The changes in thoracic sympathetic nerve activity, heart rate and frequency of phrenic nerve discharge in response to chemoreflex activation before and after bilateral microinjections of glutamate receptor antagonists into the comissural nucleus tractus solitarii (cNTS) were evaluated in the working heart-brainstem preparation of rats. Microinjections of kynurenic acid (KYN, 250 mM), (+/-)-alpha-methyl-4-carboxyphenylglycine (MCPG, 100 mM), or KYN plus MCPG into the cNTS were performed in three different groups. These microinjections into the cNTS did not affect the increase in the thoracic sympathetic nerve activity elicited by chemoreflex activation (KYN, 54 +/- 3 versus 51 +/- 2%, n = 11; MCPG, 48 +/- 5 versus 54 +/- 5%, n = 7; and KYN plus MCPG, 57 +/- 6 versus 55 +/- 3%, n = 5). The increase in the frequency of the phrenic nerve discharge in response to chemoreflex activation was also not affected by KYN (0.28 +/- 0.02 versus 0.30 +/- 0.04 Hz), MCPG (0.27 +/- 0.03 versus 0.27 +/- 0.04 Hz), or KYN plus MCPG (0.30 +/- 0.04 versus 0.20 +/- 0.03 Hz). The bradycardic response to chemoreflex activation was significantly reduced after microinjection of KYN at 2 (-220 +/- 16 versus -50 +/- 6 beats min(-1)) and 10 min (-220 +/- 16 versus -65 +/- 9 beats min(-1)) and after microinjection of KYN plus MCPG into the NTS it was abolished at 2 (-192 +/- 14 versus -2 +/- 1 beats min(-1)) and 10 min (-192 +/- 14 versus -4 +/- 2 beats min(-1)). These data support the hypothesis that the neurotransmission of the sympathoexcitatory and respiratory components of the chemoreflex in the cNTS involves neurotransmitters other than L-glutamate and also the concept that the parasympathetic component of this reflex is mediated by L-glutamate.