Loss of GATA6 from the ileum disrupts the intestinal epithelial structure and enterocyte gene expression. (A) H&E staining of ileum from control and Gata6 cKO embryos showed that mutant villi appeared shorter than those of controls. Average ileal villus length in control and GATA6 mutant ileum was determined using NDP Scan software (Hamamatsu) to measure the length of all villi per section (G6 CTL, Gata6loxP/+,117 villi,103.8 ± 7.0 μm; G6 cKO, Gata6loxP/-Villin-Cre, 88 villi, 76.4 ± 2.0 μm; n = 3 embryos/genotype, 3 sections/embryo). Villi were 26% shorter in GATA6 mutants compared with controls. (B) Control and GATA6-deficient ileum were stained for HNF4A and the number of HNF4A+ epithelial cells per section was determined. We found reduced HNF4A+ cells in GATA6 mutant ileum compared with controls (G6 CTL, Gata6loxP/+, 899 ± 43 HNF4A+ cells/section; G6 cKO, Gata6loxP/-Villin-Cre, 543 ± 85 HNF4A+ cells/section; n = 3 embryos/genotype, 3 sections/embryo). (C) Alkaline phosphatase (AP) staining was reduced or absent in GATA6 mutant ileum compared with control ileum. (D) qRT-PCR was used to determine the abundance of Apoa4 and Fabp2, small intestinal enterocyte markers, and Car1, a marker of colonocytes, in the ileum of control and GATA6 mutants. Both Apoa4 and Fabp2 levels were severely decreased in GATA6 mutant ileum compared with control ileum. Car1 transcript was induced in GATA6 mutant ileum compared with control ileum. Epithelial cells from three control (Gata6loxP/+) and four mutant ileums (Gata6loxP/-Villin-Cre) were assayed at least three times. Gapdh was used for normalization. All P-values were determined by two-sample Student’s t test: *P ≤ 0.05, **P ≤ 0.01. All error bars show SEM. Scale bars in all micrographs are 50 μm. All experiments performed using E18.5 embryos.