Figure 1. The TS mutation alters CACNA1C mRNA splicing and induces persistent expression of mutant channels in differentiating human neurons.

(a) qRT-PCR quantifying relative abundance of Cacna1c exons 8 and 8A in developing mouse cortex (n = 3 mice per timepoint from two different litters; data presented as mean ± s.e.m.; ****p<0.0001, two-way ANOVA and post-hoc Bonferroni). (b) Representative fluorescence ISH images of coronal sections through developing mouse brain at E14 depict strong expression of Cacna1c-8A transcripts in neurogenic zones lining the ventricles, in newborn neurons of the CP, and in the developing striatum. VZ, ventricular zone; SVZ, subventricular zone; IZ, intermediate zone; CP, cortical plate; GE, ganglionic eminence; and str, striatum. Scale bars, 200 μm (upper), 100 μm (lower). (c) qRT-PCR on RNA from differentiating human iPSC-derived NPCs. CACNA1C-8 transcripts are upregulated during neuronal differentiation in control cultures (NPCs: two subjects and H9 ES line, four lines total; neurons: two subjects and H9 ES line, five lines total; normalized data (to GAPDH) presented as mean across lines ± s.e.m.; **p<0.005, two-way ANOVA and post-hoc Bonferroni). (d) qRT-PCR on RNA from differentiating human TS iPSC-derived NPCs or neurons demonstrates that upregulation of exon 8 is abrogated during neuronal differentiation of TS patient-derived neurons (TS NPCs and neurons: two patients, four lines; T7643-5, T7643-7, T7643-32, and T9862-42; normalized data (to GAPDH) presented as mean across lines ± s.e.m.; n.s., not significant, two-way ANOVA). (e) The relative abundance of CACNA1C exons 8 and 8A in NPC cultures is shown separately for H9 ES line, three lines from two healthy individuals (IM23-9, NH1-1 and NH2-6), and four TS lines from two individuals (T7643-5, T7643-7, T7643-32, and T9862-42). Data points indicate individual differentiations. In all TS lines examined, exon 8A is more highly expressed than exon 8 in NPCs. (f) The ratio of exon 8A to exon 8 decreases in differentiating control neurons, while TS NPCs show an increased exon 8A/exon 8 ratio (data presented as mean ± s.e.m.; **p<0.005, ***p<0.001, ****p<0.0001, one-way ANOVA and post-hoc Bonferroni). (g, h) Single-cell qRT-PCR using Fluidigm arrays of neuronal cultures at day 45 of differentiation reveals a greater proportion of neurons expressing CACNA1C-8A in TS patients compared to controls (g), (n = 125 control neurons, n = 140 TS neurons from three control and three patient lines; ***p<0.001, χ² = 27.36, Chi-square test). (h) The percentage of neurons in patients and controls remains the same, as assessed by NCAM expression (Fluidigm arrays; data presented as mean ± s.e.m., p=0.66, n.s., not significant, unpaired t-test). (i) A working model depicting the splicing shift caused by the TS mutation in a schematized version of the CACNA1C genomic locus spanning exons 7 to 9.

Figure 1—figure supplement 1. Cacna1c exon 8 and 8A expression in mouse and human.

(a) Quantitative RT-PCR quality controls on microdissected rostral cortex across embryonic and postnatal brain development. qRT-PCR was performed on RNA extracted from the developing cerebral wall at seven embryonic and postnatal time points (n = 3 mice per time point from two different litters). Depicted here are quality controls demonstrating progressive decline of the progenitor markers Nestin and Pax6 across cortical development (data presented as mean ± s.e.m; *p<0.05, ***p<0.001, ****p<0.0001, significantly different by one-way ANOVA and post-hoc Bonferroni test; for simplicity, we only indicate significance with asterisks for comparisons with immediately preceding time point), while the GABAergic neuron-specific enzyme Gad67 progressively increases as GABAergic interneurons migrate into the cortex (data presented as mean ± s.e.m; *p<0.05, **p<0.01, significantly different by Kruskal-Wallis test and post-hoc Dunn’s correction, mean rank of each time point compared to E11.) (b) qRT-PCR specificity control for Cacna1c exon-specific primers enabling direct comparison of exon 8- and exon 8A-containing transcripts. Depicted here are serial dilutions of exon 8A-containing template. Note the linear amplification using an exon 8A-specific primer and the flat trace for the exon 8-specific primer. (c) Conversely, the specificity control using an exon 8-containing template depicts a flat trace for the exon 8A-specific primer and linear amplification using the exon 8-specific primer. (d) The ratio of Cacna1c exon 8A to exon 8 progressively decreases in the developing mouse cortex over the course of embryonic development. (n = 3 mice per time point; data presented as mean ± s.e.m.; ***p<0.001, **p<0.005, significantly different by one-way ANOVA and post-hoc Bonferroni test). (e, f) The abundance of exon 8A in Cacna1c transcripts declines at the single-cell level between E18 and P14. (e) Single-cell qRT-PCR using Fluidigm dynamic arrays reveals elevated levels of exon 8A at the single-cell level at embryonic day 18 (E18) as compared to exon 8. (n = 91 cells; individual data points correspond to single cells, bar reflects the median; ****p<0.0001, significantly different by Mann-Whitney). (f) By P14, the abundance of exon 8A in Cacna1c transcripts at the single cell level is dramatically reduced. (n = 91 cells; individual data points correspond to single cells, bar reflects the median; ****p<0.0001, significantly different by Mann-Whitney). (g, h) Cacna1c transcripts containing exon 8A are enriched in neurogenic zones in the developing cerebral wall. (g) Representative images of raw data imaged at the same exposure from fluorescence ISH experiments performed on coronal sections through the mouse brain at embryonic day 14 (E14) depict strong expression of exon 8A (top), particularly in the ventricular and subventricular zones. Exon 8 (bottom) is more broadly expressed. Scale bar, 200 μm. (h) Representative fluorescence ISH image of a coronal section through the mouse brain at E18 demonstrates strong expression of transcripts containing exon 8A in neurogenic zones of the dorsal and ventral telencephalon and in newborn neurons of the cortical plate. VZ, ventricular zone; SVZ, subventricular zone; IZ, intermediate zone; CP, cortical plate; 3V, third ventricle; and LV, lateral ventricle. Scale bar, 200 μm. (i) CACNA1C transcripts containing exon 8 are upregulated in human fetal cortex as development progresses. qRT-PCR on RNA from fetal cortex at two different embryonic time points and in adult frontal cortex shows that exon 8 is upregulated as fetal cortical development progresses. (n = 2, GW12-14; n = 2, GW22-26; n = 1 adult; for fetal time points, data presented as mean ± s.e.m). (j) The ratio of CACNA1C exon 8A to exon 8 in differentiating human NPCs is depicted in this plot. Purple bars represent merged control and TS lines at the NPC and neuron stages. Blue circles indicate TS lines, whereas yellow squares reflect controls. (As in Figure 2, control lines: H9 ES line and two subjects, four lines total for NPCs and five lines total for neurons; TS lines: two individuals with TS, four lines total for both NPCs and neurons; data presented as mean ratio across lines ± s.e.m.; ***p<0.001, significantly different by Mann-Whitney).