Fig. 5

Deletion of the trans J element increases expression frequency from the intact J allele. a Schematic illustrations of the Olfr545 alleles of mice used in b and c. One allele contains the intact J element and tauGFP-tagged Olfr545, and another allele contains the intact or deleted J element and untagged Olfr545. The white box represents the IRES sequence. b Confocal images of two-color ISH for Olfr545 (magenta) and GFP (green). The bar graph shows the percentage of co-labeled cells: 124 co-labeled cells and 107 Olfr545 single-positive cells in J/J (54,3%, from 3 mice), 213 co-labeled cells and no Olfr545 single-positive cells in J/ΔJ mouse (100%, from 3 mice). The intact J element of the GFP-tagged wild-type allele did not rescue the defective phenotype of Olfr545 expression from the mutant allele. The scale bar is 20 μm. Fisher’s exact test; p < 2.2 × 10⁻¹⁶, ***p < 0.001. c Whole-mount confocal images of the turbinate. The bar graph shows the numbers of GFP-positive OSNs on the medial face of the turbinate in half-head. The number of GFP-labeled OSNs of J/ΔJ is ~ 2 times larger than in J/J mice. The error bars represent the mean ± s.e.m. (n = 6 for J/J, n = 5 for J/ΔJ). The scale bar is 500 μm. Unpaired t-test, p = 0.0046, **p < 0.01. d Percentage of OSNs expressing class I genes in ΔJ heterozygous (J/ΔJ) or homozygous (ΔJ/ΔJ) mice compared to wild-type (J/J, set as 100%). The error bars represent the mean ± s.e.m. (n ≥ 4). One-way ANOVA in each Olfr probes. Tukey’s post hoc multiple comparisons test shows significant difference between indicated conditions. *p < 0.05, **p < 0.01, ***p < 0.001. Quantification data are summarized in Supplementary Table 2