Proc. Nail. Acad. Sci. USA Vol. 89, pp. 10267-10271, November 1992 Biochemistry

Molecular cloning of a somatostatin-28 receptor and comparison of its expression pattern with that of a somatostatin-14 receptor in rat brain (neuropeptide receptor/in situ hybridization/neuron/astrcyte/release-inhibiting factor) WOLFGANG MEYERHOF, IRIS WULFSEN, CHRISTIANE SCHONROCK, SUSANNE FEHR, AND DIETMAR RICHTER Institut fur Zelibiochemie und klinische Neurobiologie, Universitits-Krankenhaus Eppendorf, Universitit Hamburg, Martinistrasse 52, 2000 Hamburg 20, Federal Republic of Germany Communicated by Floyd E. Bloom, June 16, 1992 (receivedfor review March 27, 1992)

ABSTRACT The tetradecapeptide somatotropin-release fashion and can cause either depolarization (9) or hyperpo- inhibiting factor somatostatin-14 regulates the release of pep- larization (10). Less is known about the functional role ofthe tide hormones and also functions as neurotransmitter. The octacosapeptide that shows potencies different from those of octacosapeptide somatostatin-28, the N-terminally extended the tetradecapeptide in controlling hormone secretion from form of somatostatin-14, shows similar biological activities yet target tissues. Also, somatostatin-28 appears to bind to with different potencies. Both peptides most likely function receptors distinct from those that bind somatostatin-14, sug- through distinct receptors. Here we report on the molecular gesting at least two subtypes of somatostatin receptors (11). and functional characterization of a somatostatin-28 receptor Raynor et al. (12) characterized two receptor subtypes that (SSR-28) cloned from a rat brain cDNA library. The nucleotide mediate their activities via GTP-binding proteins, by inhib- sequence contains an open reading frame for a protein of 428 iting adenylate cyclase and by modulating Ca2+ currents; one amino acid residues with a predicted molecular mass of47 kDa. subtype additionally potentiates a delayed rectifier potassium Binding assays using radiolabeled somatostatin-14 and mem- current. cDNAs encoding two structurally different periph- branes from COS cells transfected with the cloned cDNA show eral somatostatin receptors, somatostatin receptors 1 and 2 that this receptor, SSR-28, has a higher binding affinity for (SSTR1 and SSTR2, respectively), have recently been cloned somatostatin-28 (IC5. = 0.24 nM) than for somatostatin-14 from mice and humans (1). They preferentially bind somato- (ICso = 0.89 nM). RNA blot analysis reveals a 4.4-kilobase statin-14, having less affinity for somatostatin-28. The de- mRNA in rat cerebellum and at significantly lower abundance duced amino acid sequence of SSTR1 is nearly identical to in other brain regions. In situ hybridization indicates that the previously identified rat orphan receptor rGHJP se- SSR-28 mRNA is present in the granular and Purkiqje cell quence (97% identity; ref. 13), herein we will refer to this layers ofthe cerebellum and in the large cells ofthe hypoglossal receptor as somatostatin-14 receptor (SSR-14). Here we nuclens of the brain stem. Signals for SSR-28 mRNA do not report the molecular cloning of a cDNA encoding a somato- overlap with those of a previously cloned rat receptor that statin receptor, termed somatostatin-28 receptor (SSR-28), preferentially binds somatostatin-14 (SSR-14). SSR-14 mRNA that preferentially binds the octacosapeptide.* The corre- is found in the medial cerebellar nucleus, horizontal limb ofthe sponding mRNA is found in distinct neural cell populations diagonal band, various hypothalamic nuclei, and in layers IV that are different from those containing the rat tetradecapep- and V of the cortex. In the rat cerebellum, SSR-14 and SSR-28 tide receptor. mRNAs are developmentally regulated; the levels ofthe former are highest around birth and levels of the latter are highest at MATERIALS AND METHODS the adult stage. Materials. Ifnot otherwise stated 3-month-old male Wistar The somatostatin peptide hormone family is composed of at rats were used. Peptides and analogs were obtained from least two functionally active peptides, a tetradecapeptide Bissendorf (Hannover, F.R.G.) or Bachem. RNA blot anal- (somatostatin-14) and an N-terminally elongated form that ysis (13, 14) and in situ hybridization (15) were carried out as consists of 28 amino acid residues (somatostatin-28). Both reported. peptide hormones are derived from a common prohormone Cloning of a cDNA Encoding SSR-28. Preparation of total precursor through tissue-specific proteolytic cleavage (2). cellular RNA, isolation of poly(A)+ RNA, and cDNA syn- Somatostatin-14 was initially isolated from ovine hypothal- thesis were carried out according to standard protocols (14). ami on the basis of its ability to inhibit the release of growth Receptor-encoding cDNA frgments were amplified using hormone from the anterior pituitary (3). It was subsequently primers and conditions as reported (16) and, after subcloning found to inhibit the release of other peptide hormones from into M13mpl8 vectors, were submitted to nucleotide se- the pituitary (prolactin and thyrotropin) (4) and from the quence analysis (17). A rat brain cDNA library in A ZAP pancreas (glucagon and insulin) (5, 6), to control the secretion (Stratagene) was plated at a density of5 x 104 plaque-forming of gut hormones and gastrointestinal motor activity, and to units per 15-cm diameter plate and phage DNA was trans- decrease nutrient absorption from the gut (7). Within the ferred to nylon membranes (Amersham). The filters were central nervous system, somatostatin-14 is assumed to be a hybridized with the amplified cDNA fragments labeled in the neurotransmitter or modulator, based on depolarization ex- presence of [a-32P]dCTP (specific activity, 3000 Ci/mmol; 1 periments of somatostatinergic nerve cells that cause peptide Ci = 37 GBq; Amersham; ref. 18). Membranes were finally release in a Ca2+-dependent fashion (8). At the cellular level, washed for 30 min in 0.2x standard saline citrate at 650C. somatostatin-14 influences neuronal firing rates in a complex Abbreviations: SSR-14 and SSR-28, receptors for somatostatin-14 and -28, respectively; SSTR1 and 2, somatostatin receptors 1 and 2, The publication costs of this article were defrayed in part by page charge respectively, as described in ref. 1. payment. This article must therefore be hereby marked "advertisement" *The sequence reported in this paper has been deposited in the in accordance with 18 U.S.C. §1734 solely to indicate this fact. GenBank data base (accession no. X63574). 10267 Downloaded by guest on October 3, 2021 10268 Biochemistry: Meyerhof et al. Proc. Natl. Acad Sci. USA 89 (1992) Positive phages were plaque-purified and the receptor- still lacks roughly 400 nucleotides as judged from the size of encoding plasmids were excision-rescued from A ZAP clones the corresponding 4.4-kb mRNA. by superinfection with R-408 helper phages (14). Fragments The deduced amino acid sequence displays features typical of the cDNAs, generated by restriction digests or DNase I of guanine nucleotide binding-protein-coupled receptors, deletions (19), were sequenced as reported (13). such as seven hydrophobic segments of 20-27 amino acid Ligand Binding Assays. The 3.5-kilobase-pair EcoRP frag- residues (Fig. 1). Furthermore, there are two potential ment encoding SSR-28 was subcloned into the corresponding N-linked glycosylation sites, Asn-18 and Asn-31, in the site of pcDNAI (Invitrogen); this construct was used to trans- N-terminal extracellular portion of the protein and two con- fect COS-7 cells by the calcium phosphate/glycerol method served cysteine residues (positions 117 and 192) in the first (14). Membrane fractions were prepared from 2.4 x 107 cells, and second extracellular loop (22). The deduced amino acid homogenized in 20 ml of buffer A (50 mM Tris-HCl, pH 7.6/1 sequence also contains several consensus sites for phosphor- mM EDTA/1 mM dithiothreitol/0.1 mM phenylmethylsulfo- ylation (23) by cAMP-dependent protein kinase (Ser-260), nyl fluoride), centrifuged, and taken up in 800 /ul of buffer A protein kinase C (Ser-260), or multifunctional calmodulin- (20). For ligand binding assays, 40 ,ug ofmembrane protein (1.5 dependent protein kinase II (Ser-75, Ser-245, Ser-260, and mg/ml) was incubated with 50 pM [[125I]Tyr1]somatostatin-14 Ser-346). (2200 Ci/mmol; NEN) in 300 ,ul of binding buffer [50 mM Functional Identification. Alignment of the amino acid Hepes, pH 7.5/5mM MgCl2/Trasylol (200 kallikrein inhibition sequence deduced from the rat cDNA clone with other units/mil; Bayer, Leverkusen, F.R.G.)/bacitracin (0.02 ,ug/ receptor protein sequences revealed -50%o identity with ml; Sigma)/bovine serum albumin (10 mg/ml)/phenylmethyl- SSTR1 and SSTR2, the cloned somatostatin receptors from sulfonyl fluoride (0.02 ,ug/ml)] for 1 hr at room temperature in mice and humans (1) and the rat orphan receptor rGHJP (13) the presence or absence ofunlabeled somatostatin-14 or -28 at (Fig. 1) that we refer to as SSR-14. To demonstrate that the concentrations as indicated. cloned cDNA indeed encodes a putative somatostatin recep- tor, ligand-binding assays were performed using membranes ofCOS-7 cells that transiently express the clone. Fig. 2 shows RESULTS AND DISCUSSION the binding of iodinated somatostatin-14. Binding of radioli- cDNA Cloning and Structure of a Rat Somatostatin Recep- gand was competed by using unlabeled somatostatin-14 and tor. cDNA fragments were generated by PCR amplification -28 with inhibitory concentrations for half-maximal response using oligonucleotide primers derived from sequences en- (IC50) values of0.89 nM and 0.34 nM, respectively. A number coding the C-terminal parts oftransmembrane regions III and of other unlabeled peptides (oxytocin, vasopressin, angio- VI (13, 16, 21). Several PCR amplification products proved to tensin II, thyrotropin-releasing hormone, neurotensin, sub- be subtypes of dopamine, adrenergic, adenosine, and tachy- stance P. cholecystokinin, bradykinin, enkephalins, 3-endor- kinin receptors (21). In addition, a cDNA fragment of -4 phine, bombesin, and dynorphin A) did not compete for the kilobases (kb; data not shown) encoding a receptor was binding of radiolabeled somatostatin-14 even at micromolar identified that contained an open reading frame of 1284 concentrations (data not shown). Thus, these results indicate codons specifying a protein of 428 amino acid residues with that the cloned cDNA encodes a somatostatin receptor that a molecular mass of 47.1 kDa (Fig. 1). The cDNA insert shows preferential binding affinity for the octacosapeptide; terminates with a stretch of 105 adenine bases 13 bp down- this receptor is now referred to as SSR-28. In contrast, the stream of the canonical polyadenylylation signal AATAAA; human and mouse somatostatin receptors SSTR1 and thus, the entire mRNA 3'-noncoding region has been cloned. SSTR2, when expressed in cell cultures, exhibit a higher The relatively long 5'-untranslated region of 655 nucleotides affinity for somatostatin-14 (IC50 of 1.5 nM vs. IC50 of4.7 nM

* * TMI - MAAVTYPSSVPTTLDPGNASSAWPLDTSLGNASAGTSLAGLAVWSGILISLVIFVVCGIGSVf L RLHTSSPSVJV MFPNGTAPSPTSSPSSSPGGCGEGVCSRGPGSGAADGMEEPGRNSSQNGTLSEGQGESAILISFIYSVVCjLV aLraN SMVIYVIIL4RYAKMKTrjI MDMADEPLNGSHTWLS IPFDLNGSVVSTNTSNQTEPYYDLT SNAVLTFIWFEdIILj+pj JLBYII~AKMKTIEWJI TM II TM III TM IV 83 YIL GIG AAQNAESYP LM HP TRS iFWAP iMVSAF V. ii VD: YILNLAIADEL|F4kJVTSTLRHMPFGQALL| II FLTV FJDEJVHPIKF P iEYRPTVNKEDNLVSLI #TSZkTLTVF |YILNL9AkEVAMQVH HPFGAIB R VVH IKS PRTt(MITM LIt TM V 178 EVVVFSGV---PRGMSTMQOEPAAAHARTAkII AAL LLCLCYV VRSTTRRVRAPSCQWVQAPACQ ER RRFVV IVVFSRTAANSDGTVACNML EPAQRALVGZVIYTLY ;FTM VGAICLCYVLII R -MVALKAGW-Q----- QRK KIM4 HIMIYAGLRSNQWGRSSUTIN; ESGA4YTGII 4 IL LTIINCJLCFqII13VK --SSGIRVGS---S---Kb( TM VI TM VVII 270 ANHALV{CFYLLNIV1VCPLPEEPAFFGLYFL19 -r~r-YANSCANP(IL IYR7IGRRILLRPSRRVRSQEPGSGPPEKTEEEEDEE M44MFVIC+FYVVQLVbYF-AEQDD---ATVSQLSV4 YANSCANPILYLSDN8KRS-QRIL LSWMDNAAEEPVDYYATALKSRAYSV

IIAVIF FNVS SJSMAISPTPALKGMFDF\& YANSCANPIL lbNB8SDQNVILVKVSGTDDGERSDSKQDKSRLNETT

365 EEERREEEERRMQRGQEMNGRLSQIAQPGPSGQQQRPCTGTAKEQQLLPQEATAGDKASTLSHL rSSR-28 EDFQPENLESGGVFRNGTCASRISTL rSSR-14 ETQRTLLNGDLQTS I hSSTR2

FIG. 1. Comparison of the amino acid sequences of a rat SSR-28 (rSSR-28) with rat SSR-14 (rSSR-14) and human SSTR2 (hSSTR2; ref. 1). Gaps (indicated by hyphens) have been introduced to obtain maximal similarity. Identical amino acids are boxed. The seven membrane-spanning domains, TM I to TM VII, are overlined. Potential N-linked glycosylation sites are marked by asterisks. Triangles point to putative phosphorylation sites. Numbers refer to the amino acid sequence of rSSR-28. Downloaded by guest on October 3, 2021 Biochemistry: Meyerhof et aL Proc. Nat!. Acad. Sci. USA 89 (1992) 10269

I d 7c1 l4d 28cd) '1' A

.) _

.cD X 13 e .4 k-b WI.,

FIG. 3. RNA blot analysis of SSR-14 and SSR-28 mRNAs. -11 -10 -9 -8 -7 -6 Poly(A)+ RNA (10 jug) isolated from cerebelli at the indicated time log[somatostatin] after birth [d, day(s)] was electrophoresed, blotted, and hybridized with a labeled SSR-14 (A) or SSR-28 (B) cDNA. The blot was also FIG. 2. Binding of iodinated somatostatin-14 in the presence of probed with a labeled cDNA encoding the immunoglobulin heavy unlabeled somatostatin-14 (solid circles) or somatostatin-28 (open chain binding protein, indicating that each lane contained compara- circles) to membrane fractions derived from COS-7 cells transfected ble amounts ofRNA (data not shown). Note SSR-28 mRNA was not with cDNA encoding SSR-28. detected in kidney, liver, spleen, testis, ovary, lung, heart, stomach, duodenum, andjejunum ofadult rats nor in A7r5 smooth muscle and for somatostatin-28; ref. 1), as does the rat orphan receptor GH3 anterior pituitary cells (data not shown). rGHJP (13), termed here SSR-14 (data not shown). The latter is highly homologous to the somatostatin receptor SSTR1 putative acceptor site for palmitoylation (21), a cysteine from mice and humans (1) containing only 10 deviations in a residue just after transmembrane segment VII (Fig. 1). The sequence of 391 amino acid residues. Based on the ligand absence of this cysteine residue has been identified (21) in affinities, SSTR1, SSR-14, and SSTR2, the structurally less- peptide and biogenic amine receptors, in receptors linked to related somatostatin receptor (1), can be grouped into one inositol phosphate metabolism, and in receptors activating or subfamily and SSR-28 can be placed in a second subfamily. inhibiting adenylate cyclase. These receptors may possess However, based on the divergence in their sequences, three other determinants that functionally substitute for the mod- distinct subfamilies can be proposed (SSTR1/SSR-14, ification by palmitoylation. In this context it should be noted SSTR2, and SSR-28). that SSR-28 shows the peculiarity of a contiguous stretch of The somatostatin receptors are unique for members of the 18 intiacellularly located charged residues (13 glutamic acid, guanine nucleotide binding-protein-coupled receptor family 1 aspartic acid, and 4 arginine residues at positions 358-375, in that they contain a glutamic acid residue in the second Fig. 1) near its C terminus. A search of the GenBank data transmembrane domain (Fig. 1, position 92). Interestingly, base (March 1992) with this sequence revealed that a stretch SSR-28 is the only somatostatin receptor that is missing a of glutamic acid residues is a common motif among diverse

4

.T .: : VL 4m.:.-.

i|. _,p

<_ - f ' ' "; _ , ; s .j ^ _ . D < t>,.. .' X

FIG. 4. In situ hybridization ofvarious brain sections using 35S-labeled antisense SSR-28 RNA as a hybridization probe (15). (Inset) Overview of a coronal rat brain section (for the rostro-caudal level of the coronal slices see plate 68 of ref. 28); arrows point to the anatomical localization of the hybridization signals of sections shown in A-D. h, hypoglossal nucleus; p, Purkinje cell layer; w, white matter. (A and C) Purkinje cell layer. (C) Dark region, granular layer; light region, molecular and Purkinje cell layer; arrows point to labeled cells at the interface between molecular and granular layers. (B and D) Hypoglossal nucleus. (D) Arrows point to large neuronal cells. (A and B) Dark-field images. (C and D) Bright-field images. No hybridization signals were observed using the sense RNA as a probe. (Bars: A and B, 200 ,um; C and D, 20 ,um.)

Downloaded by guest on October 3, 2021 *} ter K t .f z -; , 10270 Biochemistry: Meyerhof et alPProc. Nad. Acad Sci. USA 89 (1992)

Is A~~~~~~~~~~~~~~I 0FA- 44 44 .

W / 4 <.~~~~~~*I I

A._ a d ~ ~ / 4, Aw, 0 0 i 44 i.A T.D

g...... S 4 4a

*.,. j !~~~~~~~~~4 .' o.' : + i.

. !, <..1

Gs 4 aft _*

*+ _ H

FIG. 5. In situ hybridization ofvarious coronal rat brain sections using 35S-labeled antisense SSR-14 RNA as a hybridization probe (15). The rat SSR-14 cDNA (13) was subcloned into the EcoRI site of pGEM-3Z. 35S-labeled antisense RNA probes were obtained using T7 RNA polymerase after linearization of the plasmid at the EcoRV site within the cDNA insert. Transcripts in the sense orientation used as a negative control were synthesized with SP6 RNA polymerase after linearization ofthe plasmid at the Sst I site within the cDNA. No cross-hybridization was observed between the two somatostatin receptor-encoding cDNAs. Insets in A and B and in E and F show overviews ofbrain sections (for the rostro-caudal level of the coronal slices see plates 34 and 22 of ref. 28, respectively). Arrows point to the positions of hybridization signals shown in A-D, E, and F. c, Cortex; m, mammillary nucleus; s, suprachiasmatic nucleus; d, horizontal limb ofthe diagonal band; oc, optic chiasm. (A and C) Mammillary nucleus. (C) Hybridization signals are present in large neuronal cells (arrows), not in the dark cells (most likely glial cells, arrow heads). (B and D) Cortex. (E and G) Horizontal limb ofthe diagonal band. (F and H) Suprachiasmatic nucleus (dark arrows point to small cells, possibly astrocytes). No hybridization signals were detectable using the sense RNA as a probe. A, B, E, and F are dark-field images. C, D, G, and H are bright-field images. (Bars, A, B, E, and F, 200 um; C, D, G, and H, 20 Stm.) proteins such as the gag polyprotein of simian sarcoma virus, for somatostatin-14 and -28 in different cell populations ofthe the ryanodine receptor, and the a- and P-globulin storage rat central nervous system were subject to developmental proteins. It has been suggested that this motifmay participate regulation (10, 25-27). RNA blot and in situ hybridization in protein-protein interactions (24). data presented here are in line with those reports. In adult Distribution of mRNAs Encoing SSR-14 and SSR-28 In the rats, a 4.4-kb SSR-28 mRNA is present in the cerebellum and Rat Brain. Ligand binding studies showed distinct receptors at lower levels in cortex, hippocampus, thalamus, hypophy- Downloaded by guest on October 3, 2021 Biochemistry: Meyerhof et al. Proc. Natl. Acad. Sci. USA 89 (1992) 10271 sis, and hypothalamus (data not shown). SSR-14-encoding We thank Hannelore and Gunter Ellinghausen for technical help, mRNA is mainly detectable in the hypothalamus and cortex Drs. Klaus Hermann and Mark Darlison for critically reading the (13). Fig. 3 shows that in the cerebellum the two receptors are manuscript, and the Deutsche Forschungsgemeinschaft for financial differentially expressed. SSR-14 mRNA levels decrease support (SFB 232/B4 to W.M. and D.R.). The data presented here gradually during postnatal development, whereas SSR-28 are part of a thesis by I.W. mRNA is undetectable at birth and reaches optimal levels at 1. Yamada, Y., Post, S. R., Wang, K., Tager, H. S., Bell, G. I. the adult stage. & Seino, S. (1992) Proc. NatI. Acad. Sci. USA 89, 251-255. At the cellular level, we found no overlapping expression 2. Goodman, R. H., Montminy, M. R., Low, M. J. & Habener, of the two somatostatin receptors, SSR-14 and -28, in the J. F. (1987) in Molecular Cloning of Hormone Genes, ed. brain regions analyzed so far (Figs. 4 and 5). In situ hybrid- Habener, J. F. (Humana, Clifton, NJ), pp. 93-119. ization data indicate the presence of specific signals for the 3. Brazeau, P., Vale, W., Burgus, R., Ling, N., Butcher, M., SSR-28-encoding mRNA in the granular layer and in the Rivier, J. & Guillemin, R. (1973) Science 179, 77-79. cell of the cerebellum 4 A 4. Vale, W., Rivier, C., Brazeau, P. & Guillemin, R. (1974) Purkinje layers (Fig. and C), Endocrinology 95, 968-977. although only very small amounts of somatostatin has been 5. Vale, W., Rivier, C. & Brown, M. (1977) Annu. Rev. Physiol. found in the cerebellum (29). However, at least the afferents 39, 473-527. of the mossy fibers in rats and mice have been reported to 6. Reichlin, S. (1980) in Peptides: Integrators ofCell and Tissue contain somatostatin-like immunoreactivity (29). Thus, our Functions, ed. Bloom, F. E. (Raven, New York), pp. 1-20. data meet the requirement that somatostatin receptors are 7. Reichlin, S. (1983) in Brain Peptides, eds. Krieger, D. T., found on those cells that receive synaptic input from somato- Brownstein, M. J. & Martin, J. B. (Wiley, New York), pp. statinergic fibers. As shown in Fig. 4 B and D, magnocellular 711-752. neurons of the hypoglossal nucleus contain hybridization 8. Iversen, L. L., Iversen, S. D., Bloom, F., Dougles, C., Brown, for in this M. & Vale, W. (1978) Nature (London) 273, 161-163. signals SSR-28-encoding mRNA; region binding 9. Dodd, J. & Kelly, J. (1978) Nature (London) 273, 674-677. sites for somatostatin-28 but not for somatostatin-14 have 10. Pittman, Q. & Siggins, G. (1981) Brain Res. 121, 402-408. been reported (25). 11. Srikant, C. B. & Patel, Y. C. (1981) Nature (London) 294, The SSR-14 mRNA is present in different structures ofthe 259-260. brain that overlap with the locations of the binding sites for 12. Raynor, K., Wang, H.-L., Dichter, M. & Reisine, T. (1991) the tetradecapeptide. However, the intensity oflabeling with Mol. Pharmacol. 40, 248-253. ligand or with antisense receptor RNAs varies considerably 13. Meyerhof, W., Paust, H.-J., Schonrock, C. & Richter, D. in some brain sections. For example, the mammillary nucleus (1991) DNA Cell Biol. 10, 689-694. (Fig. 5 A and C) and the horizontal limb of the diagonal band 14. Maniatis, T., Fritsch, E. F. & Sambrook, J. (1982) Molecular E and show the most intense Cloning:A Laboratory Manual (Cold Spring Harbor Lab., Cold (Broca) (Fig. 5 G) hybridization Spring Harbor, NY). signals, whereas little or no ligand binding has been reported 15. Fehr, S., Ivell, R., Koll, R., Schams, D., Fields, M. & Richter, for these regions (25). Moderate hybridization signals are D. (1987) FEBS Lett. 210, 45-50. found in layers IV and V of the cortex (Fig. 5 B and D) and 16. Libert, F., Parmentier, M., Lefort, A., Dinsart, C., Van Sande, the retrosplenial cortex, for which intense somatostatin-14 J., Maenhaut, C., Simons, M.-J., Dumont, J. E. & Vassart, G. binding has been observed (25). Weak SSR-14 signals have (1989) Science 244, 569-572. been detected in the cerebellum; however, in contrast to the 17. Sanger, F., Nicklen, S. & Coulson, A. R. (1977) Proc. Natl. cerebellar distribution of SSR-28 mRNA, SSR-14 is present Acad. Sci. USA 74, 5463-5467. in the medial cerebellar nucleus. Other brain regions such as 18. Feinberg, A. P. & Vogelstein, B. (1983) Anal. Biochem. 132, the dentate 6-13. hippocampus, gyrus, septum, colliculus, 19. Lin, H.-C., Lei, S.-P. & Wilcox, G. (1985) Anal. Biochem. 147, amygdala, and caudate-putamen show moderate to high 114-119. levels of somatostatin binding (25) but do not contain detect- 20. Kuno, T., Andresen, J. W., Kamisaki, Y., Waldman, S. A., able in situ hybridization signals for SSR-14 or SSR-28 Chang, L. Y., Saheki, S., Leitman, D. C., Nakane, M. & mRNA. The lack of SSR-14 and SSR-28 mRNAs may be Murad, F. (1986) J. Biol. Chem. 261, 5817-5823. explained by the existence of additional somatostatin recep- 21. Meyerhof, W., Darlison, M. G. & Richter, D. (1992) in New tors that could account for the reported somatostatin-binding Comprehensive Biochemistry, Neurotransmitter Receptors, sites; the low levels ofsomatostatin receptormRNA detected ed. Hucho, F. (Elsevier, Amsterdam), in press. in the hippocampus by RNA blot analysis using a longer 22. Dohlmann, H. G., Thorner, J., Caron, M. G. & Lefkowitz, be due to to as yet R. J. (1991) Annu. Rev. Biochem. 60, 653-688. probe may cross-hybridization another, 23. Kemp, B. & Pearson, R. B. (1990) Trends Biochem. Sci. 15, unidentified, somatostatin receptor subtype. 342-346. Higher magnification shows that not all cells present in the 24. Sugimoto, Y., Yatsunami, K., Tsujimoto, M., Khorana, G. & regions labeled by the SSR-14 probe express detectable Ichikawa, A. (1991) Proc. Natl. Acad. Sci. USA 88,3116-3119. amounts ofSSR-14 mRNA. In most cases, and as exemplified 25. Leroux, P., Quirion, R. & Pelletier, G. (1985) Brain Res. 347, in layers IV and V of the cortex, cells with relatively large 74-84. weakly counterstained nuclei, most likely representing neu- 26. Heiman, M. L., Murphy, W. A. & Coy, D. H. (1987) Neuroen- rons, are decorated with silver grains (Fig. SD). SSR-14- docrinology 45, 429-436. specific signals are also present in the medial preoptic area 27. Gonzalez, B. J., Leroux, P., Laquerriere, A., Coy, D. H., and the nucleus 5 F and In the Bodenant, C. & Vaudry, H. (1988) Dev. Brain Res. 40, 154-157. suprachiasmatic (Fig. H). 28. Paxinos, G. & Watson, C. (1986) The Rat Brain in Stereotaxic suprachiasmatic nucleus, small cells with compact darkly Coordinates, (Academic, New York), 2nd Ed. counterstained nuclei, which are presumed to be astrocytes, 29. Schulman, J. A. (1983) in Chemical Neuroanatomy, ed. Em- are labeled (Fig. SH). These data are consistent with previous son, P. C. (Raven, New York), pp. 209-228. ligand binding studies (30), reporting that somatostatin-14 30. Krisch, B., Buchholz, C. & Mentlein, R. (1991) Cell Tissue Res. binds to astrocytes of the suprachiasmatic nucleus. 263, 253-263. Downloaded by guest on October 3, 2021