Alpha2-Adrenoceptor-mediated inhibition of [3H]noradrenaline release and alpha2-adrenoceptor-mediated inhibition of voltage-activated Ca2+ currents were compared in cultured thoracolumbar postganglionic sympathetic neurons from newborn wildtype (WT) mice and mice in which the alpha2A/D-adrenoceptor gene had been disrupted (alpha2A/DKO). In cultures prepared from WT mice and preincubated with [3H]noradrenaline, the alpha2-adrenoceptor agonist 5-bromo-6-(2-imidazolidinylidenamino)quinoxaline (UK 14,304) reduced the (autoinhibition-free) release of [3H]noradrenaline elicited by single electrical pulses or trains of 8 pulses at 100 Hz. The maximal inhibition by UK 14,304 amounted to 70%-85%. Its concentration-response curve was shifted to the right by phentolamine (0.3 microM) and, to a smaller extent, rauwolscine (0.3 microM). Pretreatment of the cultures with pertussis toxin abolished the effect of UK 14,304. Phentolamine and rauwolscine increased the (alpha2-autoinhibited) release of [3H]noradrenaline elicited by 18, 36 or 72 pulses at 3 Hz. In cultures from alpha2A/DKO mice, UK 14,304 failed to reduce the release of [3H]noradrenaline elicited by single pulses and phentolamine and rauwolscine failed to increase the release of [3H]noradrenaline elicited by 18-72 pulses at 3 Hz. In neurons from WT mice examined with the amphotericin B-perforated configuration of the patch clamp method, UK 14,304 reduced depolarisation-evoked Ca2+ currents. The inhibition was voltage-dependent as shown by a decline at strong depolarisation during ramp-like voltage commands and by an attenuation briefly after a conditioning depolarising pulse. The maximal inhibition by UK 14,304 was 39%. Its concentration-response curve was shifted to the right by phentolamine (0.3 microM) but not significantly changed by rauwolscine (0.3 microM) and prazosin (1 microM). Pretreatment with pertussis toxin abolished the effect of UK 14,304. In neurons from alpha2A/DKO mice, UK 14,304 also reduced depolarisation-evoked Ca2+ currents, but with a smaller maximal effect, namely 18% inhibition. Its concentration-response curve was shifted to the right by rauwolscine (0.3 microM) and prazosin (1 microM) but not significantly changed by phentolamine (0.3 microM). Pretreatment with pertussis toxin abolished the effect of UK 14,304 also in cultures from alpha2A/DKO mice. It is concluded that the only presynaptic alpha2-autoreceptors that detectably depress transmitter release from cultured thoracolumbar sympathetic neurons taken from newborn mice are alpha2A/D. In contrast, the soma-dendritic alpha2-autoreceptors that inhibit voltage-gated Ca2+ channels are both alpha2A/D and non-alpha2A/D (i.e. alpha2B or alpha2c). Both presynaptic alpha2A/D- and soma-dendritic alpha2A/D- and non-alpha2A/D-autoreceptors operate through pertussis toxin-sensitive G proteins in these neurons.