Isolated segments of rat and guinea pig midcolon were used to examine the neurotransmitters responsible for ascending contraction and descending relaxation components of the peristaltic reflex. Graded radial stretch of the extreme orad end caused only descending relaxation accompanied by significant release of vasoactive intestinal peptide (VIP) in rat (82%, P less than 0.005) and guinea pig (47%, P less than 0.05). Radial stretch of the caudad end caused only ascending contraction without VIP release. VIP antiserum (1:480 to 1:60) inhibited descending relaxation in a concentration-dependent manner at all grades of stretch (40 +/- 12% to 74 +/- 15%) but augmented ascending contraction (25 +/- 7% to 108 +/- 21%). Axonal blockade with tetrodotoxin and ganglionic blockade with hexamethonium abolished both components, indicating the participation of cholinergic neurons. Atropine and the tachykinin antagonist [D-Pro2,D-Trp7,9]substance P inhibited ascending contraction but not descending relaxation; their combination abolished ascending contraction at all grades of stretch. We conclude that cholinergic neurons coupled to VIP motor neurons regulate descending relaxation and that cholinergic neurons coupled to tachykinin and cholinergic motor neurons regulate ascending contraction.

The mechanism of action of vasoactive intestinal peptide (VIP) was examined in isolated gastric and taenia coli muscle cells and compared with that of nitric oxide (NO), sodium nitroprusside (SNP), and isoproterenol. In gastric muscle cells, VIP stimulated NO production, increased adenosine 3',5'-cyclic monophosphate (cAMP) and guanosine 3',5'-cyclic monophosphate (cGMP) levels, and induced relaxation in a concentration-dependent fashion. The NO synthase inhibitor NG-nitro-L-arginine abolished NO and cGMP production and partly inhibited relaxation. The soluble guanylate cyclase inhibitor LY 83583 abolished cGMP production and partly inhibited relaxation. (R)-p-adenosine 3',5'-cyclic phosphorothioate [(R)-p-cAMPS], a preferential inhibitor of cAMP-dependent protein kinase (cAK), and KT5823, a preferential inhibitor of cGMP-dependent protein kinase (cGK), partly inhibited relaxation separately and abolished relaxation in combination. The pattern implied that VIP induced relaxation by activation of cAK and by NO-mediated stimulation of cGMP and activation of cGK. In taenia coli muscle cells, VIP did not increase NO production or cGMP levels: relaxation was accompanied by an increase in cAMP and was partly inhibited by (R)-p-cAMPS and KT5823 and abolished by a combination of both inhibitors. Isoproterenol increased only cAMP levels in both cell types, which induced relaxation by activating cAK at low concentrations of agonist and both cAK and cGK at high concentrations in a pattern identical to that observed with VIP in taenia coli muscle cells. SNP and NO increased only cGMP levels in both cell types, which induced relaxation by activating cGK only. We conclude that cAK and cGK can be activated separately and mediate relaxation independently.(ABSTRACT TRUNCATED AT 250 WORDS)

Pituitary adenylate cyclase-activating peptide (PACAP) receptors and their signaling pathways were characterized in dispersed rabbit gastric muscle cells. 125I-PACAP-27 and 125I-vasoactive intestinal peptide (VIP) binding to muscle cells were inhibited equally by PACAP and VIP (mean inhibitory concentration 0.8 to 1.3 nM) and desensitized to the same extent (70-80%) by exposure to either peptide. PACAP, like VIP, increased cytosolic free Ca2+ and the formation of L-[3H]citrulline, NO-3/NO-2, guanosine 3',5'-cyclic monophosphate (cGMP), and adenosine 3'5'-cyclic monophosphate (cAMP) and induced relaxation (mean effective concentration 1.8 +/- 0.1 nM) that was partly inhibited by NG-nitro-L-arginine (L-NNA), VIP-(10-28), and PACAP 6-38. L-[3H]citrulline and cGMP formation were blocked by nifedipine, L-NNA, and pertussis toxin (PTx), implying activation of a G protein-coupled, Ca(2+)-calmodulin-dependent nitric oxide (NO) synthase. PACAP-induced relaxation was inhibited to the same extent (46-49%) by nifedipine, L-NNA, PTx, and the protein kinase G inhibitor KT-5823; the inhibition reflected the component of relaxation mediated by the NO-cGMP pathway. The residual relaxation was abolished by the protein kinase A inhibitor H-89. The pattern of inhibition of all responses was identical to that observed with VIP. Desensitization with VIP or PACAP abolished cAMP formation but had no effect on L-[3H]citrulline and cGMP formation induced by either peptide. Receptor protection with VIP or PACAP preserved fully all responses (L-[3H]citrulline, cGMP, and cAMP formation and relaxation) to either peptide. The complete cross-competition, cross-desensitization, cross-antagonism, and cross-protection of receptors by either VIP or PACAP are consistent with interaction of both peptides with the same receptors; the receptors consist of two classes, each coupled to a distinct signaling pathway.

We have previously shown that Ca2+ mobilization in longitudinal muscle is not mediated by inositol 1,4,5-trisphosphate (IP3) and depends on an obligatory influx of Ca2+. The present study examined whether Ca2+ influx activates ryanodine-sensitive Ca2+ channels to cause Ca(2+)-induced Ca2+ release. Ryanodine bound with high affinity to longitudinal muscle cells [dissociation constant (Kd) 7.3 +/- 0.3 nM] and microsomes (Kd 7.5 +/- 0.4 nM) and induced concentration-dependent 45Ca2+ efflux [50% effective concentration (EC50) 1.3 +/- 0.5 nM], increase in cytosolic free Ca2+ (EC50 2.0 +/- 0.7 nM), and contraction (EC50 0.9 +/- 0.2 nM) but had no effect in circular muscle cells. Ryanodine binding and ryanodine-induced Ca2+ release were enhanced by caffeine and inhibited by dantrolene and ruthenium red but were not affected by IP3 or heparin. Changes in Ca2+ concentration (50-500 nM) caused Ca2+ release from permeabilized longitudinal but not circular muscle cells loaded with 45Ca2+. The contractile agonist cholecystokinin-8 elicited 45Ca2+ efflux in both circular and longitudinal muscle cells; efflux in longitudinal muscle cells was abolished by Ca2+ channel blockers and by pretreatment of the cells with ryanodine. Pretreatment with thapsigargin abolished agonist-induced 45Ca2+ efflux in both cell types. We conclude that ryanodine-sensitive IP3-insensitive Ca2+ release channels with properties similar to those in cardiac muscle are present in longitudinal but not circular muscle cells of intestine and that agonist-mediated Ca2+ influx activates these channels, leading to Ca(2+)-induced Ca2+ release.

Vasoactive intestinal peptide caused a prompt, dose-dependent relaxation of isolated gastric smooth muscle cells of the guinea pig and a significant increase in intracellular adenosine 3',5'-monophosphate coincidentally with optimum relaxation. Relaxation was augmented by a threshold concentration of isobutyl methylxanthine. The direct relaxant effect of vasoactive intestinal peptide and the distribution of nerves containing this peptide to circular smooth muscle support the view that vasoactive intestinal peptide is the neuromuscular transmitter of enteric inhibitory nerves.