1. The block by ifenprodil of voltage-activated Ca2+ channels was investigated in intracellular free calcium concentration ([Ca2+]i) evoked by 50 mM K+ (high-[K+]o) in Fura-2-loaded rat hippocampal pyramidal neurones in culture and on currents carried by Ba2+ ions (IBa) through Ca2+ channels in mouse cultured hippocampal neurones under whole-cell voltage-clamp. The effects of ifenprodil on voltage-activated Ca2+ channels were compared with its antagonist actions on N-methyl-D-aspartate- (NMDA) evoked responses in the same neuronal preparations. 2. Rises in [Ca2+]i evoked by transient exposure to high-[K+]o in our preparation of rat cultured hippocampal pyramidal neurones are mediated predominantly by Ca2+ flux through nifedipine-sensitive Ca2+ channels, with smaller contributions from nifedipine-resistant, omega-conotoxin GVIA-sensitive Ca2+ channels and Ca2+ channels sensitive to crude funnel-web spider venom (Church et al., 1994). Ifenprodil (0.1-200 microM) reversibly attenuated high-[K+]o-evoked rises in [Ca2+]i with an IC50 value of 17 +/- 3 microM, compared with an IC50 value of 0.7 +/- 0.1 microM for the reduction of rises in [Ca2+]i evoked by 20 microM NMDA. Tested in the presence of nifedipine 10 microM, ifenprodil (1-50 microM) produced a concentration-dependent reduction of the dihydropyridine-resistant high-[K+]o-evoked rise in [Ca2+]i with an IC50 value of 13 +/- 4 microM. The results suggest that ifenprodil blocks Ca2+ flux through multiple subtypes of high voltage-activated Ca2+ channels. 3. Application of the polyamine, spermine (0.25-5 mM), produced a concentration-dependent reduction of rises in [Ca2+]i evoked by high-[K+]o. The antagonist effects of ifenprodil 20 micro M on high-[K+]0-evoked rises in [Ca2+]. were attenuated by spermine 0.25 mM but not by putrescine 1 or 5 mM. In contrast,spermine 0.1 mM increased rises in [Ca2+]i evoked by NMDA and enhanced the ifenprodil (5 micro M) block of NMDA-evoked rises in [Ca2+]i.4. Similar results were obtained in mouse cultured hippocampal pyramidal neurones under whole-cell voltage-clamp. Ifenprodil attenuated both the peak and delayed whole-cell IB. with an IC% value of 18 +/- 2 micro M, whilst it attenuated steady-state NMDA-evoked currents with an IC50 of 0.8 +/- 0.2 micro M. Block of IBa by ifenprodil 10 JaM was rapid in onset, fully reversible and occurred without change in thecurrent-voltage characteristics of Ba. The ifenprodil block of IBa was enhanced on membrane depolarization and was weakly dependent on the frequency of current activation. Spermine 0.1 mM potentiated control NMDA-evoked currents but attenuated IB,. In agreement with the microspectrofluorimetric studies, co-application of spermine produced a small enhancement of the inhibitory effect of ifenprodil 10 micro M on NMDA-evoked responses whereas the reduction of I4 by ifenprodil 10 micro M in the presence of spermine was less than expected if the inhibitory effects of ifenprodil and spermine on IBa were simply additive.5. The results indicate that ifenprodil blocks high voltage-activated Ca2+ channels in rat and mouse cultured hippocampal pyramidal neurones. Although the Ca2+ channel blocking actions of ifenprodil are observed at higher concentrations than those associated with NMDA antagonist activity, Ca2+ channel blockade may contribute, at least in part, to the established neuroprotective and anticonvulsant properties of the compound.