Fig. 1: High-throughput organoid differentiation screen reveals pro-Paneth function compounds.
a, Stem-enriched to Paneth-enriched (ENR+CD) organoid differentiation-modulating small-molecule screen (top row) assayed with multiplexed functional secretion, both basal (LYZ.NS) and 10 uM carbachol-stimulated (induced) lysozyme (LYZ.S) secretion, and cell number (ATP) on day 6 (middle row). Screened compounds (433) dosed at 4 concentrations (80 nM to 10 mM) were randomly distributed across 5 plates with interspersed quality control (QC) wells including both no cell and stimulated and non-stimulated (A,B,C,D) (bottom row), with the full screen repeated with organoids derived from 3 murine donors (m1–m3). b, Replicate strictly standardized mean difference (SSMD) for each assay in the primary screen, each point representing the SSMD from 3 replicates of 3 biological donors relative to the whole-plate control. Coloured points are hits above the false positive limit- and false negative limit-determined cutoff (dotted lines) in both LYZ.NS and LYZ.S assays. c, Mean fold change of assay effect for hits in LYZ.S and LYZ.NS (yellow) or all three assays (blue) in the primary screen; only points above 1.28 s.d. (σ) (dotted lines) of all treatment mean fold changes for LYZ.S and LYZ.NS are deemed significantly increased. d, Mean fold change for each assay in the secondary validation screen (N = 8 well replicates, relative to DMSO controls); orange, treatments advanced for profiling; red, most potent compound, KPT-330. e, Flow cytometry for the mature Paneth cell fraction of all live cells in 3D-cultured intestinal organoids treated with 6 hit compounds during 6 d of culture in ENR+CD media. Paneth cells were identified as lysozyme-positive and CD24-mid cells. Means and individual values are shown (N = 4); dotted line represents the average Paneth cell fraction in control samples. One-way analysis of variance (ANOVA) post-hoc Dunnett’s multiple comparisons test: ****P < 0.0001, ***P = 0.001.
