Chromosome Engineering of Human Colon-Derived Organoids to Develop a Model of Traditional Serrated Adenoma

Background & aims: Traditional serrated adenomas (TSAs) are rare colorectal polyps with unique histologic features. Fusions in R-spondin genes have been found in TSAs, but it is not clear whether these are sufficient for TSA development, due to the lack of a chromosome engineering platform for human tissues. We studied the effects of fusions in R-spondin genes and other genetic alterations found in TSA using CRISPR-Cas9-mediated chromosome and genetic modification of human colonic organoids.

Methods: We introduced chromosome rearrangements that involve R-spondin genes into human colonic organoids, with or without disruption of TP53, using CRISPR-Cas9 (chromosome-engineered organoids). We then knocked a mutation into BRAF encoding the V600E substitution and overexpressed the GREM1 transgene; the organoids were transplanted into colons of NOG mice and growth of xenograft tumors was measured. Colon tissues were collected and analyzed by immunohistochemistry or in situ hybridization. We also established 2 patient-derived TSA organoid lines and characterized their genetic features and phenotypes. We inserted a bicistronic cassette expressing a dimerizer-inducible suicide gene and fluorescent marker downstream of the LGR5 gene in the chromosome-engineered organoids; addition of the dimerizer eradicates LGR5⁺ cells. Some tumor-bearing mice were given intraperitoneal injections of the dimerizer to remove LGR5-expressing cells.

Results: Chromosome engineering of organoids required disruption of TP53 or culture in medium containing IGF1 and FGF2. In colons of mice, organoids that expressed BRAF^V600E and fusions in R-spondin genes formed flat serrated lesions. Patient-derived TSA organoids grew independent of exogenous R-spondin, and 1 line grew independent of Noggin. Organoids that overexpressed GREM1, in addition to BRAF^V600E and fusions in R-spondin genes, formed polypoid tumors in mice that had histologic features similar to TSAs. Xenograft tumors persisted after loss of LGR5-expressing cells.

Conclusions: We demonstrated efficient chromosomal engineering of human normal colon organoids. We introduced genetic and chromosome alterations into human colon organoids found in human TSAs; tumors grown from these organoids in mice had histopathology features of TSAs. This model might be used to study progression of human colorectal tumors with RSPO fusion gene and GREM1 overexpression.