Modeling Psychomotor Retardation using iPSCs from MCT8-Deficient Patients Indicates a Prominent Role for the Blood-Brain Barrier

Abstract

Inactivating mutations in the thyroid hormone (TH) transporter Monocarboxylate transporter 8 (MCT8) cause severe psychomotor retardation in children. Animal models do not reflect the biology of the human disease. Using patient-specific induced pluripotent stem cells (iPSCs), we generated MCT8-deficient neural cells that showed normal TH-dependent neuronal properties and maturation. However, the blood-brain barrier (BBB) controls TH entry into the brain, and reduced TH availability to neural cells could instead underlie the diseased phenotype. To test potential BBB involvement, we generated an iPSC-based BBB model of MCT8 deficiency, and we found that MCT8 was necessary for polarized influx of the active form of TH across the BBB. We also found that a candidate drug did not appreciably cross the mutant BBB. Our results therefore clarify the underlying physiological basis of this disorder, and they suggest that circumventing the diseased BBB to deliver active TH to the brain could be a viable therapeutic strategy.

Keywords: MCT8; T3; blood brain barrier; disease model; iPSC; induced pluripotent stem cells; monocarboxyl transporter 8; neuronal maturation; thyroid; thyroid hormone.

Figure 1.. Properties of iPSC-Derived MCT8-Deficient Neural Cells

(A) Scheme for differentiating iPSCs to neural cultures composed of neural progenitors, astrocytes, and neurons. (B) Immunocytochemistry on CS03iCTR control line demonstrates that βIII-Tubulin neurons (red) and GFAP glia (red) produce MCT8 protein (green), with a DAPI nuclei stain (20× magnification). (C) Western blot confirms MCT8-deficient and functional MCT8 neural cultures produce MCT8, while the mutant isogenic line does not produce detectable MCT8. GAPDH housekeeping protein used as a loading control. (D and E) RNA-seq analysis shows RNA expression of (D) TH transporters and (E) deiodinases (DIO) in MCT8-deficient and functional MCT8 neural cultures. Data are expressed as fragments per kilobase of transcript per million mapped reads (FPKM). There were no significant differences among lines. (F-I) Accumulation or clearance rates of 1 nM radiolabeled T₃ or T₄ were quantified in MCT8-deficient and control neural cultures. Levels are expressed as percent of cellular T₃ or T₄ normalized to protein amount per minute. (F) T₃ uptake (ANOVA, n = 5, p < 0.0077), (G) T₃ efflux (ANOVA, n = 5, p < 0.0479), (H) T₄ uptake (ANOVA, n = 4, p < 0.0099), and (I) T₄ efflux (ANOVA, n = 3, p < 0.0309) were all significantly reduced in MCT8-deficient neural cells compared to control neural cells. (J) qRT-PCR quantification of T₃-induced KLF9 gene expression showed a significant linear relationship between mRNA expression and T₃ concentrations (at 0.1 to 1 nM) for both the healthy control (CS03iCTR, r = 0.99122, p < 0.0001) and MCT8-deficient (CS58iMCT8, r = 0.98346, p < 0.0001) neural cells. No statistically significant differences were observed in linear regression across cell lines (ANOVA). (K) qRT-PCR quantification of T₃-induced HR mRNA showed a significant linear relationship between mRNA expression and T₃ concentrations (at 0.1 to 1 nM) for both the healthy control (CS03iCTR, r = 0.93436, p < 0.0001) and MCT8-deficient (CS58iMCT8, r = 0.94338, p < 0.0001) neural cells. No statistically significant differences were observed in linear regression across cell lines (ANOVA). See also Figures S1–S3 and Tables S1 and S2.

Figure 2.. MCT8-Deficient Cells Have Normal Differentiation and T₃-Dependent Neuronal Maturation

(A) Quantifying immunocytochemistry-stained cells showed that MCT8-deficient and control iPSCs had similar neural differentiation potential into Nestin+ progenitor cells, GFAP+ astrocytes and βIII-Tubulin+ neurons. Error bars represent SD. (B) Scheme for differentiating iPSCs into neural cultures, in the presence or absence of T₃ (100 nM) throughout 30–40 days. (C) Mean of total spikes/5 min in the MCT8 intact (control) CS02iCTR, CS03iCTR and CS01iMCT8^cor lines differentiated in the presence of T₃ (solid line) was significantly higher than in the absence of T₃ (dashed line) (ANOVA, n = 3, p < 0.0384). (D) Mean of total spikes/5 min in the CS01iMCT8, CS58iMCT8 and CS03iCTR^mut MCT8-deficient lines differentiated in the presence of T₃ (solid line) was significantly higher than in the absence of T₃ (dashed line) (ANOVA, n = 3, p < 0.014). (E) Mean of total bursts/5 min in the MCT8 intact (control) lines in the presence of T₃ (solid line) was significantly higher than in the absence of T₃ on the third week of differentiation (t test, n = 3, p < 0.01). (F) Mean of total bursts/5 min in the MCT8-deficient lines in the presence of T₃ (solid line) was not significantly higher than in the absence of T₃ (dashed line). (G) Volcano plot shows differences of expression patterns in healthy control cells in response to T₃ treatment. The dotted lines represent 1.5-fold differentially expressed genes by paired Student’s t test followed by Benjamini Hochberg correction for multiple comparisons at a false discovery rate (FDR) of 0.1. See Tables S3 and S4 for complete list. See also Figure S4 and Tables S3 and S4.

Figure 3.. Human Brain Microvascular Endothelial Cells Express MCT8

(A) MCT8 RNA expression in immunopanned purified human cells from the Barres lab database. Data are expressed as FPKM. (B) Immunohistochemisty of post-mortem cerebellum from a healthy human shows that MCT8 (purple) is expressed in microvessels in the granular cell layer (GCL), purkinje cell layer (PCL) and in the molecular layer (ML). White arrows denote microvessel-like structures. No antibody was used as a negative control (20× magnification). (C) Immunofluorescence shows MCT8 protein (green) within Glucose transporter 1 (GLUT1, red)-expressing BMECs (white arrows), and within purkinje cells (yellow arrows). Scale bar, 10 μm. (D) Scheme to differentiate iPSCs into BMECs. Colonies are differentiated for 6 days in unconditioned media (UM) and then expanded for 2 days in endothelial cell media (ECM). Differentiated cultures contain mixed colonies of neural and endothelial cells. Colonies are replated onto collagenIV/fibronectin-coated tissue culture plates or Transwell inserts at day 8 of differentiation, yielding purified iBMEC monolayers. (E) Immunocytochemistry shows that purified iBMEC monolayers produce MCT8 (green) and GLUT1 (red) (20× magnification). (F) Western blot confirms purified iBMECs produce MCT8. GAPDH housekeeping protein was used as a loading control. (G and H) RNA-seq analysis shows RNA expression of (G) TH transporters and (H) deiodinases in purified human BMECs (green bar) and purified iBMECs. Data are expressed as FPKM. No significant differences were found between MCT8-deficient and control iBMECs. Error bars represent SD.

Figure 4.. Characterization of Patient-Derived iBMECs

(A) Immunocytochemistry shows the BBB markers PECAM1, GLUT1, CLAUDIN5, and OCCLUDIN in healthy control (CS03iCTR and CSi00CTR) and in MCT8-deficient (CS01iMCT8 and CS58iMCT8) iBMECs (20× magnification). (B) RNA-seq analysis depicting RNA expression of BBB marker GLUT1 and TJ molecules in control and MCT8-deficient iBMECs. No statistically significant differences were observed across cell lines. (C) Quantification of transendothelial electrical resistance (TEER) showed no statistically significant differences between control (CS03iCTR, CS00iCTR, and CS01iMCT8^cor) and MCT8-deficient (CS01iMCT8, CS58iMCT8, and CS03iCTR^mut) iBMECs. (D) Schematic of the iBMEC monolayers in a Transwell system. Fluorescent or radiolabeled molecules are added to the apical (blood) side, and the diffusion is monitored on the basolateral (brain) side over time. (E) Quantification of the paracellular permeability of sodium fluorescein showed no statistically significant differences between control (CS03iCTR, CS00iCTR, and CS01iMCT8^cor) and MCT8-deficient (CS01iMCT8, CS58iMCT8, and CS03iCTR^mut) iBMECs. Error bars represent SD. See also Figures S5 and S6 and Table S5.

Figure 5.. MCT8 Role in the Transport of THs across the BBB

(A) Quantifying uptake rate of radiolabeled T₃ (1 nM) showed that MCT8-deficient iBMECs (CS01iMCT8 and CS58iMCT8) displayed significantly lower uptake rates compared to control iBMECs (CS00iCTR and CS03iCTR) (ANOVA, n = 4, p < 0.0138). (B) Measuring T₄ (1 nM) uptake rate into iBMECs showed that MCT8-deficient iBMECs displayed significantly lower uptake rates than control iBMECs (ANOVA, n = 4, p < 0.0001). (C) Measuring T₃ transport across iBMECs showed that the permeability of radiolabeled T₃ (1 nM) across MCT8-deficient iBMECs was significantly lower than control cells (ANOVA, n = 4, p < 0.0208). (D) Measuring T₄ (1 nM) transport across iBMECs showed that the permeability of 1 nM radiolabeled T₄ across iBMECs was not significantly different across cell lines. (E) After correcting for paracellular diffusion using a fluorescein tracer, the net permeability or trans-iBMEC component of radiolabeled T₃ was abolished with MCT8-deficient cells and significantly differed from the T₃ transport across control cells (ANOVA, n = 4, p < 0.0358). (F) The net permeability of radiolabeled T₄ across MCT8-deficient iBMECs was not significantly different from controls. Error bars represent SD. See also Figure S7.

Figure 6.. TH Transport across Patient-Derived and Isogenic-Derived iBMECs by LC-MS/MS

(A) The apical-to-basolateral transport of T₃ at 1 nM across MCT8-deficient lines (CS58iMCT8, CS01iMCT8, and CS03CTR^mut, denoted by “b”) was significantly lower than in cells containing a functional MCT8 (CS00iCTR, CS03iCTR and CS01iMCT8^cor, denoted by “a”) (ANOVA, n = 4, p < 0.0001). Pairwise comparisons showed significant differences between the isogenic lines CS03iCTR and CS03iCTR^mut (p < 0.0012) and CS01iMCT8 and CS01iMCT8^cor (p < 0.0001). (B) The apical-to-basolateral transport of T₄ at 1 nM across MCT8-deficient lines was not significantly different than transport across cells containing a functional MCT8. (C) Measuring T₃ efflux from iBMECs showed that MCT8-deficient cells (b) displayed significantly lower efflux rates than cells with a functional MCT8 (a) (n = 4, ANOVA, p < 0.0001). Pairwise comparisons showed significant differences between the isogenic lines CS03iCTR and CS03iCTR^mut (p < 0.0027), CS01iMCT8 and CS01iMCT8^cor (p < 0.0053). (D) Measuring T₄ efflux from iBMECs showed that MCT8-deficient cells (b) displayed significantly lower efflux rates than cells with a functional MCT8 (a) (n = 4, ANOVA, p < 0.001). Pairwise comparisons showed significant differences between the CS01iMCT8 and the CS01iMCT8^cor isogenic lines. (E) Scheme of experimental design. Transport from the apical (blood) side to the basolateral (brain) side corresponds to the forward (Fwd) direction. Reverse (Rev) direction corresponds to the basolateral to apical transport. (F) Testing the bidirectional transport of 1 nM T₃ across iBMECs showed a statistically significant difference between Fwd and Rev directions in functional MCT8 lines (CS02iCTR, p < 0.03; CS03iCTR, p < 0.03; and CS01iMCT8^cor, p < 0.0002) and in CS58iMCT8 (p < 0.01). No significant differences were observed across cell lines in the Rev direction. (G) DITPA (100 nM) transport across iBMECs did not differ between MCT8-deficient and control lines and was ~100-fold lower compared to T₃ transport in control lines. Error bars represent SD.

Figure 7.. MCT8 Deficiency Leads to Reduced Transport of T₃ from the Blood-stream into the Brain

To reach the human CNS, T₃ crosses from the bloodstream through the blood-brain barrier (BBB) via MCT8-dependent transport by brain endothelial cells. T₃ is then transported through MCT8 (black) and other TH transporters (gray) expressed on astrocyte and neuronal cell membranes, where it is thought to promote neuronal maturation. In the MCT8-deficent brain, T₃ is unable to cross the BBB, resulting in T₃ starvation of the brain, which in turn results in reduced neuronal maturation.