Skip to main content
Endocrinology logoLink to Endocrinology
. 2025 Apr 30;166(7):bqaf085. doi: 10.1210/endocr/bqaf085

Serotonergic and Chemosensory Brain Areas and Sensory Ganglia Expressing Type 3 Deiodinase Mapped With Dio3Cre drivers

Ye Liu 1, Lily Ng 2, Chengyu Liu 3, Douglas Forrest 4,
PMCID: PMC12092337  PMID: 40302251

Abstract

Thyroid hormone (triiodothyronine, T3) promotes neurodevelopment but under strict control because unconstrained exposure to T3 impairs brain and sensory functions. Thyroid hormone–inactivating type 3 deiodinase, encoded by Dio3, critically limits T3 signaling and controls diverse neural functions. Accordingly, understanding the cellular basis of T3 action requires identification of Dio3-expressing cell types but this is difficult because of low level, transient expression within the complexity of the nervous system. Here, we derived a knock-in Dio3Cre driver that sensitively labels Dio3-expressing cells in male and female mice. In this anatomical study, we identified Dio3 expression in the immature amygdala and other brain regions associated with emotion and motivation, and in serotonergic raphe nuclei, which influence many behavioral and physiological systems. Notably, expression in circumventricular organs, including the chemosensory subfornical organ and organum vasculosum laminae terminalis, suggested regulation of centers that lack a blood–brain barrier and directly sense signaling factors in the circulation. Expression in trigeminal, dorsal root, cochleovestibular, and other sensory ganglia highlighted contributions to sensory pathways. Although Dio3 expression declines during maturation, a conditional Dio3CreERt2 driver revealed neurons with T3-inducible expression in the adult brain, suggesting ongoing homeostatic functions. These Cre drivers indicate strategically located neuronal groups for control of T3 signaling in behavioral, chemosensory and sensory systems.

Keywords: thyroid hormone, DIO3, circumventricular organ, sensory ganglion, raphe nuclei, Cre driver


Thyroid hormone is necessary for the maturation and function of the nervous system. Developmental thyroid disorders are associated with cognitive, neuroendocrine, and sensory impairments (1-5). In addition to adequate provision of the active form of thyroid hormone, triiodothyronine (T3), and its precursor thyroxine (T4) in the circulation, a key concept that has emerged is the requirement for mechanisms that limit T3 signaling within tissues to regulate the response and prevent damage from unconstrained T3 action (6, 7). A critical factor is type 3 deiodinase (DIO3), which limits T3 signaling by depletion of both T3 and T4 (8-10). Dio3-deficient mice display a range of defects in locomotor activity, behavior (11, 12), neuroendocrine function (13-15), and sensory functions, including hearing and vision (6, 16), and have developmental changes in expression of a T3 reporter transgene in the brain, consistent with Dio3 regulating T3 availability in some regions (17). However, the cellular mechanisms that constrain T3 signaling remain unclear because of difficulties in defining Dio3-expressing cell types due to low and transient Dio3 expression profiles (18, 19).

Type 3 deiodination enzyme activity displays a peak at fetal stages then a steep postnatal decline in several brain regions (11, 13, 18, 19), the retina (16, 20) and cochlea (6) in rodents, suggesting dynamic roles during developmental transitions and the maturation of neural function (21). Similar trends occur in the human fetal brain (22). Although Dio3 expression generally declines at older ages, adult-onset deletions indicate a functional requirement for Dio3 in mature brain gene expression and locomotor activity (23, 24). In some tissues where DIO3 expression may reach a detection threshold, immunohistochemical signals have been reported in portions of aged human hypothalamus (25) and in situ hybridization signals have been detected for Dio3 mRNA in regions of the immature rodent brain (11, 26) and retina (16) but without detailed cellular resolution. It is likely, given the limitations of these mapping approaches, that other Dio3-expressing cell groups remain unrecognized.

Identification of Dio3-expressing cell types, including rarer cell populations, is imperative for understanding the cellular control of T3 action in the nervous system. Therefore, in a new approach, we derived 2 knock-in Cre drivers (nonconditional and conditional) that activate fluorescent labeling of Dio3-positive cell types with high resolution and sensitivity. In this anatomical study, Cre-mediated labeling revealed Dio3 expression in immature neuronal populations including in the amygdala, nucleus accumbens (NAc), cerebellum, hypothalamus, brainstem serotonergic nuclei, and sensory ganglia. Intriguingly, expression in chemosensory circumventricular organs (CVOs) suggests a role for Dio3 in specialized centers that have an incomplete blood–brain barrier and directly sense circulating signals involved in body fluid regulation and other homeostatic functions. These Cre driver models allow sensitive detection of Dio3 expression and indicate neuronal groups in strategic locations for control of T3 signaling in systems involved in mood and anxiety, chemosensory function, and sensory function.

Materials and Methods

Dio3 CreERt2 and Dio3Cre Knock-in Alleles

Conditional Dio3CreERt2 driver

A targeting construct with a cassette expressing tamoxifen (TAM)-dependent CreERt2 recombinase (27) was used to replace the endogenous Dio3 coding sequence by homologous recombination in C57BL/6 mouse embryonic stem cells (Ozgene Pty Ltd, Perth, WA, Australia). The construct included 5′ and 3′ flanking homology arms of 3.03 and 5.38 kb, respectively, and a neomycin resistance selection marker. Founder mice derived from recombinant embryonic stem cell clones were crossed with Rosa26Flpe deleter mice (C57BL/6 background) to remove the selection marker and establish germline transmission of the Dio3CreERt2 allele. Rosa26Flpe was removed by crossing with C57BL/6J mice (Jackson Lab, RRID:IMSR_JAX: 000664). Targeting was confirmed by Southern blot and sequencing analysis of the allele. The Dio3CreERt2 allele was crossed into Rosa26Ai6/Ai6 reporter mice (28) (Ai6, Jackson Lab, RRID:IMSR_JAX:007906, C57BL/6J background) to derive Dio3+/CreERt2; Rosa26Ai6/Ai6 progeny that express ZsGreen fluorescent protein specifically in cells where CreERt2 is expressed and when activated by TAM.

Nonconditional Dio3Cre driver

The Dio3Cre mouse line was derived by CRISPR/Cas9 editing of the Dio3CreERt2 allele to delete ERt2 sequences using standard procedures (29) at the NHLBI Transgenic Core at NIH. An upstream sgRNA targeting the junction between Cre and ERt2 domains and a downstream sgRNA targeting the ERt2 C-terminus were designed to delete most (>91%) ERt2 sequences; a single-stranded DNA oligonucleotide incorporating triple stop codons at the end of the Cre domain was included. Dio3+/CreERt2 males were crossed with superovulated females to obtain zygotes for electroporation with Cas9 protein, sgRNAs, and oligonucleotide donor. Zygotes were cultured overnight to develop into 2-cell stage embryos before implanting into pseudopregnant foster mothers, yielding 2 founders with deleted ERt2 sequences. The founders had lost the C-terminal 6 codons of Cre, which was restored by a second CRISPR-mediated editing step. A founder with a restored C-terminus and 4 codon extension (Dio3Cre allele) was backcrossed with C57BL/6J mice for 3 generations to deplete possible off-target alterations, and then was crossed with Rosa26Ai6 reporter mice to generate progeny for analysis.

Sequences of CRISPR reagents are as follows: upstream sgRNA, 5-AGG GGC AAT GGT GCG CCT GC-3′; downstream sgRNA, 5′-TAG AGC CCG CAG TGG CCA AG-3′; oligonucleotide donor, 5′-CCT GGA TAG TGA AAC AGG GGC AAT GGT GCG CCT GCT GGA AGA TGG CGA TTA ATA GTG AGA AGT TCC TAT TCC GAA GTT CCT ATT CTC TAG TAA GTA TAG GAA CTT C-3′. PCR primers for screening and sequencing progeny genomic DNA: Cre-680F, 5′-CTC TGG TGT AGC TGA TGA TC-3′ and Dio3-3UTR-R, 5′-TTT GCA CGT GGG CTT CGA AG-3′, which generated 1.4 kb and 0.44 kb amplicons for the CreErt2 and Cre (ERt2-deleted) allele, respectively.

Genotyping primers

For Dio3CreERt2 the genotyping primers were WT-F, 5′-CTA CAA CAA GGT GCA CCT GG-3′, Ert2-F, 5′-TCC CAC ATC AGG CAC ATG AG-3′, and cm-R, 5′-GAG TCT CAA GTT AGC CAG AC-3′, generating 382 bp (WT) and 488 bp (CreERt2) amplicons; for Dio3Cre, WT-F, Cre-F, 5′-AGC TGG TGG CTG GAC CAA TG-3′, and cm-R, generating 382 bp (WT) and 278 bp (Cre) amplicons.

Experimental strategy

Dio3 +/Cre males were crossed with Rosa26Ai6/Ai6 females, maintaining Cre driver and Ai6 reporter alleles in separate parents to prevent confounding germline activation of Rosa26Ai6 by Cre, which can give widespread, nonrepresentative expression, as reported for other drivers with early-onset expression (30). Using this crossing strategy, we detected no germline activation whereas up to 30% of progeny displayed germline activation if Cre and Ai6 reporter alleles were present in the same parent. To avoid variations arising in progeny from imprinting of the Dio3 gene, male parents were always used as the source of the Cre or CreERt2 driver. For tissue analyses, male and female mice were analyzed and did not show obvious differences in Cre-mediated labeling. Most figures show representative data for at least 3 males (see legends).

Tamoxifen and T3 treatment

A 10 mg/mL stock solution of TAM (MilliporeSigma, cat# T5648) was prepared in corn oil (MilliporeSigma, cat# C8267), as described (31).

Adult (8 weeks old) Dio3+/CreERt2;Rosa26Ai6/Ai6 mice were injected (intraperitoneally) with TAM at a dose of 2.3 mg/30 g body weight daily for 3 consecutive days, then analyzed 7 days after the first injection. Controls included treatment with vehicle (corn oil) only in Dio3+/CreERt2;Rosa26Ai6/Ai6 mice. Additional controls included TAM treatment of Rosa26Ai6/Ai6 reporter mice without the presence of CreERt2. For T3 treatment, mice were injected (intraperitoneally) daily with a solution of T3 (MilliporeSigma Cat# T6397) (5.0 μg/30 g body weight), 30 minutes prior to TAM injection, for 3 days (32). Control groups received saline vehicle instead of T3. All studies followed approved protocols at the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) at the National Institutes of Health (NIH).

Immunostaining and Quantification

Tissues were fixed by immersion in 4% paraformaldehyde (PFA) at 4 °C for 5 hours, then cryoprotected in 30% sucrose overnight at 4 °C. Samples were embedded in optimum cutting temperature compound (Tissue-Tek) and cryosectioned at 20 μm thickness. For immunostaining, sections were blocked with 5% donkey serum in phosphate-buffered saline with 0.01% Triton X-100 at room temperature (RT) for 1 hour, incubated with primary antibodies (or lectin) overnight at 4 °C, and then further incubated with secondary antibodies at room temperature for 1 hour. Fluorescent images were acquired using a Nikon Ti2 inverted microscope for widefield imaging or a Nikon-SoRa spinning disk confocal microscope using 20× and 60× objectives. Fluorescent cell numbers were quantified using ImageJ software (RRID:SCR-003070). Groups included ≥3 mice. For each sample, 2 representative sections were used for quantification of the area of interest.

Primary antibodies (target antigen, type, dilution, source, RRID) were the following: arginine vasopressin (AVP), rabbit polyclonal, 1:300 (Millipore Cat# AB1565, RRID:AB_90782); beta III Tubulin (Tuj1), rabbit polyclonal, 1:300 (Abcam Cat# ab18207, RRID:AB_444319); Calbindin D-28 K, rabbit polyclonal, 1:300 (Millipore Cat# AB1778, RRID:AB_2068336); Ccnd1 (Cyclin D1), rabbit monoclonal, 1:300 (Lab Vision Cat# RM-9104-S0, RRID:AB_149914); corticotropin-releasing factor (CRF), rabbit monoclonal, 1:200 (Abcam Cat# ab272391, RRID:AB_3676151); NeuN, rabbit polyclonal, 1:400 (Millipore Cat# ABN78, RRID:AB_10807945); Neurofilament H (200 kDa), rabbit polyclonal, 1:400 (Millipore Cat# AB1982, RRID:AB_2313731); Tryptophan hydroxylase 2, rabbit polyclonal, 1:300 (Millipore Cat# ABN60, RRID:AB_10806898). Secondary antibodies were Alexa Fluor 647-conjugated anti-rabbit IgG, donkey polyclonal, 1:300 (Thermo Fisher Scientific Cat# A-31573, RRID:AB_2536183); Alexa Fluor 568-conjugated anti-rabbit IgG, donkey polyclonal, 1:300 (Thermo Fisher Scientific Cat# A10042, RRID:AB_2534017). Lectin: Alexa Fluor 647-conjugated Isolectin GS-IB4 (a vascular stain), 1:400 (Thermo Fisher Scientific Cat# I21412, RRID:AB_2314662).

Statistical Analysis

Statistical analysis was conducted using GraphPad Prism version 10 (GraphPad Software). Numerical data are presented as mean ± SD. Statistical significance was determined using 1-way analysis of variance followed by Tukey post hoc tests, with significance set at *P < .05.

Results

Dio3 Cre and Dio3CreERt2 Drivers for Identifying Dio3-Expressing Cell Types

To facilitate sensitive analysis of Dio3 expression, we derived 2 knock-in Cre driver mouse lines that express a nonconditional (Cre) and a conditional (CreERt2) recombinase (27) from the endogenous Dio3 gene (Fig. 1A and 1B). These Cre recombinases expressed from the natural Dio3 promoter and control sequences trigger cell-specific expression of ZsGreen fluorescent protein in Rosa26Ai6 reporter mice (28). Each Cre driver allele offers certain advantages. The nonconditional Dio3Cre driver optimizes labeling of cell lineages from early stages and avoids both inefficiency and toxicity associated with the administration of the TAM used to activate conditional CreERt2 recombinase, especially at embryonic stages (33). In contrast, the conditional Dio3CreERt2 driver allows selective (TAM-dependent) labeling at older ages without complications due to cumulative developmental labeling that can arise with a nonconditional driver.

Figure 1.

Figure 1.

Dio3 expression in embryonic tissues detected with Cre driver alleles. (A) Diagram of Dio3Cre and Dio3CreERt2 knock-in alleles in the endogenous Dio3 gene. Cre and CreERt2 cassettes are fused at the Dio3 translation start codon and replace the Dio3 coding region. 3′ut, 3′ untranslated region. (B) PCR genotyping of progeny carrying Dio3Cre and Dio3CreERt2 alleles; primers noted in A; size markers (in bp) on right. (C) Dio3Cre activates expression of ZsGreen (ZsG) fluorescent protein from a Rosa26Ai6 reporter. Dio3-expressing cells (ZsG+) detected in sagittal sections (lateral to midline) in the head and cervical region of a representative Dio3Cre;Ai6 embryo (paternal origin Cre allele). A control Rosa26Ai6/+ littermate without Cre (right) has no ZsG signals. Tissue counterstain, Hoechst. Scale bar, 1 mm for both images. (D, E) Dio3 expression (ZsG) in retinal progenitor cells (RPCs) in a vertical section of the eye (D). ZsG+ cells colocalized with Ccnd1, a marker of RPCs (E).

Abbreviations: BAT, brown adipose tissue; ChP, choroid plexus; DRG, dorsal root ganglia; GCL, ganglion cell layer; Hyp, hypothalamus; mand, mandible; max, maxilla; men, meninges; NBL, neuroblast layer of retina; Ra, raphe nuclei; sPall, subpallium; SpC, spinal cord.

The Dio3CreERt2 line was derived by homologous recombination in embryonic stem cells and the Dio3Cre line was then derived by CRISPR/Cas9-mediated deletion of the ERt2 domain of the Dio3CreERt2 allele. Subsequent studies were based on mice heterozygous for the driver allele: either Dio3Cre/+;Rosa26Ai6/+ (abbreviated in figures as Dio3Cre;Ai6) or Dio3CreERt2/+;Rosa26Ai6/Ai6 (abbreviated in figures as Dio3CreERt2;Ai6). To preclude variability due to imprinting of the Dio3 gene (34), all progeny for analysis carried a paternally derived Cre allele obtained by crossing males carrying the driver allele (Dio3Cre/+ or Dio3CreERt2/+) with females wild type for Dio3. Expression of the Dio3 allele of paternal origin is generally higher than that of maternal origin in most tissues (34).

Figure 1C shows an overview of specific labeling in the head and cervical region detected with the Dio3Cre driver at embryonic day 15.5 (E15.5). ZsGreen (ZsG) signal was localized in brain regions, including the subpallium, which gives rise to centers involved in emotion, cognition and motor control (including parts of the amygdala and basal ganglia), and hypothalamus. Expression was also detected in the tongue, craniofacial skeletal structures, including the calvaria, and brown adipose tissue. Most of these general Dio3 expression patterns are consistent with previous in situ hybridization (26, 35) and enzyme activity data in rodents (13, 18, 36), thus validating the Cre-mediated mapping approach. ZsG signals were not detected in control Rosa26Ai6/+ mice without a Cre driver. The model was further validated by analysis of the late embryonic retina during the known peak of DIO3 activity (16, 20). Consistent with previous detection of Dio3 mRNA in undifferentiated cells in the neuroblastic layer of the retina by in situ hybridization (16), ZsG signals were detected in cells that costained with Ccnd1, a marker of retinal progenitor cells (37) in the retinal neuroblastic layer (Fig. 1D and 1E).

The high sensitivity of the Cre-mediated mapping approach revealed new and detailed insights into Dio3 expression, including in previously unrecognized regions such as the brainstem raphe nuclei and sensory ganglia (eg, dorsal root ganglia) (Fig. 1C), which are considered in the next sections.

Dio3 Expression in Serotonergic and Other Neuronal Groups in the Central Nervous System

To map Dio3 expression with cellular resolution, we first screened a series of coronal brain/head sections (rostral to caudal) of Dio3Cre;Ai6 mice, at a neonatal stage when DIO3 activity is typically at or around peak levels, before expression progressively declines (Fig. 2A-2H). ZsG signals were detected in forebrain regions including the NAc, caudate putamen, bed nucleus of the stria terminalis (BST), amygdala, and median preoptic nucleus (MnPO). These results are in accord with, and extend, previous data based on in situ hybridization in neonatal rat brain, which tended to identify regions with high expression (26). Expression in several of these regions including the BST, NAc, and amygdala implicates Dio3 in centers involved in motivation and emotion (38, 39). Also consistent with previous in situ hybridization data, we detected Dio3 expression in the Purkinje neuron layer of the cerebellum (11) (Fig. 2H). Cre-mediated mapping also localized new sites of Dio3 expression including in the prelimbic cortex (Fig. 2A). Dio3 expression was detected in several hypothalamic regions, including the preoptic area, dorsomedial nucleus, ventromedial nucleus, and arcuate nucleus (Fig. 2E-2G), consistent with reports of DIO3 enzyme activity in the rodent hypothalamus (13, 18).

Figure 2.

Figure 2.

Dio3 expression in the neonatal brain identified with the Dio3Cre driver. (A-I) Representative coronal sections of brain (rostral to caudal) indicating major regions of Dio3 expression (ZsG) in Dio3Cre;Ai6 mice at P1. (I), Rosa26Ai6/+ control brain without Cre, displayed no ZsG+ signals (approximately equivalent region as in E). Tissue counterstain, Hoechst. At least 3 males examined at P1. Scale bar in panel I is the same for B-H.

Abbreviations: AHN, anterior hypothalamic nucleus; Amg, amygdala; Aq, cerebral aqueduct; ARH, arcuate hypothalamic nucleus; BST, bed nucleus of the stria terminalis; CP, caudate-putamen; DMH, dorsomedial nucleus of the hypothalamus; fi, fimbria; Gen, geniculate group (dorsal thalamus); Hyp, hypothalamus; LHA, lateral hypothalamic area; IL, infralimbic cortex; Lv, lateral ventricle; MnPO, median preoptic nucleus; MPO, medial preoptic area; NAc, nucleus accumbens; PNL, Purkinje neuron layer; PL, prelimbic cortex; RPA, nucleus raphe pallidus; RM, nucleus raphe magnus; SCN, suprachiasmatic nucleus; SG, spiral ganglion; st, stria terminalis; TG, trigeminal ganglion; TU, tuberal nucleus; vCA1, ventral CA1 (of hippocampus); 3v, third ventricle; 4v, fourth ventricle.

A striking pattern of expression was detected in the brainstem raphe nuclei. Figure 2H shows an overview of the hindbrain with signals detected in the pontine raphe magnus and raphe pallidus nuclei. The raphe nuclei, situated along the midline of the brainstem, are the major source of serotonin in the brain and influence other regions involved in many behavioral and physiological functions (40, 41). The neuronal identity of Dio3-expressing cells (ZsG+) in the raphe nuclei was demonstrated by colocalization with tryptophan hydroxylase 2, an enzyme expressed in serotonergic neurons (Fig. 3A). Figure 3A shows specific examples of Dio3 expression in neurons of the medullary raphe obscurus and raphe pallidus, located in dorsal and ventral regions, respectively, of the medulla oblongata.

Figure 3.

Figure 3.

Dio3 expression in serotonergic and other neurons in the immature brain. (A, B) Dio3 expression (ZsG) in serotonergic raphe nuclei in the brainstem, labeled by tryptophan hydroxylase 2 (TPH2) in neonatal Dio3Cre;Ai6 mice. (A) overview of the raphe nuclei along the midline of medulla oblongata, showing nucleus raphe obscurus (RO) and nucleus raphe pallidus (RPA). (B) Magnified views of the boxed regions in A, show colocalization of ZsG and TPH2 in raphe nuclei (arrows). (C-E) Dio3 expression (ZsG) in neurons in specific brain regions in mice: prelimbic cortex (C); bed nucleus of the stria terminalis (D); ventral hippocampal CA1 (E), indicated by colocalization with neuronal marker NeuN. PAn, posterior amygdalar nucleus. Arrows indicate examples of double-labeled cells (whitish). (F) Dio3 expression (ZsG) in Purkinje neurons, shown by colocalization with Purkinje neuron marker calbindin. IGL, internal granule layer. (G, H) Dio3 expression (ZsG) in hypothalamic paraventricular nucleus (PVN), indicated by colocalization with secretory neuronal markers corticotropin-releasing factor (CRF, G) and arginine vasopressin (AVP, H). Left panel in G: overview of PVN region; box, approximate area of magnified views in panels on right. Arrows indicate examples of double-labeled cells. Representative images from at least 3 males at each age.

A neuronal identity for Dio3-expressing cells in the brain is sometimes assumed but direct cell staining evidence has been limited because of low Dio3 expression and lack of reliable antibodies for immunostaining. Using Cre-mediated labeling, we demonstrated a neuronal identity for most Dio3-expressing cells (ZsG+) in all brain regions examined, by colocalization with neuronal markers, including NeuN, shown in examples of the prelimbic cortex, BST, and ventral hippocampal CA1 area (Fig. 3C-3E). In other regions, ZsG+ cell types were confirmed as specific neurons by costaining with calbindin for cerebellar Purkinje neurons (Fig. 3F) and CRF and AVP for secretory neurons of the hypothalamic paraventricular nucleus (Fig. 3G and 3H).

Expression of Dio3 in Circumventricular Organs

The CVOs reside in the midline of the brain adjacent to the third and fourth ventricles. CVOs are typically vascularized by permeable (fenestrated) capillaries creating specific, localized gaps in the blood–brain barrier that allow direct sensing of solutes and signaling factors in the circulation, unlike most brain regions where transfer of solutes is restricted by the blood–brain barrier. CVOs can serve sensory (ie, respond to factors in the circulation) or secretory (release factors into the circulation) roles. We detected Dio3 expression in 2 key chemosensory CVOs, the organum vasculosum laminae terminalis (OVLT) and subfornical organ (SFO) on the midline anterior wall of the third ventricle (Fig. 4A), which are thought to monitor osmolality of blood and cerebrospinal fluid, and to control thirst and other homeostatic functions (42, 43). Costaining with NeuN indicated that in both the OVLT and the SFO, Dio3-expressing (ZsG+) cells are neurons (Fig. 4C and 4D), suggesting a role for Dio3 in the neural control of body fluid homeostasis. OVLT and SFO neurons connect to several brain areas including the BST, preoptic area, and hypothalamic nuclei, and accordingly Dio3 may be involved in central control of a variety of homeostatic responses (44).

Figure 4.

Figure 4.

Dio3 expression in neurons of the circumventricular organs (CVOs). (A) Dio3 expression (ZsG) is observed in the sensory OVLT and SFO, but not area postrema (AP), shown in coronal brain sections from P2 Dio3Cre;Ai6 mice. OVLT and SFO are distinguished by dense vascular networks, stained with isolectin B4 (IB4); Hoechst counterstain in all panels in A and B. Dashed line outlines OVLT and AP. (B) In the secretory ME and NH, ZsG signals are detected in nerve terminals extending from hypothalamic neurons. These nerve terminals are in close proximity to portal capillaries (IB4+). No ZsG signals were detected in PG or SCO. (C, D) Magnified views of OVLT (C) and SFO (D) showing Dio3 expression (ZsG) in neurons (NeuN+) in close proximity to the capillary network. (E) High magnification view of the ME (area outlined in B), showing hypothalamic neurosecretory fibers (ZsG) colocalized with CRF. The fibers are juxtaposed to the capillary bed (IB4). Arrows indicate double-labeled ZsG + CRF + nerve fibers. At least 3 males examined at each age.

Abbreviations: AVPV, anteroventral periventricular nucleus; CVO, circumventricular organ; OVLT, organum vasculosum laminae terminalis; SFO, subfornical organ; AP, area postrema; ME, median eminence; NH, neurohypophysis; PG, pineal gland; SCO, subcommissural organ.

The median eminence (ME) and adjacent neurohypophysis (posterior pituitary) are CVO structures housing nerve terminals that originate from secretory neurons of the hypothalamus. ZsG signals were present in nerve terminal regions of both CVOs (Fig. 4B). Nerve terminals in the ME release hypothalamic factors (such as CRF) into the portal capillary system which in turn controls release of anterior pituitary hormones into the circulation. Nerve terminals in the neurohypophysis release posterior pituitary hormones including AVP (also called antidiuretic hormone) into the circulation. Dio3 expression (ZsG) was detected in both classes of hypothalamic secretory neurons, indicated by costaining with AVP and CRF in neuronal soma in the paraventricular nucleus of the hypothalamus (see Fig. 3G and 3H). CRF and ZsG signals were further colocalized under high magnification, in nerve terminals in the ME (Fig. 4E).

Dio3 expression was not observed in the area postrema, another sensory CVO in mammals, nor in the pineal gland and subcommissural organ (Fig. 4A and 4B).

Expression of Dio3 in Sensory Ganglia in the Peripheral Nervous System

T3 promotes sensory development and has major roles in sensory organs, including the cochlea and retina (6, 16). Extending these studies, we investigated Dio3 expression in sensory ganglia that relay sensory information to the brain in Dio3Cre embryos (Fig. 5A-5C). We detected ZsG signals in the dorsal root ganglia, trigeminal ganglion, geniculate ganglion, cochleovestibular, and spiral ganglion. Expression in the spiral ganglion is in accord with previous in situ hybridization data (6). These ganglia relay inputs for somatosensation and nociception (dorsal root ganglia), somatosensation for the face and jaw (trigeminal ganglion), taste (geniculate ganglion), hearing (cochleovestibular ganglion, spiral ganglion), and balance (cochleovestibular ganglion) (45). Dio3-expressing cells in these ganglia were identified as neurons by costaining with neuronal markers Tubb3 (Fig. 5A-5C) and neurofilament heavy chain (Fig. 5D-5F). The varying percentages of ZsG+ neurons in these ganglia probably reflect labeling inefficiencies due to low level Dio3 expression (∼100% of neurons in the spiral ganglion and ∼30% in the trigeminal ganglion were ZsG+). These results support a role for Dio3 in control of T3 signaling in relay pathways for diverse sensory systems.

Figure 5.

Figure 5.

Dio3 expression in neurons of sensory ganglia identified with the Dio3Cre driver. (A-C) Representative sagittal sections of Dio3Cre;Ai6 embryos (n = 3, mixed male and female at each age), showing Dio3 expression (ZsG) in sensory ganglia. Ganglia and associated nerves are identified by co-labeling with Tubb3 (βIII-Tubulin), a neuronal-specific marker. Tissue counterstain, Hoechst. Dio3 expression is evident in trigeminal (TG), geniculate (GG), and cochleovestibular (CVG) ganglia (A), the spiral ganglion (SG) with nerve fibers extending from the cochlea (B), and dorsal root ganglia (DRG) (C). Orientation arrows: D, dorsal; A, anterior. (D-F) High magnification images showing ZsG signals in neurons of the TG, SG and DRG at P1, indicated by colocalization with ganglion marker neurofilament heavy chain (NFH).

Expression of Dio3 in Adult Brain Detected Using the Dio3CreERt2 Driver

Dio3 is also considered to function in the adult brain although it has been difficult to define in which cell types because of the decline of Dio3 expression during maturation (18). We therefore took advantage of the conditional Dio3CreErt2 driver to investigate expression in adults following administration of TAM at 8 weeks of age (Fig. 6). ZsG signals were detected in regions including the NAc, BST and amygdala, circumventricular organs (OVLT, SFO) and raphe nuclei. Notably, this pattern in the adult brain resembles that detected with the Dio3Cre driver in neonatal mice (Fig. 2) indicating continuing functions for Dio3 in many of the same regions as in the neonate. Labeling was undetected in Purkinje neurons in adult mice, suggesting that expression is strongly suppressed in adult brain regions where Dio3 may primarily serve an earlier, developmental role, consistent with the major postnatal decline of DIO3 enzyme activity in the cerebellum and other brain areas (11, 18). Novel sites of expression were not obvious in the mature brain suggesting that most adult patterns are largely established at earlier stages of neuronal lineage differentiation in the fetus and neonate. These patterns were further supported by extraction of Dio3 data from a spatial transcriptomic atlas of adult mouse brain (46) (Allen ABC atlas, dataset: MERSCOPE v1 whole brain—imputed genes and reconstructed coordinates) (https://knowledge.brain-map.org/abcatlas), in which Dio3 RNA expression correlates with most of the regions indicated by the Dio3CreERt2 driver (Fig. 6 and Table 1). The few ostensible differences between the approaches presumably arise from limitations in detection of low Dio3 expression and from variable representation of smaller cell populations. For instance, Dio3 is detected in medullary raphe nuclei, relatively small neuronal clusters, in adult Dio3CreERt2 mice but is unrecorded in the ABC atlas. Conversely, in Dio3CreERt2 adult mice, some Dio3-expressing nuclei (eg, MnPO and AVPV, Fig. 6A) are inefficiently labeled, which may be because (1) activation of Cre-dependent reporters rarely approaches 100% efficiency, typically a greater limitation for conditional (TAM-inducible) than nonconditional Cre recombinases; (2) low Dio3 expression may produce inadequate levels of CreERt2 recoombinase; (3) transport of TAM might be less efficient in certain brain regions. A potential variation that cannot be entirely excluded is that expression of the Rosa26Ai6 reporter might vary; however, once activated by Cre recombination, Rosa26Ai6 expression (ie, ZsG protein) is strongly driven by a powerful CAG promoter that generally minimizes cell type variations (28).

Figure 6.

Figure 6.

Dio3 expression and T3-sensitivity of adult brain detected with a Dio3CreERt2 driver. (A) Dio3 expression (ZsG) in the adult brain revealed by tamoxifen (TAM) treatment of Dio3CreERt2/+;  Rosa26Ai6/Ai6 mice. No ZsG signal was observed in vehicle-treated controls. Coronal sections show ZsG signals in NAc, BST, amygdala (central amygdalar nucleus, CeA), and the OVLT. T3 administration (TAM, T3) enhanced Dio3 expression, indicated by increased numbers of ZsG+ cells in these regions and in posterior agranular insular cortex (AIp). Tissue counterstain, Hoechst. DLS, dorsolateral striatum (negative area even under T3 treatment). (B) High magnification of indicated regions reveals Dio3 expression (ZsG) in neurons, with double-labeled cells (ZsG+NeuN+) appearing whitish. (C, D) Dio3 expression in CVOs at adult stages, including OVLT (C) and SFO (D), both characterized by extensive IB4 + capillary networks. T3 increased ZsG+ cell numbers, particularly in the OVLT, and the adjacent MnPO and AVPV. Dashed line outlines OVLT. (E, F) Dio3 expression (ZsG+) detected in neurons (NeuN+) in the raphe nuclei in the ventral brainstem following TAM treatment. ZsG+ cell numbers increased with T3 treatment. (G) Quantification of ZsG+ cell numbers in the regions indicated, after vehicle (veh), TAM, or T3 (T3, TAM) treatment. *P < .05, **P < .01, ***P < .001 (4 male mice; 1-way analysis of variance followed by the Tukey post hoc test).

Abbreviations are as defined in previous figures. (H) Scheme of treatment with TAM and T3.

Table 1.

Dio3 expression and T3 sensitivity in selected adult brain regions in Dio3CreERt2 mice and comparison with single cell data from the Allen Brain Cell Atlas

Brain nuclei Expression (ZsG) in Dio3-CreERt2;Ai6 micea Allen ABC Atlasb: wild-type mouse brain
TAM TAM, T3 % neurons with detectable Dio3 RNAc Expression level (mean CPM)
Nucleus accumbens (NAc) + ++ ++ 2.39
Bed nucleus of the stria terminalis (BST) + ++ ++ 2.60
Vascular organ of the lamina terminalis (OVLT) + ++ +++ 1.26
Median preoptic nucleus (MnPO) +/− ++ ++ .68
Anteroventral periventricular nucleus (AVPV) +/− ++ + .45
Subfornical organ (SFO) ++ ++ +++ .87
Central amygdalar nucleus(CEA) + ++ ++ 2.35
Raphe nuclei (medulla) + ++ <.18
Agranular insular cortex, posterior (AIp) + ++ 3.40
Cerebellar Purkinje neuron <.18
Dorsolateral striatum (DLS) <.19

a Scores estimated from ZsG+ cell counts indicating Dio3 expression (TAM) and its T3-inducibility (TAM, T3) in selected brain regions (see Fig. 6).

b Allen Brain Cell Atlas (https://knowledge.brain-map.org/abcatlas, RRID:SCR_024440), dataset: MERSCOPE v1 whole brain—imputed genes and reconstructed coordinates).

c Expression grouped in bins based on % of Dio3-expressing neurons: –, 0% to 5%; +, >5% to 33%; ++, >33% to 67%; +++, >67%; (cutoff >0.2 CPM, counts per million mapped reads).

T3 Sensitivity of Dio3 Expression in the Adult Brain

In some tissues, Dio3 is inducible by T3, suggesting a homeostatic role involving a form of auto-regulation whereby elevated T3 induces expression of DIO3 enzyme which then degrades T3 and T4 to control the tissue response and protect against overstimulation. T3 can upregulate Dio3 mRNA levels, detected by quantitative PCR, in mouse cerebral cortex (47). To investigate this question at the level of specific cell types, we treated adult Dio3CreErt2 mice with T3 in addition to TAM (Fig. 6). The addition of T3 resulted in remarkable increases in ZsG+ cells in several brain regions compared with TAM alone. The increase in Dio3 (ZsG) signals largely appeared in areas where Dio3 was already expressed under normal thyroid (euthyroid) conditions, including the amygdala, NAc, BST, circumventricular organs (OVLT, SFO), and raphe nuclei, suggesting that Dio3 expression was boosted rather than initiated de novo by T3. ZsG+ cells detected in these regions were almost all neurons, as demonstrated by colocalization with NeuN (Fig. 6B, 6D, and 6F). T3 also induced additional, sparse ZsG+ signals in a few regions such as the MnPO, AVPV, and agranular insular cortex that were otherwise undetected under euthyroid conditions, probably because of low level Dio3CreERt2 expression when not boosted by T3 treatment. The results indicate that Dio3 expression is highly T3 sensitive and is dynamically involved in specific neuronal responses in regions of the adult brain.

Discussion

The Cre driver alleles reported in this study allow fine resolution mapping of Dio3-expressing cell types within the nervous system in mice, yielding insights into cellular mechanisms of T3 signaling in behavioral, chemosensory, and sensory systems.

Dio3 expression patterns are largely established at early stages in accord with the peak of DIO3 enzyme activity reported at fetal stages in several brain regions (11, 18, 19, 22). This early pattern suggests a need for DIO3 to limit T3 action soon after neuronal lineages are formed when immature tissues may be vulnerable to T3-induced damage. Although Dio3 expression later declines substantially, evidence using a conditional Dio3CreERt2 driver (Fig. 6) indicates persistent expression in at least some of the same regions in adults, suggesting ongoing neural and homeostatic functions in the mature brain.

The developmental and spatial patterns of Cre-mediated labeling extend and are consistent with reported type 3 deiodination enzyme assay profiles (11, 16, 18) and in situ hybridization data (11, 16, 26), validating the Cre driver approach. Dio3 RNA expression data extracted from a spatial transcriptomic database of adult mouse brain (46) further supports the Dio3 expression patterns mapped at mature ages (Table 1). We note that Dio3Cre and Dio3CreERt2 drivers, like any Cre driver, can underestimate the full extent of gene expression given the imperfect efficiency of activation of Cre-dependent reporters; this limitation may be exacerbated because of low Dio3 expression in adults (Table 1). However, Cre-mediated labeling has an advantage of sensitivity allowing identification of previously-unrecognized, specialized neuronal groups that express Dio3 (eg, in the raphe nuclei (Fig. 3) and circumventricular organs (Fig. 4)) (as discussed below) and also defines cellular morphology at a level not possible with in situ hybridization data. The Cre driver alleles also allow analysis of hormonal (T3) regulation of Dio3-positive cell types (Fig. 6) and will facilitate further functional studies.

The developmental and cell-specific patterns of Dio3 expression detected suggest that DIO3 critically calibrates the timing and magnitude of T3 signaling appropriate for specialized neuronal groups in both central and peripheral nervous systems. The net result of these actions is to coordinate the overall maturation of the nervous system. The expression patterns also suggest sensitive tissues in which DIO3 may protect against T3-induced toxicity in hyperthyroid conditions (4, 48).

Dio3 Expression in Specialized Neuronal Groups

The raphe nuclei in the brainstem, the main source of serotonin in the brain, are widely involved in behaviors such as mood, aggression, and emotion and may respond to various signals (41). These nuclei influence many brain regions and have therapeutic relevance as serotonin pathways represent targets for psychiatric drugs, such as selective serotonin reuptake inhibitors (49). Dio3 expression in the raphe nuclei suggests possible involvement of serotonergic neurons in the depression or anxiety associated with thyroid disorders (50), such as hyperthyroidism due to Graves disease (51, 52) or in behavioral phenotypes observed in Dio3-deficient mice (12). Dio3 expression in the NAc and amygdala (26, 38, 39), further supports an influence of T3 signaling over emotion and motivation.

Dio3 expression in Purkinje neurons highlights the role of T3 in development of the cerebellum and motor control. DIO3 enzyme activity in the fetal cerebellum (11, 18, 22) is consistent with immature Purkinje neurons being sensitive targets that rely on Dio3 to calibrate their response to T3. Purkinje neuron differentiation is impaired by hypothyroidism (53-55) or Dio3 deficiency (11), reflecting vulnerability to too little or too much T3. Manipulations of T3 receptors indicate that Purkinje neurons are a primary target for T3 but also that these cells influence nearby granule neurons (56) suggesting that Dio3 in Purkinje neurons may more generally influence cerebellar maturation (11).

Dio3 Expression in Circumventricular Organs

Dio3 expression was detected in 2 major chemosensory circumventricular organs, namely the OVLT and SFO (Fig. 4). The interfaces of these CVOs with permeable capillary networks allow sensing of solutes and signaling factors in the circulation without restriction by the blood–brain barrier. Dio3 expression in OVLT and SFO neurons suggests a role for T3 in the chemosensory functions of these CVOs, which respond to salt and water balance, other signaling factors and hormones such as angiotensin in the circulation (44). These CVOs extend projections to other brain areas including hypothalamic paraventricular and preoptic nuclei and are thought to influence central control of body fluids, thirst and other homeostatic functions. Thus, CVO neurons could potentially integrate the response of diverse physiological systems to T3 in blood or cerebrospinal fluid (57).

ZsG signals in the nerve terminals of the ME and posterior pituitary reflect Dio3 expression in hypothalamic neurosecretory neurons (Figs. 3 and 4) and suggest control of factors that regulate both the anterior and posterior pituitary gland. This function is consistent with DIO3 enzyme activity reported in the hypothalamus (13, 18) and could contribute to the altered expression in Dio3-deficient mice of anterior pituitary hormones including thyrotropin (13) and gonadotropins (58) and posterior pituitary hormones AVP and oxytocin (59). Expression of AVP and oxytocin is subject to complex hormonal control including by T3 (60, 61). Although not the focus of the present work, the hypothalamus also regulates metabolic functions. Dio3 expression in wider hypothalamic areas (Fig. 2D-2G) suggests contributions to diverse neuroendocrine and metabolic interactions (10, 25, 62, 63).

Sensory Pathways

Sensory development relies upon T3 signaling, with prominent roles in sensory organs, including the cochlea and retina. Dio3 deficiency leads to deafness (6), visual achromatopsia (16), and changes in olfactory behavior in mice (59). Our finding of Dio3 expression in sensory ganglia that relay auditory, vestibular and somatosensory information to the brain (Fig. 5) broadens the perspective of T3 as a key factor in diverse sensory systems (3, 64). Sensory ganglia in the peripheral nervous system lack a blood–brain barrier such that Dio3 may protect these ganglia against fluctuations of T3 in the surrounding cerebrospinal fluid (65).

In summary, these results obtained with Cre driver models, suggest that Dio3 controls specific neuronal groups in behavioral, chemosensory, and sensory systems during the maturation and function of the nervous system. We also observed Dio3 expression in non-neuronal tissues such as the choroid plexus and meninges, suggesting additional non-neuronal control over T3 action in the nervous system, which may deserve future study.

Acknowledgments

We thank Jeff Reece for providing facilities at the advanced light microscopy core of the National Institute of Diabetes and Digestive and Kidney Diseases at NIH. We are grateful to Hong Liu and Young-Wook Cho for helpful discussion.

Abbreviations

AVP

arginine vasopressin

BST

bed nucleus of the stria terminalis

CRF

corticotropin-releasing factor

CVO

circumventricular organ

DIO3

type 3 deiodinase

ME

median eminence

MnPO

median preoptic nucleus

NAc

nucleus accumbens

OVLT

organum vasculosum laminae terminalis

SFO

subfornical organ

T3

triiodothyronine

T4

thyroxine

TAM

tamoxifen

ZsG

ZsGreen

Contributor Information

Ye Liu, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, MD 20892, USA.

Lily Ng, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, MD 20892, USA.

Chengyu Liu, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, MD 20892, USA.

Douglas Forrest, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, MD 20892, USA.

Funding

This work was supported by the intramural research program at the National Institute of Diabetes and Digestive and Kidney Diseases (Y.L., L.N., and D.F.) and the National Heart, Lung, and Blood Institute (C.L.) at the National Institutes of Health.

Author Contributions

Y.L., L.N., and D.F. designed research; Y.L., L.N., and C.L. performed research; Y.L., L.N., C.L., and D.F. analyzed data; and Y.L. and D.F. wrote the paper with input from other authors.

Disclosures

The authors declare no competing interests.

Data Availability

Some original data generated during and/or analyzed during the current study are not publicly available but are available from the corresponding author upon reasonable request.

References

  • 1. Bernal  J.  Thyroid and Brain: Understanding the Actions of Thyroid Hormones in Brain Development and Function. 2024.  Bentham Science Publishers. [Google Scholar]
  • 2. DeLong  GR, Stanbury  JB, Fierro-Benitez  R. Neurological signs in congenital iodine-deficiency disorder (endemic cretinism). Dev Med Child Neurol. 1985;27(3):317‐324. [DOI] [PubMed] [Google Scholar]
  • 3. Ng  L, Kelley  MW, Forrest  D. Making sense with thyroid hormone–the role of T(3) in auditory development. Nat Rev Endocrinol. 2013;9(5):296‐307. [DOI] [PubMed] [Google Scholar]
  • 4. Polak  M, Legac  I, Vuillard  E, Guibourdenche  J, Castanet  M, Luton  D. Congenital hyperthyroidism: the fetus as a patient. Horm Res. 2006;65(5):235‐242. [DOI] [PubMed] [Google Scholar]
  • 5. Rastogi  MV, LaFranchi  SH. Congenital hypothyroidism. Orphanet J Rare Dis. 2010;5(1):17. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Ng  L, Hernandez  A, He  W, et al.  A protective role for type 3 deiodinase, a thyroid hormone-inactivating enzyme, in cochlear development and auditory function. Endocrinology. 2009;150(4):1952‐1960. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. Ma  H, Yang  F, York  LR, Li  S, Ding  XQ. Excessive thyroid hormone signaling induces photoreceptor degeneration in mice. eNeuro. 2023;10(9):ENEURO.0058-23.2023. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8. Bianco  AC, Kim  BW. Deiodinases: implications of the local control of thyroid hormone action. J Clin Invest. 2006;116(10):2571‐2579. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9. St Germain  DL, Galton  VA, Hernandez  A. Minireview: defining the roles of the iodothyronine deiodinases: current concepts and challenges. Endocrinology. 2009;150(3):1097‐1107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10. Salas-Lucia  F, Fekete  C, Sinko  R, et al.  Axonal T3 uptake and transport can trigger thyroid hormone signaling in the brain. eLife. 2023;12:e82683. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11. Peeters  RP, Hernandez  A, Ng  L, et al.  Cerebellar abnormalities in mice lacking type 3 deiodinase and partial reversal of phenotype by deletion of thyroid hormone receptor alpha1. Endocrinology. 2013;154(1):550‐561. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Stohn  JP, Martinez  ME, Hernandez  A. Decreased anxiety- and depression-like behaviors and hyperactivity in a type 3 deiodinase-deficient mouse showing brain thyrotoxicosis and peripheral hypothyroidism. Psychoneuroendocrinology. 2016;74:46‐56. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Hernandez  A, Martinez  ME, Fiering  S, Galton  VA, St Germain  D. Type 3 deiodinase is critical for the maturation and function of the thyroid axis. J Clin Invest. 2006;116(2):476‐484. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14. Wu  Z, Martinez  ME, St Germain  DL, Hernandez  A. Type 3 deiodinase role on central thyroid hormone action affects the leptin-melanocortin system and circadian activity. Endocrinology. 2017;158(2):419‐430. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. Wu  Z, Martinez  ME, DeMambro  V, Francois  M, Hernandez  A. Developmental thyroid hormone action on pro-opiomelanocortin-expressing cells programs hypothalamic BMPR1A depletion and brown fat activation. J Mol Cell Biol. 2023;14(9):mjac078. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. Ng  L, Lyubarsky  A, Nikonov  SS, et al.  Type 3 deiodinase, a thyroid-hormone-inactivating enzyme, controls survival and maturation of cone photoreceptors. J Neurosci. 2010;30(9):3347‐3357. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17. Hernandez  A, Quignodon  L, Martinez  ME, Flamant  F, St Germain  DL. Type 3 deiodinase deficiency causes spatial and temporal alterations in brain T3 signaling that are dissociated from serum thyroid hormone levels. Endocrinology. 2010;151(11):5550‐5558. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Kaplan  MM, Yaskoski  KA. Maturational patterns of iodothyronine phenolic and tyrosyl ring deiodinase activities in rat cerebrum, cerebellum, and hypothalamus. J Clin Invest. 1981;67(4):1208‐1214. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19. Bates  JM, St Germain  DL, Galton  VA. Expression profiles of the three iodothyronine deiodinases, D1, D2, and D3, in the developing rat. Endocrinology. 1999;140(2):844‐851. [DOI] [PubMed] [Google Scholar]
  • 20. Ientile  R, Macaione  S, Russo  P, Pugliese  G, Di Giorgio  RM. Phenolic and tyrosyl ring deiodination in thyroxine from rat retina during postnatal development. Eur J Biochem. 1984;142(1):15‐19. [DOI] [PubMed] [Google Scholar]
  • 21. Hernandez  A, Martinez  ME, Ng  L, Forrest  D. Thyroid hormone deiodinases: dynamic switches in developmental transitions. Endocrinology. 2021;162(8):bqab091. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Kester  MH, Martinez de Mena  R, Obregon  MJ, et al.  Iodothyronine levels in the human developing brain: major regulatory roles of iodothyronine deiodinases in different areas. J Clin Endocrinol Metab. 2004;89(7):3117‐3128. [DOI] [PubMed] [Google Scholar]
  • 23. Stohn  JP, Martinez  ME, St Germain  DL, Hernandez  A. Adult onset of type 3 deiodinase deficiency in mice alters brain gene expression and increases locomotor activity. Psychoneuroendocrinology. 2019;110:104439. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Wu  Z, Hernandez  A. Thyroid hormone clearance in the paraventricular nucleus of male mice regulates lean mass and physical activity. Neuroendocrinology. 2024;114(11):1066‐1076. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25. Alkemade  A, Friesema  EC, Unmehopa  UA, et al.  Neuroanatomical pathways for thyroid hormone feedback in the human hypothalamus. J Clin Endocrinol Metab. 2005;90(7):4322‐4334. [DOI] [PubMed] [Google Scholar]
  • 26. Escamez  MJ, Guadano-Ferraz  A, Cuadrado  A, Bernal  J. Type 3 iodothyronine deiodinase is selectively expressed in areas related to sexual differentiation in the newborn rat brain. Endocrinology. 1999;140(11):5443‐5446. [DOI] [PubMed] [Google Scholar]
  • 27. Indra  AK, Warot  X, Brocard  J, et al.  Temporally-controlled site-specific mutagenesis in the basal layer of the epidermis: comparison of the recombinase activity of the tamoxifen-inducible Cre-ER(T) and Cre-ER(T2) recombinases. Nucleic Acids Res. 1999;27(22):4324‐4327. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Madisen  L, Zwingman  TA, Sunkin  SM, et al.  A robust and high-throughput Cre reporting and characterization system for the whole mouse brain. Nat Neurosci. 2010;13(1):133‐140. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29. Liu  Y, Du  Y, Xie  W, Zhang  F, Forrest  D, Liu  C. Generation of conditional knockout mice by sequential insertion of two loxP sites in cis using CRISPR/Cas9 and single-stranded DNA oligonucleotides. Methods Mol Biol. 2019; 1874:191‐210. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30. de Lange  WJ, Halabi  CM, Beyer  AM, Sigmund  CD. Germ line activation of the Tie2 and SMMHC promoters causes noncell-specific deletion of floxed alleles. Physiol Genomics. 2008;35(1):1‐4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31. Ng  L, Liu  Y, Liu  H, Forrest  D. Cochlear fibrocyte and osteoblast lineages expressing type 2 deiodinase identified with a Dio2CreERt2 allele. Endocrinology. 2021;162(12):bqab179. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. Hamidi  S, Aliesky  HA, Williams  RW, Rapoport  B, McLachlan  SM. Genetic linkages for thyroxine released in response to thyrotropin stimulation in three sets of recombinant inbred mice provide evidence for shared and novel genes controlling thyroid function. Thyroid. 2013; 23(3):360‐370. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33. Ved  N, Curran  A, Ashcroft  FM, Sparrow  DB. Tamoxifen administration in pregnant mice can be deleterious to both mother and embryo. Lab Anim. 2019;53(6):630‐633. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34. Martinez  ME, Charalambous  M, Saferali  A, et al.  Genomic imprinting variations in the mouse type 3 deiodinase gene between tissues and brain regions. Mol Endocrinol. 2014;28(11):1875‐1886. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35. Martinez  ME, Pinz  I, Preda  M, Norton  CR, Gridley  T, Hernandez  A. DIO3 protects against thyrotoxicosis-derived cranio-encephalic and cardiac congenital abnormalities. JCI Insight. 2022;7(21):e161214. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36. Hall  JA, Ribich  S, Christoffolete  MA, et al.  Absence of thyroid hormone activation during development underlies a permanent defect in adaptive thermogenesis. Endocrinology. 2010;151(9):4573‐4582. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37. Barton  KM, Levine  EM. Expression patterns and cell cycle profiles of PCNA, MCM6, cyclin D1, cyclin A2, cyclin B1, and phosphorylated histone H3 in the developing mouse retina. Dev Dyn. 2008;237(3):672‐682. [DOI] [PubMed] [Google Scholar]
  • 38. Castro  DC, Bruchas  MR. A motivational and neuropeptidergic hub: anatomical and functional diversity within the nucleus Accumbens shell. Neuron. 2019;102(3):529‐552. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39. Maddox  SA, Ponomareva  OY, Zaleski  CE, et al.  Evidence for thyroid hormone regulation of amygdala-dependent fear-relevant memory and plasticity. Mol Psychiatry. 2025;30(1):201‐212. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40. Alonso  A, Merchan  P, Sandoval  JE, et al.  Development of the serotonergic cells in murine raphe nuclei and their relations with rhombomeric domains. Brain Struct Funct. 2013;218(5):1229‐1277. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41. Azmitia  EC. Serotonin neurons, neuroplasticity, and homeostasis of neural tissue. Neuropsychopharmacology. 1999;21(2 Suppl):S33––S345.. [DOI] [PubMed] [Google Scholar]
  • 42. Hicks  AI, Kobrinsky  S, Zhou  S, Yang  J, Prager-Khoutorsky  M. Anatomical organization of the rat subfornical organ. Front Cell Neurosci. 2021;15:691711. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43. Prager-Khoutorsky  M, Bourque  CW. Anatomical organization of the rat organum vasculosum laminae terminalis. Am J Physiol Regul Integr Comp Physiol. 2015;309(4):R324‐R337. [DOI] [PubMed] [Google Scholar]
  • 44. Noda  M, Sakuta  H. Central regulation of body-fluid homeostasis. Trends Neurosci. 2013;36(11):661‐673. [DOI] [PubMed] [Google Scholar]
  • 45. Vermeiren  S, Bellefroid  EJ, Desiderio  S. Vertebrate sensory ganglia: common and divergent features of the transcriptional programs generating their functional specialization. Front Cell Dev Biol. 2020;8:587699. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46. Yao  Z, van Velthoven  CTJ, Kunst  M, et al.  A high-resolution transcriptomic and spatial atlas of cell types in the whole mouse brain. Nature. 2023;624(7991):317‐332. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47. Hernandez  A, Morte  B, Belinchon  MM, Ceballos  A, Bernal  J. Critical role of types 2 and 3 deiodinases in the negative regulation of gene expression by T(3)in the mouse cerebral cortex. Endocrinology. 2012;153(6):2919‐2928. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48. Daneman  D, Howard  NJ. Neonatal thyrotoxicosis: intellectual impairment and craniosynostosis in later years. J Pediatr. 1980;97(2):257‐259. [DOI] [PubMed] [Google Scholar]
  • 49. Charnay  Y, Leger  L. Brain serotonergic circuitries. Dialogues Clin Neurosci. 2010;12(4):471‐487. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50. Bauer  M, Heinz  A, Whybrow  PC. Thyroid hormones, serotonin and mood: of synergy and significance in the adult brain. Mol Psychiatry. 2002;7(2):140‐156. [DOI] [PubMed] [Google Scholar]
  • 51. Bunevicius  R, Prange  AJ  Jr. Psychiatric manifestations of Graves' hyperthyroidism: pathophysiology and treatment options. CNS Drugs. 2006;20(11):897‐909. [DOI] [PubMed] [Google Scholar]
  • 52. Feldman  AZ, Shrestha  RT, Hennessey  JV. Neuropsychiatric manifestations of thyroid disease. Endocrinol Metab Clin North Am. 2013;42(3):453‐476. [DOI] [PubMed] [Google Scholar]
  • 53. Legrand  J. Morphogenetic actions of thyroid hormones. Trends Neurosci. 1979;2:234‐236. [Google Scholar]
  • 54. Morte  B, Manzano  J, Scanlan  T, Vennström  B, Bernal  J. Deletion of the thyroid hormone receptor alpha 1 prevents the structural alterations of the cerebellum induced by hypothyroidism. Proc Natl Acad Sci U S A. 2002;99(6):3985‐3989. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55. Picou  F, Fauquier  T, Chatonnet  F, Flamant  F. A bimodal influence of thyroid hormone on cerebellum oligodendrocyte differentiation. Mol Endocrinol. 2012;26(4):608‐618. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56. Fauquier  T, Chatonnet  F, Picou  F, et al.  Purkinje cells and Bergmann glia are primary targets of the TRalpha1 thyroid hormone receptor during mouse cerebellum postnatal development. Development. 2014;141(1):166‐175. [DOI] [PubMed] [Google Scholar]
  • 57. Dratman  MB, Crutchfield  FL, Schoenhoff  MB. Transport of iodothyronines from bloodstream to brain: contributions by blood:brain and choroid plexus:cerebrospinal fluid barriers. Brain Res. 1991;554(1-2):229‐236. [DOI] [PubMed] [Google Scholar]
  • 58. Martinez  ME, Karaczyn  A, Stohn  JP, et al.  The type 3 deiodinase is a critical determinant of appropriate thyroid hormone action in the developing testis. Endocrinology. 2016;157(3):1276‐1288. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59. Stohn  JP, Martinez  ME, Zafer  M, Lopez-Espindola  D, Keyes  LM, Hernandez  A. Increased aggression and lack of maternal behavior in Dio3-deficient mice are associated with abnormalities in oxytocin and vasopressin systems. Genes Brain Behav. 2018;17(1):23‐35. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60. Ciosek  J, Drobnik  J. Vasopressin and oxytocin release and the thyroid function. J Physiol Pharmacol. 2004;55(2):423‐441. [PubMed] [Google Scholar]
  • 61. Dellovade  TL, Zhu  YS, Pfaff  DW. Thyroid hormones and estrogen affect oxytocin gene expression in hypothalamic neurons. J Neuroendocrinol. 1999;11(1):1‐10. [DOI] [PubMed] [Google Scholar]
  • 62. Zhang  Z, Boelen  A, Bisschop  PH, Kalsbeek  A, Fliers  E. Hypothalamic effects of thyroid hormone. Mol Cell Endocrinol. 2017;458:143‐148. [DOI] [PubMed] [Google Scholar]
  • 63. Mullur  R, Liu  YY, Brent  GA. Thyroid hormone regulation of metabolism. Physiol Rev. 2014;94(2):355‐382. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 64. Freeman  S, Sohmer  H. Effect of thyroxine on the development of somatosensory and visual evoked potentials in the rat. J Neurol Sci. 1995;128(2):143‐150. [DOI] [PubMed] [Google Scholar]
  • 65. Hagen  GA, Elliott  WJ. Transport of thyroid hormones in serum and cerebrospinal fluid. J Clin Endocrinol Metab. 1973;37(3):415‐422. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

Some original data generated during and/or analyzed during the current study are not publicly available but are available from the corresponding author upon reasonable request.


Articles from Endocrinology are provided here courtesy of The Endocrine Society

RESOURCES