1. NK1 and NK2 tachykinin receptors were characterized in guinea-pig and rat bronchopulmonary systems and in the vasculature of the rat by use of radioligand binding and/or functional studies. 2. The radioligands for NK1 and NK2 receptors ([3H]-SP and [3H]-pNKA, respectively) did not label tachykinin receptors in homogenates of rat lungs or bronchi. In contrast, in the guinea-pig, [3H]-SP bound with high affinity to these tissues (KD = 0.23 +/- 0.08 nM and 0.34 +/- 0.05 nM, for lungs and bronchi, respectively). The total number of binding sites was 4.6 fold greater in bronchus (Bmax = 135 +/- 27 fmol mg-1 protein) than in lung homogenates (Bmax = 29.3 +/- 0.1 fmol mg-1 protein). Furthermore, this binding was markedly displaced by CP-96,345 (pKi = 9.5 +/- 0.1) and RP 67580 (pKi = 7.6 +/- 0.1), antagonists of NK1 receptors, slightly displaced by SR 48968 (pKi = 6.6 +/- 0.1), but not affected by actinomycin D or L-659,877, antagonists of NK2 receptors. Specific binding of [3H]-pNKA, detected in guinea-pig bronchi (KD = 5.2 +/- 0.1 nM, and Bmax = 203 +/- 19 fmol mg-1 protein) but not in lungs, was similarly (40 to 53%) displaced by RP 67580 (1 microM), CP-96,345 (10 and 100 nM) or SR 48968 (10 and 100 nM). The displacement approximately doubled (87 to 91%) when SR 48968 (10 nM) was combined with either RP 67580 (1 microM) or CP-96,345 (10 nM), but not when RP 67580 was combined with CP-96,345. 3. In urethane-anaesthetized guinea-pigs, i.v. injections of the NK1 receptor agonists SP, [Pro9]-SP, [Sar9,Met(O2)11]-SP and septide, as well as the NK2 receptor agonists NKA and [Lys5,MeLeu9,NLeu10]-NKA(4-10) (0.1-10 micrograms kg-1, i.v.), dose-dependently increased lung inflation pressure. The most potent of these peptides were septide and [Lys5, MeLeu9,NLeu10]-NKA(4-10) (EC50 = 0.38 +/- 0.07 and 0.07 +/- 0.02 microgram kg-1, respectively). Interestingly, septide was 130 fold less potent than SP in displacing [3H]-SP from its binding sites in the guinea-pig lung, whereas it was 14 fold more potent than SP as a bronchoconstrictor. RP 67580 (0.3-5 mg kg-1, i.v.) and CP-96,345 (0.01-3 mg kg-1, i.v.) dose-dependently reduced the bronchoconstriction produced by the NK1 receptor agonists. Conversely, the NK2 receptor antagonists actinomycin D (1-10 mg kg-1, i.v.) and SR 48968 (0.03-0.3 mg kg-1, i.v.) inhibited specifically the responses induced by NK2 receptor agonists.4. In pentobarbitone-anaesthetized rats, the NK1 and NK2 receptor agonists (0.01-4 microg kg-1, i.v.)produced dose-dependent hypotensive responses. The order of potency was SP = [Sar9, Met(0211]-SP = [Pro9]-SP > septide = NKA >[Lys5, MeLeu9, NLeu 10-NKA.(4-10). RP 67580 (0.13-0.5 mg kg-1,i.v.) and CP-96,345 (0.5-2 mg kg-1, i.v.) antagonized in a dose-related manner (20 to 64%) the vascular effects of both NK, and NK2 receptor agonists, whereas actinomycin D (3 mg kg-1, i.v.) and SR 48968(2 mg kg-1, i.v.) did not. RP 67580 was approximately 4 times more potent than CP-96,345.5. These studies indicate that NK1 and NK2 receptors are both present in the guinea-pig bronchopulmonary system whereas only NK1 receptors are detectable in the rat vasculature under our experimental conditions. Furthermore, NK1 receptors in the guinea-pig bronchopulmonary system are pharmacologically distinct from those present in the rat vascular system, since both agonist potencies and antagonist affinities differ between the two species.