- Email: [email protected]
The metabolic effects of oxytocin are mediated by a uterine type of receptor and are inhibited by oxytocin antagonist and by arginine vasopressin in the dog
Life Sciences, Vol. 50, pp. 739-746 Printed in t h e U S A
Pergamon P r e s s
THE METABOLIC EFFECTS OF OXYrOCIN ARE M~W~IATEDBY A UTERINE TYPE OF RECEPTOR AND ARE INHIBITED BY OXYT(~IN ANTAGONIST AND BY ARGININE VASOPRESSIN IN THE DOG. N. Altszuler I , C~R. Rosenberg 2, B. Winkler 3 , A.R. Fuchs 4, P ~ . Hill5 and V~I. Hruby 5. 1Departments of Pharmacology and 2Environmental Medicine, New York University School of Medicine, New York, NY 10016, 3Department of Biological Sciences, Q u e e ~ b o r o u g h Community College, Bayside, NY 11363, 4Department o f Obstetrics and Gynecology, Cornell Medical School, New York, NY 10021 and 5Department of Chemistry, University of Arizona, Tucson, AZ 85721. (Received in final form January 7, 1992)
Summary Infusion of oxytocin (OT) into normal dogs,in doses which produced plasma levels of O T in the physiological range, has been shown to increase plasma levels of glucose, insulin and glucagon and increase rates of glucose production and uptake. This study sought to determine whether there was a correlation between these metabolic effects and the oxytocic potency of four less potent oxytocic analogues when infused into normal dogs. The rank order of oxytocic potency of all 4 correlated well with the rise in plasma glucose levels, and in 3 of the 4 with the rise in plasma insulin levels. A n antagonist of the oxytocic effect of O T suppressed the usual OT-induced rise in plasma glucose, insulin and glucagon as well as the increased glucose production and uptake. Arginine vasopressin (AVP) infusion,which by itself did not produce any metabolic effects, blocked completely the effects of O T infusion to raise plasma glucose and insulin levels and increase glucose production and uptake. The data suggest that the metabolic effectsof O T in the dog are mediated by O T receptors that are similar to those producing the oxytocic effects. Whether the inhibition by A V P of the metabolic and hormonal effects of O T occurs at the receptor or post receptor level or via other mechanisms remains to be determined. W e previously reported that infusion of oxytocin (OT) into normal dogs increased plasma levels of glucose, insulin and glucagon and increased rates of glucose production and overall glucose uptake by tissues (1). Since only one type of O T receptor has been described, the present study was undertaken to determine whether the metaboiic effectsof O T are also mediated by a uterine-type O T receptor. A number of O T analogues of varying oxytocic potency and a potent antagonist to O T were studied with regard to their effects on plasma glucose and insulin levels. Since the uterine O T receptor has been shown to display considerable cross-reactivitywith V P in a number of species (2,3),the effects of V P infusion on the responses to O T were also investigated. Corresponding author: Dr. Norman Altszuler, Department of Pharmacology, N e w York University School of Medidne, 550 F~st Avenue, N e w York, N Y 10016 0024-3205/92 $5.00 + .00 Copyright © 1992 Pergamon Press plc A l l r i g h t s r e s e r v e d .
740
A V P Inhibits Metabolic
Effects of OT
Vol.
50, No.
I0, 1992
MateriM~ and Methods All experiments were performed on conditioned, unanesthetized normal dogs of either sex, weighing 18-25 kg. The dogs were maintained on a modified Lusk diet, wherein protein provided 23% of the calories, 38% came from carbohydrates and 39% from fat; all experiments were begun about 18 hrs following feeding. Serial blood samples were obtained from the jugular vein through a polyethylene tube inserted percutaneously through a needle shortly before each experiment. The tube was kept patent by infusion of saline, which also served to replace the fluid volume. Drugs and tracer glucose were infused into the saphenous vein via a polyethylene tube inserted as described above. Blood samples were collected in heparinized syringes, transferred to chilled tubes, centrifuged and the plasma drawn off. Aliquots of plasma were frozen for later analysis of insulin, glucagon and glucose. In experiments using radioactive glucose, a separate aliquot of plasma was deproteinized with ZnS04 and Ba(OH)2 and the supornatant frozen for later analysis of glucose concentration and specifkc activity. Glucose production and overall glucose uptake were determined, using (6-~H) glucose, which was administered as apriming injection (55 ttC), along with a continuous infusion at a constant rate (0.5 ~tC/min). Rates of glucose production and uptake in the steady state were calculated as described before (4). During periods of acute changes in plasma glucose levels, the rates of glucose production and uptake were calculated on the basis of the rapidly mixing glucose pool being 0.7 of the initial glucose pool (4). Plasma and supernatant glucose concentrations wer~ determined by the glucose oxidase method, using the Beckman Glucose Analyzerr~ and the specific activity of the glucose was determined as described before(4). OT was determined with a highly specific RIA after extraction of acidified plasma samples using minicolumns(5). Insulin was determined by radioimmunoassay(6), using porcine insulin as the standard and dextran coated charcoal to separate the free insulin from that bound to the antibody(7) and glucagon was determined by immunoassay(8) using the Unger 30 K antibody. The coefficient of variation for insulin was + 7 % for intra-assay and + 10 % for interassay; for glucagon it was + 7 % for intra-assay anal + 1 1 % for inter~ssay,. The four analogues of OT:oxypressin, mesototin, [D;Leu~].OT, and [~MpaI-Alal]OT, and the OT antagonist [Penl,Phe(-Me) ~', Thr 4, Orn~]OT were synthesized at the University of Arizona, Tucson(9,10). The latter produces a prolonged antagonism of the oxytocic effect of OT and has a pA2 of -7.3; in vivo 3-4 ~tg/rat causes a 50-80% inhibition of the uterine contractile response to 50-80mU/rat of OT (10). Abbreviations used: Mpa=mercaptp.propionic acid; Pen=penicillamine; Phe(Me)=4 -methyl-Phe; Oxytocin (Syntocinonl~) was a gift from Sandoz, Inc., East Hanover, N.J., and arginine vasopressin was obtained from Sigma Chemicals, St.
Louis, MO. The effect of peptide infusions on the concentrations of plasma glucose and insulin was calculated by means of the Simpson s rule from the area under the curve depicting the increments in plasma levels before and during the infusion period. Because levels had fallen to nearly baseline at the end of the infusions, these values represent the total response. The results are shown as mean+SEM. The data were analyzed using two factor mixed ANOVA(ll). Differences between individual pairs of means were analyzed using Tukey's test for comparison between treatment groups and Dunnett s test for differences within each treatment group(12). All statistical computations were done using SAS/STAT User's Guide, Release 6.03 Edition, SAS Institute Inc, Cary, NC;1988. Results The oxytocic potency of the analogues used are shown in Table I together with the responses they elicited on plasma glucose and insulin levels during a 40 rain infusin at l~tg/kg/min. The actual plasma values are shown in Fig 1. The oxytocic potency was positively correlated with the plasma glucose responses to each analogue (r=0.907;p <0.001). This was also the case with the plasma insulin responses,
Vol. 50, No. 10, 1992
AVP Inhibits Metabolic Effects of OT
741
although the correlation ~as n~t as good, (r=0.429;p<0.05) due to the unexpectedly large response to [~-Mpa -Ala ]OT. O T itself produces those effects at picomolar doses (see Fig 2) rather than the nanomolar doses needed for the O T analogues, and therefore values for the O T effects are not included in Table I or Fig 1. TABLE I PLASMA GLUCOSE AND INSUI~N RESPONSES TO VARIOUS ANALOGUES OF OXYTOCIN(OT)a IN NORMAL DOGS.
Compound
Relative oxytocic potency b
Plasma glucose responsesC, d mg/dL/40 sin.
Plasma insulin responsesC, e ttU/ml/40 min.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. Mesotocin 2. [~-Mpa I, AIa7]OT f 3. Oxypressin 4. [D-Leu8]OT 5. Vehicle (saline)
50% 2(~o 15% 4% 0%
186~ 1134_~46g 928~151g 71Y_24h 5~1 h
996~179g 2187±143 558~162g 92~19 h 3+1 h
a) All infused at 1 ttg/kg/min for 40 min. b) Potency related to OT=I00%; Ref for I (23),2(24),3 (25),4 (26). c) Values are M e a n + SEM, n=5, for responses calculated according to Simpson's rule. d) Glucose response vs. potency: r=0.907, p<0.001. e) Insulin response vs. potency: r=0.429, p<0.05. f) [1-~-Mercaptopropionic acid,7-Alanine]OT. g, h) Comparison of all group means within each vertical column showed that those with same superscript are not significantly different,but all others are, p<0.001 ( A N O V A and Newman-Keuls). Infusion of the O T antagonist alone for 70 sin at 20ng/kg/min had no effect on plasma glucose, insulin and glucagon levels nor on glucose production or uptake(data not shown). As shown in Table II, when the antagonist was infused for 30 sin before the O T infusion was begun, in order to achieve inhibitory levels, it had no effect on any of the parameters. The antagonist did suppress the responses to O T by 40-70% compared to the responses to the unopposed O T infusion, but did not block the responses completely (Table III). This was not unexpected since the Ki for the antagonist is ~ 5 xl0 "8 while the K d for O T is ~ 0.5 x 10 "10 and the blood levels of the antagonist at the dose used was probably in the range of I to 10 pmoles, thus well below the Ki. The suppression of the insulin response was most marked. Arginine vasopressin(AVP) infused at 250 ttU/kg/min (0.6 picomoles/kg/min) had no effect on plasma glucose, insulin or glucagon levels (1), nor on glucose turnover(unpublished data). However, infusion of A V P at this rate completely prevented the effects of OT(500ttU or 0.84 picomoles/kg/min) on plasma insu|in levels and o n glucose production and uptake (Fig. 2). The plasma O T levels achieved with infusions of 500ttU/kg/min were measured in 5 other dogs of similar body weight; the steady state levels of 26.4+3.0 ~tU/ml (44 fentomoles/ml) were attained in 15-30 sin.
742
AVP Inhibits
Metabolic
Effects
of OT
Vol.
50, No.
i0,
1992
TABLE II EFFECT OF OXYTOCIN (OT) INFUSED ALONE AND SUPERIMPOSED ON INFUSED ANTAGONIST (A+OT) IN NORMAL DOGS ON PLASMA LEVELS OF GLUCOSE, INSULIN, GLUCAGON, GLUCOSE PRODUCTION AND UPTAKE. Response measured
Compound
Time After OT Infusion Begun, min.
OT a
0
Ab+OT .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Plasma glucose mg/dL
.
.
.
.
.
.
.
-30
.
.
.
.
.
.
.
.
.
.
.
-15
.
.
.
.
.
.
.
.
.
.
0
.
.
.
.
.
.
.
.
OT A+OT
Plasma insulin ~U/ml
95+_2
94!2
OT
.
.
6-+1
.
.
.
.
.
.
.
.
.
.
.
.
30
.
.
.
.
.
.
.
.
.
.
.
.
.
60
45
.
.
.
.
.
106+3f
105+_2 f
93+0.
102+3
10~2
.
.
.
.
.
.
.
.
.
60 .
.
.
.
.
.
.
.
.
.
.
.
.
.
106+3f
106+3f
98+_2
97+2
.
29-+3d, f
5_+1
12_+2f
10+_2f
83+8
262_+20 f
348+38f
278_+19f 265+_29f
269-+52 f
256-+30f
265_+38f 240-+52f
147+_20 17+1Y28 157-+19
Glucose OT production g m / m 2 / h r A+OT
3.8-+0.1 4.1_+0.2 4.4-+0.6 3.7_+0.3 3.8+0.1 4.1-+0.2 4.8-+0.5 4.1_+0.4
.
.
.
28+3d,f 24+2d, f
46_+5c,f
A+OT
Glucose OT uptake g m / m 2 / h r A+OT
.
45
93+_2
6-+1
0T
30
15 .
10-+1
A+OT
Pl a sm a glucagon pg/ml
15
8_+1
7+1
8.8+0.6 ef 8.4-+0.4f
6.4_+0.3f 6.3_+2f
5.7+0.3 f
5.2_+0.5f
5.5+0.8 f
4.6_+0.4 9.1-+1.7e,f 7.0-+0.2f
5.6+0.1 f
6.4-+1.2f
4.4-+0.4 6.2_+1.5f
5.8_+0.5f 5.2_+0.9
All values are Means +S.E.M.; n=10. a) Oxytocin infused at 500~tU = lng/kg/min. b) Antagonist [Pen 1, Phe(Me) 2, Thr 4, Orn 8] OT infused at 20 ng/kg/min., infused alone from -30 to 0 and then with OT from 0 to 60. c) Differences between the two groups significant at p<0.001, d) Differences between the two groups significant at p<0.01, e). Differences between the two groups significant at p<0.05-0.01, f) Significantly different from own control, p<0.05 - 0.01.
TABLE III INTEGRATED INCREMENTS a OVER INITIAL LEVELS DURING THE 60 MIN. INFUSION PERIOD OF OXYTOCIN (OT) AND ANTAGONIST (A+OT). Response .
.
.
.
.
.
.
.
.
.
.
.
.
OT .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
P l a s m a glucose
mg/dL
Plasma insulin ~tU/ml Plasma glucagon g/ml luco~,e p r o d u c t i o n gm/mZ/hr Gluco~,e u p t a k e gm/mZ/hr
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
A + OT .
.
.
.
.
.
.
.
.
.
.
695!-_167b 1,352~187
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
P .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
393_+98
N.S.
247292
<0.001
11,626d=1112
5,219kl,619
<0.02
215i10.5
99+_38
<0.02
145_+_8.4
69-~.1
<0.001
a) D a t a from Table II, n=10.
b) Values are M e a n _+_SEM; statistical a n a l y s i s u s i n g s t u d e n t ' s T test.
.
.
.
.
Vol. 50, No. 10, 1992
AVP Inhibits Metabolic Effects of OT
lpGIKGIMIN 0 I,
10
20
170
150
130
A
-
/// ~2 .
.
/~'/ ,~
GLUCOSE
MESOTOCIN
/i.,
110
40
3O
~
MG/DL
.
.
.
.
.
.
.
'7'..-:.o-.--._ I O X Y P R E S S I N .... "'"'" ( 1 - ~ - M p a - 7 - A l a ) - O T
_ _ .,,.,..¢._.,. _ - ~
- -
'
- - --.-J
8-D-Leu-OT
90
uUIML 90 ° °.°°..-.'"°
I
1 "...
:"i
70
..........1\
. .: 50
.."
INSULIN
PLASMA
.... -
"... "'..: L
/ '1"° ' '
( 1-,B-Mpa-7-AlaI-OT
30
MESOTOCIN "~ O X Y P R E S S I N
10
"~___.._ _ . - J . . . . . I
I
0
10
L------
-'
J
I
I
I
20
30
40
8-D-Leu-OT
M I N U T E S
FIG. 1. Effects of infusion of OT analogues, each at 1 Ilg/kg/min, in 6 normal u n a n e s t h e t i z e d dogs. All analogues,except [D-Leu8]OT, produced significant increases in plasma glucose and insulin levels, but differed in rank order of potency. Mesotocin produced the largest glucose response as judged by the area under the curve(see Table I) and [~-Mpal,Ala7]OT produced the greatest insulin response.
743
744
AVP Inhibits
Metabolic
I }
Effects of OT
Vol.
VASOPRESSIN2501J.U/KG/MIN OXYTOCIN 500p.U/KG/MIN
50, No.
I0, 1992
I I PLASMA GLUCOSE
MG/DL 120
f ....... :j. j__
ovP÷o, T
I00 ~U/ML
50
I
lr
PLASMA INSULIN
30 I0
MGIKG/HR 400 200 0 MG/KG/HR 400 200
.....................................
f f
/
GLUCOSE PRODUCTION T
. . . . . . .
- .
.
.
.
1"
I,
.
" . . . . . . .
,
°
--.T__
0
" VP+OT
|
-L._
I
I
-30
0
I
T
._r .
.I
T
_ .r_
_
i
1
"r
I
.....
.z
1
T
"-I-
I
30
"r
OT
- .....
~
VP÷OT
..T
!
"r
r
OT
vP+OT
.t
A
I
60
MINUTES
FIG. 2. Effects of infusion of OT (500~U or 840pg/kg/min) alone and superimposed on an infusion of AVP (250~tU or 610pg/kg/min) in 6 normal, unanesthetized dogs. Control values for glucose production and uptake were obtained in each dog in the 60 min period immediately prior to start of the hormone infusions, as described in text. Saline infusion and AVP alone had no effect on these m e a s u r e m e n t s . Insulin values were significantly different between the two groups at 15rain(p<0.001) and 30min(p<0.05). Rates of glucose production and uptake were significantly different (p<0.001) between the two groups at all post control points. ~
o
n
These studies indicate that the observed hormonal and metabolic effects of OT are most likely mediated by OT receptors t h a t are similar to those mediating its uterine effects. This is based on the capability of the antagonist of the oxytocic effects of OT to suppress the rises in plasma glucose and glucagon and rises in the rates of glucose production and uptake. This is also in keeping with our findings t h a t the
Vol.
50, No. i0, 1992
AVP Inhibits Metabolic Effects of OT
745
rank order of oxytocic potency of the several O T analogues correlated well with the rank order of their effectson plasma glucose and insulin levels. The fact that the O T antagonist suppressed plasma insulin levels (by 80%) to a greater extent than plasma glucagon and glucose levels (55% and 44%, respectively)supports our contention (13) that O T has a direct effect on insulin secretion in addition to the indirect ones mediated by the OT-induced elevations in plasma glucagon and glucose levels. A noteworthy finding was that AVP completely inhibited the OT-induced increases in plasma insulin levels and rates of glucose production and uptake when infused at about equimolar concentrations. Lysine VP was also reported to inhibit OT-induced insulin release, but not glucagon release in normal dogs (14). The mechanism for this blocking action is unknown. There have not been any reports that VP inhibits oxytocic activity of OT in the usual in vitro tests although OT and VP display extensive cross-reactivity at their uterine and ovarian receptors (2,3). However, biologic antagonistic activity between the two hormones has been reported with regard to both the antidiuretic and vascular responses in vivo(15). Excess of VP suppressed uterine responses to OT (16) and equimolar or greater concentrations of VP caused appreciable suppression of milk ejection responses to OT (17), in both cases after a brief initial potentiation. The authors of the foregoing studies concluded that a vasoconstricting effect of VP was the most likely explanation for the antagonism of the OT effects, r a t h e r than competitive antagonism at the receptor level. Whether vasoconstriction by VP might also be implicated in its antagonistic effects in the present study is uncertain. Rates of VP infusion like those used here have been shown to reduce blood flow to the pancreas, a likely site for the OT action, but had no effect on the blood flow to the liver(18). However infusion of VP alone did not decrease plasma insulin levels (1) and it did block the OT-induced increase in hepatic glucose output (Fig 1), both effects speaking against a role of vasoconstriction by VP in the present findings. The fact that infusion of AVP alone had no effect on the hormonal and metabolic parameters measured here while still exerting an inhibitory action is not unique and is analogous to cholinergic and adrenergic blocking agents which themselves produce little change but block the respective agonists. Another possibility to consider for the inhibitory effect of VP against OT is an interaction at the receptor level. The bovine liver contains VP receptors of the V1 type t h a t are at a greater concentration and have a higher affinity t h a n the low concentration,albeit high affinity, OT receptors (5). However, in the dog, much larger doses of VP (0.2-2.7 U/kg as bolus into the portal vein) than OT were needed to demonstrate a glycogenolytic response(19), indicating that species differences need to be considered in evaluating responsiveness to the metabolic effects to these hormones. Also, we have found that while OT stimulated insulin secretion by a hamster beta cell line(HIT), it required concentrations 100 fold greater than those for VP(20). In the rat pancreas, in vitro, AVP and OT were approximately equivalent in stimulating glucagon and insulin secretion(21), while in the dog pancreas AVP appeared to compete with OT binding but at a 10 fold lower affinity(22). The physiologic significance of the inhibitory effect of VP on the metabolic responses to OT remains to be established. Since the two hormones are frequently secreted simultaneously in response to various stimuli, the secreted VP may serve to moderate the peripheral metabolic effects of OT. In addition, the present study points to a possible usefulness of VP as a tool to explore the specificity of responses to OT in various tissues. Acknowledgement The studies were supported in part by a grant from the U P S Foundation (NA), from Warner Lambert Foundation(ARF) and by U.S. Public Health Service Grant D K 17420 (VJH).
746
AVP Inhibits
Metabolic
Effects
of OT
Vol.
50, No.
10,
1992
References
1. N. ALTSZULER and J. HAMPSHIRE, Diabetes 30 112-114(1981). 2. M. IVANISEVIC, O. BEHRENS, H.HELMER and A.R. FUCHS, Am. J. Obstet. Gynecol. 161 1637-1643 (1989). 3. A.R. FUCHS, O. BEHRENS, H. HELMER, M. IVANISEVIC, J. GRIFO, A. VANGSTED, C.M. BARROS and M.J. FIELDS, Am. J. Obstet. Gynecol. 163 19611967 (1990). 4. N. ALTSZULER, A. BARKAI, C. BJERKNESS, B. GOTTLIEB and R. STEELE, Amer. J. Physiol. 229 1662-1667 (1975). 5. O. BEHRENS, S. BOHMER, K. GOESCHEN and A.R. FUCHS, Obstet Gynecol.77 879-884 (1991). 6. R.S. YALOW and S. BERSON, J.Clin.Invest. 39 1157-1175 (1960). 7. V. HERBERT, K.S. LAW, C.W. GOTTLIEB and S.J. BLEICHER, J. Clin. Endocrinol. Metab. 25 1375-1384 (1965). 8. G.R. FALOONA and R.M. UNGER, Methods of Hormone Immunoassay, B.M.JAFFE and H.R. BERMAN (eds),317-330, Academic Press, New York (1974). 9. V.J. HRUBY, H.I. MOSBERG, M.E. HADLEY, W.Y. CHAN and A.M. POWELL, Int. J. Peptide Protein Res 16 372-381 (1980). 10. W.Y. CHAN, V.J. HRUBY, T.W. ROCKWAY and J.HLAVACEK, J. Pharmacol. Exp. Therap. 239 84-87 (1986). 11. B.J. WINER, Statistical Principles in Experimental Design. 2nd Edition, McGraw-Hill, New York (1971). 12. J.H. ZAR, Biostatistical analysis, 2nd Edition, Prentice Hall, Englewood Cliff, N.J. (1984). 13. N. ALTSZULER, B. WINKLER, C.R. ROSENGERG, F.X. PI-SUNYER and A.R. FUCHS, Proc. Soc. Exptl. Biol. Med. (in press); 1992 (Feb). 14. B. S. STOCK and K. UVNAS-MOBERG, Acta Physiol.Scand. 130 55-61 (1987). 15. J. RUDINGER and I. KREJCI, In Berde, B. Handbook of Experimental Pharmacology XXIII 718-780 (1968). Springer Verlag, Heidelberg, Berlin,NY. 16. A.R. FUCHS, Acta Endocr (Cph) 45 272-280(1964). 17. S. KULLANDER, Acta Endocr(Cph) 44 313-324(1963). 18. A.W. COWLEY and B.J. BARBER, In: Cross, BA and Leng, G, Eds, The Neurohypophysis, Structure, Function and Control. Prog. Brain Res. vol 60 415424 (1983). Elsevier Science Publ., BV Amsterdam-NY. 19. S.J. BERGEN,Jr., R. SULLIVAN, J.G. HILTON, S.W. WILLIS,Jr and T.B. VAN ITALLIE,Am.J. Physiol. 199(1) 136-138(1960). 20. S.B. RICHARDSON, N. EYLER, S. TWENTE, M. MONACO, N. ALTSZULER and M. GIBSON, Endocrinology 126 1047-1052 (1990). 21. B.E. DUNNING, J.H. MOLTZ and C.P. FAWCETT, Am. J. Physiol 246(9) E108-114 (1984). 22. S. STOCK, J. FASTBOM, E. BJORKSTRAND, U. UNGERSTEDT and K. UVNASMOBERG, Regulatory Peptides 30 1-13 (1990). 23. P.A. JAQUENOD and R.A. BOISSONNAS, Helv. Chim. Acta 44 113-122 (1961). 24. R. WALTER, C.W. SMITH and J. ROY, Proc. Natl. Acad. Sci. USA 73 3054-3058 (1976). 25. C.H. SCHNEIDER and V. duVIGNEAUD, J. Am. Chem. Soc. 84 3005-3008 (1962). 26. R. WALTER, V.W. SMITH, J. ROY and A. FORMENTO, J. Med. Chem. 19 822825 (1976).
Pergamon P r e s s
THE METABOLIC EFFECTS OF OXYrOCIN ARE M~W~IATEDBY A UTERINE TYPE OF RECEPTOR AND ARE INHIBITED BY OXYT(~IN ANTAGONIST AND BY ARGININE VASOPRESSIN IN THE DOG. N. Altszuler I , C~R. Rosenberg 2, B. Winkler 3 , A.R. Fuchs 4, P ~ . Hill5 and V~I. Hruby 5. 1Departments of Pharmacology and 2Environmental Medicine, New York University School of Medicine, New York, NY 10016, 3Department of Biological Sciences, Q u e e ~ b o r o u g h Community College, Bayside, NY 11363, 4Department o f Obstetrics and Gynecology, Cornell Medical School, New York, NY 10021 and 5Department of Chemistry, University of Arizona, Tucson, AZ 85721. (Received in final form January 7, 1992)
Summary Infusion of oxytocin (OT) into normal dogs,in doses which produced plasma levels of O T in the physiological range, has been shown to increase plasma levels of glucose, insulin and glucagon and increase rates of glucose production and uptake. This study sought to determine whether there was a correlation between these metabolic effects and the oxytocic potency of four less potent oxytocic analogues when infused into normal dogs. The rank order of oxytocic potency of all 4 correlated well with the rise in plasma glucose levels, and in 3 of the 4 with the rise in plasma insulin levels. A n antagonist of the oxytocic effect of O T suppressed the usual OT-induced rise in plasma glucose, insulin and glucagon as well as the increased glucose production and uptake. Arginine vasopressin (AVP) infusion,which by itself did not produce any metabolic effects, blocked completely the effects of O T infusion to raise plasma glucose and insulin levels and increase glucose production and uptake. The data suggest that the metabolic effectsof O T in the dog are mediated by O T receptors that are similar to those producing the oxytocic effects. Whether the inhibition by A V P of the metabolic and hormonal effects of O T occurs at the receptor or post receptor level or via other mechanisms remains to be determined. W e previously reported that infusion of oxytocin (OT) into normal dogs increased plasma levels of glucose, insulin and glucagon and increased rates of glucose production and overall glucose uptake by tissues (1). Since only one type of O T receptor has been described, the present study was undertaken to determine whether the metaboiic effectsof O T are also mediated by a uterine-type O T receptor. A number of O T analogues of varying oxytocic potency and a potent antagonist to O T were studied with regard to their effects on plasma glucose and insulin levels. Since the uterine O T receptor has been shown to display considerable cross-reactivitywith V P in a number of species (2,3),the effects of V P infusion on the responses to O T were also investigated. Corresponding author: Dr. Norman Altszuler, Department of Pharmacology, N e w York University School of Medidne, 550 F~st Avenue, N e w York, N Y 10016 0024-3205/92 $5.00 + .00 Copyright © 1992 Pergamon Press plc A l l r i g h t s r e s e r v e d .
740
A V P Inhibits Metabolic
Effects of OT
Vol.
50, No.
I0, 1992
MateriM~ and Methods All experiments were performed on conditioned, unanesthetized normal dogs of either sex, weighing 18-25 kg. The dogs were maintained on a modified Lusk diet, wherein protein provided 23% of the calories, 38% came from carbohydrates and 39% from fat; all experiments were begun about 18 hrs following feeding. Serial blood samples were obtained from the jugular vein through a polyethylene tube inserted percutaneously through a needle shortly before each experiment. The tube was kept patent by infusion of saline, which also served to replace the fluid volume. Drugs and tracer glucose were infused into the saphenous vein via a polyethylene tube inserted as described above. Blood samples were collected in heparinized syringes, transferred to chilled tubes, centrifuged and the plasma drawn off. Aliquots of plasma were frozen for later analysis of insulin, glucagon and glucose. In experiments using radioactive glucose, a separate aliquot of plasma was deproteinized with ZnS04 and Ba(OH)2 and the supornatant frozen for later analysis of glucose concentration and specifkc activity. Glucose production and overall glucose uptake were determined, using (6-~H) glucose, which was administered as apriming injection (55 ttC), along with a continuous infusion at a constant rate (0.5 ~tC/min). Rates of glucose production and uptake in the steady state were calculated as described before (4). During periods of acute changes in plasma glucose levels, the rates of glucose production and uptake were calculated on the basis of the rapidly mixing glucose pool being 0.7 of the initial glucose pool (4). Plasma and supernatant glucose concentrations wer~ determined by the glucose oxidase method, using the Beckman Glucose Analyzerr~ and the specific activity of the glucose was determined as described before(4). OT was determined with a highly specific RIA after extraction of acidified plasma samples using minicolumns(5). Insulin was determined by radioimmunoassay(6), using porcine insulin as the standard and dextran coated charcoal to separate the free insulin from that bound to the antibody(7) and glucagon was determined by immunoassay(8) using the Unger 30 K antibody. The coefficient of variation for insulin was + 7 % for intra-assay and + 10 % for interassay; for glucagon it was + 7 % for intra-assay anal + 1 1 % for inter~ssay,. The four analogues of OT:oxypressin, mesototin, [D;Leu~].OT, and [~MpaI-Alal]OT, and the OT antagonist [Penl,Phe(-Me) ~', Thr 4, Orn~]OT were synthesized at the University of Arizona, Tucson(9,10). The latter produces a prolonged antagonism of the oxytocic effect of OT and has a pA2 of -7.3; in vivo 3-4 ~tg/rat causes a 50-80% inhibition of the uterine contractile response to 50-80mU/rat of OT (10). Abbreviations used: Mpa=mercaptp.propionic acid; Pen=penicillamine; Phe(Me)=4 -methyl-Phe; Oxytocin (Syntocinonl~) was a gift from Sandoz, Inc., East Hanover, N.J., and arginine vasopressin was obtained from Sigma Chemicals, St.
Louis, MO. The effect of peptide infusions on the concentrations of plasma glucose and insulin was calculated by means of the Simpson s rule from the area under the curve depicting the increments in plasma levels before and during the infusion period. Because levels had fallen to nearly baseline at the end of the infusions, these values represent the total response. The results are shown as mean+SEM. The data were analyzed using two factor mixed ANOVA(ll). Differences between individual pairs of means were analyzed using Tukey's test for comparison between treatment groups and Dunnett s test for differences within each treatment group(12). All statistical computations were done using SAS/STAT User's Guide, Release 6.03 Edition, SAS Institute Inc, Cary, NC;1988. Results The oxytocic potency of the analogues used are shown in Table I together with the responses they elicited on plasma glucose and insulin levels during a 40 rain infusin at l~tg/kg/min. The actual plasma values are shown in Fig 1. The oxytocic potency was positively correlated with the plasma glucose responses to each analogue (r=0.907;p <0.001). This was also the case with the plasma insulin responses,
Vol. 50, No. 10, 1992
AVP Inhibits Metabolic Effects of OT
741
although the correlation ~as n~t as good, (r=0.429;p<0.05) due to the unexpectedly large response to [~-Mpa -Ala ]OT. O T itself produces those effects at picomolar doses (see Fig 2) rather than the nanomolar doses needed for the O T analogues, and therefore values for the O T effects are not included in Table I or Fig 1. TABLE I PLASMA GLUCOSE AND INSUI~N RESPONSES TO VARIOUS ANALOGUES OF OXYTOCIN(OT)a IN NORMAL DOGS.
Compound
Relative oxytocic potency b
Plasma glucose responsesC, d mg/dL/40 sin.
Plasma insulin responsesC, e ttU/ml/40 min.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. Mesotocin 2. [~-Mpa I, AIa7]OT f 3. Oxypressin 4. [D-Leu8]OT 5. Vehicle (saline)
50% 2(~o 15% 4% 0%
186~ 1134_~46g 928~151g 71Y_24h 5~1 h
996~179g 2187±143 558~162g 92~19 h 3+1 h
a) All infused at 1 ttg/kg/min for 40 min. b) Potency related to OT=I00%; Ref for I (23),2(24),3 (25),4 (26). c) Values are M e a n + SEM, n=5, for responses calculated according to Simpson's rule. d) Glucose response vs. potency: r=0.907, p<0.001. e) Insulin response vs. potency: r=0.429, p<0.05. f) [1-~-Mercaptopropionic acid,7-Alanine]OT. g, h) Comparison of all group means within each vertical column showed that those with same superscript are not significantly different,but all others are, p<0.001 ( A N O V A and Newman-Keuls). Infusion of the O T antagonist alone for 70 sin at 20ng/kg/min had no effect on plasma glucose, insulin and glucagon levels nor on glucose production or uptake(data not shown). As shown in Table II, when the antagonist was infused for 30 sin before the O T infusion was begun, in order to achieve inhibitory levels, it had no effect on any of the parameters. The antagonist did suppress the responses to O T by 40-70% compared to the responses to the unopposed O T infusion, but did not block the responses completely (Table III). This was not unexpected since the Ki for the antagonist is ~ 5 xl0 "8 while the K d for O T is ~ 0.5 x 10 "10 and the blood levels of the antagonist at the dose used was probably in the range of I to 10 pmoles, thus well below the Ki. The suppression of the insulin response was most marked. Arginine vasopressin(AVP) infused at 250 ttU/kg/min (0.6 picomoles/kg/min) had no effect on plasma glucose, insulin or glucagon levels (1), nor on glucose turnover(unpublished data). However, infusion of A V P at this rate completely prevented the effects of OT(500ttU or 0.84 picomoles/kg/min) on plasma insu|in levels and o n glucose production and uptake (Fig. 2). The plasma O T levels achieved with infusions of 500ttU/kg/min were measured in 5 other dogs of similar body weight; the steady state levels of 26.4+3.0 ~tU/ml (44 fentomoles/ml) were attained in 15-30 sin.
742
AVP Inhibits
Metabolic
Effects
of OT
Vol.
50, No.
i0,
1992
TABLE II EFFECT OF OXYTOCIN (OT) INFUSED ALONE AND SUPERIMPOSED ON INFUSED ANTAGONIST (A+OT) IN NORMAL DOGS ON PLASMA LEVELS OF GLUCOSE, INSULIN, GLUCAGON, GLUCOSE PRODUCTION AND UPTAKE. Response measured
Compound
Time After OT Infusion Begun, min.
OT a
0
Ab+OT .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Plasma glucose mg/dL
.
.
.
.
.
.
.
-30
.
.
.
.
.
.
.
.
.
.
.
-15
.
.
.
.
.
.
.
.
.
.
0
.
.
.
.
.
.
.
.
OT A+OT
Plasma insulin ~U/ml
95+_2
94!2
OT
.
.
6-+1
.
.
.
.
.
.
.
.
.
.
.
.
30
.
.
.
.
.
.
.
.
.
.
.
.
.
60
45
.
.
.
.
.
106+3f
105+_2 f
93+0.
102+3
10~2
.
.
.
.
.
.
.
.
.
60 .
.
.
.
.
.
.
.
.
.
.
.
.
.
106+3f
106+3f
98+_2
97+2
.
29-+3d, f
5_+1
12_+2f
10+_2f
83+8
262_+20 f
348+38f
278_+19f 265+_29f
269-+52 f
256-+30f
265_+38f 240-+52f
147+_20 17+1Y28 157-+19
Glucose OT production g m / m 2 / h r A+OT
3.8-+0.1 4.1_+0.2 4.4-+0.6 3.7_+0.3 3.8+0.1 4.1-+0.2 4.8-+0.5 4.1_+0.4
.
.
.
28+3d,f 24+2d, f
46_+5c,f
A+OT
Glucose OT uptake g m / m 2 / h r A+OT
.
45
93+_2
6-+1
0T
30
15 .
10-+1
A+OT
Pl a sm a glucagon pg/ml
15
8_+1
7+1
8.8+0.6 ef 8.4-+0.4f
6.4_+0.3f 6.3_+2f
5.7+0.3 f
5.2_+0.5f
5.5+0.8 f
4.6_+0.4 9.1-+1.7e,f 7.0-+0.2f
5.6+0.1 f
6.4-+1.2f
4.4-+0.4 6.2_+1.5f
5.8_+0.5f 5.2_+0.9
All values are Means +S.E.M.; n=10. a) Oxytocin infused at 500~tU = lng/kg/min. b) Antagonist [Pen 1, Phe(Me) 2, Thr 4, Orn 8] OT infused at 20 ng/kg/min., infused alone from -30 to 0 and then with OT from 0 to 60. c) Differences between the two groups significant at p<0.001, d) Differences between the two groups significant at p<0.01, e). Differences between the two groups significant at p<0.05-0.01, f) Significantly different from own control, p<0.05 - 0.01.
TABLE III INTEGRATED INCREMENTS a OVER INITIAL LEVELS DURING THE 60 MIN. INFUSION PERIOD OF OXYTOCIN (OT) AND ANTAGONIST (A+OT). Response .
.
.
.
.
.
.
.
.
.
.
.
.
OT .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
P l a s m a glucose
mg/dL
Plasma insulin ~tU/ml Plasma glucagon g/ml luco~,e p r o d u c t i o n gm/mZ/hr Gluco~,e u p t a k e gm/mZ/hr
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
A + OT .
.
.
.
.
.
.
.
.
.
.
695!-_167b 1,352~187
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
P .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
393_+98
N.S.
247292
<0.001
11,626d=1112
5,219kl,619
<0.02
215i10.5
99+_38
<0.02
145_+_8.4
69-~.1
<0.001
a) D a t a from Table II, n=10.
b) Values are M e a n _+_SEM; statistical a n a l y s i s u s i n g s t u d e n t ' s T test.
.
.
.
.
Vol. 50, No. 10, 1992
AVP Inhibits Metabolic Effects of OT
lpGIKGIMIN 0 I,
10
20
170
150
130
A
-
/// ~2 .
.
/~'/ ,~
GLUCOSE
MESOTOCIN
/i.,
110
40
3O
~
MG/DL
.
.
.
.
.
.
.
'7'..-:.o-.--._ I O X Y P R E S S I N .... "'"'" ( 1 - ~ - M p a - 7 - A l a ) - O T
_ _ .,,.,..¢._.,. _ - ~
- -
'
- - --.-J
8-D-Leu-OT
90
uUIML 90 ° °.°°..-.'"°
I
1 "...
:"i
70
..........1\
. .: 50
.."
INSULIN
PLASMA
.... -
"... "'..: L
/ '1"° ' '
( 1-,B-Mpa-7-AlaI-OT
30
MESOTOCIN "~ O X Y P R E S S I N
10
"~___.._ _ . - J . . . . . I
I
0
10
L------
-'
J
I
I
I
20
30
40
8-D-Leu-OT
M I N U T E S
FIG. 1. Effects of infusion of OT analogues, each at 1 Ilg/kg/min, in 6 normal u n a n e s t h e t i z e d dogs. All analogues,except [D-Leu8]OT, produced significant increases in plasma glucose and insulin levels, but differed in rank order of potency. Mesotocin produced the largest glucose response as judged by the area under the curve(see Table I) and [~-Mpal,Ala7]OT produced the greatest insulin response.
743
744
AVP Inhibits
Metabolic
I }
Effects of OT
Vol.
VASOPRESSIN2501J.U/KG/MIN OXYTOCIN 500p.U/KG/MIN
50, No.
I0, 1992
I I PLASMA GLUCOSE
MG/DL 120
f ....... :j. j__
ovP÷o, T
I00 ~U/ML
50
I
lr
PLASMA INSULIN
30 I0
MGIKG/HR 400 200 0 MG/KG/HR 400 200
.....................................
f f
/
GLUCOSE PRODUCTION T
. . . . . . .
- .
.
.
.
1"
I,
.
" . . . . . . .
,
°
--.T__
0
" VP+OT
|
-L._
I
I
-30
0
I
T
._r .
.I
T
_ .r_
_
i
1
"r
I
.....
.z
1
T
"-I-
I
30
"r
OT
- .....
~
VP÷OT
..T
!
"r
r
OT
vP+OT
.t
A
I
60
MINUTES
FIG. 2. Effects of infusion of OT (500~U or 840pg/kg/min) alone and superimposed on an infusion of AVP (250~tU or 610pg/kg/min) in 6 normal, unanesthetized dogs. Control values for glucose production and uptake were obtained in each dog in the 60 min period immediately prior to start of the hormone infusions, as described in text. Saline infusion and AVP alone had no effect on these m e a s u r e m e n t s . Insulin values were significantly different between the two groups at 15rain(p<0.001) and 30min(p<0.05). Rates of glucose production and uptake were significantly different (p<0.001) between the two groups at all post control points. ~
o
n
These studies indicate that the observed hormonal and metabolic effects of OT are most likely mediated by OT receptors t h a t are similar to those mediating its uterine effects. This is based on the capability of the antagonist of the oxytocic effects of OT to suppress the rises in plasma glucose and glucagon and rises in the rates of glucose production and uptake. This is also in keeping with our findings t h a t the
Vol.
50, No. i0, 1992
AVP Inhibits Metabolic Effects of OT
745
rank order of oxytocic potency of the several O T analogues correlated well with the rank order of their effectson plasma glucose and insulin levels. The fact that the O T antagonist suppressed plasma insulin levels (by 80%) to a greater extent than plasma glucagon and glucose levels (55% and 44%, respectively)supports our contention (13) that O T has a direct effect on insulin secretion in addition to the indirect ones mediated by the OT-induced elevations in plasma glucagon and glucose levels. A noteworthy finding was that AVP completely inhibited the OT-induced increases in plasma insulin levels and rates of glucose production and uptake when infused at about equimolar concentrations. Lysine VP was also reported to inhibit OT-induced insulin release, but not glucagon release in normal dogs (14). The mechanism for this blocking action is unknown. There have not been any reports that VP inhibits oxytocic activity of OT in the usual in vitro tests although OT and VP display extensive cross-reactivity at their uterine and ovarian receptors (2,3). However, biologic antagonistic activity between the two hormones has been reported with regard to both the antidiuretic and vascular responses in vivo(15). Excess of VP suppressed uterine responses to OT (16) and equimolar or greater concentrations of VP caused appreciable suppression of milk ejection responses to OT (17), in both cases after a brief initial potentiation. The authors of the foregoing studies concluded that a vasoconstricting effect of VP was the most likely explanation for the antagonism of the OT effects, r a t h e r than competitive antagonism at the receptor level. Whether vasoconstriction by VP might also be implicated in its antagonistic effects in the present study is uncertain. Rates of VP infusion like those used here have been shown to reduce blood flow to the pancreas, a likely site for the OT action, but had no effect on the blood flow to the liver(18). However infusion of VP alone did not decrease plasma insulin levels (1) and it did block the OT-induced increase in hepatic glucose output (Fig 1), both effects speaking against a role of vasoconstriction by VP in the present findings. The fact that infusion of AVP alone had no effect on the hormonal and metabolic parameters measured here while still exerting an inhibitory action is not unique and is analogous to cholinergic and adrenergic blocking agents which themselves produce little change but block the respective agonists. Another possibility to consider for the inhibitory effect of VP against OT is an interaction at the receptor level. The bovine liver contains VP receptors of the V1 type t h a t are at a greater concentration and have a higher affinity t h a n the low concentration,albeit high affinity, OT receptors (5). However, in the dog, much larger doses of VP (0.2-2.7 U/kg as bolus into the portal vein) than OT were needed to demonstrate a glycogenolytic response(19), indicating that species differences need to be considered in evaluating responsiveness to the metabolic effects to these hormones. Also, we have found that while OT stimulated insulin secretion by a hamster beta cell line(HIT), it required concentrations 100 fold greater than those for VP(20). In the rat pancreas, in vitro, AVP and OT were approximately equivalent in stimulating glucagon and insulin secretion(21), while in the dog pancreas AVP appeared to compete with OT binding but at a 10 fold lower affinity(22). The physiologic significance of the inhibitory effect of VP on the metabolic responses to OT remains to be established. Since the two hormones are frequently secreted simultaneously in response to various stimuli, the secreted VP may serve to moderate the peripheral metabolic effects of OT. In addition, the present study points to a possible usefulness of VP as a tool to explore the specificity of responses to OT in various tissues. Acknowledgement The studies were supported in part by a grant from the U P S Foundation (NA), from Warner Lambert Foundation(ARF) and by U.S. Public Health Service Grant D K 17420 (VJH).
746
AVP Inhibits
Metabolic
Effects
of OT
Vol.
50, No.
10,
1992
References
1. N. ALTSZULER and J. HAMPSHIRE, Diabetes 30 112-114(1981). 2. M. IVANISEVIC, O. BEHRENS, H.HELMER and A.R. FUCHS, Am. J. Obstet. Gynecol. 161 1637-1643 (1989). 3. A.R. FUCHS, O. BEHRENS, H. HELMER, M. IVANISEVIC, J. GRIFO, A. VANGSTED, C.M. BARROS and M.J. FIELDS, Am. J. Obstet. Gynecol. 163 19611967 (1990). 4. N. ALTSZULER, A. BARKAI, C. BJERKNESS, B. GOTTLIEB and R. STEELE, Amer. J. Physiol. 229 1662-1667 (1975). 5. O. BEHRENS, S. BOHMER, K. GOESCHEN and A.R. FUCHS, Obstet Gynecol.77 879-884 (1991). 6. R.S. YALOW and S. BERSON, J.Clin.Invest. 39 1157-1175 (1960). 7. V. HERBERT, K.S. LAW, C.W. GOTTLIEB and S.J. BLEICHER, J. Clin. Endocrinol. Metab. 25 1375-1384 (1965). 8. G.R. FALOONA and R.M. UNGER, Methods of Hormone Immunoassay, B.M.JAFFE and H.R. BERMAN (eds),317-330, Academic Press, New York (1974). 9. V.J. HRUBY, H.I. MOSBERG, M.E. HADLEY, W.Y. CHAN and A.M. POWELL, Int. J. Peptide Protein Res 16 372-381 (1980). 10. W.Y. CHAN, V.J. HRUBY, T.W. ROCKWAY and J.HLAVACEK, J. Pharmacol. Exp. Therap. 239 84-87 (1986). 11. B.J. WINER, Statistical Principles in Experimental Design. 2nd Edition, McGraw-Hill, New York (1971). 12. J.H. ZAR, Biostatistical analysis, 2nd Edition, Prentice Hall, Englewood Cliff, N.J. (1984). 13. N. ALTSZULER, B. WINKLER, C.R. ROSENGERG, F.X. PI-SUNYER and A.R. FUCHS, Proc. Soc. Exptl. Biol. Med. (in press); 1992 (Feb). 14. B. S. STOCK and K. UVNAS-MOBERG, Acta Physiol.Scand. 130 55-61 (1987). 15. J. RUDINGER and I. KREJCI, In Berde, B. Handbook of Experimental Pharmacology XXIII 718-780 (1968). Springer Verlag, Heidelberg, Berlin,NY. 16. A.R. FUCHS, Acta Endocr (Cph) 45 272-280(1964). 17. S. KULLANDER, Acta Endocr(Cph) 44 313-324(1963). 18. A.W. COWLEY and B.J. BARBER, In: Cross, BA and Leng, G, Eds, The Neurohypophysis, Structure, Function and Control. Prog. Brain Res. vol 60 415424 (1983). Elsevier Science Publ., BV Amsterdam-NY. 19. S.J. BERGEN,Jr., R. SULLIVAN, J.G. HILTON, S.W. WILLIS,Jr and T.B. VAN ITALLIE,Am.J. Physiol. 199(1) 136-138(1960). 20. S.B. RICHARDSON, N. EYLER, S. TWENTE, M. MONACO, N. ALTSZULER and M. GIBSON, Endocrinology 126 1047-1052 (1990). 21. B.E. DUNNING, J.H. MOLTZ and C.P. FAWCETT, Am. J. Physiol 246(9) E108-114 (1984). 22. S. STOCK, J. FASTBOM, E. BJORKSTRAND, U. UNGERSTEDT and K. UVNASMOBERG, Regulatory Peptides 30 1-13 (1990). 23. P.A. JAQUENOD and R.A. BOISSONNAS, Helv. Chim. Acta 44 113-122 (1961). 24. R. WALTER, C.W. SMITH and J. ROY, Proc. Natl. Acad. Sci. USA 73 3054-3058 (1976). 25. C.H. SCHNEIDER and V. duVIGNEAUD, J. Am. Chem. Soc. 84 3005-3008 (1962). 26. R. WALTER, V.W. SMITH, J. ROY and A. FORMENTO, J. Med. Chem. 19 822825 (1976).