Coexistence of glutamate and substance P in dorsal root ganglion neurons of the rat and monkey

Abstract

Antisera to a glutamate (Glu) conjugate, to glutaminase (GLN), and to substance P (SP) have been used to investigate the issue of putative glutamergic neurons in the dorsal root ganglia (DRG) and the possible coexistence in these neurons of Glu and SP. The Glu antiserum, characterized by immunoadsorption and immunoblot tests, is highly selective for Glu out of a number of other amino acids including aspartate. Quantitative data were gathered from consecutive 4-μm-thick paraffin sections from cervical ganglia of rats with or without spinal injections of colchicine and from one monkey not treated with colchicine. Neurons containing more than one antigen could be identified on adjacent sections tested with the three different antisera. Neurons labeled by the Glu-antiserum represent 15–30% of the DRG population in untreated rats. They include most of the small neurons (with mean perikaryal area around 300–400 μm) but also larger neurons (with perikaryal area greater than 600 μm). DRG neurons labeled by either the GLN or the SP antiserum are small (mean area 335–375 μm and 356–374 μm, respectively) and account for approximately 40 and 15%, respectively, of the sampled neurons. In colchicine-treated rats the number of Glu-positive neurons increases up to about 70%. The same tratment also increases the number of SP-positive neurons but not that of GLN-positive neurons. In the monkey about half of the DRG neurons are Glu positive and, as in rats, they are mostly small. GLN-positive neurons in the same species account for about half of the DRG population and are only small. In rats, about 60–80% of Glu-positive neurons are also GLN-positive and most GLN-positive neurons are Glu positive. In colchicine-treated rats, close to 90% of SP-positive neurons are also positive for Glu and about 60% of SP-positive neurons are triple labeled. In the monkey's DRGs, whose sections were not processed for SP, most Glu- or GLN-positive neurons are also positive for the other antiserum used. Several considerations argue against the possibility that, under the present experimental conditions, the Glu antiserum is a metabolic marker. It is therefore suggested that Glu-positive neurons may use Glu as neurotransmitter. Coexistence of Glu and SP in some DRG neurons suggests that both agents may be released by terminals of primary afferents from the same DRG neuron. A role for SP as neuromediator in afferents that use another agent as a transmitter is thus supported by the present results. Our data also support previous electrophysiological data suggesting that the same primary afferent terminal may discharge dorsal horn neurons by way of a classical neurotransmitter and also induce in these a prolonged, peptidemediated, slow depolarization.