Fig 2: lncRNA loci are recurrently mutated in patients with neurodevelopmental disorders.(A) Schematic representation of CNV morbidity map analysis for candidate lncRNAs and all other iN lncRNAs loci. The 35 mouse lncRNA candidates (28 human loci) is from Figure 1—figure supplement 1H. (B) Top: Representative tracks for lncRNA E locus, also known as lnc-NR2F1. Depicted in blue are deletions and in red duplications. Arrow points to patient with focal deletion affecting the lnc-NR2F1 locus only. Bottom: Custom CGH arrays used to validate chromosomal aberration in patient 9900850 harboring focal deletion represented in green signal. (C) Genetic pedigree analysis for family with paternally inherited balanced chromosomal translocation (5;12) (q15;q15), including a summary of clinical features for patient CMS12200 and father. The mother has a normal karyotype. Listed in the box are the symptoms of the patients. (D) Top: Circa plot representing the pathogenic chromosomal event for patient CMS12200 involving chromosomes 5 and 12. Bottom: Representative chromosome ideogram and track of the balanced chromosomal break affecting patient CMS12200. Below the ideoplot is the schematic representation of predominant human isoforms for lnc-NR2F1 and the site of the break site disrupting the long isoforms. (E) The locations of the probes are in Figure 1—figure supplement 4C. Left: Metaphase spread from patient CMS12200 with the t(5;12) translocation showing FISH signals obtained with the clone RP11-608G16 (green) spanning Chromosome five breakpoint, and a Chromosome five telomere-specific probe (red). Middle: Metaphase spread from patient CMS12200 with the t(5;12) translocation showing FISH signals obtained with the clone RP11-597C7(green) proximal to Chromosome 12 breakpoint, and a Chromosome 12 centromere-specific probe (red). Right: Metaphase spread from patient CMS12200 with the t(5;12) translocation showing FISH signals obtained with the clone RP11-641O3 (green) distal to Chromosome 12 breakpoint, and a Chromosome 12 centromere-specific probe (red).

Fig 3: Mouse Lnc-Nr2f1KO reveals lnc-Nr2f1 regulates neuronal genes.(A) Schematic showing the experimental strategy for lnc-Nr2f1 overexpression. In control mouse ES cells, an inverted construct with a splice acceptor (marked in yellow) and a polyadenylation signal (marked in red) are added after the first exon of the lnc-Nr2f1. The mouse ES were infected with rtTA and Ngn2-T2A-puro and mES derived induced neurons (mES-iN) were assayed after 3 or 4 days after dox induction. (B) Graph showing that overexpression of lnc-Nr2f1 increased the neurite length in day 3 Ngn2 mouse ES derived iN cells relative to the Ctrl. The same effect was not seen with Nr2f1 overexpression. For each replicate, the individual neurite length for all neurons in each of the five 20x field was manually traced in Fiji. The sequence used for mouse lnc-Nr2f1 overexpression is available in the supplementary document (n = 3, Student t-test, Two-tailed, * indicates p=0.048 < 0.05). Error bars show s.e.m. (C) Graph showing that overexpression of Nr2f1 decreased the neurite number in day 3 Ngn2 mouse ES-iN cells relative to the Ctrl. The same effect was not seen with lnc-Nr2f1 overexpression. (n = 3, 10 field per replicate, Student t-test, Two-tailed, ** indicates p=0.0022 < 0.01). Error bars show s.e.m. (D) ß-III-tubulin staining of the day 3 Ngn2 mouse ES derived iN cells for Ctrl, lnc-Nr2f1 overexpression and Nr2f1 overexpression. Scale bar = 50 µm. Red arrow pointed at immature induced neuronal cells with short projection. Green arrow pointed at mature induced neuronal cells with longer projection. Note that the lnc-Nr2f1 overexpression condition have more mature induced neuronal cells. (E) Hierarchical clustering heatmap of day 4 Ngn2 ES-iN cells between control and lnc-Nr2f1 overexpression (OE). There are 1912 genes differentially expressed (n = 2, FDR corrected p<0.10, Fold change >1.5 fold). Listed to the right are genes which are upregulated upon lnc-Nr2f1 overexpression. (F) Schematic showing the knocking out strategy for lnc-Nr2f1. The lnc-Nr2f1 knockout mouse ES cells are generated after Cre recombinase introduction to the Ctrl line in Figure 3A. The mouse ES were infected with rtTA and Ngn2-T2A-puro and mES derived induced neurons were assayed after 3 or 4 days after dox induction. (G) Hierarchical clustering heatmap of day 4 Ngn2 ES-iN cells between wild type, lnc-Nr2f1 knockout (KO) and lnc-Nr2f1 knockout with lnc-Nr2f1 overexpression (OE). There are 348 genes differentially expressed and can be subsequently rescued with lnc-Nr2f1 overexpression (n = 2, FDR corrected p<0.10). Listed to the right are genes which are upregulated upon lnc-Nr2f1 KO. (H) Gene ontology of the upregulated genes in lnc-Nr2f1 knockout day 4 Ngn2 mouse ES- iN cells as compared to the Ctrl. (I) Gene ontology of the downregulated genes in lnc-Nr2f1 knockout day 4 Ngn2 mouse ES- iN cells as compared to the Ctrl.

Fig 4: lnc-Nr2f1 binds to distinct genomic loci regulating neuronal genes.(A) Schematic showing the location of ChIRP probe for mouse lnc-Nr2f1 (highlighted in red). Yellow lines represent the conserved exons between mouse and human lnc-Nr2f1. (B) Heatmaps representing genome-wide occupancy profile for mouse lnc-Nr2f1 in day 4 Ngn2 mouse ES- iN cells and the RNase control obtained by ChIRP. There are 14975 significant peaks called with respect to the RNase treated control. E and O represents even and odd probes respectively. (C) UCSC browser track showing the binding site within the intronic region of Nrp2. The ‘R’ represents the RNase treated control. (D) Gene ontology terms associated with genes adjacent to the high confident mES-iN ChIRP-seq peaks. Terms highlighted in red are terms related to nervous system development. (E) Percentage of mES-iN ChIRP-seq peaks which overlap with CTCF, enhancer, H3K27ac, H3K4Me3 and PolII defined in mouse cortex relative to the control. (*** represents p<0.0001, ** represents p<0.01, Chi-square test) (F) Proposal mechanism of lnc-Nr2f1 action. lnc-Nr2f1 binds to the genomic region enriched with bHLH motif and regulates the downstream neuronal genes.

Fig 5: Human lnc-NR2F1 shows isoform-specific chromatin binding.(A) Schematic showing the location of the ChIRP probes target the short isoform-specific exon (exon 7) and long isoform-specific exon (exon 11). The red line denotes the break point for the patient. (B) Overexpression of human lnc-NR2F1 isoforms in combination with Ascl1 relative to MEFs expressing Ascl1 alone. The graph quantifies the proportion of TauGFP positive cells with projections normalized to number of TauGFP cells. TauGFP cells with projections longer than three times the diameter of the cell body were counted and normalized to the total number of TauGFP positive cells. The sequences for human lnc-Nr2f1 isoforms is available in the supplementary documents (n = 3, Student t-test, two tailed, scale bar= * represents p<0.05, ** represents p<0.01). Error bars show s.e.m. (C) Heatmap representing genome-wide occupancy profile for domain ChIRP performed using probes specific to the long and short isoform-specific exon of lnc-NR2F1 in human ES derived neural progenitor cells (NPC). There are 4404 and 415 significant peaks called relative to the RNase control for the long and short isoform respectively. E, O and M represents even, odd and merge track respectively. (D) UCSC browser track showing the site within the promoter region of LOC90784 bound by the long isoform-specific exon (exon 11) but not the short isoform-specific exon (exon 7) (E) Bar graph showing the distribution of the 913 high confident long isoform-specific peaks. The long isoform-specific peaks are enriched in the introns, exons, promoters and enhancers but depleted in the intergenic regions. (F) Gene ontology terms associated with genes adjacent to the human ES derived NPC ChIRP-seq high confident peaks. Terms highlighted in red are terms related to nervous system development. (G) Venn diagram representing the peak associated gene overlap between the domain ChIRP of the long isoform-specific exon (exon 11) from human ES derived NPC and mouse mES-iN ChIRP. (p<0.0001, ?2 = 239.921, DF = 1, Chi square test) (H) Venn diagram representing overlap between genes involved in the autism risk and genes identified by the domain ChIRP of the long isoform-specific exon (exon 11) from human ES derived NPC. (p<0.0001, ?2 = 71.670, DF = 1, Chi square test).