Thyroid hormone elicits intergenerational epigenetic effects on adult social behavior and fetal brain expression of autism susceptibility genes

Abstract

Genetic mutations identified in genome-wide association studies can only explain a small percentage of the cases of complex, highly heritable human conditions, including neurological and neurodevelopmental disorders. This suggests that intergenerational epigenetic effects, possibly triggered by environmental circumstances, may contribute to their etiology. We previously described altered DNA methylation signatures in the sperm of mice that experienced developmental overexposure to thyroid hormones as a result of a genetic defect in hormone clearance (DIO3 deficiency). Here we studied fetal brain gene expression and adult social behavior in genetically normal F2 generation descendants of overexposed mice. The brain of F2 generation E13.5 fetuses exhibited abnormal expression of genes associated with autism in humans, including Auts2, Disc1, Ldlr, Per2, Shank3, Oxtr, Igf1, Foxg1, Cd38, Grid2, Nrxn3, and Reln. These abnormal gene expression profiles differed depending on the sex of the exposed ancestor. In the three-chamber social box test, adult F2 generation males manifested significantly decreased interest in social interaction and social novelty, as revealed by decrease total time, distance traveled and time immobile in the area of interaction with novel strangers. F1 generation mice, compared to appropriate controls also exhibited altered profiles in fetal brain gene expression, although these profiles were substantially different to those in the F2 generation. Likewise adult F1 generation mice showed some abnormalities in social behavior that were sexually dimorphic and milder than those in F2 generation mice. Our results indicate that developmental overexposure to thyroid hormone causes intergenerational epigenetic effects impacting social behavior and the expression of autism-related genes during early brain development. Our results open the possibility that altered thyroid hormone states, by eliciting changes in the epigenetic information of the germ line, contribute to the susceptibility and the missing-but heriTables-etiology of complex neurodevelopmental conditions characterized by social deficits, including autism and schizophrenia.

Keywords: autism; fetal brain; social behavior; thyroid hormone; transgenerational epigenetic.

FIGURE 1

Ancestry tree of the experimental animals and E13.5 brain gene expression of T3 regulated genes. (A) Ancestry of experimental animals studied that were F1 and F2 generation descendants of animals that were overexposed to thyroid hormone during development (Dio3−/− mice, in red). PGM and PGF indicate genetically normal animals with a paternal grandmother or paternal grandfather, respectively, overexposed. (B) Expression of the housekeeping gene Actin b. (C) Expression of T3 responsive genes Klf9 and Hr. Each point represents a different embryo (n = 16, 15, 15) and mean ± SEM are shown. Experimental embryos represent 3–4 different litters. ** and *** indicate P < 0.01 and 0.001, respectively, as determined by ANOVA and Tukey’s post-hoc test.

FIGURE 2

Expression of neurodevelopmental disorders genes in E13.5 PGF and PGM brains. (A) Genes showing increased expression in PGF fetal brains. (B) Genes showing a decreased in PGM brains. (C) Genes showing an increase in PGM or in both PGF and PGM. (D) Genes showing no different expression between experimental groups. Each point represents a different embryo (n = 16, 15, 15) and mean ± SEM are shown. Experimental embryos represent 3–4 different litters. *, ^**, and ^*** indicate P < 0.05, 0.01, and 0.001, respectively, as determined by ANOVA and Tukey’s post-hoc test.

FIGURE 3

Sociability of PGF males. (A–C) Second trial of the three-chamber social box test, after first stranger is introduced in the right chamber. (A) Time spent by the test mouse in each chamber. (B) Distance traveled by the test mouse in each chamber. (C) Total time, time immobile, and distance traveled by the test mouse in the area immediately around the 1st stranger in the right chamber. Each point represents a different mouse tested at approximately 18 weeks of age (n = 17, 16) and mean ± SEM are shown. Experimental mice from each group represent 4 different litters and data represent two animal cohorts that were generated and tested at different dates. ^** and ^*** indicate P < 0.01 and 0.001, respectively, as determined by the Student’s t-test.

FIGURE 4

Interest in social novelty of PGF males. (A–D) Third trial of the three-chamber social box test after a second stranger is added into the left chamber. (A) Time spent by the test mouse in each chamber. (B) Distance traveled by the test mouse in each chamber. (C) Total time, time immobile and distance traveled by the test mouse in the area immediately around the 1st stranger in the right chamber. (D) Total time, time immobile, and distance traveled by the test mouse in the area immediately around the 2nd stranger in the left chamber. Each point represents a different mouse tested at approximately 18 weeks of age (n = 17, 16) and mean ± SEM are shown. Experimental mice represent 4–7 different litters and data represent two animal cohorts that were generated and tested at different dates. * and ^** indicate P < 0.05 and 0.01, respectively, as determined by the Student’s t-test.

FIGURE 5

Sociability of F1 generation males. (A–C) Second trial of the three-chamber social box test, after first stranger is introduced in the right chamber. (A) Time spent by the test mouse in each chamber. (B) Distance traveled by the test mouse in each chamber. (C) Total time, time immobile, and distance traveled by the test mouse in the area immediately around the 1st stranger in the right chamber. Each point represents a different mouse tested at approximately 18 weeks of age (n = 22, 20) and mean ± SEM are shown. Experimental mice from each group represent 4–6 different litters and data represent two animal cohorts that were generated and tested at different dates.

FIGURE 6

Interest in social novelty of F1 generation males. (A–C) Second trial of the three-chamber social box test, after first stranger is introduced in the right chamber. (A) Time spent by the test mouse in each chamber. (B) Distance traveled by the test mouse in each chamber. (C) Total time, time immobile, and distance traveled by the test mouse in the area immediately around the 1st stranger in the right chamber. (D) Total time, time immobile, and distance traveled by the test mouse in the area immediately around the 2nd stranger in the left chamber. Each point represents a different mouse tested at approximately 18 weeks of age (n = 22, 20) and mean ± SEM are shown. Experimental mice from each group represent 4–6 different litters and data represent two animal cohorts that were generated and tested at different dates.

FIGURE 7

Sociability of F1 generation females. (A–C) Second trial of the three-chamber social box test, after first stranger is introduced in the right chamber. (A) Time spent by the test mouse in each chamber. (B) Distance traveled by the test mouse in each chamber. (C) Total time, time immobile, and distance traveled by the test mouse in the area immediately around the 1st stranger in the right chamber. Each point represents a different mouse tested at approximately 18 weeks of age (n = 17, 19) and mean ± SEM are shown. Experimental mice from each group represent 4–6 different litters and data represent two animal cohorts that were generated and tested at different dates.

FIGURE 8

Interest in social novelty of F1 generation females. (A–C) Second trial of the three-chamber social box test, after first stranger is introduced in the right chamber. (A) Time spent by the test mouse in each chamber. (B) Distance traveled by the test mouse in each chamber. (C) Total time, time immobile, and distance traveled by the test mouse in the area immediately around the 1st stranger in the right chamber. (D) Total time, time immobile, and distance traveled by the test mouse in the area immediately around the 2nd stranger in the left chamber. Each point represents a different mouse tested at approximately 18 weeks of age (n = 17, 19) and mean ± SEM are shown. Experimental mice from each group represent 4–6 different litters and data represent two animal cohorts that were generated and tested at different dates. *, ^**, and ^*** indicate P < 0.05, 0.01, and 0.001, respectively, as determined by the Student’s t-test.

FIGURE 9

Gene expression in F1 generation E13.5 brains. (A) Expression of house-keeping gene Gapdh and thyroid hormone-responsive Klf9. (B) Genes showing no significant difference in expression. (C) Genes showing a significant decrease in expression. Each point represents a different embryo (n = 18, 29) and mean ± SEM are shown. Experimental embryos represent 3–4 different litters. *, ^**, and ^*** indicate P < 0.05, 0.01, and 0.001, respectively, as determined by the Student’s t-test.