- Email: [email protected]
Inhibitory effect of galanin on postprandial gastrointestinal motility and gut hormone release in humans
GASTROENTEROLOGY
1989:97:288-l
Inhibitory Effect of Galanin on Postprandial Gastrointestinal Motility and Gut Hormone Release in Humans F. E. BAUER, A. ZINTEL, and S. R. BLOOM
M. J. KENNY,
D. CALDER,
M. A. GHATEI,
Departments of Gastroenterology and Radiology, Universittitsklinikum Charlottenburg, Freie Universitst Berlin, Berlin, Federal Republic of Germany; and Department of Medicine, Royal Postgraduate Medical School, Hammersmith Hospital, London, United Kingdom
Galanin was infused intravenously in 8 healthy volunteers at a dose of 40 pmol/kg . min for 1 h to investigate the pharmacologic effects of this peptide on postprandial gastrointestinal motility and gut peptide release in humans. Galanin strongly inhibited gastrointestinal motility. Gastric emptying was significantly delayed, with the time taken to empty 50% of the gastric contents increasing from 59.0 f 4.8 min (control infusion) to 99.3 + 4.7 min (galanin infusion). Mouth-to-cecum transit time increased from 87.5 f 8.9 to 128.3 + 18.5 min. Galanin potently suppressed the initial postprandial rise in plasma concentrations of glucose, insulin, peptide tyrosine tyrosine, neurotensin, enteroglucagon, pancreatic glucagon, somatostatin, and pancreatic polypeptide, but did not change gastric inhibitory polypeptide, motilin, peptide histidine methionine, and gastrin concentrations compared with control. The results indicate that an infusion of galanin has potent effects on the gastrointestinal tract in humans. The changes in motor activity in particular suggest that the local galaninergic innervation could have an important physiologic role in the control of human gastrointestinal propulsive motor activity.
Gfrom porcine
alanin is a 2%amino acid peptide first isolated intestine by a chemical technique that identifies C-terminal amidated peptides (1). The peptide contracts smooth muscle strips from gastric fundus, ileum, and colon of the rat (1) and inhibits canine small intestinal muscle both in vivo and in vitro (2). It was subsequently found that galanin potently inhibits insulin release in animals (3-6) but not in humans (7). Galanin is localized exclusively to nerves, and a dense network of galanin immunoreactive nerve fibers have been reported in the gas-
trointestinal tract, being particularly numerous in the small intestine of several mammalian species including humans (8-11). The localization of galanin to the enteric nervous plexus and its pharmacologic effect on gastrointestinal smooth muscle led us to investigate the effect of the peptide on human gastrointestinal motility, using the isotope method as an index of gastric emptying, and mouth-to-cecum transit time as reflected by the time to postprandial breath hydrogen rise. In addition, we examined the action of galanin on the pattern of postprandial gut peptide release.
Materials and Methods Subjects Eight healthy volunteers (6 men, 2 women, aged 22-32 yr, weight 48-73 kg) were each studied on two occasions at least 1 wk apart. All subjects gave written informed consent and the protocol was approved by the local Ethics Committee of the Universitatsklinikum Charlottenburg. After an overnight fast, subjects were seated upright in front of a y-camera. An indwelling cannula was inserted into a vein of each forearm. Each volunteer underwent two studies, receiving in random order an infusion of 0.9% saline (control) or synthetic galanin at a dose of 40 pmol/kg . min for 60 min. The peptide was dissolved in 50 ml of saline containing 1% human serum albumin immediately before the infusion. The subjects were unaware of the nature of each infusion. Five minutes after the start of the infusion subjects ate a solid mixed meal consisting of 90 g of white bread, 10 g of butter, two boiled eggs, and 200 ml of unsweetened orange juice (62 g carbohydrate, 23 g protein, and 23 g fat; 530 kcal] within 5 min. The orange juice contained 25 g of lactulose (Bifiteral;
0 1989 by the American
Gastroenterological
0018-5085/89/$3.50
Association
August 1989
GALANIN AND GI MOTILITY
100
Duphar, Hannover, F.R.G.) serving as substrate for colonic bacteria to produce hydrogen, and 400 $i (14.8 MBq) technetium 99m (t,,, = 6 h) colloid (Pertechnetat; Hoechst, Frankfurt, F.R.G.). The orange juice was drunk immediately after eating the solid meal. Blood samples were collected 15 min before the start of the infusion and thereafter at 15min intervals throughout the 3-h experimental period.
Gastric
Gamma camera (Apex 415 Elscint, Tel Aviv, Israel) scintigraphic data were collected at l-min intervals and stored on magnetic tape for later processing by an on-line computer. The stomach area was defined from the distribution of activity during the first 4 min after drinking the isotope by two independent observers unaware of the nature of the infusion. Collection of data started immediately after ingestion of the radioactive drink and was continued for 180 min. The rate of gastric emptying was calculated as the rate of loss of radioactivity’from the stomach area, after correction for isotope decay, computed against time. The 50% point of gastric emptying was defined as the time taken for radioactivity over the stomach area to decline to 50% of the baseline maximum. In addition, the percentage of radioactivity remaining in the stomach after 60 min (i.e., at the end of the infusion period) was calculated.
Mouth-to-Cecum
Transit
Time
Mouth-to-cecum transit time was estimated by the breath hydrogen technique (12,13). Twenty-milliliter samples of end-tidal exhaled air were collected using a 20-ml syringe with a special three-way tap. Collection was performed at 5-min intervals from -15 to 180 min and samples were analyzed for hydrogen content using an electronic monitor (HZ-Exhalation Analyzer; Stimotron, Wendelstein, F.R.G.). The time taken for the breath hydrogen concentration to exceed the baseline value by at least 3 ppm for three consecutive readings was taken as a measure of mouth-to-cecum transit time.
Peptide
Assays
Ten-milliliter blood samples were collected from the contralateral cannula at 15-min intervals from -15 to 180 min into lithium-heparin tubes containing 4000 KIU aprotinin (Trasylol; Bayer AG, Leverkusen, F.R.G.). Plasma was separated by immediate centrifugation for 10 min at 3000 rpm and stored at -20°C until assayed. Aprotinin effectively prevented enzymatic degradation of galanin in plasma for up to 4 h (unpublished observation). Plasma concentrations of peptide tyrosine tyrosine, neurotensin, insulin, pancreatic glucagon, enteroglucagon, pancreatic polypeptide, gastric inhibitory polypeptide, motilin, peptide histidine methionine, gastrin, and somatostatin were measured by established radioimmunoassays as previously described (14-17). Details of the galanin assay have been reported elsewhere (18). In plasma the sensitivity of
f g
90
ii -c
80
E
c E E dE ; D(
70 60 50 40
Figure
Emptying
261
0
10
20
30 minutes
40
50
60
1. Gastric emptying measured by the percentage of isotope technetium 99m colloid (Pertechnetat) remaining in the stomach during a 60-min infusion of 40 pmoll kg . min synthetic porcine galanin (solid circles] compared with control saline (open circles) in 8 healthy volunteers after a solid mixed meal (530 kcal). Results expressed as mean + SEM for each time point. [The areas under the curve (AUC) were significantly different, p < 0.02.] After 1 h in controls 48.9% + 3.4% of radioactivity remained in the stomach compared with 78.5% + 2.4% during galanin infusion.
the galanin variability
assay was 10 pmol/L, with an intraassay of 7.5% + 0.4% and an interassay variability of
12.7% * 0.4% (mean ? SEM; n = 12).
Statistical
Analysis
Gastric emptying data, mouth-to-cecum travsit time, and plasma peptide concentrations are expressed as mean and standard error of the mean (mean c SEM). The Wilcoxon paired test was used to assess statistical significance of the galanin infusion data compared with control infusion data. This analysis was performed directly on the mouth-to-cecum transit time data, on the area under the curves for the gastric emptying data, and on the incremental integrated responses (i.e., the area under the curve minus baseline values) for each peptide.
Results All 8 volunteers experienced transitory bitter taste and hypersalivation during the galanin infusion. Two volunteers described mild abdominal discomfort during the last 30 min of infusion. Galanin was undetectable in basal plasma and in all plasma samples throughout the saline infusion. During the galanin infusion plasma concentrations reached a plateau value of 634 -+ 51 pmol/L but fell to an undetectable level 15 min after the infusion ceased.
Action of Galanin Motility
on Gastrointestinal
The rate of gastric emptying was significantly delayed by infusion of galanin when compared with saline infusion (p < 0.02) [Figure 1). The time to 50% decline of radioactivity over the stomach area increased from 59.0 -+ 4.8 min during the control study
262
BAUER ET AL.
GASTROENTEROLOGY
to 99.3 ? 4.7 min during the 60-min galanin infusion. One hour after the orange juiceilactuloselisotope drink, 48.9% ? 3.4% of the initial counts remained in the area of the stomach during the control test, whereas 78.5% + 2.4% remained when galanin was infused. Mouth-to-cecum transit time increased markedly from 67.5 + 6.9 min (control) to 126.3 2 18.5 min (galanin infusion) (p < 0.02). There was no individual correlation between 50% gastric emptying and mouth-to-cecum transit time. Action of Galanin on Peptides the Gastrointestinal Tract
Localized
Vol. 97, No. 2
P
in 30
Galanin strongly suppressed the postprandial rise in the plasma concentrations of peptide tyrosine tyrosine, neurotensin, somatostatin, and enteroglucagon. The plasma peptide concentrations before, during, and after a 60-min infusion of galanin (40 pmol/kg . min) or saline are shown in Figure 2. The incremental integrated responses, percentage of plasma peptide reduction, and the significance levels are presented in Table 1. No significant influence of galanin was seen on the plasma levels of gastric inhibitory polypeptide, motilin, peptide histidine methionine, and gastrin. Actions of Galanin the Pancreas
on Peptides
Localized
20
Galanin NaCl
10
p
z-3
in
During the galanin infusion the responses of insulin, pancreatic glucagon, and pancreatic polypeptide to the mixed meal were markedly blunted compared with controls. The plasma concentrations, incremental integrated responses, percentage of peptide suppression, and the significance levels of the changes are shown in Figure 3 and Table 1. The lower insulin level was associated with a significantly lower plasma glucose concentration (p < 0.05) which lasted throughout the 60-min galanin infusion (Figure 3). In summary, where a change was shown, the effect of galanin infusion on intestinal and pancreatic peptides was inhibitory. Discussion After its isolation from the gut, galanin was subsequently localized to the enteric nervous system (8-11). Galanin nerves in the rat were found to be predominantly intrinsic in origin as they were not destroyed by extrinsic denervation (10). In humans, galanin immunoreactive nerve fibers were found scattered throughout the stomach. In the small intestine more fibers were found in the longitudinal than
Figure
2. Plasma concentration of neurotensin, enteroglucagon, peptide YY, and somatostatin before, during, and after pora 60-min infusion of 40 pmolikg . min synthetic cine galanin (solid circles) compared with control saline (open circles] in 8 healthy volunteers after a mixed meal (530 kcal). Results are expressed as mean 2 SEM. Bar indicates infusion period.
in the circular muscle, the greatest number being in the myenteric plexus. Galanin has also been reported to be colocalized with vasoactive intestinal polypeptide (10). Currently, the function of the galaninergic innervation is not clear, but in view of its location and the inhibitory action of exogenous galanin it would seem possible that it is involved in the regulation of motility. The inhibitory actions of galanin on gastrointestinal motility in humans seen in this study are supported by a number of animal studies. These pharmacologic experiments on animals, both in vivo and in vitro, using synthetic porcine galanin suggest possible mechanisms for this motility effect. Galanin injected into dogs caused inhibition of intestinal phasic activity that was unaffected by treatment with atropine, hexamethonium, yohimbine, or naloxone and was therefore considered to be a direct effect of galanin on smooth muscle (2). The acetylcholine output produced by electrical stimulation,
August 1989
GALANIN AND GI MOTILITY
Table 1. Action of Galanin on Peptides in Gastrointestinal IIR (pmol/L
PYY
Neurotensin Somatostatin Enteroglucagon Insulin Pancreatic glucagon Pancreatic polypeptide
2.63 10.60 1.91 19.85 74.61 1.89 90.43
Tract and Pancreas
15 min)
Control
Galanin
+ 1.18 2 4.53 + 2.75 ‘- 9.37 2 14.67 2 1.23 +- 32.31
263
-3.50 -1.88 -3.26 -11.88 -11.20 0.55 -24.0’2
2 ? ? + + 2 +
0.97 1.16 2.01 5.52 6.26 1.35 18.03
Reduction (%I 235 118 273 160 115 71 127
Significance level p p p p p p p
< < < < < < <
0.02 0.02 0.05 0.02 0.02 0.05 0.02
IIR, incremental integrated responses; PYY, peptide tyrosine tyrosine. Significant changes in plasma peptide concentrations in 8 healthy volunteers during a 60-min infusion of 40 pmol/kg min of synthetic porcine galanin compared with control saline after a solid mixed meal (530 kcal). Results are expressed as incremental integrated responses (mean + SEM) showing net changes of plasma peptide concentrations after correction for baseline values. The percentage reduction of plasma peptide concentrations and the significance levels are shown.
vasoactive intestinal polypeptide, or by substance P from guinea pig small intestinal muscle strips was significantly depressed by galanin, suggesting that
pco.05 pmol/l ‘““1
insulin
300
Galanin Na Cl
200
pco.02 pmol/l 250 1 PP
Galanin NaCl
10
1
infusion period
PCO.05
-15
Figure 3. Plasma concentration of glucose, insulin, pancreatic polypeptide, and pancreatic glucagon before, during, and after a 60-min infusion of 40 pmol/kg min synthetic porcine galanin (solid circles) compared with control saline (open circles) in 8 healthy volunteers after a solid mixed meal (530 kcal). Values are mean * SEM. Bar indicates infusion period.
galanin has an inhibitory effect on myenteric cholinergic neurons (19). In view of the colocalization of galanin and vasoactive intestinal polypeptide in gut neurons (lo), it is possible to speculate that their opposing actions may provide a local regulatory mechanism for acetylcholine release. Finally, galanin has been reported to modulate synaptic transmission. A number of electrophysiologic studies on myenteric neurons of the guinea pig intestine have shown that (a) galanin mimics slow inhibitory postsynaptic potentials (20,21), (b) galanin abolishes fast excitatory postsynaptic potentials and suppresses the excitatory postsynaptic potentials-like response to acetylcholine, suggesting an inhibition of synaptic transmission at nicotinic cholinergic synapses (22), and (c) galanin blocks both fast and slow excitatory postsynaptic potentials, suppressing the excitation evoked by histamine, vasoactive intestinal polypeptide, acetylcholine, forskolin, substance P, and calcitonin gene-related peptide (21). The reduction in the postprandial plasma concentrations of gastrointestinal peptides caused by the galanin infusion may possibly be secondary to the inhibition of gastric emptying and intestinal motility. It is of significance that the rise in plasma glucose after a meal was actually lower during the galanin infusion, so that the lower insulin response could be considered appropriate for the delayed glucose absorption. The reduced response of neurotensin, peptide tyrosine tyrosine, and enteroglucagon could also be explained by slower transit producing lower concentrations of nutrients in the intestine. However, it cannot be excluded that these changes in the postprandial pattern of gastrointestinal peptides may be due, at least in part, to a direct or indirect action of galanin on endocrine cells. The reduced incremental responses of insulin, glucagon, and pancreatic polypeptide seen in this study are of interest in the light of the hypothesis that galanin
264
BALJER ET AL.
may be involved in the regulation of islet cell hormones. A number of animal studies have shown that galanin acts on p-cells, causing an inhibition of basal and stimulated insulin secretion (23-26). In dogs, a galanin infusion significantly reduced the insulin response to oral glucose or a mixed meal (4). However, it is not clear that the dose of galanin used in the experiments reported here has the same effect in humans. A previous study showed that a galanin infusion caused no significant change in plasma insulin concentrations in humans after intravenous glucose (7). In conclusion, the distribution of galaninlike immunoreactivity in nerves of the stomach and small intestine and the marked effect that a pharmacologic dose of galanin has on motor activity, suggests that galanin may be involved in the physiologic control of human gastrointestinal motility. A therapeutic role for galanin, or an analogue, in hypermotility disorders or diarrhea associated with ileostomy, dumping, and short bowel syndrome could appear worthy of investigation. References 1. Tatemoto K, Rokaeus A, Jornvall H, McDonald TJ, Mutt V. Galanin-a novel biologically active peptide from porcine intestine. Febs Lett 1983;164:124-8. 2. Fox JET, McDonald TJ, Kostolanska F, Tatemoto K. Galanin: an inhibitory neural peptide of the canine small intestine. Life Sci 1986;39:103-10. 3. McDonald TJ, Dupre J, Tatemoto K, Greenberg GR, Radziuk J, Mutt V. Galanin inhibits insulin secretion and induces hyperglycemia in dogs. Diabetes 1985;34:192-6. 4. McDonald TJ, Dupre J, Greenberg GR, et al. The effect of galanin on canine plasma glucose and gastroenteropancreatic hormone response to oral nutrients and intravenous arginine. Endocrinology 1986;119:2340-5. 5. Manabe T, Yoshimura T, Kii E, et al. Galanin-induced hyperglycemia: effect on insulin and glucagon. Endocr Res 1986; 12:93-8. 6. Lindskog S, Ahren B. Galanin: effects on basal and stimulated insulin and glucagon secretion in the mouse. Acta Physiol Stand 1987;129:305-9. 7. Bauer FE, Ginsberg L, Venetikou M, MacKay DJ, Burrin JM, Bloom SR. Growth hormone release in man induced by galanin, a new hypothalamic peptide. Lancet 1986;ii:192-5. 8. Rokaeus A, Melander T, Hokfelt T, et al. A galanin-like peptide in the central nervous system and intestine of the rat. Neurosci Lett 1984;47:161-6. 9. Ekblad E, Rokaeus A, Hakanson R, Sundler F. Galanin nerve fibres in the rat gut: distribution, origin and projections. Neuroscience 1985;16:355-63. 10. Bishop AE, Polak JM, Bauer FE, Christofides ND, Carlei F, Bloom SR. Occurrence and distribution of a newly discovered peptide, galanin, in the mammalian enteric nervous system. Gut 1986;27:849-57. 11. Melander T, Hokfelt T, Rokaeus A, Fahrenkrug J, Tatemoto K, Mutt V. Distribution of galanin-like immunoreactivity in the
GASTROENTEROLOGY
gastrointestinal tract of several Tissue Res 1985:239:253-70.
mammalian
Vol. 97, No. 2
species.
Cell
12. Bond JH, Levitt MD. Investigation of small bowel transit time in man utilising pulmonary breath hydrogen (H,) measurements J Lab Clin Med 1975;85:546-55. 13. Read NW, Al-Janabi MN, Bates TE, et al. Interpretation of the breath hydrogen profile obtained after ingesting a solid meal containing unabsorbable carbohydrate. Gut 1985;26:834-42. 14. Bloom SR, Long RG. Radioimmunoassay of gut regulatory peptides. London, Philadelphia, Toronto: WB Saunders, 1982. Y, Williams SJ, Bishop AE, Polak JM, Bloom SR. 15. Yiangou Peptide
histidine-methionine
immunoreactivity
in plasma
and tissue from patients with vasoactive intestinal peptidesecreting tumours and watery diarrhea syndrome. J Clin Endocrinol Metab 1987;64:131-9. 16. Adrian TE, Savage AP, Bacarese-Hamilton AJ, Wolfe K, Besterman HS, Bloom SR. Peptide YY abnormalities in gastrointestinal diseases. Gastroenterology 1986;90:379-84. 17. Albano JDM, Ekins RP, Maritz G, Turner RC. A sensitive, precise radioimmunoassay of serum insulin relying on charcoal separation of bound and free hormone moieties. Acta Endocrinol (Copenh) 1972;70:487-509. ND, et al. Distribution and 18. Bauer FE, Adrian TE, Christofides molecular heterogeneity of galanin in human, pig, guinea pig, and rat gastrointestinal tracts. Gastroenterology 1986;91:87783. 19. Yau WM, Dorsett JA, Youther ML. Evidence for galanin as an inhibitory neuropeptide on myenteric cholinergic neurons (abstr). Gastroenterology 1986;91:1072. 20. Palmer JM, Schemann M, Tamura K, Wood JD. Galanin mimics slow synaptic inhibition in myenteric neurons. Eur J Pharmacol 1986;124:379-80. M, Winkelmann C, Wood 21. Tamura K, Palmer JM, Schemann JD. Mechanism of inhibitory action of galanin on myenteric neurones of guinea-pig small intestine. Dig Dis Sci 1987; 32:929. 22. Tamura K, Palmer JM, Wood JD. Galanin suppresses nicotinic synaptic transmission in the myenteric plexus of guinea-pig small intestine. Eur J Pharmacol 1987;136:445-6. 23. Ahren B, Arkhammar P, Berggren PO. Galanin inhibits glucose-stimulated insulin release by a mechanism involving hyperpolarization and lowering of cytoplasmatic free Ca’+ concentration. Biochem Biophys Res Commun 1986;140: 1059-63. A, Timmers K. A direct inhibitory effect of the 24. Rokaeus peptide galanin on insulin secretion. J Clin Endocrinol Metab 1987;64:27. 25. Dunning BE, Ahren B, Veith RC, Bottcher G, Sundler F, Taborsky GJ. Galanin: a novel pancreatic neuropeptide. Am J Physiol 1986;251:E127-33. ML, 26. Silvestre RA. Monge L, Miralles P, Moreno P, Villanueva Marco J. Effect of galanin, a new intestinal peptide, on pancreatic hormone secretion by the perfused rat pancreas (abstr). Diabetologia 1986;29:594A.
Received March 14, 1988. Accepted January 18, 1989. Address requests for reprints to: Professor F.E. Bauer, Department of Clinical Pharmacology, Georg-August-Universitat Gottingen, Robert-Koch-Strasse 40, 3400 Gottingen, Federal Republic of Germany. The authors thank A. Mohnhaupt for statistical advice.
1989:97:288-l
Inhibitory Effect of Galanin on Postprandial Gastrointestinal Motility and Gut Hormone Release in Humans F. E. BAUER, A. ZINTEL, and S. R. BLOOM
M. J. KENNY,
D. CALDER,
M. A. GHATEI,
Departments of Gastroenterology and Radiology, Universittitsklinikum Charlottenburg, Freie Universitst Berlin, Berlin, Federal Republic of Germany; and Department of Medicine, Royal Postgraduate Medical School, Hammersmith Hospital, London, United Kingdom
Galanin was infused intravenously in 8 healthy volunteers at a dose of 40 pmol/kg . min for 1 h to investigate the pharmacologic effects of this peptide on postprandial gastrointestinal motility and gut peptide release in humans. Galanin strongly inhibited gastrointestinal motility. Gastric emptying was significantly delayed, with the time taken to empty 50% of the gastric contents increasing from 59.0 f 4.8 min (control infusion) to 99.3 + 4.7 min (galanin infusion). Mouth-to-cecum transit time increased from 87.5 f 8.9 to 128.3 + 18.5 min. Galanin potently suppressed the initial postprandial rise in plasma concentrations of glucose, insulin, peptide tyrosine tyrosine, neurotensin, enteroglucagon, pancreatic glucagon, somatostatin, and pancreatic polypeptide, but did not change gastric inhibitory polypeptide, motilin, peptide histidine methionine, and gastrin concentrations compared with control. The results indicate that an infusion of galanin has potent effects on the gastrointestinal tract in humans. The changes in motor activity in particular suggest that the local galaninergic innervation could have an important physiologic role in the control of human gastrointestinal propulsive motor activity.
Gfrom porcine
alanin is a 2%amino acid peptide first isolated intestine by a chemical technique that identifies C-terminal amidated peptides (1). The peptide contracts smooth muscle strips from gastric fundus, ileum, and colon of the rat (1) and inhibits canine small intestinal muscle both in vivo and in vitro (2). It was subsequently found that galanin potently inhibits insulin release in animals (3-6) but not in humans (7). Galanin is localized exclusively to nerves, and a dense network of galanin immunoreactive nerve fibers have been reported in the gas-
trointestinal tract, being particularly numerous in the small intestine of several mammalian species including humans (8-11). The localization of galanin to the enteric nervous plexus and its pharmacologic effect on gastrointestinal smooth muscle led us to investigate the effect of the peptide on human gastrointestinal motility, using the isotope method as an index of gastric emptying, and mouth-to-cecum transit time as reflected by the time to postprandial breath hydrogen rise. In addition, we examined the action of galanin on the pattern of postprandial gut peptide release.
Materials and Methods Subjects Eight healthy volunteers (6 men, 2 women, aged 22-32 yr, weight 48-73 kg) were each studied on two occasions at least 1 wk apart. All subjects gave written informed consent and the protocol was approved by the local Ethics Committee of the Universitatsklinikum Charlottenburg. After an overnight fast, subjects were seated upright in front of a y-camera. An indwelling cannula was inserted into a vein of each forearm. Each volunteer underwent two studies, receiving in random order an infusion of 0.9% saline (control) or synthetic galanin at a dose of 40 pmol/kg . min for 60 min. The peptide was dissolved in 50 ml of saline containing 1% human serum albumin immediately before the infusion. The subjects were unaware of the nature of each infusion. Five minutes after the start of the infusion subjects ate a solid mixed meal consisting of 90 g of white bread, 10 g of butter, two boiled eggs, and 200 ml of unsweetened orange juice (62 g carbohydrate, 23 g protein, and 23 g fat; 530 kcal] within 5 min. The orange juice contained 25 g of lactulose (Bifiteral;
0 1989 by the American
Gastroenterological
0018-5085/89/$3.50
Association
August 1989
GALANIN AND GI MOTILITY
100
Duphar, Hannover, F.R.G.) serving as substrate for colonic bacteria to produce hydrogen, and 400 $i (14.8 MBq) technetium 99m (t,,, = 6 h) colloid (Pertechnetat; Hoechst, Frankfurt, F.R.G.). The orange juice was drunk immediately after eating the solid meal. Blood samples were collected 15 min before the start of the infusion and thereafter at 15min intervals throughout the 3-h experimental period.
Gastric
Gamma camera (Apex 415 Elscint, Tel Aviv, Israel) scintigraphic data were collected at l-min intervals and stored on magnetic tape for later processing by an on-line computer. The stomach area was defined from the distribution of activity during the first 4 min after drinking the isotope by two independent observers unaware of the nature of the infusion. Collection of data started immediately after ingestion of the radioactive drink and was continued for 180 min. The rate of gastric emptying was calculated as the rate of loss of radioactivity’from the stomach area, after correction for isotope decay, computed against time. The 50% point of gastric emptying was defined as the time taken for radioactivity over the stomach area to decline to 50% of the baseline maximum. In addition, the percentage of radioactivity remaining in the stomach after 60 min (i.e., at the end of the infusion period) was calculated.
Mouth-to-Cecum
Transit
Time
Mouth-to-cecum transit time was estimated by the breath hydrogen technique (12,13). Twenty-milliliter samples of end-tidal exhaled air were collected using a 20-ml syringe with a special three-way tap. Collection was performed at 5-min intervals from -15 to 180 min and samples were analyzed for hydrogen content using an electronic monitor (HZ-Exhalation Analyzer; Stimotron, Wendelstein, F.R.G.). The time taken for the breath hydrogen concentration to exceed the baseline value by at least 3 ppm for three consecutive readings was taken as a measure of mouth-to-cecum transit time.
Peptide
Assays
Ten-milliliter blood samples were collected from the contralateral cannula at 15-min intervals from -15 to 180 min into lithium-heparin tubes containing 4000 KIU aprotinin (Trasylol; Bayer AG, Leverkusen, F.R.G.). Plasma was separated by immediate centrifugation for 10 min at 3000 rpm and stored at -20°C until assayed. Aprotinin effectively prevented enzymatic degradation of galanin in plasma for up to 4 h (unpublished observation). Plasma concentrations of peptide tyrosine tyrosine, neurotensin, insulin, pancreatic glucagon, enteroglucagon, pancreatic polypeptide, gastric inhibitory polypeptide, motilin, peptide histidine methionine, gastrin, and somatostatin were measured by established radioimmunoassays as previously described (14-17). Details of the galanin assay have been reported elsewhere (18). In plasma the sensitivity of
f g
90
ii -c
80
E
c E E dE ; D(
70 60 50 40
Figure
Emptying
261
0
10
20
30 minutes
40
50
60
1. Gastric emptying measured by the percentage of isotope technetium 99m colloid (Pertechnetat) remaining in the stomach during a 60-min infusion of 40 pmoll kg . min synthetic porcine galanin (solid circles] compared with control saline (open circles) in 8 healthy volunteers after a solid mixed meal (530 kcal). Results expressed as mean + SEM for each time point. [The areas under the curve (AUC) were significantly different, p < 0.02.] After 1 h in controls 48.9% + 3.4% of radioactivity remained in the stomach compared with 78.5% + 2.4% during galanin infusion.
the galanin variability
assay was 10 pmol/L, with an intraassay of 7.5% + 0.4% and an interassay variability of
12.7% * 0.4% (mean ? SEM; n = 12).
Statistical
Analysis
Gastric emptying data, mouth-to-cecum travsit time, and plasma peptide concentrations are expressed as mean and standard error of the mean (mean c SEM). The Wilcoxon paired test was used to assess statistical significance of the galanin infusion data compared with control infusion data. This analysis was performed directly on the mouth-to-cecum transit time data, on the area under the curves for the gastric emptying data, and on the incremental integrated responses (i.e., the area under the curve minus baseline values) for each peptide.
Results All 8 volunteers experienced transitory bitter taste and hypersalivation during the galanin infusion. Two volunteers described mild abdominal discomfort during the last 30 min of infusion. Galanin was undetectable in basal plasma and in all plasma samples throughout the saline infusion. During the galanin infusion plasma concentrations reached a plateau value of 634 -+ 51 pmol/L but fell to an undetectable level 15 min after the infusion ceased.
Action of Galanin Motility
on Gastrointestinal
The rate of gastric emptying was significantly delayed by infusion of galanin when compared with saline infusion (p < 0.02) [Figure 1). The time to 50% decline of radioactivity over the stomach area increased from 59.0 -+ 4.8 min during the control study
262
BAUER ET AL.
GASTROENTEROLOGY
to 99.3 ? 4.7 min during the 60-min galanin infusion. One hour after the orange juiceilactuloselisotope drink, 48.9% ? 3.4% of the initial counts remained in the area of the stomach during the control test, whereas 78.5% + 2.4% remained when galanin was infused. Mouth-to-cecum transit time increased markedly from 67.5 + 6.9 min (control) to 126.3 2 18.5 min (galanin infusion) (p < 0.02). There was no individual correlation between 50% gastric emptying and mouth-to-cecum transit time. Action of Galanin on Peptides the Gastrointestinal Tract
Localized
Vol. 97, No. 2
P
in 30
Galanin strongly suppressed the postprandial rise in the plasma concentrations of peptide tyrosine tyrosine, neurotensin, somatostatin, and enteroglucagon. The plasma peptide concentrations before, during, and after a 60-min infusion of galanin (40 pmol/kg . min) or saline are shown in Figure 2. The incremental integrated responses, percentage of plasma peptide reduction, and the significance levels are presented in Table 1. No significant influence of galanin was seen on the plasma levels of gastric inhibitory polypeptide, motilin, peptide histidine methionine, and gastrin. Actions of Galanin the Pancreas
on Peptides
Localized
20
Galanin NaCl
10
p
z-3
in
During the galanin infusion the responses of insulin, pancreatic glucagon, and pancreatic polypeptide to the mixed meal were markedly blunted compared with controls. The plasma concentrations, incremental integrated responses, percentage of peptide suppression, and the significance levels of the changes are shown in Figure 3 and Table 1. The lower insulin level was associated with a significantly lower plasma glucose concentration (p < 0.05) which lasted throughout the 60-min galanin infusion (Figure 3). In summary, where a change was shown, the effect of galanin infusion on intestinal and pancreatic peptides was inhibitory. Discussion After its isolation from the gut, galanin was subsequently localized to the enteric nervous system (8-11). Galanin nerves in the rat were found to be predominantly intrinsic in origin as they were not destroyed by extrinsic denervation (10). In humans, galanin immunoreactive nerve fibers were found scattered throughout the stomach. In the small intestine more fibers were found in the longitudinal than
Figure
2. Plasma concentration of neurotensin, enteroglucagon, peptide YY, and somatostatin before, during, and after pora 60-min infusion of 40 pmolikg . min synthetic cine galanin (solid circles) compared with control saline (open circles] in 8 healthy volunteers after a mixed meal (530 kcal). Results are expressed as mean 2 SEM. Bar indicates infusion period.
in the circular muscle, the greatest number being in the myenteric plexus. Galanin has also been reported to be colocalized with vasoactive intestinal polypeptide (10). Currently, the function of the galaninergic innervation is not clear, but in view of its location and the inhibitory action of exogenous galanin it would seem possible that it is involved in the regulation of motility. The inhibitory actions of galanin on gastrointestinal motility in humans seen in this study are supported by a number of animal studies. These pharmacologic experiments on animals, both in vivo and in vitro, using synthetic porcine galanin suggest possible mechanisms for this motility effect. Galanin injected into dogs caused inhibition of intestinal phasic activity that was unaffected by treatment with atropine, hexamethonium, yohimbine, or naloxone and was therefore considered to be a direct effect of galanin on smooth muscle (2). The acetylcholine output produced by electrical stimulation,
August 1989
GALANIN AND GI MOTILITY
Table 1. Action of Galanin on Peptides in Gastrointestinal IIR (pmol/L
PYY
Neurotensin Somatostatin Enteroglucagon Insulin Pancreatic glucagon Pancreatic polypeptide
2.63 10.60 1.91 19.85 74.61 1.89 90.43
Tract and Pancreas
15 min)
Control
Galanin
+ 1.18 2 4.53 + 2.75 ‘- 9.37 2 14.67 2 1.23 +- 32.31
263
-3.50 -1.88 -3.26 -11.88 -11.20 0.55 -24.0’2
2 ? ? + + 2 +
0.97 1.16 2.01 5.52 6.26 1.35 18.03
Reduction (%I 235 118 273 160 115 71 127
Significance level p p p p p p p
< < < < < < <
0.02 0.02 0.05 0.02 0.02 0.05 0.02
IIR, incremental integrated responses; PYY, peptide tyrosine tyrosine. Significant changes in plasma peptide concentrations in 8 healthy volunteers during a 60-min infusion of 40 pmol/kg min of synthetic porcine galanin compared with control saline after a solid mixed meal (530 kcal). Results are expressed as incremental integrated responses (mean + SEM) showing net changes of plasma peptide concentrations after correction for baseline values. The percentage reduction of plasma peptide concentrations and the significance levels are shown.
vasoactive intestinal polypeptide, or by substance P from guinea pig small intestinal muscle strips was significantly depressed by galanin, suggesting that
pco.05 pmol/l ‘““1
insulin
300
Galanin Na Cl
200
pco.02 pmol/l 250 1 PP
Galanin NaCl
10
1
infusion period
PCO.05
-15
Figure 3. Plasma concentration of glucose, insulin, pancreatic polypeptide, and pancreatic glucagon before, during, and after a 60-min infusion of 40 pmol/kg min synthetic porcine galanin (solid circles) compared with control saline (open circles) in 8 healthy volunteers after a solid mixed meal (530 kcal). Values are mean * SEM. Bar indicates infusion period.
galanin has an inhibitory effect on myenteric cholinergic neurons (19). In view of the colocalization of galanin and vasoactive intestinal polypeptide in gut neurons (lo), it is possible to speculate that their opposing actions may provide a local regulatory mechanism for acetylcholine release. Finally, galanin has been reported to modulate synaptic transmission. A number of electrophysiologic studies on myenteric neurons of the guinea pig intestine have shown that (a) galanin mimics slow inhibitory postsynaptic potentials (20,21), (b) galanin abolishes fast excitatory postsynaptic potentials and suppresses the excitatory postsynaptic potentials-like response to acetylcholine, suggesting an inhibition of synaptic transmission at nicotinic cholinergic synapses (22), and (c) galanin blocks both fast and slow excitatory postsynaptic potentials, suppressing the excitation evoked by histamine, vasoactive intestinal polypeptide, acetylcholine, forskolin, substance P, and calcitonin gene-related peptide (21). The reduction in the postprandial plasma concentrations of gastrointestinal peptides caused by the galanin infusion may possibly be secondary to the inhibition of gastric emptying and intestinal motility. It is of significance that the rise in plasma glucose after a meal was actually lower during the galanin infusion, so that the lower insulin response could be considered appropriate for the delayed glucose absorption. The reduced response of neurotensin, peptide tyrosine tyrosine, and enteroglucagon could also be explained by slower transit producing lower concentrations of nutrients in the intestine. However, it cannot be excluded that these changes in the postprandial pattern of gastrointestinal peptides may be due, at least in part, to a direct or indirect action of galanin on endocrine cells. The reduced incremental responses of insulin, glucagon, and pancreatic polypeptide seen in this study are of interest in the light of the hypothesis that galanin
264
BALJER ET AL.
may be involved in the regulation of islet cell hormones. A number of animal studies have shown that galanin acts on p-cells, causing an inhibition of basal and stimulated insulin secretion (23-26). In dogs, a galanin infusion significantly reduced the insulin response to oral glucose or a mixed meal (4). However, it is not clear that the dose of galanin used in the experiments reported here has the same effect in humans. A previous study showed that a galanin infusion caused no significant change in plasma insulin concentrations in humans after intravenous glucose (7). In conclusion, the distribution of galaninlike immunoreactivity in nerves of the stomach and small intestine and the marked effect that a pharmacologic dose of galanin has on motor activity, suggests that galanin may be involved in the physiologic control of human gastrointestinal motility. A therapeutic role for galanin, or an analogue, in hypermotility disorders or diarrhea associated with ileostomy, dumping, and short bowel syndrome could appear worthy of investigation. References 1. Tatemoto K, Rokaeus A, Jornvall H, McDonald TJ, Mutt V. Galanin-a novel biologically active peptide from porcine intestine. Febs Lett 1983;164:124-8. 2. Fox JET, McDonald TJ, Kostolanska F, Tatemoto K. Galanin: an inhibitory neural peptide of the canine small intestine. Life Sci 1986;39:103-10. 3. McDonald TJ, Dupre J, Tatemoto K, Greenberg GR, Radziuk J, Mutt V. Galanin inhibits insulin secretion and induces hyperglycemia in dogs. Diabetes 1985;34:192-6. 4. McDonald TJ, Dupre J, Greenberg GR, et al. The effect of galanin on canine plasma glucose and gastroenteropancreatic hormone response to oral nutrients and intravenous arginine. Endocrinology 1986;119:2340-5. 5. Manabe T, Yoshimura T, Kii E, et al. Galanin-induced hyperglycemia: effect on insulin and glucagon. Endocr Res 1986; 12:93-8. 6. Lindskog S, Ahren B. Galanin: effects on basal and stimulated insulin and glucagon secretion in the mouse. Acta Physiol Stand 1987;129:305-9. 7. Bauer FE, Ginsberg L, Venetikou M, MacKay DJ, Burrin JM, Bloom SR. Growth hormone release in man induced by galanin, a new hypothalamic peptide. Lancet 1986;ii:192-5. 8. Rokaeus A, Melander T, Hokfelt T, et al. A galanin-like peptide in the central nervous system and intestine of the rat. Neurosci Lett 1984;47:161-6. 9. Ekblad E, Rokaeus A, Hakanson R, Sundler F. Galanin nerve fibres in the rat gut: distribution, origin and projections. Neuroscience 1985;16:355-63. 10. Bishop AE, Polak JM, Bauer FE, Christofides ND, Carlei F, Bloom SR. Occurrence and distribution of a newly discovered peptide, galanin, in the mammalian enteric nervous system. Gut 1986;27:849-57. 11. Melander T, Hokfelt T, Rokaeus A, Fahrenkrug J, Tatemoto K, Mutt V. Distribution of galanin-like immunoreactivity in the
GASTROENTEROLOGY
gastrointestinal tract of several Tissue Res 1985:239:253-70.
mammalian
Vol. 97, No. 2
species.
Cell
12. Bond JH, Levitt MD. Investigation of small bowel transit time in man utilising pulmonary breath hydrogen (H,) measurements J Lab Clin Med 1975;85:546-55. 13. Read NW, Al-Janabi MN, Bates TE, et al. Interpretation of the breath hydrogen profile obtained after ingesting a solid meal containing unabsorbable carbohydrate. Gut 1985;26:834-42. 14. Bloom SR, Long RG. Radioimmunoassay of gut regulatory peptides. London, Philadelphia, Toronto: WB Saunders, 1982. Y, Williams SJ, Bishop AE, Polak JM, Bloom SR. 15. Yiangou Peptide
histidine-methionine
immunoreactivity
in plasma
and tissue from patients with vasoactive intestinal peptidesecreting tumours and watery diarrhea syndrome. J Clin Endocrinol Metab 1987;64:131-9. 16. Adrian TE, Savage AP, Bacarese-Hamilton AJ, Wolfe K, Besterman HS, Bloom SR. Peptide YY abnormalities in gastrointestinal diseases. Gastroenterology 1986;90:379-84. 17. Albano JDM, Ekins RP, Maritz G, Turner RC. A sensitive, precise radioimmunoassay of serum insulin relying on charcoal separation of bound and free hormone moieties. Acta Endocrinol (Copenh) 1972;70:487-509. ND, et al. Distribution and 18. Bauer FE, Adrian TE, Christofides molecular heterogeneity of galanin in human, pig, guinea pig, and rat gastrointestinal tracts. Gastroenterology 1986;91:87783. 19. Yau WM, Dorsett JA, Youther ML. Evidence for galanin as an inhibitory neuropeptide on myenteric cholinergic neurons (abstr). Gastroenterology 1986;91:1072. 20. Palmer JM, Schemann M, Tamura K, Wood JD. Galanin mimics slow synaptic inhibition in myenteric neurons. Eur J Pharmacol 1986;124:379-80. M, Winkelmann C, Wood 21. Tamura K, Palmer JM, Schemann JD. Mechanism of inhibitory action of galanin on myenteric neurones of guinea-pig small intestine. Dig Dis Sci 1987; 32:929. 22. Tamura K, Palmer JM, Wood JD. Galanin suppresses nicotinic synaptic transmission in the myenteric plexus of guinea-pig small intestine. Eur J Pharmacol 1987;136:445-6. 23. Ahren B, Arkhammar P, Berggren PO. Galanin inhibits glucose-stimulated insulin release by a mechanism involving hyperpolarization and lowering of cytoplasmatic free Ca’+ concentration. Biochem Biophys Res Commun 1986;140: 1059-63. A, Timmers K. A direct inhibitory effect of the 24. Rokaeus peptide galanin on insulin secretion. J Clin Endocrinol Metab 1987;64:27. 25. Dunning BE, Ahren B, Veith RC, Bottcher G, Sundler F, Taborsky GJ. Galanin: a novel pancreatic neuropeptide. Am J Physiol 1986;251:E127-33. ML, 26. Silvestre RA. Monge L, Miralles P, Moreno P, Villanueva Marco J. Effect of galanin, a new intestinal peptide, on pancreatic hormone secretion by the perfused rat pancreas (abstr). Diabetologia 1986;29:594A.
Received March 14, 1988. Accepted January 18, 1989. Address requests for reprints to: Professor F.E. Bauer, Department of Clinical Pharmacology, Georg-August-Universitat Gottingen, Robert-Koch-Strasse 40, 3400 Gottingen, Federal Republic of Germany. The authors thank A. Mohnhaupt for statistical advice.