Thyroid hormones and brain

Research topics

Relative role of thyroid hormone transporters and deiodinases in the brain action of thyroid hormones

Thyroid hormones cross the blood-brain barrier through integral membrane transporter proteins, the monocarboxylate 8 (MCT8) transporter and the organic anion transporter polypeptide 1C1 (OATP1C1). MCT8 is encoded by the SLC16A2 gene located in the X crhomosome. It is specific for T3, T4, and other thyroid hormone derivatives. Disruption of the gene causes an X-linked syndrome with altered thyroid hormone metabolism and action, and profound neuromotor and cognitive impairement (Allan-Herndon-Dudley syndrome, or AHDS). It is thought that the syndrome is due to deficient transport of T4 and T3 through the blood-brain barrier from fetal stages. OATP1C1 is encoded by the SLCO1C1 gene, located in chromosome 12, with high affinity for T4 but very low for T3. A neurodegenerative syndrome has recently been described caused by mutation in this gene, and the relationship with lack of T4 transport is suspected but not demonstrated. 

The active thyroid hormone is T3, which binds to nuclear receptors and regulates gene expression. T4, the main hormone produced by the thyroid gland, is a prohormone which generates T3 in tissues by the action of deiodinases. These are selenoenzymes which remove iodine atoms from the iodothyronine molecule generating the active product T3, from T4 (types 1 and 3 deiodinases, or DIO1 and DIO2), or inactive metabolites from T4 or T3 (Type 3 dedinase, or DIO3). Brain T3 derives in part from the circulation, and in part from DIO2-catalyzed, T4 deiodination in the astrocytes. During early development, most brain T3 is derived from T4 which, as stated above, enters the brain through MCT8 and OATP1C1. OATP1C1 is expressed at very low levels in the human blood-brain barrier, and therefore the human brain is strictly dependent on MCT8. In contrast, the rodent brain expresses similarly MCT8 and OATP1C1 so that MCT8 knock out mice do not have neurological phenotype, because lack of MCT8 is compensated by OATP1C1. 

Alternative mdels of AHDS have been generated. One, the MCT8 and DIO2 double KO (dKO), in which T3 transport and T3 generation from T4 are compromised. Another, the MCT8 and OATP1C1 dKO, in which T4 and T3 transport is abolished. In either case similar T3 deficiency and brain hypothyroidism should occur. However, a series of theoretical and experimental considerations cast doubts on this model. We compared the effects of the dKO models, and systemic hypothyroidism on gene expression in the cerebral cortex and the striatum using RNA-Seq. We analyzed the following groups: wild type mice, systemic hypothyroid mice after chemical thyroidectomy, MCT8 and DIO2 dKO, and MCT8 and OATP1C1 dKO. The two dKO cause hypothyrodism limited to the cerebrum, whereas systemic hypothyroidism affects all tissues.The results show that the impact of systemic hypothyroidism is much more profound in terms of gene expression than cerebral hypothyroidism. In addition, there is only a partial overlap between the two models of cerebral hypothyroidism. The results obtained by us strongly question current hypothesis on the mechanisms of disease in AHDS.

Keywords: brain development, thyroid hormones, membrane transporters, neonatal hypotonia, hypothyroidism, gene expression.

Analysis of the cellular coexpression of thyroid hormone transporters, deiodinases and receptors in single cells of the human brain

Our research interest is the mechanisms of action of thyroid hormones in the human brain during development. Along this line we are interested in the identification of the thyroid hormone sensitive cells during cortex development. Recently, the accumulation of transcriptomics datasets on single cells isolated from human samples makes it possible to perform in silico studies to analyze the co-expression of thyroid hormone transporters, thyroid hormone receptors, and thyroid hormone activating and inactivating enzymes in single cells, as a mean to identify the sensitive cells. One of these databases was generated recently by Tomasz J. Nowakowski and coworkers (Nowakowski et al, Spatiotemporal gene expression trajectories reveal developmental hierarchies of the human cortex, Science 358: 1318-1323, 2017). This database contains transcriptomics data of 4260 single cortical cells from 73 human fetal subjects, and is deposited in the database on Genotypes and Phenotypes (dbGaP) ( under the label “STUDIES OF HUMAN DEVELOPMENTAL NEUROGENESIS phs000989.v3.p1.” 

We are performing in silico analysis of this dataset. Clustering and differential expression analysis will enable to identify groups of cells showing evidence of regulation by thyroid hormones. For this we will focus on genes involved in thyroid hormone production and transport, their nuclear receptors and genes known to be regulated by thyroid hormones in humans and mice. Some preliminary data indicate that DIO2 is present in radial glia cells that also express OATP1C1. The MCT8 transporter is associated with the thyroid hormone receptor alpha and with many of the T3 target genes. Interestingly the TRbeta is present in a cluster containing interneurons (Fig 1). The data will shed light as to the specific role of thyroid hormones on cortical development and the mechanisms involved.

Keywords: Fetal human development, radial glia, interneurons, Deiodinases, thyroid hormone receptors, single cell genomics.


Fig 1: DIO2  and THRB expression (brown color) in single cells of the developing human cerebral cortex. DIO2 is the enzyme that generates the active thyroid hormone T3 from the T4 precursor by deiodination. It is expressed in clusters of cells with the identitity of radial glia cells and astrocytes. THRB is the thyroid hormone nuclear receptor beta subtype, and is specifically expressed in a cluster of cells with the identity of interneurons. The data were extracted from the databases generated by Novakowski et al. (Science 358, 1318–1323, 2017). The cell clusters were generated by unbiased clustering and weighted gene coexpression network analysis after RNA-Seq of 4261 individual cells. Our analysis indicates that the radial glia is a local source of the active thyroid hormone, and that the thyroid hornone receptor beta subtype is involved in differentiation of interneurons. (RG: Radial glia; IN, Interneurons; NE, excitatory neurons).