Release of Vasoactive Intestinal Polypeptide (VIP) by Intraduodenal Stimuli 0. B. SCHAFFALITZKY DE MUCKADELL, J. FAHRENKRUG, J. J. HOLST & K. B. LAURITSEN Dept. of Clinical Chemistry, Bispebjerg Hospital, Copenhagen, Denmark
Scand J Gastroenterol Downloaded from informahealthcare.com by McMaster University on 11/21/14 For personal use only.
Schaffalitzky de Muckadell, 0. B., Fahrenkrug, J., Holst, J. J. & Lauritsen, K. B. Release of vasoactive intestinal polypeptide (VIP) by intraduodenal stimuli, Scand. J. Gastroent. 1977. 12. 793-199 The effect of intraduodenal infusion of amino acids, glucose, fat, HCI, ethanol, or saline on plasma VIP concentration was investigated in 7 normal subjects, 5 post-vagotomy patients, and 12 anaesthetized pigs. Furthermore, the concentrations of VIP in plasma after ingestion of a mixed meal were measured in 6 normal subjects. In normal subjects the median peripheral concentration of VIP in the basal state was 4.3pmol x 1-I (range 0- 12.0). No significant changes occurred after amino acids, glucose, saline, or ingestion of a meal. In contrast infusion of HCI, fat, or ethanol resulted in a rise in plasma VIP concentration in all the subjects studied. The peak values (medians and ranges)after HCI, fat, or ethanol were 9.8 (5.9-12.6), 7.5 (2.4-10.2), and 12.6 (7.8-16.8)pmol x l-', respectively. Truncal vagotomy did not change the response of HCI. The results from measurements in portal plasma of pigs confirmed the findings in peripheral plasma of normal subjects and showed that the levels of VIP in portal plasma are 1.6-2.9 times higher than the levels of VIP in arterial plasma. The pH threshold to release of VIP was pH 1.1-2.1, and the effect of HCI was not abolished by ganglionic blockade. Key-words: Gastrointestinal hormones: radioimmunoassay; vasoactive intestinal polypeptide Ove B. Schaffalitzky de Muckadell, M.D., Dept. of Clinical Chemistry, Bispebjerg Hospital, Bispebjerg Bakke 23, DK-2400 Copenhagen N V , Denmark
Vasoactive intestinal polypeptide (VIP) was iso- MATERIALS AND METHODS lated from porcine intestinal wall in 1970 (12). Subjects and investigative procedures Originally the peptide was demonstrated to be localThirteen healthy subjects (two females) and five ized to endocrine cells in the gut (11). In later postvagotomy patients (one female) volunteered for studies however, VIP was found in neurons of sevthe study after the nature, purpose, and possible eral organs including the brain (1, 8, 9, 14). High been carefully explained to them. In the risks had concentrations were found in the gastrointestinal group of normal subjects the age ranged from 20-35 tract of different mammalian species (8). years with a median of 28 years. In the five former VIP has a broad spectrum of biological actions (10, 12, 13), but the physiological significance of duodenal ulcer patients a truncal vagotomy and pyloroplasty had been performed 10- 1 1 years earthese effects still remains to be clarified. At present electric stimulation of the vagal nerves lier. The age ranged from 4 1 to 50 years with a (15) or intravenous calcium (3) are the only known median of 43 years. a ) Effect of NaCI, glucose, f a t , amino acids, stimuli to cause a release of VIP into the circulation. ethanol, or HCI on VIP concentration in peripheral In this study we have investigated the effects of of normal subjects. Seven subjects were each plasma various intraduodenal stimuli on the plasma concenon seven different days with at least five examined tration of immunoreactive VIP in man and pig.
Scand J Gastroenterol Downloaded from informahealthcare.com by McMaster University on 11/21/14 For personal use only.
794
0. B. SchaQalitzky de Muckadell, J. Fahrenkrug, J. J. Holst & K.B. Lauritsen
days' interval. The studies were performed in randomized order after an overnight fast. A nasoduodenal polyethylene tube was positioned under fluoroscopic control with the openings of the tube placed proximally in the second part of the duodenum. After each experiment the position of the tube was controlled. On the test day one of the following solutions was infusedintothe duodenum over a period of 3 min: 0.15 rnol x 1-INaCl 40 ml, 0.3 rnol x 1-' glucose 40 ml, isotonic fat emulsion (Intralipid@ 20 per cent,Vitrum, Stockholm, Sweden) 40ml, isotonic amino acid mixture (corresponding to a mixture of ct and Ij casein ( 5 ) 40m1, 1.4mol x 1-' glucose 40m1, ethanol (vodka 86 proof) 60m1, or 0.1 rnol x 1-' HCI (made isotonic with NaCI) 40ml. All solutions were pre-heated to 37 "C and except for HCI adjusted to pH 7.0 with 0.15 rnol x 1-' HCI or NaOH. Blood samples for VIP assay were withdrawn from an antecubital vein at 20, 10, and 0 rnin before, and 5 , 10, 15, 20, 30, 45, and 60min after the onset of the infusion. b ) Effect of HCl on VIP concentration in peripheral plasma of post-vagotomy patients. HCI was infused and blood samples were drawn as described above. c ) Effect of a meal on VIP concentration in peripheral plasma of normal subjects. Six healthy subjects were given a meal consisting of a Hamburger, mashed potatoes, a fried egg, and gravy (corresponding to 78g fat, 48g protein, and 35g carbohydrate). Peripheral venous blood was collected at 60, 40, 20, and 0 rnin before and 10, 20, 30,45,60,75,90, 120, 150, 180,240,and 300min after the start of the meal. Animals and investigative procedures Further studies were performed in pigs, since the assay is developed against porcine VIP. Twelve female pigs of Danish landrace (weighing 2 1-30 kg) were prepared as previously described (16). Infusion and drainage of the duodenum were accomplished by two catheters (Foley 22 CH); one was inserted through an incision in the stomach into the duodenal bulb and kept in place with a slightly inflated balloon by means of a tight ligature around the pylorus. Care was taken not to damage or ligate the gastro-epiploic vessels. The other catheter was inserted via a duodenostomy close to the ligament of
Treitz and anchored by a tight ligature around the duodenum and inflation of the balloon. Blood samples were obtained through polyethylene catheters placed in the portal vein and the femoral artery. a ) Effect of intraduodenal infusion of NaCI, glucose, fat, amino acids, or HCl on plasma VIP concentration in pigs. The study included seven test periods in each of six pigs. After a basal period of 15min, 50ml of the following solutions were infused into the duodenum over a one min period in the order mentioned: 0.15mol x 1-I NaCI, 0.3 rnol x 1.' glucose, isotonic fat emulsion (Intralipid@), isotonic amino acid mixture, 0.1 rnol x 1-I HCI, 1.5mol x 1-' NaCI, or 3.0mol X 1-I glucose. All solutions were pre-warmed to 37 "C and except for HCI adjusted to pH 7.0. Fifteen rnin after the infusion, the test substance was drained from the duodenum. The next test substance was not introduced into the duodenum until 15 min-and in the case of HCI 30min-after drainage of the foregoing solution. During each stimulation blood samples from the portal vein were obtained for VIP assay at 0, 5 , and 15 min, and during HCI at 0, 5, 7.5, 10, 12.5, 15, and 30min. During the intraduodenal acidification blood was drawn simultaneously from the portal vein and femoral artery. b ) Effect of intraduodenal perfusion with citrate buffers with various pH. To investigate the pH threshold of VIP release citrate buffers with a pH of 4.0, 3.0, 2.0, or 1.0 and made isotonic with saline were perfused through the duodenum at a rate of 30ml per minute. The study included four test periods of 20min in each of three pigs. The concentration of citrate was 60.4, 41.5, 30.3, 5.0mmol x 1-' respectively, and thus 60 mmol of titrable acid (end point pH 7 ) were infused at all pH levels. Blood was drawn from the portal vein 0, 10, 15, and 20min after the start of the perfusion. c ) Effect of ganglionic blockade on VIP release induced by HCl inpigs. Ten mg hexamethonium per kg bodyweight was given intravenously, and 30min later lOOml 0.1 rnol x 1-' HCI was infused into the duodenum over 5min. Portal blood was drawn 10 and 0 min before and 10, 15, and 20min after the start of the infusion. Laboratory analysis
Blood was collected in ice-chilled glass tubes
Scand J Gastroenterol Downloaded from informahealthcare.com by McMaster University on 11/21/14 For personal use only.
Release of VIP
containing 500 K.I.U. aprotinin (Trasylol? Bayer Leverkusen, West Germany) and 50 I.U. heparin per ml blood. The samples were centrifuged at 4 "C and plasma stored at -20°C until assayed. The concentration of VIP in plasma was measured radioimmunochemically (4). Antibodies were raised against highly purified natural porcine VIP (kindly donated by Prof. V. Mutt, Karolinska Institutet, Stockholm, Sweden) covalently coupled to bovine serum albumin. The antiserum used (56036 ) reacted with an effective equilibrium constant of 3.5 x 10" 1 xmol-', and no cross-reactivity was found with porcine gastric inhibitory peptide, porcine pancreatic glucagon, porcine enteroglucagon, human pancreatic polypeptide, synthetic bovine substance P, porcine natural secretin, or synthetic ovine somatostatin in concentrations below 105pmol x 1-'. '251-VIP was prepared by a chloramine T method to a specific radioactivity of approximately 900pCi per nmol peptide. Highly purified natural porcine VIP was used as standard and dissolved in VIP-free plasma prepared by charcoaling the plasma from each subject or animal. Bound and free peptide moieties were separated by plasmacoated charcoal. The lowest concentration of VIP to be distinguished from zero with 95 per cent confidence was 3.3pmol x 1-'. The within and between assay reproducibility expressed as coefficient of variation was 0.09 and 0.13 at a level of 18pmol x 1-'. All samples were assayed in triplicate. Results are given as medians and ranges (in parentheses). Statistical analyses were performed using Wilcoxon matched-pairs signed-ranks test, the Friedman two-way analysis of variance and the on-sample runs test ( 17). Differences with p values of less than 0.05 were considered significant.
795
U Amino acids
UGlucose 5
n -
>
c
I
d
X
uGlucose 25
u Na CI
-io
0
20
LQ
60
minutes
Fig. 1. Concentrations of vasoactive intestinal polypeptide (VIP) in peripheral plasma after intraduodenal infusion ( 11 ) of amino acids, isotonic glucose (glucose 5 ) , hypertonic glucose (glucose 2 5 ) or isotonic NaCI. Medians and ranges (hatched area)of 7 normal subjects.
plasma VIP concentration in all subjects studied (Fig. 2). HCI significantly raised the median plasma VIP concentration within 5 min, and a peak value of 9.8 (5.9-12.6)pmol x 1-' was reachedafter 10min. The VIP levels remained elevated for 60min. The RESULTS intraduodenal fat load raised plasma VIP to 7.5 Effect OfNaCl,glucose,fat, amino acids, ethanol, or (2.4-10.2)pmol x 1-' at 20min. The median conHCl on VIP concentration in peripheral plasma of centrations measured in the 30, 45, and 60min normal subjects. The median peripheral concentra- samples did not differ significantly from basal level. tion of VIP in the basal state was 4.3 Intraduodenal infusion of ethanol caused a signifi(0-12.0)pmol x 1-'. N o significant changes oc- cant rise in median VIP levels within lOmin, and a curred after amino acids, isotonic glucose, hyper- peak of 12.6 (7.8-16.8)pmol x 1-' was reached at tonic glucose, or NaCl (Fig. 1). In contrast, infusion 45min. VIP levels did not return to prestimulated of HCI, fat, or ethanol resulted in a significant rise in values within the 60-min observation period.
796
0. B. Schaffalitzky a'e Muckaa'ell,J. Fahrenkrug, J. J. Holst & K . B. Lauritsen
aHCI
15
a Fat Scand J Gastroenterol Downloaded from informahealthcare.com by McMaster University on 11/21/14 For personal use only.
'OI 0 5
-
51 Ethanol
-20
I
0'
20
1
I
20
40
minutes
Fig. 3. Concentrations of vasoactive intestinal polypeptide (VIP) in peripheral plasma after intraduodenal infusion of 0. lmol x I-' HCI (1). Medians and ranges (hatched area) of 5 postvagotomy patients.
UNaCI
-20
0
40
60
minutes
centration in response to ingestion of a meal (Fig. 4). Individual maximum increment was 3.0 (0.5-17.3)pmol x l-l, and peak values occurred at different times 20-300min after the start of the meal.
Fig. 2. Concentrations of vasoactive intestinal polypeptide (VIP) in peripheral plasma after intraduodenal infusion of 0.1 mol x I-' HCI, fat (20% soya bean oil emulsion), ethanol or isotonic NaCI. Medians and ranges (hatched area) of 7 normal subjects.
Effect of HCI on VIP concentration in peripheral plasma after truncal vagotomy. Median peripheral concentration ofVIP was 5.8 (0-1 1.6)pmol x 1-' in the basal state and rose significantly to 10.2 (7.5-13.0)pmol x 1 - I after infusion of HCI (Fig. 3). The response pattern did not differ from that found in normal subjects. Effect of a meal on VIP concentration in peripheralplasma of normal subjects. In three subjects the postprandial VIP concentrations increased significantly when compared to the median basal level of the same subject. However, the group as a whole showed no significant changes in plasma VIP con-
I
i,
1
1
1
z
1
hours
Fig. 4. Concentrations of vasoactive intestinal polypeptide (VIP) in peripheral plasma after ingestion of a mixed meal (bar). Medians and ranges (hatched area) of 6 normal subjects.
Release of VIP
797
Table I. Concentration (pmol x I-') of vasoactive intestinal polypeptide (VIP) in portal plasma before and after the infusion of 50ml of various solutions into the duodenum of anaesthetized pigs. Results are given as medians and ranges. n=6.
Scand J Gastroenterol Downloaded from informahealthcare.com by McMaster University on 11/21/14 For personal use only.
Basal HCI (0.1 mol x I-') Fat ( l o g soya bean oil) Amino acids Glucose (0.3 rnol x I-') Glucose (3mol x I - ' ) Saline ( O . l S r n o l x I - ' ) Saline (1.5molx 1.')
19.4 11.5 18.4 20.5 17.4 17.7 22.1
5 min
(7.9-27.4) (6.8-28.8) (7.1-34.4) (10.c~39.1) (7.1-27.7) (10.5-38.0) (11.5-29.2)
15min
28.6* ( 1 1.5-105.9) 11.7 (7.5-28.4) 18.2 (7.1-28.3) 21.0 (5.2-34.4) 16.4 (5.3-58.1) 22.0 (10.6-38.4) 17.8 (12.G41.5)
28.6* 20.3* 17.6 17.6 17. I 20.2 21.5
(9.7-79.4) (10.3-44.6) (8.8-28.6) (7.9-34.0) (5.3-39.7) (12.4-34.1) (13.2-41.5)
*Significantly different from basal values (p < 0.05).
Effect of intraduodenal infusion of NaCI, glucose, fat, amino acids or HCI on plasma VIP concentration in pigs. T h e results obtained from measurements in portal plasma of pigs (Table I and Fig. 5 )
confirmed the findings in peripheral plasma of normal subjects. The ratio between concentrations of VIP in portal venous plasma and arterial plasma was 1.6 (1.3-2.0) in the basal state. indicating that VIP is released from the splanchnic bed. HCI raised the ratio t o a maximum value of 2.9 (1.7-4.9).
Effect of buffers with various pH on plasma VIP concentration in pigs. Only the buffer with p H adjusted to 1.0 resulted in a significant increase in the portal concentration of VIP (Fig. 6). When buffer with a p H of 2 or 1 was infused, the median
60
n 5 40 40
T
d
I -L
0
30
€
20
20 10
HC I 0
0
15
30
minutes
Fig. 5. Concentrations of vasoactive intestinal polypeptide (VIP) in portal (0)and arterial (0) plasma after intraduodenal infusion of 0.1mol x 1.' HCI (bar). Medians and ranges (hatched area) of 6 anaesthetized pigs.
0
20
io
60
80
minutes
Fig. 6. Effect of intraduodenal pH on concentrations of vasoactive intestinal polypeptide (VIP) in portal plasma of 3 anaesthetized pigs. Individual values are shown.
0.B. Schaffalitzky de Muckadell. J. Fahrenkrug, J. J. Holst & K . B. Lauritsen
798
150-
Scand J Gastroenterol Downloaded from informahealthcare.com by McMaster University on 11/21/14 For personal use only.
100-
50-
0I
-10
i,
115
30
minutes
Fig. 7. Changes in the portal venous concentrations of vasoactive intestinal polypeptide (VIP) in 3 pigs after ganglionic blockade ( lOmg/kg Hexamethonium)and intraduodenal infusion of 0.1 mol x 1-' HCI ( 1). Individual values are shown.
pH of the duodenal aspirate was 2.1 or 1.1, respectively. Eflect of ganglionic blockade on the release of VIP induced by HCI in pigs. The response to intraduodenal infusion of HCI was not abolished by hexamethonium (Fig. 7). DISCUSSION The present study has shown that intraduodenal HCI, fat, or ethanol are stimuli to VIP release. In contrast, amino acids, isotonic glucose, hypertonic glucose, isotonic saline or hypertonic saline did not affect the VIP concentrations either in peripheral or in portal plasma. The lack of significant changes in the median VIP concentration in peripheral plasma after ingestion of a mixed meal might seem surprising, since the meal contained fat and since increased vagal activity, at least by electrical stimulation, is known to release VIP (15). However, some inactivation of VIP during the passage through the liver occurs (3, 7) and changes in concentration of VIP in portal plasma are
3-4 times greater than the changes in peripheral plasma (Fig. 5). Thus the release of VIP after a meal might be detectable in portal plasma only, and escape notice in peripheral plasma. In addition, the response to the meal is blurred out by interindividual variation in gastric emptying rate. Immunocytochemical and radioimmunochemical studies (8) have shown that VIP in the gut is predominantly localized to neurons in the mucosa, the submucosa, and the smooth muscle layers. The localization of VIP to neurons suggests that VIP is released into the intercellular space at the site of action and not primarily released into the circulation. It cannot be excluded therefore that VIP might be released locally without a concomitant increase in circulating levels of VIP. In fact, it is possible, that all changes in the plasma concentration of VIP only reflect overflow of VIP released from nerve terminals in innervated tissues. However, VIP release from VIP-containing endocrine cells may contribute in some species (1 1). VIP stimulates pancreatic bicarbonate secretion and inhibits pentagastrin-stimulated acid and pepsin secretion (10). Thus the finding that VIP is released by acid or fat in the gut might indicate a physiological role of the peptide as a modulator of pancreatic and gastric secretions. The concentrations of VIP in arterial plasma after these stimuli, however, may not be sufficiently high to affect these secretions. If however circulating VIP merely represents an overflow of locally released VIP, it is reasonable to believe that much higher concentrations are encountered at the receptor site in innervated tissue. The pH threshold for VIP-release after intraduodenal acid is above the levels normally encountered in the second part of the duodenum. Thus under physiological circumstances, only the most proximal part of the duodenum would be expected to be exposed to this high acidity. However, the experiments were performed in anaesthetized animals, and it cannot be excluded that anaesthesia might have suppressed the release of VIP. On the other hand, a local release of VIP which escapes notice in plasma may take place at higher pH values. Ethanol has been reported to release secretin from the duodenum ( 18) and thus to stimulate pancreatic bicarbonate secretion. We have been unable to confirm this finding (6). It is of interest that ethanol
Release of VIP
stimulates VIP-release, which by its action on the pancreatic bicarbonate secretion might explain the earlier observed enhancing effect of ethanol on the concentration of bicarbonate in duodenal aspirates I?\
Scand J Gastroenterol Downloaded from informahealthcare.com by McMaster University on 11/21/14 For personal use only.
ILJ.
Hexamethonium, in a dose which completely abolishes the release of VIP that follows electric stimulation of the vagal nerves (Fahrenkrung, Schaffalitzky de Muckdell, and Holst: Unpublished results), did not abolish the VIP-release induced by HCI in the gut. Thus, if VIP is released from nerve fibres, no ganglionic (cholinergic) transmission via nicotinic receptors seems to be implied. Furthermore, an intact vagal innervation is not necessary for the VIP release after intraduodenal acid, since the vagotomized patients responded normally to this stimulus.
ACKNOWLEDGEMENTS The skilful technical assistance of Hanne Dam, Lene Poulsen, Nina Rasmussen, and Anne Sylvest is gratefully acknowledged. This study was supported by grants from The Danish Medical Research Foundation (jm. 5 125273, 5 12-5676, 5 12-7221), The Danish Hospital Foundation for Medical Research, Region of Copenhagen, The Faroe Islands, and Greenland (jm. 76/17,46 and 75/76,40) and King Christian den X’s Fond. Received 20 May 1977 Accepted 20 June 1977
799
REFERENCES 1. Bryant, M. G., Bloom, S. R., Polak, J. M., Albuquerque, R. H., ~ d l i 1.~&, pearse, A. G. E. L~~~~~ 1976, i, 99 1-993 2. CaDitaine, Y., de Barros Mott, C., Gullo, L. & Sarles, H.Biologie et Gastro-Enterologie 197 1,3, 193-198 3. Ebeid, A. M., Murray, P., Sooters, P. B. & Fischer, J. E. Amer. J. Surg. 1977, 133, 140-144 4. Fahrenkrug, J. & Schaffalitzky de Muckadell, 0. B. J. Lab. Clin. Med., 1977, 89,1379-1388 5. Fahrenkrug, J., Schaffalitzky de Muckadell, 0. B. & Holst, J. J. Scand. J. Gastroent. 1977, 22, 273-279 6. Fahrenkrug, J. & SchafTalitzky de Muckadell, 0. B. Europ. J. Clin. Invest. 1977, 7, 199-201 7. Kitamura, S.,Yoshida, T. & Said, S . I. Proc. Soc. Exp. Biol. [N.Y.] 1975, 148, 25-29 8. Larsson, L.-I., Fahrenkrug, J., Schaffalitzky de Muckadell, 0. B., Sundler, F., Hikanson, R. & Rehfeld, J. F. Proc. Nut. Acad. Sci (Wash.) 1976, 73, 3 197-3200 9. Larsson, L.-I., Fahrenkrug, J. & SchafTalitzky de Muckadell, 0. B. Science, in press 10. Makhlouf, G. M. & Said, S. I. pp. 599-610 in Thompson, J. C., Gastroentestinal Hormones. Austin and London, University of Texas Press, 1975 11. Polak, J. M., Pearse, A. G. E., Geraud, J.-C. & Bloom, S . R. Gut. 1974, 15. 720-724 12. Said, S. 1. & Mutt, V. Science 1970. 169, 1217 13. Said, S. I. pp. 591-597 in Thompson, J. C. (ed.) Gastroentestinal Hormones, Austin and London, University of Texas Press, 1975 14. Said, S. I. & Rosenberg, R. N. Science, 1976, 192, 907-908 15. SchafTalitzky de Muckadell, 0. B., Fahrenkrug, J. & Holst, J. J. Gastroenterology 1977, 72, 373-375 16. SchafTalitzky de Muckall, 0. B., Fahrenkrug, J. & Holst, J. J. Scand. J. Gastroent. 1977, 12, 267-272 17. Siegel, S. New-York: McGraw-Hill Book Company, Inc., 1956. 18. Straus, E., Urback, H.-J. & Yalow,R. S. NewEngl. J. Med. 1975, 292, 103 1-1032
Scand J Gastroenterol Downloaded from informahealthcare.com by McMaster University on 11/21/14 For personal use only.
Schaffalitzky de Muckadell, 0. B., Fahrenkrug, J., Holst, J. J. & Lauritsen, K. B. Release of vasoactive intestinal polypeptide (VIP) by intraduodenal stimuli, Scand. J. Gastroent. 1977. 12. 793-199 The effect of intraduodenal infusion of amino acids, glucose, fat, HCI, ethanol, or saline on plasma VIP concentration was investigated in 7 normal subjects, 5 post-vagotomy patients, and 12 anaesthetized pigs. Furthermore, the concentrations of VIP in plasma after ingestion of a mixed meal were measured in 6 normal subjects. In normal subjects the median peripheral concentration of VIP in the basal state was 4.3pmol x 1-I (range 0- 12.0). No significant changes occurred after amino acids, glucose, saline, or ingestion of a meal. In contrast infusion of HCI, fat, or ethanol resulted in a rise in plasma VIP concentration in all the subjects studied. The peak values (medians and ranges)after HCI, fat, or ethanol were 9.8 (5.9-12.6), 7.5 (2.4-10.2), and 12.6 (7.8-16.8)pmol x l-', respectively. Truncal vagotomy did not change the response of HCI. The results from measurements in portal plasma of pigs confirmed the findings in peripheral plasma of normal subjects and showed that the levels of VIP in portal plasma are 1.6-2.9 times higher than the levels of VIP in arterial plasma. The pH threshold to release of VIP was pH 1.1-2.1, and the effect of HCI was not abolished by ganglionic blockade. Key-words: Gastrointestinal hormones: radioimmunoassay; vasoactive intestinal polypeptide Ove B. Schaffalitzky de Muckadell, M.D., Dept. of Clinical Chemistry, Bispebjerg Hospital, Bispebjerg Bakke 23, DK-2400 Copenhagen N V , Denmark
Vasoactive intestinal polypeptide (VIP) was iso- MATERIALS AND METHODS lated from porcine intestinal wall in 1970 (12). Subjects and investigative procedures Originally the peptide was demonstrated to be localThirteen healthy subjects (two females) and five ized to endocrine cells in the gut (11). In later postvagotomy patients (one female) volunteered for studies however, VIP was found in neurons of sevthe study after the nature, purpose, and possible eral organs including the brain (1, 8, 9, 14). High been carefully explained to them. In the risks had concentrations were found in the gastrointestinal group of normal subjects the age ranged from 20-35 tract of different mammalian species (8). years with a median of 28 years. In the five former VIP has a broad spectrum of biological actions (10, 12, 13), but the physiological significance of duodenal ulcer patients a truncal vagotomy and pyloroplasty had been performed 10- 1 1 years earthese effects still remains to be clarified. At present electric stimulation of the vagal nerves lier. The age ranged from 4 1 to 50 years with a (15) or intravenous calcium (3) are the only known median of 43 years. a ) Effect of NaCI, glucose, f a t , amino acids, stimuli to cause a release of VIP into the circulation. ethanol, or HCI on VIP concentration in peripheral In this study we have investigated the effects of of normal subjects. Seven subjects were each plasma various intraduodenal stimuli on the plasma concenon seven different days with at least five examined tration of immunoreactive VIP in man and pig.
Scand J Gastroenterol Downloaded from informahealthcare.com by McMaster University on 11/21/14 For personal use only.
794
0. B. SchaQalitzky de Muckadell, J. Fahrenkrug, J. J. Holst & K.B. Lauritsen
days' interval. The studies were performed in randomized order after an overnight fast. A nasoduodenal polyethylene tube was positioned under fluoroscopic control with the openings of the tube placed proximally in the second part of the duodenum. After each experiment the position of the tube was controlled. On the test day one of the following solutions was infusedintothe duodenum over a period of 3 min: 0.15 rnol x 1-INaCl 40 ml, 0.3 rnol x 1-' glucose 40 ml, isotonic fat emulsion (Intralipid@ 20 per cent,Vitrum, Stockholm, Sweden) 40ml, isotonic amino acid mixture (corresponding to a mixture of ct and Ij casein ( 5 ) 40m1, 1.4mol x 1-' glucose 40m1, ethanol (vodka 86 proof) 60m1, or 0.1 rnol x 1-' HCI (made isotonic with NaCI) 40ml. All solutions were pre-heated to 37 "C and except for HCI adjusted to pH 7.0 with 0.15 rnol x 1-' HCI or NaOH. Blood samples for VIP assay were withdrawn from an antecubital vein at 20, 10, and 0 rnin before, and 5 , 10, 15, 20, 30, 45, and 60min after the onset of the infusion. b ) Effect of HCl on VIP concentration in peripheral plasma of post-vagotomy patients. HCI was infused and blood samples were drawn as described above. c ) Effect of a meal on VIP concentration in peripheral plasma of normal subjects. Six healthy subjects were given a meal consisting of a Hamburger, mashed potatoes, a fried egg, and gravy (corresponding to 78g fat, 48g protein, and 35g carbohydrate). Peripheral venous blood was collected at 60, 40, 20, and 0 rnin before and 10, 20, 30,45,60,75,90, 120, 150, 180,240,and 300min after the start of the meal. Animals and investigative procedures Further studies were performed in pigs, since the assay is developed against porcine VIP. Twelve female pigs of Danish landrace (weighing 2 1-30 kg) were prepared as previously described (16). Infusion and drainage of the duodenum were accomplished by two catheters (Foley 22 CH); one was inserted through an incision in the stomach into the duodenal bulb and kept in place with a slightly inflated balloon by means of a tight ligature around the pylorus. Care was taken not to damage or ligate the gastro-epiploic vessels. The other catheter was inserted via a duodenostomy close to the ligament of
Treitz and anchored by a tight ligature around the duodenum and inflation of the balloon. Blood samples were obtained through polyethylene catheters placed in the portal vein and the femoral artery. a ) Effect of intraduodenal infusion of NaCI, glucose, fat, amino acids, or HCl on plasma VIP concentration in pigs. The study included seven test periods in each of six pigs. After a basal period of 15min, 50ml of the following solutions were infused into the duodenum over a one min period in the order mentioned: 0.15mol x 1-I NaCI, 0.3 rnol x 1.' glucose, isotonic fat emulsion (Intralipid@), isotonic amino acid mixture, 0.1 rnol x 1-I HCI, 1.5mol x 1-' NaCI, or 3.0mol X 1-I glucose. All solutions were pre-warmed to 37 "C and except for HCI adjusted to pH 7.0. Fifteen rnin after the infusion, the test substance was drained from the duodenum. The next test substance was not introduced into the duodenum until 15 min-and in the case of HCI 30min-after drainage of the foregoing solution. During each stimulation blood samples from the portal vein were obtained for VIP assay at 0, 5 , and 15 min, and during HCI at 0, 5, 7.5, 10, 12.5, 15, and 30min. During the intraduodenal acidification blood was drawn simultaneously from the portal vein and femoral artery. b ) Effect of intraduodenal perfusion with citrate buffers with various pH. To investigate the pH threshold of VIP release citrate buffers with a pH of 4.0, 3.0, 2.0, or 1.0 and made isotonic with saline were perfused through the duodenum at a rate of 30ml per minute. The study included four test periods of 20min in each of three pigs. The concentration of citrate was 60.4, 41.5, 30.3, 5.0mmol x 1-' respectively, and thus 60 mmol of titrable acid (end point pH 7 ) were infused at all pH levels. Blood was drawn from the portal vein 0, 10, 15, and 20min after the start of the perfusion. c ) Effect of ganglionic blockade on VIP release induced by HCl inpigs. Ten mg hexamethonium per kg bodyweight was given intravenously, and 30min later lOOml 0.1 rnol x 1-' HCI was infused into the duodenum over 5min. Portal blood was drawn 10 and 0 min before and 10, 15, and 20min after the start of the infusion. Laboratory analysis
Blood was collected in ice-chilled glass tubes
Scand J Gastroenterol Downloaded from informahealthcare.com by McMaster University on 11/21/14 For personal use only.
Release of VIP
containing 500 K.I.U. aprotinin (Trasylol? Bayer Leverkusen, West Germany) and 50 I.U. heparin per ml blood. The samples were centrifuged at 4 "C and plasma stored at -20°C until assayed. The concentration of VIP in plasma was measured radioimmunochemically (4). Antibodies were raised against highly purified natural porcine VIP (kindly donated by Prof. V. Mutt, Karolinska Institutet, Stockholm, Sweden) covalently coupled to bovine serum albumin. The antiserum used (56036 ) reacted with an effective equilibrium constant of 3.5 x 10" 1 xmol-', and no cross-reactivity was found with porcine gastric inhibitory peptide, porcine pancreatic glucagon, porcine enteroglucagon, human pancreatic polypeptide, synthetic bovine substance P, porcine natural secretin, or synthetic ovine somatostatin in concentrations below 105pmol x 1-'. '251-VIP was prepared by a chloramine T method to a specific radioactivity of approximately 900pCi per nmol peptide. Highly purified natural porcine VIP was used as standard and dissolved in VIP-free plasma prepared by charcoaling the plasma from each subject or animal. Bound and free peptide moieties were separated by plasmacoated charcoal. The lowest concentration of VIP to be distinguished from zero with 95 per cent confidence was 3.3pmol x 1-'. The within and between assay reproducibility expressed as coefficient of variation was 0.09 and 0.13 at a level of 18pmol x 1-'. All samples were assayed in triplicate. Results are given as medians and ranges (in parentheses). Statistical analyses were performed using Wilcoxon matched-pairs signed-ranks test, the Friedman two-way analysis of variance and the on-sample runs test ( 17). Differences with p values of less than 0.05 were considered significant.
795
U Amino acids
UGlucose 5
n -
>
c
I
d
X
uGlucose 25
u Na CI
-io
0
20
LQ
60
minutes
Fig. 1. Concentrations of vasoactive intestinal polypeptide (VIP) in peripheral plasma after intraduodenal infusion ( 11 ) of amino acids, isotonic glucose (glucose 5 ) , hypertonic glucose (glucose 2 5 ) or isotonic NaCI. Medians and ranges (hatched area)of 7 normal subjects.
plasma VIP concentration in all subjects studied (Fig. 2). HCI significantly raised the median plasma VIP concentration within 5 min, and a peak value of 9.8 (5.9-12.6)pmol x 1-' was reachedafter 10min. The VIP levels remained elevated for 60min. The RESULTS intraduodenal fat load raised plasma VIP to 7.5 Effect OfNaCl,glucose,fat, amino acids, ethanol, or (2.4-10.2)pmol x 1-' at 20min. The median conHCl on VIP concentration in peripheral plasma of centrations measured in the 30, 45, and 60min normal subjects. The median peripheral concentra- samples did not differ significantly from basal level. tion of VIP in the basal state was 4.3 Intraduodenal infusion of ethanol caused a signifi(0-12.0)pmol x 1-'. N o significant changes oc- cant rise in median VIP levels within lOmin, and a curred after amino acids, isotonic glucose, hyper- peak of 12.6 (7.8-16.8)pmol x 1-' was reached at tonic glucose, or NaCl (Fig. 1). In contrast, infusion 45min. VIP levels did not return to prestimulated of HCI, fat, or ethanol resulted in a significant rise in values within the 60-min observation period.
796
0. B. Schaffalitzky a'e Muckaa'ell,J. Fahrenkrug, J. J. Holst & K . B. Lauritsen
aHCI
15
a Fat Scand J Gastroenterol Downloaded from informahealthcare.com by McMaster University on 11/21/14 For personal use only.
'OI 0 5
-
51 Ethanol
-20
I
0'
20
1
I
20
40
minutes
Fig. 3. Concentrations of vasoactive intestinal polypeptide (VIP) in peripheral plasma after intraduodenal infusion of 0. lmol x I-' HCI (1). Medians and ranges (hatched area) of 5 postvagotomy patients.
UNaCI
-20
0
40
60
minutes
centration in response to ingestion of a meal (Fig. 4). Individual maximum increment was 3.0 (0.5-17.3)pmol x l-l, and peak values occurred at different times 20-300min after the start of the meal.
Fig. 2. Concentrations of vasoactive intestinal polypeptide (VIP) in peripheral plasma after intraduodenal infusion of 0.1 mol x I-' HCI, fat (20% soya bean oil emulsion), ethanol or isotonic NaCI. Medians and ranges (hatched area) of 7 normal subjects.
Effect of HCI on VIP concentration in peripheral plasma after truncal vagotomy. Median peripheral concentration ofVIP was 5.8 (0-1 1.6)pmol x 1-' in the basal state and rose significantly to 10.2 (7.5-13.0)pmol x 1 - I after infusion of HCI (Fig. 3). The response pattern did not differ from that found in normal subjects. Effect of a meal on VIP concentration in peripheralplasma of normal subjects. In three subjects the postprandial VIP concentrations increased significantly when compared to the median basal level of the same subject. However, the group as a whole showed no significant changes in plasma VIP con-
I
i,
1
1
1
z
1
hours
Fig. 4. Concentrations of vasoactive intestinal polypeptide (VIP) in peripheral plasma after ingestion of a mixed meal (bar). Medians and ranges (hatched area) of 6 normal subjects.
Release of VIP
797
Table I. Concentration (pmol x I-') of vasoactive intestinal polypeptide (VIP) in portal plasma before and after the infusion of 50ml of various solutions into the duodenum of anaesthetized pigs. Results are given as medians and ranges. n=6.
Scand J Gastroenterol Downloaded from informahealthcare.com by McMaster University on 11/21/14 For personal use only.
Basal HCI (0.1 mol x I-') Fat ( l o g soya bean oil) Amino acids Glucose (0.3 rnol x I-') Glucose (3mol x I - ' ) Saline ( O . l S r n o l x I - ' ) Saline (1.5molx 1.')
19.4 11.5 18.4 20.5 17.4 17.7 22.1
5 min
(7.9-27.4) (6.8-28.8) (7.1-34.4) (10.c~39.1) (7.1-27.7) (10.5-38.0) (11.5-29.2)
15min
28.6* ( 1 1.5-105.9) 11.7 (7.5-28.4) 18.2 (7.1-28.3) 21.0 (5.2-34.4) 16.4 (5.3-58.1) 22.0 (10.6-38.4) 17.8 (12.G41.5)
28.6* 20.3* 17.6 17.6 17. I 20.2 21.5
(9.7-79.4) (10.3-44.6) (8.8-28.6) (7.9-34.0) (5.3-39.7) (12.4-34.1) (13.2-41.5)
*Significantly different from basal values (p < 0.05).
Effect of intraduodenal infusion of NaCI, glucose, fat, amino acids or HCI on plasma VIP concentration in pigs. T h e results obtained from measurements in portal plasma of pigs (Table I and Fig. 5 )
confirmed the findings in peripheral plasma of normal subjects. The ratio between concentrations of VIP in portal venous plasma and arterial plasma was 1.6 (1.3-2.0) in the basal state. indicating that VIP is released from the splanchnic bed. HCI raised the ratio t o a maximum value of 2.9 (1.7-4.9).
Effect of buffers with various pH on plasma VIP concentration in pigs. Only the buffer with p H adjusted to 1.0 resulted in a significant increase in the portal concentration of VIP (Fig. 6). When buffer with a p H of 2 or 1 was infused, the median
60
n 5 40 40
T
d
I -L
0
30
€
20
20 10
HC I 0
0
15
30
minutes
Fig. 5. Concentrations of vasoactive intestinal polypeptide (VIP) in portal (0)and arterial (0) plasma after intraduodenal infusion of 0.1mol x 1.' HCI (bar). Medians and ranges (hatched area) of 6 anaesthetized pigs.
0
20
io
60
80
minutes
Fig. 6. Effect of intraduodenal pH on concentrations of vasoactive intestinal polypeptide (VIP) in portal plasma of 3 anaesthetized pigs. Individual values are shown.
0.B. Schaffalitzky de Muckadell. J. Fahrenkrug, J. J. Holst & K . B. Lauritsen
798
150-
Scand J Gastroenterol Downloaded from informahealthcare.com by McMaster University on 11/21/14 For personal use only.
100-
50-
0I
-10
i,
115
30
minutes
Fig. 7. Changes in the portal venous concentrations of vasoactive intestinal polypeptide (VIP) in 3 pigs after ganglionic blockade ( lOmg/kg Hexamethonium)and intraduodenal infusion of 0.1 mol x 1-' HCI ( 1). Individual values are shown.
pH of the duodenal aspirate was 2.1 or 1.1, respectively. Eflect of ganglionic blockade on the release of VIP induced by HCI in pigs. The response to intraduodenal infusion of HCI was not abolished by hexamethonium (Fig. 7). DISCUSSION The present study has shown that intraduodenal HCI, fat, or ethanol are stimuli to VIP release. In contrast, amino acids, isotonic glucose, hypertonic glucose, isotonic saline or hypertonic saline did not affect the VIP concentrations either in peripheral or in portal plasma. The lack of significant changes in the median VIP concentration in peripheral plasma after ingestion of a mixed meal might seem surprising, since the meal contained fat and since increased vagal activity, at least by electrical stimulation, is known to release VIP (15). However, some inactivation of VIP during the passage through the liver occurs (3, 7) and changes in concentration of VIP in portal plasma are
3-4 times greater than the changes in peripheral plasma (Fig. 5). Thus the release of VIP after a meal might be detectable in portal plasma only, and escape notice in peripheral plasma. In addition, the response to the meal is blurred out by interindividual variation in gastric emptying rate. Immunocytochemical and radioimmunochemical studies (8) have shown that VIP in the gut is predominantly localized to neurons in the mucosa, the submucosa, and the smooth muscle layers. The localization of VIP to neurons suggests that VIP is released into the intercellular space at the site of action and not primarily released into the circulation. It cannot be excluded therefore that VIP might be released locally without a concomitant increase in circulating levels of VIP. In fact, it is possible, that all changes in the plasma concentration of VIP only reflect overflow of VIP released from nerve terminals in innervated tissues. However, VIP release from VIP-containing endocrine cells may contribute in some species (1 1). VIP stimulates pancreatic bicarbonate secretion and inhibits pentagastrin-stimulated acid and pepsin secretion (10). Thus the finding that VIP is released by acid or fat in the gut might indicate a physiological role of the peptide as a modulator of pancreatic and gastric secretions. The concentrations of VIP in arterial plasma after these stimuli, however, may not be sufficiently high to affect these secretions. If however circulating VIP merely represents an overflow of locally released VIP, it is reasonable to believe that much higher concentrations are encountered at the receptor site in innervated tissue. The pH threshold for VIP-release after intraduodenal acid is above the levels normally encountered in the second part of the duodenum. Thus under physiological circumstances, only the most proximal part of the duodenum would be expected to be exposed to this high acidity. However, the experiments were performed in anaesthetized animals, and it cannot be excluded that anaesthesia might have suppressed the release of VIP. On the other hand, a local release of VIP which escapes notice in plasma may take place at higher pH values. Ethanol has been reported to release secretin from the duodenum ( 18) and thus to stimulate pancreatic bicarbonate secretion. We have been unable to confirm this finding (6). It is of interest that ethanol
Release of VIP
stimulates VIP-release, which by its action on the pancreatic bicarbonate secretion might explain the earlier observed enhancing effect of ethanol on the concentration of bicarbonate in duodenal aspirates I?\
Scand J Gastroenterol Downloaded from informahealthcare.com by McMaster University on 11/21/14 For personal use only.
ILJ.
Hexamethonium, in a dose which completely abolishes the release of VIP that follows electric stimulation of the vagal nerves (Fahrenkrung, Schaffalitzky de Muckdell, and Holst: Unpublished results), did not abolish the VIP-release induced by HCI in the gut. Thus, if VIP is released from nerve fibres, no ganglionic (cholinergic) transmission via nicotinic receptors seems to be implied. Furthermore, an intact vagal innervation is not necessary for the VIP release after intraduodenal acid, since the vagotomized patients responded normally to this stimulus.
ACKNOWLEDGEMENTS The skilful technical assistance of Hanne Dam, Lene Poulsen, Nina Rasmussen, and Anne Sylvest is gratefully acknowledged. This study was supported by grants from The Danish Medical Research Foundation (jm. 5 125273, 5 12-5676, 5 12-7221), The Danish Hospital Foundation for Medical Research, Region of Copenhagen, The Faroe Islands, and Greenland (jm. 76/17,46 and 75/76,40) and King Christian den X’s Fond. Received 20 May 1977 Accepted 20 June 1977
799
REFERENCES 1. Bryant, M. G., Bloom, S. R., Polak, J. M., Albuquerque, R. H., ~ d l i 1.~&, pearse, A. G. E. L~~~~~ 1976, i, 99 1-993 2. CaDitaine, Y., de Barros Mott, C., Gullo, L. & Sarles, H.Biologie et Gastro-Enterologie 197 1,3, 193-198 3. Ebeid, A. M., Murray, P., Sooters, P. B. & Fischer, J. E. Amer. J. Surg. 1977, 133, 140-144 4. Fahrenkrug, J. & Schaffalitzky de Muckadell, 0. B. J. Lab. Clin. Med., 1977, 89,1379-1388 5. Fahrenkrug, J., Schaffalitzky de Muckadell, 0. B. & Holst, J. J. Scand. J. Gastroent. 1977, 22, 273-279 6. Fahrenkrug, J. & SchafTalitzky de Muckadell, 0. B. Europ. J. Clin. Invest. 1977, 7, 199-201 7. Kitamura, S.,Yoshida, T. & Said, S . I. Proc. Soc. Exp. Biol. [N.Y.] 1975, 148, 25-29 8. Larsson, L.-I., Fahrenkrug, J., Schaffalitzky de Muckadell, 0. B., Sundler, F., Hikanson, R. & Rehfeld, J. F. Proc. Nut. Acad. Sci (Wash.) 1976, 73, 3 197-3200 9. Larsson, L.-I., Fahrenkrug, J. & SchafTalitzky de Muckadell, 0. B. Science, in press 10. Makhlouf, G. M. & Said, S. I. pp. 599-610 in Thompson, J. C., Gastroentestinal Hormones. Austin and London, University of Texas Press, 1975 11. Polak, J. M., Pearse, A. G. E., Geraud, J.-C. & Bloom, S . R. Gut. 1974, 15. 720-724 12. Said, S. 1. & Mutt, V. Science 1970. 169, 1217 13. Said, S. I. pp. 591-597 in Thompson, J. C. (ed.) Gastroentestinal Hormones, Austin and London, University of Texas Press, 1975 14. Said, S. I. & Rosenberg, R. N. Science, 1976, 192, 907-908 15. SchafTalitzky de Muckadell, 0. B., Fahrenkrug, J. & Holst, J. J. Gastroenterology 1977, 72, 373-375 16. SchafTalitzky de Muckall, 0. B., Fahrenkrug, J. & Holst, J. J. Scand. J. Gastroent. 1977, 12, 267-272 17. Siegel, S. New-York: McGraw-Hill Book Company, Inc., 1956. 18. Straus, E., Urback, H.-J. & Yalow,R. S. NewEngl. J. Med. 1975, 292, 103 1-1032
