Medullo-Spinal Modulation of Sympathetic Output and Spinal Afferent Input

Spinal interneurons infected by renal injection of pseudorabies virus in the rat

The potency of spinal sympathetic reflexes is increased after spinal injury, and these reflexes may result in life-threatening hypertensive crises in humans. Few, if any, primary afferents project directly to sympathetic preganglionic neurons (SPN). Therefore, spinal sympathetic interneurons (IN) must play a major role in generating dysfunctional sympathetic activity after spinal cord injury. Furthermore, these IN are potentially aberrant targets, either for ascending and descending axons that may sprout after spinal cord injury or for axons that regenerate after spinal cord injury. We identified IN via the transsynaptic retrograde transport of pseudorabies virus (PRV) injected into the kidneys of rats. The proportion of infected IN ranged from approximately 1/3 to approximately 2/3 of the number of infected SPN. IN were heavily concentrated among the SPN in spinal lamina VII. However, IN were located in all lamina of the dorsal horn. The longitudinal distribution of infected IN was closely correlated with the longitudinal distribution of infected SPN. Few infected IN were found rostral or caudal to the longitudinal range of infected SPN. Infected IN were heterogeneous in both their sizes and the extent of their dendritic trees. The strong correlation between longitudinal distributions of infected IN and SPN supports physiological data demonstrating a segmental organization of spinal sympathetic reflexes. The paucity of infected IN in segments distant from SPN suggests that multisegmental sympathetic reflexes are mediated by projections onto IN rather than onto SPN themselves. The morphological heterogeneity of IN probably manifests the variety of systems that affect spinal sympathetic regulation.

Locations and morphologies of sympathetically correlated neurons in the T(10) spinal segment of the rat

We precisely localized and morphologically characterized sympathetically correlated neurons in the acutely transected spinal cord of the rat. We have shown that these neurons are likely members of the spinal networks that generate sympathetic activity after spinal cord transection. In humans with injured spinal cords, these networks are responsible for hypertensive crises that occur in response to ordinarily innocuous stimuli. We recorded from neurons in the dorsal horn of the T(10) spinal segment of anesthetized rats after acute spinal cord transection at C(2). Neurons with activities closely correlated to renal sympathetic nerve activity (RSNA) were considered to be putative components of spinal sympathetic systems. These neurons had receptive fields on the left flank and abdomen. After characterizing their ongoing activities, receptive fields, and degrees of correlation with RSNA, we juxtacellularly labeled neurons with biotinamide and subsequently reconstructed their somas and dendrites histologically. Confirming our earlier studies, sympathetically correlated neurons were found in dorsal horn laminae III, IV, and V. For the first time, we also identified sympathetically correlated neurons in laminae I and II. The dendrites of all sympathetically correlated neurons projected to multiple lamina. By virtue of the positions of their somas and the broad projections of their dendrites, we concluded that sympathetically correlated neurons may receive direct input both from supraspinal systems and from nociceptive and non-nociceptive primary afferents.

Spinal interneurons play a minor role in generating ongoing renal sympathetic nerve activity in spinally intact rats

The purpose of the present study was to determine whether spinal interneurons play a role in the regulation of sympathetic activity in spinally intact rats. In acutely spinally transected rats, we have described a population of spinal interneurons that, by virtue of correlations between their ongoing firing rates and the magnitude of ongoing renal sympathetic nerve activity (RSNA), are candidates for generators of sympathetic activity. Further evidence for a sympathetic role for these neurons comes from our observation that cervical spinal stimulation that reduces RSNA also reduces their discharge rates. In chloralose-anesthetized, spinally intact and spinally transected rats, we recorded ongoing RSNA and the ongoing activities of T(10) dorsal horn and intermediate zone interneurons, and we determined the incidence of sympathetically related neurons in these rats by cross-correlating their activities with RSNA. The incidence of correlated neurons was much smaller in spinally intact than in spinally transected rats. We stimulated the dorsolateral, C(2-3) spinal cord before and after acute C(1) spinal transection. Dorsolateral cervical stimulation in spinally transected rats reduced both RSNA and the activities of most T(10) interneurons, but stimulation in spinally intact rats increased RSNA while still reducing the activities of most T(10) interneurons. Both the low incidence of sympathetically correlated spinal neurons in intact rats and the dissociation between the effects of cervical stimulation on RSNA and the discharge rates of spinal interneurons argue against these neurons playing a major role in regulating sympathetic activity in intact rats.

Identification of spinal interneurons antecedent to adrenal sympathetic preganglionic neurons using trans-synaptic transport of herpes simplex virus type 1

Control of sympathetic preganglionic neurons appears to be mediated, in part, through polysynaptic pathways using spinal interneurons. To identify spinal interneurons antecedent to adrenal sympathetic preganglionic neurons, we injected herpes simplex virus type 1 into the adrenal gland of hamsters as this virus is an effective trans-synaptic tracer of neural pathways. After a three day survival period, immunocytochemistry was used to visualize virus-infected spinal cord cells. Infected sympathetic preganglionic neurons with somata that were either kite-shaped, elliptical or fusiform and that had extensive dendrite arbors were identified as well as a group of smaller round cells with finer processes. For comparison, in additional hamsters, labelling with the retrograde tracer Fluoro-Gold and histochemical reactions for the enzyme nicotinamide adenine dinucleotide phosphate-diaphorase were used to identify sympathetic preganglionic neurons. Sympathetic preganglionic neurons identified with Fluoro-Gold or herpes virus were present mostly in the nucleus intermediolateralis, pars intermediolateralis and nucleus intermediolateralis, pars funicularis of the spinal cord. The smaller herpes virus-infected cells were found mostly medial to the preganglionic neurons in lamina VII and also dorsally in lamina V of the spinal cord. Assessing immunoreactivity for glial fibrillary acidic protein demonstrated that the smaller herpes virus-infected cells were not reactive astrocytes. Furthermore, these cells were immunoreactive for two neuronal markers, neuron-specific enolase and for microtubule-associated protein 2. These findings suggest that these smaller round cells with finer processes are distinct from sympathetic preganglionic neurons and astrocytes and may be interneurons antecedent to the sympathetic preganglionic neurons.

Interaction between medullary and cervical regulation of renal sympathetic activity

We have reported that electrical or glutamate stimulation of the dorsolateral surface of the cervical spinal cord elicits a 40-60% decrease in renal sympathetic activity (RSA) in anesthetized rats. Because evoked sympatho-inhibition was observed, however, only after transection of the cervical spinal cord at C1, we suggested that unidentified supraspinal neurons affect the regulation of RSA by cervical neurons. In the present experiments, we tested the hypothesis that the modulatory supraspinal neurons are located in the ventrolateral medulla by observing the effects of rostroventral, lateral, medullary (RVLM) injections of the GABAergic agonist, muscimol, on baseline RSA and on our ability to inhibit that activity by cervical stimulation. GABAergic inhibition in the RVLM of chlorolose anesthetized rats elicited changes in RSA that were similar to those observed after transection of the spinal cord, including a 41% decrease in mean arterial pressure and a 44% increase in RSA. Moreover, after muscimol inhibition of RVLM neurons, electrical or glutamate stimulation of the dorsolateral cervical spinal cord elicited a decrease in RSA in otherwise intact rats. These results suggest that neurons in the RVLM interact with neurons in the cervical spinal cord in the regulation of RSA.

Role of cervical neurons in propriospinal inhibition of thoracic dorsal horn neurons

We previously reported that electrical or glutamate stimulation of the cervical spinal cord elicits a 40-60% decrease in renal sympathetic nerve activity (RSA) in the anesthetized rats. This sympatho-inhibition was possible, however, only after transection of the spinal cord at C1 or GABAergic inhibition of neurons in the rostral ventrolateral medulla. We postulated that cervical neurons inhibit RSA by inhibiting the activity of spinal interneurons that are antecedent to sympathetic preganglionic neurons (SPNs), and that these interneurons may be, in turn, excited by afferent signals. In this study, we tested the hypothesis that cervical neurons can inhibit visceroceptive thoracic spinal neurons. We recorded the spontaneous and evoked activity of 45 dorsal horn neurons responsive to splanchnic stimulation before, during, and after chemical or electrical stimulation of the cervical spinal cord in chloralose-anesthetized spinal rats. Cervical spinal stimulation that inhibited RSA also inhibited the spontaneous and/or evoked activity of 44 dorsal horn neurons. In addition to inhibiting splanchnic-evoked neuronal responses, cervical stimulation also inhibited responses, in the same neurons, evoked by noxious heat or light brushing of receptive dermatomes. We concluded that cervical neurons participate in propriospinal inhibition of afferent transmission and that this inhibitory system may be involved in controlling the access of afferent information to SPNs.

Coexistence of autonomic and somatic mechanisms in the pressor areas of medulla in cats

The effects of electrical stimulation and microinjection of sodium glutamate (0.5 M) in the sympathetic pressor areas of the dorsal medulla (DM), ventrolateral medulla (VLM), and parvocellular nucleus (PVC) on the knee jerk, crossed extension, and evoked potential of the L5 ventral root produced by intermittent electrical stimulation were studied in 98 adult cats anesthetized with chloralose and urethane. During electrical and glutamate stimulation of these pressor areas, in addition to the rise of systemic arterial blood pressure marked inhibition of the spinal reflex was produced, indicating presence of neuronal perikarya responsible for these actions. Mild to moderate augmentation of spinal reflexes was also observed during brain stimulation but only in a few cases. The magnitude of the somatic effects among the pressor areas of the VLM, DM, and PVC subsequent to glutamate activation was about the same. Induced spinal reflex inhibition, independent from the baroreceptor and vagal influence, remained essentially unaltered after acute midcollicular decerebration. The inhibition was also observed in cats decerebellated 8-10 days in advance. The inhibition was not affected after bilateral electrolytic- or kainic-acid-induced lesions in the paramedian reticular nucleus (PRN). On the contrary, PRN-induced spinal reflex inhibition was attenuated after bilateral lesions in the DM or VLM. Data suggest that there coexists neuronal subpopulations in the VLM, DM, and PVC that can affect both the sympathetic pressor systems and spinal reflexes.