Cell Tissue Res (1992) 269:275-279
Cell&Tissue Research 9 Springer-Verlag 1992
Pituitary adenylate cyclase-activating polypeptide (PACAP): occurrence in rodent pancreas and effects on insulin and glucagon secretion in the mouse Tord Fridolf1, Frank Sundler 2, and Bo Ahr~n 3 1 Department of Pharmacology, S61vegatan 10, a Department of Medical Cell Research, Biskopsgatan 5, and 3 Department of Surgery, Lund University, S-223 62 Lund, Sweden Received November 18, 1991 / Accepted March 31, 1992
Summary. Pituitary adenylate cyclase-activating polypeptide (PACAP) is a neuropeptide that occurs in several tissues, e.g., in the gut. We have studied PACAP-like immunoreactivity in the pancreas o f rat and mouse, and the effects of PACAP-38 on basal and stimulated insulin and glucagon secretion in the mouse. Immunofluorescence staining demonstrated the presence of PACAP-like immunoreactivity in nerve fibers in both the rat and mouse pancreas. The nerve fibers were seen in the exocrine pancreas and surrounding the islets. Occasionally, the nerve fibers occurred within the islets. Most PACAPpositive nerve fibers innervated the intrapancreatic ganglia, although no nerve cell bodies contained PACAPlike immunoreactivity. In-vivo experiments in mice revealed that basal plasma glucagon levels were increased by PACAP-38 injected intravenously at dose levels exceeding 1.8 nmol/kg. Furthermore, PACAP-38 (7 nmol/ kg) potentiated the plasma glucagon response to the cholinergic agonist carbachol (0.16 gmol/kg). This potentiation was reduced to simple addition by pretreatment with a combined e- and/%adrenergic blockade by phentolamine (35 gmol/kg) and propranolol (8.5 gmol/ kg). Moreover, PACAP-38 inhibited a carbachol-induced increase in the level of plasma insulin in the absence but not in the presence of adrenergic blockade. PACAP-38 increased basal plasma insulin levels and increased basal plasma glucose levels 6 rain and 10 rain, respectively, after injection of the peptide. We conclude that PACAP-like immunoreactivity exists in nerve fibers innervating the mouse and rat pancreas, particularly the intrapancreatic ganglia, and that PACAP-38 augments both basal and carbachol-stimulated glucagon secretion in the mouse. Key words: Pituitary adenylate cyclase-activating peptide (PACAP) - Immunocytochemistry - Pancreas, endocrine, exocrine - Insulin secretion - Glucagon secretion - Mouse (NMRI) - Rat (Sprague-Dawley) Correspondence to: T. Fridolf
Pituitary adenylate cyclase-activating polypeptide (PACAP)-38 and PACAP-27, which corresponds to the N-terminal 27 amino acid residues of PACAP-38, have recently been isolated from ovine hypothalamic tissue and shown to stimulate adenylate cyclase in rat pituitary cell cultures (Miyata et al. 1989, 1990). The N-terminal 28 amino acids in PACAP-38 bear a structural relationship to the peptides belonging to the glucagon peptide superfamily, i.e., vasoactive intestinal peptide (VIP), peptide histidine isoleucine (PHI), secretin, helospectin I and II, helodermin, and glucagon (Parker et al. 1984; Robberecht et al. 1985; Ahr6n and Lundquist 1988; Ahr6n and Falck 1991). VIP, PHI, secretin and helospectin I have previously been shown to stimulate glucagon secretion markedly in mice, whereas they exert only slight influences on insulin secretion (Ahr6n and Falck 1991; Ahr6n and Lundquist 1986, 1988; Ahr~n 1991). We have now studied whether PACAP-38 also affects insulin and glucagon secretion in vivo, in mice. Since it has been shown that PACAP-like immunoreactivity occurs in peripheral tissues, e.g., in the respiratory tract (Uddman et al. 1991) and the gut (Sundler et al. 1992), we have also studied whether PACAP-like immunoreactivity occurs in the pancreas of the rat and the mouse.
Materials and methods Animals Female mice of the NMRI strain, weighing 25-30 g, and male Sprague-Dawley rats, weighing 200-250 g (both species from ALAB, Stockholm, Sweden), were used. The animals were fed on a standard pellet diet (Astra-Ewos, S6dertfilje, Sweden) and tap water ad libitum before and during the experiments.
Immunocytochemistry Tissue specimens of the duodenal and the splenic portions of the pancreas were taken from decapitated animals and fixed overnight by immersion in Stefanini's solution (2% formaldehyde and 0.2%
276
Fig.
l a-e. Sections from mouse pancreas immunostained for PACAP-27. Immunoreactive nerve fibers appear as delicate fiber elements running in the exocrine pancreas (a), as fibers located in nerve bundles (b), and as fibers penetrating into the islets (c). a x 2 5 0 ; b , c x300 Fig. 2. Section from rat pancreas immunostained for PACAP-27. Note immunoreactive nerve fibers in the exocrine pancreatic parenchyma. • 300 Fig. 3. Section from rat pancreas immunostained for PACAP-27. Immunoreactive nerve fibers innervate a ganglion, x 250
277 picric acid in 0.1 M phosphate-buffered saline, pH 7.2). After rinsing in sucrose-enriched (10%) buffer, the specimens were frozen on dry ice. Sections were cut at a thickness of 15 pm in a cryostat and collected on chromalum-coated slides. The sections were then processed for the immunocytochemical demonstration of PACAP using the indirect immunofluorescence technique (Coons et al. 1955). The PACAP antiserum (code 88121-3; kindly donated by Dr. A. Arimura, Tulane University, Belle Chasse, La., USA) was raised in a rabbit against synthetic ovine PACAP-27 (conjugated with thyroglobulin) and used at a dilution of 1:640. Control sections were incubated with PACAP antiserum preabsorbed with excess antigen (10 pg PACAP-27 per ml diluted antiserum). These sections did not show any immunoreactivity. The antiserum showed no cross-reactivity with glucagon, VIP, PHI, helodermin, helospectin I, growth hormone-releasing factor (GRF) or gastrinreleasing peptide (GRP) (Peninsula Europe, Merseyside, St. Helens, UK). However, cross-reaction with other peptides or proteins containing the amino acid sequence recognized by the PACAP antiserum cannot be excluded. It is appropriate, therefore, to refer to the immunoreactive material as PACAP-like.
In-vivo experiment in mice Ovine PACAP-38 (Peninsula Europe, Merseyside, St. Helens, U K ; dissolved in saline + 0.1% gelatine) was injected intravenously into the tail vein of unesthetized mice at dose levels of 1.8, 3.5, 7 and 14 nmol/kg alone or together with carbachol (British Drug House, Poole, U K ; 0.16 gmol/kg) or D-glucose (British Drug House; 2.8 mmol/kg). In the experiments with the adrenoceptor antagonists, phentolamine (CIBA-Geigy, Basle, Switzerland; 35 #mol/kg) and DL-propranolol (Sigma, St. Louis, Mo., USA; 8.5 ~tmol/kg) were given intraperitoneally 20 rain before intravenous (i.v.) injection. Blood was sampled from the retro-orbital venous plexus at 2, 6 and 10 min after the i.v. injection. At 2 min after i.v. injection of glucose and carbachol, the increase in plasma insulin levels is maximal (Ahr6n and Lundquist 1981) and, similarly, plasma glucagon levels are elevated 2 min after the i.v. injection of carbachol (Ahr~n and Lundquist 1986). The volume load was 10 lal/g. Controls were given the saline-gelatine vehicle. Plasma was separated and frozen at - 2 0 ~ C until analysis.
Results
Immunocytochemistry Figs. 1, 2 and 3 demonstrate the immunocytochemical distribution of PACAP-like immunoreactivity in the mouse and rat pancreas. In both species, PACAP-like immunoreactivity was found in nerve fibers within the pancreas. Numerous fibers innervated intrapancreatic ganglia. Fibers were also seen in the parenchyma of the exocrine pancreas, and occasionally around blood vessels and within the islets9Generally, the density of PACAP-immunoreactive nerve fibers was higher in the mouse than in the rat pancreas. No nerve cell bodies containing PACAP-like immunoreactivity were seen.
Basal insulin, glucagon, and glucose Fig. 4 shows the plasma levels of insulin, glucagon and glucose 2 rain after injection of four different dose levels (1.8, 3.5, 7, 14 nmol kg) of PACAP-38 in mice. At dose levels above 1.8 nmol/kg, PACAP-38 increased the plasma glucagon levels ( P < 0.001) without affecting the plasma levels of insulin and glucose. Fig. 5 shows the effect of PACAP-38 (7 nmol/kg) 2, 6, and 10 rain after intravenous administration in mice. At 2 rain after injection of PACAP-38, the plasma glucagon levels were raised, whereas the plasma levels of insulin and glucose were not affected. At 6 rain after injection of the peptide, the plasma levels of insulin were increased. Plasma glucose levels were significantly increased 10 rain after administration of PACAP-38.
120 100 1
Determination of insulin, glucagon and glucose Insulin was determined radioimmunochemically by the use of a guinea-pig anti-porcine insulin antiserum (MILAB, Malm6, Sweden), lzSI-labeled porcine insulin, and porcine insulin as standard (Novo Res, Bagsvaerd, Denmark). Glucagon was determined radioimmunochemically by the use of a rabbit anti-porcine glucagon antiserum specific for pancreatic glucagon, lzSI-labeled porcine glucagon (MILAB, Malm6, Sweden) and porcine glucagon as standard. The separation of bound and free radioactivity was performed by the dextran-coated charcol technique (Herbert et al. 1965). Finally, plasma glucose levels were determined by the glucose oxidase technique (Bruss and Black 1978).
Statistics The results are reported as means + SEM. The data in Fig. 5 are reported as delta values for each group of animals injected with PACAP-38 (7 nmol/kg) at 2, 6 and 10 rain compared with its control group. Student's t-test was used for a simple comparison of significance, and a one way analysis of variance followed by the Newman-Keuls test was used for multiple comparison tests of significance. P < 0.05 was considered significant.
[] ,-T_,
1.8 nmoi/kg 3.5 nmoi/kg
~m_ 60
70 nmot/kg
c
14.0 nmot/kg
40 20
c
0 4000 3000
O3 [3_
c 2000 IOOO cD
0 Glucose (mmoi/t)
93 10.6 +-0.3 tO.2
9.8 "0.2
10.4 10.5 *-0.3 *-0.2
Fig. 4. Plasma levels of insulin (upper panel) and glucagon (lower panel) 2 rain after intravenous injection of the saline-gelatine vehicle (open bars) and 4 different dose levels of PACAP-38 (as indicated). Each group consists of 15-16 animals. Plasma levels of glucose in each group are shown. Means ___SEM are given. Asterisks indicate the probability level of random differences between PACAP-38 injected animals and controls; * P
Cell&Tissue Research 9 Springer-Verlag 1992
Pituitary adenylate cyclase-activating polypeptide (PACAP): occurrence in rodent pancreas and effects on insulin and glucagon secretion in the mouse Tord Fridolf1, Frank Sundler 2, and Bo Ahr~n 3 1 Department of Pharmacology, S61vegatan 10, a Department of Medical Cell Research, Biskopsgatan 5, and 3 Department of Surgery, Lund University, S-223 62 Lund, Sweden Received November 18, 1991 / Accepted March 31, 1992
Summary. Pituitary adenylate cyclase-activating polypeptide (PACAP) is a neuropeptide that occurs in several tissues, e.g., in the gut. We have studied PACAP-like immunoreactivity in the pancreas o f rat and mouse, and the effects of PACAP-38 on basal and stimulated insulin and glucagon secretion in the mouse. Immunofluorescence staining demonstrated the presence of PACAP-like immunoreactivity in nerve fibers in both the rat and mouse pancreas. The nerve fibers were seen in the exocrine pancreas and surrounding the islets. Occasionally, the nerve fibers occurred within the islets. Most PACAPpositive nerve fibers innervated the intrapancreatic ganglia, although no nerve cell bodies contained PACAPlike immunoreactivity. In-vivo experiments in mice revealed that basal plasma glucagon levels were increased by PACAP-38 injected intravenously at dose levels exceeding 1.8 nmol/kg. Furthermore, PACAP-38 (7 nmol/ kg) potentiated the plasma glucagon response to the cholinergic agonist carbachol (0.16 gmol/kg). This potentiation was reduced to simple addition by pretreatment with a combined e- and/%adrenergic blockade by phentolamine (35 gmol/kg) and propranolol (8.5 gmol/ kg). Moreover, PACAP-38 inhibited a carbachol-induced increase in the level of plasma insulin in the absence but not in the presence of adrenergic blockade. PACAP-38 increased basal plasma insulin levels and increased basal plasma glucose levels 6 rain and 10 rain, respectively, after injection of the peptide. We conclude that PACAP-like immunoreactivity exists in nerve fibers innervating the mouse and rat pancreas, particularly the intrapancreatic ganglia, and that PACAP-38 augments both basal and carbachol-stimulated glucagon secretion in the mouse. Key words: Pituitary adenylate cyclase-activating peptide (PACAP) - Immunocytochemistry - Pancreas, endocrine, exocrine - Insulin secretion - Glucagon secretion - Mouse (NMRI) - Rat (Sprague-Dawley) Correspondence to: T. Fridolf
Pituitary adenylate cyclase-activating polypeptide (PACAP)-38 and PACAP-27, which corresponds to the N-terminal 27 amino acid residues of PACAP-38, have recently been isolated from ovine hypothalamic tissue and shown to stimulate adenylate cyclase in rat pituitary cell cultures (Miyata et al. 1989, 1990). The N-terminal 28 amino acids in PACAP-38 bear a structural relationship to the peptides belonging to the glucagon peptide superfamily, i.e., vasoactive intestinal peptide (VIP), peptide histidine isoleucine (PHI), secretin, helospectin I and II, helodermin, and glucagon (Parker et al. 1984; Robberecht et al. 1985; Ahr6n and Lundquist 1988; Ahr6n and Falck 1991). VIP, PHI, secretin and helospectin I have previously been shown to stimulate glucagon secretion markedly in mice, whereas they exert only slight influences on insulin secretion (Ahr6n and Falck 1991; Ahr6n and Lundquist 1986, 1988; Ahr~n 1991). We have now studied whether PACAP-38 also affects insulin and glucagon secretion in vivo, in mice. Since it has been shown that PACAP-like immunoreactivity occurs in peripheral tissues, e.g., in the respiratory tract (Uddman et al. 1991) and the gut (Sundler et al. 1992), we have also studied whether PACAP-like immunoreactivity occurs in the pancreas of the rat and the mouse.
Materials and methods Animals Female mice of the NMRI strain, weighing 25-30 g, and male Sprague-Dawley rats, weighing 200-250 g (both species from ALAB, Stockholm, Sweden), were used. The animals were fed on a standard pellet diet (Astra-Ewos, S6dertfilje, Sweden) and tap water ad libitum before and during the experiments.
Immunocytochemistry Tissue specimens of the duodenal and the splenic portions of the pancreas were taken from decapitated animals and fixed overnight by immersion in Stefanini's solution (2% formaldehyde and 0.2%
276
Fig.
l a-e. Sections from mouse pancreas immunostained for PACAP-27. Immunoreactive nerve fibers appear as delicate fiber elements running in the exocrine pancreas (a), as fibers located in nerve bundles (b), and as fibers penetrating into the islets (c). a x 2 5 0 ; b , c x300 Fig. 2. Section from rat pancreas immunostained for PACAP-27. Note immunoreactive nerve fibers in the exocrine pancreatic parenchyma. • 300 Fig. 3. Section from rat pancreas immunostained for PACAP-27. Immunoreactive nerve fibers innervate a ganglion, x 250
277 picric acid in 0.1 M phosphate-buffered saline, pH 7.2). After rinsing in sucrose-enriched (10%) buffer, the specimens were frozen on dry ice. Sections were cut at a thickness of 15 pm in a cryostat and collected on chromalum-coated slides. The sections were then processed for the immunocytochemical demonstration of PACAP using the indirect immunofluorescence technique (Coons et al. 1955). The PACAP antiserum (code 88121-3; kindly donated by Dr. A. Arimura, Tulane University, Belle Chasse, La., USA) was raised in a rabbit against synthetic ovine PACAP-27 (conjugated with thyroglobulin) and used at a dilution of 1:640. Control sections were incubated with PACAP antiserum preabsorbed with excess antigen (10 pg PACAP-27 per ml diluted antiserum). These sections did not show any immunoreactivity. The antiserum showed no cross-reactivity with glucagon, VIP, PHI, helodermin, helospectin I, growth hormone-releasing factor (GRF) or gastrinreleasing peptide (GRP) (Peninsula Europe, Merseyside, St. Helens, UK). However, cross-reaction with other peptides or proteins containing the amino acid sequence recognized by the PACAP antiserum cannot be excluded. It is appropriate, therefore, to refer to the immunoreactive material as PACAP-like.
In-vivo experiment in mice Ovine PACAP-38 (Peninsula Europe, Merseyside, St. Helens, U K ; dissolved in saline + 0.1% gelatine) was injected intravenously into the tail vein of unesthetized mice at dose levels of 1.8, 3.5, 7 and 14 nmol/kg alone or together with carbachol (British Drug House, Poole, U K ; 0.16 gmol/kg) or D-glucose (British Drug House; 2.8 mmol/kg). In the experiments with the adrenoceptor antagonists, phentolamine (CIBA-Geigy, Basle, Switzerland; 35 #mol/kg) and DL-propranolol (Sigma, St. Louis, Mo., USA; 8.5 ~tmol/kg) were given intraperitoneally 20 rain before intravenous (i.v.) injection. Blood was sampled from the retro-orbital venous plexus at 2, 6 and 10 min after the i.v. injection. At 2 min after i.v. injection of glucose and carbachol, the increase in plasma insulin levels is maximal (Ahr6n and Lundquist 1981) and, similarly, plasma glucagon levels are elevated 2 min after the i.v. injection of carbachol (Ahr~n and Lundquist 1986). The volume load was 10 lal/g. Controls were given the saline-gelatine vehicle. Plasma was separated and frozen at - 2 0 ~ C until analysis.
Results
Immunocytochemistry Figs. 1, 2 and 3 demonstrate the immunocytochemical distribution of PACAP-like immunoreactivity in the mouse and rat pancreas. In both species, PACAP-like immunoreactivity was found in nerve fibers within the pancreas. Numerous fibers innervated intrapancreatic ganglia. Fibers were also seen in the parenchyma of the exocrine pancreas, and occasionally around blood vessels and within the islets9Generally, the density of PACAP-immunoreactive nerve fibers was higher in the mouse than in the rat pancreas. No nerve cell bodies containing PACAP-like immunoreactivity were seen.
Basal insulin, glucagon, and glucose Fig. 4 shows the plasma levels of insulin, glucagon and glucose 2 rain after injection of four different dose levels (1.8, 3.5, 7, 14 nmol kg) of PACAP-38 in mice. At dose levels above 1.8 nmol/kg, PACAP-38 increased the plasma glucagon levels ( P < 0.001) without affecting the plasma levels of insulin and glucose. Fig. 5 shows the effect of PACAP-38 (7 nmol/kg) 2, 6, and 10 rain after intravenous administration in mice. At 2 rain after injection of PACAP-38, the plasma glucagon levels were raised, whereas the plasma levels of insulin and glucose were not affected. At 6 rain after injection of the peptide, the plasma levels of insulin were increased. Plasma glucose levels were significantly increased 10 rain after administration of PACAP-38.
120 100 1
Determination of insulin, glucagon and glucose Insulin was determined radioimmunochemically by the use of a guinea-pig anti-porcine insulin antiserum (MILAB, Malm6, Sweden), lzSI-labeled porcine insulin, and porcine insulin as standard (Novo Res, Bagsvaerd, Denmark). Glucagon was determined radioimmunochemically by the use of a rabbit anti-porcine glucagon antiserum specific for pancreatic glucagon, lzSI-labeled porcine glucagon (MILAB, Malm6, Sweden) and porcine glucagon as standard. The separation of bound and free radioactivity was performed by the dextran-coated charcol technique (Herbert et al. 1965). Finally, plasma glucose levels were determined by the glucose oxidase technique (Bruss and Black 1978).
Statistics The results are reported as means + SEM. The data in Fig. 5 are reported as delta values for each group of animals injected with PACAP-38 (7 nmol/kg) at 2, 6 and 10 rain compared with its control group. Student's t-test was used for a simple comparison of significance, and a one way analysis of variance followed by the Newman-Keuls test was used for multiple comparison tests of significance. P < 0.05 was considered significant.
[] ,-T_,
1.8 nmoi/kg 3.5 nmoi/kg
~m_ 60
70 nmot/kg
c
14.0 nmot/kg
40 20
c
0 4000 3000
O3 [3_
c 2000 IOOO cD
0 Glucose (mmoi/t)
93 10.6 +-0.3 tO.2
9.8 "0.2
10.4 10.5 *-0.3 *-0.2
Fig. 4. Plasma levels of insulin (upper panel) and glucagon (lower panel) 2 rain after intravenous injection of the saline-gelatine vehicle (open bars) and 4 different dose levels of PACAP-38 (as indicated). Each group consists of 15-16 animals. Plasma levels of glucose in each group are shown. Means ___SEM are given. Asterisks indicate the probability level of random differences between PACAP-38 injected animals and controls; * P
