Fig. 4

Reduced H3K9ac levels at the cL1 insertion site in penetrant Axin^cL1 mice. a ChIP-qPCR analyses of H3K9ac levels using primers specific to the exon 6 splicing site of the Axin1^cL1 allele. Arrows indicate relative locations of the primers used for ChIP-qPCR analyses (upper panel). Data are presented as means ± SEM, n = 3, *p < 0.05. b ChIP-qPCR analyses of H3K9ac levels using primers specific to the exon 7 splicing site of the Axin1^cL1 and Axin1⁺ alleles. Arrows indicate relative locations of the primers used for ChIP-qPCR analyses (upper panel). Data are presented as means ± SEM, n = 3, *p < 0.05. c ChIP-qPCR analyses of H3K9ac levels using primers specific to the exon 6 splicing site of the Axin1⁺ allele. Arrows indicate relative locations of the primers used for ChIP-qPCR analyses (upper panel). Data are presented as means ± SEM, n = 3. d Schematic illustration showing the effect of reduced H3K9ac levels on splicing. Briefly, higher H3K9ac levels ensure correct splicing, which excludes cL1 from the transcript, whereas with lower H3K9ac levels, the cL1 tends to be retained and included in the transcript. e Comparison of the molecular mechanisms underlying the kinky tail phenotype between Axin^Fu and Axin^cL1 mice. The kinky tail phenotype in Axin^Fu mice results from a shorter transcript isoform initiated from intron 6, and the phenotypic severity is inversely correlated with DNA methylation status, whereas the kinky tail phenotype in Axin^cL1 mice is caused by a longer transcript isoform with cL1 intron retention, and the penetrance of the phenotype is fixed at 70–80%, and inversely correlated with H3K9ac levels at the cL1 insertion site