- Email: [email protected]
Cyclic hexa- and pentapeptide somatostatin analogues with reduced gastric inhibitory activity
Peptides, Vol. 5, pp. 857-860. 1984. ~"Ankho International inc. Printed in the U.S.A.
0196-9781/84 $3.00 + .00
Cyclic Hexa- and Pentapeptide Somatostatin Analogues with Reduced Gastric Inhibitory Activity B A R R Y H . H I R S T , *~ E D U A R D O A R I L L A , * ' - ' D A V I D H . C O Y ? AND BERNARD SHAW*
*Department o f Physiological Sciences, Medical School, University o f Newcastle upon Tyne Newcastle upon Tyne NEI 7RU, U.K. and ?Department o f Medicine, Tulane University School o f Medicine, New Orleans, LA 70112 R e c e i v e d 23 J a n u a r y 1984 HIRST. B. H.. E. ARILLA, D. H. COY AND B. SHAW. Cyclic"hexa- and pentapeptide somatostatin analogues with reduced gastric" inhibitory activity. PEPTIDES 5(5) 857-860, 1984.--The gastric inhibitory activity of cyclic hexa- and pentapeptide analogues of somatostatin was investigated in conscious cats with gastric fistdae. Gastric acid and pepsin secretions were stimulated by pentagastrin. Cycio~Phe-Phe-I)-Trp-Lys-Thr-Phe) showed no inhibition of acid secretion at molar doses up to 50-times the ID~ for somatostatin. This peptide inhibited pepsin secretion at the highest dose (50 p.g kg-1 hr-~. and its potency is approximately 0.005 compared with somatostatin (1.0). Cyclo(Pro-Phe-D-Trp-Lys-Thr-Phe) inhibited acid (~50%) and pepsin (-85%) secretions, but the inhibition was not dose-related being similar with doses of 10 to 50 jzg kg-I hr-~. The cyclic pentapeptide, cyclo(7-aminoheptanoyI-Phe-D-Trp-Lys-Thr),was inactive in the dose range studied, with a potency <0.01. Cyclo[7-aminoheptanoyl-Phe-D-Trp-Lys-Thr(Bzl)]has been described as a somatostatin antagonist with respect to inhibition of growth hormone, insulin and glucagon release in rats [2]. Up to 60-fold molar excesses of this peptide failed to antagonise the inhibitory activity of somatostatin in the stomach. The results demonstrate that residues outside the central 6-11 region of somatostatin are very important for its gastric activity. The lack of gastric antagonistic activity of the pentapeptide antagonist indicates that these residues are likely to he involved in receptor recognition/binding. Gastric acid secretion
Pepsin secretion
Peptides
Somatostatin
SHORT. cyclic analogues of somatostatin have been developed which retain marked inhibitory activity against growth hormone (GH), insulin and glucagon secretion [9--11]. The development o f these short peptides greatly increases the ease with which large numbers of structural analogues of somatostatin can by synthesized in order to better define structure-function relationships. Indeed, studies based on such rationale have led to the synthesis of the first somatostatin antagonist, a cyclic pentapeptide [2]. Somatostatin is also a potent inhibitor of gastric acid and pepsin secretions [3]. Short-chain somatostatin analogues can be equally useful in defining structure-activity relationships of somatostatin in the stomach. In addition, the availability o f a somatostatin analogue antagonistic towards its actions in the stomach would help to elucidate its physiological role in this organ. We report the inhibitory activity of cyclic hexa- and penta-peptide analogues of somatostatin against gastric secretions, and investigate the activity of the recently described somatostatin antagonist in the stomach.
Somatostain antagonist
METHOD Experiments were carried out in six conscious cats prepared with cannulated gastric fistulae under full surgical anaesthesia at least one year earlier. F o o d was withheld from the animals for 36 hr before experiments, but free access to water was allowed. Gastric secretions were collected continuously by gravity drainage and divided into 15 rain samples. Acid output was determined by electrometric titration o f a 1.0 ml sample o f gastric juice to pH 7.0 with 0.15 M-NaOH (Radiometer, Copenhagen, Denmark). Peptic activity was measured by a haemoglobin digestion method [8]. A percutaneous needle was inserted into a cephalic vein, and 0.9% NaCl infused at 10 ml hr -~. All compounds were added to this infusion.
Materials Gastric secretion was stimulated by pentagastrin (Peptavion®; I.C.I. Pharmaceuticals, Macclesfield, U.K.) infused
1Requests for reprints should be addressed to Dr. B. H. Hirst. :On leave from: Department of Biochemistry, School of Medicine, University de AIcala de Henares, Aicala de Henares, Madrid, Spain.
857
858
HIRST, ARILLA, COY AND SHAW TABLE 1 DOSE-DEPENDENT EFFECTS OF SOMATOSTATINAND CYCLIC HEXA-AND PENTA-PEPTIDEANALOGUESON GASTRICACID AND PEPSIN SECRETION
Peptide Somatostatin
Dose (v.g kg-l hr-1)
Acid Inhibition (%)*
Pepsin Inhibition (c~),
0 0.5 1.0 2.0 4.0
0.4 31.5 44.2 50.5 73.9
± 8.0(6) ± 7.8(6)t __ 12.0(6)* -- 11.9(6)* ± 2.8(6)+
21.4 --_ 16.6(6) 66.7 ± 13.0(6)~ 79.5 ± 9.0(6)* 90.3 ± 3.0(6)t 92.7 ± 3.7(6)t
CHPI
0 10.0 25.0 50.0
7.6 7.7 7.0 -9.2
__ 4.7(6) ± 14.2(6) ± 15.8(6) ± 19.8(5)
8.0 -35.2 37.8 44.9
± ± ± ±
21.0(6) 53.0(6) 23.5(6) 10.8(5)~
CHP2
0 I0.0 25.0 50.0
3.9 46.1 49.8 52.7
± ± ± ±
13.2(5) 13.0(5)~" 7.3(5)~ 12.1(4)~
-19.0 80.9 87.0 85.3
± ± ± --
26.9(5) 6.4(5)* 3.0(5)t 2.4(5)~,
CPP2
0 10.0 25.0 50.0
-7.2 ± 19.5 ± -12.8 ± -9.5 ±
17.1(5) 11.1(5) 12.7(5) 17.0(5)
-I.0 -36.6 -24.3 -7.2
± ± ± ±
35.0(5) 80.0(5) 35.7(5) 25.0(5)
Gastric secretions were stimulated by pentagastrin, 8 t~g kg-' hr-'. *Mean -.+ I s.e. mean(n). ~'p<0.05.
continuously at 8/zg kg-' h r - ' . Somatostatin and somatostatin analogues (Fig. 1) were synthesized either by solid-phase methodology (somatostatin, CCPI and CCP2; [2]) or by conventional fragmental-condensation procedures (CHPI and CHP2). Peptides were purified by either preparative HPLC (CPPI and CPP2) or counter-current distribution (CHPI and CHP2), and were homogeneous as judged by TLC in various solvent systems. Amino acid analysis of the acid hydrolysates gave the expected values. CHPI and CHP2 were supplied by Dr. R. H. Andreatta, Ciba-Geigy, Basel, Switzerland.
CHP1
6 7 8 9 i0 ii cyclo(Phe-Phe-D-Trp-Lys-Thr-Phe)
CHP2
6 7 8 9 i0 ii cyclo(Pro-Phe-D-Trp-Lys-Thr-Phe)
CPP1
7 8 9 i0 eyclo[Aha-Phe-D-Trp-Lys-Thr(Bzl)]
CPP2
7 8 9 I0 eye Io (Aha-Phe-D-Trp-Lys-Th r)
Experimental Protocols The assessment of the inhibitory action of a range of doses of somatostatin and analogues was based on the percent inhibition of gastric acid and pepsin secretions stimulated by pentagastrin, 8 ~g kg-' h r - ' . These methods are reproducible and have been described in detail [1, 4-7]. The control response to pentagastrin alone was determined in each animal. In experiments with somatostatin or analogues, pentagastrin was infused alone for 45 rain, and then increasing doses of somatostatin or analogue were added to the pentagastrin infusion for subsequent 30 rain periods. The acid output in the 15 min immediately before infusion of somatostatin differed by less than 10% that in the control experiment (see Table 1). The inhibition of acid and pepsin output produced by each dose of somatostatin analogue was calculated by subtracting the output during the second 15 rain of infusion of each dose from the control response at the same period of time in the same animal, and expressed as a percentage of the control response. The antagonistic activity of CPP1 was investigated by in-
FIG. I. Primary structure of short-chain cyclic somatostatin analogues. Numbering of amino acid residues refers to native somatostatin-14. Aha=7-aminoheptanoyl, Bzl=benzyl.
fusing it during somatostatin-inhibition of gastric secretion stimulated by pentagastrin. Pentagastrin, 8 ~g kg- ' h r - ' , and somatostatin, 2 or I0/zg kg- ' h r - ' , were infused together for 2.5 hr. CPPi was added to this infusion for the middle 1 hr. Significance of results was estimated by analysis of variance. RESULTS AND DISCUSSION The two cyclic hexapeptides (CHPI and 2) and the cyclic pentapeptide (CPP2) were very weak inhibitors of gastric
SOMATOSTATIN A N A L O G U E S : GASTRIC EFFECTS secretions in comparison to the native peptide. Somatostatin is a potent inhibitor of gastric acid secretion with an IDa,, of 1.2-1.7 nmol kg -~ hr ' [4.6] (and see Table l). Pepsin secretion stimulated by pentagastrin is about 5-times more sensitive to inhibition by somatostatin (Table i and [4,6]). CHPI constitutes cyclised residues 6-11 of [D-Trp"lsomatostatin (Fig. 1). This peptide showed no significant inhibition of acid secretion at molar doses up to 50-times the ID.~, for somatostatin (Table l). At the highest dose, significant inhibition of pepsin secretion was achieved (Table I). Thus the potency of CHPI in the stomach is approximately 0.005 compared with somatostatin (1.0). In comparison, CHP1 has a relatively greater (0.27) potency against GH release in the rat [10l. The very low activity of CHPI in the stomach may be partially explained by the substitution of the cystine disulphide bridge with a peptide link, since similar replacement in the full ring structure yields an analogue with a potency of 0.22 [5]. However, we have previously reported that a cyclic octapeptide consisting of residues 6-11 (with D-Trp ~ as CHPI) but cyclised through a cystine link (cyclo[Cys-Phe-Phe-D-TrpLys-Thr-Phe-Cys]) was also inactive (potency<0.05) against gastric secretions I6]. Substituting the Phe" residue in CHPI for Pro to give CHP2 (Fig. 1) increased gastric inhibitory activity (Table 1). Similarly, GH release inhibitory activity is increased by the same modification [10]. The increased activity of the proline substituted hexapeptide has been argued to be the result of favourable conformational influences of this residue [10]. However, CHP2 demonstrated reduced activity against gastric secretions in the cat (Table 1), while showing a relative potency of 5-25, compared with somatostatin, as an inhibitor of GH, insulin and glucagon release in rats in vivo [10]. A five-fold range of doses (10-50/zg kg-' hr-') resulted in approximately 50% inhibition of acid, and approximately 85% inhibition of pepsin secretion (Table 1). This dose-unrelated relationship is unusual, and we have not observed it with any other somatostatin analogue [ 1,4--6]. It may be related to the overall reduced activity of the short-chain cyclic analogues, but we can offer no definite explanation at the present time. CPP2 consists of residues 7-10 of somatostatin cyclised through 7-amino-heptanoyi (Fig. 1). This peptide was inactive (potency<0.01) in the dose range studied (Table 1). CPP2 was also inactive as an inhibitor of GH release in rats [2]. CPPI, equivalent to CPP2 but with the benzyl protecting group on T h r " retained (Fig. 1), has been described as an antagonist of somatostatin with respect to its actions against GH, insulin and glucagon release in the rat [2]. In the stomach, CPPI in doses ranging from 2-50/~g-' hr -~ did not alter somatostatin inhibition of acid or pepsin secretion. For example, somatostatin 2 p.g kg-' hr -~ almost completely inhibited pepsin secretion stimulated by laentagastrin, with a less marked effect on acid secretion. Addition of CPPI at 50 /zg kg -~ hr -~ (60-fold molar excess compared with somatostatin) did not alter the gastric secretions (Fig. 2). Similarly, CPPI at 50 p.g kg -1 hr -1 ( 10-fold molar excess) failed to influence the more marked inhibition of gastric secretion produced by somatostatin 10/xg kg -~ hr-J (Fig. 3). Doses of CPPI of 6--10/zg kg -x were sufficient to antagonise the actions of endogenous somatostatin on GH, insulin and glucagon release in rats, while molar excesses of 2-10-fold antagonised the effects of exogenous somatostatin against these secretions [2]. The lack of antagonistic activity of CPPI in the stomach, although disappointing, may not be surprising when the reduced activity of the cyclic hexapeptide analogues in this organ is considered.
859
,Pentagastrin 8)ug kg-'h-' o A u ,Somatostotin 2)ug kg-'h-' e A ' CPPI 50jug kg-lh -1 • ' 600r
/b //
•
2oo 0
/7 "
I600 1400
~
./
3.S 1200 ~E 3 ~ I000 O~ C ~" 800
"~_~
a~
600:
£L ~, 4 0 0 200 t
0
I
30
I
60
I
90
I
t
120
150
Time (rain) FIG. 2. Failure of CPP1 to influence somatostatin-inhibition o~ pentagastrin-stimulated gastric secretions. Experiments are illustrated as pentagastrin, 8 ~g kg-~ hr -R, infused alone for 150 min (©), pentagastrin, 8 p.g kg-' hr-', infused together with somatostatin, 2 ~.g kg-~ hr-~, for 150 min (A), and CPP1, 50 p.g kg-~ hr-~, infused during the middle hour of an infusion of pentagastrin, 8 ~g kg-~ hr-~, plus somatostatin, 2/zg kg-~ hr-j (0). Values illustrated as mean 15 acid or pepsin outputs (n--6).
The data reported here illustrate the marked differences in the structural requirements for somatostatininhibition of feline gastric exocrine secretions compared with rat endocrine secretions. Short-chain cyclic analogues of somatostatin, while retaining marked inhibitory activity against pituitary and pancreatic secretions in the rat [9-11], show only low activity against fefine gastric secretions. The failure of the cyclic pentapeptide CPP1 to act as a somatostatin antagonist in the stomach exemplifies these differences. Comparisons of the gastric and endocrine activities of these somatostatin analogues are, however, complicated by several experimental differences. The gastric studies were carried out in cats, while the endocrine studies in rats. However, short-chain bicyclic analogues, although not entirely analogous to the short-chain analogues described here, also show reduced gastric activity in the dog [11]. Somatostatin and analogues were administered by continuous infusion of the peptides in the gastric studies, compared with bolus injections in the pituitary and pancreatic studies.
860
HIRST, A R I L L A , COY AND SHAVv
, Pentagostrin BAg kg'lh -t I
o,
Somatostatin lO~g kg-~h -~ ,' CPPI 5 0 ~ g kg-lh -1 •'
600 .c_ 500 ~Se ~ 400 o
o~ 300
~- :~00 ,oo
3
o
1600 1400 / 12oo ,ooo
,- ~
800
U'J ..~
0L I
0
I
30
I
60
I
90
I
120
This experimental difference may be of particular importance when peptides of differing duration of action are being compared. Indeed, the duration of action of CHP2 in inhibiting GH release in the rat after SC or oral administration is prolonged compared with somatostatin [10]. However, increased duration of action of these analogues would be expected to increase their potency compared with somatostatin, which contrasts to the observed decrease in potency. A third difference between the gastric and endocrine studies is the degree to which the various secretions are stimulated. Gastric acid secretion was stimulated at near-maximal rates by pentagastrin [I], while the endocrine secretions were either unstimulated, or only generally stimulated by anaesthesia. However, gastric pepsin was secreted at only approximately one-third of the maximum rate [l], but the shortchain cyclic analogues also demonstrated low activity against this secretion, as was seen for acid secretion. In summary, comparison of the gastric potency of the somatostatin analogues with their reported potency against endocrine secretions is not simple. However, considering that in each system somatostatin is included as an internal standard, such comparisons are useful, and the general conclusion that short-chain cyclic analogues are relatively inactive in the stomach compared with the pituitary and pancreas appears valid. We conclude that our results suggest that residues outside the central 6- I 1 region of somatostatin are more important for gastric than pituitary or pancreatic somatostatin receptors. The lack of antagonistic activity of CPPI in the stomach indicates that these residues are more likely to be involved in receptor recognition/binding rather than activation.
1
150 ACKNOWLEDGEMENTS
Time (min) FIG. 3. Failure of CPP! to influence somatostatin-inhibition of pentagastrin-stimulated gastric secretions. Exper/ments illustrated as pentagastrin, 8 tzg kg-' hr-', infused alone for 150 rain (©), and ~ n tagastrin, 8 ~g kg-' hr-t, infused with somatostatin, 10 ~g kg-t hr-', for 150 min and CPPI, 50 t~g kg-~ hr-t, added during the middle hour (0). Values illustrated as mean-'-1 s.c. mean (n=6).
We thank Ken EUiott and David Stephenson for technical assistance, and Dr. R. H. Andreatta (Ciba-Geigy, Basel, Switzerland) for generously supplying CHP 1 and 2. This work was supported in part by NIH grant AM-18370. E. A. was supported by a grant from the British Council.
REFERENCES 1. Brown, M. P., D. H. Coy, A. Gomez-Pan, B. H. Hirst, M. Hunter, C. Meyers, J. D. Reed, A. V. Schally and B. Shaw. Structure-activity relationships of eighteen somatostatin analogues on gastric secretion. J Physiol 277: 1-14, 1978. 2. Fries, J. L., W. A. Murphy, J. Sueiras-Diaz and D. H. Coy. Somatostatin antagonist analog increases GH, insulin and glucagon release in the rat. Peptides 3:811-814, 1982. 3. Gomez-Pan, A., J. D. Reed, M. Albinus, B. Shaw, R. Hall, G. M. Besser, D. H. Coy, A. J. Kastin and A. V. Schally. Direct inhibition of gastric acid and pepsin secretion by growth hormone-releasing inhibiting hormone in cats. Lancet h 888890, 1975. 4. Hirst, B. H., J. M. Conion, D. H. Coy, J. Holland and B. Shaw. Comparison of gastric exocrine inhibitory activities and plasma kinetics of somatostatin-28 and somatostatin-14 in cats. Regul Pep 4: 22%237, 1982. 5. Hirst, B. H., J. D. Reed, B. Shaw, C. F. Hayward and J. S. Morley. Non-reducible cyclic, and azaphenylalanyl" analogues of somatostatin. Fur J Pharmacol 65: 151-156, 1980. 6. Hirst, B. H., B. Shaw, C. A. Meyers and D. H. Coy. Structure-activity studies with somatostatin: the role of tryptophan in position 8. Regul Pep h 97-113, 1980.
7. Reed, J. D., B. H. Hirst, A. Gomez-Pan, D. H. Coy, A. V. Schally and C. Meyers. Inhibition of gastric secretion by stereoisomers of somatostatin. Metabolism 27:1411-1413, 1978. 8. Shaw, B. and C. L. Wright. The pepsinogens of cat gastric mucosa and the pepsins derived from them. Digestion 4: 142152, 1976. 9. Vale, W., C. Rivier, M. Brown and J. Rivier. Pharmacology of thyrotropin releasing factor (TRF), luteinizing hormone releasing factor (LRF), and somatostatin. In: Hypothalamic Peptide Hormones and Pituitary Regulation, edited by J. C. Porter. New York: Plenum Press, 1977, pp. 123-156. 10. Veber, D. F., R. M. Freidinger, D. S. Periow, W. J. Paleveda, F. W. Holly, R. G. Strachan, R. F. Nutt, B. H. Arison. C. Homnick, W. C. Randall, M. S. Glitzer, R. Saperstein and R. Hirschmann. A potent cyclic hexapeptide analogue of somatostatin. Nature 292: 55-58, 1981. 11. Veber, D. F., F. W. Holly, W. J. Paleveda, R. F. Nutt. S. J. Bergstrand, M. Torchiana. M. S. Glitzer, R. Saperstein and R. Hirschmann. Conformationally restricted bicyclic analogs of somatostatin. Proc Nat/Ac'ad Sci USA 75: 2636-2640, 1978.
0196-9781/84 $3.00 + .00
Cyclic Hexa- and Pentapeptide Somatostatin Analogues with Reduced Gastric Inhibitory Activity B A R R Y H . H I R S T , *~ E D U A R D O A R I L L A , * ' - ' D A V I D H . C O Y ? AND BERNARD SHAW*
*Department o f Physiological Sciences, Medical School, University o f Newcastle upon Tyne Newcastle upon Tyne NEI 7RU, U.K. and ?Department o f Medicine, Tulane University School o f Medicine, New Orleans, LA 70112 R e c e i v e d 23 J a n u a r y 1984 HIRST. B. H.. E. ARILLA, D. H. COY AND B. SHAW. Cyclic"hexa- and pentapeptide somatostatin analogues with reduced gastric" inhibitory activity. PEPTIDES 5(5) 857-860, 1984.--The gastric inhibitory activity of cyclic hexa- and pentapeptide analogues of somatostatin was investigated in conscious cats with gastric fistdae. Gastric acid and pepsin secretions were stimulated by pentagastrin. Cycio~Phe-Phe-I)-Trp-Lys-Thr-Phe) showed no inhibition of acid secretion at molar doses up to 50-times the ID~ for somatostatin. This peptide inhibited pepsin secretion at the highest dose (50 p.g kg-1 hr-~. and its potency is approximately 0.005 compared with somatostatin (1.0). Cyclo(Pro-Phe-D-Trp-Lys-Thr-Phe) inhibited acid (~50%) and pepsin (-85%) secretions, but the inhibition was not dose-related being similar with doses of 10 to 50 jzg kg-I hr-~. The cyclic pentapeptide, cyclo(7-aminoheptanoyI-Phe-D-Trp-Lys-Thr),was inactive in the dose range studied, with a potency <0.01. Cyclo[7-aminoheptanoyl-Phe-D-Trp-Lys-Thr(Bzl)]has been described as a somatostatin antagonist with respect to inhibition of growth hormone, insulin and glucagon release in rats [2]. Up to 60-fold molar excesses of this peptide failed to antagonise the inhibitory activity of somatostatin in the stomach. The results demonstrate that residues outside the central 6-11 region of somatostatin are very important for its gastric activity. The lack of gastric antagonistic activity of the pentapeptide antagonist indicates that these residues are likely to he involved in receptor recognition/binding. Gastric acid secretion
Pepsin secretion
Peptides
Somatostatin
SHORT. cyclic analogues of somatostatin have been developed which retain marked inhibitory activity against growth hormone (GH), insulin and glucagon secretion [9--11]. The development o f these short peptides greatly increases the ease with which large numbers of structural analogues of somatostatin can by synthesized in order to better define structure-function relationships. Indeed, studies based on such rationale have led to the synthesis of the first somatostatin antagonist, a cyclic pentapeptide [2]. Somatostatin is also a potent inhibitor of gastric acid and pepsin secretions [3]. Short-chain somatostatin analogues can be equally useful in defining structure-activity relationships of somatostatin in the stomach. In addition, the availability o f a somatostatin analogue antagonistic towards its actions in the stomach would help to elucidate its physiological role in this organ. We report the inhibitory activity of cyclic hexa- and penta-peptide analogues of somatostatin against gastric secretions, and investigate the activity of the recently described somatostatin antagonist in the stomach.
Somatostain antagonist
METHOD Experiments were carried out in six conscious cats prepared with cannulated gastric fistulae under full surgical anaesthesia at least one year earlier. F o o d was withheld from the animals for 36 hr before experiments, but free access to water was allowed. Gastric secretions were collected continuously by gravity drainage and divided into 15 rain samples. Acid output was determined by electrometric titration o f a 1.0 ml sample o f gastric juice to pH 7.0 with 0.15 M-NaOH (Radiometer, Copenhagen, Denmark). Peptic activity was measured by a haemoglobin digestion method [8]. A percutaneous needle was inserted into a cephalic vein, and 0.9% NaCl infused at 10 ml hr -~. All compounds were added to this infusion.
Materials Gastric secretion was stimulated by pentagastrin (Peptavion®; I.C.I. Pharmaceuticals, Macclesfield, U.K.) infused
1Requests for reprints should be addressed to Dr. B. H. Hirst. :On leave from: Department of Biochemistry, School of Medicine, University de AIcala de Henares, Aicala de Henares, Madrid, Spain.
857
858
HIRST, ARILLA, COY AND SHAW TABLE 1 DOSE-DEPENDENT EFFECTS OF SOMATOSTATINAND CYCLIC HEXA-AND PENTA-PEPTIDEANALOGUESON GASTRICACID AND PEPSIN SECRETION
Peptide Somatostatin
Dose (v.g kg-l hr-1)
Acid Inhibition (%)*
Pepsin Inhibition (c~),
0 0.5 1.0 2.0 4.0
0.4 31.5 44.2 50.5 73.9
± 8.0(6) ± 7.8(6)t __ 12.0(6)* -- 11.9(6)* ± 2.8(6)+
21.4 --_ 16.6(6) 66.7 ± 13.0(6)~ 79.5 ± 9.0(6)* 90.3 ± 3.0(6)t 92.7 ± 3.7(6)t
CHPI
0 10.0 25.0 50.0
7.6 7.7 7.0 -9.2
__ 4.7(6) ± 14.2(6) ± 15.8(6) ± 19.8(5)
8.0 -35.2 37.8 44.9
± ± ± ±
21.0(6) 53.0(6) 23.5(6) 10.8(5)~
CHP2
0 I0.0 25.0 50.0
3.9 46.1 49.8 52.7
± ± ± ±
13.2(5) 13.0(5)~" 7.3(5)~ 12.1(4)~
-19.0 80.9 87.0 85.3
± ± ± --
26.9(5) 6.4(5)* 3.0(5)t 2.4(5)~,
CPP2
0 10.0 25.0 50.0
-7.2 ± 19.5 ± -12.8 ± -9.5 ±
17.1(5) 11.1(5) 12.7(5) 17.0(5)
-I.0 -36.6 -24.3 -7.2
± ± ± ±
35.0(5) 80.0(5) 35.7(5) 25.0(5)
Gastric secretions were stimulated by pentagastrin, 8 t~g kg-' hr-'. *Mean -.+ I s.e. mean(n). ~'p<0.05.
continuously at 8/zg kg-' h r - ' . Somatostatin and somatostatin analogues (Fig. 1) were synthesized either by solid-phase methodology (somatostatin, CCPI and CCP2; [2]) or by conventional fragmental-condensation procedures (CHPI and CHP2). Peptides were purified by either preparative HPLC (CPPI and CPP2) or counter-current distribution (CHPI and CHP2), and were homogeneous as judged by TLC in various solvent systems. Amino acid analysis of the acid hydrolysates gave the expected values. CHPI and CHP2 were supplied by Dr. R. H. Andreatta, Ciba-Geigy, Basel, Switzerland.
CHP1
6 7 8 9 i0 ii cyclo(Phe-Phe-D-Trp-Lys-Thr-Phe)
CHP2
6 7 8 9 i0 ii cyclo(Pro-Phe-D-Trp-Lys-Thr-Phe)
CPP1
7 8 9 i0 eyclo[Aha-Phe-D-Trp-Lys-Thr(Bzl)]
CPP2
7 8 9 I0 eye Io (Aha-Phe-D-Trp-Lys-Th r)
Experimental Protocols The assessment of the inhibitory action of a range of doses of somatostatin and analogues was based on the percent inhibition of gastric acid and pepsin secretions stimulated by pentagastrin, 8 ~g kg-' h r - ' . These methods are reproducible and have been described in detail [1, 4-7]. The control response to pentagastrin alone was determined in each animal. In experiments with somatostatin or analogues, pentagastrin was infused alone for 45 rain, and then increasing doses of somatostatin or analogue were added to the pentagastrin infusion for subsequent 30 rain periods. The acid output in the 15 min immediately before infusion of somatostatin differed by less than 10% that in the control experiment (see Table 1). The inhibition of acid and pepsin output produced by each dose of somatostatin analogue was calculated by subtracting the output during the second 15 rain of infusion of each dose from the control response at the same period of time in the same animal, and expressed as a percentage of the control response. The antagonistic activity of CPP1 was investigated by in-
FIG. I. Primary structure of short-chain cyclic somatostatin analogues. Numbering of amino acid residues refers to native somatostatin-14. Aha=7-aminoheptanoyl, Bzl=benzyl.
fusing it during somatostatin-inhibition of gastric secretion stimulated by pentagastrin. Pentagastrin, 8 ~g kg- ' h r - ' , and somatostatin, 2 or I0/zg kg- ' h r - ' , were infused together for 2.5 hr. CPPi was added to this infusion for the middle 1 hr. Significance of results was estimated by analysis of variance. RESULTS AND DISCUSSION The two cyclic hexapeptides (CHPI and 2) and the cyclic pentapeptide (CPP2) were very weak inhibitors of gastric
SOMATOSTATIN A N A L O G U E S : GASTRIC EFFECTS secretions in comparison to the native peptide. Somatostatin is a potent inhibitor of gastric acid secretion with an IDa,, of 1.2-1.7 nmol kg -~ hr ' [4.6] (and see Table l). Pepsin secretion stimulated by pentagastrin is about 5-times more sensitive to inhibition by somatostatin (Table i and [4,6]). CHPI constitutes cyclised residues 6-11 of [D-Trp"lsomatostatin (Fig. 1). This peptide showed no significant inhibition of acid secretion at molar doses up to 50-times the ID.~, for somatostatin (Table l). At the highest dose, significant inhibition of pepsin secretion was achieved (Table I). Thus the potency of CHPI in the stomach is approximately 0.005 compared with somatostatin (1.0). In comparison, CHP1 has a relatively greater (0.27) potency against GH release in the rat [10l. The very low activity of CHPI in the stomach may be partially explained by the substitution of the cystine disulphide bridge with a peptide link, since similar replacement in the full ring structure yields an analogue with a potency of 0.22 [5]. However, we have previously reported that a cyclic octapeptide consisting of residues 6-11 (with D-Trp ~ as CHPI) but cyclised through a cystine link (cyclo[Cys-Phe-Phe-D-TrpLys-Thr-Phe-Cys]) was also inactive (potency<0.05) against gastric secretions I6]. Substituting the Phe" residue in CHPI for Pro to give CHP2 (Fig. 1) increased gastric inhibitory activity (Table 1). Similarly, GH release inhibitory activity is increased by the same modification [10]. The increased activity of the proline substituted hexapeptide has been argued to be the result of favourable conformational influences of this residue [10]. However, CHP2 demonstrated reduced activity against gastric secretions in the cat (Table 1), while showing a relative potency of 5-25, compared with somatostatin, as an inhibitor of GH, insulin and glucagon release in rats in vivo [10]. A five-fold range of doses (10-50/zg kg-' hr-') resulted in approximately 50% inhibition of acid, and approximately 85% inhibition of pepsin secretion (Table 1). This dose-unrelated relationship is unusual, and we have not observed it with any other somatostatin analogue [ 1,4--6]. It may be related to the overall reduced activity of the short-chain cyclic analogues, but we can offer no definite explanation at the present time. CPP2 consists of residues 7-10 of somatostatin cyclised through 7-amino-heptanoyi (Fig. 1). This peptide was inactive (potency<0.01) in the dose range studied (Table 1). CPP2 was also inactive as an inhibitor of GH release in rats [2]. CPPI, equivalent to CPP2 but with the benzyl protecting group on T h r " retained (Fig. 1), has been described as an antagonist of somatostatin with respect to its actions against GH, insulin and glucagon release in the rat [2]. In the stomach, CPPI in doses ranging from 2-50/~g-' hr -~ did not alter somatostatin inhibition of acid or pepsin secretion. For example, somatostatin 2 p.g kg-' hr -~ almost completely inhibited pepsin secretion stimulated by laentagastrin, with a less marked effect on acid secretion. Addition of CPPI at 50 /zg kg -~ hr -~ (60-fold molar excess compared with somatostatin) did not alter the gastric secretions (Fig. 2). Similarly, CPPI at 50 p.g kg -1 hr -1 ( 10-fold molar excess) failed to influence the more marked inhibition of gastric secretion produced by somatostatin 10/xg kg -~ hr-J (Fig. 3). Doses of CPPI of 6--10/zg kg -x were sufficient to antagonise the actions of endogenous somatostatin on GH, insulin and glucagon release in rats, while molar excesses of 2-10-fold antagonised the effects of exogenous somatostatin against these secretions [2]. The lack of antagonistic activity of CPPI in the stomach, although disappointing, may not be surprising when the reduced activity of the cyclic hexapeptide analogues in this organ is considered.
859
,Pentagastrin 8)ug kg-'h-' o A u ,Somatostotin 2)ug kg-'h-' e A ' CPPI 50jug kg-lh -1 • ' 600r
/b //
•
2oo 0
/7 "
I600 1400
~
./
3.S 1200 ~E 3 ~ I000 O~ C ~" 800
"~_~
a~
600:
£L ~, 4 0 0 200 t
0
I
30
I
60
I
90
I
t
120
150
Time (rain) FIG. 2. Failure of CPP1 to influence somatostatin-inhibition o~ pentagastrin-stimulated gastric secretions. Experiments are illustrated as pentagastrin, 8 ~g kg-~ hr -R, infused alone for 150 min (©), pentagastrin, 8 p.g kg-' hr-', infused together with somatostatin, 2 ~.g kg-~ hr-~, for 150 min (A), and CPP1, 50 p.g kg-~ hr-~, infused during the middle hour of an infusion of pentagastrin, 8 ~g kg-~ hr-~, plus somatostatin, 2/zg kg-~ hr-j (0). Values illustrated as mean 15 acid or pepsin outputs (n--6).
The data reported here illustrate the marked differences in the structural requirements for somatostatininhibition of feline gastric exocrine secretions compared with rat endocrine secretions. Short-chain cyclic analogues of somatostatin, while retaining marked inhibitory activity against pituitary and pancreatic secretions in the rat [9-11], show only low activity against fefine gastric secretions. The failure of the cyclic pentapeptide CPP1 to act as a somatostatin antagonist in the stomach exemplifies these differences. Comparisons of the gastric and endocrine activities of these somatostatin analogues are, however, complicated by several experimental differences. The gastric studies were carried out in cats, while the endocrine studies in rats. However, short-chain bicyclic analogues, although not entirely analogous to the short-chain analogues described here, also show reduced gastric activity in the dog [11]. Somatostatin and analogues were administered by continuous infusion of the peptides in the gastric studies, compared with bolus injections in the pituitary and pancreatic studies.
860
HIRST, A R I L L A , COY AND SHAVv
, Pentagostrin BAg kg'lh -t I
o,
Somatostatin lO~g kg-~h -~ ,' CPPI 5 0 ~ g kg-lh -1 •'
600 .c_ 500 ~Se ~ 400 o
o~ 300
~- :~00 ,oo
3
o
1600 1400 / 12oo ,ooo
,- ~
800
U'J ..~
0L I
0
I
30
I
60
I
90
I
120
This experimental difference may be of particular importance when peptides of differing duration of action are being compared. Indeed, the duration of action of CHP2 in inhibiting GH release in the rat after SC or oral administration is prolonged compared with somatostatin [10]. However, increased duration of action of these analogues would be expected to increase their potency compared with somatostatin, which contrasts to the observed decrease in potency. A third difference between the gastric and endocrine studies is the degree to which the various secretions are stimulated. Gastric acid secretion was stimulated at near-maximal rates by pentagastrin [I], while the endocrine secretions were either unstimulated, or only generally stimulated by anaesthesia. However, gastric pepsin was secreted at only approximately one-third of the maximum rate [l], but the shortchain cyclic analogues also demonstrated low activity against this secretion, as was seen for acid secretion. In summary, comparison of the gastric potency of the somatostatin analogues with their reported potency against endocrine secretions is not simple. However, considering that in each system somatostatin is included as an internal standard, such comparisons are useful, and the general conclusion that short-chain cyclic analogues are relatively inactive in the stomach compared with the pituitary and pancreas appears valid. We conclude that our results suggest that residues outside the central 6- I 1 region of somatostatin are more important for gastric than pituitary or pancreatic somatostatin receptors. The lack of antagonistic activity of CPPI in the stomach indicates that these residues are more likely to be involved in receptor recognition/binding rather than activation.
1
150 ACKNOWLEDGEMENTS
Time (min) FIG. 3. Failure of CPP! to influence somatostatin-inhibition of pentagastrin-stimulated gastric secretions. Exper/ments illustrated as pentagastrin, 8 tzg kg-' hr-', infused alone for 150 rain (©), and ~ n tagastrin, 8 ~g kg-' hr-t, infused with somatostatin, 10 ~g kg-t hr-', for 150 min and CPPI, 50 t~g kg-~ hr-t, added during the middle hour (0). Values illustrated as mean-'-1 s.c. mean (n=6).
We thank Ken EUiott and David Stephenson for technical assistance, and Dr. R. H. Andreatta (Ciba-Geigy, Basel, Switzerland) for generously supplying CHP 1 and 2. This work was supported in part by NIH grant AM-18370. E. A. was supported by a grant from the British Council.
REFERENCES 1. Brown, M. P., D. H. Coy, A. Gomez-Pan, B. H. Hirst, M. Hunter, C. Meyers, J. D. Reed, A. V. Schally and B. Shaw. Structure-activity relationships of eighteen somatostatin analogues on gastric secretion. J Physiol 277: 1-14, 1978. 2. Fries, J. L., W. A. Murphy, J. Sueiras-Diaz and D. H. Coy. Somatostatin antagonist analog increases GH, insulin and glucagon release in the rat. Peptides 3:811-814, 1982. 3. Gomez-Pan, A., J. D. Reed, M. Albinus, B. Shaw, R. Hall, G. M. Besser, D. H. Coy, A. J. Kastin and A. V. Schally. Direct inhibition of gastric acid and pepsin secretion by growth hormone-releasing inhibiting hormone in cats. Lancet h 888890, 1975. 4. Hirst, B. H., J. M. Conion, D. H. Coy, J. Holland and B. Shaw. Comparison of gastric exocrine inhibitory activities and plasma kinetics of somatostatin-28 and somatostatin-14 in cats. Regul Pep 4: 22%237, 1982. 5. Hirst, B. H., J. D. Reed, B. Shaw, C. F. Hayward and J. S. Morley. Non-reducible cyclic, and azaphenylalanyl" analogues of somatostatin. Fur J Pharmacol 65: 151-156, 1980. 6. Hirst, B. H., B. Shaw, C. A. Meyers and D. H. Coy. Structure-activity studies with somatostatin: the role of tryptophan in position 8. Regul Pep h 97-113, 1980.
7. Reed, J. D., B. H. Hirst, A. Gomez-Pan, D. H. Coy, A. V. Schally and C. Meyers. Inhibition of gastric secretion by stereoisomers of somatostatin. Metabolism 27:1411-1413, 1978. 8. Shaw, B. and C. L. Wright. The pepsinogens of cat gastric mucosa and the pepsins derived from them. Digestion 4: 142152, 1976. 9. Vale, W., C. Rivier, M. Brown and J. Rivier. Pharmacology of thyrotropin releasing factor (TRF), luteinizing hormone releasing factor (LRF), and somatostatin. In: Hypothalamic Peptide Hormones and Pituitary Regulation, edited by J. C. Porter. New York: Plenum Press, 1977, pp. 123-156. 10. Veber, D. F., R. M. Freidinger, D. S. Periow, W. J. Paleveda, F. W. Holly, R. G. Strachan, R. F. Nutt, B. H. Arison. C. Homnick, W. C. Randall, M. S. Glitzer, R. Saperstein and R. Hirschmann. A potent cyclic hexapeptide analogue of somatostatin. Nature 292: 55-58, 1981. 11. Veber, D. F., F. W. Holly, W. J. Paleveda, R. F. Nutt. S. J. Bergstrand, M. Torchiana. M. S. Glitzer, R. Saperstein and R. Hirschmann. Conformationally restricted bicyclic analogs of somatostatin. Proc Nat/Ac'ad Sci USA 75: 2636-2640, 1978.