Identification and Characterization in Rat Brain Membranes (Receptor Binding/Somatostatin Analogs/Neuropeptides/Neuropharmacology) COIMBATORE B

Proc. Nati Acad. Sci. USA Vol. 78, No. 6, pp. 3930-3934, June 1981 Neurobiology

Somatostatin receptors: Identification and characterization in rat brain membranes (receptor binding/somatostatin analogs/neuropeptides/neuropharmacology) COIMBATORE B. SRIKANT AND YOGESH C. PATEL* Fraser Laboratories, McGill University, Departments of Medicine, Neurology and Neurosurgery, Royal Victoria Hospital, and Montreal Neurological Institute, Montreal, Quebec H3A lAl, Canada Communicated by Brenda Milner, February 23, 1981

ABSTRACT We have identified and characterized specific sin, and bombesin were purchased from Bachem Fine Chem- receptors for tetradecapeptide somatostatin (SRIF; somatotropin icals (Torrance, CA). Other analogs of SRIF were provided by release-inhibiting factor) in rat brain using ['25I-Tyr"]SRIF as the D. Coy and C. Meyers (New Orleans, LA) and J. Rivier (La radioligand. These receptors are present in membranes obtained Jolla, CA). Synthetic 13H-endorphin was obtained from N. Ling from a subfraction of synaptosomes. Membranes derived from (La Jolla, CA), and [Met]enkephalin was from G. Tregear (Mel- cerebral cortex bind SRIF with high affinity (K. = 1.25 X 1010 bourne, Australia). Vasoactive intestinal polypeptide (VIP) was M-') andhave a maximum bindingcapacity (Bmw) of0. 155 X 10-12 a gift from S. I. Said (Dallas, TX). All other chemicals used were mol/mg. Neither opiates nor other neuropeptides appear to in- of analytical grade. fluence the binding of SRIF to brain membranes. Synthetic ana- Brain Membranes. Adult male logs with greater biological potency than SRIF-[D-Trp8]SRIF, Preparation of Sprague- [D-Cys14]SRIF, and [D-Trps, D-Cys'4]SRIF-bind to the recep- Dawley rats (150-200 g) were killed by decapitation. The cer- tors with greater avidity than SRIF, whereas inactive analogs ebral cortex was dissected and homogenized (10%, wt/vol) in [(2H)Ala3]SRIF and [Ala6]SRIF exhibit low binding. The ratio of 0.32 M sucrose/20 mM Tris-HCl, pH 7.5, with a Dounce ho- receptor density to endogenous somatostatin is high in the cortex, mogenizer, and membrane fractions were prepared by a mod- thalamus, and striatum, low in the hypothalamus, and extremely ification of the method of Cotman and Matthews (16). The ho- low in the brain stem and cerebellum. Thus, SRIF receptors in mogenate was centrifuged at 1000 X g for 5 min to remove the the brain appear to be a distinct, new class of receptors with a nuclear debris (fraction P1). The supernatant obtained was cen- regional distribution different from that ofendogenous somatostatin. trifuged at 10,000 x g for 45 min. The resulting pellet containing the crude mitochondrial fraction (P2) was resuspended in The tetradecapeptide somatostatin (SRIF; somatotropin re- 0.32 M sucrose, applied to the top of a discontinuous Ficoll lease-inhibiting factor) originally discovered as a growth hor- (Pharmacia, Uppsala, Sweden) gradient (8-20%) in 0.32 M su- mone release-inhibiting factor in extracts ofsheep hypothalami crose, and sedimented at 63,580 X g for 45 min in a Beckman (1) has since been shown to be distributed throughout the ex- L5-65 ultracentrifuge. Subcellular fractions sedimenting in 20% trahypothalamic brain and in peripheral tissues and to exert a (vol/vol) Ficoll (P3) and at the different density interfaces 16- wide spectrum ofbiological actions (24). Although the precise 20% (P4), 12-16% (P5), 8-12% (P6), and 8% Ficoll/sucrose in- function of somatostatin in the central nervous system is un- terface (P7) were carefully collected and pelleted in 0.32 M su- known, the extensive distribution of somatostatin-containing crose at 10,000 X g. The synaptosomal fractions P5 and P6 were neurons in the brain (5), its synaptosomal localization (6), the hypoosmotically ruptured by swelling in 5 mM Tris-HCI (pH effects of microiontophoretically applied SRIF on the sponta- 7.5) twice. The final pellets obtained by centrifugation at 10,000 neous electrical activity of neurons (7), and the behavioral ef- x g were resuspended in 50 mM Hepes-KOH (pH 7.5). Ali- fects (8) of intracisternally injected SRIF (8) suggest that the quots of the membranes were resuspended in an identical vol- peptide may act as a neurotransmitter or neuromodulator (7, ume of 50 mM Tris-HCl (pH 7.5) for determination of protein 9). There is good evidence to suggest that the action of SRIF content (17) because Hepes interferes in this assay. Membrane on growth hormone secretion is mediated through binding to fractions from other areas ofthe brain were also prepared in this specific receptors which have been demonstrated in rat pitui- manner to determine the regional distribution ofSRIF receptors. tary tumor cells in culture (10) and in bovine pituitary plasma Histological Techniques. Membrane pellets isolated from membranes (11). Similar binding of SRIF to subcellular com- the cerebral cortex were fixed in 2.5% (vol/vol) glutaraldehyde ponents of neural tissue has not been reported. Therefore, the in 0.1 M sodium cacodylate for 2 hr, washed with 0.1 M sodium present study was designed to identify and characterize specific cacodylate/7% (wt/vol) sucrose, postfixed (2 hr) in Karnovsky's receptors for SRIF in rat brain synaptosomal membrane frac- reduced osmium (18), dehydrated, and embedded in epon; sec- tions that have been shown to possess specific binding sites for tions were prepared for electron microscopy. other neuropeptides (12-15). Preparation of Radioiodinated Peptide. Because SRIF does not contain a tyrosine or histidine residue, it is necessary to use MATERLALS AND METHODS suitable synthetic analogs for radioiodination. Three tyrosinated Synthetic cyclic SRIF was obtained from Ayerst Laboratories analogs ofSRIF, [Tyr']SRIF, Tyr-SRIF, and [Tyr"]SRIF, have (Montreal, Canada). Tyrosinated analogs of SRIF, thyrotropin- luteinizing hormone- Abbreviations: SRIF, synthetic tetradecapeptide somatostatin (soma- releasing hormone (TRH; thyroliberin), totropin release-inhibiting factor); VIP, vasoactive intestinal polypep- releasing hormone (LHRH; luliberin), substance P, neuroten- tide; TRH, thyrotropin-releasing hormone; LHRH, luteinizing hormone-releasing hormone (luliberin). The publication costs ofthis article were defrayed in part by page charge * To whom reprint requests should be addressed at: Room M3-10, payment. This article must therefore be hereby marked "advertise- Royal Victoria Hospital, 687 Pine Avenue West, Montreal, Quebec ment" in accordance with 18 U. S. C. §1734 solely to indicate this fact. H3A lAl, Canada. 3930 Downloaded by guest on October 2, 2021 Neurobiology: Srikant and Patel Proc. Natl. Acad. Sci. USA 78 (1981) 3931

been developed for this purpose (19) and have been shown to Table 1. Integrity of tracer after incubation with synaptosomal possess similar biological activities to the native peptide (19, 20). cell membranes (fraction P5)* All three were selected for our SRIF binding experiments. They % intact were radioiodinated with Na1'2I (New England Nuclear) to high tracer specific activities (=1050 Ci/mmol; 1 Ci = 3.7 x 1010 becque- Tracer A B rels) by a modification of the chloramine-T technique as de- scribed (3). [125I-Tyr']SRIF 35 42 Stability of Radioiodinated SR-IF Analogs During Incuba- 125I-Tyr-SRIF 53 73 71 97 tion. Because SRIF and radioiodinated SRIF analogs are rapidly [125I-Tyr11]SRIF degraded by aminopeptidases and endopeptidases in plasma * In the absence (A) and presence (B) of antiproteolytic agents: Tra- and tissue extracts (21, 22), and because brain synaptosomal sylol, phenylmethylsulfonyl fluoride, and bacitracin. fractions are known to be associated with these enzymes (23), we first investigated the stability of the radioiodinated analogs teolytic agents were compared, [125I-Tyr']SRIF showed low during incubation with the synaptosomal membrane fractions specific binding (<1%) and high nonspecific binding (.14%). P5 and P6. Approximately 130,000 cpm (0.75 nM) of each 1I-Tyr-SRIF exhibited paradoxical binding in that more ra- analog was added to 0.50 ,ug of membrane protein and incu- dioligand was bound in the presence of 10 AuM SRIF than in its bated for 1 hr at 300C in the absence and presence of the fol- absence. Because of this phenomenon, it was not possible to lowing enzyme inhibitors: Trasylol (aprotinin; Bayer, Federal detect any specific binding with this radioligand. On the other Republic ofGermany) (500 kallikrein inhibitor units/ml), phen- hand, the best binding parameters were obtained with [125I- ylmethylsulphonyl fluoride (0.02 jig/ml), and bacitracin (0.02 Tyr"]SRIF. For these reasons, [125I-Tyrl]SRIF was the radioli- ,ug/ml). The final reaction volume was adjusted to 100 ,i1 with gand chosen for all subsequent binding experiments. 50 mM Hepes-KOH (pH 7.5) containing bovine serum albumin Identification of Membrane Fractions Containing Specific (10 mg/ml) and MgCl2 (5 mM). At the end of the incubation, Receptors for SRIF. The specific and nonspecific binding of the free radioligand was separated from bound radioligand by ['0I-Tyr"1]SRIF to the different subcellular fractions of the rat centrifugation and subjected to chromatoelectrophoresis (3) to cerebral cortex are compared in Table 2. Fraction P5 exhibited determine the percentage of intact tracer remaining. the highest specific and lowest nonspecific binding (9.67% and Determination of Specific Binding of Radioligand. About 4.12%, respectively). Morphologically this subfraction was en- 0.60 nM ofthe appropriate radioligand with 50 ,ug ofmembrane riched in synaptosomes (Fig. 1). Less than 0.5% specific binding protein were incubated with each antiproteolytic agent at 300C was observed in fraction P6. Both these fractions contained high until equilibrium had been reached. Radioligand bound to the concentrations of endogenous somatostatin, 9.7 and 11.0 ng/ membrane was separated by centrifugation (2500 x g), washed mg of protein, respectively, which could be released by hy- twice with 1 ml of 50 nM Hepes-KOH, pH 7.5/1% albumin, poosmotic lysis of the synaptosomes (Table 2). Removal of the and the radioactivity in the resulting pellet was assayed in a endogenous somatostatin in this manner resulted in a 7-fold in- Beckman L-4000 autogamma spectrometer. Specific binding crease in the specific binding of ['2I-Tyr"]SRIF to the mem- was defined as the difference in the amount ofradioligand bound branes in P5, whereas no increase was seen in fraction P6. Frac- in the absence and presence of 10 pM SRIF. Calculation of the tion P7, which consists mainly of myelin (16), exhibited 2.1% binding data and determination of kinetic constants were per- specific binding but, because of the high nonspecific binding formed by Scatchard analysis (24). (>11%), was not investigated further in the present study. All Measurement of Endogenous Somatostatin. Endogenous equilibrium binding studies were performed with membranes somatostatin immunoreactivity in brain tissues and in the dif- from fraction P5. ferent membrane fractions was extracted in 1 M acetic acid at Binding of [l251-Tyrl]SRIF at Equilibrium. The binding of 0°C by sonication and measured by radioimmunoassay (3). [125I-Tyr"]SRIF to rat brain membrane fraction P5 was time dependent and had reached equilibrium by 40 min (Fig. 2A). RESULTS Nonspecific binding in the presence of 10 ,AM SRIF reached a maximal level by 20 min and remained constant thereafter. Stability of Radioiodinated SRIF Analogs During Incuba- When specific binding under these conditions was plotted semi- tion and Selection ofRadioligand for Binding Experiments. All logarithmically, the linear result (Fig. 2B) suggested pseudo- three radioiodinated tyrosine (125I-Tyr) analogs of SRIF under- went extensive degradation during incubation with brain mem- Table 2. Comparison of endogenous somatostatin branes (fractions P5 and P6). The integrity of the three radioli- immunoreactivity and the binding of ['25I-Tyrll]SRIF to gands during incubation with synaptosomal membranes obtained subcellular fractions of rat cerebral cortex fraction that exhibited maximum specific Endogenous % from fraction P5 (the [125I yrl bound,t binding) are compared in Table 1. ['25I-Tyr']SRIF was partic- somatostatin,* % _['21-TrSRIF ularly susceptible to proteolytic damage, with only 35% of the Fraction ng/mg Specific Nonspecific radioactivity remaining as intact ligand after incubation with the P1 6.61 0.08 ± 0.01 10.00 ± 1.10 synaptosomal membranes in the absence of enzyme inhibitors. P2 9.93 1.1 ± 0.15 12.00 ± 2.07 Under these conditions 125I-Tyr-SRIF and [125I-Tyr"]SRIF P3 4.50 0 7.00 0.73 were relatively more stable (53% and 71%, respectively, ofeach P4 7.40 0.52 + 0.2 5.38 ± 0.85 ligand remaining intact). The stability of all three radioligands P5 9.10 (1.20)t 9.67 ± 1.39 4.12 ± 0.98 ± was enhanced by the addition of the protease inhibitors Tra- P6 11.20 (1.86)t 0.52 ± 0.18 5.70 1.56 sylol, phenylmethylsulfonyl fluoride, and bacitracin. In partic- P7 4.10 2.10 0.76 11.04 3.24 ular, ['25I-Tyr11]SRIF was found to be almost completely pro- * Endogenous somatostatin was determined in these fractions after tected against degradation, with 97% of the radioactivity extraction in 1 M acetic acid. remaining as intact radioligand after incubation. t Mean ± SEM of three determinations, each in triplicate. When the bindings of the three radioligands to the brain t Values in parentheses indicate the endogenous somatostatin content membranes (fractions Pl-P7) in the presence of the antipro- in these fractions after hypoosmotic rupture of synaptosomes. Downloaded by guest on October 2, 2021 3932 Neurobiology: Srikant and Patel Proc. Natl. Acad. Sci. USA 78 (1981)

+-uwm w ti.']; =S2 s. ^

FIG. 1. Morphological appearance ofsynaptosomal subfraction P5. Numerous synaptosomes, some containing intraterminal mitochondria, are seen. (x 10,700.) first-order kinetics (25). The half-time for approach to equilib- to inhibit ['25I-Tyr"]SRIF binding compared to SRIF (Ta- rium binding was 9 min, which corresponds to a pseudo-first- ble 3). On the other hand, two biologically inactive analogs order rate constant (kobs) of0.077 min1. To test the reversibility [(2H)Ala3]SRIF and [Ala6]SRIF (19) were much less effective ofthe binding of['25I-Tyr11]SRIF to the membrane fraction P5, the receptor-bound radioligand at equilibrium was removed 12 A- from the free radioligand and resuspended in the incubation medium with and without added unlabeled SRIF (0.1 MM), and the time course of dissociation was followed at the same tem- perature. The rates ofdissociation under these conditions were identical, the half-time for dissociation being 26 min, which corresponds to a rate constant koff = 0.027 mind. When SRIF was added to the incubation medium, the spe- an,4 A A_ cific binding of[12I-Tyr"]SRIF was inhibited in a dose-depen- Cl) dent manner over a concentration range of0.1 nM to 1 ,.M (Fig. 3). Fig. 3 Inset shows the Scatchard analysis (24) ofthe binding o~~~ data. The linearity of the plot indicates a single class of noninteracting sites with a maximum binding capacity of0. 155pmol/ 20 40 60 mg. The affinity constant was calculated to be 1.25 x 101 M-1. The inhibition curves obtained with [Tyr"]SRIF and SRIF were 1.0 B identical, suggesting that both these peptides had comparable affinity for binding to the receptors (data not shown). 0.6- Specificity of Binding of ['25I-Tyr"]SRIF. Various neuro- E0.4_ peptides were tested for their ability to inhibit the binding of [5I-Tyrl"]SRIF to the brain membranes. In contrast to 0.5 AM SRIF, which produced 90% inhibition of['5I-Tyr"]SRIF bind- 0.2_ ing, up to 1 AM TRH, LH-RH, [Met]enkephalin, 13H-endorphin, substance P, VIP, and bombesin were without effect (Fig. 0 10 20 30 4). These peptides also did not alter the ability ofSRIF to inhibit Time, min the specific binding of [125I-Tyr"]SRIF. Neurotensin (1 AuM) exhibited slight inhibition of ['2I-Tyr"]SRIF binding, but this FIG. 2. Rate of binding of [125I-Tyrll]SRIF to rat brain synapto- effect was <0.1% that of SRIF. somal membranes. Specific binding (0) represents the difference between [125I-Tyrl]SRIF bound in the absence (0) and presence (A) of 10 The specificity ofSRIF binding was further investigated with pM unlabeled SRIF at each time point (mean of three experiments in several synthetic SRIF analogs. The biologically active analogs triplicate). (B) Specific binding observed in A was plotted semiloga- [D-Trp8]SRIF, [D-Cys'4]SRIF, and [D-Trp , D-Cys 4]SRIF (19, rithmically as a function of time. B,, Specific binding at time t, Bm., 26, 27) possessed greater ability (5-, 2-, and 3-fold, respectively) pecific binding at equilibrium. Downloaded by guest on October 2, 2021 Neurobiology: Srikant and Patel Proc. Nati. Acad. Sci. USA 78 (1981) 3933

Table 3. Potency of SRIF and its analogs in inhibiting ['25I-Tyr11ISRIF binding Relative potency, a Analog IC50,* nM % of SRIF :a 0 SRIF 1.27 100 [D-Trp8]SRIF 0.27 470 [D-Cys14ISRIF 0.67 190 [D-Trp8, D-Cys14]SRIF 0.43 295 [(2H)Ala3]SRIF 270 0.5 [Ala6]SRIF 740 0.2 *The concentration of peptide required to inhibit 50% of [125I- 1o-11 1lo10 09 lo-, 10-7 10-6 Tyr11]SRIF binding was estimated from inhibition curves. SRIF, M DISCUSSION FIG. 3. Inhibition ofthe specific bindingof["2'1-Tyr11ISRIF to syn- We have identified and characterized receptors for SRIF in a aptosomal membranes in the presence ofincreasing concentrations of subfraction of synaptosomal membranes isolated from rat cer- SRIF. Each point represents the mean of three experiments in tripli- ebral cortex that interact with SRIF in a specific, saturable, and cate. Increasing concentrations ofSRIF, [Tyr11]SRIF, and [125I-Tyr]- SRIF inhibited the binding ofthe ['"I-Tyr'1]SRIEF to the same- extent reversible manner. Synaptosomes have been shown to sedi- (details not shown). (Inset) Scatchard plot of the binding data. Slope ment in a broad band in continuous gradients of sucrose and oftheplotgivesaKa= 1.25 x 1010M-',andtheinterceptontheabscissa Ficoll and appear to be heterogenous (28, 29). They also have gives a binding capacity of 0.155 pmol/mg of membrane protein. been subfractionated in discontinuous gradients (30). The two subcellular fractions P5 and P6 identified in the present study appear to be enriched in synaptosomes as shown morphologi- in inhibiting binding of the radioligand. cally for P5 (Fig. 1) and as reported for the fraction correspond- Regional Distribution of Endogenous Somatostatin and ing to P6 (16). Furthermore, both fractions contained a high SRIF Receptors in Brain. The specific binding of ['"I- concentration of endogenous somatostatin, which decreased Tyr"]SRIF to synaptosomal membranes (fraction P5) isolated substantially after hypoosmotic rupture of the synaptosomes. from different areas ofthe rat brain was compared with the en- Thus, our finding of a marked difference in SRIF binding be- dogenous somatostatin concentration in these brain regions tween the two fractions provides further evidence for the het- (Table 4). The highest density of SRIF binding sites was found erogeneity of synaptosomes. The high density of receptor sites in the cerebral cortex (155 fmol/mg). Thalamus, hypothalamus, in synaptosomes suggests that SRIF may function as a neuro- and striatum contained intermediate receptor densities, whereas transmitter. The subcellular component in the heavier synap- the medulla, pons, and cerebellum exhibited low specific bind- tosomal fraction P5 that contains specific receptors for SRIF ing. When the receptor concentration in the different tissues remains to be identified. was examined in relation to the endogenous somatostatin con- ['25I-Tyr"]SRIF was found to be the most suitable radioli- tent, marked differences were observed. The ratio of receptor gand for studying receptor binding. In the two previous reports to endogenous somatostatin was high in the thalamus, cortex, of SRIF receptors in the pituitary, ['25I-Tyr']SRIF was used as and striatum, low in the hypothalamus, and extremely low in the radioligand (10, 11). This radioligand was found to undergo the medulla/pons and cerebellum. up to 20% degradation during incubation with pituitary tumor cells (10). In the present study, up to 60% degradation of [125I- Tyr']SRIF was observed during incubation with synaptosomal _A *--A membrane fraction P5. The high nonspecific binding of ['sI- Tyr']SRIF encountered in the present investigation (>14%) and that reported for bovine pituitary plasma membrane (18%) (11) could be due to binding of damaged fragments of the ra-

0 Table 4. Distribution of endogenous somatostatin and SRIF receptors in rat brain aC Endogenous 0 -o somatostatin, SRIF specifically Receptor/ ng/mg of bound, fmol/mg of endogenous Region protein. protein* somatostatint Cortex 6.72 ± 1.7 155 ± 17.6 23.1 Thalamus 2.9 + 0.3 89.5 ± 6.0 30.9 Hypothalamus. 36.7 ± 2.0 79 + 4.3 2.15 Striatum 2.34 ± 0.9 61.5.-+- 4.8 26.3 Medulla/pons 3.27 ± 0.3 1.8 + 0.16 0.6 'lo-1Plo- 10-91e -8 10-7 10-6 Cerebellum 0.51 ± 0.08 -0.1 0.02 M Peptide, * The binding capacity is based on-the specific binding obtained with membranes prepared from fraction P5 and is expressed as fmol/mg FIG. 4. Specificity of ['26I-Tr11JSRIF binding to brain mem- of membrane protein present in this fraction. The striking dissocia- branes. The binding of ["I-Tyr1 ISRIF (% of maximum)..is plotted as tion in the binding of SRIF to fractions P5 and P6 observed for cer- a function ofthe concentrations of SREF (o), neurotensin (o), and any ebral cortex (Table 2) was also observed between these fractions pre- of [Metlenkephalin, PH-endorphin, substance P, VIP, TRH, LH-RH, pared from other regions ofthe brain. and bombesin (A). t fmol SRIF bound/ng of endogenous somatostatin. Downloaded by guest on October 2, 2021 3934 Neurobiology: Srikant and Patel Proc. Natl. Acad. Sci. USA 78 (1981)

dioligand to sites that are not specific for SRIF. With [125I- 3. Patel, Y. C. & Reichlin, S. (1978) Endocrinology 102, 523-530. Tyr' ]SRIF, which was more stable than the other two radioli- 4. Reichlin, S., Sapperstein, R., Jackson, I. M. D., Boyd, A. E. & gands tested, we have obtained up to 10% specific binding, Patel, Y. C. (1976) Annu. Rev. Physiol. 38, 389-424. which was significantly higher than that reported 5. Elde, R., Hockfelt, R., Johansson, O., Schultzberg, M., Ef- for pituitary fendic, S. & Luft, R. (1978) Metabolism 77, Suppl. 1, 1151-1160. tumor cells (10). 6. Epelbaum, I., Brazeau, P., Tsang, D., Brawer, J. & Martin, J. Kinetic analysis ofthe binding data indicate that the receptors B. (1977) Brain Res. 126, 309-322. in rat brain synaptosomal membranes contain a single class of 7. Renaud, L. P., Martin, J. B. & Brazeau, P. (1975) Nature (Lon- noninteracting sites. The affinity ofbindingfor SRIF, [Tyr'1]SRIF, don) 225, 233-235. and ['25I-Tyr"]SRIF was identical, which suggests that the re- 8. Kastin, A. J., Coy, D. H., Jacquet, Y., Schally, A. V. & Plotni- koff, N. P. (1978) Metabolism 27, Suppl. 1, 1247-1252. ceptors do not distinguish between the radioiodinated and non- 9. Barker, J. L. (1977) in Peptides in Neurobiology, ed. Gainer, H. radioiodinated ligands (data not shown). The binding parame- (Plenum, New York), pp. 295-343. ters of rat brain SRIF receptors in these studies appear to be 10. Schonbrunn, A. & Tashjian, A., Jr. (1978) J. Biol. Chem. 253, comparable to those reported for pituitary GH4C1 tumor cells 6473-6483. (10) in culture, despite the use of different radioiodinated an- 11. Leitner, J. W., Rifkin, R. M., Maman, A. & Sussman, K. E. alogs of SRIF as'ligands. Such a similarity between the brain (1979) Biochem. Biophys. Res. Commun. 87, 919-927. 12. Pert, C. B. & Snyder, S. H. (1973) Science 179, 1011-1014. SRIF receptors and the cytosol binding protein reported by 13. Taylor, D. P. & Pert, C. B. (1979) Proc. Natl. Acad. Sci. USA 76, Ogawa et al. (31) was not observed here. Somatostatin has been 660-664. shown to act as a partial agonist/antagonist to opiate receptors 14. Moody, T. W., Pert, C. B., Rivier, J. & Brocon, M. (1978) Proc. (32). In GH4C1 cells, a clonal strain of pituitary tumor cells, Natl. Acad. Sci. USA 75, 5372-5376. TRH has been reported to alter,the specific binding of SRIF to 15. Uhl, G..R., Bennett, J. P., Jr. & Snyder, S. H. (1977) Brain Res. its receptors (33). In the present study, neither opiates nor TRH 130, 299-313. 16. Cotman, C. W. & Matthews, D. A. (1971) Biochim. Biophys. or other neuropeptides appeared to influence the binding of Acta. 249, 380-391. SRIF to brain membranes, suggesting that SRIF receptors are 17. Lowry, 0. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J. distinct from,those of other peptides. (1951) J. Biol. 'Chem. 193, 263-275. The ability ofbiologically active analogs to inhibit the specific 18. Karnovski, M. J. (1965)J. Cell Biol. 27, 137 (abstr.). binding of ['"I-Tyr"]SRIF suggests that the action-of SRIF is 19. Vale, W., Rivier, J. & Ling, N. (1978) Metabolism 27, Suppl. 1, probably mediated through binding to its receptors. Additional 1391-1401. 20. Ghashi, S., Sawano, S., Kokubu, R., Gondo, M. & Sakakibara, evidence for this comes from the observation that some biolog- K. (1976) Endocrinol. Jpn. 23, 43438. ically inactive analogs tested were significantly less effective in 21. Marks, N. & Stem, F. (1975) FEBS Lett. 55, 220-224. inhibiting radioligand binding. 22. Marks, N., Stem, F. & Bennuck, M. (1976) Nature (London) 261, The distribution of SRIF receptors in the rat brain differs 511-512. from that ofopiate (12), VIP (13), bombesin (14), carnosine (34), 23. Burbach, J. P. H., Loeber, J. G. & Verhoff, J. (1979) Biochem. f3adrenergic (35), and muscarinic/cholinergic (36) receptors. Biophys. Res. Commun. 86, 1296-1303. 24. Rodbard, D., Hult, D. & Faden, V. B. (1974) in Statistical Anal- Furthermore, the regional distribution of endogenous soma- ysis ofImmunoassays and Immunoradiometric Assays in RIA and tostatin and its receptors in the brain is not parallel. Maximum Related Procedures in Medicine, eds. Robard, D. & Hult, D. (In- receptor density occurs in the cortex, whereas the concentration ternational Atomic Energy Agency, Vienna), pp. 165-192. ofendogenous somatostatin is highest in the hypothalamus. The 25. Kahn, C. R. (1975) in Membrane Biology, ed. Kohn, E. D. ratio of receptor concentration to endogenous somatostatin is (Plenum, New York), Vol. 3, pp. 81-146. high in the cortex, thalamus, and striatum, low in the hypo- 26. Rivier, J., Brown, M. & Vale, W. (1975) Biochem. Biophys. Res. Commun. 65, 746-751. thalamus, and extremely low in the brainstem and cerebellum. 27. Meyers, C., Arimura, A., Gordin, A., Femandez-Durango, R., One possible explanation for the dissociation between endog- Coy, D. H., Schally, A. V., Drouin, J., Beaulieu, M. & Labrie, enous somatostatin and receptor density in the hypothalamus F. (1977) Biochem. Biophys. Res. Commun. 74, 630-636. is the predominant localization ofhypothalamic somatostatin in 28. Iverson, I. L. & Snyder, S. H. (1968) Nature (London) 220, 796- neurosecretory neurons which terminate on the capillaries of 798. the portal vessels (37). However, interaction between SRIF and 29. Snodgrass, S. R., Hedley-Whyte, E. T. & Lorenzo, A. V. (1973) J. Neurochem. 20, 771-782. other peptigeric neurons (e.g., TRH) (38) at an axo-axonic level 30. Gfeller, E., Kuhar, M. J. & Snyder, S. H. (1971) Proc. Natl. in the median eminence and somatostatin-containing terminals Acad. Sci. USA 68, 155-159. shown in various hypothalamic nuclei (5) may account for the 31. Ogawa, N., Thompson, J., Friessen, H. G., Martin, J. B. & Bra- intermediate concentration of SRIF receptors found in the hy- zeau, P. (1977) Biochem. J. 165, 269-277. pothalamus. The low ratio of receptors to endogenous soma- 32. Terenius, L. (1976) Eur. J. Pharmacol. 38, 211-213. tostatin in the brain stem could be due to an upward 33. Schonbrunn, A. & Tashjian, A., Jr. (1980) J. Biol. Chem. 255, projection 190-198. offibers of somatostatin-containing neurons with cell bodies lo- 34. Hirsch, J. D., Grillo, M. & Margolis, F. L. (1978) Brain Res. 158, cated in the pons and medulla. 407-422. We thank Dr. H. H. Zingg for designing the program for; the com- 35. Bylund, D. B. & Snyder, S. H. (1976) Mol. Pharmacol. 12, 568- putation of data and Mrs. M. Correia for secretarial help. Tbis study 580. was supported by grants from the Medical.Research Council of Canada 36. Yamaura, H. I. & Snyder, S. H. (1974) Proc. Natl. Acad. Sci. (MA 6196) and the U.S. Public Health Service (AM 21373). USA 71, 1725-1729. 37. Patel, Y. C., Hoyte, K. & Martin, J. B. (1979) Endocrinology 1. Brazeau, P., Vale, W. & Burgus, R. (1973) Science 179, 77-79. 105, 712-715. 2. Brownstein, M., Arimura, A. & Sato, H. (1975) Endocrinology 38. Hirooka, Y., Hollander, C. S., Suzuki, S., Ferdinand, P. & Juan, 96, 1456-1461. S.-I. (1978) Proc. Natl. Acad. Sci. USA 75, 4509-4513. Downloaded by guest on October 2, 2021