We describe the characteristics of long-term potentiation (LTP) in the intact mouse. Perforant path stimulation evokes both a population excitatory postsynaptic potential (pop-EPSP) and a population spike potential (pop-spike) from the hippocampal dentate gyrus in urethane anesthetized animals. LTP, as measured by increased pop-spike amplitude and pop-EPSP slope, was successfully induced and reliably maintained at a stable level for at least 12 h, the longest time tested. The LTP-inducing stimulus (3 trains of 400 Hz, 8 0.4 ms pulses/train) used in two strains of mice was less by half than that used in rat. These parameters for inducing LTP were also successfully applied to obtain LTP in two different transgenic mouse strains: one bearing a F1/Gap-43 promoter-lacZ fusion gene and another which overexpresses the S100 beta gene. We also examined the effects of protein synthesis inhibitors, cycloheximide (CXM) and anisomycin (ANI). When CXM or ANI was given 30 min before LTP induction, there was no persistent loss of LTP at the 4 h time point. However, if CXM was given 4 h before LTP induction, significant decay of the potentiated responses occurred 90 min after induction. Half of the animals receiving CXM but not ANI showed a complete and sudden elimination of the entire response after the LTP-inducing stimulus. It was speculated that loss of a constitutively-expressed housekeeping protein, for example a calcium buffering protein, with an estimated half-life of 2 h would lead to an inability to buffer LTP-induced alterations, such as intracellular calcium elevation, increasing intracellular calcium to toxic levels.(ABSTRACT TRUNCATED AT 250 WORDS)