~ Pergamon Neuroscience Vol. 73, No. 2, pp. 509-518, 1996 Copyright © 1996 IBRO. Published by ElsevierScienceLtd Printed in Great Britain S0306-4522(96)00063 -2 0306-4522/96 $15.00 + 0.00 DIRECT EVIDENCE OF AN EXTENSIVE GABAERGIC INNERVATION OF THE SPINAL DORSAL HORN BY FIBRES DESCENDING FROM THE ROSTRAL VENTROMEDIAL MEDULLA M. ANTAL,*t M. PETKO,* E. POLG,/~R, * C. W. HEIZMANN~ and J. STORM-MATHISEN§ • Department of Anatomy, University Medical School, Debrecen, H-4012, Hungary ~Department of Pediatrics, Division of Clinical Chemistry, University of Zurich, Zurich, CH-8032, Switzerland §Anatomical Institute, University of Oslo, Oslo, N-0317, Norway Abstract--A long line of studies emphasizes the contribution of serotonergic fibres descending from the rostral ventromedial medulla in the control of spinal nociceptive information processing. A growing body of evidence, however, suggests that the relative contribution of serotonin to the mediation of spinal neuronal activity from the rostral ventromedial medulla may require re-evaluation. It has recently been substantiated that, in addition to the serotonergic fibres, the spinal dorsal horn receives an abundant non-serotonergic projection from the rostral ventromedial medulla. Furthermore, stimulation in the rostral ventromedial medulla could result in a powerful inhibition of nociceptive spinothalamic tract cells without any detectable serotonin release in the dorsal horn. After labelling raph~spinal axons and axon terminals in the rat by iontophoretic injections of the anterograde axonal tracer Phaseolus vulgaris leucoagglutinin into the central region of the rostral ventromedial medulla (nucleus raphe magnus) and revealing GABA and glycine immunoreactivites of the labelled raphe-spinal terminals and their postsynaptic targets by postembedding immunocytochemical methods, here we demonstrate an extensive GABAergic projection from the rostral ventromedial medulla to the spinal dorsal horn. We show that the majority of the labelled raphe--spinal terminals in laminae I IIo and IV-V contain GABA and some of the GABA-immunoreactive terminals are also immunoreactive for glycine. We also disclose that GABA-immunoreactive raphe-spinal terminals establish synaptic contacts primarily with GABA- and glycine-negative, presumably excitatory, spinal neurons, including Calbindin-D28 k- as well as parvalbumin- immunoreactive cells in both laminae I-IIo and IV V. The results suggest that volleys in fibres descending from the rostral ventromedial medulla may evoke GABA release from raphe spinal terminals, and the released GABA, in some cases probably acting together with glycine, might play a crucial, as yet mostly unidentified, role in the inhibition of nociceptive information processing in the dorsal horn of the spinal cord. Copyright © 1996 IBRO. Published by Elsevier Science Ltd. Key words: descending pain control, inhibitory amino acids, neural tracing, calcium-binding proteins, immunocytochemistry, rat. Stimulation of the rostral ventromedial medulla (RVM), which includes the nucleus raphe magnus (NRM) and the adjacent nuclei of the reticular formation [nucleus reticularis paragigantocellularis, nucleus reticularis gigantocellularis pars alpha (GiA), nucleus reticularis magnocellularis], produces inhi- bition of nociceptive neurons in the spinal dorsal horn, resulting in both analgesia and a powerful attenuation of pain behaviour. 16'17'45 It has generally been assumed that inhibition of spinal nociceptive transmission evoked by RVM stimulation is due to the activation of a spinally projecting cell population tTo whom correspondence should be addressed. Abbreviations: GiA, gigantocellular reticular nucleus pars alpha; 5-HT, serotonin; NRM, nucleus raphe magnus; PHA-L, Phaseolus vulgaris leucoagglutinin; RVM, ros- tral ventromedial medulla. that releases serotonin (5-HT) H'~2 and suppresses nociceptive information processing in the dorsal horn through 5-HTm and 5-HT 3 receptors. TM Evidence indicating that the antinociceptive effect of RVM stimulation is mediated by 5-HT released into the dorsal horn, however, remains inconclusive. A grow- ing body of experimental data suggests that RVM is not a homogeneous population of serotonergic cells, 1°'24"25'37'39 and that multiple neurotransmitter systems are possibly involved when raphe-spinal neurons are activated. 9'~4'33Thus, it is quite likely that neurotransmitters other than 5-HT participate in RVM-produced antinociception and are released following stimulation in the absence of, or in addition to, 5-HT. Recently, it has been proposed that GABA- ergic and glycinergic neurons of the RVM also contribute to the raphe-spinal inhibitory influences 509