Know your place

The image is reproduced with permission from van de Wetering, M. et al. Cell 111,241-250 (2002) © Elsevier Science.

Activating mutations in the Wnt pathway are the only known genetic alterations that cause intestinal epithelial cells to develop premalignant lesions (polyps). Back-to-back reports in Cell by Clevers and colleagues now provide insight into how dysregulation of the Wnt pathway might cause colorectal cancer (CRC).

Activation of the Wnt pathway induces the translocation of β-catenin to the nucleus to interact with TCF transcription factors. To understand the role of β-catenin–TCF complexes in CRC, the authors used dominant-negative TCF-4 (dnTCF-4) to inhibit TCF transactivation in CRC cell lines; this caused cell-cycle arrest in G1. Analysis of complementary DNA from the dnTCF-4-expressing CRC cells showed small subsets of genes that were up- and downregulated. Those that were downregulated were normally expressed in the proliferative compartment of colon crypts, whereas genes that were markedly upregulated localized to the top of the crypts (where differentiated cells are usually found), or were absent when polyps arose.

Of the genes upregulated by dnTCF-4, the cyclin-dependent kinase inhibitor p21 ^CIP1/WAF1 was the only cell-cycle regulator. When the authors induced p21^CIP1/WAF1 in CRC cells, G1 arrest and differentiation occurred. Conversely, c-MYC was the only dnTCF-4-downregulated gene that overrode the growth arrest induced by dnTCF-4 or p21^CIP1/WAF1, by binding to and repressing the promoter of p21^CIP1/WAF1. So levels of p21^CIP1/WAF1 are key in regulating differentiation or proliferation.

In the second paper, the role of Eph–ephrin signalling in mediating cell positioning in the small intestine was studied. EphB2 and EphB3 were downregulated in response to TCF-4 inhibition. In wild-type embryos, both genes are expressed in the proliferative intervillus pockets, whereas the ephrin-B1 ligand is expressed on adjacent differentiated villus cells. The authors propose that the interaction between cells at the boundary of these two compartments restricts cell intermingling in newborns, consistent with established roles for Eph–ephrin signalling. In adults, the pattern of EphB–ephrin-B1 expression is more complex. And in polyps, high levels of EphB2 and EphB3, but not ephrin-B ligands, were expressed. Ephrin-B1-expressing normal cells surrounded, but didn't mix with, EphB-expressing polyps. So, an attractive model put forward by the authors is that “β-catenin–TCF signalling couples cell positioning with cell proliferation, cell-cycle arrest and differentiation.”