Comparison of blood and brain mercury levels in infant monkeys exposed to methylmercury or vaccines containing thimerosal

Abstract

Thimerosal is a preservative that has been used in manufacturing vaccines since the 1930s. Reports have indicated that infants can receive ethylmercury (in the form of thimerosal) at or above the U.S. Environmental Protection Agency guidelines for methylmercury exposure, depending on the exact vaccinations, schedule, and size of the infant. In this study we compared the systemic disposition and brain distribution of total and inorganic mercury in infant monkeys after thimerosal exposure with those exposed to MeHg. Monkeys were exposed to MeHg (via oral gavage) or vaccines containing thimerosal (via intramuscular injection) at birth and 1, 2, and 3 weeks of age. Total blood Hg levels were determined 2, 4, and 7 days after each exposure. Total and inorganic brain Hg levels were assessed 2, 4, 7, or 28 days after the last exposure. The initial and terminal half-life of Hg in blood after thimerosal exposure was 2.1 and 8.6 days, respectively, which are significantly shorter than the elimination half-life of Hg after MeHg exposure at 21.5 days. Brain concentrations of total Hg were significantly lower by approximately 3-fold for the thimerosal-exposed monkeys when compared with the MeHg infants, whereas the average brain-to-blood concentration ratio was slightly higher for the thimerosal-exposed monkeys (3.5 +/- 0.5 vs. 2.5 +/- 0.3). A higher percentage of the total Hg in the brain was in the form of inorganic Hg for the thimerosal-exposed monkeys (34% vs. 7%). The results indicate that MeHg is not a suitable reference for risk assessment from exposure to thimerosal-derived Hg. Knowledge of the toxicokinetics and developmental toxicity of thimerosal is needed to afford a meaningful assessment of the developmental effects of thimerosal-containing vaccines.

Figure 1

Weight gain of infant monkeys during study. Error bars indicate SE.

Figure 2

Comparison of predicted and observed mean blood total Hg concentrations during and after four weekly oral doses (20 μg/kg) of MeHg. Error bars indicate SD.

Figure 3

A semilogarithmic plot of washout of total Hg in blood and the brain after four weekly oral doses (20 μg/kg) of MeHg. The data were collected from groups of infant monkeys sacrificed 2, 4, 7, and 28 days after the last dose. The lines represent nonlinear regression fit of the data to a monoexponential model; the regression estimate (± SE) of T_1/2 is T_1/2 = 19.1 ± 5.1 days (r = 0.81) for blood and T_1/2 = 59.5 ± 24.1 days (r = 0.59) for brain.

Figure 4

A semilogarithmic plot of the washout of organic and inorganic Hg in the brain after four weekly oral doses (20 μg/kg) of MeHg. The data were collected from groups of infant monkeys sacrificed at 2, 4, 7, and 28 days after the last dose. The lines represent nonlinear regression fit of the data to a monoexponential model. The regression estimate (± SE) for organic Hg is T_1/2 = 58.4 ± 25.0 days (r = 0.57). The half-life of inorganic Hg is too long (> 120 days) to be accurately estimated from the present data (i.e., r is not significantly different from 0).

Figure 5

Comparison of predicted and observed mean blood total Hg concentration during and after four weekly im injections of vaccine containing thimerosal (20 μg/kg Hg). Error bars indicate SD.

Figure 6

A semilogarithmic plot of washout of total Hg in blood and the brain after four weekly im injections of vaccine thimerosal (20 μg/kg Hg). The data were collected from groups of infant monkeys sacrificed at 2, 4, 7, 10, 17, and 21 days after the last dose. The lines represent nonlinear regression fit of the data to a monoexponential model. The regression estimate (± SE) of T_1/2 is 24.2 ± 7.4 days (r = 0.74) for brain and 6.9 ± 1.7 days (r = 0.82) for blood.

Figure 7

A semilogarithmic plot of washout of organic and inorganic Hg in the brain after four weekly im injection of vaccines containing thimerosal (20 μg/kg Hg). The data were collected from groups of infant monkeys sacrificed at 2, 4, 7, and 28 days after the last dose. The lines represent nonlinear regression fit of the data to a monoexponential model. The regression estimate (± SE) of T_1/2 for organic Hg is T_1/2 = 14.2 ± 5.2 days (r = 0.76). The half-life of inorganic Hg is too long (> 120 days) to be accurately estimated from the present data (i.e., r is not significantly different from 0).