Brain Research, 126 (1977) 309-323

~(') Elsevier/North-Holland Biomedical Press, Amsterdam - Printed in The Netherlands

309

S U B C E L L U L A R D I S T R I B U T I O N OF R A D I O I M M U N O A S S A Y A B L E SOMATOSTATIN IN RAT BRAIN

JACQUES EPELBAUM*, PAUL BRAZEAU**, DAVID TSANG, JAMES BRAWER** and JOSEPH B . MARTIN*** Departments of Medicine and Psychiatry, Montreal General Hospital, and Departments Of Anatomy attd Obstetrics' and Gynecology, McGill University, Montreal (Canada)

(Accepted September 3rd, 1976)

SUMMARY A specific, sensitive and reproducible radioimmunoassay is described for measurement of somatostatin in brain tissue. The sensitivity of the assay is 10 pg/tube and recovery of synthetic somatostatin added to brain homogenates was 95.8 ~_ 6.2 ~/0. Dilution of tissue extracts from various brain regions showed parallelism in standard curves with labelled somatostatin. Somatostatin is shown to be widespread in the central nervous system with highest concentrations in hypothalamus, particularly the median eminence. Subcellular preparations of medial basal hypothalamus, preoptic area and amygdala indicate that over 70 ~ of somatostatin immunoreactivity is localized to the synaptosome fraction. Recovery of activity in the P2 pellet prior to separation on sucrose gradient is approximately 100~,~. It is hypothesized that somatostatin, in addition to being released into blood vessels of the median eminence, may also be liberated from nerve terminals in other brain regions.

INTRODUCTION Recently, somatostatin (somatotropin-release inhibiting factor: SRIF) was isolated from ovine hypothalamic extracts and shown to have activity in inhibiting growth hormone (GH) release both "in vivo"4, 5 and "in vitro"4,5, 46. The structure of somatostatin was characterized as: H-Ala-Gly-Cys-Lys-Asn-Phe-Phe-Trp-Lys-Thr-Phe-Thr-Ser-Cys-OH L 1 * Killam Fellow, Department of Neurology and Neurosurgery, McGill University. Scholar, Medical Research Council of Canada. *** Associate, Medical Research Council of Canada.

**

310 The peptide has also been shown to be active in inhibiting the release ot thyrotropinTM 48, prolactin4S, glucagon16-18,2(;, insulin16-1s,26 and gastrin 7 in a variety of physiologic conditions in many species, including man. Somatos'atin is not only localized to the hypothalamus; substantial amounts of both immunoreactive8,3a,35 and bioactivea7 SRIF are detectable in extrahypothalamic regions of the central nervous system (CNS) and in the upper gastrointestinal tract '',14. Cytoimmunologic studies of brain have confirmed that SRIF is present in axons and nerve terminals in the median eminencer~,lz,22,z:~,26,a~. Furthermore, some workers have described positive staining in magnocellular perikarya of the paraventricular and supraoptic nuclei and in the preoptic and anterior hypothalamic regions v3,'a. However, attempts to localize SRI F in other regions of the brain have not been succesful using this technique. A clearer understanding of the subcellular localization of somatostatin within brain regions would prove useful in the further investigation of the role of the peptide in brain function. In the present investigations, a sensitive, specific and reproducible radioimmunoassay for SRIF was used to monitor the activity of the peptide during preparation of the subcellular components using the technique described by Whittaker '5°,5~ to disrupt nerve cells. This method, which has been used to study thyrotropin-releasing hormone (TRH) ~ and luteinizing hormone-releasing hormone (LHRH) ~,av,~'2 distribution in hypothalamus, has shown that the majority of these peptides reside within synaptosomes or nerve terminal fractions of hypothalamus. The present results indicate that SRIF activity in the hypothalamus and in various other brain regions is also primarily localized in nerve terminals*. MATERIALS AND METHODS

(1) Antibody production Synthetic SRIF was combined with methylated BSA as described by Weir4% One mg of methylated BSA was dissolved in 0.8 ml distilled water and mixed with 5-7 mg somatostatin (AY-24,910 obtained from Ayerst Laboratories, Montreal, courtesy of Dr. M. G6tz and Dr. Hans Immer) in 0.7 ml of saline. The solution was emulsified in equal volumes of complete Freund's adjuvant and injected into young rabbits in multiple intradermal sites. Booster injections were given intramuscularly every 2-3 weeks after the initial priming injection. Twelve to 14 days after each immunization, the animal was bled and serum tested for immunoreactivity.

(2) Preparationof labelled antigen Tyrosine-l-SRIF (Ay-25-511), provided by Ayerst Laboratories, Montreal, was iodinated with lz5I using the lactoperoxidase technique of Thorell and Johansson4~. Ten/~1(0.5 mg/ml) of tyrl-SRIF diluted in 0.002 N NH4 acetate buffer (pH 4.6) was added to 25 pl 0.4N Na acetate buffer at pH 5.6, followed by 2 mCi 1251 (Amersham). * An abstract of this work was publishedin the Progr. of the 58th Annualmeetingof the Endocrine Society, p. 137, 1976.

311

Twenty #g lact0per0xidase diluted in 0.002 N NH4 acetate buffer was then added followed by two aliquots of 10/~l, 1:15,000 H2Oz. One hundred #1 Na Azide (200 /~g/ml), diluted in 0.002 N N H4 acetate buffer was added 30 sec after the first addition of HzO2. Fifty/A 10 To human serum albumin, diluted in buffer, was added before the separation procedure. The reaction mixture was kept chilled on ice after iodination until purification on a CMC-52 column (0.5 cm × 15 cm). The labelled antigen obtained by this technique was found to bind in excess of 80~ with a 1:200 dilution of antiserum to SRIF.

(3) Tracer purification The iodination products were eluted on a CMC-52 column with 0.002 Nfollowed by 0.2 N NH4 acetate buffer, (pH 4.6). Labelled tyrl-SRlF was stored at 4 °C in 0.5 ~ Trasylol (Boehringer Laboratories, Montreal). Individual peaks of radioactivity were tested for immunoreactivity, at antibody dilutions of l :250, 1:1000 and 1:5000.

(4) Assay procedure The following reagents were added in duplicate to disposable 12 mm × 75 mm polystyrene tubes and kept in iced water throughout the assay: 100/~1 of phosphosaline buffer, pH 7.2, containing 0.05 M EDTA, 0.1 ~ sodium azide and 0.1 ~ human serum albumin; 100 #l of diluted standards (range 4-10,000 pg/tube) dissolved in phosphosaline buffer, pH 7.2; 50/~l iodinated tyrl-SRIF, freshly diluted in phosphosaline buffer; 100/~1 of rabbit anti-SRlF serum (Rabbit BSA 4) in a final dilution of 1:5000; and 100 ttl of a solution consisting of one part of 0.25 ~i normal rabbit serum and one part of 1/10 diluted goat anti-rabbit gamma-globulin. The tubes were incubated for 3 days at 4 °C; on the third day 1.0 ml of 95~ ethanol was added to enhance separation of bound and free label. After 30 rain centrifugation at 3000 rev/ rain, the supernatant was discarded and the precipitate counted.

(5) Specificity of SRIF antibody The specificity of the antiserum was evaluated by determining its cross-reactivity with 19 peptides, amines and proteins in 5 concentrations ranging from 200 pg/ml to 2 #g/ml. Substances tested included TRH, LHRH, melanocyte-inhibiting factor (MIF), glucagon, insulin, rat growth hormone (rGH), rat plolactin (rPRL), rat thyroid stimulating hormone (rTSH), rat [uteinizing hormone (rLH), vasopressin, oxytocin, neurotensin, substance P, bacterotoxin, norepinephrine, dopamine, serotonin, chicken broth and methylated BSA. Several SRIF analogs, provided by Ayerst Laboratories, were also tested for cross-reactivity to partially define the immunogenic site of the SRIF molecule.

(6) Brain extraction for SRIF Brain tissues from 15 to 20 rapidly decapitated male Sprague-Dawley rats (200-250 g) were carefully dissected and pooled in 0.2 N acetic acid, 0.5 ml/fragment. All animals were killed between 9:00 and 10:00 a.m A fragment of the medial basal hypothalamus (MBH) containing the median eminence (ME), the medial and poste-

312 rior part of the arcuate nucleus and a portion of the ventromedial nucleus (VM N) was first dissected from the brain. A rostral fragment (APO) containing structures located superior to the optic chiasm, including the suprachiasmatic recess of the lllrd ventricle, the lamina terminalis, part of the suprachiasmatic nuclei and the prcoptic area was then removed. The amygdaloid complex was dissected by removal of a cube of tissue; the anterior and posterior extent of the tissue taken corresponded to the posterior optic chiasm and the mammillary bodies, respectively. This block of tissue was divided into the corticomedial (CMA) and the basolateral (BLA) part. Fragments of parietal cortex (CX) of comparable size were also removed as were pineal glands and portions of cervical spinal cord. All tissues were homogenized at 600-800 rev/min using a glass "Potter" homogenizer with clearance of 0.15 mm. The homogenate was centrifuged at 3000 × g for 30 min, the supernatant decanted and frozen until assayed. Brain tissues of each origin were diluted at concentrations ranging from 1/3 to 1/1250 fragment equivalents depending on the origin of tissue and concentration of SRIF in each tissue. Fragments of MBH, CMA, BLA, APO and CX were compared (in standard curves) for parallelism with SRIF. For preparation of subcellular tissue, brain fragments from 10 rats were pooled in 0.32 M sucrose (10 o/w/v) and carefully homogenized on ice with a Teflon pestle in a glass homogenizer (with clearance of 0.15 mm). The subsequent fractionation procedure was carried out with a slight modification of that originally described by Whittaker, the exact protocol of which is reported elsewhere 37. A portion of the synaptosome preparation was fixed in glutaraldehyde, embedded in plastic and examined by electron microscopy. The cytoplasmic enzyme marker lactate dehydrogenase (LDH) was assayed according to the method of Johnson and Whittaker 2'~. The protein concentration of each fraction was determined by the method of Lowry'~s. After pipetting aliquots of each fraction for protein and enzyme titration, the samples were extracted with 0.2 N acetic acid and frozen. On the day of radioimmunoassay, each ~ample was centrifuged at 1000 x g for 10 rain, the pellet discarded and the supernatant neutralized and radioimmunoassayed in duplicate for SRIF. RESULTS

(1) Iodination of antigen The elution pattern oftyrl-SRIF (on CMC 52) is shown in Fig. 1. These results were reproducible on 10 consecutive weekly iodinations. From the elution pattern and the corresponding immunoreactivity and optical density recordings, it is evident that the first peak corresponds to free iodine and the second and third to human serum albumin and lactoperoxidase. The principle immunoreactivity was found in two large peaks that presumably correspond to monoiodo- and diiodo-tyrl-SRIF, respectively. The largest peak was found to have the greatest immunoreactivity.

(2) Specificity of antibody for SRIF None of the hypothalamic peptides tested showed cross-reactivity with the anti-

313 .80. .60, .40 .20_

lo-'. .08~

zq E 06.

o

04_ o

.02_ 00

A_

600t 1

-600

4001

_400

200L

~

,oo

~

200 %0

[:) 8O x -- 60 u

8o

I

40

I

[

40

20

I

Id/V,~'/

""

\"~~

l

0 9 i

l

0002N 0.2N

17

25

33

41

49

57

63

71

rnl

NH4 Acetate: pH 4 6

Fig. I. Elution pattern of tyrl-SR1F on carboxymethyl-cellulose (CMC 52) column. Upper part shows optical density (280 nm) readings and lower part radioactivity (solid line) and immunoreactivity (dotted line). SR1F serum at concentrations of 100 WOO times greater than that o f SRIF (Fig. 2). There was also no cross-reactivity with the biogenic amines, or with pituitary or pancreatic hormones. Interestingly, no cross-reactivity was evident with methylated bovine serum albumin, although this was the coupling protein used to induce immunogenicity to SRIF.

314 100 90TRF LRF MIF DOPAMINE 5-HT NE OLUCAGON INSULIN OXYTOCIN VASOPRESSIN SUBSTANCE P NEUROTENSIN BACTEROTOXlN rC:,H rLH r PRL rTSH CHICKEN SOUP CH3-BSA

80-

70-

60-

50-

4030ABI: 1:5000

20-

~ .

BO/T; 44.5 ~o

10 0

i

004

v

.01

i

,

,

.04

.1

.4

t

1.0

i

4.0

1010

i

40.0

I

100.0

1.000.0

~Junits or NG/0.5 ML

Fig. 2. Standard curve for somatostatin (SRIF). None of the substances tested showed cross-reactivity

with synthetic SRIF. For abbreviations see text. The results with SRIF analogs indicate that both the C O O H and NH2 radicals are needed for full immunogenicity, the NH~ being the most crucial (Table I). This point is emphasized by the fact that the A ~ B/Bo obtained with des-(NH~)I-SRIF was 9.8, whereas, that of the des-(COOH)14 SRIF was 38.1. Furthermore, it is apparTABLE I Immunoreactivity of SR1F analogs Analog

Displacement of label* (zX % B/Bo)

SRIF Tyrl-SRIF (AY-25,511) Des(NH~)I-SRIF (AY-25,081) I)es(COOH)14-SRIF (AY-25,088) Des-Ala-Gly-N-acetyl-SRIF

53.8 66.1 9.8 38.1 60.8

* Calculated as difference between the ~ binding obtained with 10 pg standard SRIF or analog and the binding obtained with 10 ng standard SRIF or analog, respectively.

315

l .o\\

o~ooSO

\

SOMATOSTATI N

CMAX~,.

40

30

20

10

0"

.004

i

w

.ol

.o4

IP

.1

.4

1.o

a.o

1o.o

NG/O.S ML 1:1280

!:320

1:80

1:20

1:6

1:3

FRAGMENT equiv./0.5 ML

Fig. 3. Displacement of labelled tyrl-SRIF from antibody by dilutions of extracts from various brain regions HPT, bypothalamus; CMA, corticomedial amygdala; BLA, basolateral amygdala; APO, anterior hypotbalamic-preoptic area; CX, cortex.

ent that the first two amino acid residues (Ala-Gly) are not required for immunoreactivity provided the NH2 terminal is present. The displacement of bound antigen by acidic extracts of various brain regions was parallel to that of synthetic SRIF (Fig. 3). The displacement curves indicate that largest amounts of immunoreactive SRIF are present in the hypothalamus, followed by CMA, BLA, APO and CX when expressed in terms of fragment equivalents.

(3) Sensitivity, recovery and reproducibility The sensitivity of the assay ranged from 10 to 4000 pg/tube. In a series of 12 determinations, recovery of known amounts of synthetic SRIF added to brain extracts prior to homogenization was 95.8 Jz 6.2 ~. Intra-assay and inter-assay variabilities were ~_ 3 °/o and ~ 4 ~, respectively.

316 TABLE II

Regional distribution of S R I F in brain Fragments of brain were homogenized in 0.2 N acetic acid, centrifuged at 3000 ~.. g for 30 rain, the pellet discarded and the supernatant neutralized and immunoassayed in duplicate. Abbreviations: APO, preoptic area; CMA, corticomedial amygdala; BLA, basolateral amygdala; CX, cortex.

Brain region

No. of experiments

Wet weight (rag)

pg SRIF/ fragment

pg SRIF, mg wet weight

Total MBH M.E.** Remaining hypothalamus*** APO CMA BLA CX Spinal cord Pineal gland

7 7

16.7 ~ 0.8* 1.4 q 0.2

22239 :i: 2160 17656 1036

1397 208 15470 ! 2421

7 7 7 7 7 7 7

13.9 13.3 30.5 34.1 42.0 34.0 2.3

3837 4414 8212 12214 3750 3696 250

~ 1.4 % 0.5 ~. 1.2 t:: 1.9 J 4.4 _~: 3.2 -[:: 0.3

j 604 -~ 387 i_ 574 t 1472 :~: 404 ~:: 508

266 338 :: 272 351 93 : 114

45 37 23 23 16 18

* Mean 5: S.E. ** Median eminence (M.E.). *** Hypothalamus excluding medial basal hypothalamus (MBH).

(4) Regional distribution of SRIF (Table I1) More than two-thirds of hypothalamic SRIF is found in the MBH fragment, which contains principally the ME, arcuate nucleus and part of the VMN (Table II). The majority of this activity is localized in the ME. The remainder of the peptide is distributed evenly in the anterior fragment of hypothalamus and preoptic area (APO) and in the rest of the hypothalamus. Amygdaloid SRIF amounted to more than onethird of the whole hypothalamic content and was equally distributed between the TABLE I11

Subcellular distribution of SR1F in different brain regions involved in the control of GH secretion Brain fragments from 10 rats were homogenized on ice in 0.32 M sucrose ( 1 0 ~ v/w) and fractionated by the method of Whittaker ~°. Numbers in parentheses indicate separate experiments. At each step of the fractionation, aliquots were extracted, as described in the text, for radioimmunoassay.

H S P $2 P2

Protein (t~g/rng eq.)

LDH (l~M NADH~/ h/eq.)

SRIF (pg/eq.) MBH

APO

Amygdala

88.53+ (7) 65.62~ 18.44± 22.53~ 39.28:t-

l15.00:L (4) 97.35 i 12.70± 50.15± 40.35 ~:

15331 ± 1 5 7 4 (7) 14815 ~:2013 2588 ± 883 1 3 6 2 i 698 15401 ± 1 7 8 7

5 2 7 6 ~ 724 (6) 4528 ± 452 1 3 7 0 i 474 499:t: 112 6200:~ 1138

12076 ~ 2607 (5) 10943 ~ 807 2324 ~_ 680 6 8 6 ± 194 9755 ± 1812

* M e a n ± S.E.

6.92* 10.44 0.85 4.75 5.75

10.00 10.55 1.52 7.61 9.70

317 LACTATE DEHYDROGE NASE A v v

e~

MEDIOBASAL HYPOTHALAMUS

PREOPTIC AREA

AMYGDALOtD COMPLEX

1oo

c~

80

u

6O

tr

v

SOMATOSTATI N -RIA •

40-

±

a~ 20

[~

: before A

L~:A

~=B

L~=C

Fig. 4. Subcellular distribution of P2 precipitate separated on sucrose gradient. The predominant lactate dehydrogenase activity and immunoassayable somatostatin activity are in band B, which corresponds to the synaptosome fraction.

B L A and the C M A subdivisions o f the amygdala. There was a small but consistently detectable a m o u n t o f SRIF in the parietal cortex as well as in the spinal cord; no activity was detected in single determinations o f pineal glands.

Fig. 5. Electron photomicrograph of material separated in band B on sucrose gradient to show a synaptosome containing vesicles and mitochondria. × 58,000.

318

(5) Subeellular distribution of SRIF (Table 111) The majority of the hormone present in crude acidic extracts was recovered in acidic extracts of sucrose homogenates of each brain region (Table 1II). In the MBH, APO and amygdala, approximately 90 ~ of the SRIF found in the sucrose homogenate was recovered in $1 (total extract after removal of the nuclear pellet and larger membrane debris). When $1 was further fractionated, the bulk of the hormone was recovered in the crude mitochondrial fraction, P2, which contains synaptosomes, mitochondria and small membrane debris. More than 70 700 of the P2 synaptosomes separated on sucrose gradient were present in band B, as indicated by LDH distribution (Fig. 4) and by electron microscopy (Fig. 5). Eighty per cent of the peptide was recovered in the B band. DISCUSSION Histidine or tyrosine residues are necessary to iodinate a peptide molecule; as native SRIF is lacking these two amino acids in its primary structure, it was necessary to use tyrl-SR1F as substrate for the iodination. This procedure has also been used by other workers 1,35. That this substitution is acceptable was shown recently by studies in which tyrl-SRIF was found to bind in an equivalent manner to SRIF in a specific radioreceptor assay3L The technique of generating antibody to cyclized SRIF by using the peptide coupled to methylated bovine serum albumin in aqueous solution has proven very effective; of 8 rabbits immunized using this method, 3 have shown moderate immunoreactivity (15-25 ~o binding at 1:2500 dilution), and one has developed an antiserum with higher binding affinity (30-45 ~,~binding at a dilution of 1:5000). This latter antiserum, which is the one presently used in our laboratory, has been demonstrated to be effective in blocking stress-induced inhibition of growth hormone secretion in the rat 44. The specificity of the antiserum has been shown by the failure of various substances to displace the binding of P25-tyrl-SRIF to the antibody. Although it is impossible to rule out the occurrence of a reaction between the antiserum and some as yet unidentified substance present in the brain, the parallelism obtained with SRIF standard curves and acidic brain extracts permits the conclusion that the antiserum to SRIF is specific. Furthermore. the recovery of somatostatin added during the homogenization of a cortex extract approximated 100 ~. From these results, it appears that the extraction procedure used for brain samples is adequate and that degradative enzymatic activities known to be present in brain homogenates are inactivated by the acidic extraction2°,21. However, we cannot exclude the possibility that internal cleavage of the SRIF molecule occurs and that such fragments might cross-react with the antibody in our system. Our findings concerning the immunogenicity of the SRIF molecule are quite preliminary; nevertheless, it seems that the antiserum reacts with the ring portion of SRIF, since des (Ala-Gly)-N-acetyl SRIF and native SRIF elicit a similar displacement of the antibody-antigen complex. As previously described for certain antibodies raised against LHRH 3~, the NH~ terminal plays an important role in the

319 immunogenicity of the molecule. The COOH terminal, though necessary, seems to be less important, as des (COOH14)-SRIF shows immunoactivity only slightly less than that of native SRIF. These results enable comparisons between immunologic and biologic properties of the SRIF molecule at least insofar as biological activity is defined by effects on suppression of growth hormone secretion. The ring part of the peptide is necessary in both instances but the antibody does not recognize the precise sites necessary for biologic activity since des (NH2h-SRIF, has almost full biologic potency (unpublished data). The sensitivity of the assay for SRIF described in this paper is of the same order of magnitude as that of other radioimmunoassays for T R H z4, L H R H 32 and SR1F 1. Such assays have provided valuable tools for further elucidation of the functions and distributions of these small molecular weight peptides. The concentrations of SRIF found in various brain regions are approximately 70-80 ~; of the amounts previously reported by Brownstein et al. s. This slight discrepancy could be due to sex or strain differences as these investigators used OsborneMendel female rats, whereas in the present experiment, male Sprague-Dawley rats were used. Furthermore, animals were killed at different times during the day and preliminal y data from our laboratory indicate a marked variation in content of SRIF in brain taken at different times during the day. As in previous reports using bioassay 47 and radioimmunoassay z, the highest concentration of SRIF is in the hypothalamus; in this region the bulk of the activity is found in the MBH, particularly in ME, but the hormone is also present in the APO and in the remainder of the hypothalamus. The amygdala is the second most concentrated region we have tested with equal amounts present in the CMA and BLA. The hypothalamic distribution also agrees with cytoimmunological data which have shown that immunoreactive material is found in the external zone of the M E, in the anterior supraoptic and paraventriclar nuclei and in the APO12,22, T M . There are no cytoimmunological data concerning the amygdaloid distribution of somatostatin; studies done in collaboration with E. Zimmerman have failed to demonstrate the peptide in amygdala, cortex, brain stem or spinal cord (unpublished data). This failure could be explained by the relatively low concentration of SRIF in these regions as compared to the ME. The subcellular fractionation was monitored by enzyme marker and protein titration. The values obtained in these studies are in good agreement with the original data of Whittaker19, 51. Electron microscopic examination of the sucrose gradient showed a good yield of synaptosomes, over 70 ~ , in the B band; the LDH distribution parallels that of the synaptosomes. The good correlation between our controls and those in the literature provides evidence of the validity of our fractionation technique. The subcellular localization of SRIF in the MBH, APO and amygdala is totally parallel. Over 80 ~o of the hormone present in the homogenate is concentrated in the crude mitochondrial fraction (Pz) and 6 0 - 7 0 ~ in the synaptosomal band of the sucrose purification gradient. Such a distribution suggests that most SRIF is concentrated in nerve terminals in the brain. The possibility cannot be excluded that some soluble SRIF may have been originally present in the $1 and subsequently degraded

320 during the fractionation by the enzymatic activities known to be present in brain homogenates even at low temperatureZO, 2l. However, it should be stressed that the S R I F content of the sucrose homogenate is equivalent to the values obtained with the crude acidic extraction. Furthermore, the recovery rate along the fractionation was high, much greater than in similar studies with L H R H or T R H , which have shown that these hormones are also mainly localized in the nerve terminals of the MBHZ,4L The results, obtained by biochemical separation of various neural components, are in agreement with cytoimmunological data using Arimura's antibody22,a< 4t which also show that immunoreactive S R I F is predominantly found in nerve axons and terminals in median eminence. The localization of S R I F in the hypothalamus, amygdala and preoptic area is particularly significant since these regions have been shown in the rat to exert control over G H secretion from the adenohypophysis. For example, electrical stimulation of the hypothalamus, particularly of the ventromedial nucleus, elicits G H release 1'~',2,~--3~. whereas stimulation of the preoptic area and the corticomedial amygdala causes G H inhibition; on the other hand. stimulation of the basolateral amygdala causes G H release29, a0. The presence of SRIF in nerve terminals of the amygdaloid complex suggests that release may occur at synapses indicating a potential neurotransmitter role for the peptide. In recent papers, Renaud et al. as-4° have described certain tuberoinfundibular neurons in the V M H nucleus with axonal branches that project both to the ME and to the amygdala. Thus, it is possible that the same neuron, according to Dale's w i n ciplO 1, may liberate S R I F as a hormone into the portal blood in the ME and at the same time act as a neurotransmitter in the amygdala. Such a mechanism would provide a double control system for G H secretion: somatostatin may directly inhibit the secretion of G H f r o m adenohypophysis and also act within the amygdala or preoptic area in a neuronal feedback loop to control the activity ot extrahypothalamic regions involved in G H control. ACKNOWLEDGEMENTS We thank the Medical Research Council of Canada for financial support and Dr. M. G6tz and Dr. H. Immer, Ayerst Laboratories, Montreal, who generously provided synthetic somatostatin and several analogs. Mrs. Adete D ' A m a t o typed the manuscript.

REFERENCES 1 Arimura, A., Sato, H., Coy, D. H. and Schally, A. V., Radioimmunoassay for GH-release inhibiting hormones, Proc. Soc. exp. Biol. N.Y., 148 (t976) 784-792. 2 Arimura, A., Sato, H., Dupont, A., Nishi, N. and Schally, A. V., Somatostatin: abundance of immunoreactive hormone in rat stomach and pancreas, Science, 189 (1975) 1007-1009. 3 Barnea, A., Ben Jonathan, N., Colston, C., Johnston, J. M. and Porter, J. C., Differential subcellular compartmentalization of thyrotropin releasing hormone (TRH) and gonadotropin releasing hormone (LH-RH) in hypothalamic tissue, Proc. nat. Acad. Sei. (Wash.), 72 (1975) 3153-3157.

321 4 Brazeau, P., Vale, W., Burgus, R., Ling, N., Butcher, M., Rivier, J. and Guillemin, R., Hypotha[amic polypeptide that inhibits the secretion of immunoreactive pituitary growth hormone, Science, 197 (1973) 77-79. 5 Brazeau, P., Vale, W., Burgus, R. and Guillemin, R., Isolation of somatostatin (a somatropin release inhibiting factor) of ovine hypothalamic origin, Canad. J. Biochem., 52 (1974) 1067-1072 6 Brazeau, P., Vale, W., Rivier, J. and Guillemin, R., Biological potency of synthetic somatostatin in the oxidized and reduced form on the inhibition of secretion of growth hormone, Life Sci., 15 (1974) 351 357. 7 Bloom, S. R., Mortimer, C. H., Thorner, M. O., Besser, G. M., Hall, R., Gomez-Pan, A., Roy, V. M., Russell, R. C. A., Coy, D. H., Kastin, A. J. and Schally, A. V., Inhibition of gastrin and gastric-acid secretion by growth-hormone release-inhibiting hormone, Lancet, ii (1974) 1106 1109. 8 Brownstein, M., Arimura, A., Sato, H., Schally, A. V. and Kizer, J. S., The regional distribution of somatostatin in the rat brain, Endocrinology, 96 (1975) 1456-1461. 9 Burgus, R., Ling, N., Butcher, M. and Guillemin, R., Primary structure of somatostatin, a hypothalamic peptide that inhibits the secretion of pituitaly growth hormone, Proc. nat. Acad. Sci. (Wash.), 70 (1973) 684-686. 10 Burgus, R., Brazeau, P. and Vale, W., Isolation and determination of the primary structure of somatostatin (a somatotropin release inhibiting factor) of ovine hypotha[amic origin. In S. Raiti (Ed.), Advances hi Human Growth Homone, D H E W Publ. No. (Nil])74-612, U.S. Gov't Printing Office, Washington, D.C., 1973, pp. 144-158. l l Dale, H. A., Pharmacology and nerve endings, Proc. roy. Soc. B, 28 (1935) 319-332. 12 Dubois, M. P., Barry, J. et Leonardelli, J., Mise en 6vidence par immunofluorescence et r6partition de la somatostatine (SRIF) dans l'6minence m6diane des vertdbr6s, mammif6res, oiseaux, amphibiens, poissons, C.R. Aead. Sei. (Paris), 272 (1974) 1899 1901. 13 Dubois, M. P. et Kolodziecjezyk, E., Centres hypothalamiques du rat secr6tant de la somatostatin: r6paltition des p6rycarions en systSmes magno et parvocellulaires (6tude immunocytologique), C.R. Aead. Sci. (Paris.), 281 (1975) 1737-1740. 14 Dubois, M., Immunoreactive somatostatin is present in discrete cells of the endocrine pancreas, Proc. nat. Acad. Sci. (Wash.), 72 (1975) 1340-1342. 15 Frohman, L A. and Bernardis, L. L., Growth hormone and insulin levels in weanling rats with ventromedial lesions, Endocrinology, 82 (1968) 1125-1132. 16 Gerich, J. E., Lorenzi, M., Schneider, V. and Fotsham, P. H., Effect of somatostatin on plasma glucose and insulin responses to glucagon and tobbutamide in man, J. clin. Endocr., 39 (1974) 1057-1061. 17 Gerich, J. E., Lorenzi, M., Schneider, V., Kwan, C. W., Kavam, J. H., Guillemin, R. and Forsham, P., Inhibition of pancreatic glucagon responses to arginine by somatostatin in normal man and in insulin-dependent diabetics, Diabetes, 23 0974) 876-887. 18 Gerich, J. E., Lorenzi, M., Schneider, V., Karam, J. H., Rivier, J., Guillemin, R. and Forsham, P. H., Effects of somatostatin on plasma glucose and glucagon levels in human diabetes mellitus: pathophysiologic and therapeutic implications, New Engl. J. Med., 291 (1974) 544-548. 19 Gray, E. G. and Whittaker, V. P., The isolation of nerve endings from brain: an electron microscopy study of cell fragments derived by homogenisation and centrifugation, J. Anat. (Lond.), 96 (1962) 79 88. 20 Griffith, E. C. and Hooper, K. C., Peptidase activity in different areas of the rat hypothalamus, Acta endoer. (Kbh.), 77 (1974) 10 18. 21 Griffith, E. C., Hooper, K. C., Jeffcoate, S. L. and Holland, D. T., The presence of peptidase in the rat hypothalamus inactivating luteinizing hormone releasing hormone (LH-RH). Acta emlocr. (Kbh.), 77 (1974) 435-442. 22 H6kfelt, T., Efendic, S., Johansson, O., Luft, R. and Arimura, A., lmmunohistochemical localization of somatostatin (growth-hormone release-inhibiting factor) in the guinea pig brain, Brain Research, 80 (1974) 165-169. 23 H6kfe[t, T., Efendic, S., Hellstr6m, C., Johansson, O., Luft, R. and Arimura, A., Cellular localization of somatostatin in endocrine like cells and neurons of the rat with special references to the A1 cells of the pancreatic islets and to the hypothalamus, Acta endocr. (Kbh.), 80, Suppl 200 (1975) 1 41. 24 Jackson, 1. M. D. and Reich[in, S., Thyrotropin releasing hormone (TRH): distribution in hypothalamic and extrahypothalamic brain tissues of mammalians and submammalian chordates, Endocrinology, 95 (1974) 854-862.

322 25 Johnson, N. K. and Whittaker, V. P., Lactate dehydrogenase as a cytoplasmic marker in brain, Biochem. J., 88 (1963) 404--409. 26 Koerker, D. J., Ruch, W., Chideckel, E., Palmer, J., Goodner, C., Ensinck, J. and Gale, C. C.. somatostatin: hypothalamic inhibitor of the endocrine pancreas, Science, 184 (1974)492-483. 27 Kordon, C., Kerdelhu6, B., Pattou, E. and Jutisz, M., Immunohistochemical localization of LHR H in axons and nerve terminals of the rat median eminence, Proc. Soc. exp. Biol. (N. Y. ), 147 (1974) 122-127. 28 Lowry, O. H., Rosenbrough, J., Farr, A. L. and Randall, R. J., Protein measurements with the Folin phenol reagent, J. biol. Chem., 193 (1951) 265-275. 29 Martin, J. B., Plasma G H response to hypothalamic or extra hypothalamic electrical stimulation, Endocrinology, 91 (1972) 107-115. 30 Martin, J. B., Inhibitory effect of somatostatin (SRIF) on the release of growth hormone (GH) induced in the rat by electrical stimulation, Endocrinology, 94 (1974) 497-502. 3l Martin, J. B., Kontor, J. and Mead, P., Plasma G H responses to hypothalamic, hippocampal and amygdaloid electrical stimulation: effects of variation of stimulation parameters and treatment with a-methyl-para-tyrosine (aMT), Endocrinology, 92 (1973) 1354-1361. 32 Nett, T. M., Akbar, A. M., Niswender, G. D., Hedlund, M. J. and White, W. F., A radioimmunoassay for gonadotropin-releasing hormone (Gn-RH) in serum, J. clin. Endocr., 36 (1973) 880-885. 33 Ogawa, N., Friessen, H. G., Martin, J. B. and Brazeau, P., Radioreceptor assay for somatostatin, Program 58th Ann. Meet. Endocr. Sot., (1976) A-196 (Abstr.). 34 Palkovits, M., Brownstein, J., Arimura, A., Sato, H., Schally, A. V. and Kizer, J. S., Somatostatin content of the hypothalamic ventromedial and arcuate nuclei and the circumventricular organs in the rat, Brain Research, 109 (1976) 430434. 35 Patel, Y. C., Weir, G. C. and Reichlin, S., Anatomic distribution of somatostatin (SRIF) in brain and pancreatic islets as studied by radioimmunoassay (RIA), Program 57th Ann. t~/[eet. Endocr. Soc., June 18-20, (1975) 127 (Abstract). 36 Pelletier, G., Leclerc, R., Dube, D., Labrie, F., Puviane, R., Arimura, A. and Schally, A. V., Localization of growth-hormone release inhibiting hormone (somatostatin) in the rat brain, Amer. J. Anat., t42 (1975) 397-401. 37 Ramirez, V. D., Gautron, J. P., Epelbaum, J., Pattou, E., Zamora, A., and Kordon C., Distribution of LH-RH in subcellular fractions of the basomedial hypothalamus, Molec. Cell. Endocr., 3 (1975) 339-350. 38 Renaud, L. P., Tuberoinfundibular neurons in the basomedial hypothalamus of the rat: electrophysiological evidence for axon collaterals to hypothalamic and extra hypothalamic areas, Brah; Research, 105 (1976) 59-72. 39 Renaud, L. P., Influence of amygdala stimulation on the activity of identified tuberoinfundibular neurons in the rat hypothalamus. J. Physiol. (Lond.), 260 (1976) 237-252. 40 Renaud, L. P. and Martin, J. B., Electrophysiological studies of connections of hypothalamic ventromedial nucleus neurons in the rat: evidence for a role in neuroendocrine regulation, Brain Research, 93 (1975) 145-151. 41 Setalo, G., Vigh, S., Schally, A. V., Arimura, A. and Flerk6, B., GH-RIH containing neural elements in the hypothalamus, Brahl Research, 90 (1975) 352-356. 42 Shin, S. H., Morris, A., Snyder, J., Hymer, W. C. and Milligan, J. V., Subcellular localization of LH releasing hormone in the rat hypothalamus, Neuroendocrinology, 16 (1974) 191 -201. 43 Slier, T. M., Yen, S. S. C., Vale, W. and Guillemin, R., Inhibition by somatostatin on the release of TSH induced in man by thyrotropin-releasing factor, J. clin. Endocr., 38 (1974) 724-744. 44 Terry, L. C., Willoughby, J. O., Brazeau, P. and Martin, J. B., Antiserum to somatostatin prevents stress-induced inhibition of growth hormone secretion in the rat, Science, 192 (1976) 565-566. 45 Thorell, J. I. and Johansson, B. G., Enzymatic iodination of polypeptides with 12:JI to high specific activity, Biochim. biophys. Acta (Amst.), 251 (1971) 363-369. 46 Vale, W., Brazeau, P., Grant, G., Nussey, A., Burgus, R., Rivier, J., Ling, N. et Guillemin, R., Premi+res observations sur le mode d'action de la somatostatine, urt facteur hypothalamique qui inhibe la s6cr6tion de l'hormone de corissance, C.R. Acad. Sci. (Paris.) 275 (1972) 2913-2916. 47 Vale, W., Rivier, C., Palkovits, M., Saavedra, J. M. and Brownstein, J., Ubiquitous brain distribution of inhibitors of adenohypophysial secretion, Program. 56th Ann. Meet. Endocr. Soc., (1974) A-128 (Abstract). 48 Vale, W., Rivier, C., Brazeau, P. and Guillemin, R., Effects of somatostatin on the secretion of thyrotropin and prolactin, Endocrinology, 95 (1974) 968-976.

323 49 Weir, D. M., Immunological methods applied to the study of tissue antigens and antibodies. In D. M. Weir (Ed.), Blackwell, London, Handbook of Experimental Immunology, 1973, 35.5. 50 Whittaker, V. P., Michaelson, I. A. and Kirkland, R. S. A., The separation of synaptic vesicles from nerve endings particles (synaptosomes), Biochern. J., 90 (1964) 293-303. 51 Whittaker, V. P., The synaptosome, Handb. Neurochem., 2 (1969) 327-339.

Subcellular distribution of radioimmunoassayable somatostatin in rat brain.

Brain Research, 126 (1977) 309-323 ~(') Elsevier/North-Holland Biomedical Press, Amsterdam - Printed in The Netherlands 309 S U B C E L L U L A R D...
963KB Sizes 0 Downloads 0 Views