Elevated intrinsic reactivity of seryl hydroxyl groups within the linear peptide triads His-Xaa-Ser or Ser-Xaa-His.

Chemical modification studies of peptide hormones and random peptides have revealed that seryl hydroxyl groups had enhanced reactivity toward acylating reagents when they occurred in the linear triads His-Xaa-Ser or Ser-Xaa-His (Xaa = any amino acid). O-acylation of serine within these triads was achieved by reaction with N-hydroxysuccinimide esters of biotin (NHS-biotin) and succinic anhydride. Seryl residues not occurring in His-Xaa-Ser/Ser-Xaa-His triads showed no reactivity towards NHS-biotin under reaction conditions described. Results of histidine replacement studies and studies of the pH dependence of O-biotinylation indicated that the increased nucleophilicity of the seryl hydroxyl group was due to intramolecular interaction between the seryl and histidyl residues. Our findings provide strong evidence that such triads represent novel consensus motifs in peptides.

Characterization of neurokinin binding sites in rat brain membranes using highly selective ligands.

Neurokinin binding sites are distributed throughout the central and peripheral nervous system and three neurokinin binding sites have been described until now. The endogenous tachykinins substance P, neurokinin A and eledoisin as well as the highly selective neurokinin ligands [Arg6, Sar9, Met(O2)11]SP6-11 and [Sar9, Met(O2)11]SP (for neurokinin-1), MEN 10,376 and [beta Ala8]NKA(4-10) (for neurokinin-2) and senktide and [MePhe7]NKB (for neurokinin-3) were used for displacement experiments. Neurokinin-1 and -3 binding sites were demonstrated in membrane preparations of rat striatum, frontal cortex, hypothalamus, hippocampus and amygdala by displacing [125I]-BH-substance P and [125I]-BH-eledoisin, respectively. The highly selective neurokinin-2 ligand MEN 10,376 was iodinated to measure neurokinin-2 binding sites, but no specific binding was found in membranes of all brain regions, the spinal cord, the stomach, the urinary bladder or the guinea-pig lung, probably due to loss of binding properties. We conclude that neurokinin-1 and neurokinin-3 binding sites are distributed in several brain regions of the rat brain and selective neurokinin ligands are important tools to characterize neurokinin binding sites.

Possible existence of a new tachykinin receptor subtype in the guinea pig ileum.

The guinea pig ileum possesses NK-1 and NK-3 tachykinin receptors. As expected, [Pro9]SP and senktide, which are selective agonists of NK-1 and NK-3 receptors, respectively, were found to be highly potent in contracting the guinea pig ileum. Surprisingly, similar observations were made with septide, SP-O-CH3, [Apa9-10]SP, or [Pro9,10]SP although, in contrast to [Pro9]SP, these four peptides showed a low affinity for 3H-[Pro9]SP-specific NK-1 binding sites on membranes from the guinea pig ileum. They were also devoid of affinity for NK-2 and NK-3 binding sites. GR 71251, a compound which has been described as a NK-1 antagonist, was more potent in inhibiting the septide- than the [Pro9]SP-evoked contracting response. Altogether, these results suggest that septide, [Apa9-10]SP, and [Pro9,10]SP exert their high contracting activity in the guinea pig ileum by acting on a new subtype of tachykinin receptors.

Ontogeny and characterization of [125I]Bolton Hunter-eledoisin binding sites in rat spinal cord by quantitative autoradiography.

The distribution and characteristics of [125I]Bolton Hunter-eledoisin binding sites in rat lumbar spinal cord were studied during postnatal development by in vitro receptor autoradiography. At three, six and 10 days of age, specific [125I]eledoisin binding was distributed throughout the dorsal and ventral horns of the spinal cord. In contrast, from day 24 onwards, specific binding of [125I]eledoisin was confined to superficial layers of the dorsal horn, with negligible amounts of specific binding in the ventral horn. [125I]Eledoisin binding to neonatal (three day) and adult (eight to 12 weeks) spinal cord sections was characterized using tachykinin agonists. In both dorsal and ventral horns of neonatal spinal cord, the rank order of potency of agonists indicated that the majority (64%) of specific [125I]eledoisin binding was to neurokinin-3 binding sites. The identity of the non-neurokinin-3 sites labelled by [125I]eledoisin remains to be determined. In adult rat spinal cord, [125I]eledoisin appeared to bind exclusively to neurokinin-3 binding sites. These results suggest that major changes take place in the localization of neurokinin-3 receptors during postnatal ontogeny of the rat spinal cord. These changes may reflect an important role for tachykinins in neuronal plasticity of the developing spinal cord.

Radioiodinated substance P, neurokinin A, and eledoisin bind predominantly in NK1 receptors in guinea pig lung.

In homogenates of guinea pig lung, binding of 125I-Bolton-Hunter-labeled substance P (BHSP), Bolton-Hunter-labeled eledoisin (BHELE), and [125I]iodohistidyl neurokinin A (INKA) was investigated. Equilibrium dissociation constants (derived from "cold" saturation experiments) for BHSP, INKA, and BHELE were 0.96 +/- 0.15, 1.61 +/- 0.26, and 1.98 +/- 0.12 nM, respectively. Specific binding of all three radioligands was increased 2-3-fold by 10 microM phosphoramidon. The rank order of potency of unlabeled tachykinins in competing against BHSP was substance P (SP) greater than [Sar9,Met(O2)11]-SP greater than SP methyl ester greater than neuropeptide gamma greater than neurokinin A greater than or equal to neurokinin B = kassinin greater than or equal to eledoisin greater than or equal to scyliorhinin II much greater than neuropeptide K, indicating binding to sites with the general characteristics of NK1 receptors. Similar rank potency orders were observed for INKA and BHELE, showing binding to NK1 sites, rather than to NK2 or NK3 sites, which are labeled with high affinity by these radioligands in other tissues. For all radioligands, competition curves for SP and the NK1-selective agonist [Sar9,Met(O2)11]-SP could be resolved into two components, representing high and low affinity binding sites. These were present in the approximate ratios 2:3 (for BHSP), 1:1 (for INKA), and 8:1 (for BHELE). Other agonist competition curves also yielded high and low affinity components. The data suggest that BHSP and INKA bind partly and BHELE predominantly to high affinity NK1 receptors. The nature of the low affinity site(s) could be another tachykinin receptor or a low affinity state of the NK1 receptor. Binding to a "classical" NK2 receptor is unlikely, because selective NK2 receptor antagonists and analogs were very weak competitors. Our data suggest that, in addition to the NK1 receptor, another type of tachykinin receptor may exist in this tissue. The inability to detect NK2 binding sites is strikingly at variance with functional studies.

The presence of NK3 tachykinin receptors on rat uterus.

The NK3 agonist, senktide, induced a potent contraction of rat uterus in the presence of tetrodotoxin, atropine and indomethacin, or the tachykinin receptor antagonists L-659877 and [D-Pro4,D-Trp7,9,10]substance P (4-11). Additional contractile and radioligand binding studies with receptor selective agonists and antagonists confirmed the presence of NK3 receptors and also revealed the presence of NK1 and NK2 receptors. The rat uterus is the second peripheral tissue in which a post-synaptic, non-neuronal NK3 receptor has been identified.

Control of vascular permeability and vascular smooth muscle in the respiratory tract by multiple neuropeptides.

Recent evidence suggests that a population of primary afferent neurons which terminate in the upper and lower airways contains several coexisting peptides. Thus, substance P was found to coexist with at least three structurally related tachykinins, viz. neurokinin A, neuropeptide K and an eledoisin-like peptide. Furthermore, calcitonin gene-related peptide which is structurally not related to tachykinins is also present in the same neurons which innervate a variety of peripheral tissues. Substance P, neurokinin A, eledoisin-like peptide and calcitonin gene-related peptide can simultaneously be released from central and peripheral branches of primary sensory neurons, whereas a release of neuropeptide K could not be demonstrated. Local inflammatory mediators like bradykinin or histamine also released multiple peptides in the airways and caused vascular protein leakage which is partly dependent on intact sensory nerves. In experimental animals, tachykinins as well as calcitonin gene-related peptide caused vasodilatation, whereas tachykinins but not calcitonin gene-related peptide caused an increase in vascular protein leakage. In humans, intranasally administered substance P (but not calcitonin gene-related peptide) lead to nasal obstruction as measured by rhinomanometry, and to secretion. It is suggested that activation of peripheral endings of perivascular primary sensory neurons causes vasodilatation and increase in vascular protein leakage by simultaneous action of several neuropeptides including tachykinins and calcitonin gene-related peptide.

Tachykinin binding sites in the interpeduncular nucleus of the rat: normal distribution, postnatal development and the effects of lesions.

Tachykinin binding sites in the basal midbrain were labeled in adult and neonatal rats using 125I-Bolton Hunter (BH) substance P (SP) and 125I-BH eledoisin as ligands. In the adult, binding was very low in the tegmentum and raphe adjacent to the interpeduncular nucleus (IPN). Within the IPN, no binding with either ligand was seen in the target subnuclei of the habenular SP and substance K projections, the lateral subnuclei and the cap of the rostral subnucleus. Labeling with 125I-BH-SP was very light and was restricted primarily to the central subnucleus of the IPN while 125I-BH-eledoisin labeling was very dense over the dorsal, the ventral sector of the rostral, the intermediate and the central subnuclei. Lesions of major afferents to the IPN, the fasciculus retroflexus or the locus coeruleus, had no effect on the distribution or density of the binding of either ligand. In rats 0, 4 or 7 days or age, 125I-BH-SP binding was very dense in the ventral tegmental region, the raphe and in the dorsal, rostral and central subnuclei. 125I-BH-eledoisin binding was extremely dense in the raphe and in the dorsal, rostral, intermediate and central subnuclei but was less dense in the ventral tegmentum. Adult levels of binding in the midbrain were established by 11 days of age. Neonatal lesions restricted to the fasciculus retroflexus had no effect on the density of labeling with either ligand in animals allowed to reach adulthood.

Pharmacological analysis of 125I-Bolton and Hunter labelled eledoisin binding sites in rat spinal cord by quantitative autoradiography.

Using the technique of quantitative autoradiography it has been possible to investigate and compare the pharmacological characteristics of 125I-Bolton-Hunter conjugated eledoisin (125I-BHE) binding sites in rat spinal cord with those in the rat cortex. 125I-BHE specific binding sites were discretely localised in the outer layers of the rat spinal cord. The rank order of affinity of unlabelled tachykinins in competing for 125I-BHE specific binding sites in rat spinal cord (NKB greater than ELE greater than NKA greater than SP) was identical to that found in the rat cortex suggesting the presence of 'NK-3 like' receptors in rat dorsal horn.

Comparison of the bronchoconstrictor and cardiovascular effects of some tachykinins in guinea pigs.

The effects of three tachykinins [substance P (SP), physalaemin (PH), and eledoisin-related peptide (ERP)] were investigated in anesthetized paralyzed guinea pigs. We measured airway resistance (R) and dynamic thoracic elastance (E) with a computerized technique, and blood pressure via a carotid artery. Tachykinins injected iv or intra-aortically (ia) induced dose-dependent increases in R and E, 4 times greater on iv than ia injection. They did not give rise to tachyphylaxis. As a bronchoconstrictor, PH was 5.0X and ERP 1.8X more potent than SP; time to peak response was longer for PH than for ERP and SP; and hypotensive responses, which were of similar magnitude for all three substances, lasted longest after PH. Bronchoconstrictor responses were unaltered by bilateral vagotomy, atropine, mecamylamine, and mepyramine. Morphine reduced PH-induced increases in R (P less than 0.01) and E (P less than 0.05), which were not reversed by naloxone, and capsaicin treatment 1 week before the experiments reduced both SP- and PH-induced increases in E (P less than 0.05). [D-Pro2,D-Trp7,9]-SP reduced ERP-induced increases in R and E, and [D-Arg1,D-Pro2,D-Trp7,9,Leu11]-SP reduced both SP- and PH-induced increases in R and E. We conclude that PH is the most potent bronchoconstrictor of the tachykinins tested. Tachykinin-induced bronchospasm is 'non-reflex' arising via a direct effect on airway smooth muscle; the release of histamine, acetylcholine, or other tachykinins is not involved in the responses. [D-Pro2,D-Trp7,9]-SP is more effective at SP-E receptors, and [D-Arg1,D-Pro2,D-Trp7,9,Leu11]-SP at SP-P receptors; both types of receptor are located all along the airways.

Sulfhydryl reagents have different effects on substance P and neurokinin B binding sites on cortical synaptosomes in the rat.

The effects of several protein modifying reagents, including phenoxybenzamine, N-ethylmaleimide (NEM), 5,5'-dithiobis (2-nitro)--benzoic acid (DTNB), p-chloromercuryphenyl sulfonic acid (PCMP) and p-bromophenacylbromide (PBPB), on the binding of 125I-Bolton Hunter substance P (125I-BHSP), 125I-Bolton Hunter eledoisin (125I-BHELE) and 3H-neurokinin B (3H-NKB)4 to cortical synaptosomes were examined. PCMP (10(-4) M), DTNB (10(-4) M) and NEM (10(-3) M) were without effect on the 125I-BHSP binding but reduced markedly the specific binding of 125I-BHELE or 3H-NKB. Although phenoxybenzamine and PBPB inhibited the 125I-BHSP binding when used in high concentrations (10(-4) M and 10(-3) M), they were more potent in inhibiting the 125I-BHELE or 3H-NKB binding. These results indicate that the NKB binding site is more sensitive to alkylating reagents than the SP binding sites and that these reagents can be used to distinguish SP and NKB receptors in the brain.

[3H]neurokinin B and 125I-Bolton Hunter eledoisin label identical tachykinin binding sites in the rat brain.

[3H]Neurokinin B ([3H]NKB) of high specific activity (75 Ci/mmol) was synthesized for study of its binding to crude synaptosomes from the rat cerebral cortex. The specific binding of [3H]NKB (75% of total binding) was temperature dependent, saturable, and reversible. Scatchard analyses and Hill plots showed the existence of a single population of noninteracting binding sites (KD = 4.3 nM; Bmax = 123 fmol/mg of protein). Competition studies indicated the following rank order of potencies among tachykinins: NKB greater than eledoisin (E) greater than kassinin greater than physalaemin greater than neurokinin A (NKA) greater than substance P (SP), a result suggesting that NKB might be the endogenous ligand for [3H]NKB binding sites. It is of interest that 127I-Bolton Hunter (BH) NKA (127I-BHNKA) was much more potent than NKA in inhibiting the specific binding of [3H]NKB, which raises certain questions concerning the use of 125I-BHNKA as a ligand for NKA binding sites in the brain. These results, as well as those obtained with different SP analogues, show a close similarity to those obtained previously with 125I-BHE binding to cortical synaptosomes. This suggested that the two ligands labeled identical binding sites. In addition, using either [3H]NKB or 125I-BHE as ligands, similar displacement curves were obtained with increasing concentrations of NKB and 127I-BHE. The similarity of the [3H]NKB and 125I-BHE binding sites was further confirmed by comparison of their localization on rat brain sections by autoradiography. The distribution of binding sites for [3H]NKB and 125I-BHE was identical throughout the brain, and the highest density of binding sites for the two ligands was found in layers IV and V of the cerebral cortex, the paraventricular nucleus of the hypothalamus (magnocellular part), and the ventral tegmental area.

Quantitative autoradiographic analysis of the distribution of binding sites for [125I]Bolton Hunter derivatives of eledoisin and substance P in the rat brain.

[125I]Bolton and Hunter eledoisin binds to a single class of non-interacting sites in rat cerebral cortex tissue sections with an apparent Kd of 9.9 nM and a Bmax of 244 fmol/mg protein. When concentrations of up to 23 nM [125I]Bolton and Hunter eledoisin were used, [125I]Bolton and Hunter eledoisin binding was specific, saturable and reversible. Kassinin, eledoisin and neurokinin B were more potent than substance P and neurokinin A in inhibiting the specific binding of [125I]Bolton and Hunter eledoisin to cerebral cortex tissue sections. These kinetic and pharmacological characteristics are consistent with results obtained from binding studies on cortical synaptosomes. When the localization of [125I]Bolton and Hunter substance P and [125I]Bolton and Hunter eledoisin binding sites were compared, differences in many areas of the brain were noted. Large differences were seen in the paraventricular and supraoptic hypothalamic nuclei, and in layers IV and V of the cerebral cortex, which were densely labeled by [125I]Bolton and Hunter eledoisin, but not by [125I]Bolton and Hunter substance P. In contrast, nuclei of the septum (diagonal band of Broca, septohippocampal nucleus, dorsal part of the lateral septal nucleus), the rostrodorsal part of the hippocampus and other discrete nuclei [endopyriform nucleus, anterior cortical amygdaloid nucleus, the vermis columns (9-10), the dorsal tegmental nucleus, the hypoglossal and ambiguus nucleus] had high levels of [125I]Bolton and Hunter substance P binding but were only labeled weakly by [125I]Bolton and Hunter eledoisin. Thus, the two ligands seem to label different sites, since these binding sites have different biochemical and pharmacological properties, and are localized in different anatomical structures.

Response of splanchnic-driven neurons to substance P and eledoisin-related peptide.

The effects of iontophoretically applied substance P (SP) and eledoisin-related peptide (ERP) on the extracellularly recorded activity of spontaneous or D,L-homocysteic acid (DLH)-driven thoracic (T7-T8) spinal neurons, identified by orthodromic activation by the greater splanchnic nerve, were examined in chloralose/urethane anesthetized cats. Of 24 neurons tested, SP and/or ERP (18-120 nA) produced a weak (ca 30%) excitatory effect on 5 neurons, a weak inhibitory effect on 9 neurons, and no effect on the remaining neurons. When compared to the action of either DLH or GABA, we found both the excitatory and inhibitory effects of SP and/or ERP were delayed and prolonged. These data do not support a role for SP as a major neurotransmitter in spinal circuits in which visceral (splanchnic) afferents participate.

Characterization and autoradiographic localization of multiple tachykinin binding sites in gastrointestinal tract and bladder.

Binding sites for the [125I]Bolton-Hunter-labeled tachykinins substance K (BHSK), eledoisin (BHE) and substance P (BHSP) were investigated using crude membrane suspensions and autoradiography. In smooth muscle membranes from guinea-pig small intestine and rat duodenum, specific binding of BHSK was saturable and reversible, showing a single class of sites with a KD of 1 to 3 nM and maximum number of specific binding sites of 1 to 2 fmol/mg of wet weight tissue. Pharmacological characterization of this binding revealed a novel receptor site (K) with affinity for substance K greater than kassinin greater than or equal to eledoisin greater than neuromedin K greater than substance P greater than physalaemin. Inhibition of the binding of BHSK in membranes from mouse urinary bladder exhibited a similar K-type pattern. In rat duodenum and mouse bladder membranes, the binding of BHE was inhibited by substance K greater than kassinin greater than eledoisin greater than neuromedin K greater than substance P greater than physalaemin indicating the same receptor site as for BHSK. On the other hand, in rat cerebral cortex membranes BHE binding was inhibited by neuromedin K = kassinin = eledoisin greater than physalaemin greater than substance K greater than substance P indicating a definitive tachykinin E receptor site. The same displacement pattern of BHE binding was also detected in longitudinal muscle membranes from the guinea-pig small intestine. In mouse bladder membranes and in rat and guinea-pig intestine, the binding of BHSP was inhibited by substance P greater than physalaemin greater than substance K greater than or equal to eledoisin = kassinin greater than neuromedin K indicating a definitive tachykinin P receptor site. Autoradiographic binding sites for both BHSK and BHSP were seen in circular muscle of the rat stomach, small intestine and colon and in circular and longitudinal muscle of the guinea-pig small intestine and colon. Binding sites for BHSK, but not for BHSP, were seen in the muscularis mucosae of the gastric fundus and colon of the rat. Binding sites for BHSP, but not for BHSK, were seen in mucosa of guinea-pig colon and were densely clustered over ganglia of the myenteric and submucous plexuses in rat and guinea-pig colon. The guinea-pig intestine probably contains all three types of tachykinin binding sites whereas rat duodenum and mouse bladder contain only K and P sites. Some tissues classified previously as SP-P or SP-E may actually contain P and/or K sites.(ABSTRACT TRUNCATED AT 400 WORDS)

Effects of tachykinins on gastric acid and pepsin secretion and on gastric outflow in the Atlantic cod, Gadus morhua.

Gastric acid and pepsin responses to substance P, physalaemin, eledoisin, and an eledoisin-related peptide, [Lys6]eledoisin-(6-11), were measured in gastrically and intestinally perfused cods. The intestinal perfusion maintains water balance and inhibits drinking. During basal conditions acid secretion was stimulated (approximately equal to 25%) by low doses (less than 0.13 nmol X kg-1 X h-1) of physalaemin and eledoisin. High doses (greater than 16 nmol X kg-1 X h-1) were inhibitory. Median and very high doses of substance P and eledoisin-related peptide, respectively, tended to stimulate acid secretion. All tachykinins were extremely efficacious pepsigogues. Physalaemin and eledoisin were the most potent (D50 approximately 10(-10) mol X kg-1 X h-1) but produced fading and submaximal responses at high doses. The fading persisted despite endogenous acidification produced by histamine stimulation. Relative to physalaemin, the potencies of substance P and eledoisin-related peptide were 0.04 and 0.001. The results suggest that some tachykinin may be a physiological stimulator of pepsin secretion and that the effect on acid secretion results from activation of both stimulatory and inhibitory pathways. The inhibitory component probably includes a cholinergic link. Gastric volume outflow increased during infusion of physalaemin, eledoisin, and (slightly) substance P. The response, which was not related to acid secretory rate (and conceivably not to volume secretion), suggests that a tachykinin may be involved also in the regulation of drinking.

Influence of the amino acids of substance P in the recognition of its receptor: affinities of synthesized SP analogues for the specific 125I-BHSP binding site on rat brain synaptosomes.

Substance P analogues have been synthesized, by solid-phase methodology, in order to get a better knowledge of the structural requirements for the 125I-BHSP binding on rat brain synaptosomes. Assuming that the core of SP exists in an alpha-helicoidal structure three major points should be underlined: the SP receptor recognizes probably the side of the helix bearing the two side chains of Phe and Phe; the arginine guanidinium interacts with either a carboxylate or a phosphate function of the binding site; the C-terminal tripeptide undergoes a conformational change allowing the interactions of the C-terminal amide with a carboxylate and that of the sulfur atom with an electrophile of the binding site. The specificity of these peptides have been further estimated by comparing their binding potencies to those observed for the 125I-BHE specific binding on rat cortical synaptosomes and their bioactivities on guinea-pig ileum.

Pharmacological characterisation of two tachykinin binding sites in the rat cerebral cortex.

The pharmacological properties of two types of tachykinin receptor were characterised on rat cortical synaptosomes using 125I-Bolton Hunter substance P (125I-BHSP) or with 125I-Bolton Hunter eledoisin (125I-BHE). Shorter SP C-terminal fragments, such as SP (6-11) or (pGlu)-SP (6-11), were more potent than SP itself or longer SP C-terminal fragments in competing for 125I-BHE binding; their efficacy was comparable to that of eledoisin. In contrast, longer SP C-terminal fragments exhibited a higher affinity than shorter ones for the 125I-BHSP binding sites as previously reported. SP N-terminal fragments were devoid of activity on either type of binding sites. SP methyl ester inhibited 125I-BHSP binding but was without effect on 125I-BHE binding whilst, DiMe-C7, a metabolically stable tachykinin analog, had the opposite selectivity. Eledoisin related peptide (ERP) was less effective than either SP or eledoisin on 125I-BHSP and 125I-BHE binding sites respectively. Finally, the undecapeptide or octapeptide SP antagonists, which are weak inhibitors of 125I-BHSP binding, had negligable activity on 125I-BHE binding sites.

Effects of glutamate, substance P and eledoisin-related peptide on solitary tract neurones involved in respiration and respiratory reflexes.

Recent studies have implicated glutamate and substance P in synaptic transmission in the nuclei tractus solitarii and in central regulation of cardiorespiratory functions. Consequently, in chloralose-anaesthetized cats that were artificially ventilated, we examined the effects of the microiontophoretic application of both chemicals (and the substance P homologue, eledoisin-related peptide) on single neurones of the nuclei tractus solitarii implicated in the control of respiration and respiratory tract reflexes. These neurones were functionally identified as either respiratory neurones or presumed reflex interneurones, and showed functional properties comparable to those previously documented for each of these two types. The iontophoretic application of glutamate produced an excitation of rapid onset in 23 or 25 reflex interneurones tested, but the respiratory neurones showed a differential sensitivity: one type (n = 32) was "glutamate-sensitive" and showed rapid excitation with glutamate applications of less than 30 nA, the other type of respiratory neurone (n = 26) was termed "glutamate-insensitive" since it either showed excitation only with applications of 60 nA or more or showed no response even with currents up to 94 nA. Each neurone studied was clearly of one type or the other. Glutamate could increase the number of spikes per rhythmic burst and the burst duration of respiratory neurones, it facilitated evoked activity in the reflex interneurones and in those respiratory neurones having a superior laryngeal nerve or vagus nerve afferent input, and the magnitude of the excitatory responses to glutamate varied directly with the amount of ejecting current. Substance P and eledoisin-related peptide also had excitatory effects on respiratory neurones and reflex interneurones, but compared with glutamate-induced effects the excitation was slower in onset and more prolonged in after-discharge. Both rhythmic and evoked activity could be facilitated, and the magnitude of the effect varied directly with the magnitude of the ejecting current. In showing that both glutamate and substance P (and its analogue, eledoisin-related peptide) have excitatory effects on the activity of respiratory neurones and reflex interneurones, this study provides evidence suggesting that these neurones have receptors for these neural chemicals, supportive of a role for each chemical in the regulation of respiration and respiratory tract reflexes.