B-lymphocytes from a population of children with autism spectrum disorder and their unaffected siblings exhibit hypersensitivity to thimerosal

Abstract

The role of thimerosal containing vaccines in the development of autism spectrum disorder (ASD) has been an area of intense debate, as has the presence of mercury dental amalgams and fish ingestion by pregnant mothers. We studied the effects of thimerosal on cell proliferation and mitochondrial function from B-lymphocytes taken from individuals with autism, their nonautistic twins, and their nontwin siblings. Eleven families were examined and compared to matched controls. B-cells were grown with increasing levels of thimerosal, and various assays (LDH, XTT, DCFH, etc.) were performed to examine the effects on cellular proliferation and mitochondrial function. A subpopulation of eight individuals (4 ASD, 2 twins, and 2 siblings) from four of the families showed thimerosal hypersensitivity, whereas none of the control individuals displayed this response. The thimerosal concentration required to inhibit cell proliferation in these individuals was only 40% of controls. Cells hypersensitive to thimerosal also had higher levels of oxidative stress markers, protein carbonyls, and oxidant generation. This suggests certain individuals with a mild mitochondrial defect may be highly susceptible to mitochondrial specific toxins like the vaccine preservative thimerosal.

Figure 1

This figure shows the typical responses of a family of cells to thimerosal, measured on day 5 postinoculation. ASD = black, Twin = red, Sibling = blue, Control = green. (a) The percent control LDH values are plotted against increasing thimerosal concentrations for the 4 cell types. Arrows indicate LDH-G₅₀ values with ASD = 314 nM, Twin = 648 nM, Sib = 373 nM, and Cont = 1000 nM. (b) The percent control XTT values are plotted against increasing thimerosal concentrations. XTT-G₅₀ values are obtained using the mitochondrial XTT assay. (c) The percentage of cellular LDH that is accessible to lactate (i.e., percentage of dead and dying cells) is plotted against thimerosal. (d) Cell number and XTT reduction rates as a function of increasing thimerosal concentrations are shown. The plots obtained are “hockey stick” in shape.

Figure 2

(a) shows the thimerosal concentration that induced a 50% growth inhibition at day 5, measured by the LDH method (LDH-G₅₀). In (a) the color-coded cells are ranked in terms of sensitivity to thimerosal. Four family groups are shown whose autistic B-cell LDH-G₅₀'s fall greater than 2 standard deviations below the mean: A, B, C, and D. Underlined letters denote those cells believed to have a heightened sensitivity to thimerosal (i.e., those falling outside two standard deviations of the control population). The distribution of the control population is indicated by the green lines showing mean + SD, mean, mean – SD, and mean − 2SD. Distribution of Cells Types. In the upper insert, (b) we show the distribution of the four cells types: ASD, unaffected Twin, unaffected Sibling, and external age/sex matched control, in the four quartiles of the ranked distribution. It is noteworthy that the ASD derived cells are more clustered in the left hand side of the distribution and the external controls are distributed to the right hand side. Test for Systemic Errors. In the second insert, (c) we show that there is no systemic correlation between low LDH-G₅₀'s in cells drawn from families from an ASD background and their respective controls. The graph in (c) shows the average and standard deviation of the difference between the appropriate age/sex matched control and each of the ASD familial LDH-G₅₀. The positive values indicate that cells from the affected families are more sensitive than the external controls. The P values indicate the results from a one-tailed t-test, n = 11, with, (*) indicating <0.05, and (***) indicating <0.005.

Figure 3

This figure shows two methods to estimate the size of the thimerosal sensitive population, by using a pseudo-Jackknife statistical procedure (a) and by simulation (b). In (a) the ranked data set shown in Figure 2 was treated as a bimodal population. We calculate the mean and standard deviation where the size of the hypersensitive population, n, was increased from 0 to 44 and the bulk population fell from 44 to 0. The ratios of the two SDs divided by their means were plotted against n. The dashed line is the ratio of the SD/mean of the control population. The insert shows the line-shape generated when the two ratios are plotted against each other. This pseudo-Jackknifing procedure indicates that the size of the hypersensitive population is at least 8 and could be as high as 11. The ranked data were also fitted by simulation to two populations with means of ≈380 and ≈100 nM, respectively. The simulations indicated that the two populations had the same population distribution when n was 9, (b).

Figure 4

This figure shows the concentration of thimerosal which induced a 50% inhibition of growth at day 5, measured by the XTT method (XTT-G₅₀). In (a) we have ranked the color-coded cells in terms of sensitivity to thimerosal and have highlighted four family groups: A, B, C, and D. Underlined letters denote cell lines we believe have a heightened sensitivity to thimerosal. The distribution of the control population is indicated by the green line that indicated mean + SD, mean, mean – SD, and mean − 2SD. Distribution of Cells Types. In the upper insert, (b) we show the distribution of the four cells types, ASD, unaffected Twin, unaffected Sibling, and external age/sex matched control, in the four quartiles of the ranked distribution. There is less clustering of the ASD derived cells, but as with the LDH assay, the external controls are distributed to the right hand side of the rankings. Test for Systemic Errors. In the second insert, (c) we show that there is no systemic correlation between low XTT-G₅₀'s in cells drawn from families from an ASD background and their respective controls. The plot in (c) shows the average and standard deviation of the difference between the appropriate age/sex matched control and each of the ASD familial XTT-G₅₀. The positive values indicate that cells from the affected families are more sensitive than the external controls. The P values indicate the results from a one-tailed t-test, n = 11, with (*) indicating <0.05 and (***) indicating <0.005.

Figure 5

This figure shows data from Family B only, part of the same data set presented in Figure 1. (a) shows the rate of XTT reduction per million cells versus the % control cell growth for increasing thimerosal concentrations. (b) shows the rate of lactate production per million cells versus the % control cell growth for increasing thimerosal concentrations. (c) shows the XTT/lactate ratio versus % control cell growth for increasing thimerosal concentrations. (d) shows the amount of nMoles of carbonyl per mg of protein produced versus the % control cell growth for increasing thimerosal concentrations.