Proc. Nail. Acad. Sci. USA Vol. 89, pp. 5113-5117, June 1992 Neurobiology in C6-2B glioma cell mitochondria: Regulation by a mitochondrial binding inhibitor receptor (glial cells/steroidogenesis//diazepam binding inhibitor) VASSILIOS PAPADOPOULOS*, PATRIZIA GUARNERIt, KARL E. KRUEGERt, ALESSANDRO GUIDOTTIt, AND ERMINIO COSTAt *Department of Anatomy and Cell Biology and tFidia-Georgetown Institute for the Neurosciences, Georgetown University School of Medicine, 3900 Reservoir Road NW, Washington, DC 20007 Contributed by Erminio Costa, March 6, 1992

ABSTRACT The C6-2B glioma cell line, rich in mitochon- been proposed that glial cells can synthesize de novo drial receptors that bind with high affinity to benzodiazepines, (6, 7). In fact, glial cells can convert to preg- imidazopyridines, and isoquinolinecarboxamides (previously nenolone, which then may be further metabolized to preg- called peripheral-type receptors), was investi- nenolone sulfate, 3f3-hydroxy-5-pregnene-20-one, 3/3- gated as a model to study the significance of the polypeptide hydroxy-5-androstene-17-one, 3a-hydroxy-5a-pregnane-20- diazepam binding inhibitor (DBI) and the putative DBI pro- one, and 5a-pregnane-3a,21-diol-20-one, which positively or cessing products on mitochondrial receptor-regulated ste- negatively modulate the GABA-gating of Cl- channels (4, 8). roidogenesis. DBI and its naturally occurring fragments have Although detailed characterization of this biosynthetic appa- been found to be present in high concentrations in C6-2B ratus in glial cells is far from complete, the glioma cells, to compete against specific isoquinolinecarboxa- side-chain-cleavage (P450scc), which participates in mide or 4'-chlorodiazepam binding to mitochondrial recogni- biosynthesis by converting cholesterol to preg- tion sites with high affinity, and to stimulate mitochondrial nenolone, has been identified immunocytochemically in pri- pregnenolone formation. These data suggest that this cell type mary glial cultures and in specific rat brain regions (7, 9). may express both the receptor and the putative ligand Two important findings prompted the present investiga- to regulate steroidogenesis. Therefore, we propose to term this tion: (i) the observations that in adrenal and testis receptors mitochondrial receptor MDR (mitochondrial DBI receptor) to located on the outer membrane of mitochondria and which indicate its responsiveness to DBI in steroid biosynthesis. In the recognize benzodiazepines, isoquinolinecarboxamides, and present work, we show that mitochondria of C6-2B cells imidazopyridines with high affinity-previously defined as convert (22R)-22-hydroxycholesterol to pregnenolone by a peripheral-type benzodiazepine receptors-regulate intram- mechanism blocked by aminoglutethimide. Immunoblotting itochondrial cholesterol transport, the rate-limiting step in high of steroidogenesis (10-12), and (ii) the demonstration that di- confirmed the presence of relatively levels cytochrome azepam binding inhibitor (DBI), a polypeptide abundant in P-450 cholesterol side-chain-cleavage enzyme in C6-2B cell steroidogenic cells (1, 13, 14), can stimulate pregnenolone mitochondria. Furthermore, isoquinolinecarboxamide binding formation in adrenocortical and testicular Leydig cell mito- sites associated with the 18-kDa mitochondrial polypeptide chondria by an interaction with these mitochondrial recep- subunit of the MDR are abundant in C6-2B glioma cell tors (15, 16). The present experiments show that rat glioma mitochondria (B 30 pmol/mg ) and are coupled to C6-2B cells and mitochondria prepared from these cells the regulation of steroid biosynthesis. Occupancy of MDRs contain the P-45Scc enzyme and thereby convert cholesterol with nanomolar concentrations of the naturally occurring to pregnenolone. This enzymatic reaction can be activated by polypeptide, DBI, as well as its naturally occurring processing nanomolar amounts of DBI and the DBI fragment tetratria- product tetratriacontaneuropeptide [DBI-(17-50)] increases contaneuropeptide [TTN, DBI-(17-50)]. Since DBI, which is pregnenolone formation. and octadecaneuropep- synthesized and processed in glial cells (17), competes tide [DBI-(33-50)], which exhibit a higher affinity for r-ami- against the binding of the mitochondrial receptor ligands nobutyric acid type A receptors but a low affinity for MDR, 1-(2-chlorophenyl)-N-methyl-N-(1-methylpropyl)-3-isoquin- were ineffective in stimulating pregnenolone synthesis. These olinecarboxamide (PK 11195) and 4'-chlorodiazepam (4'CD) findings provide evidence that C6-2B cells exhibit a significant with an affinity comparable to its in the steroidogenic steroidogenic activity which resembles that found in peripheral effect, it appears to be the natural agonist for the receptor that endocrine organs and they suggest that MDRs and DBI are controls steroidogenesis in glial cells. Therefore, we suggest involved in the regulation of glial cell steroidogenesis. that this glial mitochondrial receptor be termed MDR (mito- chondrial DBI receptor). The specific interactions of steroids with putative membrane binding sites associated with the type A receptor for 'y-ami- METHODS nobutyric acid (GABAA receptor) (1-4) has prompted a reexamination of the rapid effects of various steroids on Cell Culture. The C6-2B subclone of the rat glioma C6 cell neuronal function as well as the investigation of the mecha- line was obtained from Gary Brooker at this institute (18). nisms that control the synthesis and release of steroids from These cells were adapted for growth in serum-free Aim V cells in the central nervous system. It appears that steroids of peripheral origin can be taken up from the circulation (5) by Abbreviations: GABA, y-aminobutyric acid; DBI, diazepam binding brain cells, where they bind to specific receptors to form inhibitor; MDR, mitochondrial DBI receptor(s); P450,,c, cholesterol transcription activating factors. In addition, it has recently side-chain-cleavage cytochrome P450 (EC 1.14.15.6); 4'CD, 4'- chlorodiazepam (Ro 5-4864); PK 11195, 1-(2-chlorophenyl)-N- methyl-N-(1-methylpropyl)-3-isoquinolinecarboxamide; PK 14105, The publication costs of this article were defrayed in part by page charge 1-(2-fluoro-5-nitrophenyl)-N-methyl-N-(1-methylpropyl)-3-isoquin- payment. This article must therefore be hereby marked "advertisement" olinecarboxamide; TTN, tetratriacontaneuropeptide; ODN, octade- in accordance with 18 U.S.C. §1734 solely to indicate this fact. caneuropeptide.

5113 Downloaded by guest on September 30, 2021 5114 Neurobiology: Papadopoulos et al. Proc. Natl. Acad. Sci. USA 89 (1992) medium (GIBCO). Culture dishes (150 mm diameter) were gel, and subjected to electrophoresis. were then pretreated with 10 ml ofDulbecco's modified Eagle's medium transferred to nitrocellulose membranes (0.45 pm; Hoefer) (DMEM) supplemented with 10% fetal bovine serum and blotted with a rabbit anti-bovine adrenal P-450SCc anti- (GIBCO) for 4 hr and washed with phosphate-buffered saline. body, obtained from Oxygene (Dallas) or normal rabbit Cells were then plated and grown in Aim V medium. This immunoglobulin. The antigens were detected with goat anti- procedure did not significantly alter the growth rate of the rabbit IgG coupled to horseradish peroxidase (Bio-Rad) using C6-2B cells compared with their growth in serum-containing 3-amino-9-ethylcarbazole and hydrogen peroxide as sub- media. strate (19). Measurement of Mitochondrial Steroid Biosynthesis. Mito- DBI Extraction, Purification, and Assay. DBI was extracted chondria from C6-2B cells were prepared as previously and purified from rat brain as previously described (20). The described (12) and resuspended at a concentration of 2.0 mg purity of the preparations used (>95%) was established by of protein per ml in buffer A (10 mM potassium phosphate SDS/PAGE, amino acid composition, and chromatographic buffer, pH 7.0/0.25 M sucrose/5 mM MgCl2/20 mM KCI/15 characteristics on reverse-phase HPLC (14, 20). DBI-like mM triethanolamine HCI) containing 5 ,uM (WIN- immunoreactivity of cell extracts was determined by radio- 24540; Sterling-Winthrop Group), an inhibitor of preg- immunoassay (14, 20). Peptides were synthesized and char- nenolone (11). Various concentrations of MDR acterized as previously described (21). Protein was quantified ligands to be tested were then added in buffer A. The mixture by using the procedure of Bradford (22) with gammaglobulin was preincubated for 5 min at 370C, and the reaction was as a standard. started by addition of 15 mM malate (neutralized with 1 M NaOH) and 0.5 mM NADP. The incubation was continued RESULTS for 15 min at 37TC and the reaction was stopped by the Cholesterol Side-Chain Cleavage in C6-2B Mitchodria. addition of 4 vol of cold ethanol. [3H]Pregnenolone [1500 The first step in steroid biosynthesis is the conversion of cpm, 22.6 Ci/mmol (1 Ci = 37 GBq); NEN/DuPont] was cholesterol to pregnenolone, which is catalyzed by P-450.C in added to each sample to monitor the recovery of the extrac- inner mitochondrial membranes. This side-chain-cleavage tion. The samples were then extracted three times with 5 vol reaction occurs via multiple hydroxylation steps of the cho- of n-hexane and the ethanolic phase was evaporated to lesterol side chain. The rate of the side-chain-cleavage reac- dryness. Pregnenolone was measured by a specific radioim- tion is limited by the availability of cholesterol to the intra- munoassay using the antibody provided by Radioassay Sys- mitochondrial P450,,,. To overcome this limitation (22R)- tems Laboratories (Carson, CA) and the conditions recom- 22-hydroxycholesterol, the first intermediate of the side- mended by this supplier. chain-cleavage reaction, can be supplied to mitochondria to Binding Studies in Sterokdogenic Buffer at 37C. [3H]PK attain maximal P450cC activity (23). When mitochondria of 11195 was obtained from NEN/DuPont. [3H]PK 11195 bind- C6-2B cells are incubated with (22R)-22-hydroxycholesterol ing studies on 10 ,g of protein from the mitochondria of a dose-dependent stimulation of pregnenolone formation is C6-2B glioma cells were performed in 250 Al ofbuffer at 37°C observed (Fig. 1). The synthesis of pregnenolone was inhib- in conditions identical to those under which the effects of ited by addition of aminoglutethimide, a P-450sCC inhibitor MDR ligands on steroidogenesis were studied. Samples were (24). Glial cell mitochondrial pregnenolone formation was incubated for 15 min with 2 nM [3H]PK 11195 and then they also increased, by 3- to 5-fold, upon addition of either 25- or were filtered on Whatman GF/C filters pretreated for 1 hr 20-hydroxycholesterol (not shown). Aminoglutethimide in- with 0.1% polyethyleneimine which were then washed with hibited pregnenolone formation from 25-hydroxycholesterol 15 ml of cold (4°C) 10 mM Hepes buffer, pH 7.4. Nonspecific but failed to affect the conversion of binding was determined in the presence of 10 ,M PK 11195 20-hydroxycholesterol (gift from C. Gudrdmy, Pharmuka, Gennevilliers, France). Total binding was approximately 10%6 of the total free radi- oligand included in the assay, and specific binding was 70-80% of the total binding. Ligands to be tested for their displacing activity were added to the mitochondrial suspen- .~~ sion 5 min before the addition of [3H]PK 11195. [3H]PK 11195 _.~~~~~~~~~~~~~ binding at 37°C reached steady-state conditions by 10 min. Scatchard analysis demonstrated a single class of binding sites with an apparent dissociation constant, Kd, of approx- I~~~~ imately 2 nM and an apparent maximal binding, Bm.x, of approximately 40 nmol/mg of protein. These values are similar to those of the [3H]PK 11195 binding measured in the same mitochondria at 4°C (see Fig. 2). Similar studies were performed using [3H]4'CD (NEN/DuPont) as ligand. Binding Studies in Hepes at 40C and Photolabeling of MDR. Radioligand binding experiments were performed in 10 mM Hepes, pH 7.4, at 4°C with 10 ,g of mitochondrial protein in a final volume of 400 ,ul. Samples were incubated for 2 hr to attain equilibrium and then filtered as reported before. Pho- tolabeling of C6-2B glial cell MDRs with 10 nM 1-(2-fluoro- FIG. 1. P-450. activity in C6-2B mitochondria. Mitochondria 5-nitrophenyl)-N-[3H]methyl-N-(1-methylpropyl-3-isoquino- from C6-2B cells were incubated with the indicated concentrations of linecarboxamide ([3H]PK 14105; Research Products Interna- (22R)-22-hydroxycholesterol in the presence (0) or absence (e) of tional) was performed as previously described (11). 0.76 mM aminoglutethimide, a PA450.cc inhibitor. After 15 min of Immunoblot Analyses. Y-1 adrenocortical (12), MA-10 incubation at 370C pregnenolone was quantified by radioimmunoas- Leydig (11), and C6-2B cells were washed twice with cold say. (Inset) Immunoblotting of the P-450Occ obtained with antibodies to the bovine adrenal-enzyme. Samples applied-on different lanes are phosphate-buffered saline and mitochondria were prepared signified as C6, C6-2B glioma cell mitochondria; Y-1, Y-1 mouse as previously described (12). Western blot analyses were adrenocortical cell mitochondria; and MA-10, MA-10 mouse Leydig performed by loading samples (25 ,ug of mitochondrial pro- cell mitochondria. Equal amounts ofprotein (25 Iug) were loaded onto tein) onto SDS/10% polyacrylamide gels, using a 4% stacking each lane. Downloaded by guest on September 30, 2021 Neurobiology: Papadopoulos et al. Proc. Natl. Acad. Sci. USA 89 (1992) 5115 mains (25) is identified in both C6-2B and rat adrenal mito- chondria by specific covalent labeling with the photoaffinity probe [3H]PK 14105 (Fig. 2 Inset). These results verify that C6-2B mitochondria contain an MDR protein with binding affinity characteristic for the isoquinolinecarboxamides [3H]PK 11195 and [3H]PK 14105 similar to those of mito- chondria present in other steroidogenic systems. ~~~~14- Regulation of Pregnenolone Synthesis in C6-2B Mitochon- G6 Ad dria by Synthetic and Natural Ligands of MDR. Two MDR synthetic ligands (4'CD and clonazapam) with various affin- . ities for MDR and the natural ligand DBI were tested for their effects on pregnenolone production by C6-2B mitochondrial fractions. The benzodiazepine 4'CD (Fig. 3A) and the 9-kDa polypeptide DBI (Fig. 3A) stimulated pregnenolone synthesis at low nanomolar concentrations with potencies consistent t) ~~1() 2() 30 with their abilities to displace [3H]PK 11195 (Fig. 3B) from B, prnol/mng protein MDR under the same experimental conditions used to study steroidogenesis. FIG. 2. Binding characteristics of MDR located in C6-2B mito- It is important to note that 4'CD binds with much higher chondria. Scatchard analysis ofthe binding isotherm of[3H]PK 11195 affinity to glial MDRs at 0°C than at 37°C, whereas the affinity in the concentration range of 0.07-20 nM. B, bound ligand; F, free of [3H]PK 11195 is virtually unaffected by the temperature ligand (M). (Inset) Autoradiogram ofrat adrenal (Ad) and C6-2B (C6) and buffer incubation conditions. The time course of the mitochondria photolabeled with [3H]PK 14105 and electrophoresed action of maximally stimulated concentrations of DBI and on a 15% polyacrylamide gel (11). 4'CD on mitochondrial pregnenolone synthesis indicates that the response is virtually linear up to 15 min and a plateau is into pregnenolone, thus suggesting a functional analogy be- reached shortly after that time (data not shown). The tween adrenal (24) and glial P-450scc. of these ligands is linear with concentration from 10-10 to Antibodies raised against purified bovine adrenal PA450,c 10-7 M, and it plateaus at 2-3 times the basal level (Fig. 3A); reveal the presence of an immunoreactive protein in C6-2B however, the efficacy tends to decline for concentrations mitochondrial fractions with the same molecular size as the above 10-6 M. In contrast, the benzodiazepine clonazepam, one present in Y-1 adrenocortical and testicular MA-10 a structural analogue of 4'CD, which binds with nanomolar Leydig cell mitochondria (Fig. 1 Inset). These findings sug- affinity to GABAA receptors but fails to bind MDR in gest that C6-2B cells express a mitochondrial cytochrome concentrations as high as 10-6 M (Fig. 3B), did not stimulate P-450 immunochemically similar to the enzyme present in pregnenolone formation (Fig. 3A). DBI-like immunoreactiv- well-characterized steroidogenic cells. ity was found in significant amounts (225 + 20 pmol/mg of Characterization ofMDR in C6-2B Cells. In adrenal cortical protein, n = 3) in C6-2B cell extracts, corresponding to a level cells and in testicular Leydig cells the intramitochondrial 50-fold higher than that required to obtain a maximal ste- transport ofcholesterol can be accelerated by specific ligands roidogenic response in mitochondrial preparations. The spec- of MDR acting as on this receptor (11, 12, 16). ificity of the steroidogenic effect of DBI was determined by Scatchard analysis of the [3H]PK 11195 binding to C6-2B using TTN [DBI-(17-50)] and octadecaneuropeptide [ODN, mitochondrial fractions at 4°C typically revealed a single DBI-(33-50)], two naturally occurring processing products of class of binding sites with an apparent Kd of 4 nM and a Bmax DBI. TTN, which is known to displace [3H]PK 11195 or of about 25-40 pmol/mg of protein (Fig. 2). The 18-kDa [3H]4'CD from glial cells and from brain mitochondria (17, 21, protein that was purified, sequenced, and shown to be a 26), stimulated C6-2B mitochondrial pregnenolone formation mitochondrial protein participating in the structure of the at low nanomolar concentrations (Fig. 4). In contrast, ODN, isoquinolinecarboxamide and benzodiazepine binding do- which exhibits selectivity for GABAA receptors but low

201 -1 r. A 1CK)o B Clonazepam 15 * DBI * * 8M0 boa.) 0- 4'CD E .aQ I 60o DBI bo I [0 0. r. ~~~~~4'CD ._ 40 5 Conazepam 2I00 "<

0 l ---. l l 9 |2 el 0.Ln .. ... -. ..I..., -10 -9 -8 -7 -6 -9 -8 -7 -6 -5 Ligand, log M

FIG. 3. Stimulation ofpregnenolone synthesis by specific MDR ligands. (A) C6-2B mitochondria were incubated with various concentrations of DBI (e), 4'CD (o), and clonazepam (m) for 15 min at 37°C. Pregnenolone was then extracted and measured by radioimmunoassay. Each point is the mean SEM of three experiments. *, P < 0.05 compared with synthesis with no added ligand. (B) Specific binding of 2 nM [3H]PK 11195 to C6-2B mitochondria under steroidogenic conditions was measured in the presence of the specific concentrations of each competing ligand and is expressed relative to the specific binding in the absence of a competitor. The points are the mean of three experiments; the SEM for each point is less than 5% of the mean. Downloaded by guest on September 30, 2021 5116 Neurobiology: Papadopoulos et al. Proc. Natl. Acad. Sci. USA 89 (1992) In rat adrenocorticotropin (ACTH), perhaps by

_15 means of cAMP production, appears to acutely control DBI processing and chronically control its gene expression (28, 29). However, the physiological stimuli that regulate DBI c ODN (f15 mn t 3°C biosynthesis in glial cells and whether DBI and TTN are ~10 physiological modulators of cholesterol accessibility to P450cC (via an interaction with MDR) remain to be eluci- -4) -1 -1D-B-I- dated. This important question may be addressed by decreas- ing the content of DBI in C6-2B glioma cells, incubating the cells with appropriate antisense oligonucleotides, and mea- 01 suring whether the steroidogenic capacity of the cells is decreased in proportion to the decrease of DBI. The finding that drugs (i.e., 4'CD) that bind with high -11 -10 -9 -8 -7 affinity to the MDR stimulate pregnenolone formation in Peptide, log M C6-2B cells highlights an important model system in which to study the pharmacological profile (agonist, antagonist, partial FIG. 4. Stimulation of glial cell mitochondrial steroidogenesis by of action. The studies described here reveal the DBI and TTN. C6-2B mitochondria were incubated with various agonist) drug

concentrations of DBI (e), TTN (v), or ODN (,) for 15 min at 370C. possibility that effects on glial steroidogenesis mediated by At the end of the incubation pregnenolone was extracted and MDR may give rise to changes in production ofneurosteroids measured by radioimmunoassay. Each value is the mean ± SEM of active at several possible levels of neuronal function, includ- three experiments. ing synaptic mechanisms as has been demonstrated for GABAA receptors (1-4) and glutamate receptors (30). affinity for MDR (17, 21), did not stimulate pregnenolone In the present report we have also demonstrated that the formation (Fig. 4). It should be noted that DBI and TTN were MDRs in C6-2B glioma cells not only have pharmacological slightly more efficacious than 4'CD. properties similar to those of adrenal or testis but also have an 18-kDa subunit with biochemical properties identical to DISCUSSION those of the receptor subunit cloned from rat adrenal cortex mitochondria. Thus the C6-2B glioma cell line is likely to be Previous reports have suggested that primary glial cells an excellent model system to gain insight on the molecular enriched in oligodendrocytes (7) have steroidogenic capacity. mechanisms (second and third messenger) that control MDR It was therefore to be expected that an established glioma cell gene expression, transcription, translation, or posttransla- line possess a similar activity. The present findings and the tional modifications [e.g., phosphorylation (31)] after drug recent results of Guarneri et al. (27) demonstrate that C6-2B treatment or modification of the culture conditions. glioma cells indeed have steroidogenic capacity and at the In summary, these findings demonstrate that C6-2B glioma same time express significant amounts of the polypeptide cells can metabolize cholesterol to pregnenolone, the parent DBI, which may control the steroidogenic activity of these component of all major steroids. Moreover, these data up- cells by interacting with a mitochondrial receptor, MDR, that hold the proposal that DBI may be the physiological modu- also recognizes imidazopyridines, isoquinolinecarboxam- lator of cholesterol accessibility to P450acc by means of its ides, and benzodiazepines with high affinity (6, 7). This claim action on MDR. is upheld by the following: (i) Mitochondria of C6-2B cells produce pregnenolone when (22R)-22-hydroxycholesterol is We gratefully acknowledge Sterling-Winthrop Group for the gen- supplied as a substrate, and this activity is blocked by the erous gift of trilostane, Hoffmann-La Roche for the gift of4'CD, and P-450scc inhibitor aminoglutethimide. (ii) Immunoblots of Dr. Gudrdmy for the gift of PK 11195. We are also grateful to Drs. C6-2B mitochondria reveal a protein immunoreactive with an S. M. Paul (National Institute of Mental Health, Bethesda, MD) and antibody specific for P-450scc. (iii) Pregnenolone synthesis in H. I. Yamamura (University of Arizona, Tucson) for critically DBI and natural reviewing this manuscript. V.P. was supported by a Fidia Research C6-2B mitochondria is stimulated by TTN, Foundation grant, National Science Foundation Grant DCB-9017752 ligands that potently compete with [3H]PK 11195 and and National Institutes of Health Grant DK-43358. K.E.K. was [3H]4'CD binding to C6-2B mitochondria. (iv) This stimula- partially supported by National Institutes of Health Grant MH- tion is dose related at low nanomolar concentrations. 44284. It has been reported that at 4TC under nonsteroidogenic conditions DBI and TTN are able to displace [3H]PK 11195 1. Costa, E. & Guidotti, A. (1991) Life Sci. 49, 325-344. or [3H]4'CD from glial cells or crude rat brain and adrenal 2. Majewska, M. D., Harrison, N. L., Schwartz, R. D., Barker, homogenate preparations (14, 17, 21, 26). The present study J. L. & Paul, S. M. (1986) Science 232, 1004-1007. demonstrates that when DBI is tested on purified C6-2B 3. Gee, K. W., Bolger, M. B., Brinton, R. E., Coirini, H. & mitochondria under steroidogenic conditions at 37°C, it ex- McEwen, B. S. (1988) J. Pharmacol. Exp. Ther. 246, 803-812. 4. Puia, G., Santi, M. R., Vicini, S., Pritchett, D. B., Purdy, hibits nanomolar affinity for MDRs, in good agreement with R. H., Paul, S. M., Seeburg, P. H. & Costa, E. (1990) Neuron its stimulatory potency on steroid biosynthesis, via P-450SCC. 4, 759-765. The observation that DBI and TTN stimulate glial cell 5. McEwen, B. S., De Kloet, E. R. & Rostene, W. (1986) Physiol. mitochondrial steroid biosynthesis is significant when con- Rev. 66, 1121-1188. sidering that C6-2B cells contain levels of DBI above the 6. Hu, Z. Y., Bourreau, E., Jung-Testas, I., Robel, P. & Baulieu, amount required for steroidogenic potency. Moreover, the E. E. (1987) Proc. Natl. Acad. Sci. USA 84, 8215-8219. expression of an 18-kDa mitochondrial receptor subunit that 7. Jung-Testas, I., Hu, Z. Y., Baulieu, E. E. & Robel, P. (1989) is covalently photolabeled by PK 14105 and the ability ofDBI Endocrinology 125, 2083-2091. to compete with [3H]PK 11195 binding to this receptor further 8. Mienville, J. M. & Vicini, S. (1989) Brain Res. 489, 190-194. the 9. Le Goascogne, C., Robel, P., Gouezou, M., Sananes, N., suggest that this peptide is the endogenous agonist for Baulieu, E. E. & Waterman, M. (1987) Science 237, 1212-1215. mitochondrial receptors that control glial steroidogenesis. 10. Yanagibashi, K., Ohno, Y., Nakamichi, N., Matsui, T., Ha- Thus the term "MDR" for this class of glial mitochondrial yashida, K., Takamura, M., Yamada, K., Tou, S. & Kawa- receptors is upheld. mura, M. (1989) J. Biochem. (Tokyo) 106, 1026-1029. It can be anticipated that changes of DBI content in glial 11. Papadopoulos, V., Mukhin, A. G., Costa, E. & Krueger, K. E. cells may result in changes in the amount of steroids formed. (1990) J. Biol. Chem. 265, 3772-3779. Downloaded by guest on September 30, 2021 Neurobiology: Papadopoulos et al. Proc. Natl. Acad. Sci. USA 89 (1992) 5117

12. Krueger, K. E. & Papadopoulos, V. (1990) J. Biol. Chem. 265, Bennet, C. D. & Costa, E. (1983) Proc. Natl. Acad. Sci. USA 15015-15022. 80, 3531-3533. 13. Gray, P. W., Glaister, D., Seeburg, P. H., Guidotti, A. & 21. Berkovich, A., McPhie, P., Campagnone, M., Guidotti, A. & Costa, E. (1986) Proc. Natl. Acad. Sci. USA 83, 7547-7551. Costa, E. (1990) Mol. Pharmacol. 37, 164-172. 14. Bovolin, P., Schlichting, J., Miyata, J., Ferrarese, C., Guidotti, 22. Bradford, M. M. (1976) Anal. Biochem. 72, 248-254. A. & Alho, H. (1990) Regul. Pept. 29, 267-281. 23. Hall, P. F. (1984) Int. Rev. Cytol. 86, 53-95. 15. Besman, M. J., Yanagibashi, K., Lee, T. D., Kawamura, M., 24. Gower, D. B. (1974) J. Steroid Biochem. 5, 501-523. Hall, P. F. & Shively, J. E. (1989) Proc. Natl. Acad. Sci. USA 25. Sprengel, R., Werner, P., Seeburg, P. H., Mukhin, A. G., 86, 4897-4901. Santi, M. R., Grayson, D. R., Guidotti, A. & Krueger, K. E. (1989) J. Biol. Chem. 264, 20415-20421. 16. Papadopoulos, V., Berkovich, A., Krueger, K. E., Costa, E. & 26. Slobodyansky, E., Guidotti, A., Wambebe, C., Berkovich, C. Guidotti, A. (1991) Endocrinology 129, 1481-1488. & Costa, E. (1989) J. Neurochem. 53, 1276-1284. 17. Guidotti, A., Alho, H., Berkovich, A., Cox, D. H., Ferrarese, 27. Guarneri, P., Papadopoulos, V., Pan, B. & Costa, E. (1992) C., Slobodyansky, E., Santi, M. R. & Wambebe, C. (1989) in Proc. Natl. Acad. Sci. USA 89, 5118-5122. Allosteric Modulation of Amino Acid Receptors: Therapeutic 28. Massotti, M., Slobodyansky, E., Konkel, D., Costa, E. & Implications, eds. Barnard E. A. & Costa E. (Raven, New Guidotti, A. (1991) Endocrinology 129, 591-596. York), pp. 100-123. 29. Cavallaro, S., Massotti, M., Mocchetti, I., Krueger, K. E., 18. De Vellis, J. & Brooker, G. (1973) in Tissue Culture of the Casalotti, S., Guidotti, A. & Costa, E. (1991) Soc. Neurosci. Nervous System, ed. Sato, G. (Plenum, New York), pp. 231- Abstr. 17, 1477. 245. 30. Wu, F., Gibbs, T. T. & Farb, D. H. (1991) Mol. Pharmacol. 40, 19. Graham, R. C., Ludholm, U. & Karnovsky, M. J. (1965) J. 333-336. Histochem. Cytochem. 13, 150-158. 31. Whalin, M. E., Papadopoulos, V., Costa, E. & Krueger, K. E. 20. Guidotti, A., Forchetti, C. M., Corda, M. G., Konkel, D., (1991) Soc. Neurosci. Abstr. 17, 917. Downloaded by guest on September 30, 2021