High biological activity of the synthetic replicates of somatostatin-28 and somatostatin-25

High biological activity of the synthetic replicates of somatostatin-28 and somatostatin-25

Regulatory Peptides, 1 (1981) 255-264 © Elsevier/North-Holland Biomedical Press 255 HIGH BIOLOGICAL ACTIVITY OF THE SYNTHETIC REPLICATES OF SOMATOST...

716KB Sizes 0 Downloads 0 Views

Regulatory Peptides, 1 (1981) 255-264 © Elsevier/North-Holland Biomedical Press



PAUL BRAZEAU, NICHOLAS LING, FREDERICK ESCH, PETER BOHLEN, ROBERT BENOIT and ROGER GUILLEMIN Laboratories for Neuroendocrinology, Salk Institute for Biological Studies, 1:'.O. Box 85800, San Diego, CA 92138, U.S.A. Accepted for publication 25 August 1980


We have isolated from extracts of ovine hypothalami two molecules characterized as somatostatin-28 and somatostatin-4-28 (referred to as somatostatin-25). They were reproduced by solid phase synthesis. In equimolar ratio and depending upon the experimental conditions, synthetic somatostatin-28 and somatostatin-25 are 3 - 1 4 times more potent than somatostatin14 to inhibit the basal in vitro secretion of growth hormone or as stimulated by prostaglandin (PGE~). In early studies in vivo, somatostatin-28 and somatostatin-25 are also more potent than somatostatin-14 in inhibiting the secretion of growth hormone acutely stimulated in the rat by injection of morphine; somatostatin-28 is also longer-acting than somatostatin- 14. These results suggest that somatostatin-14, as originally isolated, is a biologically active fragment of a larger molecule of greater specific activity;it should be considered as another form of somatostatin with high biological activity present in some tissues and likely secreted by the tissues along with somatostatin-14 and possibly other somatostatin-peptides of diverse sizes. somatostatin; precursors; growth hormone; bio-assay; hypothalamus; synthetic peptides; morphine

Author to whom correspondence should be addressed: Paul Brazeau, Laboratories for Neuroendocrinology, Salk Institute for Biological Studies, P.O. Box 85800, San Diego, CA 92; "~8, U.S.A.


In our early publications on the isolation and characterization of hypothalain'; somatostatin [ 1,2], we had noted and reported the existence of several chromatographic zones of biological activity other than the one corresponding to what we eventually isolated as the tetradecapeptide (somatostatin-14). Later on, several groups [3-61 made the same observation and conclusively showed that these represented several forms of immunoreactive somatostatin of larger molecular weights than that of the tetradecapeptide; on that account, these were considered as precursors of somatostatin-14. Similar data and conclusions were reported when deafing with extracts of the brain excluding hypothalamus, of the pancreas or of the gut. Recently, Pradayrol et al. [ 7,81 isolated and characterized an immunoreactive somatostatin from porcine intestinal extracts as a polypeptide containing 28 amino acid residues; somatostatin-28 was shown to contain a sequence identical to that of ovine hypothalamic somatostatin-14 as its COOH-terminal and a novel amino acid sequence of 14 residues as its NH2-~erminal, the complete primary structure as reported by Pradayrol et al. being H-Ser-Ala-Asn-Ser-AsnPro-AlaoMet-Ala-Pro-Arg-Glu-Arg-Lys-Ala-Gly-[Cys-Lys-Asn-Phe-Phe-TrpLys-Thr-Phe-Thr-Ser-Cysl-OH. We have recently isolated from extracts of ovine hypothalami two molecules corresponding to somatostatin-28 and somatostatin-4-28 (which we will call subsequently somatostatin-25). During the purification of these substances, we had been surprised by the biological activity of these natural products, apparently more active than the tetradecapeptide to inhibit in vitro the secretion of growth hormone. Somatostatin-28 and somatostatin-25 were reproduced by total synthesis to quantitatively ascertain their biological activities when related to that of the tetradecapeptide. MATERIALS AND METHODS

The synthetic peptides were prepared and purified by the usual techniques used in this laboratory [9]. The primary structures of synthetic somatostatin-28 and somatostatin-25 were verified by re-characterization of the amino acid sequence on a Beckman 890C sequencer. For the bioassays in vitro, monolayer tissue cultures of rat adenohypophyses were prepared and used in the assay as follows: The anterior pituitaries from usually 2 5 - 3 0 male SpragueoDawley rats (175 g) were removed after decapitation and dissected with microforceps, collected in Hepes buffer (pH ?.35) and dispersed at 37"C using two consecutive enzymatic solutions. The first solution consisted of a 25-ml solution

257 in Hepes buffer (pH 7.35) of collagenase (CLS, I; Worthington), 4 mg per ml, and dispase (Protease grade II, Boehringer Mannheim), 2 mg per ml. After an exposure of 70 min in this solution, ~he digested tissues were separated by gentle centrifugation (50 X g, 5 min) and resuspended in a solution of neuraminidase (Sigma N-2876) (8/zg/ml) and 200/~g/ml EDTA in Hepes buffer pH 7.35 for 10 min. After this second digestion, the cells were washed twice with the plating medium (defined below), and plated in multiwell-plates (Falcon No. 3008; 2 X l0 s cells per ml) using the following defined medium: F-I 2/DMEM/BGjb * (6 : 3 : I) (Gibco) with 2 g BSA/I, 2.38 g Hepes/l, 50 mg gentamycin/I plus the following supplements: cortisol 100/zg/l, insulin I/~g/l, Ta 0.4/~g/l, PTH 0.2 Fg/l, glucagon 10 ng/l, EGF 0.1 /~g/1, F G F 0.2 /tg/l and transferrin 10 mg/l. Plating was done with this medium with 2% fetal calf serum added to ensure rapid fixation of the cells to the plates. Two days later the cells were washed and reincubated for 2 more days using the medium as above without fetal calf serum. On the 4th day the cells were washed twice with the above medium. The peptides to be assayed were then added in concentrations ranging from 10 -12 M to 10 -9 M, each dose in triplicate wells for a secretion period of 3 h. The assays, when performed in that way in the above medium, allow detection of 10 -13 M of somatostatin-14 (EDso: 2 X 10 -1° M). In some experiments the inhibitory effects of somatostatin-28, -25 and -14 were studied by comparing the secretion of growth hormone stimulated by the addition of 10 -6 M prostaglandin (PGE2). Radioimmunoassays of growth hormone as well as prolactin secreted in the in vitro system were conducted with the reference materials and reagents prepared by Albert Parlow, NIH Hormone Distribution Program. In the case of GH-RIA an anti-mouse GH antiserum was used as first antibody (gift from Dr. Sinha, Scripps Clinic, La Jolla, CA). For in vivo studies, male Sprague-Dawley rats, 100 ± 10 g body weight, were accustomed to handling over a period of two days. To evaluate the relative potency of the somatostatin peptides, on the day of experiment the rats received three subcutaneous injections of the peptides in saline at doses specified below (see Table llI), 30 rain apart. 10 min after the last injection of somatostatin peptides the animals received an intraperitoneal injection of * F-12 = Ham's F-12 nutrient mixture (Gibco, ~30-1700); DMEM = Dulbecco's modifield Eagle's medium (Gibco, 430-1600); BGjb = BGjb medium, Fitton4ackson modification (Gibco, 320-2591); cortisol = hydrocortison (Sigma, H-4001); insulin = porcine insulin (Sigma, 13,505); T3 = 3,3',5-triido L-thyronine (Sigma, T2877); PTH = parathyroid h o r m o n e (Sigma, PO892); glucagon (Sigma, G4250); EGF = epithelial growth factor, and FGF = fibroblast growth factor (gifts from Dr. Denis Gospodarowicz, San Francisco).

Protocol number (a) 23707


23724 23761


Culture conditions

Defined medium

Defined medium

Defined medium

Defined medium

Defined medium

$25 $25 $25

$28 $28 $28

$25 $25

S28 $28 $28

S28 $28 $28 $28

9 10 11 12

10, 9 11, 10 12, 11

10, 9 11, 10 12, 11

11, 10 12, 11

10, 9 11, 10 12, 11

10, 11, 12, 13,

(10 "x M; ~x)

Doses of Peptides in molarity

7 4 3

6 6 2

7 6

21 14 3

9 8 8 8

Potency (H)

3-23 2-25 2-6

2-23 2-33 I-4

3-37 2-37

6-230 4-150 2-8

5-17 4-23 4-25 4-20


Confidence limits

0.2394 0.3504 0.2167

0.2807 0.3341 0.1632

0.2995 0.3162

0.3118 0.3377 0.2382

0.1387 0.2038 0.2292 0.1779

Index of precision








Average potency

$28 = somatostatin-28; $25 = somatostatin-25; FCS = fetal calf serum. (a): 4 points bioassays where the reference standard is somatostatin-14 (potency defined as 1). The average potency (H) is the arithmetical mean of all potencies (H), unweighted, as obtained in each experiment. The protocol numbers are laboratory codes identifying each experiment.

In vitro activity of somatostatin-28 and somatostatin-2$ as assayed in equimolar concentrations with somatostatin-I 4


t~ Oe


Defined medium + 10% FCS

14 16 9

9, 8 10, 9 11, 10


Defined medium + 10% FCS $2.5 $25 $25

$25 $25 $25


Defined medium + 2% FCS

2 3 4

11, 10 12, 11

10, 9

9 20

$28 $28 $28


Defined medium + 2% FCS

18 14 10

11, 10 12, 11 13, 12


10, 9 11, 10

$28 $28 $28


Defined medium + 2% FCS

11, 9


7 3 4 11

$28 $28



Defined medium

11, 9

9 10 11 12

3 4 6



Defined medium

10, 11, 12, 13,

10, 9 11, 10 12, 10

$25 $25 $25 $25


Defined medium

5-93 7-59 5-19

4-25 8-90

2-4 3-8 3-14

1-3 2-8 2-11

9-57 5-89 3-85



3-36 1-18 2-25 3-112

0.2907 0.2148 0.1522

0.1987 0.2315

0.0916 0.1668 0.1929

0.1372 0.2324 0.2318

0.1879 0,2848 0.3344



0.2934 0.3588 0.3483 0.3466





tO t.n


200/Jg of morphine sulfate in safine, a pharmacological agent, well known to acutely stimulate the secretion of growth hormone; the animals were decapitated 20 min after morphine injection and blood was collected in refrigerated glass tubes with heparin (100/~l) added for the measurement of growth hormone by the same radioimmunoassay as above. To evaluate the duration of action of the somatostatin-peptides, rats accustomed to handling as above, on the day of the experiment were given three subcutaneous injections of the peptides in saline at doses specified below (see Table lID, 30 min apart, but timed in such a way that the last injection would be either 2 h or 30 mi~ prior to decapitation. Animals and blood samples were then processed as described above. The statistical analyses of all the results were conducted for analysis of variance, multiple comparison tests, regression and potency calculatior, s, using a modification of the program EXBIOL. RESULTS AND DISCUSSION

As can be seen from the results presented in Table 1, somatostatin-28 and somatosta~in-25 give linear responses between doses of 10 -9 M and 10 -13 M which are parallel to those due to similar doses o f the tetradecapeptide. Depending on the experiment, and with 95% con~dence limits overlapping between experiments, somatostatin-28 and somatostatin-25 are 3 - 1 4 times more active than the tetradecapeptide, in equimolar ratios, to inhibit the basal secretion of growth hormone in the in vitro system. The statistical 'Jariations of the potency as calculated in this series of experiments seem to be in direct correlation with the level of basal growth hormone secretion by the cells in vitro. In this series of in vitro experiments, a nonweighted average of the calculated potencies is 9.4 for somatostatin-28 and 6.8 for somatostatin-25 when compared to that of somatostatin-14 as the reference standard. As shown in Table II, somatostatin-28 and somatostatin-25 again are more active than somatostatin-14 and in the same statistical ratio as above to inhibit the considerable secretion of growth hormone stimulated by the addition of prostaglandin, PGE2, to the culture medium simultaneously with the peptides. As shown in Table III, both somatostatin-28 and somatostatin-25 are considerably more active than the tetradecapeptide to inhibit the secretion of growth hormone as induced in vivo by the acute administration of morphine. In fact, at the doses used here the activity of somatostatin-28 and somatostatin-25 is so great that no true potency can be reliably calculated. This will require other experiments after well defined conditions of doses and timing

261 TABLE II In vitro inhibition of the secretion of growth hormone by somatostatin-28 ~nd somatostatin-25 as assayed in equimolar concentrations with somatostatin-14 on PGE2 (10 -6 M) pretreated pituitary cells Legend: see Table I. Protocol number (a)

Doses of peptides in molarity I0T M M; x

Potency (H)

Confidence limits (95%)

Index of precision (•)


$28 $28 $28

10, 9 11, I0 12, 11

6 7 10

4-8 4-18 4-32

0.0897 0.2209 0.2583


$25 $25 $25

10, 9 11, 10 12, 11

3 4 4

2-4 2-8 2-9

0.0897 0.2209 0.2583

Average potency (H)

for sampling have been determined, to meet the strict statistical criteria necessary for calculation of potency. On the basis of the results reported here, the potency of somatostatin-28 and somatostatin-25 in vivo has to be at least 10 times that of somatostatin-14 on a molar basis. Early results also show somatostatin-28 to be statistically longer-acting in vivo than somatostatin-14, at equimolar doses, in inhibiting the secretion of growth hormone stimulated by the injection of morphine. As shown in Table Ill, somatostatin-28 is still statistically highly active when administered 2 h prior to decapitation while somatostatin-14 is no more active. Under the same conditions, somatostatin-25 is not active. Obviously, more extensive experiments will have to be carried out with multiple doses, different timings, etc., to answer the question of the duration of activity of somatostatin28 and somatostatin-25. It would appear that, at equimolar doses; somato. star;n-28 is somewhat longer-acting than the tetradecapeptide. In none of the in vitro experiments described above have we observed any effect of somatostatin-28 or somatostatin-25 (or of the tetradecapeptide for that matter) on the secretion of prolactin. It would thus appear that rather than considering somatostatin-28 as a precursor of somatostatin-I 4 one should instead consider the tetradecapeptide originally isolated to be a fragment, biologically active, of a larger molecule of much greater activity. What, then, is the "true" somatostatin-product secreted? A part of the answer is the rejoinder that this may not be the right question, as such. There is already well established evidence that the precursor molecule pro-

262 TABLE !!i In vivo inhibition by somatostatin-28 and somatostatin-25 of the release of growth hormone induced by morphine sulfate; comparison of specific activity with somatostatin-14 and study of duration of activity • Protocol No.



Plasma rGH ng/ml ± S.E.M.


Saline S14, 1.0/~g S14, 10~g $28, 0.2 #g $28, 2.0;tg

12 12 12 12 12

773 531 435 405 380

± 44 ± 40 ± 39 ± 23 ± 23

** ** ** **

Saline S14, 1.0 ttg S14, 10#g $25,0.2/2g $25,2.0 #g

18 12 12 12 i2

736 6!2 377 441 259

± 25 ± 59 ± 26 ± 42 + 19

* ** ** **




Saline S14, 10#g, $28, 20/~g, S14, I0/~g, $28, 20 ttg,

1/2 h 1/2 h 2h 2h

17 12 12 12 12

686 ± 19 384 ± 30 301 ± 29 670±93 486 ± 55

Saline SI4, 10/~g, $25, 17 ttg, S!4, 10 rig, $25,17/Jg,

I/2h I/2 h 2h 2h

15 11 1! 11 11

817 ± 57 368±40 273 ± 20 685 ± 65 818 ± 66

P value

** ** ** ** **

P value calculated by the multiple comparison test of Dunnett for significance of the difference from values for the saline control group. - : not significant; *: 5%, **: 1%; n: number of rats per experimental group. Other codes as in legend to Table I.

opiocortin is processed quite differently in adenohypophyseal cells and in cells of the pars intermedia, leading to statistically different amounts of corticotropin (ACTH), ~-iipotropin (LPH) and ~-endorphin, as well as of melanotropin (a-MSH) to be found in extracts of the several tissues or secreted by these same tissues [ 10]. Thus, it should not seem surprising if somatostatin-14 and somatostatin-28 were to be processed and secreted differently in and by the different tissues involved, such as hypothalamic neurons, gastric mucosa, duodenum, endocrine pancreas. In fact, early results by ourselves and others show that both somatostatin-28 and somatostatin- 14 are found in hypothalamic extracts, somatostatin- 14 is to be found almost exclusively in extracts of the pancreas or of the stomach, while

263 somatostatin-28 appears to be the prevalent form in the gut. Similarly, hepatic portal blood contains almost exclusively somatostatin-14, but hydrochloric acid stomach perfusate contains both soma tostatin-28 and somatostatin-14, as judged from gel permeation studies. Thus, the true secretion product may be defined 0niy as astatiStical m i x t u r e of both somatostatin-28 and somatostatin-14, of variable composition. It is impossible to know, at this moment, whether somatostatin-25 is an artifact of the extraction method, storage o f the starting material, etc., or corresponds to a product processed in vivo and actively secreted as such. Should this latter proposal be ascertained, the true somatostatin secretory product may well include, besides somatostatin-28 and somatostatin-14, in variable ratios several forms with NH2-terminal extensions of several possible lengths from the tetradecapeptide. Ongoing studies by several groups who have succeeded in isolating and cloning the mRNA for somatostatin from various tissues should help in clarifying the maximal length of the peptide sequence extending the tetradecapeptide and still endowed with biological activity. ACKNOWLEDGEMENTS Research supported by grants from NIH (AM 18811-06, HD 09690-06), the W.R. Hearst Foundation and the National Foundation March of Dimes. We gladly acknowledge the skillful technical assistance of D. Martineau, J. Addison, G. Sturdevant, D. Davis, M. Regno and M. Mercado. REFERENCES 1 Brazeau, P., Vale, W., Burgus, R., Long, N., Butcher, M., Rivier, J. and Guillemin, R., A hypothalamic polypeptide that inhibits the secretion of pituitary growth hormone, Science, 179 (I 973) 77- 79. 2 Brazeau, P., Vale, W., Guillemin, R. and Burgus, R., Isolation of somatostatin (a somatotropin release inhibiting factor) of ovine hypothalamic origin, Can. J. Biochem., 52 (1974) 1067-1072. 3 Schally, A.V., Dupont, A., Arimura, A., Redding, T.W., Nishi, N., Linthicum, G.L. and Schlesinger, D.H. Isolation and structure of somatostatin from porcine hypothalami, Biochemistry, 15 (1976) 509-514. 4 MiUar, R.P., Somatostatin immunoreactive peptides of higher molecular weight in ovine hypothalamic extracts, L Endocrinol., 77 (1978) 429-430. 5 Lauber, M., Camier, M. and Cohen, P., Higher molecular weight forms of immunoreactive somatostatin in mouse hypothalamic extracts: Evidence of processing in vitro, Proc. Natl. Acad. Sci. U.S.A., 76 (1979) 6004-6008. 6 Zysnar, E., Conlon, ].M., Schusdziarra, V. and Unger, R., Properties of somatostatinlike immunoreactive polypeptides in canine extrahypothalamic brain and stomach, Endocrinology, 105 (1979) 1426-1431.

264 7 Pradayrol, L., Chayvialle, I.A., Cadquist, M. and Mutt, V., Isolation of a porcine intestinal peptide with C-terminal somatostatin, Biochem. Biophys. Res. Commun., 85 (1978) 701-708. 8 Pradayrol, L., $6rnvaU, H., Mutt, V., and Ribet, A., N-terminally extended somatostatin: the primary structure of somatostatin-28, FEBS Lett., 109 (1980) 55-58. 9 Ling, N., Esch, F., Davis, D., Mercado, M., Regno, M., B6hlen, P., Brazeau, P. and GuiUemin, R., Solid phase synthesis of somatostatiw28, Biochem. Biophys. Res. Commun. (in press). I 0 Eipper, B.A. and Mains, R.E., Existence of a common precursor to ACTH and endorphin in the anterior and intermediate lobe of the rat pituitary, J. Supramol. Struct., 8 (1978) 247-262.