Nitric oxide synthase neurons in the rodent spinal cord: distribution, relation to Substance P fibers, and effects of dorsal rhizotomy.

The indirect immunofluorescent method was employed to investigate the distribution of neuronal nitric oxide synthase-like immunoreactivity(nNOS-LI) in the spinal cord of the golden hamster and to compare it to data obtained from rats. Immunoreactive neurons were found throughout the cervico-sacral extent in the dorsal horn (mainly in laminae I-III) and in the preganglionic autonomic regions, i.e., the sympathetic intermediolateral nucleus (IML), lateral funicle (LF), intercalated region (IC), the area surrounding the central canal (CA), and the sacral preganglionic parasympathetic cell group. While the distribution of immunoreactive cells was generally similar in both species, some differences were observed. For example in the hamster LF, a higher percentage of stained neurons was seen than in the IML, while the situation was rather inverse in the rat. In order to study the coincidence of nNOS-LI in the population of preganglionic sympathetic neurons (PSN) that innervate the superior cervical ganglion (SCG), these were identified by retrograde axonal transport of fluoro-gold (FG) following unilateral injection into the SCG. PSN were localized ipsilateral to the injection site mainly in the IML and LF of spinal segments C7-Th4. The portion of double-labeled neurons of the IML were lower in hamster (17% in C7, 34% in C8) of FG-labeled cells) than in rat (47% in C8, 77% in Th2), while in the LF of segments C8-Th2 in both species the majority of FG-neurons contained nNOS. While only very few double-labeled neurons were detected in the IC in hamster and rat, a striking difference was observed in the CA, where no double-labeled neurons were found in hamster, but up to 50% in rat. Double immunofluorescence detection of nNOS and substance P (SP) showed that in both the autonomic regions and the dorsal horn, SP-LI fibers and puncta were present in close spatial relationship to nNOS-LI cell bodies. These results were basically identical in the hamster and rat. Unilateral transection of the dorsal roots of segments C6-Th2 in rats resulted in a clear reduction of SP-LI structures in the dorsal horn 5 days after rhizotomy, but not in the autonomic regions. Compared to the unlesioned side, the numbers of nNOS-LI neurons in the superficial laminae of the dorsal horn were reduced to 32-46% in the lesioned segments, and to 53% and 87%, respectively, in the two segments cranial to the rhizotomized segments but remained unchanged caudally to the lesion. Numbers of nNOS-LI cell bodies in the autonomic regions were not altered following dorsal root transection. The present study provides data on the widespread distribution of nNOS in the spinal cord of golden hamster and describes the partial coincidence of the enzyme in PSN. The effects of dorsal rhizotomy on nNOS-LI neurons in the dorsal horn reveal that primary-afferent fibers provide a stimulatory influence on neurons of the dorsal horn to generate the gaseous neuroactive substance, nitric oxide.