Figures & data
Figure 1. Recurrent mutations of the TCF3 gene in BL affect TCF3 alternative splicing.
(A) Top: TCF3 alternative splicing. Inclusion of exon 18a produces E12, and inclusion of exon 18b produces E47. Bottom: Recurrent mutations of TCF3 in BL. Such mutations are underlined and shown in red arrows and letters. (B) Effects of BL-related TCF3 mutations on TCF3 alternative splicing. Splicing of transcripts produced from the transfected mutated TCF3 minigenes was determined by 32P RT-PCR. The exon 18a inclusion (E12) percentages were calculated by dividing exon 18a signal by the sum of exons 18a and 18b signals. Results are indicated in a bar graph with means ± SD from three independent experiments. (*) p < 0.05, (**) p < 0.01.
Figure 2. Isoform-specific regulation of TCF3’s downstream target genes PTPN6 and CCND3.
(A) Left: TCF3 knockdown efficiency in Raji cells determined by 32P RT-PCR. Right: Expression of PTPN6 and CCND3 in Raji cells after isoform-specific TCF3 knockdown by RT-qPCR. (B) Left: TCF3 knockdown efficiency in 293T cells determined by 32P RT-PCR. Right: Expression of PTPN6 and CCND3 in 293T cells after isoform-specific TCF3 knockdown by RT-qPCR. (C) Expression of PTPN6 and CCND3 in Namalwacells after isoform-specific TCF3 knockdown by RT-qPCR. Data were normalized to ACTB and GAPDH. Results are indicated in bar graphs with means ± SD from three independent experiments. (*) p < 0.05, (**) p < 0.01.
Figure 3. Recurrent TCF3 mutations can disturb hnRNPH1 binding to exon 18b.
(A) Knockdown efficiency of hnRNPH1 in Raji cells by Western blot. The numbers indicate the related hnRNPH1 signal normalized to β-Actin (ACTB). (B) Top: E12/E47 mRNA expression in Raji cells determined by 32P RT-PCR after hnRNPH1 knockdown. Bottom: Averages of E12 percentage and quantification bar graph. (**) p < 0.01. (C) Expression of PTPN6 and CCND3 in Raji cells determined by RT-qPCR after hnRNPH1 knockdown. Data were normalized to ACTB and GAPDH. (**) p < 0.01. (D) Representative results of the gel shift assay showing reduced hnRNPH1 binding to exon 18b probe containing G1663C mutation compared with wildtype (WT) control. (E) Binding curves of hnRNPH1 to exon 18b probes containing the indicated mutations. Three independent experiments were performed. (F) Bar graph of Kd values for each mutation calculated from Fig. 3E. Data are shown as means ± SEM.
Figure 4. Summary of the effects of recurrent TCF3 mutations in BL, and a model of their effects on TCF3 AS regulation.
(A) Summary of the effects of BL-related recurrent TCF3 mutations. Six recurrent TCF3 mutations can alter TCF3 AS to produce more E47 (first row). Among them, G1663 C and G1681A tend to reduce hnRNPH1 binding (darker shaded box). Some of them also affect TCF3 protein function (last two rows). (B) A model explaining the effects of BL-related TCF3 mutations on TCF3 AS. Top: TCF3 AS is regulated by long-range cooperative actions of hnRNPH1/F, PTBP and DDX21, on intronic/exonic splicing regulatory elements, to achieve a balanced E12/E47 ratio. Bottom: G1663C and G1681A mutations tend to reduce hnRNPH1 binding, thus increasing E47 levels. Other recurrent mutations might increase E47 levels by altering interactions with other splicing regulators.
