Tachykinins as mediators of slow EPSPs in guinea-pig gall-bladder ganglia: involvement of neurokinin-3 receptors

1. The effects of endogenous tachykinins and related peptides on intact guinea-pig gall-bladder neurones were investigated with single-electrode voltage- and current-clamp recording techniques. 2. Pressure ejection of substance P (100 microM) caused a long lasting membrane depolarization that was associated with a decrease in input resistance. In cells that were voltage-clamped to their resting membrane potential, substance P activated an inward current. 3. The reversal potentials of the substance P-induced depolarization and inward current were congruent to 0 mV. In a low-Na+ solution, the substance P-induced depolarization and inward current were reduced in amplitude. 4. Substance P increased the excitability of neurones, as evidenced by a greater anodal break activity and an increase in the number of action potentials generated during a depolarizing current pulse. 5. Substance P, neurokinin A (NKA) and neurokinin B (NKB) were applied by superfusion to determine the relative potencies of these tachykinins. NKB was the most potent, with an EC50 of 24 nM. The EC50 values for NKA and substance P were 47.8 and 281 nM, respectively. 6. The neurokinin-3 (NK-3) receptor agonist senktide depolarized neurones with an EC50 of 6.3 nM. Neither the NK-1 receptor agonist [Sar9,Met(O2)11]-substance P nor the NK-2 receptor agonist [beta-Ala8]-NKA(4-10) caused a measurable depolarization. 7. The NK-3 antagonist [Trp7,beta-Ala8]-NKA (4-10) inhibited the responsiveness of gall-bladder neurones to substance P with a KB (dissociation constant of receptor antagonist) of 49 nM, and depressed both capsaicin-induced depolarizations and stimulus-evoked slow EPSPs. 8. These data indicate that tachykinins mediate slow EPSPs in guinea-pig gall-bladder ganglia by activating NK-3 receptors on gall-bladder neurones. It is proposed that in response to inflammation or high intraluminal pressure, tachykinins may be released within ganglia by sensory fibres and act directly on intrinsic neurones to facilitate ganglionic transmission.

Immunohistochemical identification of neurons in ganglia of the guinea pig sphincter of Oddi

The sphincter of Oddi is a smooth muscle sphincter that regulates the flow of bile into the duodenum. To identify potential chemical coding in sphincter of Oddi neurons, immunohistochemistry and histochemistry were employed to assay for putative neurotransmitters and related synthetic enzymes in wholemount preparations, with and without colchicine treatment. Immunoreactivities for enkephalin-endorphin (ENK-END), substance P (SP), nitric oxide synthase, vasoactive intestinal peptide (VIP), neuropeptide Y (NPY), and calcitonin gene-related peptide (CGRP) were demonstrated within the ganglionated plexus. Roughly half of the neurons in the sphincter of Oddi expressed immunoreactivity for both SP and ENK-END, but not for nitric oxide synthase. About 25% of the neurons expressed nitric oxide synthase immunoreactivity as well as NADPH-diaphorase activity. This contingent of neurons was made up of two subgroups: one that expressed immunoreactivity for VIP, the other for NPY. Neurons that expressed CGRP immunoreactivity were sparse in sphincter of Oddi ganglia; however, many axons immunoreactive for both CGRP and SP were present in the ganglionated plexus. The CGRP/SP fibers are probably visceral afferents that may influence ganglionic output through axon reflex circuits. These results, along with studies of the actions of these neuroactive compounds on sphincter tone, support the view that ganglia of the sphincter of Oddi are largely comprised of excitatory (SP/ENK-END-immunoreactive) and inhibitory (nitric oxide synthase/VIP- or NPY-immunoreactive) neurons, and that sphincter of Oddi tone is controlled by the regulation of the outputs of these two groups of cells.

Actions of cholecystokinin and norepinephrine on vagal inputs to ganglion cells in guinea pig gallbladder

Previous studies have demonstrated that all guinea pig gallbladder neurons receive nicotinic synaptic input and that cholecystokinin (CCK) and norepinephrine have presynaptic facilitory and inhibitory effects, respectively, on these fast synaptic events. The current study was undertaken to determine the sources of the cholinergic terminals that provide nicotinic input to gallbladder neurons. To stimulate potential extrinsic inputs to gallbladder neurons, a stimulating electrode was placed on the nerve bundles that pass along the cystic duct. Stimulation of these nerves elicited fast excitatory postsynaptic potentials (EPSPs) in gallbladder neurons that were sensitive to hexamethonium, facilitated by CCK, and inhibited by norepinephrine. After vagotomy, most neurons (14 of 18) did not exhibit any nicotinic input. However, some neurons (3 of 18) did exhibit fast EPSPs in response to fiber tract stimulation, but not cystic nerve stimulation, indicating that interganglionic communication does exist amongst gallbladder neurons. These results demonstrate that the vagus nerves provide the major nicotinic input to gallbladder neurons. Furthermore, these data suggest that vagal terminals within gallbladder are a site of neurohormonal modulation of gallbladder ganglionic output by CCK, norepinephrine, and possibly other compounds.

Protein kinase A mediates activation of ATP-sensitive K+ currents by CGRP in gallbladder smooth muscle

The signal transduction mechanisms underlying the activation of ATP-sensitive potassium (KATP) current by calcitonin gene-related peptide (CGRP) in gallbladder smooth muscle were examined with intracellular microelectrode recording and whole cell patch-clamp techniques. In the intact gallbladder preparation, the adenylyl cyclase activator forskolin hyperpolarized the membrane potential and abolished spontaneous action potentials. This response was inhibited by the KATP channel blocker glibenclamide. CGRP (10 nM), forskolin (10 microM), the membrane-permeable adenosine 3',5'-cyclic monophosphate (cAMP) analogue adenosine 3',5'-cyclic monophosphothioate (Sp-cAMP[S]; 500 microM), and the catalytic subunit of protein kinase A (100 U/ml) activated glibenclamide-sensitive currents in enzymatically dissociated gallbladder smooth muscle cells. CGRP activation of potassium currents was prevented by dialysis of the cell cytoplasm with guanosine 5'-O-(2-thiodiphosphate) (5 mM) or a specific peptide inhibitor of protein kinase A (2.3 microM). Okadaic acid (5 microM), a phosphatase inhibitor, slowed the deactivation of the KATP current, following removal of CGRP. The results of this study indicate that CGRP hyperpolarizes gallbladder smooth muscle by elevation of cAMP and subsequent stimulation of protein kinase A.

Activation of ATP-sensitive potassium currents in guinea-pig gall-bladder smooth muscle by the neuropeptide CGRP

1. The actions of a neuroactive peptide, calcitonin gene-related peptide (CGRP), and the ATP-sensitive potassium (K+ATP) channel activator lemakalim on guinea-pig gall-bladder smooth muscle cells were investigated using intracellular recording from intact preparations and whole-cell patch clamp recording from acutely dissociated myocytes. 2. CGRP and lemakalim caused a glibenclamide-sensitive hyperpolarization of the plasma membrane of intact cells with an associated suppression of spontaneous action potentials. 3. In isolated smooth muscle cells, CGRP (10 nM) and lemakalim (10 microM) activated currents that were glibenclamide sensitive, voltage independent and potassium selective. 4. External TEA+ at 1.0 and 10.0 mM inhibited glibenclamide-sensitive, CGRP-activated currents by 3.8 and 66.5%, respectively. 5. Increases in the intracellular ATP concentration from 0.1 to 5.0 mM reduced the glibenclamide-sensitive potassium current in the presence of CGRP (10 nM) or lemakalim (10 microM) by > 60%. The increase in the intracellular ATP also reduced the steady-state glibenclamide-sensitive current by > 80%. 6. These findings indicate that CGRP activates K+ATP channels to hyperpolarize the membrane of gall-bladder smooth muscle cells. This hyperpolarization may be an important mechanism underlying the relaxant effects of CGRP on the gall-bladder.

Sympathetic input to ganglia of the guinea pig sphincter of Oddi

Intracellular recording and immunohistochemical staining techniques were used to establish whether sphincter of Oddi (SO) ganglia are a target of sympathetic input to this region. Norepinephrine (0.01-10.0 microM) decreased the amplitude of the nicotinic fast excitatory postsynaptic potential (EPSP) evoked by stimulation of interganglionic fiber tracts, with a half-maximal inhibitory concentration (EC50) of 300 nM. Norepinephrine did not alter the responsiveness of the neurons to acetylcholine. The alpha 2-adrenoreceptor agonist UK-14304 mimicked the norepinephrine-induced effect with a EC50 of 2.5 nM, whereas alpha 1- and beta-adrenoreceptor agonists had no effect on the EPSP. The alpha 2-adrenoreceptor antagonist idazoxan (1.0 microM) inhibited the UK-14304 response, with a dissociation constant of 1.0 nM. Release of endogenous catecholamines, by the addition of tyramine (100 microM) to the bath, caused an idazoxan-sensitive decrease in the amplitude of the fast EPSP. In the minority of SO neurons that exhibited inhibitory postsynaptic potentials (IPSPs), norepinephrine caused a hyperpolarization of the membrane potential. The IPSP and the norepinephrine-induced hyperpolarization were inhibited by alpha 2-adrenoreceptor antagonists. Desipramine (1.0 microM), an uptake inhibitor, reversibly increased the amplitude of the IPSP. Immunoreactivities for tyrosine hydroxylase and dopamine beta-hydroxylase were coexistent in nerve fibers and nonexistent in cell bodies in the ganglionated plexus of the SO. The results of this study indicate that norepinephrine acts pre- and postsynaptically as an inhibitory neurotransmitter in SO ganglia.

Ionic basis of the action potential of guinea pig gallbladder smooth muscle cells

Smooth muscle cells in the intact guinea pig gallbladder had a resting membrane potential of about -45 mV and had spontaneous action potentials that consisted of a rapid depolarization, a transient repolarization, a plateau phase, and a complete repolarization. These action potentials lasted approximately 570 ms and occurred at a frequency of approximately 0.4 Hz. Action potentials were abolished by the dihydropyridine (DHP)-sensitive Ca2+ channel blocker nifedipine (1.0 microM) and were enhanced by the DHP-sensitive Ca2+ channel agonist BAY K 8644 (0.5 microM). The K+ channel blockers tetraethylammonium chloride (5.0 mM) and 4-aminopyridine (4-AP; 2.0 mM) prolonged the action potential, whereas charybdotoxin (100 nM), a blocker of calcium-activated potassium channels, had no effect. Whole cell currents were characterized in enzymatically isolated smooth muscle cells from the same preparation. 4-AP, a blocker of voltage-dependent K+ channels, suppressed 70% of the outward current at 0 mV. Charybdotoxin (100 nM) reduced an additional 15% of the current at 0 mV. Single calcium-activated potassium channels were identified. The potential for half-activation of these channels, at a cytosolic Ca2+ concentration of 100 nM, was 66.8 mV. A fivefold increase in cytosolic Ca2+ resulted in a shift of the activation curve by -53 mV. External tetraethylammonium chloride (200 microM) reduced the mean single channel current by 48% at 0 mV. The whole cell outward current was abolished by replacement of intracellular K+ for Cs+. Ca2+ currents were inhibited by nifedipine and were increased by BAY K 8644. We conclude that DHP-sensitive voltage-dependent Ca2+ channels are responsible for the depolarization of the action potentials and that the repolarization is due to primarily 4-AP-sensitive K+ current.

Physiological and morphological properties of neurons in sphincter of Oddi region of the guinea pig

Intracellular recordings and dye injections were used to investigate neurons located in ganglia of the sphincter of Oddi (SO) region in guinea pigs. Four types of neurons were encountered based on physiological properties. The two most abundant types, tonic and phasic, had similar membrane characteristics and morphologies but yet could be differentiated by their spiking characteristics. Tonic cells spiked throughout a depolarizing current pulse and were sometimes spontaneously active. Phasic cells fired only a single action potential at the onset of a current pulse regardless of stimulus amplitude or duration. Both tonic and phasic cells had Dogiel type I morphologies. They typically had a single long process and several very short processes emanating from the soma. NADPH diaphorase activity was demonstrated in cells with similar morphologies, indicating that nitric oxide may be an intrinsic transmitter in some of these cells. Cells with a prolonged afterhyperpolarization (AH cells), similar to the type 1/AH cells of the gut, were rarely encountered. This finding was consistent with the observation that very few calbindin D-immunoreactive neurons exist in this region. Action potentials could not be generated in the fourth type of neuron, called nonspiking neurons, even though they did receive synaptic input. Most tonic and phasic cells received at least one nicotinic fast excitatory postsynaptic potential (EPSP). In addition, both slow EPSPs and inhibitory postsynaptic potentials were observed. Most AH cells received only slow excitatory synaptic input.

NADPH-diaphorase and VIP are co-localized in neurons of gallbladder ganglia

Histochemical staining was used to demonstrate that intramural neurons of the gallbladder contain NADPH-diaphorase, and therefore are likely to produce nitric oxide. A subset of the neurons in the gallbladders of the guinea pig, gerbil, opossum, dog, and human stained positively for the enzyme. In the guinea pig, all neurons that were immunoreactive for vasoactive intestinal peptide (VIP), also contained NADPH-diaphorase. Furthermore, neurons that were immunoreactive for neuropeptide Y, which have been shown to be immunoreactive for substance P and somatostatin as well, rarely contained NADPH-diaphorase. It is suggested that the VIP/NADPH-diaphorase-containing neurons represent intrinsic inhibitory motor neurons of the gallbladder, and that these neurons may have a role in the relaxation of the muscularis during gallbladder filling.

Noradrenaline as a presynaptic inhibitory neurotransmitter in ganglia of the guinea-pig gall-bladder

1. The effects of noradrenaline on guinea-pig gall-bladder ganglia were investigated with intracellular recording techniques. 2. Noradrenaline (0.01-100 microM) decreased the amplitude of the fast excitatory postsynaptic potential (EPSP) that was evoked by stimulation of interganglionic fibre tracts. High concentrations of noradrenaline (10-100 microM) caused an inhibition ranging from 93-100%. The noradrenaline concentration that resulted in half-maximal inhibition (EC50) of the EPSP was 280 nM. 3. Experiments with selective agonists and antagonists indicated that the alpha 2-adrenoreceptor was involved in the inhibition of the EPSP. Clonidine (0.001-100 microM) reduced the EPSP in a concentration-dependent manner with an EC50 of 30 nM. Yohimbine (100-300 nM) caused a rightward shift of the noradrenaline concentration-effect relationship, with a dissociation equilibrium constant of 1.4 nM. 4. Release of endogenous catecholamines by tyramine (100 microM) in the presence of desipramine (1.0 microM), caused a yohimbine-sensitive decrease in the amplitude of the EPSP. Treatment with tyramine did not affect the amplitude of the EPSP in tissue that had undergone prior chemical sympathectomy with 6-hydroxydopamine. 5. Electrical stimulation of the vascular plexus (1-3 s; 10-20 Hz; 10 mA) decreased the amplitude of the EPSP. In some cases suprathreshold responses were reduced to subthreshold EPSPs following stimulation of the vascular plexus. Yohimbine (300 nM) reversibly inhibited the effects of vascular plexus stimulation. 6. Noradrenaline did not modify the responses of gall-bladder neurones to exogenously applied acetylcholine. Also, application of noradrenaline, by superfusion (0.001-100 microM) or by pressure microejection (1.0 mM), had no effect on the resting membrane potential, membrane conductance, or action potential characteristics of gall-bladder neurones. 7. Immunoreactivity for type A monoamine oxidase (MAO-A) was found in the vascular plexus and the ganglionated plexus of the gall-bladder. 8. These results show that noradrenaline has an alpha 2-adrenoreceptor-mediated presynaptic inhibitory effect on fast synaptic transmission in the ganglia of the guinea-pig gall-bladder. It is proposed that vagal terminals may be an important target of this adrenergic inhibitory input to the gall-bladder.

Transmitter diversity in ganglion cells of the guinea pig gallbladder: an immunohistochemical study

Several neurotransmitters have been reported to exist in the ganglionated plexus of the guinea pig gallbladder. These include substance P, neuropeptide Y (NPY), calcitonin gene-related peptide, vasoactive intestinal peptide (VIP), acetylcholine, norepinephrine, serotonin, and dopamine. To determine which neuropeptides are intrinsic to gallbladder ganglia, we performed immunohistochemistry on colchicine-treated preparations. In separate, single-labeled preparations, a majority of neurons contained substance P-, NPY-, or somatostatin-like immunoreactivity. In double-labeled preparations, a large majority of the neurons that contained substance P-like immunoreactivity also contained NPY-like immunoreactivity and somatostatin-like immunoreactivity. Immunoreactivity for VIP was present in a small percentage of the gallbladder neurons which did not contain substance P-like immunoreactivity. Additional experiments were done to test for the presence of other compounds, known to exist in the neurons of the gut. Although immunoreactivity was found in control preparations of small intestine, the ganglionated plexus of the gallbladder lacked immunoreactivity for galanin, dynorphin, enkephalin, gastrin-releasing peptide, or gamma-aminobutyric acid. We conclude that ganglia of the guinea pig gallbladder contain at least two populations of neurons, based on transmitter phenotype. One of these populations appears to contain substance P, NPY, and somatostatin. Another population, which represents a small contingent of the total population of neurons, contains VIP.

Structure of neurons and ganglia of the guinea pig gallbladder: light and electron microscopic studies

This study was undertaken to examine the morphological features of cells within ganglia of the guinea pig gallbladder, and to examine the ultrastructure of the ganglionated plexus. Gallbladder neurons are large, with a relatively simple form, having only one or two major processes. Neurobiotin often filled axons to their varicose arbors on smooth muscle in close proximity to the interganglionic connectives. With the exception of connective tissue clefts that sometimes penetrated into them, ganglia were devoid of intercellular spaces, capillaries, or connective tissue elements such as collagen and basal laminae. However, ganglia were surrounded by a single, continuous basal lamina that was enclosed within a fibroblast and collagen capsule. Within ganglia, neurons were insulated by the processes of cells that resembled the astrocyte-like glia of enteric ganglia. Although few classical synapses were observed, numerous sites of direct apposition were identified between vesicle-rich profiles and processes of gallbladder neurons. Direct appositions between vesicle-rich profiles and the ganglion-limiting basal laminae were also observed. Vesiculated profiles contained small clear vesicles and large dense-core vesicles. Within interganglionic connectives, axons were unmyelinated and were isolated from one another by processes of glia that resembled Schwann cells. As was seen in the ganglia, direct appositions between vesicle-rich profiles and the connective-limiting basal laminae were observed. The results of this study demonstrate that gallbladder ganglia are similar, ultrastructurally, to enteric ganglia in the CNS-like composition of the neuropil. However, the greater degree of glial investment, lesser degree of innervation, and simpler neurons indicated differences from the enteric nervous system that may be functionally significant.

The role of cholecystokinin in ganglionic transmission in the guinea-pig gall-bladder

1. The effects of cholecystokinin (CCK) on intact guinea-pig gall-bladder ganglia were investigated with intracellular, single-electrode current- and voltage-clamp recording techniques. 2. Cholecystokinin octapeptide (CCK-8; 0.01-100 nM) increased the amplitude of the fast excitatory postsynaptic potential (EPSP) that was evoked by stimulation of interganglionic fibre tracts. In most cases, neurones that exhibited subthreshold EPSPs in normal Krebs solution fired action potentials in the presence of CCK-8. In a low Ca2+/high Mg2+ solution, CCK-8 caused a 3-fold increase in the amplitude of fast EPSPs. 3. The amplitude of the evoked excitatory postsynaptic current (EPSC) was increased by CCK-8 (0.01-100 nM) in a concentration-dependent manner. The effect was maximal at 1.0 nM. 4. Cholecystokinin octapeptide caused a 3-fold increase in the quantal content of the EPSP in a low Ca2+/high Mg2+ solution, but had no effect on the quantal size. 5. The specific CCK-A receptor antagonist, MK-329 (formerly L-364,718; 1.0 nM), reversibly blocked the facilitatory effect of CCK-8 on ganglionic transmission. However, the specific CCK-B receptor antagonist, L-365,260 (10 nM), did not alter the presynaptic facilitatory effect of CCK-8. 6. The response of gall-bladder neurones to exogenously applied ACh was not modified by CCK-8. 7. Application of CCK-8, by superfusion (0.001-100 nM) or by pressure microejection (100 microM), had no effect on the membrane potential, membrane conductance, action potential, or threshold of gall-bladder neurones. 8. Immunohistochemistry was employed to determine whether the actions of CCK could be elicited by release of the peptide from nerve terminals within the ganglionated plexus of the gall-bladder. Immunoreactivity for CCK was not detected in the ganglionated plexus of the gall-bladder, but CCK immunoreactivity was plentiful in control preparations of intestinal myenteric and submucosal plexuses. 9. These results show that CCK has a presynaptic facilitatory effect on fast synaptic transmission in guinea-pig gall-bladder ganglia, and that this effect is mediated by presynaptic CCK-A receptors. Furthermore, it appears that such an effect would normally occur in response to hormonal CCK, rather than CCK that is released from nerve terminals.

Use of stereoisomers of zacopride to analyze actions of 5-hydroxytryptamine on enteric neurons

Two subtypes of excitatory 5-hydroxytryptamine (5-HT) receptor, 5-HT1P and 5-HT3, are found on type 2-AH neurons of the guinea pig myenteric plexus. The 5-HT1P receptor mediates a slow and the 5-HT3 receptor a fast depolarization of these cells, however, the role of these receptors in the physiology of the gut is unknown. Renzapride (BRL 24924), a substituted benzamide, has previously been found to antagonize responses of myenteric neurons mediated by both 5-HT1P and 5-HT3 receptors. The effects on myenteric type 2-AH neurons of a structurally similar benzamide, zacopride, which unlike renzapride has S and R stereoisomers, were investigated to gain further insight into 5-HT receptor function. In contrast to renzapride, S-, but not R-zacopride, was found to mimic the 5-HT1P receptor-mediated slow response to 5-HT. Desensitization of 5-HT1P receptors with 5-HT inhibited slow depolarizing responses to S-zacopride, and desensitization with S-zacopride antagonized slow responses to 5-HT. Responses to S-zacopride were also inhibited by renzapride and the 5-HT1P receptor antagonist N-acetyl-5-hydroxytryptophyl-5-hydroxytryptophan amide (5-HTP-DP). S-zacopride, like renzapride and 5-HT, presynaptically inhibited nicotinic fast excitatory postsynaptic potentials, an effect that can be mediated by 5-HT1P or 5-HT1A receptors. Both S and R stereoisomers of zacopride antagonized 5-HT3 receptor-mediated fast responses to 5-HT. Unlike 5-HTP-DP, neither zacopride or its stereoisomers nor renzapride inhibited the binding of 5-[3H]HT to 5-HT1P receptors. [3H]zacopride (5-10 nM) was found to bind to a site in the gut from which it could be displaced by a 1,000-fold excess of renzapride and S-zacopride (but not R-zacopride) greater than 5-HTP-DP much greater than the 5-HT3 receptor antagonist ICS 205-930. These observations suggest that, in addition to 5-HT3 receptors, there is a benzamide binding site on myenteric neurons that interacts with, but is distinct from, the 5-HT recognition site of 5-HT1P receptors. Benzamides may affect coupling of the 5-HT1P receptor to its effector.

Evaluation of the activity of chemically identified enteric neurons through the histochemical demonstration of cytochrome oxidase

The measurement of the density of the reaction product produced by the histochemical demonstration of cytochrome oxidase activity provides a method for the visual identification of physiologically active enteric neurons. The current study utilized the cytochrome oxidase technique in order to evaluate the metabolic history of neurons in different regions of the bowel and in chemically identified types of neuron. In addition, the effect of drugs or neurotoxins commonly used in the immunocytochemical identification of enteric neuronal phenotypes was also analyzed. Cytochrome oxidase activity was visualized with a blue-black reaction product resulting from the cobalt-intensified oxidation of 3,3'-diaminobenzidine. Peptides or 5-hydroxytryptamine (5-HT) were localized with biotinylated secondary antibodies and alkaline phosphatase-labeled avidin. Bound avidin or endogenous alkaline phosphatase was visualized with a red reaction product in the presence or absence, respectively, of levamisole. Use of measured without interference from a simultaneously demonstrated histo- or immunochemical marker. A multi-peptidergic class of cholinergic submucosal secretomotor neuron containing neuropeptide Y (NPY) and calcitonin gene related peptide (CGRP) immunoreactivities was found to be less metabolically active than the average of all submucosal neurons. In contrast, a non-cholinergic submucosal secretomotor neuron containing dynorphin (which is also known to contain vasoactive intestinal peptide) immunoreactivity was more metabolically active than submucosal neurons that do not contain this peptide. On average, submucosal neurons were more metabolically active than those of the myenteric plexus, and levels of metabolic activity in the myenteric plexus were found to be higher in the duodenum and the cecum than in the jejunum-ileum or colon. Myenteric neurons characterized by CGRP or NPY immunoreactivities or by endogenous alkaline phosphatase activity, were all less metabolically active than the average of all neurons in myenteric ganglia. Colchicine, which stimulates intestinal motility, was observed to increase cytochrome oxidase activity in enteric neurons, suggesting that an effect on the enteric nervous system contributes to its action on the bowel. The neurotoxins, 6-hydroxydopamine and 5,7-dihydroxytryptamine (5,7-DHT) were each found to stimulate neuronal metabolic activity. 5,7-DHT appeared to activate excitatory subtypes of 5-HT receptor since its effects were blocked or mimicked by compounds that act as antagonists or agonists, respectively, at these receptors.(ABSTRACT TRUNCATED AT 400 WORDS)

Intracellular recording from neurones of the guinea-pig gall-bladder

1. Intracellular recordings were made from neurones of the guinea-pig gall-bladder in vitro. Intracellular injection of horseradish peroxidase revealed a simple structure, consisting of a soma and a single process, but no discernible dendritic arborization. 2. The resting membrane potential was -50.5 +/- 0.4 mV and the input resistance was 80 M omega. 3. Gall-bladder neurones spiked only once at the onset of depolarizing current pulses. Action potentials were blocked by tetrodotoxin, but a Ca2(+)-dependent spike could be elicited in the presence of tetrodotoxin and tetraethylammonium. 4. Action potential after-hyperpolarizations had a duration of 172 +/- 3.7 ms and reversed at a membrane potential of -93 mV; this reversal potential was linearly related to the logarithm of the external potassium concentration. The initial phase of the after-hyperpolarization was inhibited by tetraethylammonium (1-10 mM) and was not affected by 3,4-diaminopyridine. The late phase of the after-hyperpolarization was blocked by apamin (10 nM) or curare (500 microM). Both the early and late phases of the after-hyperpolarization were inhibited when the preparation was perfused with a calcium-free, high-magnesium solution. The calcium-free, high-magnesium solution had no effect on the membrane potential or input resistance of these cells. 5. Fast excitatory synaptic responses and antidromic responses were elicited in gall-bladder neurones by focal stimulation of fibre tracts. High-frequency fibre tract stimulation often resulted in prolonged, calcium-dependent, depolarizations that were associated with a decrease in input resistance. 6. 5-Hydroxytryptamine and substance P caused depolarizations that were associated with a decrease in input resistance. Bethanechol caused hyperpolarizations that were associated with a decrease in input resistance and which were blocked by atropine.

Development of synaptic transmission at autonomic synapses in vitro revealed by cytochrome oxidase histochemistry

We have studied the development of synaptic transmission by innervating sympathetic neurons in vitro and monitoring synaptic activity with both physiological recording and cytochrome oxidase histochemistry. The onset of synaptic transmission was reflected in increased cytochrome oxidase reaction product within individual neurons. Within 24 hours of co-culture, relatively low frequency suprathreshold potentials were recorded in approximately 20% of the innervated neurons. At this stage the cytochrome oxidase activity of innervated neurons, as assessed by optical density of the histochemical reaction product, was increased twofold compared with uninnervated neurons. Over the next 2-4 days of innervation, changes in the pattern and extent of synaptic activity and superthreshold events were accompanied by a net fourfold increase in cytochrome oxidase activity levels compared with noninnervated neurons. The increase in density of cytochrome oxidase reaction product observed after innervation was reversed completely by blockade of synaptic transmission. Differences in the efficacy of synaptic input provided to the sympathetic neurons by appropriate versus inappropriate presynaptic sources was determined by co-culturing sympathetic neurons with explants that contained either preganglionic neurons or somatic motor neurons. Although sympathetic neurons innervated by motor neuron explants had increased levels of cytochrome oxidase activity compared with noninnervated controls, the density of cytochrome oxidase reaction product was even greater in sympathetic neurons innervated by preganglionic explants. We conclude that both the onset of innervation of sympathetic neurons as well as the subsequent maturation of synaptic function is directly reflected in graded increases in cytochrome oxidase reaction product.

Blockade of 5-HT-mediated enteric slow EPSPs by BRL 24924: gastrokinetic effects

Two types of 5-hydroxytryptamine (5-HT) receptor, 5-HT1P and 5-HT3, have been identified physiologically on enteric neurons impaled by intracellular microelectrodes. Activation of 5-HT1P receptors evokes a long-lasting membrane depolarization associated with an increased input resistance, whereas stimulation of 5-HT3 receptors results in a brief depolarization during which the input resistance falls. Slow excitatory postsynaptic potentials (EPSPs) in myenteric type II-hyperpolarizing afterpotential (AH) neurons have been demonstrated to be mediated by 5-HT1P receptors. The current experiments were done to determine whether the substituted benzamide, BRL 24924, is a specific antagonist at 5-HT1P receptors and can be used as a probe to investigate the role played by serotoninergic neurons in the control of gastrointestinal motility. Intracellular microelectrodes were used to analyze the effects of BRL 24924 on guinea pig myenteric neurons. Microejection of BRL 24924 mimicked neither the long-lasting nor the brief response to 5-HT; however, BRL 24924 (0.5-1.0 microM) reversibly antagonized both the long-lasting 5-HT1P receptor-mediated responses of myenteric neurons to 5-HT and 5-HT-mediated slow EPSPs. A greater than 10-fold higher concentration of BRL 24924 was required to reduce the short-lived responses mediated by 5-HT3 receptors. BRL 24924 did not affect the response of myenteric neurons to substance P. These results indicate that BRL 24924 is primarily a 5-HT1P antagonist. Unlike other 5-HT1P agonists or antagonists, BRL 24924 did not block the binding of 5-[3H]HT to 5-HT1P receptors. This observation suggests that specific antagonism of physiological responses to 5-HT by BRL 24924 may be the result of an action on the coupling of the 5-HT1P receptor to its effector mechanism. BRL 24924 (0.5-1 mg/kg) and another 5-HT1P antagonist, N-acetyl-5-hydroxytryptophyl-5-hydroxytryptophan amide (5 mg/kg), significantly increased the rate of emptying of a 51Cr-labeled liquid meal from the murine stomach. In contrast, the 5-HT3 antagonist, ICS 205-930 (0.1-0.5 mg/kg), did not affect the rate of gastric emptying. These observations are consistent with the hypothesis that intrinsic inhibitory neurons of the murine stomach are activated by serotoninergic axons acting through 5-HT1P receptors. Antagonism of an excitatory drive to neurons in a relaxant pathway may thus explain the gastrokinetic effects of BRL 24924.

Immunocytochemical analysis of potential neurotransmitters present in the myenteric plexus and muscular layers of the corpus of the guinea pig stomach

Recent electrophysiological studies of neurons of the myenteric plexus of the corpus of the guinea pig stomach have revealed that slow synaptic events are extremely rare. In contrast, they are commonly encountered in similar investigations of myenteric ganglia of the guinea pig small intestine. The current immunocytochemical analysis of the myenteric plexus and innervation of the muscularis externa of the corpus of the guinea pig stomach was undertaken in order to determine whether putative neurotransmitters capable of mediating slow synaptic events are present in gastric ganglia. A major difference between the small intestine and the stomach was found in the innervation of the musculature. Whereas the longitudinal muscle layer of the small intestine contains very few nerve fibers and is innervated mainly at its interface with the myenteric plexus, the longitudinal muscle of the corpus of the stomach contained as many varicose substance P (SP)-, vasocative intestinal polypeptide (VIP)-, and neuropeptide Y (NPY)-immunoreactive axons as the circular muscle layer. These putative neurotransmitters were also present in the ganglia of the myenteric plexus, where varicose SP-, VIP-, and NPY-immunoreactive fibers encircled nonimmunoreactive neurons. Varicose 5-hydroxytryptamine (5-HT)-immunoreactive terminal axons were essentially limited to the myenteric plexus and were found both in ganglia and in interganglionic connectives, where they were particularly numerous; 5-HT-immunoreactive neurons appeared to be more abundant in the stomach than in the small intestine. Tyrosine hydroxylase (TH)- and calcitonin-gene-related-peptide (CGRP)-immunoreactive axons were also more common in the myenteric plexus than in the musculature, but of these, only the TH-immunoreactive neurites tended, like those of the other putative transmitters, to encircle neurons in myenteric ganglia. Evidence was obtained that, as in the small intestine, at least some of the SP-, VIP-, NPY-, and 5-HT-immunoreactive fibers in the stomach are derived from intrinsic gastric myenteric neurons. In contrast, unlike the small intestine, gastric myenteric ganglia appeared to lack intrinsic CGRP-immunoreactive neurons; therefore, the CGRP-immunoreactive gastric axons are probably of extrinsic origin.(ABSTRACT TRUNCATED AT 400 WORDS)

Structure, afferent innervation, and transmitter content of ganglia of the guinea pig gallbladder: relationship to the enteric nervous system

Although a well-developed plexus of nerves and ganglia is known to be present in the wall of the gallbladder, little has previously been learned about the function or organization of this innervation. The current study was undertaken in order to evaluate the hypothesis that the ganglionated plexus of the gallbladder is analogous to elements of the enteric nervous system (ENS). The ganglionated plexus of the gallbladder was found to resemble closely the submucosal plexus of the small intestine in its organization into two irregular anastomosing and interwoven networks of ganglia, in the numbers of neurons per ganglion, and in the manifestation of histochemically demonstrable acetylcholinesterase activity in virtually all ganglion cells. In common with enteric ganglia, laminin immunoreactivity was observed to be excluded from the interiors of gallbladder ganglia, which were surrounded by a periganglionic laminin-immunoreactive sheath. As in the submucosal plexus, intrinsic substance P-, vasoactive intestinal polypeptide (VIP)-, and neuropeptide Y (NPY)-immunoreactive neurons were seen in the ganglionated plexus of the gallbladder. Extrinsic nerves in the gallbladder that degenerated following chemical sympathectomy with 6-hydroxydopamine (6-OHDA), and which contained NPY, tyrosine hydroxylase (TH), and dopamine-beta-hydroxylase (DBH) immunoreactivities, formed a perivascular plexus closely associated with blood vessels. Endogenous catecholamines could also be demonstrated in these perivascular nerves by aldehyde-induced histofluorescence. In addition to perivascular nerves, paravascular nerve bundles were observed that were loosely associated with vessels, did not degenerate following administration of 6-OHDA, and contained NPY immunoreactivity. Other paravascular nerves, probably visceral sensory axons, coexpressed substance P and calcitonin-gene-related peptide (CGRP) immunoreactivities. The ganglionated plexus of the gallbladder resembled enteric ganglia in having intrinsic 5-hydroxytryptamine (5-HT)-immunoreactive cells and highly varicose nerve fibers. The 5-HT-immunoreactive gallbladder axons were, like those of the gut, resistant to 6-OHDA, and separate from fibers that expressed TH immunoreactivity. Differences between the ganglionated plexus of the gallbladder and enteric ganglia of the small intestine included in the gallbladder are 1) the presence of TH-immunoreactive cells that contain an endogenous catecholamine, but not DBH; 2) DBH-immunoreactive neurons, some of which coexpress substance P immunoreactivity, but which contain neither a catecholamine nor TH immunoreactivity; 3) an apparent absence of CGRP-immunoreactive cell bodies.(ABSTRACT TRUNCATED AT 400 WORDS)

Characterization and localization of a peripheral neural 5-hydroxytryptamine receptor subtype (5-HT1P) with a selective agonist, 3H-5-hydroxyindalpine

Peripheral neural 5-hydroxytryptamine (5-HT) receptors are different from both classes 5-HT1 and 5-HT2, which have been described from studies of 5-HT receptors in the brain. Recently, it has been shown that, as in the CNS, there is more than a single type of neural receptor for 5-HT in the enteric nervous system. One of these, called 5-HT1P, has a high affinity for 3H-5-HT, initiates a long-lasting depolarization of enteric neurons associated with an increase in membrane resistance, and is the physiological receptor through which enteric serotoninergic neurons mediate slow EPSPs. The other receptor, called 5-HT3 (5-HT2P), does not bind 3H-5-HT with high affinity, and initiates a brief depolarization of enteric neurons with decreased input resistance, but a physiological action of 5-HT mediated by these receptors has not yet been identified. Hydroxylated indalpines have been found to be agonists at 5-HT1P receptors. We have now examined 5-HT1P receptors using 5-hydroxyindalpine (5-OHIP) as a probe. The action of 5-OHIP on enteric neurons was determined electrophysiologically and compared with that of 5-HT; the binding of 3H-5-OHIP to isolated enteric membranes was studied by rapid filtration, and to frozen sections of tissue by radioautography. 3H-5-OHIP binding was compared with that of 3H-5-HT. 5-OHIP, like 5-HT, induced a triphasic response in most enteric neurons: an initial short-lived depolarization, during which input resistance fell, followed by recovery, and then a long-lasting depolarization, during which the input resistance increased. 5-OHIP bound saturably, reversibly, and with high affinity to enteric membranes (Kd = 7.6 +/- 0.7 nM; Bmax = 76 +/- 14 fmol/mg protein). Binding of 3H-5-OHIP was not inhibited by agents that bind to alpha- or beta-adrenoceptors, nicotinic or muscarinic receptors, histamine H1 or H2 receptors, or 5-HT1(A,B,C, or D), 5-HT2, or 5-HT3 receptors, but was displaced by substances, such as hydroxylated indoles and a dipeptide of 5-hydroxytryptophan (5-HTP-DP), that antagonize the binding of 3H-5-HT to enteric membranes or tissue sections. It is concluded that 5-OHIP is an agonist at peripheral neural 5-HT1P receptors and can be used to label these receptors selectively outside the brain. Radioautographs demonstrated enteric 5-HT1P receptors in the lamina propria of the intestinal mucosa and in the submucosal and myenteric plexuses. Extraenteric 5-HT1P receptors were also found in the skin and heart. It is suggested that 5-HT1P receptors may be found on subtypes of primary afferent nerve fibers.

Differences in synaptic inputs to preganglionic neurons in the dorsal and lateral band subdivisions of the cat sacral parasympathetic nucleus

In the cat, preganglionic neurons (PGNs) found in the dorsal portion of the sacral parasympathetic nucleus (dorsal band or DB cells) participate in bowel control, while those found along the lateral edge (lateral band or LB cells) influence bladder function. In order to determine whether differences in the synaptic inputs exist between these two populations, HRP was applied to the sacral ventral rootlets of cats, and the S2 cord segment was prepared for sequential light and electron microscopy. When measured with light microscope, the LB somata had greater cross-sectional areas than did the DB cells. Ultrastructurally, the LB cells had a significantly greater percentage of their membrane apposed by synaptic active sites than did the DB cells. Also, the proximal dendrities of the labelled neurons received greater synaptic input than did the somata. No difference was found in the proportion of terminals containing dense cored vesicles (DCVs) when comparing LB and DB somata; however, the LB proximal dendrites had a higher proportion of their surface contacted by DCV-containing terminals than did the DB dendrites. These ultrastructural results offer evidence that these two populations of preganglionic neurons differ with respect to their synaptic input as well as their peripheral targets.

Distribution and ultrastructure of ventral root afferents to lamina I of the cat sacral spinal cord

Following ventral root injury-filling with horseradish peroxidase at sacral spinal levels S1, S2, and S3, small fascicles of axons can be observed to enter the ventral portion of lamina I and arborize primarily in the dorsolateral region just medial to the tract of Lissauer (TL). Some axons enter the TL and turn in a rostrocaudal direction. Labeled axons studied by electron microscopy are both myelinated and unmyelinated, and terminals in lamina I contain round clear, and dense cored vesicles and contact smaller diameter dendrites and vesicle-containing elements. The concentration of these putative ventral root afferents appears to be largest in the S2 segment, with smaller numbers found within S1 and S3.

Origin and morphology of nerve fibers in the aganglionic colon of the lethal spotted (ls/ls) mutant mouse

The lethal spotted mutant mouse (ls/ls) develops congenital megacolon because of the absence of ganglia in the terminal colon. This aganglionosis results from a failure of neural crest cells to colonize this area during fetal life. We have postulated that the microenvironment of the aganglionic segment of bowel is abnormal. Our hypothesis suggests that this abnormal enteric microenvironment fosters the sprouting of neuritic processes. We further propose that neural and glial precursors cease to migrate once they have extended their definitive processes. As a result, the area distal to the site where neurite extension is favored does not become colonized by neural or glial precursors. A prediction of this hypothesis is that the aganglionic tissue should be innervated by axons from neurons located both in the more proximal ganglionated bowel and in ganglia located outside the gut. Neurons and their processes in control and ls/ls terminal gut were located by the histochemical demonstration of acetylcholinesterase (AChE) activity and their structure was classified as intrinsic (enteric) or extrinsic in type by electron microscopy. In ls/ls mice the submucosal plexus was much more severely affected than the myenteric plexus. No submucosal ganglia were found within 30 mm of the anus. In contrast, myenteric ganglia extended to within 4 mm of the anus on the mesenteric side of the gut and to within 15 mm on the antimesenteric side. Rostral to the areas that were absolutely aganglionic, both plexuses were hypoganglionic, especially the submucosal plexus, which was hypoganglionic throughout the entire colon. Both the aganglionic and caudal hypoganglionic zones of the ls/ls bowel were penetrated by large nerve trunks that had the ultrastructural characteristics of extra-enteric peripheral nerve. Unusual ganglia, outside the enteric musculature in the adventitia of the colon, were connected to these trunks. The location of the cell bodies of origin of the nerve fibers in the terminal colon of control mice and in the aganglionic segment of the bowel in ls/ls mice was determined by following the retrograde transport of tracers injected as close as possible to the anus. An extrinsic innervation originating from the inferior mesenteric ganglion and dorsal root ganglia (L6-S1) was found in both types of animal. In control but not ls/ls mice retrograde labeling was also observed in the sacral parasympathetic nucleus of the spinal cord. In addition, neuritic processes were traced to neurons in myenteric ganglia. In control mice, these labeled neurons were present in ganglia within the injection site as well as in bowel rostral and caudal to it.(ABSTRACT TRUNCATED AT 400 WORDS)

Peripheral neural serotonin receptors: identification and characterization with specific antagonists and agonists

Serotonin (5-hydroxytryptamine, 5-HT) has been shown to be a neurotransmitter in the enteric nervous system (ENS). Although 5-HT is a mediator of slow excitatory postsynaptic potentials evoked by stimulation of interganglionic connectives, the precise role it plays in the physiology of the gut is unclear. Research has been hampered by an inadequate knowledge of the types of 5-HT receptor in the ENS and thus the lack of well-characterized antagonists. We now report the identification of two classes of enteric neural 5-HT receptor, the effects of activating these receptors on myenteric type II/AH neurons, and their characterization with specific agonists and antagonists. One class, which we propose to call 5-HT1P, is characterized by a high affinity for [3H]5-HT in radioligand binding assays. This class of receptor mediates a slow depolarization of myenteric type II/AH neurons associated with an increase in input resistance. Agonists at this receptor include, in addition to 5-HT (in order of potency), 5- and 6-hydroxyindalpine and 2-methyl-5-HT. 5-HT1P-mediated responses are specifically antagonized by 5-hydroxytryptophyl-5-hydroxytryptophan amide. The other class of 5-HT receptor, which we propose to call 5-HT2P, appears not to have a high affinity for [3H]5-HT. This receptor mediates a brief depolarization of myenteric II/AH neurons associated with a fall in input resistance. 2-Methyl-5-HT, at low concentrations, is a specific agonist at this receptor and ICS 205-930 is a specific antagonist. Binding of [3H]5-HT to enteric membranes is inhibited by 5-HT1P receptor agonists and antagonists but not by the 5-HT2P receptor antagonist ICS 205-930 or by MDL 72222, another compound reported to be an antagonist of 5-HT at peripheral receptors.