MCT8 deficiency in Purkinje cells disrupts embryonic chicken cerebellar development

  1. Veerle M Darras1
  1. 1Laboratory of Comparative Endocrinology, Department of Biology, KU Leuven, Leuven, Belgium
  2. 2Medical Research Council Centre for Developmental Neurobiology, King’s College London, London, UK
  1. Correspondence should be addressed to V M Darras; Email: Veerle.Darras{at}
  1. Figure 1

    In vivo monitoring of transfection and validation of MCT8 knockdown. (A) Co-expression of RFP allows accurate in vivo monitoring of the transfection efficiency in the left cerebellar lobe, as checked at E6. (B) Scheme of the dissected hindbrain depicting the two cerebellar lobes, which are unfolded (white arrows in C) and coronally sectioned (dotted line). (C) The transfected side (TRS) is marked by a diffuse RFP signal, whereas the opposing side is untransfected (UTS). Scale bar: 200 µm. (D, E and F) The majority of transfected cells migrate from the ventricular zone towards the cerebellar mantle region (delineated by white dotted line). ISH analysis after electroporation with the control RFP vector reveals strong MCT8 mRNA expression (black arrow) at both the TRS (white arrow) and UTS and excludes the possible influence of the electroporation procedure. Scale bar: 50 µm (G, H and I) Transfection with the MCT8-RNAi vector results in a clear reduction of MCT8 mRNA expression (black dotted arrow) as seen at the TRS (white dotted arrow). Scale bar: 50 µm. (J, K and L) Pronounced RFP signal is still found in the more mature cerebellum at E18 in the soma as well as the dendritic tree of PCs, the latter being visualised by co-electroporation of the pCAβ-eGFP-m5 plasmid. Scale bar: 20 µm. C, caudal; Cb, cerebellum; D, dorsal; F, frontal; H, hindbrain; IV, fourth ventricle; TRS, transfected side; UTS, untransfected side; V, ventral.

  2. Figure 2

    Decreased RORα and LHX1/5 expression in MCT8-RNAi-transfected PCs at E6. (A and B) Post-mitotic PCs at the TRS of the E6 cerebellum show markedly reduced RORα mRNA expression (black arrow) in contrast to expression at the UTS, indicating a negative influence of MCT8 knockdown on early PC dendritic differentiation. Scale bar: 50 µm. (C, D and E) Immunolabelling of the PC-specific differentiation marker LHX1/5 (green) reveals strong overlap of LHX1/5- and RFP-expressing cells (visible as yellow cells) in the control condition (n = 6), as seen on detailed pictures of the TRS and UTS (black dotted boxes in C). (F, G and H) The fraction of LHX1/5-positive cells in the MCT8-RNAi-transfected population (n = 5) is significantly decreased as compared to that in the control RFP-transfected population (***P < 0.001, two-tailed Student’s t-test). High magnification views (white boxes) show yellow cells as a result of strong overlap of LHX1/5 and RFP signal (white arrows). This overlap is greatly reduced in the MCT8-RNAi condition compared to the control condition. Scale bar: 20 µm. L, left; R, right; TRS, transfected side; UTS, untransfected side.

  3. Figure 3

    TA3 administration at E4 and E5 partially rescues LHX1/5 expression in MCT8-RNAi-transfected cells at E6. (A, B and C) Administration of TA3 at E4 and E5 after electroporation of the control RFP vector at E3 did not increase the number of transfected LHX1/5-expressing cells at E6 (n = 7) as compared to the vehicle solution only (n = 5, P = 0.68, two-tailed Student’s t-test). (D, E and F) In contrast, administration of TA3 partially restored the fraction of LHX1/5-expressing cells within the MCT8-RNAi-transfected population (n = 8) as compared to vehicle (n = 10, ***P < 0.001, two-tailed Student’s t-test). Scale bar: 20 µm.

  4. Figure 4

    Reduced GC precursor proliferation and EGL thickness in the E10 cerebellum after MCT8-RNAi transfection. (A) EdU-positive cells (green) identify the EGL in the dorsolateral cerebellum (white arrows), whereas PCs expressing the control RFP vector or the MCT8-RNAi vector (red) reside exclusively in the underlying cerebellar anlage. Scale bar: 200 µm. (B, C, D and E) High magnification views of TRS and UTS (white dotted boxes in A) clearly show the EdU-positive cells in the EGL (region enclosed by white dotted lines in B, C, D and E) in both conditions. Scale bars: 20 µm. (F and G) EGL thickness as well as the number of EdU-positive cells in the EGL is similar in the UTS of both conditions and in the TRS of the cerebellum transfected with the control RFP vector, but is significantly lower after transfection with the MCT8-RNAi vector (n = 5–7, **P < 0.01, two-way repeated measures ANOVA, followed by a Bonferroni post hoc test). (H and I) PAX6 expression (green) is less prominent at the TRS in the MCT8-RNAi condition. Scale bar: 50 µm. Cb, cerebellum; IV, fourth ventricle; TRS, transfected side; UTS, untransfected side.

  5. Figure 5

    Stalled maturation and migration of post-mitotic GCs in the MCT8-RNAi-transfected cerebellum at E18. (A and D) EGL thickness (white line) was measured in the inner region of folium V to VII. (B and E) The number of DAPI-positive cells (white arrows) in the ML (between white dotted lines) was counted in regions characterised by a high amount of transfected PCs (white dotted boxes in A and D) as measure for presumptive migrating granule cells. (C and F) Thickness of the Axonin-1-positive layer (white line) was measured and the number of DAPI-positive cells within this layer was counted on confocal images. Scale bar: 50 µm (A and D); 20 µm (C and F). (G, H, I and J) Thickness of the EGL and Axonin-1-positive layer as well as the number of DAPI-positive cells in the ML and the Axonin-1-positive layer of the EGL were significantly increased in the MCT8-RNAi condition (n = 8) as compared to controls (n = 9) (*P < 0.05 and **P < 0.01, two-tailed Student’s t-test). EGL, external germinal layer; IGL, internal granular layer; ML, molecular layer; PCL, Purkinje cell layer.

  6. Figure 6

    Smaller and less complex dendritic tree of MCT8-RNAi-transfected PCs at E18. (A and B) Representative images showing the dendritic tree (visualised by co-injection of the pCAβ-eGFP-m5 plasmid) in control and MCT8-deficient PCs. Scale bars: 20 µm. (C) The dendritic tree area was significantly reduced in the MCT8-RNAi condition (n = 5) as compared to the control condition (n = 6, **P < 0.01, Mann–Whitney U test). (D) The length of the longest dendrite, measured from the bottom of the primary dendrite up to the highest point, served as a measure for PC height, and was significantly smaller compared to the control condition (*P < 0.05, unpaired two-tailed Student’s t-test). (E) The sum of total intersections was markedly lower in the MCT8-RNAi condition as determined by Sholl analysis (*P < 0.05, unpaired two-tailed Student’s t-test). (F) Further analysis revealed less branching points at 70 (***P < 0.001), 80 (***P < 0.001) and 90 µm (*P < 0.05) distance from the PC soma centre (control: n = 6, MCT8-RNAi: n = 5, two-way repeated measures ANOVA, followed by a Bonferroni post hoc test).

  7. Figure 7

    Theoretical model summarising the importance of MCT8 in PCs for cell-autonomous and non-autonomous development. During cerebellar development, T3 is taken up by MCT8-expressing Purkinje cells (PCs). T3 excess is prevented by deiodinase type 3 (D3)-dependent inactivation, whereas a sufficient amount can bind to the TH receptor α1 (TRα1), ultimately regulating the expression of TH-responsive factors like RORα and LHX1/5, which govern cell-autonomous PC differentiation. In the absence of MCT8, TA3 can enter the PC via a presently unknown transporter and mimic T3 action. The nuclear transcription factor RORα also affects the downstream expression of sonic hedgehog (SHH), a diffusible factor that stimulates granule cell precursor (GCP) proliferation in the external germinal layer (EGL). GCPs and post-mitotic granule cells (GCs) express PAX6, whereas GCs in the initial stage of migration express Axonin-1 (Ax-1) and mature during inward migration to form the internal granular layer.

| Table of Contents