Thyroid hormone promotes fetal neurogenesis

Abstract

Maternal low thyroxine (T4) serum levels during the first trimester of pregnancy correlate with cerebral cortex volume and mental development of the progeny, but why neural cells during early fetal brain development are vulnerable to maternal T4 levels remains unknown. In this study, using iPSCs obtained from a boy with a loss-of-function mutation in MCT8 - a transporter previously identified as critical for thyroid hormone uptake and action in neural cells - we demonstrate that thyroid hormone induces transcriptional changes that promote the progression of human neural precursor cells along the dorsal projection trajectory. Consistent with these findings, single-cell, spatial, and bulk transcriptomics from MCT8-deficient cerebral organoids and cultures of human neural precursor cells underscored the necessity for optimal thyroid hormone levels for these cells to differentiate into neurons. The controlled intracellular activation of T4 signaling occurs through the transient expression of the enzyme type 2 deiodinase, which converts T4 into its active form, T3, alongside the coordinated expression of thyroid hormone nuclear receptors. The intracellular activation of T4 in neural precursor cells results in transcriptional changes important for their division mode and cell cycle progression. Thus, T4 is essential for fetal neurogenesis, highlighting the importance of adequate treatment for mothers with hypothyroidism.

Keywords: Embryonic stem cells; Endocrinology; Neuroscience; Thyroid disease; Transcriptomics.

Figure 1. scRNA-seq analysis of D50 Control- and MCT8-COs.

(A) Schematic representation of generating D50 MCT8-COs and obtaining single-cell suspension for sc-RNA-seq analysis. (B) UMAP plot showing the cell clusters identified by principal component analysis in the indicated COs. (C) UMAP plots showing the distribution of markers for NPCs and neurons. (D) Dot plot showing the relative expression levels of the gene markers used to identify each cell population. (E) UMAP plot showing the cell types identified by manual curation in control COs. (F) Dot plot showing the relative expression levels of SLC16A2, THRB, and THRA in each cell population of the dorsal projection trajectory. (G) Interpretation of sequential gene expression changes during dorsal projection trajectory progression. (H) UMAP plot showing the cell types identified by manual curation in MCT8-COs. (I) Histogram of the relative number of cells in control and MCT8-COs. (J) Volcano plots showing the distribution of differentially expressed genes in MCT8-deficient versus control COs. iPSCs, induced pluripotent stem cells; COs, cortical organoids; NPCs, neural precursor cells; IPCs, intermediate precursor cells; DPT, dorsal projection trajectory; UMAP, uniform manifold approximation and projection. The χ² test was used for multiple comparisons and pairwise cell proportions. ****P < 0.0001. Differentially expressed gene thresholds: P < 0.05 and average log₂(fold change) = 0.26.

Figure 2. Spatial transcriptomics analysis of D50 Control- and MCT8-COs.

(A) Schematic representation of obtaining spatial transcriptomic data. (B) Image showing the cell segmentation on the indicated COs, based on transcript localization data. Scale bars: 100 μm. (C) Heatmap showing the relative expression levels of the genes used to identify the indicated neural cell types. (D) Heatmap of the differentially expressed genes in the indicated neural cells between control versus MCT8-COs. (E) Same as in B, except the cells are classified into the indicated types. (F) Distribution of distances between the indicated cells and groups. (G) Histogram of the distribution of cell densities. Considering a radius of 100 μm, if only 1 cell was in contact with another cell, the distribution was considered sparse. If 1 cell was in contact with 5 or more cells, it was considered dense. (H) Violin plots show the expression levels of the indicated genes, considering the indicated cellular densities. **P < 0.01, ***P < 0.001 by multiple-testing P-value adjustments using the Benjamini-Hochberg method. The analysis of the differentially expressed genes was performed using the findMarkers function and Wilcoxon’s test. (I) Ligand-receptor interaction between the indicated neural cells in control and MCT8-deficient COs. Differentially expressed gene threshold: P < 0.05; the CellPhoneDB toolkit was used to assess significant receptor-ligand dyads.

Figure 3. Control- and MCT8-deficient iPSC-derived neurons.

(A) Schematic representation of generating NPCs from iPSCs and the subsequent differentiation of the NPCs into neurons. (B) Bright-field and confocal fluorescence images showing iPSC-derived NPCs stained for SOX2 (magenta), NESTIN (yellow), MCT8 (magenta), and with DAPI (blue; nuclear). (C) Relative mRNA levels of the indicated genes in control and MCT8-NPCs after 1, 6, and 12 days of neurodifferentiation. (D) Bright-field images showing NPCs after 6 days of neurodifferentiation. (E) Principal component plot illustrating differences between control and MCT8-deficient NPCs. (F) Volcano plots showing the distribution of differentially expressed genes in control versus MCT8-NPCs; each point represents the average of 5 control and 5 MCT8-deficient samples of pooled NPCs for each transcript. (G) Heatmap depicting the top 20 differentially expressed genes related to neurogenesis between control versus MCT8-NPCs identified by bulk RNA-seq. (H) Venn comparison of differentially expressed genes belonging to the gene set related to neurogenesis between control versus MCT8-NPCs identified by bulk RNA-seq (blue) and scRNA-seq analysis (green) and between control versus MCT8-IPCs identified by scRNA-seq analysis (purple); common differentially expressed genes were identified (gray box). (I) Bright-field images of control and MCT8-NPCs after 12 days of neurodifferentiation. TUJ1, green; DAPI, blue (nuclear). (J) The upper 2 panels are MCT8 staining in red, TUJ1 in green, and DAPI (blue); the lower panels are RBFOX3 staining in green, NEUROD1 in red, and DAPI in blue. (K) MCT8-NPCs after being treated with 60 nM T3 during the 12 days of neurodifferentiation. TUJ1, green; SOX2, red. (L) SOX2 staining (red) and DAPI (blue) on the indicated cells and treatments. Scale bars: 50 μm (B) and 100 μm (D, I, and J). Differentially expressed gene thresholds: P < 0.05 and average log₂(fold change) = 1.5 in the Partek Flow platform. Expression values are mean ± SD of n = 3–6 RNA samples, each of them consisting of 2 pooled 6-well plates of NPCs from either control or MCT8-NPCs. **P < 0.01, ***P < 0.001 by 2-tailed Student’s t test for comparing DIO2 deiodination and relative mRNA expression between D1, D6, and D12 of neurodifferentiation.

Figure 4. Studies on the presence and effect of the DIO2 pathway in iPSC-derived NPCs and COs.

(A) Schematic representation of treating iPSC-derived NPCs with T4^I125 to measure T3^I125 production. (B) Representative chromatograms of the medium after control and MCT8-NPCs were incubated with T4^I125 for 24 hours. (C) Quantitation of the DIO2 deiodination in control and MCT8-NPCs; n = 5 DIO2 assays. (D) Volcano plots showing the distribution of differentially expressed genes in Control + 1 nM T4 versus Control NPCs; each point represents the average of 5 WT + 1 nM T4 and 5 WT samples of pooled NPCs for each transcript. (E) Interpretation of the findings in A–D. (F) Schematic of the generation and timing of CO generation, starting with iPSCs to a culture of embryoid bodies, followed by neural induction, neuroepithelial bud expansion, and maturation. (G) Quantitation of DIO2 deiodination in control COs during their first 20 days in culture. n = 4 DIO2 assays per time point, each consisting of 4 pooled COs from control COs. (H) Relative SOX2 mRNA levels in control COs during their first 20 days in culture. Expression values are mean ± SD of n = 3–6 RNA samples, each of them consisting of 4 pooled COs from control COs. (I) Schematic representation of the experiment. COs were treated with 1 nM T4 from D7 to D50 and then dissociated into a single-cell suspension for scRNA-seq. (J) UMAP plot showing the cell types identified. (K) Histogram of the relative number of cells in T4-COs and control COs. (L) Histograms of the relative number of cells in clusters of NPCs. The identification number of each cell cluster is indicated at the bottom right corner of each rectangle. (M) Volcano plots showing the distribution of differentially expressed genes in T4-COs versus control COs. (N) Gene set enrichment analysis reveals gene ontology terms enriched in T4-COs. (O) Histograms of the relative number of cells undergoing the indicate cell cycle phase in clusters 1 and 9 of control and NPCs. ***P < 0.001 by 2-tailed Student’s t test for comparing DIO2 deiodination in iPSC-derived NPCs. Differentially expressed gene thresholds: P < 0.05 and average log₂(fold change) = 0.26.

Figure 5. Studies on the presence and effect of the DIO2 pathway in primary mouse NPCs.

(A) Schematic representation of the protocol to isolate mNPCs, propagate them in culture, and differentiate them into neurons. (B and C) Bright-field images showing representative neurospheres containing mNPCs (B) and the mNPCs cultured on poly-D-lysine–coated plates (C). (D–F) Confocal images showing mNPCs expressing Nestin (green) and Sox2 (red) (D) and Mct8 (red) (E). The insets in E depict 2 dividing mNPCs that exhibited higher intensity of Mct8 immunofluorescence. (F) Quantitation of the DIO2 deiodination in mNPCs; n = 6 DIO2 assays. The background condition includes all assay reagents except for the cell samples. (G) Bright-field images showing representative mNPCs after 2 days of neurodifferentiation. (H–J) Relative mRNA levels of the indicated genes in mNPCs under the indicated conditions; n = 4. (K) Bright-field images showing representative mNPCs after 4 days of neurodifferentiation. Note the neuronal process extension. (L) Tuj1⁺ cells under the indicated conditions and the quantitation of the percentage of Tuj1⁺ cells. Values are the mean ± SD of 5 replicates. Scale bars: 300 μm (B) and 25 μm (C–L). **P < 0.01; ***P < 0.001 by 2-tailed Student’s t test for comparing DIO2 deiodination in iPSC-derived NPCs.