The rates of accumulation of newly synthesized catecholamines and endogenous catecholamine levels in mice were determined after treatment with cycloheximide, acetoxycycloheximide, puromycin, and anisomycin. The rates of accumulation were found to be decreased by all antibiotics tested, weakening the assumption that their amnestic effects are due solely to inhibition of protein synthesis. The hypothesis that protein synthesis is essential for longterm memory has developed from the observation in several species that treatment with an inhibitor of protein synthesis before or shortly after training leads to amnesia of the training experience (1-6). There is evidence, however, that important side-effects on the central adrenergic system may be common to all inhibitors of protein synthesis and that these side-effects may contribute to the amnesia. For example, in mice and rats the amnesia caused by puromycin, acetoxycycloheximide (AXM), and cycloheximide (CXM) is attenuated by adrenergic stimulants or monoamine oxidase inhibitors (7-9). Also consistent with this possibility are the reports that AXM, CXM, and anisomycin (Ani) cause a reduction of tyrosine hydroxylase activity (10,11). This further suggests that the amnestic effects of the antibiotics may depend in part on a reduction of the functional pools of newly synthesized catecholamines (CAs) (12, 13). We found support for involvement of these functional pools by observing that a large amnestic dose of AXM strikingly reduced the rate of accumulation of CAs from circulating tyrosine for at least 12 hr, greatly exceeding in degree and duration what might be anticipated from the loss of tyrosine hydroxylase activity in Vitro. We also found that the antibiotic, in spite of its inhibitory effect on the rate of accumulation of newly synthesized CAs, increased the endogenous levels of CAs for 4 hr after treatment, indicating that it interferes with aspects of CA metabolism in addition to biosynthesis (14).The present experiments extend this approach to CXM, Ani, puromycin, and to the smallest dose of intracerebrally injected AXM found to be amnestic in mice (15). We find all of these antibiotics to have substantial effects on central CAs. In some ways, these effects resemble those produced by inhibitors of tyrosine hydroxylase and dopamine f3-hydroxylase, agents reported to have the same amnestic properties as those of CXM in active or passive avoidance (16, 17) and food-motivated discrimination tasks (18).Male Swiss-Webster and ICR mice (28-32 g) were housed as previously described (14). CXM (120 mg/kg) and Ani (25-30 mg/kg; gift of Pfizer Inc.) were injected subcutaneously in 0.2 ml of water. AXM (10,ug/injection site; gift of the National Cancer Institute) or puromycin (90 gg/injection site; neutralized with NaOH) were injected bitemporally in 12 ,ul of water under light Evipal (150 mg/kg) anesthesia. Controls were injected in the same way with 12 gl of water per injection site. The inhibitor of tyrosine hydroxylase, a-methyl-para-tyrosine (methyl ester-HCl...