Fig. 7

Ubiquitous deletion of Dll4 has a greater inhibitory effect in the Apc ^Min/+ tumorigenesis and neoplastic transformation than the endothelial-specific deletion. a, b, f, g, k, l Immunofluorescence stainings of small (a, f, k) and large (b, g, l) intestinal tumor cryosections (10 μm) from Apc ^Min/+ endoDll4 ^-/- and ubiqDll4 ^-/- mice versus the controls (CT) at 18 weeks of age. Representative images of staining density for PCNA (in green) and PECAM-1 (in red) (a, b), for Lgr5 (in green) (f, g), and for Cyclin D1 (in green) (k, l). The nuclei were counterstained with DAPI (in blue). Scale bars = 100 μm. c, h, m Graphic bars represent the small (SI) and large intestinal (LI) tumor proliferation index (c), and the relative tumor Lgr5 (h) and Cyclin D1 (m) density (%) ± SEM in the animals described above. One experiment with n = 6 per group and 6 fields per animal. d, e Graphic bars represent the proportion (%) of hyperplasias and adenomas with low and high-grade dysplasia obtained in the histopathological analysis (H&E) of the macroscopic small (d) and large (e) intestinal lesions from Apc ^Min/+ endoDll4 ^-/-, ubiqDll4 ^-/- and control (CT) mice at 18 weeks of age. One experiment with n = 12 per group. i, j RT-PCR analysis of Lgr5, Bmi1, Cdkn1b, Cdkn1c, Myc, Ccnd2 relative expression in the Apc ^Min/+ endoDll4 ^-/- and ubiqDll4 ^-/- small (i) and large (j) intestinal tumor samples from mice at 18 weeks of age. One experiment with n = 3 per group. *P < 0.05; **P < 0.01