BACKGROUND & AIMS: Inappropriate activation of beta-catenin in adult tissues is associated with a wide variety of cancers, especially in the digestive tract. Classic transgenic and knockout murine models in which beta-catenin is activated in large fields of cells have provided experimental support in favor of a role for this molecule in tumorigenesis. However, these models do not reproduce the sporadic nature of the majority of human cancers, beginning with the activation of an oncogene at random in a single cell. METHODS: We used the "hit and run" strategy to generate a mouse model in which the expression of an activated form of beta-catenin occurs sporadically in vivo. RESULTS: Sporadic, multifocal lesions were observed in the stomach of 3% of mice aged 8 months and older. These lesions were associated with loss of Sonic hedgehog (Shh), and a causal relationship between beta-catenin activation and Shh inhibition was established in gastric cells in vitro. No lesion was detected in the intestine or in the liver. In addition, one third of female mutant mice developed benign perimammary papillomas. Mutant mice were also hypersensitive to chemically induced premalignant skin lesions. CONCLUSIONS: These results challenge the view that activation of beta-catenin induces malignant cancerogenesis, because they show in mice that sporadically activated beta-catenin is sufficient for tumor initiation, yet without further malignant progression, and that it sensitizes cells to environmental hits. This model represents a powerful tool to investigate the interplay between genetic and environmental factors in tumor progression.
BACKGROUND & AIMS: Inappropriate activation of beta-catenin in adult tissues is associated with a wide variety of cancers, especially in the digestive tract. Classic transgenic and knockout murine models in which beta-catenin is activated in large fields of cells have provided experimental support in favor of a role for this molecule in tumorigenesis. However, these models do not reproduce the sporadic nature of the majority of humancancers, beginning with the activation of an oncogene at random in a single cell. METHODS: We used the "hit and run" strategy to generate a mouse model in which the expression of an activated form of beta-catenin occurs sporadically in vivo. RESULTS: Sporadic, multifocal lesions were observed in the stomach of 3% of mice aged 8 months and older. These lesions were associated with loss of Sonic hedgehog (Shh), and a causal relationship between beta-catenin activation and Shh inhibition was established in gastric cells in vitro. No lesion was detected in the intestine or in the liver. In addition, one third of female mutant mice developed benign perimammary papillomas. Mutant mice were also hypersensitive to chemically induced premalignant skin lesions. CONCLUSIONS: These results challenge the view that activation of beta-catenin induces malignant cancerogenesis, because they show in mice that sporadically activated beta-catenin is sufficient for tumor initiation, yet without further malignant progression, and that it sensitizes cells to environmental hits. This model represents a powerful tool to investigate the interplay between genetic and environmental factors in tumor progression.