a, Schematic representation of the Il22ra1tm1a(EUCOMM)Wtsi allele in targeted ES cells. The position of PCR primers and Southern blot probes is indicated. b, Verification of construct integration into JM8.N4 ES cells by genomic PCR. DNA from untargeted ES cells (Bruce4) served as a control. c, Southern blot after BamHI digestion, confirming the correct integration of the construct, which introduced a novel BamHI recognition site into the genomic DNA. d, Germline transmission was assessed by genomic PCR in the F1 generation. e, LacZ reporter activity was assessed in F1 mice. f, Gene expression of the IL22RA1 binding partners in colonic epithelial cells of the indicated mouse strains was assessed by qRT–PCR (Il22+/+, n = 5; Il22−/−, n = 4; Il22ra1fl/fl;Vil1-cre, n = 4; Ahrfl/fl;Rorc(γt)-cre, n = 3; mean ± s.e.m.). g, Representative immunofluorescence of IL22RA1 (red) in colonic epithelial cells. Scale bars, 50 µm. h, i, Il22ra1fl/fl;Lgr5creERT2/+;R26R-Confetti or Il22ra1fl/+;Lgr5creERT2/+;R26R-Confetti mice were fed with a tamoxifen-containing diet for two weeks. h, Labelling efficiency of colon crypts at different time points after stopping the tamoxifen treatment (day 1: Il22ra1fl/fl, n = 8; Il22ra1fl/+, n = 10; day 7: Il22ra1fl/fl, n = 7; Il22ra1fl/+, n = 8; day 14: Il22ra1fl/fl, n = 7; Il22ra1fl/+, n = 7 visual fields; mean ± s.e.m.). i, Fraction of Confetti+ crypts (mean) with one, two or three colours. j, The indicated mouse strains were fed with tamoxifen-containing diets and subjected to AOM–DSS treatment as indicated. Body weight is shown as percentage of initial weight (Il22ra1fl/+, n = 6; Il22ra1fl/fl, n = 10; mean ± s.e.m.). Data in f–j are representative of two biologically independent experiments.