Extended Data Fig. 9: Relationship between Sox2 and EC cell identity.

(a) Representative expression of endogenous Sox2 protein in the antrum of mouse stomach by immunofluorescence (left). Colocalization of Sox2 and TdTomato in the antrum of Sox2^TdTomato reporter mice by immunofluorescence (right). Higher magnification of the isthmus region is shown on the right of the image. Closed arrowheads point to cells with lower levels of endogenous Sox2 staining in the nucleus that is colocalized with cytoplasmic TdTomato signal. Open arrowheads point to cells with high levels of nuclear Sox2 expression and high cytoplasmic TdTomato signal, scale bars, 50 μm, n = 5 animals examined. (b) Colocalization of Sox2 and TdTomato in the corpus of Sox2^TdTomato reporter mice by immunofluorescence. Higher magnification of the base and isthmus region are shown on the right. Closed arrowheads point to nuclear Sox2 staining identified in both the isthmus and base of corpus glands, scale bars, 50 μm, n = 5 animals examined. (c) RNA sequencing of primary Sox2^neg and Sox2^pos cells taken from the corpus. RPKM values of gland base/chief markers Lgr5, Pgc, Gif, and Tnfrsf19 are plotted data are presented as mean values, n = 3 biologically independent samples. (d) Sox2 expression levels in sorted Sox2^neg, Sox2^pos, and Sox2^high cells by RNA sequencing data are presented as mean values, n = 3 biologically independent samples. (e) RNA sequencing of primary Sox2^pos and Sox2^high cells taken from the antrum. RPKM values of isthmus markers Hmgb2, Mki67, and Set are plotted data are presented as mean values, n = 3 biologically independent samples. (f) Single-cell gene expression of ChgA and/or ChgB in primary antrum EE cells embedded using UMAP embedding. EC cell population is circled. (g) MDS plot comparing Sox2^high and Sox2^low cells from the glandular stomach to ChgA^pos and ChgA^neg cells in the intestine. (h) Representative image showing colocalization of SOX2 and 5⁻HT expression in human antrum by immunofluorescence. Higher magnification of outlined area is shown below. Closed arrowheads point to cells that are double positive for both nuclear SOX2 and cytoplasmic 5-HT, scale bars, 100 μm, n = 3 biologically independent experiments. (i) Percentage of 5-HT positive cells relative to total epithelial nuclei present in the field in the stomachs shown in Fig. 7j. Data are presented as mean values + /- SD, n = 3 biologically independent samples for each group, 5 fields quantified per replicate.