Ancestral exposure to thiacloprid changes the thyroid size in exposed progeny
In order to reveal the effect of thia exposure on thyroid morphology, we dissected thyroid gland and both lobes were weighted. Weights of both lobes were added and the total weight of thyroid was compared to body weight. In F1 females exposed to thiacloprid we observed a significant 18% decrease in thyroid weight compared to the control (Fig. <link rid="fig1">1</link>–1). Female mice from the F3 generation, derived from dose “0.6,” show a 1.3-fold increase in thyroid weight, while female F3 mice derived from dose “6” did not show significant changes (Fig. 1–2). In F1 males, the progeny derived from exposure to dose “6” show a tendency of decreases in thyroid weights (0.8-fold; p = 0.1) (Fig. 1–3). In the F3 generation, males derived from exposure dose “0.6” show a significant increase (1.4-fold) in thyroid weight while no significant changes were detected for dose “6” (Fig. 1–4).
To analyze morphology, the thyroids were fixed and embedded in paraffin, sections were cut and stained with Hematoxylin and Eosin (H&E). Images were taken using nanozoomer (Fig. 2 − 1, Fig. <link rid="fig2">2</link>–2). The images were analysed and the data were averaged and plotted. Specifically, we calculated the averaged size of the lumen called colloids, as well as the total volume of colloid. Our analysis showed that in F1 females no significant changes in the number of follicles, in the mean area of follicles, nor the mean volume were observed (Fig. 2–3). In F3 females, the mean area of follicles and the volume has decreased by 32% and 40% respectively in the exposed group compared to control (Fig. 2–4). In F1 males, we observed a 42% increase in the number of colloids in the exposed group compared to control for dose “6”. However, this increase did not pass the significance threshold (p = 0.18) cut off (Fig. 2C). The area and volume of colloids in the exposed group significantly decreased by 27% and 40%, respectively, compared to the control (Fig. 2–5). In F3 males, the average number of colloids increased by 43% in the exposed group compared to the control (p = 0.057). The mean area and volume of colloids in the exposed group decreased by 47% and 63%, respectively, compared to the control, but these changes were not significant (Fig. 2–6).
To conclude, our analysis showed that ancestral exposure to thiacloprid leads to a decrease in thyroid weights in F1 females. Both males and female F3 progeny showed an increase in thyroid weights at dose “0.6”. No significant changes in thyroid weights were observed for dose “6” in F3 for both sexes. No significant changes in the morphology were observed in F1 females exposed to dose “6”. In F3 progeny, females had a significant decrease in the mean area and the mean volume of thyroid. In F1 males we observed decrease in the mean area and volume of glands and a tendency to in the number of colloids in F3.
Thyroid hormonal imbalance in serum of F1 and F3
Since the production of thyroid hormones is the major function of the thyroid, we performed the analysis of the most abundant hormone, T4, using blood serum of control and treated groups using ELISA. In F1 females, we found that T4 levels have significantly increased by 3.4% in the treated group (Fig. 3 − 1). In the F3 generation, there is tendency to increase by 3.5% (p = 0.095) in the thia “6” derived group. In F1 males, we found an increase in T4 by 3.3% but values did not pass significant cut off (p = 0.27) between the two groups. In F3 males, there was a significant 5.3% decrease of T4 levels in treated mice (Fig. 3 − 1). Because TSH is important for the regulation of T3 and T4 hormones, we decided to measure TSH by ELISA using blood serum. In F1 females, there was a significant 24% TSH increase while in F1 males no significant changes were observed (Fig. 3 − 2). We did not perform the analysis in F3. Since hypothalamus plays a key role in the production of TSH and regulation of hormonal biosynthesis is also linked to the activity of the brain cholinergic system, we decided to analyse the impact of thiacloprid on gene expression in the hypothalamus. To this end, we dissected hypothalamuses, extracted RNA, prepared cDNA and performed RT-qPCR analysis. We analysed genes encoding several key acetylcholine receptors such as Chrna1, Chrna3, Chrna7, and Chrnb2 and Chrnb4, genes encoding specific subunits of TSH (Tshb) and FSH (Fshb) hormones as well as the gene encoding a common subunit of these hormones, Cga. Since the importance of the imprinted gene Gnas for hormonal regulation was reported, we also included this gene for the analysis. In F1 female brains, there was a decrease in expression of genes encoding cholinergic receptors, however the changes were not significant, in F3 females we observed a significant decrease in Chrna7 expression (Fig. <link rid="fig3">3</link>–3). In F1 females we observed a 3.3-fold significant decrease in Tshb, the gene had a tendency to decrease in F3 females (Fig. <link rid="fig3">3</link>–3). In F1 males, we observed a significant decrease in Chrna1 and significant 3-fold increase in Chrnb2 expression, this gene had a tendency to decrease by 2.6-fold in F3 (Fig. 3–4). Both F1 males and females had a tendency to increase in Gnas.
To sum up, exposure to thia increases both T4 and TSH levels in F1 females suggesting that a major function of thyroid hormone production has been disrupted upon exposure in females. F3 males have a decrease in T4 hormone levels, suggesting that effects induced in first generation males were not rescued in third generations. The changes in thyroid hormones were associated with hypothalamus gene expression alterations suggesting that hormonal disfunctions could be induced by the perturbation of nervous system regulatory mechanisms.
Quantification of the DIO1 protein shows a decrease in F1 and F3 female thyroids.
The active form of thyroid hormone is T3 which originates from the conversion of T4 by enzymes called deiodinases. Type 1 and type 2 deiodinase (DIO1 and DIO2) convert T4 into T3, whereas D3 degrades T4 and T3 into inactive metabolites. DIO1 is expressed in the thyroid, liver and kidneys and it is suggested that DIO1 is a sensitive marker of peripheral thyroid status in the mouse 26. Thus, we decided to analyse DIO1 protein levels using immunofluorescence microscopy. We prepared thyroid tissue sections and incubated them with a primary antibody against DIO1 and a secondary fluorescent antibody. The images were taken with the same exposure time and objective aperture and the signals were measured and corrected according to the size of the analysed area. We performed this analysis only in females (Fig. 4 − 1, Fig. 4 − 2). The results showed a significant decrease of DIO1 protein in the treated group compared to the control in F1 and F3 by 53 and 37% respectively (Fig. 4 − 3), suggesting a possible impact on T4 conversion into T3 which is consistent with the elevated T4 levels found in females.
Changes in GNAS protein levels following thia exposure
GNAS complex locus gives rise to maternally, paternally, and biallelically expressed transcripts. The gene encodes for different isoforms of the G-protein alpha subunit, which is a key element of the signal transduction pathways involving the activation of adenylyl cyclase. Since a large number of thyroid pathologies in human have been linked to GNAS deficiency 27,28,29,30 we decided to analyze the protein levels of GNAS. First, we prepared protein extracts from thyroid tissue from several replicates and then quantified protein levels of GNAS by western blot in both female and male thyroids for both generations (Fig. 5 − 1). Full-length gels and blots are included in a Supplementary Information file (Supplementary Figure S2). The analysis showed that GNAS has a tendency to increase by 3-fold in F1 females (p = 0.0635) and a tendency to decrease in F3 females by 0.3-fold (p = 0.074) (Fig. 5 − 2). In F1 males we observed a tendency to decrease in GNAS amounts by 0.2-fold in (p = 0.057) and a nonsignificant decrease in F3 was detected (Fig. 5 − 2).
Next, we prepared tissue sections and performed immunofluorescent analysis of GNAS in females. The analysis showed that GNAS protein is mostly localized at the membranes. The analysis detected a significant increase of GNAS levels in F1, and a non-significant decrease in F3 (Fig. 5 − 3, Fig. 5 − 4), the directions of the changes are consistent with our western blot data.
Gene expression analysis reveals dramatic changes in the expression of genes involved in thyroid development and hormone biosynthesis and transport
To reveal the effects of thia exposure on gene expression in the thyroid gland, we chose several target genes based on their importance for thyroid gland development, hormone biosynthesis, hormone transport and remodelling functions (Supplementary table S1). We extracted RNA, synthesized cDNAs and performed RT-qPCR analysis in F1 and F3 groups from both sexes. In F1 females, we found that Egfr and Foxe1, which are important for thyroid gland development, were strongly upregulated by 4.8 and 7.8 folds, respectively. These changes were not significant in F3 (Fig. 6 − 1 and Fig. 6 − 2). Id4 was upregulated in F1 by 2- fold and decreased in F3 0.7-fold. Another factor, Tbx1 significantly increased in F3 by 2.3-fold. The most dramatic changes in female thyroid were observed in thyroid hormone biosynthesis genes. We observed a dramatic increase in Duox2, Iyd, Tg and Tpo in F1 by 7.3, 3.6, 7.8-folds, respectively. Duox2, Tg and Tpo were significantly increased in F3 by 3.4, 3.2 and 4.2-folds respectively. Slc16a2 encoding the transporter MCT8 has increased by 5.7-fold in F1 and tend to increase in F3 by 2.4-fold (Fig. 6 − 1 and Fig. 6 − 2).
In F1 males, the expression of the Foxe1 was similarly upregulated by 1.9-fold in the treated group while no changes were detected in F3. Id4 was upregulated in F1 and decreased in F3. Pax8 had decreased by 1.8-fold in F1. We observed a similar increase in Duox2 by 1.6-fold in F1 and a decrease in Duox2 and Tpo in F3 males by 2 and 1.6-fold, respectively. We also observed that gene expression of Dio1 decreased in F3 males by 2-fold. Slc16a2 had a tendency to increase in F1 by 1.6-fold and significantly decreased in F3 by 1.6-fold (Fig. 7 − 1 and 7 − 2).
In conclusion, our analysis shows that there are significant changes in the expression of genes involved in thyroid development, hormone biosynthesis and transport in F1 and F3 thyroids from both sexes. The genes encoding proteins important for hormone biosynthesis were strongly upregulated in both F1 and F3 females which is consistent with the observed T4 hormone increase in females.
Analysis of H3K4me3 occupancy in sperm revealed alterations at Dio1, Mmp2 and Tg loci.
Because the previous results show that some genes have an altered expression in the third generation of exposed animals, we decided to look whether these expression changes have been transmitted through generations via epigenetic changes in sperm. It is suggested that regions which retained histones in the sperm are responsible of the transmission of the induced alterations to subsequent generations 31. In order to check whether H3K4me3 marks are preserved in sperm, we used the most recent dataset of sperm H3K4me3 occupancy in mice32. We found that most of the thyroid genes we used in our gene expression analysis display H3K4me3 at their promoters except for Tpo. To reveal whether the histone containing fraction shows modified H3K4me3 occupancies, we performed ChIP followed by qPCR in F1, F2 and F3 sperm (Fig. 8 − 1, Fig. 8 − 2, Fig. 8 − 3). An equal amount of immunoprecipitated DNA and Input were used for qPCR. We narrowed the analysis to the promoters of the genes that were differentially expressed in thyroid. In F1 sperm we did not observe a significant difference between the control and the treated groups except for a tendency to decrease for Dio1 and Slca16a2 promoters. In F2 sperm, the promoters of Tg and Mmp2 genes have increased H3K4me3 occupancy and a tendency to increase for Slc16a2, and Peg3. Dio1 has a tendency to have decreased H3K4me3 occupancy in F1 and have decreased occupancy in F3 which is consistent with Dio1 protein levels in F1 and F3 in females and with gene expression in F3 males (Fig. 8 − 1 and Fig. 8 − 2, Fig. 8 − 3).
To conclude, our data shows that changes in gene expression of Dio1, Mmp2 and Tg in F3 could be explained by increased histone H3K4me3 occupancy in the sperm at their promoters.