Brain Research, 596 (1992) 202-208

202

© 1992 Elsevi.~r Science Publishers B.V. All rights reserved 0006-8993/92/$05.00

BRES 18274

Cultured astrocytes express mRNA for peptidylglycine-a-amidating monooxygenase, a neuropeptide processing enzyme Robyn S. Klein and Lloyd D. Fricker Departments of Molecular Pharmacology and Neuroscience, Albert Einstein £ollege of Medicine, Bronx, IVY 10461 (USA) (Accepted 23 June 1992)

Key words: Neuroglia; Prohormone processing; Neuropeptide biosynthesis

Cultured astrocytes have been previously found to express several neuropeptides, as well as the neuropeptide processing enzyme carboxypeptidase E (CPE). To investigate whether cultured astrocytes contain additional peptide-processing enzymes, Northern blots were screened for peptidylglycine-a-amidating monooxygenase (PAM) mRNA. PAM is involved with the formation of amide groups on the C-terminus of numerous peptide hormones and neurotransmitters. Primary cultures of astrocytes contain moderate levels of PAM mRNA, as determined by Northern blot analysis. The level of PAM mRNA in cultured hypothalamic astrocytes is similar to the level expressed in cultured hypothalamic neurons. The relative abundance of PAM mRNA differs up to 6-fold between astrocytes cultured from various brain regions. Astrocytes cultured from hypothalamus have high levels of PAM mRNA, those cultured from striatum, frontal cortex, and hippocampus have moderate levels, and those cultured from cerebellum have low levels. To investigate whether all cultured astrocytes express PAM mRNA, in situ hybridization analysis of cultured astrocytes was performed. Interestingly, virtually all of the astrocytes cultured from either hypothalamus or cerebellum express PAM mRNA, in contrast to a previous finding that only 20-40% of similarly cultured astrocytes express CPE. The presence of PAM mRNA in cultured astrocytes suggests that these cells have the capacity to produce amidated neuropeptides.

INTRODUCTION Most secreted neuropeptides are synthesized as inactive precursor molecules that gain bioactivity through post-translational processing 3l. This processing typically includes proteolysis by trypsin-like endopeptidases, followed by removal of basic C-terminal residues by CPE s,~4,2~. The fate of the cleavage products depends on their content: peptides containing an exposed glycine at their C-termini are converted into the amide by PAM (EC 1.14.17.3) 5.9. This step is important for many peptide hormones and neurotransmitters; approximately half of all peptides with a known biological function contain a C-terminal amide 9. PAM was originally purified from bovine neurointermediate pituitary, which is rich in a-amidated peptides 5'~. Until recently, it was believed that a single enzyme was responsible for the amidation of peptides with a C-terminal glycine. It is now understood that PAM is a precursor molecule containing two separable

enzymatic activities involved in amidation s,25,a4. PAM undergoes posttra~slational processing to form a peptidylglycine a-hydroxylating monooxygenase (PHM) and a peptidyl-a-hydroxylglycine a-amidating lyase (PAL) 25. PHM, which is contained in the N-terminus of PAM, is responsible for the hydroxylation of the CH2 group of C-terminal glycine in a reaction involving copper, ascorbate, and molecular oxygen25'38. The conversion of the a-hydroxy-glycine intermediate into an amidated peptide is accomplished by PAL at physiological pH 12A9,2s. Various forms of PHM and PAL proteins arise through post-translational processing of the PAM precursor, as well as alternative splicing of PAM mRNA 13'32. Amidating activity is present in secretory granules of a variety of tissues, including the pituitary, hypothalamus, parotid gland, heart, neonatal pancreas, and cultured corticotrophic tumor cells 9'1°'35. In the adult rat, the level of amidating activity is highest in the heart, anterior and intermediate pituitary, and submaxillary

Correspondence: L.D. Pricker, Department of Molecular Pharmacology, F205, Albert Einsteia College of Medicine, 1300 Morris Park Ave., Bronx, NY 10461, USA. Fax: (1) (718) 829-8705.

203 gIand4. Within rat brain, the level of amidating activity is highest in the hypothalamus, moderate in many other brain regions, and lowest in the cerebellum4. The distribution of PAM mRNA generally parallels the distribution of amidating activity4. Astrocytes are involved in a variety of intercellular interactions in the central nervous system. They adjust the extracellular environment through neurotransmitter and ion uptake systems, and by mediating inflammatory responses 1*3*16. Astrocytes enable proper metabolic and structural organization of the developing brain through the secretion of trophic factors”? In recent studies, astrocytes have been shown to express neuropeptides and neuropeptide processing enzymes; proenkephalin (PE)17~2g*30~36, endothelin23, and angiotensinogen33 have been detected in astrocytes in culture and/or in viva. Somatostatin2’ and nerve growth factor (NGF)” are produced by cultured astrocytes. In addition, cultured astrocytes produce CPE3’ and an unidentified metalloendopeptidase22. In situ hybridization and immunocytochemical analyses have detected PAM mRNA and PAM immunoreactivity throughout the rat brain, in both neuronal and non-neuronal cell types, including ependyma, oligodendroglia, Bergmann astrocytes, and subpial and subependymal fibers26p28. Astrocytes derived from different brain regions do not appear to be identical. Evidence of regional heterogeneity of astrocytes includes differential expression of surface glycoproteins, neurotransmitters and cellular responses to various neurohormones”? Astrocytes also demonstrate regional heterogeneity in their expression of PE, somatostatin, and CPE29q3697. Only a subpopulation of astrocytes demonstrate expression of PE mRNA in culture2” and in vivo3’. In addition, CPE mRNA expression is observed in only a fraction of cultured astrocytes 2o . The percentages of cultured astrocytes expressing CPE mRNA varies from 20-40%. depending on the brain region from which the cells are derived2’. In this study, we report the presence of PAM mRNA in primary cultures of astrocytes from various regions of rat brain. In contrast to the expression of PE and CPE mRNAs, cultured astrocytes ap pear to be more homogeneous in their expression of PAM mRNA. MATERIALS AND METHODS Cell culture Primary cultures of astrocytes were prepared from neonatal Sprague-Dawley rat brains, and neurons were prepared from embryonic (17 day) rat hypothalami, as previously described2’. Briefly, after dissection and dissociation, cells were plated on poly-o-lysinecoated, 60 mm diameter, Falcon Petri dishes or onto 12 mm circular coverslips in 24 we!! tissue culture plates. Astrocytes were plated at a density of 1 x lo6 cells/dish or 1 x 105ce!!s/coverslip and neurons

wereplatedat a density of

3 x lo6 cells/dish. Cells were grown in 45% (v/v) minima! essential medium (GIBCO)/&% (v/v) Ham’s F-12 medium (Gibco)/lO% (v/v) fetal calf serum (KC Biological) with 5 pg/m! insulin. To minimize astrocyte growth, the neuronal cultures were treated with 0.1 mM cytosine arabinoside after 3 days in culture. To ensure that cells would be at the same stage of the cell tW!e, astrocyte cultures were treated with 0.1 mM 2’-deoxy-5-fluorouridine after 10 days in culture. Astrocyte culture medium was replaced with ice-cold medium after 3 days in culture and once a week thereafter. Immunocytochemistry Astrocytes were labeled by immunocytochemical staining after 3 weeks in culture, as previously described*‘. Cells were tested for immunoreactivity to four different protein markers: glia! fibrillary acidic protein (GFAP), found intracellularly in astrocytes; A2B5, distributed on the surfaces of neurons and astrocytic precursors; fibronectin, a cell surface marker for fibroblasts; and galactocerebroside, found on the cell surfaces of oligodendrocytes. The results of this labeling demonstrated that the astrocyte cultures from frontal cortex and striatum were > 95% type I astrocytes (GFAP+, A2B5-, fibronectin-, galactocerebroside- ). Astrocytes cultured from hypothalamus, hippocampus and cerebellum were relatively pure, except for a small percentage of fibroblasts. The hypothalamic neuronal cultures consisted of 95% neurons and 5% mostly type I astrocytes (GFAP+, A2BS- ), after 2 weeks in culture. Northern blot analysis RNA was extracted from neurona! and astrocytic cultures after 2 and 3 weeks in culture, respectively, as previously described3’. Briefly, total RNA was prepared by phenol/chloroform extractions of cell homogenates, followed by ethanol precipitation. RNA was quantified by measuring the absorption at 260 and 280 nm. The ratio between absorption at 260 and 280 nm was > 1.7 for a!! samples. Ten pg of total nucleic acid was denatured at 50°C for 15 min and loaded onto a 1.5% agarose gel containing 2% formaldehyde. After electrophoresis, the gel was stained with ethidium bromide to visualize the 18 and 28s rRNA, and the nucleic acid in the gel was transferred to Gene Screen Plus membranes (New England Nuclear) by capillary blotting. The membranes were baked at 80°C in a vacuum oven for 2 h, and then prehybridized for 1 hour at 65°C in 50% formamide, 0,5% sodium dodecy! sulfate, 1 mM EDTA, 0.5 M NaC! and 100 &ml

denatured salmon sperm DNA. 32P-!abe!ed riboprobe (see below) was added to give a final concentration of 10” cpm/m!, and the membranes were incubated overnight at 65°C. Following hybridization, the membranes were washed with several changes of 02X standard saline citrate (SSC) containing 0.1% sodium dodecy! sulfate at 70°C, dried, and exposed to Kodak The membranes were also hybridized oligomeric probe (described below) in Sarkosy!, 1 xDenhardt’s solution at washing in 2xSSC at 50°C

XAR-5 X-ray film at -70°C. with 10” cpm/m! 18s rRNA 2~ SSC, 10 mM EDTA, 0.1% 45°C overnight, followed by

Probes The plasmid containing PAM cDNA was provided by Dr. Betty Eipper (Department of Neuroscience, Johns Hopkins University School of Medicine, Baltimore, MD) and consists of a 1.1 kb fragment of rat atria! PAM-1 cDNA in a Bluescript SK vector (M13-) flanked by T7 and T3 promoters I1 . The plasmid was linearized with restriction endonucleases EcoRI or 22~01,and T7 or T3 polymerases were used to generate ‘antisense’ or ‘sense’ cRNA probts, respectively. The ‘antisense’ probe is complementary to PAM mRNA. The plasmid containing lB15 cDNA was a gift of Dr. James Douglass (Vo!!mn Institute, Oregon Health Sciences University, Portland, OR) and consisted of 680 bp of the translated region of lB15 cDNA in the pSP-65 vector, as described’. The proopiomelanocortin (PGMC) probe was prepared from a plasmid containing a 920 bp mouse PGMC cDNA insert in the pSP-65 vector”; this plasmid was also a gift of Dr. James Douglass. The 18s rRNA probe was synthesized bY the Albert Einstein College DNA Synthesis Facility, and is comP!ementary to the first 45nucleotides of the 5’-end of rat ribosoma! 18s

204

-Astrocytes-

RNA6. This oligonucleotide was end-labelled with 32p using T4 polynucleotide kinase, cRNA probes were labeled with [32P]UTP for use in Northern blot analysis and with 35S-UTP for use in in situ hybridization experiments.

In situ hybridization and autoradiography In situ hybridization experiments were performed as previously described 2°. Briefly, astrocytes were fixed with 4% paraforma!dehyde in phosphate buffered saline (PBS), washed with PBS and stored under 70% ethanol at 4°C until hybridization. Prehybridization treatments of cells included acetylation with 0.25% v / v of acetic anhydride in 0.1 M triethanolamine, washes in 0.2xSSC, dehydration through increasing volumes of ethanol and treatment with 50 # g / m l denatured, sonicated salmon sperm DNA for 1 h at the hybridization temperature (see below). To control for non-specific binding, several coverslips were pretreated with 0.3 mg/ml RNase A. Astrocytes were hybridized with 'antisense' or 'sense' PAM cRNA probes for 2 days at 42°C. The hybridization buffc:r contained 50% formamide, I x Denhardt's solution, 3 × SSC, 10 taM dithiothreitol (DTT), 10% dextran sulfate, 0.1 mg/ml yeast tRNA, 50 mM sodium phosphate and 1 X 106 cpm of probe. After hybridization, cells were washed in decreasing concentrations of SSC and treated with 30 # g / m l RNase A in 0.5 M NaCI, l0 mM Tris pH 7.5, l mM EDTA for 30 min at 37°C. After several more washes in 0.2xSSC, cells were dehydrated in ethanols containing 0.3 M NH4Ac, mounted on glass slides and dipped in Kodak NTB-2 photographic emulsion, diluted l:l with 0.3 M NH4Ac at 42°C. Slides were exposed for 24 days at 4°C and then developed in D-19 developer and fixed in Kodak fix. Astrocytes were counterstained with hematoxylin and eosin.

RESULTS

Northern blot analysis of total RNA extractc~d/rom primary cultures of frontal cortex astrocytes reveals a single major band of PAM mRNA (figure 1). This species of mRNA is approximately 3.7 kb in length. RNA prepared from astrocytes cultured from various brain regions displays heterogeneity in the level of PAM mRNA. Astroeytes from all regions examined contain the 3.7 kb PAM mRNA as the predominant species (Fig. 1). In the striatum and hippocampus, another species of PAM mRNA with an approximate size of 4.4 kb is present in low levels; this species of PAM mRNA is not detectable in RNA isolated from frontal cortical cultures (Fig. 1). The cerebellum contains low but detectable levels of two species of PAM mRNA, approximately 3.7 and 4,2 kb in length (Fig. 1). The relative abundance of the various sizes of PAM mRNA varied slightly from one blot to another. For quantitation, the total densities of all bands were measured for this, and for two other Nothern blots (not shown). To control for the amount of RNA loade~ in each lane of the Northern blot and subsequently transferred to the filter, the Northern blots were rehybridized with 18S oligonucleotide probe. This probe detects only 18S rRNA ,Jnder the conditions employed for the hybridization and wash (data not shown). When expressed relative to the level of 18S rRNA, the level of

LL I

U3 I

C)

-28S -18S

Fig. 1. Northern blot analysis of PAM mRNA in astrocytes cultured from various brain regions. The Northern blot was hybridized with PAM cRNA probe as described in Materials and Methods.

PAM mRNA is similar in astrocytes and neurons cultured from the hypothalamus (Fig. 2A). The level of PAM mRNA it~ hypothalamic cultures is 1- to 6-fold higher than in astrocytes cultured from the frontal 4,

PAM mRNA 3

F,Ctx Hyp. Str, Hip, Cer. Astrocytes . . . . . .

Hyp.

I Neurons

tw

ImRNA J ~g F.Ctx Hyp. Str. Hip. Cer. Hyp. Astrocytes I Neurons Fig. 2, Comparison of the relative amounts of PAM mRNA and 1B15 mRNA in cultured astrocytes and neurons. Three Northern blots using separate preparations of RNA were analyzed as described in Materials and Methods. The radioactivity on the film was quantified with a Quantimet 920 image analysis system (Cambridge Institute), Data was adjusted to the level of I8S RNA for each lane of the Northern blots. Error bars show S,E.M.

205

Hypothalamic Astrocytes

Cerebellar Astrocytes

|

Fig. 3. In situ hybridization and immunocytochemistry analysis of cultured hypothalamic (A,C,E) and cerebellar (B,D,F) astrocytes. Astrocytes were hybridized with 'sense' (A,B) or 'antisense' (C,D) PAM cRNA probes. Cells were exposed to emulsion for 24 days. Following in situ hybridization, cells were stained with hematoxylin and eosin and only the nuclei are visible in these photographs. Cells were also tested for immunoreactivity with anti-GFAP antisera (E,F).

206 cortex, striatum, hippocampus, and cerebellum (Fig. 2A). When the same Northern blots were probed with IB15 riboprobe, a single band of approximately 1 kb wa: aetected in both glial and neuronal RNA (data not shown/. The relative abundance of 1B15 mRNA also shows some variation between cultures, with astrocytes cultured from frontal cortex and striatum containing higher levels of 1B15 mRNA than astrocytes cultured from the oilier regions or hypothalamic neuronal cultures (Fig. 2B). This distribution is different from that of PAM mRNA, and from the previously reported distribution of CPE 37 and somatostatin 29 mRNA. The Northern blots were also probed with POMC riboprobe: POMC mRNA was not detectable in cultures of astrocytes from any of the regions examined (data not shown). Previous studies of cultured astrocytes using quantitative in situ hybridization have detected a large variation between individual cells in their expression of PE and CPE mRNAs 2°. In order to determine whether PAM mRNA is also expressed heterogeneously by astrocytes, we performed in situ hybridization analysis on our cultures. Astrocytes hybridized with the control 'sense' PAM cRNA probe show a low level of nonspecific background labeling (Fig. 3A,B). Astrocytes pretreated with RNase A also do not demonstrate specific hybridization signals when hybridized with 'antisense' PAM probe (data not shown). In contrast, hybridization with 'antisense' PAM cRNA probes demonstrates substantially more grains over all cells in both the hypothalamic (Fig. 3C) and cerebellar (Fig. 3D) cultures. The level of PAM mRNA expression per cell appears to be lower in the cerebellar cultures (Fig. 3D). The immunocytochemical analysis indicates that 93-98% of the cells were GFAP positive, depending on the region from which they were derived (Fig. 3E,F). Quantitative analysis of the expression of PAM mRNA by hypothalamic astrocytes was performed by counting the numbers of grains over approximately 200 cells hybridized with either 'sense' or 'antisense' cRNA PAM probes. The cells hybridized with 'sense' probe, defined as the background level, contained an average grain density of 8.8 + 0.4 (S.E.M.) grains per cell (Fig. 4). Cells hybridized with the 'antisense' probe showed only a single population of cells, although this population was highly variable (Fig. 4). The average grain density was 30 + 1 (S.E.M.) grains per cell, with a standard deviation of 18. A population of cells expressing only background levels of grains was not detected by this analysis of 193 randomly selected cells hybridized with antisense PAM probe. In contrast, previ-

8O m

sense

7O

SO 0 to.

0 t.

50 4O

o JD 3O

E .-i

20

Z

10 0

0

25

SO

Ioo

OJk.

ontisense [

4o~

u

75

o o

3O

E •'!

10,

Z O, 0

2S

50

75

Number of groins per cell

1004.

Fig. 4. Frequency histograms of the grains per cell for cultured hypothalamic astrocytes hybridized with sense (top) or antisense (bottom) PAM cRNA probes, as described in Materials and Methods. For each histogram, emulsion-coated cells were imaged using a Quantimet 920 image analysis system (Cambridge Institute). The boundaries of the cells were determined using Nomarski optics, and the area of grains determined using the image analysis system. Then, the total grain area was divided by the average grain area to arrive at the total number of grains. This method allows the resolution of overlapping grains. The cells were randomly selected for this analysis; n--198 for the sense histogram, n--193 for the antisense histogram.

ous studies with CPE and PE mRNAs showed two populations of cells; one with only background levels of grains, and the other with levels of grains several-fold higher than background 2°. DISCUSSION These results demonstrate that cultured astrocytes express PAM mRNA, further supporting the proposal that these cells have the capacity to produce peptide hormones. In a recent in vivo study, PAM immunoreactivity was localized to several glial cell types including Schwann cells, oligodendrocytes, ependymal cells and astrocytes, as well as neurons 26. Previous studies have established the presence of PE, NGF, CPE, and somatostatin mRNAs in similarly cultured astroc y t e s 17'18'24'29'36'37 and the presence of PE and angiotensinogen mRNAs in astrocytes in brain sections 3°'33. The finding that PAM mRNA is expressed in cultured astrocytes raises the possibility that amidated forms of PE, such as metorphamide and amidorphin, are produced by astrocytes. Recently, astrocytes in culture and in vivo have been reported to contain Met-enkephalin and high molecular weight Met-enkephalin-containing peptides 2'24'3°. Little characterization of these peptides was performed, and the presence of amidated peptides was not examined.

207 The forms of PAM mRNA detected in this study are consistent with the 3.8 and 4.2 kb forms which are found in the central nervous system of the adult rat 4'32. PAM transcripts of 3.6 kb have also been reported in other tissues 4'32. The functional significance of the various sizes of PAM mRNA is not known, although the different forms presumably arise from differential splicing of exons within the coding regions 13'32. The regional variation in glial PAM mRNA levels, relative to the levels of 18S rRNA, is similar to the regional variation of PAM mRNA in brain tissue 4. PAM mRNA has been reported to be approximately 1to 2-fold more abundant in hypothalamus than in cortex, striatum, or hippocampus 4, much like the distribution of PAM mRNA in astrocytes (Fig. 2). Of all brain regions examined, the level of PAM mRNA is lowest in the cerebellum, as found for astrocytes cultured from this region. The similarity of the distribution of PAM mRNA in brain tissue and in cultured astrocytes, along with the observation that cultured hypothalamic astrocytes and neurons have similar levels of PAM mRNA raises the possibility that gila are a substantial source of PAM mRNA in the brain. The regional variations of PAM mRNA levels in cultured astrocytes is similar but not identical to the regional variations of CPE mRNA. Of all the brain regions examined, the lowest levels of both CPE and PAM mRNA are found in cerebellar astrocytes. CPE mRNA is highest in astrocytes cultured from the frontal cortex, with the relative density approximately 20-30fold higher than in astrocytes cultured from cerebellum aT. Astrocytes cultured from hypothalamus, striaturn, and hippocampus contain approximately half as much CPE mRNA compared to astrocytes from the frontal cortex 37. The regional variation of 1B15 mRNA in cultured astroeytes is fairly small, consistent with the tissue distribution of this mRNA7. Other investigators have used the tissue levels of 1B15 mRNA as a control for the amount of RNA loaded onto the gel, and normalized the amount of other species of mRNA to the level of 1B15 mRNA29'36. Since the tissue levels of 1B15 mRNA vary 1- to 2-fold, relative to the level of 18S rRNA, the accuracy of 1B15 as a control for the amount of mRNA is questionable. Astrocytes demonstrate specificity in their expression of prohormones. In a previous study, PE mRNA was found in astrocytes cultured from various brain regions, whereas prodynorphin mRNA was not detectable on the same Northern blots 36. The Northern blots analyzed in the present study were rehybridized to POMC probes. In all brain regions examined, POMC mRNA in cultured astrocytes was undetectable (data not shown). Thus, only one member of the opioid gene

family (PE) is expressed in detectable levels by cultured astrocytes. Other prohormone mRNAs have also been examined in astrocytes. Somatostatin mRNA is present in cultured cerebellar astrocytes, but not in astrocytes cultured from cortex or striatum 29. In the same study, cholecystokinin mRNA was not detectable in astrocytes from these three brain regions. Using a combination of in situ hybridization and immunocytochemistry with neuronal and glial markers, angiotensinogen mRNA has been localized to astrocytes in certain hypothalamic, midbrain, and brainstem nuclei 33. Other than these nuclei, few astrocytes were found to express angiotensinogen mRNA33. Interestingly, astrocytes cultured from the same brain region appear to be relatively homogeneous in their expression of PAM mRNA. This is in contrast with the expression of PE and CPE mRNAs, which are expressed by subpopulations of cultured astrocytes 2°. The differences in the expression of CPE, PAM, and PE mRNAs in cultured astrocytes indicates that these mRNAs are not coordinately expressed in astrocytes. This is consistent with the tissue distribution of PAM, CPE and PE mRNAs. While PAM is found in almost all tissues examined, CPE is predominantly found in neuroendocrine tissues, and PE is produced by specific cells in the brain and neuroendocrine tissues 4'9'14. Taken together, these results suggest that the astrocyte expression of prohormone mRNAs is not an artifact of the culturing conditions since only specific prohormone mRNAs are expressed, and prohormone processing enzymes are expressed along with the prohormone mRNAs. The physiological significance of this expression is not known. It is possib!e that the peptides are involved in development and/or maintenance of the neuronal architecture. Glia in this and other studies were cultured from neonatal rats. In preliminary studies, we have found that glia cultured from embryonic rats (day 17) express PE, CPE, and PAM mRNAs, consistent with a developmental role for neuropeptides. Further studies are needed to address this important issue. Acknowledgements. The data in this manuscript are from a thesis to be submitted in partial fulfillment of the requirements for the Degree of Doctor of Philosophy in the Sue Goiding Graduate Division of Medical Sciences, Albert Einstein College of Medicine, Yeshivw University (R.S.K.). This work was supported by in part by NIDA Grant DA-04494, an Alfred P. Sloan fellowship, and an Irma T. Hitschl fellowship (to L.D.F.), and by PHS Training Grant GM07260 (R.S.K.). The authors wish to thank Dr. Ruth Angeletti for helpful discussions.

REFERENCES 1 Barres,B.A.,Chun, L.Y.and Corey,D.P., Ion channel expression b~, white matter glia: 1. type 2 astrocytes and oligodendro~tes, Gila, 1 (1988) 10-30.

208 2 Batter, D.K., Vilijn, M-H. and Kessler, J.A., Cultured astrocytes release proenkephalin, Brain Res., 563 (1991) 28-32. 3 Bignami, A. and Dahi, D., The astroglial response to stabbing. Immunofluorescence studies with antibodies to astrocyte specific protein (GFA) in mammalian and submammalian vertebrates, Neuropathol. Appl. Neurobiol., 2 (1976) 99-110. 4 Braas, K.M., Stoffers, D.A.. Eloper, B.A. and May, V., Tissue specific expression of rat peptidylglycine a-amidating monooxygenase activity and mRNA, Mol. Endocrmol., 3 (1989) 1387-1398. 5 Bradbury, A.F. and Smyth, D.G., Modification of the N- and C-termini of bioactive peptides: a~idation and acetylation. In L.D. Fricker (Ed.), Peptide Biosynthesis and Processing, CRC Press, Boca Raton, 1991, pp. 231-250. 6 Chan, Y.L., Gutell, R., Nollers, H.F. and Wool, I.G., The nucleotide sequence of rat 18S ribosomal ribonucleic acid gene and a proposal for the secondary structure of 18S ribosomal ribonucleic acid, J. Biol. Chem., 259 (1984) 224-230. 7 Danielson, P.E., Forss-Peters, S., Brow, M.A., Calavetta, L., Douglass, J., Milner, R.J. and Sutcliffe, J.G., plBlS: A cDNA clone of the rat mRNA encoding cyclophilin, DNA, 7 (1988) 261-267. 8 Devi, L., Peptide processing at monobasic sites. In L.D.Fricker (Ed.) Peptide Biosynthesis and Processing, CRC Press, Boca Raton, 1991, pp. 175-198. 9 Eipper, B.A., Mains, R.E., Peptide a-amidation, Annu. ReL'. Physiol., 50 (1988) 333-44. 10 Eipper, B.A., Myers, A.C. and Mains, R.E., Peptidyl-glycine-~amidation activity in tissues and serum of the adult rat, Endocrinology, 116 (1985) 2497-2504. 11 Eipper, B.A., Park, L.P., Dickerson, I.M., Keutmann, H,T., Thiele, E.A., Rodriguez, H., Schofield, P.R. and Mains, R.E., Structure of the precursor to an enzymc mediating COOH-termihal amidation in peptide biosynthesis, Mol. Endocrinol., 1 (1987) 777-90. 12 Eipper, B.A., Perkins, S.N., Husten, E.J., Johnson, R.C., Keutmann, H.T. and Mains, R.E., Peptidyl-a-hydroxyglycine aamidating lyase: purification, characterization and expression, J. Biol. Chem., 266 (1991) 7827-7833. 13 Eipper, B.A., Green, C.B.R., Campbell, T.A., Stoffers, D.A., Keutmann, H.T., Mains, R.E. and Ouafik, L.H., Alternative splicing and endoproteolytic processing generate tissue-specific forms of pituitary peptidylglycine a-amidating monoox~genase, J. Biol. Chem., 267 (1992) 4008-4015. 14 Fricker, L.D., Peptide processing exopeptidases: amino- and carboxypeptidases involved with peptide biosynthesis. In L.D. Fricker (Ed.), Peptide Biosynthesis and Processing, CRC Press, Boca Raton, 1991, pp. 199-230. 15 Giulian, K., Vaca, K. and Johnson, B., Secreted peptides as regulators of neuron-glia and glia-glia interactions in the developing nervous system, J. Neurosci., 21 (1988) 487-500. 16 Hansson, E., Astroglia from defined brain regions as studied with primary cultures, Prog. Neurobiol., 30 (1988) 369-397. 17 Hauser, K.F., Osborne, J.G,, Steine-Martin, A. and Melner, M.H., Cellular localization of proenkephalin mRNA and enkephalin peptide products in cultured astrocytes, Brain Res., 522 (1990) 347-353. 18 Houlgatte, R., Mallat, M., Brachet, P. and Prochiantz, A., Secretion of nerve growth factor in cultures of glial cells and neurons derived from different regions of the mouse brain, J. Neurosci. Res., 24 (1989) 143-152. 19 Katopodis, A.G., Ping, D. and May, S.W., A novel enzyme from bovine neurointermediate pituitary catalyzes dealkylation of ahydrox'jlglycine derivatives, thereby functioning sequentially with peptidylglycine ot-amidating monooxygenase in peptide amidation, Biochemistry, 29 (1990) 6115-6120. 20 Klein, R.S. and Fricker, L.D., Heterogeneous expression of car-

boxypeptidase E and proenkephalin mRNAs in cultured astrocytes, Brain Res., 569 (1992) 300-310. 21 Lindberg, I. and Hutton, J.C., Peptide processing proteinases with selectivity for paired basic residues. In L.D. Fricker (Ed.), Peptide Biosynthesis and Processing, CRC Press, Boca Raton, 1991, pp. 141-174. 22 Lucias, R. and Mentlein, R., Degradation of the neuropeptide somatostatin by cultivated neuronal and glial cells, J. Biol. Chem., 266 (1991) 18907-18913. 23 MacCumber, M.W., Ross, C.A. and Snyder, S.H., Endothelin in brain: receptors, mitogenesis, and biosynthesis in glial cells, Proc. Natl. Acad. Sci. USA, 87 (1990) 2359-2363. 24 Melner, M.H., Low, K.G., Allen, R.G., Nielsen, C.P., Young, S.L. and Saneto, R.P., The regulation of proenkephalin expression in a distinct population of glial cells, EMBO J., 9 (1990) 791-796. 25 Perkins, S.N., Husten, E.J. and Eipper, B.A., The 108-kDa peptidylglycine a-amidating monooxygenase precursor contains two separable enzymatic activities involved in peptide amidation, Biochem. Biophys. Res. Commun., 171 (1990)926-932. 26 Rhodes, C.H., Xu, R.Y. and Angeletti, R.A., Peptidylglycine a-amidating monooxygenase (PAM) in Schwann cells and glia as well as neurons, J. Histochem. Cytochem., 38 (1990) 1301-1311. 27 Roberts, J.L., Seeburg, P.H., Shine, J., Herbert, E., Baxter, J.D. and Goodman, H.M. Corticotropin and /]-endorphin: construction and analysis of recombinant DNA complementary to mRNA for the common precursor, Proc. Natl. Acad. Sci. USA, 76 (1979) 2153-2157. 28 Schafer, M.K.-H., Stoffers, D.A., Eipper, B.A. and Watson, S.W., Expression of peptidylglycine a-amidating monooxygenase (EC 1.14.17.3) in the rat central nervous system, J. Neurosci., 12 (1992) 222-234. 29 Shinoda, H., Marini, A.M., Cosi, C. and Schwartz, J.P., Brain region and gene specificity of neuropeptide gene expression in cultured astrocytes, Science, 245 (1989) 415-417. 30 Spruce, B.A., Curtis, R., Wilkin, G.P. ~'nd Glover, D.M., A neuropeptide precursor in cerebellum: I~roenkephalin exists in subpopulations of both neurons and astrocytes, EMBO J., 9 (! 990) 1787-1795. 31 Steiner, D.F., The biosynthesis of biologically active peptides: a perspective. In L.D. Fricker (Ed.), Peptide Biosynthesis and Pro. cessing, CRC Press, Boca Raton, 1991, pp. 1-16. 32 Stoffers, D.A., Green, C.B.R, and Eipper, B.A., Alternative mRNA splicing generates multiple forms of peptidyl-glycine aamidating monooxygenase in rat atrium, Proc. NatL Acad. Sci. USA, 86 (1989) 735-739. 33 Stornetta, R.L., Hawelu-Johnson, C.L., Guyenet, P.G. and Lynch, K.R., Astrocytes synthesize angiotensinogen in brain, Science, 242 (1988) 1444-1446. 34 Takahashi, K., Okamoto, H., Seino, H. and Noguchi, M., Peptidylglycine a-amidating reaction: evidence for a two-step mechanism involving a stable intermediate at neutral pH, Biochem. Biophys. Res. Commun., 169 (1990)524-530, 35 Thiele, E.A., Marek, K.L. and Eipper, B.A., Tissue-specific regulation of peptidyl-glycine a-amidating monoxygenase expression, Endocrinology, 125 (1989) 2279-2288. 36 Vilijn, M-H,, Vaysse, P.J.-J., Zukin, R.S. and Kessler, J.A., Expression of preproenkephalin mRNA by cultured astrocytes and neurons, Proc. Natl. Acad. ScL USA, 85 (1988) 6551-6555. 37 Vilijn, M.H., Das, B., Kessler, J.A. and Fricker, L.D., Cultured astrocytes and neurons synthesize and secrete carboxypeptidase E, a neuropeptide-processing enzyme, J. Neurochem., 53 (1989) 1487-1493. 38 Young, S.D. and Tamburini, P.P., Enzymatic peptidyl a-amidation proceeds through formation of an a-hydroxylglycine intermediate, J. Am. Chem. Soc., 111 (1989) 1933-1934.