Introduction of CRC driver mutations in murine intestinal organoids.

A. Experimental design for sequentially introducing CRC driver mutations into the genomes of primary murine intestinal organoids with Cre recombination and CRISPR-Cas9. Cre recombinase and specific gRNAs were introduced via AAV (adeno-associated virus). K, KrasG12D activation; A, Apc loss; T, Trp53 loss. B. RT-qPCR analysis of Trp53 and Apc mRNAs expression levels in SKC, KA, KT, KAT and KTA organoid lines. Data is shown as mean ± SEM. (n = 3 biologically independent experiments). C. RT-qPCR analysis of Cdkn1a and Axin2 mRNAs (downstream targets of Trp53 and Apc, respectively) in SKC, KA, KT, KAT and KTA organoid lines. Data is shown as mean ± SEM. (n = 3 biologically independent experiments). Statistical analysis of RT-qPCR was done by One-way ANOVA, adjusted P values: *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.

Mutation accumulation drives histopathological progression in engineered organoids.

A. Representative bright field images of different organoid lines showing the growth during five consecutive days. Scale bar, 1000 μm. B. Comparison of organoid size at day 5 after single cell seeding. The analysis was performed with N = 2 clonal organoid lines except for SKC and K organoids, and n >= 18 organoids per genotype were measured. C. Histological analysis of different organoid lines by H&E staining, epithelial thickening in KA organoids is indicated by the black bar (epi) and focal dysplasia in KT, nuclear enlargement and disorganized epithelial cells in KA, KT, KAT and KTA organoids is shown by black arrowheads. Scale bar, 200 μm (left panel, overview), 50 μm (middle panel, individual organoid) and 20 μm (right panel, magnification). D. Comparison of organoid epithelium thickness. The analysis was performed with N = 2 clonal organoid lines except for SKC and K, and n >= 6 organoids per genotype were measured. E. Comparison of organoid formation rate at day 5 after single cell seeding. The analysis was performed with N = 2 clonal organoid lines except for SKC and K, and n >= 2 independent formation assays were performed. F. Growth curve of different organoid lines. The analysis was performed with N = 2 clonal organoid lines except for SKC and n >= 3 independent replicates. G. Ki67 immunostaining and proliferative index of mutant organoids. Scale bar, 50 μm. The analysis was performed with N = 2 clonal organoid lines and n = 9 distinct views. Statistical analysis of B, D, E and G were done by One-way ANOVA, P values: ns, non-significant, P > 0.05; *, P < 0.05; **, P < 0.01; ****, P < 0.0001.

Mutation burden-dependent tumorigenesis of engineered organoids.

A. Tumor outgrowth of organoid lines with different mutations in immunodeficient mice, showing the combined data from two independent experiments (see Methods for criteria defining outgrowth). B. Representative images of subcutaneous tumors developed from injected organoids with different mutations. All tumors were collected upon termination of the first experiment, with varying endpoint times. Scale bar, 1 cm. C. Representative H&E staining images showing the tumor morphology of KA, KT, KAT, and KTA tumors, necrosis was indicated with black asterisks. Scale bar, 200 μm (left panel, overview), 50 μm (middel panel), and 20 μm (right panel) for magnification. D. Representative images of Ki67 immunostaining showing proliferating cells in KA, KT, KAT, and KTA tumors from the first experiment. Scale bar, 100 μm. E. Ki67 index in indicated tumors. The analysis was performed with N >= 2 clonal organoid lines and n = 9 distinct views, statistical analysis was done by One-way ANOVA, adjusted P values: ns, non-significant, P > 0.05; ****, P < 0.0001.

Transcriptional changes during stepwise tumorigenesis.

A. Principal component analysis (PCA) of transcriptomic profiles from different organoid lines, and organoids with different genotypes are labeled by colors. PC1 and PC2 explain 36% and 19% of the total variance, respectively. B. Bar plot showing the number of differentially expressed genes (DEGs) along the mutagenesis, all comparisons were performed relative to SKC organoids. The DEGs were defined as genes with Log2 Fold Change (LFC) >= 1, adjusted P values (padj) <= 0.05. Red bars represent up-regulated genes, and blue bars represent down-regulated genes. C. GSEA summary showing the key biological pathway dysregulations upon KrasG12D mutation and subsequent Apc or Trp53 loss. Red and blue dots represent activation and suppression of indicated pathways, respectively. The color is scaled as indicated by adjusted P values. D. Venn plots showing the overlap of DEGs from KAT vs. SKC and KTA vs. SKC comparisons. Up- and down-regulated genes are colored as red and blue. E. Raincloud plots showing the comparable magnitudes in expression changes (indicated by paired lines connecting two points in two groups) of the overlapped 813 up-regulated and 988 down-regulated genes, suggesting mutation order minimally impacts overall transcriptional output, significant analysis was done by Wilcoxon test, ns: P > 0.05. F. Summary dynamics of transcriptional changes in selected gene sets, as indicated. The bar corresponding to each stage is colored by the log transformed gene expression z-score of each gene set.

Comparative transcriptomic analysis revealing conserved and context-specific gene expression changes induced by the same genetic mutation.

A. MA-plot showing the DEGs between KAT and KTA organoids, with 179 up-regulated genes and 118 down-regulated genes, defined as genes with Log2 Fold Change (LFC) >= 1, adjusted P values (padj) <= 0.05. B. GSEA enrichment showing the transcriptomic differences between KAT and KTA organoids, with interferon responses being most significantly activated and lipid metabolism being suppressed in KAT organoids compared to KTA organoids. C. Correlation of gene expression changes showing the LFC of up- and down-regulated genes (marked as red and blue dots) in indicated transcriptional comparison upon Apc (up panel) and Trp53 (lower panel) loss at different genetic background. D. Venn plots showing the overlap of the consistent up- (left panel) and down- (right panel) regulated genes when acquiring Apc or Trp53 mutation at different genetic background. E. RT-qPCR analysis of immune-related genes in KAT and KTA tumors derived from immunodeficient mice, and data is shown as mean and SEM. (n = 10 KAT tumors and n = 9 KTA tumors). Significance analysis was done by Wilcoxon test, P values are indicated in non-significant groups; *, P < 0.05; **, P < 0.01.

Tumor formation of KAT and KTA organoids in immunocompetent mice and investigation of clinical relevance.

A. Tumor outgrowth of KAT and KTA organoids in immunocompetent mice, showing the combined data from two independent experiments. B. Growth kinetics of KAT tumors (of the same isogenic clone) derived from both immunodeficient (n = 3 tumors) and immunocompetent (n = 3 tumors) hosts overtime, data from the second experiment. C. Representative images of subcutaneous KAT tumors (of the same isogenic clone) derived from both immunodeficient (n = 3 tumors, the second experiment) and immunocompetent (n = 3 tumors, the second experiment) hosts at the experimental endpoint. Scale bar, 1 cm. D. Representative images of CD4 and CD8 (T cell) immunostaining with positive cells indicated by black arrows. Scale bar, 2 mm (up panel, overview) and 200 μm (down panel, magnification). E. Bar plot showing the NES distribution of IFN-alpha and IFN-gamma signatures in ATK CRCs, which were generated from transcriptome comparison of KAT and KTA organoids (see Table S2, named organoid hallmark geneset here), red dots represent IFN-high IP tumors (n = 2) and blue dots represent IFN-low IP tumors (n = 12). Statistical analysis was done by Wilcoxon test, and P values are indicated in the figure.