Neuropeptides, 'gaseous' messengers and classic transmitters : electrophysiological and histochemical studies on coexistence and interactions in the nervous system

Abstract

The distribution and functional role of neuropeptides were studied, with special reference to galanin (GAL) and the free radical nitric oxide (NO). In developing systems GAL expression was observed already at E14 in trigeminal and dorsal root ganglion (DRG) neurons and at E15 in sensory epithelia in ear, eye and nose, as well as at E19 during bone formation. Also GAL-R1 receptor mRNA was expressed in the sensory ganglia of embryos but was detected later than the ligand. Thus GAL may have a developmental role in several sensory systems and during bone formation. The GAL-R1 receptor was present in CGRP neurons and was down-regulated by peripheral nerve injury and inflammation. GAL did not change the membrane current in DRG neurons from normal rats, but caused an inward current in DRG neurons from sciatic nerve transected (SNT) rats. Similar results were obtained with CCK-8S. In normal rats neurotensin (NT) receptor mRNA was expressed in about 25% of the small DRG neuron profiles, and NT evoked an outward current in neuropeptide Y (NPY)-insensitive C-type (small) neurons, while NPY induced an outward current in NT-insensitive C-type neurons. In axotomized rats NT only induced an inward current in C-type neurons. These results give evidence for functional peptide receptors in the DRG neurons and that nerve injury causes distinct changes in receptor phenotype, perhaps reflecting attempts of the DRG neurons to cope with, and to counteract the consequences of nerve injury.

In the central nervous system, GAL/GMAP-positive fibers in the dorsal and many in the ventral hippocampal formation were noradrenergic. GAL-R2 receptor mRNA was expressed in the granule cell layer in the dentate gyrus. However, especially in the ventral hippocampus some GAL/GMAP fibers were DBH-negative, and did not disappear after 6-OH-DA treatment. This suggests that GAL/GMAP are present in multiple hippocampal systems and that most GALergic fibers are noradrenergic arising from locus coeruleus(LC). GAL-evoked hyperpolarization of LC neurons, and this was accompanied by a decrease in membrane resistance as recorded in a slice preparation. Binding studies revealed GAL binding sites in the LC, but GAL-R1 and GAL-R2 mRNAs were not detectable. Ultrastructural immunocytochemistry showed GAL in many neuronal somata and dendritic processes within the nucleus. These findings suggest that GAL exerts its inhibitory effect in the LC, probably via an increase in K+ conductance, and that endogenous GAL, possibly released from the LC cell bodies and dendrites, may act on autoreceptors or receptors on adjacent neurons in LC.

In the dorsal raphe (DR), 5-HT/GAL/NO synthase (NOS)-coexistence neurons were found mainly in the midline. All GAL neuron profiles contained 5-HT, and the proportion of 5-HT/GAL neurons was high while that of 5-HT/NOS neurons was low. Some 5-HT/GAL/NOS neurons were found to project to the striatum. However, only in a few cases could GAL-LI be shown to coexist with 5-HT in striatal fibers with no NOS/5-HT coexistence. The present findings strongly suggest that many DR 5-HT neurons can synthesize and release two additional messenger molecules, GAL and NO. However, little GAL and NOS seem to be transported anterogradely to the nerve terminals in the striatum, supporting the concept of release at the soma/dendrite level. GAL caused a hyperpolarization of 5-HT of DR neurons accompanied by a decrease in membrane resistance. The 5-HT-induced outward current was enhanced and prolonged by preincubation with low concentrations of GAL. The dose-response curve to 5-HT was changed by GAL with a shift to the left. These results suggest that GAL exerts its effect in the DR directly by acting at receptors on the membrane of 5-HT-sensitive neurons via an increase in K+ conductance and indirectly by enhancing the 5-HT action.

A small number of NOS-positive, non-noradrenergic neurons were observed within the LC. NOS inhibitors enhanced the EPSP in LC, an effect that was reversed by coadministration of L-arginine. Application of NO-donors increased levels of cGMP as seen with immunohistochemistry as well as induced a hyperpolarization and reduced the EPSP/EPSC in LC neurons. Prior application of hemoglobin prevented the action of NO-donors and enhanced the EPSP/EPSC. Application of the membrane permeable cGMP analog 8-bromo-cGMP mimicked the action of NO-donors. Preincubation with the guanylate cyclase inhibitor ODQ reduced NO-donor-induced hyperpolarization. These results suggest a role for NO in synaptic transmission in the LC, which may act through the NO-cGMP pathway by stimulating guanylyl cyclase and increasing endogenous levels of cGMP. Even if only few NOS-positive neurons are present in the LC, NO may after release from cell bodies, dendrites and axonal processes via diffusion influence many noradrenergic LC neurons.