Tumorigenesis: TCF4 feeds c-Jun

Mutations in the canonical Wnt pathway proteins APC or β-catenin, which often result in constitutive activation of the transcription factor TCF4, are a frequent occurrence in colorectal cancers. Nateri et al. now report (Nature 437, 281; 2005) that canonical Wnt signalling joins forces with a non-canonical Wnt pathway — acting through JNK — to exacerbate Wnt's contribution to gut tumorigenesis.

Apc^Min/+ mice display numerous large intestinal adenomas (a, c), whereas mice that have lost c-Jun expression (Apc^Min/+;c-jun^ΔG) in the gut (b, d) are characterized by the presence of cysts (dashed circle). (c, d) Intestinal sections stained for c-Jun.

The authors identified TCF4 to be a binding partner of phosphorylated c-Jun — a component of the AP-1 transcription factor — using a genetic screen in yeast. They confirmed the c-Jun–TCF4 interaction biochemically, and showed that c-Jun and TCF4 form a ternary complex with β-catenin. JNK stimulates activation of c-Jun, and phosphorylation of c-Jun results in the increased transcriptional activation of several target genes, including c-jun. Chromatin immunoprecipitation analysis of the native c-jun promoter revealed that binding of c-Jun and β-catenin–TCF4 to this promoter is dependent on JNK activity.

So is there crosstalk between JNK and the canonical Wnt pathway? c-Jun and TCF4 cooperate in the induction of c-jun activity and this requires both the TCF and AP-1 sites of the c-jun promoter acting in cis, and depends on amino-terminal c-Jun serine residue phosphorylation. Furthermore, when the authors knocked down β-catenin expression by siRNA, they observed a reduction in the transcriptional activation of c-jun by c-Jun–TCF4, suggesting that β-catenin was required for the activation of the c-jun promoter by these transcription factors.

To test the physiological relevance of the c-Jun–TCF4 interaction, Nateri et al. crossed mice heterozygous for the Apc^Min mutation, which causes intestinal tumours by activating TCF4/β-catenin, with mice expressing a mutant form of c-Jun that cannot bind TCF4. The lifespan of these compound mutant mice was extended compared with wild-type controls, and the average size of their adenomas was reduced. Furthermore, when Apc^Min heterozygous mice were bred with transgenic mice that had lost c-Jun expression specifically in the gut, tumours did not develop until nine months of age. Thus, it appears that β-catenin signalling was no longer able to induce tumour formation in the absence of c-Jun, and suggests that c-Jun repression protects Apc^Min heterozygous mice against the development of intestinal tumours.

Collectively, the findings by Nateri et al. point to a mechanism whereby the interaction between TCF4 and phospho-c-Jun controls the transcriptional activation of c-jun by recruiting β-catenin. JNK inhibitors have proven successful in in vivo models of neurodegeneration, and the c-Jun–TCF4 interaction should help to invigorate effective therapeutic targeting of c-Jun-related diseases, including colon cancers.