3,4-methylenedioxymethamphetamine (ecstasy)-induced learning and memory impairments depend on the age of exposure during early development

Abstract

Use of 3,4-methylenedioxymethamphetamine (MDMA; ecstasy) has increased dramatically in recent years, yet little is known about its effects on the developing brain. Neonatal rats were administered MDMA on days 1-10 or 11-20 (analogous to early and late human third trimester brain development). MDMA exposure had no effect on survival but did affect body weight gain during treatment. After treatment, body weight largely recovered to 90-95% of controls. MDMA exposure on days 11-20 resulted in dose-related impairments of sequential learning and spatial learning and memory, whereas neonatal rats exposed on days 1-10 showed almost no effects. At neither stage of exposure did MDMA-treated offspring show effects on swimming ability or cued learning. Brain region-specific dopamine, serotonin, and norepinephrine changes were small and were not correlated to learning changes. These findings suggest that MDMA may pose a previously unrecognized risk to the developing brain by inducing long-term deleterious effects on learning and memory.

Fig. 1.

Latency (seconds) to escape from a straight swimming channel averaged across four trials (mean ± SEM). A one-between, three-within factor ANOVA (age by treatment group by gender by trial) showed no effect of treatment, age, or gender and no interaction between treatment and other factors.

Fig. 2.

Errors (a) and latency (b) (seconds) to escape from the multiple-T (Cincinnati) water maze (mean ± SEM averaged across days, trials, and gender). Overall ANOVAs with age as a factor showed that age was part of a significant interaction with treatment; therefore, follow-up ANOVAs within each age group were performed. For errors, the P1–P10 ANOVA showed no treatment effect and one interaction (treatment by day by trial; F_(9,126) = 1.99;p < 0.05). The P11–P20 ANOVA showed a treatment effect (F_(3,42) = 3.04;p < 0.05) and interactions between treatment by day (F_(9,126) = 2.68;p < 0.01), treatment by day by gender (F_(9,126) = 1.98; p< 0.05), and treatment by trial by gender (F_(3,42) = 3.04; p< 0.05). These interactions all reflected MDMA-induced increases in errors, but the effects were larger on certain days and trials and in females. For simplicity of presentation, the main effect of treatment is shown, because it captures the essence of the principal effects seen in the MDMA groups. For latency, the P1–P10 ANOVA showed a treatment main effect (F_(3,42) = 3.17;p < 0.05) and a treatment by day interaction (F_(9,126) = 2.76; p< 0.01). These effects represented differences among the MDMA groups (data not shown). The P11–P20 ANOVA for latency showed a treatment effect (F_(3,42) = 3.89;p < 0.02) and no interactions. *p < 0.05, **p < 0.01, †p < 0.10 compared with saline controls.

Fig. 3.

Latency (seconds) to escape from the Morris water maze during cued learning trials (4 trials per day averaged across 5 d; mean ± SEM). A one-between, four-within ANOVA (age by treatment group by gender by day by trial) showed no main effect of treatment, age, or gender and no treatment interactions with day or trial. Data are shown for males and females combined.

Fig. 4.

Morris water maze spatial learning results in the P11–P20 treatment groups (mean ± SEM) averaged across four trials per day and 5 d. a–c, Cumulative distance from the hidden platform during acquisition (a), reversal (b), and reduced platform (c) trials. d–f, Path length to find the platform during the same three phases of testing.g–i, Latency (seconds) to find the platform during the three phases of testing. No treatment effects were found in the Morris water maze in the P1–P10 treatment groups (data not shown). Treatment effects were obtained on measures of cumulative distance during acquisition trials (a) (F_(3,42) = 4.88; p< 0.01) and reduced platform trials (c) (F_(3,42) = 8.74, p< 0.0001), but not on reversal. A similar pattern was obtained for path-length analyses (acquisition:F_(3,42) = 4.70; p< 0.01; reduced: F_(3,42) = 7.55;p < 0.001; reversal was not significant). For latency, treatment effects were obtained on all three phases (acquisition: F_(3,42) = 3.31;p < 0.05; reversal:F_(3,42) = 3.05; p< 0.05; reduced: F_(3,42) = 7.90;p < 0.001). Data for males and females are combined for presentation. Several interactions between treatment and day and gender were also obtained. These indicated that the effects of MDMA were largest in females and on early and middle days and smallest on the last day of testing. *p < 0.05, **p < 0.01, †p < 0.10 compared with saline control.

Fig. 5.

Morris water maze results of memory (probe) trials in the P11–P20 treatment groups (mean ± SEM) averaged across trials; one probe trial was administered on days 2, 4, and 6. There were no treatment main effects among the P1–P10 treatment groups (data not shown). For the P11–P20 treatment groups, treatment main effects were found on average distance from the target during the probe trials for acquisition (F_(3,42) = 6.17;p < 0.002), reversal (F_(3,42) = 5.88; p< 0.002), and reduced (F_(3,42) = 5.58;p < 0.01). Data for males and females were combined for presentation. No interactions with treatment were obtained. *p < 0.05, **p < 0.01 compared with saline controls. Similar patterns were found for the P11–P20 treatment groups on platform site crossings and for the percentage of time spent in the target quadrant (data not shown).

Fig. 6.

Cumulative distance from the platform learning curves for the P11–P20 groups for each day of testing during the acquisition phase of Morris water maze testing (mean ± SEM) averaged across genders. Inset, Additional details for each trial for day 1 of testing. There were no significant group differences on day 1 and no differences on trial 1 of day 1, demonstrating that MDMA animals showed no pre-existing performance differences before finding the hidden platform for the first time.