Jeffrey A. Love

Autonomic pathways regulating pancreatic exocrine secretion

The parasympathetic (PNS) and sympathetic (SNS) and nervous systems densely innervate the exocrine pancreas. Efferent PNS pathways, consisting of central dorsal motor nucleus of the vagus (DMV) and peripheral pancreatic neurons, stimulate exocrine secretion. The DMV integrates cortical (olfactory, gustatory) and gastric, and intestinal vagal afferent input to determine central PNS outflow during cephalic, gastric and intestinal phases of exocrine secretion. Pancreatic neurons integrate DMV input with peripheral enteric, sympathetic, and, possibly, afferent axon reflexes to determine final PNS input to all exocrine effectors. Gut and islet hormones appear to modulate both central and peripheral PNS pathways. Preganglionic sympathetic neurons in the intermediolateral (IML) column of the spinal cord receive inputs from brain centers, some shared with the PNS, and innervate postganglionic neurons, mainly in prevertebral ganglia. Sympathetic innervation of the exocrine pancreas is primarily indirect, and inhibits secretion by decreasing blood flow and inhibiting transmission in pancreatic ganglia. Interactions between SNS and PNS pathways appear to occur in brain, spinal cord, pancreatic and prevertebral ganglia, and at neuroeffector synapses. Thus, the PNS and SNS pathways regulating the exocrine pancreas are directly or indirectly antagonistic at multiple sites: the state of exocrine secretion reflects the balance of these influences. Despite over a century of study, much remains to be understood about the connections of specific neurons forming pancreatic pathways, their processes of neurotransmission, and how disruption of these pathways contributes to pancreatic disease.

Morphology and histochemistry of the rabbit pancreatic innervation.

Stimulation of extrinsic nerves markedly alters pancreatic endocrine and exocrine secretion, yet little is known of the neurochemical organization and physiologic roles of specific neural pathways within the pancreas. Here we report histochemical staining for acetylcholinesterase (AChE), NADPH-diaphorase (NADPH-d), nitric oxide synthase (NOS), and several neuropeptides to identify the neurotransmitter content of rabbit pancreatic nerves. An extensive network of AChE-positive nerve fibers was found throughout the islets, acini, ducts, ganglia, and blood vessels. All pancreatic neurons were AChE positive, two thirds were NADPH-d positive, and many were NOS positive. Ganglia in the head/neck region were connected to the duodenal myenteric plexus by AChE- and NADPH-d-positive fibers, and NADPH-d-positive pancreatic neurons appeared to send processes toward both the duodenum and pancreas. Many pancreatic neurons were vasoactive intestinal peptide (VIP) positive, and VIP nerve terminals were abundant in ganglia, acini, islets, and ducts. Pituitary adenylate cyclase-activating peptide (PACAP-38)-positive fibers also were observed within acini and passing through ganglia. Substance P (SP)-, calcitonin gene-related peptide (CGRP)-, and dopamine beta-hydroxylase (DBH)-positive fibers were abundant along blood vessels and ducts, and varicose fibers were observed in pancreatic ganglia. Fine galanin-positive fibers were also occasionally observed running with blood vessels and through ganglia. Thus the rabbit pancreas receives a dense, diverse innervation by cholinergic, adrenergic, and peptidergic nerves and cholinergic pancreatic neurons, most also containing VIP or NOS or both, appear to innervate both endocrine and exocrine tissue, and may mediate local communication between the duodenum and pancreas.

Noradrenergic innervation of rabbit pancreatic ganglia

Sympathetic nerve stimulation indirectly regulates pancreatic endocrine and exocrine secretion, in part, through actions on the cholinergic parasympathetic innervation of the secretory tissues. Earlier work identified noradrenergic nerves in pancreatic ganglia and demonstrated the effects of exogenous norepinephrine (NE) on synaptic transmission but no quantitative studies of ganglionic NE content and release exist. Therefore, the distribution and density of catecholamine (CA)-containing nerves in rabbit pancreatic ganglia were studied using paraformaldehyde/glutaraldehyde (FAGLU) staining and HPLC analysis of CA concentrations. Neural release of [3H]NE was measured in ganglia isolated from the head/neck or body regions of the pancreas. CA-containing nerves densely innervated most ganglia (86%) from both regions, while neural and non-neural CA-containing cell bodies were rarely found. Ganglia from the head/neck region contained significantly higher concentrations of NE. Both 40 mM K+ and veratridine evoked Ca2+-dependent [3H]NE release and tetrodotoxin inhibited 80% of veratridine-stimulated release. omega-Conotoxin GVIA alone antagonized veratridine-stimulated release by 40% but the addition of nifedipine or omega-agatoxin IVA caused no further inhibition. There were no apparent regional differences in the Ca2+-dependence or toxin-sensitivity of NE release. In conclusion, ganglia throughout the rabbit pancreas receive a dense, functional noradrenergic innervation and NE release is dependent upon N- but not P/Q- or L-type voltage-dependent Ca2+ channels. These noradrenergic nerves may indirectly regulate pancreatic secretion through actions on ganglionic transmission.

Catecholamines and 5-hydroxytryptamine in tissues of the rabbit exocrine pancreas.

Objectives: Norepinephrine (NE), dopamine (DA), epinephrine (Epi), and 5-hydroxytryptamine (5-HT) all modulate pancreatic exocrine secretion, yet their concentrations in specific tissues of the exocrine pancreas are unknown. Methods: Concentrations of catecholamines and 5-HT in rabbit pancreatic ganglia, acini, ducts and ampullae, and arteries and veins were measured using HPLC. Results: Concentrations of NE in ganglia from the head/neck region were significantly higher than those from the body (1620 ± 220 vs. 778 ± 179 pmol/mg protein). Acini contained little NE, DA, or 5-HT (9 ± 2, 0.9 ± 0.2, 13 ± 5 pmol/mg protein). Ducts and ampullae contained NE (314 ± 74 and 156 ± 24 pmol/mg protein), DA (43 ± 14 and 13 ± 4 pmol/mg protein), Epi (63 ± 29 and 39 ± 6 pmol/mg protein), and 5-HT (696 ± 151 and 3563 ± 288 pmol/mg protein). Arteries and veins contained the highest concentrations of NE (1962 ± 463 and 736 ± 80 pmol/mg protein, respectively). Conclusions: Pancreatic ganglia and blood vessels, rather than acini, are the main sites of noradrenergic sympathetic innervation of the rabbit exocrine pancreas. These nerves preferentially target ganglionic transmission in the head/neck versus the body. Serotonergic nerves provide little or no innervation of rabbit pancreatic ganglia or acini.

Electrical properties and synaptic potentials of rabbit pancreatic neurons.

Pancreatic ganglia receive innervation from a wide variety of extrinsic nerves and supply the predominant innervation to pancreatic acini, islets, and ducts. This study used intracellular recordings to investigate the electrical properties and synaptic potentials of rabbit pancreatic neurons. Neurons had a mean resting membrane potential of -54+/-0.4 mV and generated action potentials with a mean overshoot of 10+/-0.4 mV and a mean after-spike hyperpolarization (ASH) of 11+/-0.5 mV with duration of 210+/-19 ms. Action potentials exhibited a high threshold (-15+/-1 mV) for intracellular stimulation and a phasic firing pattern was observed in response to prolonged depolarizing currents. Stimulation of attached nerve bundles evoked multiple fast excitatory postsynaptic potentials (fEPSPs) which were abolished by hexamethonium in 75% of neurons, while a non-cholinergic fEPSP was observed in 25% of the neurons. Repetitive stimulation (3-30 Hz) evoked muscarinic slow EPSPs with a mean amplitude of 8+/-2 mV and duration of 5+/-1 s in a small subset (21%) of neurons. Exogenous muscarine evoked a mean slow depolarization of 10+/-1 mV amplitude in 22% of neurons tested. Following repetitive nerve stimulation non-cholinergic late, slow EPSPs with a mean amplitude of 4.3+/-0.4 mV were recorded in 32% of neurons. Nicotinic transmission was subject to inhibition mediated by presynaptic muscarinic receptors at low (0.5 Hz) stimulus frequencies in 80% of neurons. At higher frequencies (> or =1 Hz), either facilitation or depression of nicotinic transmission was observed depending on the ganglion studied. A population (9%) of neurons exhibited spontaneous, low-amplitude pacemaker-like potentials. Spontaneous fEPSPs and action potentials were also observed and these occasionally occurred in rhythmically timed bursts. Thus, distinct subpopulations of pancreatic neurons could be identified on the basis of both their intrinsic electrical properties and the receptors mediating and/or modulating synaptic transmission. These neurons function as critical sites of integration for synaptic input from extrinsic pancreatic nerves and thereby determine the postganglionic firing patterns presented to the pancreatic exocrine and endocrine secretory cells.

Veratridine-stimulated amylase secretion from rabbit pancreatic lobules: role of cholinergic and noncholinergic receptors.

Stimulation of pancreatic nerves results in marked increases in exocrine secretion. However, the neurotransmitters and pre- and postsynaptic receptors, which determine synaptic transmission between nerves and acinar cells, are poorly defined. We used rabbit pancreatic lobules, which contain nerve terminals and secrete independently of the influences of vascular perfusion or gastrointestinal hormones, to study the role of cholinergic and noncholinergic nerves in regulating amylase secretion. Pancreatic nerves were stimulated by veratridine (Ver; 50-200 microM), an activator of voltage-dependent sodium channels, in a concentration-dependent and tetrodotoxin-sensitive manner, resulting in an increase of 138+/-15% in amylase secretion above basal at 100 microM. This stimulation was unaffected by either hexamethonium (100 microM) or the combination of phentolamine and propranolol (10 microM). Atropine (5 microM) inhibited Ver-stimulated secretion by approximately 65-70%. Bethanechol (Bch; 0.01-100 microM) increased amylase secretion in a concentration-dependent manner (EC50, 6.2 microM), with a maximal stimulation of 177+/-15% above basal. Antagonism of Bch-stimulated secretion with 4-diphenylacetoxy-N-methyl-piperidine, pirenzepine (Pzp), or methoctramine (Met) resulted in IC50 values of 7.9 nM, 282 nM, and 79.8 microM, respectively. Ver-stimulated secretion was unaffected by Pzp (0.1 and 1 nM) or Met (1 and 100 nM) at concentrations that had no significant effect on Bch-stimulated secretion. Thus cholinergic nerves, activating postsynaptic M3 receptors, provided the predominant stimulatory innervation of rabbit pancreatic acini. Nonadrenergic, noncholinergic nerves also made a significant contribution to secretion. Adrenergic nerves did not appear to innervate acini or the excitatory cholinergic nerves directly.

Role of adrenergic receptors in veratridine-stimulated amylase secretion from rabbit pancreatic lobules.

Sympathetic inhibition of pancreatic enzyme secretion has been attributed to vasoconstriction and direct inhibition of acinar cells. We observed both adrenergic inhibition and facilitation of cholinergic transmission in rabbit pancreatic ganglia, which innervate acini. Here we used pancreatic lobules to determine whether adrenergic receptors also regulate synaptic transmission between pancreatic nerves and acini. Stimulation of pancreatic nerve terminals with veratridine (Ver), an activator of voltage-dependent Na+ channels, resulted in a 102% increase in amylase secretion, which was unaffected by &agr; and &bgr; receptor antagonists but inhibited 65% by atropine. At a concentration of 10 &mgr; M, norepinephrine inhibited (38%) and epinephrine potentiated (40%) Ver-stimulated secretion. At the same concentration, the &agr;2 agonist clonidine (Clon) inhibited (39%), whereas the nonselective &bgr; agonist isoproterenol (Iso) and the selective &bgr;3 agonist BRL 37344 potentiated (71 and 67%, respectively) nerve-stimulated secretion. The effects of Clon and Iso and BRL 37344 were antagonized by yohimbine and propranolol, respectively. Phenylephrine, dobutamine, and terbutaline had no effect. Neither basal, bethanechol-stimulated, nor noncholinergic nerve-stimulated secretion was significantly altered by Clon or Iso. Thus, cholinergic nerve terminals innervating pancreatic acini exhibit both inhibitory &agr;2 and atypical facilitatory &bgr; adrenergic receptors. The apparent lack of adrenergic innervation suggests that adrenergic receptors on the nerve terminals of cholinergic pancreatic neurons are under hormonal control by circulating catecholamines. These results provide further evidence that intrinsic pancreatic neurons, which supply most, if not all, of the cholinergic innervation of acini, are important sites of sympathetic regulation of pancreatic exocrine secretion.

Alpha-adrenergic modulation of synaptic transmission in rabbit pancreatic ganglia

Pancreatic ganglia contain noradrenergic nerve terminals whose role in ganglionic transmission is unknown. Intracellular recordings from rabbit pancreatic neurons were used to study the effects of alpha-adrenergic agonists and antagonists on ganglionic transmission and to determine if endogenously released norepinephrine contributed to synaptic depression. Significant regional differences in alpha adrenergic effects were observed. In neurons from ganglia of the head/neck region norepinephrine or selective alpha(2) agonists presynaptically inhibited ganglionic transmission and this effect was antagonized by the alpha(2) antagonist yohimbine. In the majority of cells membrane hyperpolarization accompanied presynaptic inhibition during superfusion of alpha(2) agonists. Repetitive nerve stimulation evoked a presynaptic post-train depression (PTD) of ganglionic transmission in all neurons tested. A combination of nisoxetine (selective inhibitor of the norepinephrine transporter) and tyramine (releaser of endogenous catecholamines) increased PTD. Pretreatment with clonidine inhibited synaptic transmission and abolished PTD while yohimbine did not affect it. Pretreatment with guanethidine (>or=3.5 h) also failed reduce PTD while neurons unresponsive to alpha(2) adrenoceptor agonists routinely exhibited PTD, implying the presence of other inhibitory neurotransmitters sharing a common presynaptic mechanism with alpha(2) agonists. In the majority of neurons from ganglia of the body region superfusion of norepinephrine or the selective alpha(1) agonist phenylephrine evoked membrane depolarization and facilitated ganglionic transmission. These effects were antagonized by the alpha(1) antagonist prazosin. The remaining neurons exhibited either alpha(2)-mediated synaptic inhibition or no-response. In conclusion, inhibitory alpha(2) and excitatory alpha(1) adrenoceptors exist in pancreatic ganglia and predominate in the head/neck and body, respectively. Norepinephrine, released during repetitive nerve stimulation, may contribute to synaptic depression in the head/neck region and appeared to share a common mechanism with other, unidentified neurotransmitters mediating synaptic depression in both regions. These differences indicate a functional heterogeneity of pancreatic sympathetic innervation that may reflect the reported regional differences in exocrine and endocrine cells.

Short-term synaptic plasticity in rabbit pancreatic ganglia

The extrinsic innervation of the pancreas converges on a plexus of intrinsic pancreatic ganglia whose cholinergic neurons innervate acini, ducts, islets and blood vessels. Therefore, understanding ganglionic transmission is essential for understanding neural control of pancreatic secretion. Intracellular recordings of nicotinic fast excitatory postsynaptic potentials (fEPSPs) and action potentials (APs) were used to characterize and compare transmission in ganglia from the head/neck and body regions of the rabbit pancreas. Paired-pulse facilitation (PPF) or depression (PPD) of fEPSPs was observed in ganglia from both regions with PPF peaking and disappearing at shorter inter-stimulus intervals than PPD. PPF was most frequent in the head/neck (60%) and PPD (50%) in the body. Repetitive stimulation (10 Hz/5 s) evoked multiple forms of mid- and post-train plasticity. Facilitation during the first 1-2 s of train stimulation was reduced or reversed with continued stimulation due to development of synaptic depression and mid-train depression was of greater magnitude in the head/neck region. A brief (approximately 10 s) post-train augmentation was followed by a 1-2 min post-train depression that appeared to result from inhibition of ACh release. Regional differences in the frequency, magnitude, or duration of all forms of synaptic plasticity suggested regional differences in the extrinsic innervation patterns and possibly the function of pancreatic ganglia. In conclusion, rabbit pancreatic ganglia exhibit multiple forms of short-term synaptic plasticity that markedly alter the probability of postsynaptic firing, consistent with these ganglia being critical sites of synaptic integration and autonomic regulation of pancreatic secretion.