.4* 1544+ 115.4"

*P < 0.05 as comparedto the respectivecontrolsby ANOVA. regional difference in intracellular cAMP levels following drug treatment. Therefore, cAMP levels were directly measured in striatal and hypothalamic astrocyte cultures 30 min after treatment with either isoproterenol or forskolin. As shown in Table II, cAMP levels were increased in striatal cultures with the two treatments approximately 75- and 23-fold, respectively. In hypothalamic cultures isoproterenol increased cAMP levels 56fold and forskolin 29-fold, indicating that these astrocytes were also capable of elevating their cAMP levels following drug treatment. DISCUSSION These studies indicate that astrocytes from two brain regions differ with regard to mechanisms regulating preproenkephalin mRNA. Specifically, stimulation of intracellular cAMP elevated levels of PPE mRNA in striatal but not in hypothalamic astrocytes. The gene encoding preproenkephalin in a number of species including rat is well characterized and has been shown to include a cAMP-responsive element. Moreover, stimulation of cAMP in a variety of cell types such as adrenomedullary cells, C6 glioma cells, as well as striatal and cortical astrocytes elevates levels of PPE mRNA. Therefore our finding that PPE mRNA levels in hypothalamic astrocytes did not change following stimulation of cAMP is an interesting exception. Since hypothalamic astrocytes contain fewer/~-adrenergic receptors than astrocytes from other brain regions6, is was plausible to hypothesize that the lack of response to isoproterenol reflected the lower level of receptors. However, direct stimulation of intracellular cAMP with forskolin or 8-bromo-cAMP, thus bypassing the receptor, failed to elevate PPE mRNA in hypothalamic astrocytesl An alternative hypothesis was that cAMP in confluent hypothalamic astrocytes was already at maximal levels, precluding a response to any of the treatments. However,

direct measurements of intracellular cAMP levels after these treatments demonstrated as much as a 60-fold increase in response to isoproterenol. Finally, i t was possible that PPE mRNA in hypothalamic astrocytes was already maximally stimulated in the basal state. This seems unlikely, however, since basal le,vels of the mRNA in hypothalamic astrocytes were lower than the stimulated levels in striatal astrocytes. The lack of stimulation of PPE mRNA in hypothalamic astrocytes therefore must reflect a basic difference in the cascade of intracellular events which transpires after elevation of levels of cAMP. For example, there might be inhibition of the trans-acting protein CREB TM which activates the cAMP responsive element. Alternatively, tome other trans-acting factor might inhibit the cAMP response in these cells. It is also possible that longer treatment with cAMP-stimulating agents may be necessary to produce an increase of PPE mRNA in hypothalamic astrocytes. For example, Stachowiak et al. 2° have shown that preproenkephalin mRNA levels in cultured bovine chromaffin ceils do not reach a maximum until 12 h after treatment under similar conditions. Regardless, further investigation of the anomalous regulation of PPE mRNA in cultured hypothalamic astrocytes shoui~ provide new insights into mechanisms regulating preproenkephalin gene expression. The region-specific difference we have demonstrated, as well as those shown by other investigators (for reviews, see refs. 8, 23), support the idea that there are unique astrocyte populations in different areas of the brain. However, it is possible that these regional differences may reflect c~llular heterogeneity within astrocyte populations in each brain region. For example, Raft and his colleagues have been able to separate astrocytes from the optic nerve into two populations based on morphology, antigenic expression, and response to growth factors (for review, see ref. 16). As knowledge of astrocyte neurobiology increases, it seems likely that other subclassifications will be described. Thus it is conceivable that the difference in preproenkephalin mRNA regulation that we observe results from varied ratios of astrocyte subpopulations within a particular brain region. Regardless of the cellular or molecular mechanisms responsible, the regional difference in astrocytic preproenkephalin mRNA regulation may have important functional implications. If the heterogeneity that we observe in vitro reflects the situation in vivo, PPE mRNA and its product(s) might play a role in region-specific specialization in the central nervous system. In support of this hypotheses, Spruce et al. t9 have recently shown that astrocytes in vivo do display heterogeneity with respect to proenkephrW~n and PPE mRNA expression both within ~nd between brain regions. Further, cultured astrocytes

69 do not process proenkephalin intracellularly to metenkephalin but rather secrete large quantities of unprocessed proenkephalin which is cleaved extracellularly (manuscript in preparation). A fuller understanding of the expression and regulation of preproenkephalin m R N A awaits the elucidation of the role of astrocytic

proenkephalin (and peptide products) in brain function.

REFERENCES

14 Privat, A. and Rataboul, P., Fibrous and protoplasmic astrocytes. In S. Federoff and A. Vernadakis (Eds.), Astrocyte$: Biochemistry, Physiology and Pharmacology of Astrocytes, Vol. 2, Academic Press, FL, 1986, pp. 105-130. 15 Raff, M.C., Abney, E.R., Cohen, J., Lindsey, R. and Noble, M., Two types of astrocytes in cultures of developing rat white matter: differences in morphology, surface gangliosides and growth characteristics, J. Neurosci., 3 (1983) 1289-1300. 16 Raff, M.C., Glial cell diversification in the rat optic nerve, Science, 243 (1989) 1450-1455. 17 Schwartz, J.P., Chronic exposure to opiate agonists increases proenkephalin biosynthesis in NG108 cells, Mol. Brain Res., 3 (1988) 141-146. 18 Shinoda, H., Marini, A., Cosi, C. and Schwartz, J., Brain region and gene specificity of neuropeptide gene expression, Science, 245 (1989) 415-417. 19 Spruce, B.A., Curtis, R., Wilkin, G.P. and Glover, D.M., A neuropeptide precursor in cerebellum: proenkephaiin exists in subpopulations of both neurons and astrocytes, EMBO J., 9 (1990) 1787-1795. 20 Stachowiak, M.K., Hone, J.S. and Viveros, O.H., Coordinate and differential regulation of phenylethanolamine N-methyltransferase, tyrosine hydroxylase and proenkephalin mRNA's by neural and hormonal mechanisms in cultured bovine adrenal medullary cells, Brain Res., 510 (1990) 277-288. 21 Stornetta, R.L., Hawelu4ohnson, C.L., Guyenet, P.C. and Lynch, K.R., Astrocytes synthesize angiotensinogen in brain, Science, 242 (1988) 1444-1446. 22 Vilijn, Id.-H., Vaysse, P.J.-J., Zukin, R.S. and Kessler, J.A., Expression of preproenkephalin mRNA by cultured astrocytes and neurons, Proc. Natl. Acad. Sci. U.S.A., 85 (1988) 65516555. 23 Wflkin, G.E, Marriott, D.R. and Cholewinski, A.J., Astrocyte heterogeneity, Trends Neurosci., 13 (1990) 43-46. 24 Yoshikawa, K. and Sabol, S., Expression of the enkephalin precursor gene in C6 rat glioma cells: regulation by fl-adrenergic agonists and glucocorticoids, Mol. Brain Res., 1 (1986) 75-83. 25 Yoshikawa, J. an~'lSabol, S.L., Glucocorticoids and cyclic AMP synergisticallyregu;ate the abundance of proenkephafin messenger RNA in neuroblastoma-glioma hybrid cells, Biochem. Biophys. Res. Commun., 139 (1986) 1-10. 26 Yoshikawa, K., Williams, C. and Saboi, S.L., Rat brain preproenkephalin mi?NA, cDNA cloning, primary structure, and distribution in the central nervous system, 1. Biol. Chem., 259 (1984) 14301-14308.

1 Barbin, G., Katz, D.M., Chamak, B., Glowinski, J. and Prochiantz, A., Brain astrocytes express region-specific glycoproteins in culture, Gila, 1 (1988) 96-103. 2 ChomczTnski, P. and Saeehi, N., Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform ext-'~ction, Anal. Biochem., 162 (1987) 156-159. 3 Comb, M., Bilaberg, N.C., Seasholtz, A., Herbert, E. and Goodman, H., A cyclic AMP- and phorbol ester-inducible DNA element, Nature, 323 (1986) 353-356. 4 Danielson, P.E., Forss-Petters, S., Brow, M.A., Calavetta, L., Douglass, J., Milner, R.J. and Sutcliffe, J.G., plB15: a cDNA clone of ~he rat mRNA encoding eyclophilin, DNA, 7 (1988) 261-267. 5 Denis-Donini, S., Glowinskt, J. and Prochiantz, A., Gfial heterogeneity may define the three-dimensional shape of mouse mesencephalic dopaminergic neurones, Nature, 307 (1984) 641643. 6 Ernsberger, P., lacovitti, L. and Reis, D.J., Astrocytes cultured from specific brain regions differ in their expression of adrenergic binding site~;, Brain Res., 517 (1990) 202-208. 7 Goodman, R.H., Regulation of neuropeptide gene expression, Annu. Rev. Neurosci., 13 (1990) 111-117. 8 Hansson E., Astrocytes from defined brain regions as studied with primary cultures, Prog. Neurobiol., 30 (1988) 369-397. 9 Kilpatrick, D.L., Howells, R.D., Fleminger, G. and Udenfriend, S., Denervation of rat adrenal glands markedly increases proenkephalin mRNA, Proc. Natl. Acad. Sci. U.S.A., 81 (1984) 7221-7223. 10 LaGamma, E.F., White, J.D., Adler, J.E., Krause, J.E., McKelvy, J.F. and Black, I.B., Depolarization regulates adrenal preproenkephalin mRNA, Proc. Natl. Acad. Sci. U.S.A., 82 (1985) 8252-8255. 11 Melner, M., Low, K.G., Allen, R.G., Nielsen, C.P., Young, S.L. and Saneto, R.E, The regulation of proenkephalin expression in a distinct population of glial cells, EMBO J., 9 (1990) 791-796. 12 Montminy, M.R. and Bilezikjian, LM., Binding of nuclear proteins to the cyclic-AMP responsive element of the somatostatin gene, Nature, 328 (1987) 175-178. 13 Naranjo, J.R., Mocchetti, I., Schwartz, J.P. and Costa, E., Permissive effects of dexamethasone on the increase of proenkephalin mRNA induced by depolarization of chromaffin cells, Proc. Natl. Acad. Sci. U.S.A., 83 (1986) 1513-1517.

Acknowledgements. We thank Dr. Steven Sabol for providing the rat preproenkephalin cDNA, and Dr. James Douglass for providing the rat cyclophilin (1B15) cDNA. This work was supported by NIH Grants NS20013 and NS20778.