Familial adenomatous polyposis is caused by an inherited mutation of adenomatosis polyposis coli ( APC ) and predisposes individuals to the development of colorectal polyps and colon cancer. Apc +/− mutant mice, however, have more polyps in the small intestine than in the colon. Aoki et al. now show that if the homeobox transcription factor gene Cdx2 — expression of which is reduced in most human colon cancers — is also mutated in these mice, the polyposis phenotype is similar to that found in humans. The authors show that disruption of Cdx2 causes fast transition through the G1–S phase of the cell cycle, leading to chromosomal instability and increased polyposis.

First, the authors observed that Apc+/Δ716Cdx2+/− mice had more colon polyps than Apc+/Δ716 mice and that these polyps had reduced expression of Cdx2. The polyps in Apc+/Δ716Cdx2+/− mice also showed loss of heterozygosity (LOH) at the Apc locus. In most Apc mouse models, LOH is caused by chromosomal instability and Aoki et al. showed that Cdx2+/− and Apc+/Δ716Cdx2+/− mice had higher chromosome instability than Apc+/Δ716 mice.

Next, Aoki et al. decided to fill in the gaps — how does reduction of CDX2 cause chromosomal instability? They observed that a human colon cancer cell line (DLD-1) with reduced CDX2 underwent accelerated G1–S transition compared with DLD-1 cells with normal levels of CDX2. Because activation of the phosphoinositidyl 3-kinase (PI3K) pathway accelerates G1 progression, the authors examined proteins in this pathway and found that mammalian target of rapamycin (mTOR) — a downstream effector of PI3K — was activated. Overexpression of CDX2 in DLD-1 cells suppressed mTOR activity and reduced the speed of G1–S transition, confirming that mTOR activity is regulated by CDX2. In addition, expression of constitutively active AKT — an upstream effector of mTOR — led to high chromosome instability in DLD-1 cells and in the colon polyp cells of Apc+/Δ716Cdx2+/− mice. When low CDX2-expressing DLD-1 cells were treated with inhibitors of mTOR, chromosome instability was reduced in a dose-dependent manner, but only if the cells were exposed in the G1-early–S phase of the cell cycle.

Future studies should address what the direct target of CDX2 is that activates mTOR. Confirmation of this pathway of tumorigenesis in human polyposis would provide a rationale for testing mTOR inhibitors for prevention and treatment of cancers that involve chromosomal instability.