Ij. Llewellynsmith et al., GLUTAMATE-IMMUNOREACTIVE AND GABA-IMMUNOREACTIVE SYNAPSES ON SYMPATHETIC PREGANGLIONIC NEURONS CAUDAL TO A SPINAL-CORD TRANSECTION IN RATS, Neuroscience, 80(4), 1997, pp. 1225-1235
Spinal cord injury destroys bulbospinal amino acid-containing pathways
to sympathetic preganglionic neurons and severely disrupts blood pres
sure control, resulting in resting or postural hypotension and episodi
c hypertension. Almost all immunoreactivity far the excitatory amino a
cid L-glutamate has been reported to disappear from autonomic areas of
the cord caudal to a transection, apparently depriving autonomic neur
ons of their major excitatory input. However, the magnitude of the neu
rogenic episodic hypertension after cord injury suggests that excitato
ry inputs to sympathetic preganglionic neurons must still be present.
Moreover, the hypotension associated with high spinal injuries may ref
lect a enhanced role for inhibitory transmitters, such as GABA. This a
pparent contradiction regarding the presence of glutamate and lack of
information about GABA prompted the present investigation. In rats sev
en days after spinal cord transection, we examined identified sympathe
tic preganglionic neurons caudal to the injury for the presence of syn
apses or direct contacts from Varicosities that were immunoreactive fo
r the amino acids, L-glutamate and GABA. Adrenal sympathetic pregangli
onic neurons were retrogradely labelled with cholera toxin B subunit a
nd amino acid immunoreactivity was revealed with post-embedding immuno
gold labelling. In single ultrathin sections, 46% (98/212) of the syna
pses or direct contacts on adrenal sympathetic preganglionic neurons w
ere immunoreactive for glutamate and 39% (83/214) were immunoreactive
for GABA. Analysis of inputs with the physical disector yielded simila
r results for the two amino acids. The proportions of glutamatergic or
GABAergic synapses on cell bodies and dendrites were similar. When al
ternate ultrathin sections were stained to reveal glutamate or GABA im
munoreactivity, either one or the other amino acid occurred in 78.4% (
116/148) of inputs; 4.1% (6/148) of inputs contained both amino acids
and 17.5% (26/148) of inputs contained neither. These results demonstr
ate that nerve fibres immunoreactive for the neurotransmitter amino ac
ids, glutamate and GABA, provide most of the input to sympathetic preg
anglionic neurons caudal to a spinal cord transection. Synapses contai
ning glutamate and GABA could provide the anatomical substrate For the
exaggerated sympathetic reflexes and the low sympathetic tone that re
sult from spinal cord injury. (C) 1997 IBRO. Published by Elsevier Sci
ence Ltd.