Gestational exposure to methylmercury retards choice in transition in aging rats.

Developmental exposure to methylmercury has behavioral effects that extend into adulthood and aging. In this study, methylmercury's prolonged effects on the acquisition of choice and sensitivity to changes in reinforcement rates were studied. Pregnant female rats were exposed to drinking water containing 0, 0.5, or 6.4 ppm Hg as methylmercury, resulting in about 40 and 500 microg/kg/day of mercury intake. Maternal exposure began at least 4 weeks before mating, and continued to postnatal day 16. Then all mercury exposure ended. The behavior of 1.7- and 2.3-year-old offspring was maintained under various concurrent schedules of reinforcement. Thus, one reinforcement schedule maintained left-lever responding and a separate one maintained right-lever responding. The animal could switch ("changeover") between the two levers at any time. For the first 30 min of a 3-h session, the left and right levers each produced reinforcement at the same rate and left:right response ratios were about 1:1. After 30 min, either the left lever became richer than the right; the right lever became richer than the left, or there was no change. Terminal reinforcer ratios (left:right) used were 9:1, 4:1, 3:1, 1:1, 1:3, 1:4, and 1:9. Response rates on the two levers were tracked continuously through a session. This novel procedure for examining choice, and its acquisition, in a single session, was validated through many comparisons with the extant literature. Both response rates and changeover rates were influenced by the reinforcer ratios for the 1.7-year-olds. Changeover rates were not influenced by reinforcement rate for their 2.3-year-old littermates. For the 1.7-year-olds, there was no effect of methylmercury on changeover or response rates and there was no interaction between exposure and reinforcer ratio. In controls and most methylmercury-exposed rats, response ratios (the measure of choice) approximately matched reinforcer ratios by the end of the single session. This is commonly interpreted as reflecting sensitivity to reinforcement rates. Methylmercury exposure did not affect this measure systematically. The single-session transition from baseline (response ratios about 1:1) to terminal performance was retarded in many methylmercury-exposed rats relative to controls, especially in the older rats. The 2.3-year-old control rats required about 20 to 25 reinforcers to complete one half of the 9:1 and 4:1 transitions, respectively, and exposed rats required about twice as many. Thus, prenatal methylmercury exposure specifically retarded the acquisition of choice in older rats. Methylmercury did not interfere with the final expression of choice. Moreover, two rate measures, lever-press rates and changeover rates, were not systematically affected by methylmercury. The acquisition of choice appears to be very sensitive to subtle consequences of developmental methylmercury exposure. The specific tactics greatly reduced the time required to study behavior in transition from a month in previous reports to a single session here.