Cortical interneuron development is affected in 4H leukodystrophy 41 2 collected on 1 batch of neurons per cell line. All analyses were either done automated or blinded. Neuronal morphology, myelination and the number of GAD65/67+, CTIP2+, OLIG2+ and MBP+ cells were analysed using Columbus software (Perkin Elmer). Synaptic analyses and the percentage of myelinated axons were determined in ImageJ (imagej.nih.gov/ij) using the NeuronJ and SynaptoCount plugins (Supplementary Methods). Statistical analyses were done in IBM SPSS statistics software version 26 (IBM). Data was tested for normality with a Shapiro-Wilk test. Significance was tested with independent samples ttests (for parametric data) or a Mann-Whitney U test (for non-parametric data). To compare the effect of compounds on MEA data or the effect of SAG treatment paired samples t-test or Wilcoxon signed rank tests were used. All statistical tests were two-tailed and tested against alpha-level P < 0.05. Data availability The authors confirm that the data supporting the findings of this study are available within the article and its supplementary material. RESULTS In vitro products present with iPSC and cerebellar cell identity As 4H patients with POLR3B variants often have cerebellar atrophy, we differentiated iPSCs into cerebellar neurons as described previously (Holmes & Heine, 2017b). We first wanted to establish that reprogramming and differentiation products derived from patient and control fibroblasts were unique cell populations, with relatively little variation due to source material. Further, we tested the validity of the reprogramming protocols by estimating how close each derived cell type resembled published expression profiles. To investigate sample similarity, t-Distributed Stochastic Neighbor Embedding (t-SNE) was performed with perplexity at 5. Cluster analysis showed that samples group together based on cell types (fibroblast, iPSC, product) regardless of cell source (patient or control) (Fig. 1A). The relatively high similarity of samples from the same cell type, regardless of their source, suggests that reprogramming methods were effective at generating a consistent cell type. To estimate the accuracy of cell reprogramming, pair-wise Spearman’s rank correlations were computed for 23 samples from this study and two expression profiles obtained from ENCODE. One was an ENCODE iPSC expression profile (ENCSR722POQ) and the other an ENCODE granule cell expression profile (ENCSR313IUO), both containing two biological
RkJQdWJsaXNoZXIy MTk4NDMw