Fig. 4: IMP1 directly regulates a network of microtubule genes.

a Volcano plot showing proteins that are both differentially expressed upon IMP1 knock down in neurons and whose cognate mRNA is bound by IMP1 in our CLIP experiments. Proteins significantly up- or down-regulated in IMP1 siRNA (siIMP1) vs non-targeting control (siCTRL) are highlighted in orange and pink respectively. IMP1, CLASP1 and PEBP1 (Fig. 2a) are highlighted with blue dots and blue squares. Selection was based on log2 FC < −1 (down-regulated proteins) or log2 FC > 1 (upregulated proteins) and p value < 0.05. A two-sided one-sample Student’s t test was performed, for transcripts containing at least one IMP1 binding site. For MS, n = 3 independent iPSC lines for each condition (biological replicates) were used; for iCLIP, n = 3 technical replicates from 3 independent iPSC lines (biological replicates) were used. b Number of upregulated and downregulated proteins in neurons, from (a) and NPCs, from Supplementary Fig. 4g, reported as a barplot. For MS on neuronal samples, replicates are as stated above. For MS on NPCs, n = 2 independent iPSC lines (biological replicates) for each condition were used; for iCLIP on NPCs, n = 2 technical replicates from 2 independent iPSC lines + 3 technical replicates from 1 independent iPSC line (biological replicates) were used. c Cumulative distribution plot of log2 FC in protein expression between IMP1 KD and control in neurons, as detected by MS, was plotted for RNA classes with 0, 1–2, 3–5 and 6 or more peaks in iCLIP experiments. For MS, n = 3 independent iPSC lines for each condition (biological replicates); for iCLIP, n = 3 technical replicates from 3 independent iPSC lines (biological replicates). The reported p values were calculated using a two-sided Kolmogorov–Smirnov test comparing 1–2 peaks, 3–5 peaks or 6 and + peaks to 0 peaks. d Number of IMP1 peaks per gene detected by iCLIP for GO categories related to microtubules, neuronal or other pathways classified by PANTHER. Number of genes in each category is in the same range (between 26 and 30). Data presented as boxplots where the centre line is the median, limits are the interquartile range and whiskers correspond to 1.5 times the interquartile range. Outliers are not displayed for clarity. Red dots represent the mean. The categories “Cytoskeleton organization”, “Microtubule-based process”, “Neuron projection development” contain many common genes. Black dots represent transcripts which are present in more than one of these categories, while orange dots represent transcripts in a single category. “Positive regulation protein kinase activity” with “Small molecule biosynthetic process” include different transcripts, represented by blue dots. Each category is represented by a number below the corresponding bar. P values were calculated using a two-sided Mann–Whitney U test and added to the plot. Comparisons were performed within each cell type. e STRING analysis of the interaction network between proteins that are downregulated in IMP1 siRNA treated neurons, where the cognate encoding transcripts are also bound by IMP1 (as identified in 4a). The analysis reports on in the GO category “microtubule-based process” (PANTHER). Network protein-protein interaction (PPI) enrichment p value < 1.0e–16. P value was calculated using a hypergeometric test. Source data are provided as a Source Data file.