Structural requirements for galanin interaction with receptors from pancreatic beta cells and from brain tissue of the rat

The binding activity of several galanin fragments and analogs was measured on specific receptors present in rat brain and the rat pancreatic beta cell line Rin m 5F. In both tissues it was observed that: 1) galanin(3-29), galanin(10-29) and [Ile2]-galanin were ineffective for inhibiting [125I] galanin binding and 2) active peptides had the following rank order of potency: galanin(1-29) greater than [Ac-Trp2]-galanin(2-29) greater than galanin(2-29) greater than galanin(1-15) greater than [Phe2]-galanin greater than [Tyr2]-galanin. It was concluded that the N-terminal portion of galanin is very important for interaction with central or peripheral receptors. The aromatic amino acid in position 2 (Trp in native galanin) plays a crucial role.     

Galanin binding sites in rat gastric and jejunal smooth muscle membrane preparations.

Receptors for galanin in membranes from the rat gastric and jejunal smooth muscle were studied using [125I] radioiodinated synthetic porcine galanin. Specific binding was time and temperature dependent. At 32 degrees C radioligand was degraded in the presence of smooth muscle membranes in a time-dependent manner. At optimal experimental conditions, the equilibrium binding analyses showed the presence of a single population of high affinity binding sites in both the rat stomach and jejunum (Kd value of 2.77 +/- 0.78 nM and 4.93 +/- 1.74 nM for stomach and jejunal smooth muscle membranes, respectively). The concentration of the high affinity binding sites was 58.19 +/- 11.04 and 32.36 +/- 5.68 fmol/mg protein, for gastric and jejunal preparations, respectively. Specific binding was completely inhibited by 10(-6) M of nonradioactive galanin; was 75% blocked by 1 microM of galanin(9-29); it was 10% blocked by 1 microM of galanin(15-29). Galanin(1-15) at a concentration of 1 microM was ineffective for inhibiting [125I]galanin binding. Deletion of four C-terminal amino acid residues from galanin(9-29) to give galanin(9-25) also resulted in almost complete loss of affinity. Radioiodinated galanin and N-terminally deleted fragments had receptor binding potency in the following order: galanin(1-29) greater than galanin(9-29) greater than galanin(15-29). We conclude that the C-terminal part of the galanin chain is important for the rat gastric and jejunal smooth muscle membrane receptor recognition and binding and that N-terminal amino acid sequences are probably not so important, since galanin(1-15) was not active but galanin(9-29) retained most of the receptor binding activity.     

Subunit composition of G(o) proteins functionally coupling galanin receptors to voltage-gated calcium channels.

The neuropeptide galanin is widely expressed in the central nervous system and other tissues and induces different cellular reactions, e.g. hormone release from pituitary and inhibition of insulin release from pancreatic B cells. By microinjection of antisense oligonucleotides we studied the question as to which G proteins mediate the galanin-induced inhibition of voltage-gated Ca2+ channels in the rat pancreatic B-cell line RINm5F and in the rat pituitary cell line GH3. Injection of antisense oligonucleotides directed against alpha 01, beta 2, beta 3, gamma 2 and gamma 4 G protein subunits reduced the inhibition of Ca2+ channel current which was induced by galanin, whereas no change was seen after injection of cells with antisense oligonucleotides directed against alpha i, alpha q, alpha 11, alpha 14, alpha 15, beta 1, beta 4, gamma 1, gamma 3, gamma 5, or gamma 7 G protein subunits or with sense control oligonucleotides. In view of these data and of previous results, we conclude that the galanin receptors in GH3 and in RINm5F cells couple mainly to the G(0) protein consisting of alpha 01 beta 2 gamma 2 to inhibit Ca2+ channels and use alpha 01beta 3 gamma 4 less efficiently. The latter G protein composition was previously shown to be used by muscarinic M4 receptors to inhibit Ca2+ channels.     

Solubilization and molecular characterization of active galanin receptors from rat brain.

Galanin receptors were solubilized from rat brain using the zwitterionic detergent 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonic acid (CHAPS). Binding of 125I-galanin to the soluble fraction was time- and temperature-dependent, saturable, and reversible. Scatchard analysis of binding data indicated that the soluble extract contained a single class of galanin binding sites with a Kd of 0.8 nM and a Bmax of 26 fmol/mg of protein. Unlabeled galanin and its fragments galanin(2-29) and galanin(1-15) antagonized the binding of 125I-galanin to CHAPS-solubilized extracts with relative potencies similar to those observed with membrane receptors. Galanin(3-29) was found inactive. Binding of 125I-galanin to CHAPS extracts was inhibited by guanine nucleotides with the following rank order of potency: GMP-P-(NH)P greater than GTP greater than GDP. Molecular analysis of the soluble galanin receptor by covalent cross-linking of 125I-galanin to CHAPS extracts using disuccinimidyl tartrate and further identification on SDS-PAGE indicated that the soluble galanin binding site behaves as a protein of Mr 54,000. After incubation of CHAPS extracts with 125I-galanin, gel filtration on Sephacryl S-300 followed by ultracentrifugation on sucrose density gradient revealed a binding component with the following hydrodynamic parameters: Stokes radius, 5 nm; s20,w, 4.5 S; Mr, 98,000; frictional ratio, 1.6. GMP-P(NH)P treatment of CHAPS extracts gave rise to a molecular form with the following characteristics: Stokes radius, 4 nm; s20,w, 3.3 S; Mr, 57,000; frictional ratio, 1.4.(ABSTRACT TRUNCATED AT 250 WORDS)     

Design of chimeric peptide ligands to galanin receptors and substance P receptors.

Several chimeric peptides were synthesized and found to be high-affinity ligands for both galanin and substance P receptors in membranes from the rat hypothalamus. The peptide galantide, composed of the N-terminal part of galanin and C-terminal part of substance P (SP), galanin-(1-12)-Pro-SP-(5-11) amide, which is the first galanin antagonist to be reported, recognizes two classes of galanin binding sites (KD(1) less than 0.1 nM and KD(2) approximately 6 nM) in the rat hypothalamus, while it appears to bind to a single population of SP receptors (KD approximately 40 nM). The chimeric peptide has higher affinity towards galanin receptors than the endogenous peptide galanin-(1-29) (KD approximately 1 nM) or its N-terminal fragment galanin-(1-13) (KD approximately 1 microM), which constitutes the N-terminus of the chimeric peptide. Galantide has also higher affinity for the SP receptors than the C-terminal SP fragment-(4-11) amide (KD = 0.4 microM), which constitutes its C-terminal portion. Substitution of amino acid residues, which is of importance for recognition of galanin by galanin receptors, such as [Trp2], in the galanin portion of the chimeric peptide or substitution of ([Phe7] or [Met11]-amide) in the SP portion of chimeric peptide both cause significant loss in affinity of the analogs of galantide for both the galanin- and the SP-receptors. These results suggest that the high affinity of the chimeric peptide, galantide, may in part be accounted for by simultaneous recognition/binding to both receptors.(ABSTRACT TRUNCATED AT 250 WORDS)     

Structural requirements for galanin action in the guinea-pig ileum.

Galanin fragments and galanin analogues were tested on neurally evoked muscle contractions in guinea-pig ileum in vitro. Galanin fragments inhibited the neurally evoked circular muscle contractions with the following order of potency: Galanin(1-29), galanin(2-29), galanin(1-15). In contrast, galanin(3-29), galanin(10-29), galanin(21-29), [D-Trp2]galanin, [Phe2]galanin and [Tyr2]galanin were ineffective. Galanin(1-29), galanin(2-29) and galanin(1-15) did not affect the neurally evoked longitudinal muscle contractions. These results indicate that (1) the two N-terminal amino acid residues of the galanin molecule are essential for the inhibitory action of galanin on neurally-evoked circular muscle contraction and (2) for the full potency also the C-terminal end is required.     

Glucagon-like peptide-1 improves insulin and proinsulin binding on RINm5F cells and human monocytes

Glucagon-like peptide-1-(7---36) amide (GLP-1) is a potent incretin hormone secreted from distal gut. It stimulates basal and glucose-induced insulin secretion and proinsulin gene expression. The present study tested the hypothesis that GLP-1 may modulate insulin receptor binding. RINm5F rat insulinoma cells were incubated with GLP-1 (0.01-100 nM) for different periods (1 min-24 h). Insulin receptor binding was assessed by competitive ligand binding studies. In addition, we investigated the effect of GLP-1 on insulin receptor binding on monocytes isolated from type 1 and type 2 diabetes patients and healthy volunteers. In RINm5F cells, GLP-1 increased the capacity and affinity of insulin binding in a time- and concentration-dependent manner. The GLP-1 receptor agonist exendin-4 showed similar effects, whereas the receptor antagonist exendin-(9---39) amide inhibited the GLP-1-induced increase in insulin receptor binding. The GLP-1 effect was potentiated by the adenylyl cyclase activator forskolin and the stable cAMP analog Sp-5, 6-dichloro-1-beta-D-ribofuranosyl-benzimidazole-3', 5'-monophosphorothioate but was antagonized by the intracellular Ca(2+) chelator 1,2-bis(0-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid-AM. Glucagon, gastric inhibitory peptide (GIP), and GIP-(1---30) did not affect insulin binding. In isolated monocytes, 24 h incubation with 100 nM GLP-1 significantly (P<0.05) increased the diminished number of high-capacity/low-affinity insulin binding sites per cell in type 1 diabetics (9,000+/-3,200 vs. 18,500+/-3,600) and in type 2 diabetics (15,700+/-2,100 vs. 28,900+/-1,800) compared with nondiabetic control subjects (25,100+/-2,700 vs. 26,200+/-4,200). Based on our previous experiments in IEC-6 cells and IM-9 lymphoblasts indicating that the low-affinity/high-capacity insulin binding sites may be more specific for proinsulin (Jehle, PM, Fussgaenger RD, Angelus NK, Jungwirth RJ, Saile B, and Lutz MP. Am J Physiol Endocrinol Metab 276: E262-E268, 1999 and Jehle, PM, Lutz MP, and Fussgaenger RD. Diabetologia 39: 421-432, 1996), we further investigated the effect of GLP-1 on proinsulin binding in RINm5F cells and monocytes. In both cell types, GLP-1 induced a significant increase in proinsulin binding. We conclude that, in RINm5F cells and in isolated human monocytes, GLP-1 specifically increases the number of high-capacity insulin binding sites that may be functional proinsulin receptors.     

Discrimination of galanin receptor subtypes in RINm5F cells by structurally different galanin radioligands

Galanin (GAL) is a biologically active peptide that is involved in a variety of physiological functions. The purpose of this study was to evaluate whether porcine and rat galanin radioligands could be used as probes to discriminate GAL receptors (GALR) subtypes using a cell line, RINm5F, that express multiple GALR subtypes. Data from parallel equilibrium binding experiments using the same RINm5F membrane homogenates reveal that [125I]pGAL labels 20% more GALRs with a 2-fold lower affinity than those values identified when using [125I]rGAL. Competition studies using various GAL peptides showed different rank order of potencies depending on the radioligand used. Preincubation of RINm5F membranes with GppNHp, a non-hydrolizable GTP analog, prior to radioligand labeling suggests that a portion of GALRs is precoupled to G proteins. In addition, receptors labeled by [125I]rGAL appear more sensitive to GppNHp-induced uncoupling of G proteins than those labeled by [125I]pGAL. In conclusion, our data suggest that pGAL and rGAL radioligands define different pharmacological profiles of GALRs, and hence, these ligands can be used as pharmacological tools to discriminate GALR subtypes. Additionally, our data suggests that GALRs exist in a precoupled state with their respective G-proteins prior to interaction with the agonist.     

Galanin activates nucleotide-dependent K+ channels in insulin-secreting cells via a pertussis toxin-sensitive G-protein.

The effects of galanin (7-70 nM) on ATP-sensitive K+ channels (KATP channels), membrane potential and the release of insulin have been studied in the insulinoma cell line, RINm5F. Single-channel currents have been recorded from excised outside-out membrane patches as well as intact insulin-secreting cells and it is shown that galanin, added to the outside of the membrane, specifically activates KATP channels. Studies carried out using the fluorescent probe bisoxonol demonstrate that galanin hyperpolarizes RINm5F cells. Galanin was also found to abolish glyceraldehyde-stimulated immunoreactive insulin release from the insulinoma cells. Both the galanin-evoked hyperpolarization and inhibition of insulin release were abolished in cells pre-exposed to pertussis toxin. The possibility that the gating of KATP channels could be mediated by a G-protein was studied in patch-clamp experiments by adding F- to the solution bathing the inside of the cell membranes (open-cell), in order to generate the alumino-fluoride complex AlF4-. F- (1-10 mM) evoked dose-dependent activation of KATP channels and this effect was fully reversible. F- was also able to activate K+ channels inhibited by ATP. That the fluoride activation of KATP channels is mediated by the complex AlF4- was indicated by experiments in which AlCl3 (10 microM) was found to enhance further the activation of K+ channels evoked by 1 mM F- and by results showing that F(-)-stimulation of KATP channels was (i) abolished in the continued presence of F- by the Al3+ chelator deferoxamine (0.5 mM) and (ii) could be mimicked by VO4(3-) which has a structure similar to that of the AlF4- complex.     

Linear and cyclic N-terminal galanin fragments and analogs as ligands at the hypothalamic galanin receptor.

The neuropeptide galanin (1-29) binds with high affinity to hypothalamic receptors (KD approximately 0.9 nM) and regulates feeding behavior. The N-terminal fragments (1-16), (1-16)NH2 are high affinity (KD approximately 6 nM) full agonists in vivo and in vitro. L-Ala substitutions show that amino acid residues Gly1, Trp2, Asn5, Tyr9, and Gly12 are important for the high affinity binding of galanin (1-16). Shortening the fragment (1-16) to galanin (1-7) causes a gradual drop of affinity: galanin (1-15), (1-14), and (1-13) have submicromolar KD values and galanin (1-12) has KD approximately 3 microM. Cyclic analogs of galanin (1-12) of different ring size were synthesized by condensing Gly1 and Gly12 without or with spacer groups. These analogs, independent of ring size, had a lower affinity than the linear galanin (1-12). Derivatization of the N-terminus of galanin (1-29), (1-16), and (1-12) all resulted in a large drop of affinity for the receptors, suggesting again the importance of the free N-terminal Gly.     

Prostacyclin release induced by neurokinins in cultured human endothelial cells

The effects of neurokinins (NK) and related peptides on the secretion of 6-keto-prostaglandin F1 alpha, a stable metabolite of prostacyclin, were measured. These peptides enhanced three- to five-fold the basal secretion rate with the following rank order of potency (based on threshold concentrations for a significant output): substance P (SP) greater than or equal to NKA greater than SP 4-11 greater than or equal to [pGlu6]SP 6-11 = SP 7-11.NKB and SP 1-9 were inactive. Ac[Arg6, Sar9, Met(O2)11]SP, a NK1 receptor selective agonist, was more potent than other selective agonists for the NK2 and NK3 receptor subtypes. These results suggest that the NK receptors, which mediate the release of prostacyclin from human endothelial cells, belong to the NK1 subtype.     

Involvement of pertussis toxin-sensitive G-proteins in the hormonal inhibition of dihydropyridine-sensitive Ca2+ currents in an insulin-secreting cell line (RINm5F).

Adrenaline inhibits insulin secretion via pertussis toxin-sensitive mechanisms. Since voltage-dependent Ca2+ currents play a key role in insulin secretion, we examined whether adrenaline modulates voltage-dependent Ca2+ currents of the rat insulinoma cell line, RINm5F. In the whole-cell configuration of the patch-clamp technique, dihydropyridine- but not omega-conotoxin-sensitive Ca2+ currents were identified. Adrenaline via alpha 2-adrenoceptors inhibited the Ca2+ currents by about 50%. Somatostatin which also inhibits insulin secretion was less efficient (inhibition by 20%). The hormonal inhibition of Ca2+ currents was not affected by intracellularly applied cAMP but blocked by the intracellularly applied GDP analog guanosine 5'-O-(2-thiodiphosphate) and by pretreatment of cells with pertussis toxin. In contrast to adrenaline and somatostatin, galanin, another inhibitor of insulin secretion, reduced Ca2+ currents by about 40% in a pertussis toxin-insensitive manner. Immunoblot experiments performed with antibodies generated against synthetic peptides revealed that membranes of RINm5F cells possess four pertussis toxin-sensitive G-proteins including Gi1, Gi2, Go2, and another Go subtype, most likely representing Go1. In membranes of control but not of pertussis toxin-treated cells, adrenaline via alpha 2-adrenoceptors stimulated incorporation of the photo-reactive GTP analog [alpha-32P]GTP azidoanilide into pertussis toxin substrates comigrating with the alpha-subunits of Gi2, Go2, and the not further identified Go subtype. The present findings indicate that activated alpha 2-adrenoceptors of RINm5F cells interact with multiple G-proteins, i.e. two forms of Go and with Gi2. These G-proteins are likely to be involved in the adrenaline-induced inhibition of dihydropyridine-sensitive Ca2+ currents and in other signal transduction pathways contributing to the adrenaline-induced inhibition of insulin secretion.     

Islet cell surface antibodies in genetically obese hyperglycaemic (ob/ob) mice

A quantitative method for circulating islet cell surface antibodies (ICSA), based on the binding of¹²⁵I-protein A to insulin-producing RINm5F cells, was used to evaluate ICSA in plasma of 4- to 40-week-old Aston obese hyperglycaemic (ob/ob) mice and normal control (+/+) mice. RINm5F cells bound 2502±l196 c.p.m.¹²⁵I-protein A per l0⁵ cells (mean±S.D.,n=54) after incubation with +/+ plasma. ICSA positive plasma (defined as¹²⁵I-protein A binding, mean±2 S.D. of +/+ plasma) was detected in 3 out of 54+/+ mice and 3 out of 54 ob/ob mice. ICSA were not observed in ob/ob mice before the onset of diabetes (7 weeks of age), but were detected at 9, 20 and 40 weeks. At 20 weeks¹²⁵I-protein A binding produced by ob/ob plasma was 35% greater than +/+ plasma (P<0.05). The low occurrence of ICSA in ob/ob mice (6%) suggests that factors other than ICSA are responsible for B-cell dysfunction and eventual islet degeneration observed in Aston ob/ob mice.     

Inhibitory action of galanin on gastric acid secretion in pentobarbital-anesthetized rats

The effects of galanin and two galanin fragments, GAL(9-29) and GAL(15-29), were studied for potential effects on pentagastrin- and bethanechol-stimulated gastric acid secretion in a pentobarbital-anesthetized rat experimental model. At a dose of 10 micrograms/kg/h, galanin potently inhibited pentagastrin-stimulated gastric acid secretion whereas inhibition of bethanechol-stimulated gastric acid secretion was not statistically significant. Simultaneous iv infusion of galanin and atropine did not affect the inhibitory action of the former. In similar experiments, a GAL(15-29) fragment was completely inactive whilst GAL(9-29) retained only about 5% potency. These results indicate that galanin probably induces its inhibitory effects by acting directly on the parietal cells rather than through a cholinergic pathway. They also demonstrate that the rat gastric acid inhibitory activity of galanin depends critically on the integrity of the first fourteen N-terminal amino acids.     

The processing of receptor-bound [125I-Tyr11]somatostatin by RINm5F insulinoma cells

The peptide somatostatin (SRIF) is secreted by delta cells of the endocrine pancreas and inhibits the secretion of insulin from pancreatic beta cells. We have previously shown that [125I-Tyr11]SRIF binds to specific, high affinity receptors on RINm5F insulinoma cells and that these receptors mediate the action of SRIF to inhibit insulin release. In the present study we investigated the processing of receptor-bound [125I-Tyr11]SRIF in this clonal cell line. Surface-bound and internalized peptides were distinguished by the ability of an acid/salt solution (0.2 M acetic acid, 0.5 M NaCl, pH 2.5) to dissociate only exposed ligand-receptor complexes. Surprisingly, greater than 80% of saturably bound [125I-Tyr11]SRIF was removed by this acid wash independent of the time or temperature of the binding incubation. In contrast, the processing of receptor-bound [125I]EGF (epidermal growth factor) in RINm5F cells was markedly temperature-dependent. Although over 90% of saturably bound [125I]EGF was dissociated by acid after a 4 degrees C binding incubation, less than 10% was removed by acid treatment after 37 degrees C binding. The radioactivity released upon dissociation of receptor-bound [125I-Tyr11]SRIF was analyzed by high performance liquid chromatography and shown to consist of a mixture of intact peptide (40%) and [125I]tyrosine (60%). However, neither the rate of [125I-Tyr11]SRIF dissociation nor its degradation were affected by NH4Cl, methylamine, or leupeptin at concentrations which inhibited the lysosomal degradation of [125I] EGF. Of 11 other protease inhibitors tested, only the metalloendoprotease inhibitor, phosphoramidon, substantially reduced the degradation of receptor-bound [125I-Tyr11]SRIF. These data indicate that, unlike [125I] EGF, receptor-bound [125I-Tyr11]SRIF is not rapidly internalized by RINm5F cells and is degraded by a nonlysosomal process which may involve a metalloendoprotease.     

Preferential activation by galanin 1–15 fragment of the GalR1 protomer of a GalR1–GalR2 heteroreceptor complex

The three cloned galanin receptors show a higher affinity for galanin than for galanin N-terminal fragments. Galanin fragment (1–15) binding sites were discovered in the rat Central Nervous System, especially in dorsal hippocampus, indicating a relevant role of galanin fragments in central galanin communication. The hypothesis was introduced that these N-terminal galanin fragment preferring sites are formed through the formation of GalR1–GalR2 heteromers which may play a significant role in mediating galanin fragment (1–15) signaling. In HEK293T cells evidence for the existence of GalR1–GalR2 heteroreceptor complexes were obtained with proximity ligation and BRET² assays. PLA positive blobs representing GalR1–GalR2 heteroreceptor complexes were also observed in the raphe-hippocampal system. In CRE luciferase reporter gene assays, galanin (1–15) was more potent than galanin (1–29) in inhibiting the forskolin-induced increase of luciferase activity in GalR1–GalR2 transfected cells. The inhibition of CREB by 50 nM of galanin (1–15) and of galanin (1–29) was fully counteracted by the non-selective galanin antagonist M35 and the selective GalR2 antagonist M871. These results suggested that the orthosteric agonist binding site of GalR1 protomer may have an increased affinity for the galanin (1–15) vs galanin (1–29) which can lead to its demonstrated increase in potency to inhibit CREB vs galanin (1–29). In contrast, in NFAT reporter gene assays galanin (1–29) shows a higher efficacy than galanin (1–15) in increasing Gq/11 mediated signaling over the GalR2 of these heteroreceptor complexes. This disbalance in the signaling of the GalR1–GalR2 heteroreceptor complexes induced by galanin (1–15) may contribute to depression-like actions since GalR1 agonists produce such effects.     

Resistance to alloxan of tumoral insulin-producing cells

Rat pancreatic islets and insulin-producing cells of the RINm5F line were incubated for 5 min at 7 or 23 degrees C in media containing 3H2O and either L-[1-14C]glucose or [2-14C]alloxan. In the islets the intracellular distribution space of [2-14C]alloxan represented, at 7 and 23 degrees C respectively, 11.4 +/- 1.0 and 25.5 +/- 2.3% of the intracellular 3H2O space. In the RINm5F cells, the distribution space of [2-14C]alloxan failed to be affected by the ambient temperature and represented, after correction for extracellular contamination, no more than 5.2 +/- 0.5% of the intracellular 3H2O space. Preincubation for 30 min at 7 degrees C in the presence of alloxan (10 mM) failed to affect subsequent D-[U-14C]glucose oxidation in the tumoral cells, whilst causing a 70% inhibition of glucose oxidation in the islets. It is proposed that RINm5F cells are resistant to the cytotoxic action of alloxan, this being attributable, in part at least, to poor uptake of the diabetogenic agent.     

Galanin can inhibit insulin release by a mechanism other than membrane hyperpolarization or inhibition of adenylate cyclase

Studies on the mode of action of galanin to inhibit insulin release in RINm5F cells have shown that basal and glyceraldehyde-stimulated release were both inhibited. Galanin was inhibitory at concentrations in the low nanomolar range. Binding studies with 125I-labeled galanin indicated that the RINm5F cells exhibit a single set of sites estimated to be of the order of 30,000 sites/cell. Displacement of 125I-galanin by galanin from the receptor sites occurred over a similar concentration range to that which inhibited insulin release. Half-displacement was achieved with 2 nM galanin. Measurements of bis-(1,3-diethylthiobarbiturate) trimethineoxonol (bis-oxonol) fluorescence showed that galanin hyperpolarized the RINm5F cell plasma membrane. Measurements of intracellular free calcium, [Ca2+]i by means of the fluorescent indicator fura-2 showed that galanin decreased [Ca2+]i. As galanin did not inhibit either basal or glyceraldehyde-stimulated insulin release in the presence of the Ca2+ channel blocker nitrendipine, the hyperpolarization and reduction of Ca2+ entry appear to be a possible explanation for the galanin effects. However, quantitatively, the effects on membrane potential and [Ca2+]i appear to be insufficient to account for the potent inhibition of insulin release. Furthermore, evidence for an additional mechanism of action was obtained from experiments with 12-O-tetradecanoylphorbol-13-acetate (TPA), a phorbol ester which stimulates insulin secretion by at least two mechanisms, one Ca2+ dependent and one Ca2+ independent. TPA-stimulated insulin release was inhibited by galanin over the same concentration range as for the inhibition of glyceraldehyde-stimulated release. Galanin inhibited TPA-stimulated release in the presence of maximally effective concentrations of nitrendipine and in the absence of extracellular Ca2+. These effects cannot be explained by hyperpolarization of the plasma membrane and consequent reduction of Ca2+ entry via the voltage-dependent Ca2+ channels. One suggested mechanism for the action of galanin is inhibition of adenylate cyclase. However, it was found that galanin inhibits insulin release even in the presence of 8-Br-cAMP, an agent which effectively bypasses adenylate cyclase. Therefore, an additional mechanism for the inhibitory effect of galanin must be present. All of the effects of galanin were sensitive to pertussis toxin. These data suggest two G-protein-dependent actions of galanin, one to hyperpolarize the plasma membrane and one at a distal point in stimulus-secretion coupling, close to the exocytotic event.