The possibility was explored that the recently defined 5-HT1D binding sites could be negatively coupled to adenylate cyclase in calf substantia nigra. 2) 5-HT inhibited forskolin-stimulated adenylate cyclase activity in a concentration-dependent manner (EC50 value = 24.0 nmol/l, Emax = 22.7% inhibition) in the presence of GTP (10 mumol/l), which was required for this inhibitory effect. 3) The following 5-HT receptor agonists inhibited adenylate cyclase activity (in decreasing order of potency): 5-carboxamidotryptamine greater than 5-HT greater than 5-methoxytryptamine greater than 5-methoxy-3-(1,2,3,6-tetrahydro-4-pyridinyl)-1H indole (RU 24969) greater than or equal to N,N-dipropyl-5-carboxamidotryptamine greater than 8-hydroxy-2(di-n-propylamino)-tetralin (8-OH-DPAT) greater than buspirone greater than ipsapirone; the latter two compounds apparently behaved as partial agonists. 4) Other compounds displaying agonist activity in this system were: metergoline greater than methysergide greater than or equal to rauwolscine greater than or equal to cyanopindolol greater than or equal to yohimbine greater than (+/-)-4(3-tert-butyl-amino-2-hydroxypropoxy)-indol-2 carbonic acid isopropylester (21-009) greater than corynanthine. 5) Methiothepin, mianserin and spiperone displaced the concentration-effect curve of 5-HT to the right without depressing the Emax value. The same held true for the partial agonists ipsapirone, buspirone and corynanthine. 6) The rank order of potency of agonists as well as of antagonists in this system was in full agreement with their affinities at 5-HT1D binding site. A highly significant correlation was found between both parameters (r = 0.94, P = 0.0001).(ABSTRACT TRUNCATED AT 250 WORDS)