Maternal Thyroid Hormone Programs Cardiovascular Functions in the Offspring

Abstract

Background:

Maternal thyroid hormone (TH) plays an essential role for fetal development, especially for the cardiovascular system and its central control. However, the precise consequences of altered TH action during the different periods in pregnancy remain poorly understood.

Methods:

To address this question, we used mice heterozygous for a mutant thyroid hormone receptor α1 (TRα1) and wild-type controls that were born to wild-type mothers treated with 3,3′,5-triiodothyronine (T3) during the first or the second half of pregnancy. We then phenotyped the offspring animals as adults by in vivo measurements and postmortem tissue analyses.

Results:

Maternal T3 treatment in either half of the pregnancy did not affect postnatal growth development. Serum thyroxine and hypophyseal thyrotropin subunit beta or deiodinase type II expression was also not affected in any group, only TRα1 mutant males exhibited a reduction in serum T3 levels after the treatment. Likewise, hepatic deiodinase type I was not altered, but serum selenium levels were reduced by the maternal treatment in wild-type offspring of both genders. Most interestingly, a significant increase in heart weight was found in adult wild-types born to mothers that received T3 during the first or second half of pregnancy, while TRα1 mutant males were protected from this effect. Moreover, we detected a significant increase in heart rate selectively in male mice that were exposed to elevated maternal T3 in the second half of the pregnancy.

Conclusion:

Taken together, our findings demonstrate that maternal TH is of particular relevance during the second half of pregnancy for establishing cardiac properties, with specific effects depending on TRα1 or gender. The data advocate routinely monitoring TH levels during pregnancy to avoid adverse cardiac effects in the offspring.

Introduction

Thyroid hormone (TH) is important for development (1). As the fetal thyroid gland does not produce hormones until late pregnancy in humans, around embryonic day 15 in mice (2), the embryonal demand for the hormone needs to be met by the maternal thyroid gland (3). Consequently, defects in maternal thyroid function can severely impact the fetus, especially the developing brain (4), as well as pregnancy outcomes including miscarriage and growth retardation. Moreover, the offspring's setting of the hypothalamic/pituitary/thyroid (HPT) axis can be permanently altered (5,6).

The HPT axis regulates circulating THs via an endocrine feedback loop. Thyrotropin-releasing hormone (TRH) is secreted by the hypothalamus, stimulating the anterior pituitary to produce thyrotropin (TSH), which drives the thyroid gland to release THs. THs in turn negatively regulate TRH and TSH, thereby affecting their own production and release (7). This tightly controlled mechanism is required as TH has potent effects on many physiological functions, including energy homeostasis (8) and cardiac contractility or heart rate (9 –11). The active hormone 3,3′,5-triiodothyronine (T3) acts via nuclear TH receptors (TRs), namely TRα1 and TRβ (12), which bind to target genes and lead to activation or suppression of gene expression (13).

Despite TH's major role in development, it is currently controversial whether pregnant women should be routinely screened for thyroid function, and whether to substitute in subclinical cases (14,15). As maternal thyroid dysfunction can impair cognitive and motor development in the offspring (16 –18), the majority of clinical studies have focused primarily on these endpoints. However, recent clinical and preclinical data have demonstrated that maternal TH is also crucial for the central control of cardiovascular functions in the offspring (19,20), suggesting that these endpoints should also be considered. Nevertheless, the precise consequences of altered gestational TH signaling on the cardiovascular functions of the offspring are still incompletely understood.

Here we used mice heterozygous for the dominant negative TRα1R384C mutation (TRα1+m) in combination with maternal pharmacological T3 treatments to characterize the temporal role of prenatal TRα1 signaling for the fetal programming of cardiovascular properties. Taking advantage of the fact that mutant TRα1 can be reactivated by elevating T3 levels in vivo (21), we set up an experimental paradigm, in which TRα1 signaling in the TRα1+m embryo was selectively restored during the first or second half of pregnancy. The animals were compared with wild-type littermates (TRα1++) exposed to high T3 in the same periods as well as untreated controls.

Our findings show that maternal TH is important for establishing cardiac properties. Especially in the second half of pregnancy, elevated TH levels lead to increased heart rate and weight in the male offspring, associated with permanent changes in pacemaker channel gene expression and DNA methylation.

Materials and Methods

Animal husbandry and study design

Mice were housed at 21°C on a 12-hour light/dark cycle with free access to food and water. Animal procedures were conducted according to EU directives (2010/63/EU) and approved by Ministerium für Energiewende, Landwirtschaft, Umwelt, Natur und Digitalisierung, Schleswig-Holstein, Germany.

Wild-type C57BL/6NCr female mice were purchased at 10 weeks from Charles River, Germany, and mated to heterozygous TRα1+m males to obtain TRα1++ or TRα1+m offspring. Dams were treated either from the first day after positive plug (E0.5) until E12.5 or from E12.5 until E18.5 with T3 via drinking water (0.5 μg/mL; T6397, Sigma Aldrich, Germany) containing 0.01% BSA (A7906, Sigma Aldrich). If not stated otherwise, adult offspring were analyzed at four to five months.

Gene expression measurement by quantitative real-time PCR

RNA was isolated using the RNeasy mini or fibrous tissue mini kit (QIAGEN, Germany) including a DNase digestion step to prevent genomic DNA contamination. cDNA was synthesized of 1 μg RNA using the Molecular Biology RevertAid cDNA Kit (Thermo Fisher Scientific, Germany). QuantStudio (Thermo Fisher Scientific, Germany) and GoTaq^® qPCR Master Mix (Promega, Germany) were used for quantitative real-time PCR (qPCR). Reference genes were determined using NormFinder (https://moma.dk/normfinder-software, Denmark), revealing Cyclophilin over Rpl32 and/or Hprt as most stable. A standard curve was used to correct PCR efficiency. Specificity of amplification was verified by melting curve. Primer sequences are available on request.

Heart rate and blood pressure analysis

For blood pressure measurements, a noninvasive tail-cuff system with a temperature-controlled platform (100°F) was used (Model SC-100; Hatteras Instrument Incorporated, USA). Blood pressure (diastolic, systolic, and mean arterial pressure) was measured 15 times in 1 session and averaged over 3 consecutive days.

Serum parameter measurements: total T3 and thyroxine enzyme linked immunosorbent assay

Serum concentrations of total T3 and total thyroxine (T4) were determined by enzyme linked immunosorbent assay (ELISA) (tT3: DNOVO53, NovaTec Immundiagnostica GmbH, Dietzenbach, Germany, detection limit 0.05 ng/mL, intra- and interassay variance <11%; tT4: ELISA, EIA-1781; DRG Instruments GmbH, Marburg, Germany, detection limit 0.5 ng/mL, intra- and interassay variance <4.5%).

Western blot analysis

Hearts were homogenized in RIPA buffer with protease inhibitors (Roche Diagnostics GmbH, Germany). Fifty micrograms of protein were separated on a 12% sodium duodecyl sulfate-polyacrylamide gel (Bio-Rad Laboratories, Germany), transferred onto a membrane (PVDF; Merck Millipore, Germany), and probed with the rabbit polyclonal anti-HCN2 (1:1000 dilution; APC-30, Alomone, Israel) or rabbit polyclonal anti-HCN4 antibodies (1:1000 dilution; APC-052, Alomone, Israel) followed by a peroxidase-conjugated secondary goat anti-rabbit-IgG antibody (P0448, DAKO, Denmark; 1:5000 dilution), visualized using an ECL system (Chemi Doc Touch; Bio-Rad Laboratories, Germany), and normalized to total protein.

DIO1 and GPX1 activity assays

DIO1 activity was analyzed by a nonradioactive assay (22). Briefly, liver homogenates were mixed with dimethylsulfoxide or propylthiouracil (PTU), and incubated with a substrate mix at 37°C. The released iodide was determined by a Sandell/Kolthoff reaction and the background from PTU-inhibited controls subtracted. Interassay variance was below 20%. GPX1 activity was determined by a coupled enzymatic assay (23). Inter- and intra-assay coefficient of variability was below 15%.

Trace element measurements

Total concentrations of copper and selenium in serum and liver homogenates were determined by total reflection X-ray fluorescence using a benchtop analyzer (S4 T-STAR; Bruker Nano GmbH, Berlin, Germany), as described (24), with a Seronorm standard (Sero AS, Billingstad, Norway).

DNA methylation by bisulfite pyrosequencing

Genomic DNA was extracted using the QIAamp DNA Mini Kit (QIAGEN), quantified (Quantus Fluorometer, Promega, Germany), and converted with sodium-bisulfite using the EpiTect Fast DNA Bisulfite Kit (QIAGEN). PCR was performed using the PyroMark PCR Kit (QIAGEN) and primers generated with the PyroMark Assay Design Software 2.0 (QIAGEN). The PCR product was pyrosequenced on a PyroMark Q48 Autoprep with Q48 Advanced CpG Reagents (QIAGEN). A standard curve of 0%, 25%, 50%, 75%, and 100% methylated DNA was used to quantify absolute CpG methylation.

Statistical analysis

Data were analyzed using two-way analysis of variance (ANOVA) with offspring genotype and maternal treatment as independent variables followed by Holm/Sidak's multiple comparison test using Prism 7 software (GraphPad Software, Inc., San Diego, CA, USA). Males and females were tested separately. Values are mean ± standard error of the mean. Statistical significance was defined as *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001, and details can be found in Supplementary Table S1.

Result

Experimental setup

We used mice TRα1+m mice born to wild-type mothers as model system for impaired TH signaling in utero. As this particular mutation can be reactivated by elevating TH levels (21,25), we treated the mother with 0.5 mg/L T3 in drinking water either in the first or in the second half of pregnancy (Fig. 1A), a treatment shown to result in 8–10-fold elevated serum T3 levels in adult animals and sufficient to restore embryonal TRα1 signaling selectively in these periods (4,26 –28). As the respective TRα1++ littermates were exposed to elevated TH during the same periods, we obtained the entire range of impaired or enhanced TH signaling in the developing embryos.

FIG. 1.

(A) Scheme of T3 treatment during pregnancy. Wild-type dams were mated with TRα1+m mutant males and treated with T3 in drinking water either in the first or second half of pregnancy or remained untreated. (B) Water and (C) food intake in the respective dams measured every other day and normalized to body weight. (D) Body weight of the male and female offspring in both genotypes determined every other week. (E) Body length of male and female offspring in both genotypes measured at the age of 16 weeks. All values are mean ± SEM and analyzed with two-way ANOVA with Holm/Sidak's multiple comparison test. (n = 6 for each group). ANOVA, analysis of variance; SEM, standard error of the mean; T3, 3,3′,5-triiodothyronine; TRα1, thyroid hormone receptor α1. Color images are available online.

T3 treatment has no effect on offspring's development

Throughout pregnancy, daily water intake per body weight was comparable with nonpregnant adult mice, indicating sufficiently high daily oral T3 uptake to reactivate the mutant TRα1 in utero (4,26 –28), with no effect of the T3 treatment on dams' water and food intake (Fig. 1B, C). Litter sizes were comparable (untreated [n = 5]: 7.4 ± 2.4; T3 E0.5–12.5 [n = 6]: 8.0 ± 1.7; T3 E12.5–18.5 [n = 7]: 9.0 ± 1.2). Postnatal weight development was not affected by maternal T3 treatment (Fig. 1D; Supplementary Table S1); however, body weight was lower in TRα1+m as expected (21). Total body length at the age of 16 weeks reflected this pattern (Fig. 1E).

T3 treatment during pregnancy decreases offspring's serum selenium

Maternal thyroid disease can affect the HPT-axis setting (29), however, our maternal T3 treatment caused reduced T3 only in the adult TRα1+m male offspring (Fig. 2A; Supplementary Table S1). Neither serum total T4 (Fig. 2B) nor pituitary thyrotropin subunit beta (Tshb) or deiodinase type II (Dio2) mRNA expression was altered (Fig. 2C, D), suggesting that the HPT-axis was largely unaffected. Deiodinase type I (Dio1) mRNA and DIO1 activity were higher in TRα1+m as expected (30); however, changes by maternal treatment in either period were not significant (Fig. 2E, F). Maternal T3 reduced Spot14 mRNA in female but not male offspring (Fig. 2G), while hepatic GPX1 activity, as well as selenium and copper levels, were not altered in either condition (Supplementary Fig. S1A; Fig. 2H), concurring with previous studies (31). Remarkably, serum selenium levels were strongly reduced in males of both genotypes and wild-type females upon either maternal T3 treatment (Fig. 2I), while serum copper levels, also regulated by TH (32), were not altered (Fig. 2J).

FIG. 2.

(A) Serum total T3 in adult male and female offspring of control dams or dams treated with T3 either in the first or second half of pregnancy. (B) Serum total T4 in adult male and female offspring. (C) mRNA expression of Tshb in adult male and female offspring. (D) mRNA expression of Dio2 in adult male and female offspring. (E) mRNA expression of hepatic Dio1 in male and female adult offspring. (F) DIO1 activity in male and female adult offspring. (G) Hepatic expression of Spot14 in adult male and female offspring. (H) Selenium levels of male and female offspring in serum and (I) in the liver. (J) Serum copper levels in male and female offspring. All values are mean ± SEM and analyzed with two-way ANOVA with Holm/Sidak's multiple comparison test. *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001 (n = 6 for each group). T4, thyroxine; Tshb, thyrotropin subunit beta; Dio1, deiodinase type I; Dio2, deiodinase type II. Color images are available online.

Gender- and genotype-specific cardiovascular alterations in the offspring

As maternal TH can affect the cardiovascular set point (4,20), we analyzed adult offspring using a noninvasive tail-cuff system. T3 treatment in the second half of pregnancy led to a significant increase of heart rate in wild-type and TRα1+m male offspring, while not affecting wild-type females and even showing the reverse in TRα1+m females (Fig. 3A; Supplementary Table S1). Heart rate was generally lower in TRα1+m consistent with previous studies (33 –35). Blood pressure was unaffected except for a mild increase in wild-type males and a decrease in TRα1+m females treated in the second half of pregnancy (Fig. 3B–D). Remarkably, wild-type male and female offspring displayed significant hypertrophy when mothers were treated in the second half of pregnancy (Fig. 3E). While heart weight was lower in TRα1+m as expected, it was further lowered by maternal T3 treatment selectively in males.

FIG. 3.

(A) Heart rate (beats per minute) measured in restrained but conscious adult male and female offspring of control dams or dams treated with T3 either in the first or second half of pregnancy using a noninvasive tail-cuff system. (B) Systolic, (C) diastolic, and (D) mean arterial blood pressure (MAP) analyzed in these adult offspring. (E) Heart weight (mg) in relation to body weight (g) in adult offspring. All values are shown as ± SEM of all animals and analyzed with two-way ANOVA with Holm/Sidak's multiple comparison test. *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001 (n = 6 for each group). Color images are available online.

Cardiac gene expression after maternal T3 treatment

Maternal T3 treatment led to a significant upregulation of cardiac alpha myosin heavy chain (α-Mhc), cardiac beta myosin heavy chain (β-Mhc) (35 –37), hyperpolarization-activated cyclic nucleotide-gated channel 2 and 4 (Hcn2 and Hcn4), and sarcoplasmic/endoplasmic reticulum calcium ATPase 2 (Serca2) in adult male offspring of both genotypes (Fig. 4A–G; Supplementary Table S1), with the exception of Hcn2, which was only affected by T3 in the second half of pregnancy. Females were less affected, primarily TRa1+m offspring treated in the second half of pregnancy—an effect not caused by protective deiodinase III (Dio3) expression (Fig. 4H).

FIG. 4.

(A) α-Mhc expression in hearts of adult male and female offspring of control dams or dams treated with T3 either in the first or second half of pregnancy. (B) β-Mhc expression in these animals. (C) Pacemaker Hcn2 expression in these offspring. (D) Hcn4 expression in these offspring. (E) HCN2 and (F) HCN4 protein levels in the hearts of these animals. (G) Serca2 expression in these mice. (H) Dio3 expression in the hearts of these offspring. All values are shown as ± SEM of all animals and analyzed with two-way ANOVA with Holm/Sidak's multiple comparison test. *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001 (n = 6 for each group). α-Mhc, cardiac alpha myosin heavy chain; β-Mhc, cardiac beta myosin heavy chain; Hcn2, hyperpolarization-activated cyclic nucleotide-gated potassium channel 2; Serca2, sarcoplasmic/endoplasmic reticulum calcium ATPase 2; Dio3, deiodinase III. Color images are available online.

Maternal T3 treatment in the second half of pregnancy leads to altered DNA methylation in Hcn2 and Hcn4 genes

Given that pacemaker Hcn2 and Hcn4 expression correlated strongly with the heart rate (Supplementary Fig. S1C, D, and Supplementary Table S1), we hypothesized that altered fetal programming by DNA methylation in these genes could underlie the persistently elevated expression. Using the publicly available genome-wide methylation data sets, GSE21415 (38) and ENCSR397YEG (39), we identified CpG sites in these genes that are actively methylated/demethylated during cardiac development, expecting that they might be most sensitive to reprogramming by maternal TH. Subsequently, we measured DNA methylation in intron four of the Hcn2 gene, observing a consistent decrease in three of five CpG sites in male offspring of both genotypes induced by T3 treatment in the second half of pregnancy (Fig. 5A–C; Supplementary Fig. S1E, F, and Supplementary Table S1). Within exon 1 of Hcn4, three of the four CpG sites showed a significant increase in DNA methylation after T3 treatment in the second half of pregnancy (Fig. 5D–F; Supplementary Fig. S1G). For the affected CpG sites, Hcn2 expression correlated negatively and Hcn4 positively with DNA methylation (Supplementary Fig. S2), suggesting an altered chromatin state in these regulatory regions.

FIG. 5.

(A) Absolute genomic cardiac CpG DNA methylation of Hcn2 at the CpG site chr10:79,192,307 within the fourth intron of adult male and female offspring, whose mothers were treated either in the first half or in the second half of pregnancy compared with the control group quantified using bisulfite pyrosequencing. (B) Absolute genomic cardiac CpG DNA methylation of Hcn2 at the CpG site chr10: 79,192,303 in these animals. (C) Absolute genomic cardiac CpG DNA methylation of Hcn2 at the CpG site chr10:79,192,274 in these animals. (D) Absolute genomic cardiac CpG DNA methylation of Hcn4 at the CpG site chr9:58,672,034 within the fourth intron of adult male and female offspring, whose mothers were treated either in the first half (orange) or in the second half of pregnancy (red) compared with the control group (blue) quantified using bisulfite pyrosequencing. (E) Absolute genomic cardiac CpG DNA methylation of Hcn4 at the CpG site chr9: 58,672,042 in these animals. (F) Absolute genomic cardiac CpG DNA methylation of Hcn4 at the CpG site chr9: 58,672,044 in these mice. Data were analyzed using two-way ANOVA with offspring genotype and maternal treatment as independent variables followed by Holm/Sidak's multiple comparison test. Males and females were tested separately. Values are mean ± SEM. Statistical significance was defined as *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001. Color images are available online.

Discussion

Several studies have shown that maternal TH affects embryonal development (40,41) and particularly the cardiovascular system (20,42,43). However, the precise window of TH action and its mechanistic targets have remained unclear. Our study now provides a phenotypical roadmap on the cardiovascular consequences of altered TH signaling during the first or second half of murine pregnancy, corresponding to the first and second trimester in humans.

Consequences of maternal T3 treatment on early development and HPT axis

While altered TH levels during the entire pregnancy can affect body mass and body length of the offspring (44,45), this was not observed in our study, likely due to the shorter treatment being insufficient to induce catabolic conditions. TRα1+m offspring were lighter than their control littermates as expected (21), underlining the well-established importance of TRα1 for postnatal development (46,47).

Although maternal TH can interfere with the offspring's HPT axis, this was generally not observed in our paradigm. Likewise, Dio1 as a sensitive hepatic T3 marker was not affected by maternal T3 treatment, and only elevated in TRα1+m mice as observed previously (30). However, there was an interesting tendency of increased Dio1 mRNA and activity despite the somewhat reduced serum TH levels in the maternally T3-treated offspring. This may suggest altered hepatic programming, concurring with a study revealing that altered perinatal hepatic TH action by abolishing Dio2 expression leads to permanent epigenetic modifications in the liver (48).

Interestingly, serum selenium levels were reduced in wild-type offspring upon maternal T3 treatment in the second half of pregnancy. While T3 is known to positively regulate serum selenium levels in the adult animal (31), our findings rather suggest that selenium metabolism may also be epigenetically reprogrammed by maternal T3. A similar effect was observed in congenital hypothyroid children (49). Despite sufficient T4 substitution, their serum selenium levels remained lower than controls, suggesting that even a minor transient developmental hypothyroidism can cause permanent alterations in adult selenoprotein physiology. As hepatic selenium levels, DIO1, and GPX1 enzyme activity were unaffected, the liver is likely not the origin of the effect; therefore, further studies addressing this are warranted.

Sexual dimorphic effects on cardiac functions after maternal T3

Although variations in maternal TH function affect cardiovascular parameters in humans (42) and mice (4,20), blood pressure was generally unaffected. In contrast, heart weight was increased in wild-types of mothers treated in the second half of pregnancy with T3; an effect that was abolished in male TRα1+m offspring, suggesting a protective role of the mutant TRα1. This concurs with previous studies showing reduced heart weight in the offspring of hypothyroid mothers (50), and the stronger phenotype in offspring exposed to high T3 in the second half of pregnancy matches with the onset of TRα1 expression at E13.5 in mice (51). While the precise reason for the hypertrophy remains to be established, it can be speculated that maternal T3 either acts in the developing fetal brain, given that a central mechanism also drives cardiac hypertrophy in adult hyperthyroidism (52), or directly in the fetal heart. However, we found elevated α- and β-Mhc in the treated offspring, although these genes are usually oppositely regulated by T3. Together with the fact that hyperthyroidism-induced hypertrophy is usually reversed after reestablishing euthyroidism (53), the precise mechanism seems to be different than the classic cardiac T3 signaling cascades.

In contrast to the hypertrophy, which was similar in males and females, we observed a sexual dimorphism in heart rate: Male offspring of mothers treated with T3 in the second half of pregnancy had increased heart rate in both genotypes, although heart rate was generally lower in TRα1+m mice as expected (33), while a tachycardia was absent in females. This effect correlated with the level of pacemaker gene expression, especially HCN4 protein, suggesting that these genes may drive the tachycardia. Why the effect was only apparent in males remains unknown, but it could be speculated that the previously described cross talk between TH and sex hormones (54) or the well-known sexual dimorphism in deiodinase regulation (55) might contribute to differences in local TH action during cardiac development.

Epigenetic alterations in pacemaker genes associated with maternal T3

Based on the persistently elevated level of Hcn2 and Hcn4 gene expression, we hypothesized that altered fetal programming by DNA methylation might contribute to the observed phenotype. This concept of an interaction between TH and epigenetic programming has been established recently in several elegant studies, which demonstrated that (i) Dio3 is an important gatekeeper for epigenetic programming in the brain (56), (ii) the sensitivity to TH action in the offspring is also epigenetically controlled (57), and (iii) alterations in maternal TH indeed affect DNA methylation patterns of several genes (58). It is therefore not surprising that we observed a similar effect of maternal TH on DNA methylation of Hcn2 and Hcn4, especially since intronic regions are known to be of regulatory importance for the regulation of Hcn genes (59). While some CpG sites were affected in both genders, others were affected only in males, an effect often observed in fetal programming (60). T3 treatment in the second half of pregnancy had a more pronounced effect on DNA methylation, concurring with the important role of TRα1 in this developmental period of the heart (61). It remains to be elucidated whether these DNA methylation changes indeed causally affect gene expression, and how far increased HCN2 and HCN4 contribute to tachycardia in males.

Conclusion

In conclusion, our study reveals that maternal TH signaling—in particular in the second half of pregnancy—is important for establishing cardiac properties in the offspring. These findings are of clinical relevance, as a large proportion of pregnant women display thyroid disorders (62), but there is a controversial discussion about whether pregnant women should be routinely tested (63). Unfortunately, the current arguments are largely based on studies with psychological and motoric endpoints. Our study now advocates for the inclusion of cardiovascular endpoints in these studies and underlines the necessity to study endocrine disrupting compounds during pregnancy (64).

Footnotes

Acknowledgments

We thank the staff of the animal facility (GTH) of the University of Lübeck for their help. The support of Cathleen Geissler and Henriette Kirchner with the epigenetic quantifications is greatly appreciated.

Authors' Contributions

M.P., R.O., Q.S., and J.R. performed the experiments; L.S. and J.M. designed the research and supervised the experiments; M.P. and J.M. drafted the article; and all authors edited and approved the article.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

This work was supported by the German Research Foundation DFG (grant numbers MI1242/3-2 to J.M. and OE723/2-1 to R.O.), the GRK1957 “Adipocyte-Brain-Crosstalk,” the TR-SFB296 “LocoTact” (P12, P14, and P17 to J.M. and L.S.), and Research Unit FOR-2558 “TraceAge” (Scho 849/6-2 to L.S.).

Supplementary Material

Supplementary Figure S1

Supplementary Figure S2

Supplementary Table S1

References

Yen

. 2001. Physiological and molecular basis of thyroid hormone action. Physiol Rev, 81:1097–1142.

Nilsson

, Fagman

. 2017. Development of the thyroid gland. Development, 144:2123–2140.

de Escobar

, Obregon

, del Rey

. 2007. Iodine deficiency and brain development in the first half of pregnancy. Public Health Nutr, 10:1554–1570.

Harder

, Dudazy-Gralla

, Muller-Fielitz

, Hjerling Leffler

, Vennstrom

, Heuer

, Mittag

. 2018. Maternal thyroid hormone is required for parvalbumin neurone development in the anterior hypothalamic area. J Neuroendocrinol, 30:e12573.

Korevaar

TIM

, Medici

, Visser

, Peeters

. 2017. Thyroid disease in pregnancy: new insights in diagnosis and clinical management. Nat Rev Endocrinol, 13:610–622.

Moog

, Entringer

, Heim

, Wadhwa

, Kathmann

, Buss

. 2017. Influence of maternal thyroid hormones during gestation on fetal brain development. Neuroscience, 342:68–100.

Feldt-Rasmussen

, Effraimidis

, Klose

. 2021. The hypothalamus-pituitary-thyroid (HPT)-axis and its role in physiology and pathophysiology of other hypothalamus-pituitary functions. Mol Cell Endocrinol, 525:111173.

Udovcic

, Pena

, Patham

, Tabatabai

, Kansara

. 2017. Hypothyroidism and the heart. Methodist Debakey Cardiovasc J, 13:55–59.

Klein

. 2011. Thyroid and the heart: the intimacy is strained. Thyroid, 21:469–470.

10.

Dillmann

. 2010. Cardiac hypertrophy and thyroid hormone signaling. Heart Fail Rev, 15:125–132.

11.

Danzi

, Klein

. 2020. Thyroid abnormalities in heart failure. Heart Fail Clin, 16:1–9.

12.

O'Shea

, Williams

. 2002. Insight into the physiological actions of thyroid hormone receptors from genetically modified mice. J Endocrinol, 175:553–570.

13.

Chatonnet

, Guyot

, Benoit

, Flamant

. 2013. Genome-wide analysis of thyroid hormone receptors shared and specific functions in neural cells. Proc Natl Acad Sci U S A, 110:E766–E775.

14.

Cooper

, Pearce

. 2017. Subclinical hypothyroidism and hypothyroxinemia in pregnancy—still no answers. N Engl J Med, 376:876–877.

15.

Casey

, Thom

, Peaceman

, Varner

, Sorokin

, Hirtz

, Reddy

, Wapner

, Thorp

Jr , Saade

, Tita

, Rouse

, Sibai

, Iams

, Mercer

, Tolosa

, Caritis

, VanDorsten

, Eunice Kennedy Shriver National Institute of Child

Health

, Human Development Maternal-Fetal Medicine Units

Network

. 2017. Treatment of subclinical hypothyroidism or hypothyroxinemia in pregnancy. N Engl J Med, 376:815–825.

16.

McLeod

, McIntyre

. 2010. Subclinical hypothyroidism and related biochemical entities in pregnancy: implications and management. Obstet Med, 3:139–144.

17.

Heindel

, Zoeller

. 2003. Thyroid hormone and brain development: translating molecular mechanisms to population risk. Thyroid, 13:1001–1004.

18.

de Escobar

, Ares

, Berbel

, Obregon

, del Rey

. 2008. The changing role of maternal thyroid hormone in fetal brain development. Semin Perinatol, 32:380–386.

19.

Rytter

, Andersen

, Bech

, Halldorsson

, Henriksen

, Laurberg

, Olsen

. 2016. Maternal thyroid function in pregnancy may program offspring blood pressure, but not adiposity at 20 y of age. Pediatr Res, 80:7–13.

20.

Mittag

, Lyons

, Sallstrom

, Vujovic

, Dudazy-Gralla

, Warner

, Wallis

, Alkemade

, Nordstrom

, Monyer

, Broberger

, Arner

, Vennstrom

. 2013. Thyroid hormone is required for hypothalamic neurons regulating cardiovascular functions. J Clin Investig, 123:509–516.

21.

Tinnikov

, Nordstrom

, Thoren

, Kindblom

, Malin

, Rozell

, Adams

, Rajanayagam

, Pettersson

, Ohlsson

, Chatterjee

, Vennstrom

. 2002. Retardation of post-natal development caused by a negatively acting thyroid hormone receptor alpha1. EMBO J, 21:5079–5087.

22.

Renko

, Hoefig

, Hiller

, Schomburg

, Kohrle

. 2012. Identification of iopanoic acid as substrate of type 1 deiodinase by a novel nonradioactive iodide-release assay. Endocrinology, 153:2506–2513.

23.

Flohe

, Gunzler

. 1984. Assays of glutathione peroxidase. Methods Enzymol, 105:114–121.

24.

Hughes

, Fedirko

, Jenab

, Schomburg

, Meplan

, Freisling

, Bueno-de-Mesquita

, Hybsier

, Becker

, Czuban

, Tjonneland

, Outzen

, Boutron-Ruault

, Racine

, Bastide

, Kuhn

, Kaaks

, Trichopoulos

, Trichopoulou

, Lagiou

, Panico

, Peeters

, Weiderpass

, Skeie

, Dagrun

, Chirlaque

, Sanchez

, Ardanaz

, Ljuslinder

, Wennberg

, Bradbury

, Vineis

, Naccarati

, Palli

, Boeing

, Overvad

, Dorronsoro

, Jakszyn

, Cross

, Quiros

, Stepien

, Kong

, Duarte-Salles

, Riboli

, Hesketh

. 2015. Selenium status is associated with colorectal cancer risk in the European prospective investigation of cancer and nutrition cohort. Int J Cancer, 136:1149–1161.

25.

Dudazy-Gralla

, Nordstrom

, Hofmann

, Meseh

, Schomburg

, Vennstrom

, Mittag

. 2013. Identification of thyroid hormone response elements in vivo using mice expressing a tagged thyroid hormone receptor alpha1. Biosci Rep, 33:e00027.

26.

Wallis

, Sjogren

, van Hogerlinden

, Silberberg

, Fisahn

, Nordstrom

, Larsson

, Westerblad

, Morreale de Escobar

, Shupliakov

, Vennstrom

. 2008. Locomotor deficiencies and aberrant development of subtype-specific GABAergic interneurons caused by an unliganded thyroid hormone receptor alpha1. J Neurosci, 28:1904–1915.

27.

Johann

, Cremer

, Fischer

, Heine

, Pensado

, Resch

, Nock

, Virtue

, Harder

, Oelkrug

, Astiz

, Brabant

, Warner

, Vidal-Puig

, Oster

, Boelen

, Lopez

, Heeren

, Dalley

, Backes

, Mittag

. 2019. Thyroid-hormone-induced browning of white adipose tissue does not contribute to thermogenesis and glucose consumption. Cell Rep, 27:3385–3400 e3383.

28.

Venero

, Guadano-Ferraz

, Herrero

, Nordstrom

, Manzano

, de Escobar

, Bernal

, Vennstrom

. 2005. Anxiety, memory impairment, and locomotor dysfunction caused by a mutant thyroid hormone receptor alpha1 can be ameliorated by T3 treatment. Genes Dev, 19:2152–2163.

29.

Zoeller

, Tan

, Tyl

. 2007. General background on the hypothalamic-pituitary-thyroid (HPT) axis. Crit Rev Toxicol, 37:11–53.

30.

Vujovic

, Nordstrom

, Gauthier

, Flamant

, Visser

, Vennstrom

, Mittag

. 2009. Interference of a mutant thyroid hormone receptor alpha1 with hepatic glucose metabolism. Endocrinology, 150:2940–2947.

31.

Mittag

, Behrends

, Hoefig

, Vennstrom

, Schomburg

. 2010. Thyroid hormones regulate selenoprotein expression and selenium status in mice. PLoS One, 5:e12931.

32.

Mittag

, Behrends

, Nordstrom

, Anselmo

, Vennstrom

, Schomburg

. 2012. Serum copper as a novel biomarker for resistance to thyroid hormone. Biochem J, 443:103–109.

33.

Mittag

, Davis

, Vujovic

, Arner

, Vennstrom

. 2010. Adaptations of the autonomous nervous system controlling heart rate are impaired by a mutant thyroid hormone receptor-alpha1. Endocrinology, 151:2388–2395.

34.

Klein

, Danzi

. 2007. Thyroid disease and the heart. Circulation, 116:1725–1735.

35.

Danzi

, Klein

. 2002. Thyroid hormone-regulated cardiac gene expression and cardiovascular disease. Thyroid, 12:467–472.

36.

Morkin

. 2000. Control of cardiac myosin heavy chain gene expression. Microsc Res Tech, 50:522–531.

37.

Small

, Krieg

. 2004. Molecular regulation of cardiac chamber-specific gene expression. Trends Cardiovasc Med, 14:13–18.

38.

Liang

, Song

, Ghosh

, Morien

, Qin

, Mahmood

, Fujiwara

, Igarashi

, Nagase

, Held

. 2011. Genome-wide survey reveals dynamic widespread tissue-specific changes in DNA methylation during development. BMC Genomics, 12:231.

39.

, Hariharan

, Gorkin

, Dickel

, Luo

, Castanon

, Nery

, Lee

, Zhao

, Huang

, Williams

, Trout

, Amrhein

, Fang

, Chen

, Li

, Visel

, Pennacchio

, Ren

, Ecker

. 2020. Spatiotemporal DNA methylome dynamics of the developing mouse fetus. Nature, 583:752–759.

40.

Jansen

, Korevaar

TIM

, Mulder

, White

, Muetzel

, Peeters

, Tiemeier

. 2019. Maternal thyroid function during pregnancy and child brain morphology: a time window-specific analysis of a prospective cohort. Lancet Diabetes Endocrinol, 7:629–637.

41.

Springer

, Jiskra

, Limanova

, Zima

, Potlukova

. 2017. Thyroid in pregnancy: from physiology to screening. Crit Rev Clin Lab Sci, 54:102–116.

42.

Godoy

, Korevaar

, Peeters

, Hofman

, de Rijke

, Bongers-Schokking

, Tiemeier

, Jaddoe

, Gaillard

. 2014. Maternal thyroid hormones during pregnancy, childhood adiposity and cardiovascular risk factors: the Generation R Study. Clin Endocrinol, 81:117–125.

43.

Cappola

, Desai

, Medici

, Cooper

, Egan

, Sopko

, Fishman

, Goldman

, Cooper

, Mora

, Kudenchuk

, Hollenberg

, McDonald

, Ladenson

. 2019. Thyroid and cardiovascular disease: research agenda for enhancing knowledge, prevention, and treatment. Thyroid, 29:760–777.

44.

Tapia-Martinez

, Franco-Colin

, Ortiz-Butron

, Pineda-Reynoso

, Cano-Europa

. 2019. Hypothyroid offspring replacement with euthyroid wet nurses during lactation improves thyroid programming without modifying metabolic programming. Arch Endocrinol Metab, 63:199–207.

45.

Tapia-Martinez

, Torres-Manzo

, Franco-Colin

, Pineda-Reynoso

, Cano-Europa

. 2019. Maternal thyroid hormone deficiency during gestation and lactation alters metabolic and thyroid programming of the offspring in the adult stage. Horm Metab Res, 51:381–388.

46.

van Gucht

ALM

, Moran

, Meima

, Visser

, Chatterjee

, Visser

, Peeters

. 2017. Resistance to thyroid hormone due to heterozygous mutations in thyroid hormone receptor alpha. Curr Top Dev Biol, 125:337–355.

47.

Tylki-Szymanska

, Acuna-Hidalgo

, Krajewska-Walasek

, Lecka-Ambroziak

, Steehouwer

, Gilissen

, Brunner

, Jurecka

, Rozdzynska-Swiatkowska

, Hoischen

, Chrzanowska

. 2015. Thyroid hormone resistance syndrome due to mutations in the thyroid hormone receptor alpha gene (THRA). J Med Genet, 52:312–316.

48.

Fonseca

, Fernandes

, McAninch

, Bocco

, Abdalla

, Ribeiro

, Mohacsik

, Fekete

, Li

, Xing

, Wang

, Gereben

, Bianco

. 2015. Perinatal deiodinase 2 expression in hepatocytes defines epigenetic susceptibility to liver steatosis and obesity. Proc Natl Acad Sci U S A, 112:14018–14023.

49.

Chanoine

, Neve

, Wu

, Vanderpas

, Bourdoux

. 2001. Selenium decreases thyroglobulin concentrations but does not affect the increased thyroxine-to-triiodothyronine ratio in children with congenital hypothyroidism. J Clin Endocrinol Metab, 86:1160–1163.

50.

Korevaar

, Chaker

, Jaddoe

, Visser

, Medici

, Peeters

. 2016. Maternal and birth characteristics are determinants of offspring thyroid function. J Clin Endocrinol Metab, 101:206–213.

51.

Wallis

, Dudazy

, van Hogerlinden

, Nordstrom

, Mittag

, Vennstrom

. 2010. The thyroid hormone receptor alpha1 protein is expressed in embryonic postmitotic neurons and persists in most adult neurons. Mol Endocrinol, 24:1904–1916.

52.

Herrmann

, Harder

, Oelkrug

, Chen

, Gachkar

, Nock

, Resch

, Korkowski

, Heuer

, Mittag

. 2020. Central hypothyroidism impairs heart rate stability and prevents thyroid hormone-induced cardiac hypertrophy and pyrexia. Thyroid, 30:1205–1216.

53.

Hoefig

, Harder

, Oelkrug

, Meusel

, Vennstrom

, Brabant

, Mittag

. 2016. Thermoregulatory and cardiovascular consequences of a transient thyrotoxicosis and recovery in male mice. Endocrinology, 157:2957–2967.

54.

Silva

, Bertasso

, Pietrobon

, Lopes

, Santos

, Peixoto-Silva

, Carvalho

, Claudio-Neto

, Manhaes

, Cabral

, Kluck

GEG

, Atella

, Oliveira

, Moura

, Lisboa

. 2019. Effects of maternal bisphenol A on behavior, sex steroid and thyroid hormones levels in the adult rat offspring. Life Sci, 218:253–264.

55.

Kohrle

, Schomburg

, Drescher

, Fekete

, Bauer

. 1995. Rapid stimulation of type I 5'-deiodinase in rat pituitaries by 3,3',5-triiodo-L-thyronine. Mol Cell Endocrinol, 108:17–21.

56.

Hernandez

, Stohn

. 2018. The type 3 deiodinase: epigenetic control of brain thyroid hormone action and neurological function. Int J Mol Sci, 19:1804.

57.

Anselmo

, Scherberg

, Dumitrescu

, Refetoff

. 2019. Reduced sensitivity to thyroid hormone as a transgenerational epigenetic marker transmitted along the human male line. Thyroid, 29:778–782.

58.

Martinez

, Duarte

, Stohn

, Karaczyn

, Wu

, DeMambro

, Hernandez

. 2020. Thyroid hormone influences brain gene expression programs and behaviors in later generations by altering germ line epigenetic information. Mol Psychiatry, 25:939–950.

59.

Kuratomi

, Kuratomi

, Kuwahara

, Ishii

, Nakao

, Saito

, Takano

. 2007. NRSF regulates the developmental and hypertrophic changes of HCN4 transcription in rat cardiac myocytes. Biochem Biophys Res Commun, 353:67–73.

60.

Grigore

, Ojeda

, Alexander

. 2008. Sex differences in the fetal programming of hypertension. Gend Med, 5 Suppl A:S121–S132.

61.

Mai

, Janier

, Allioli

, Quignodon

, Chuzel

, Flamant

, Samarut

. 2004. Thyroid hormone receptor alpha is a molecular switch of cardiac function between fetal and postnatal life. Proc Natl Acad Sci U S A, 101:10332–10337.

62.

Korevaar

, Medici

, de Rijke

, Visser

, de Muinck Keizer-Schrama

, Jaddoe

, Hofman

, Ross

, Visser

, Hooijkaas

, Steegers

, Tiemeier

, Bongers-Schokking

, Visser

, Peeters

. 2013. Ethnic differences in maternal thyroid parameters during pregnancy: the Generation R study. J Clin Endocrinol Metab, 98:3678–3686.

63.

Poppe

, Bisschop

, Fugazzola

, Minziori

, Unuane

, Weghofer

. 2021. 2021 European Thyroid Association Guideline on thyroid disorders prior to and during assisted reproduction. Eur Thyroid J, 9:281–295.

64.

Gore

, Chappell

, Fenton

, Flaws

, Nadal

, Prins

, Toppari

, Zoeller

. 2015. EDC-2: The Endocrine Society's Second Scientific Statement on Endocrine-Disrupting Chemicals. Endocr Rev, 36:E1–E150.