Wnt Signaling in Pancreas Development and Cancer

During early somite stages of development, the broad endoderm-derived foregut and hindgut territories become further subdivided along the antero-posterior axis into organs. One of these organs is the pancreas which serves two major physiological functions: the production of digestive enzymes (exocrine pancreas); and the regulation of blood sugar (endocrine cells of the islet of Langerhans). Understanding how the molecular hierarchies operate during pancreatic primordium induction and organogenesis is of medical interest because the pancreas is the target of two important diseases: diabetes mellitus and pancreatic cancer. Among all the signaling pathways studied, we focused on the complex Wnt signaling pathway which plays multiple roles in early endoderm specification, pancreas differentiation and pancreatic tumor formation. During the antero-posterior patterning of the endoderm, we found that in mice the Wnt β-catenin pathway is active in posterior endoderm from 7.5 dpc to 9.0 dpc. Moreover we demonstrated in chick that Wnts are sufficient to posteriorize endoderm. To assess whether Wnts act directly on endoderm, we developed a conditional genetic approach to inactivate and overactivate the canonical Wnt pathway in definitive endoderm by using the Foxa-3-Cre transgenic line. β-catenin loss- and gain-of-function alleles specifically in endoderm induce only low numbers of recombined cells. This is in part due to the low and mosaic expression of the recombinase by the Foxa-3-Cre during gastrulation but it also strongly suggests that the cells that have impaired β-catenin function are counter-selected. As a consequence, embryos observed at 9.5 dpc and 10.5 dpc did not present any anterior or posterior shift of the pancreatic marker Pdx1. We developed a new mouse line (Sox17-CreGFP) expressing Cre earlier and in a less mosaic fashion in endoderm to pursue the project. Many laboratories including ours, have investigated the role of the Wnt/β-catenin signaling pathway in pancreas development. However, the effect on differentiation is controversial. In contrast, the contribution of non-canonical Wnt signaling to pancreas development has not been studied and remains to be elucidated. We took advantage of a mutant affecting the balance between the β-catenin and the non-caninocal planar cell polarity (PCP) pathways to investigate the role of these pathways in pancreas development. Inversin is a protein that switches off canonical Wnt signaling by targeting Dishevelled for proteasomal degradation and promotes the PCP pathway. Inv/Inv mice exhibit left right asymmetry defects and polycystic kidney disease. In addition their pancreas is cystic. These phenotypes suggest a role of the PCP pathway in the control of duct width and are reminiscent of the phenotype of mutants inactivating cilia genes in the pancreas (Tg737orpk, Kif3, Pkhd1). However, unlike cilia mutants, Inv/Inv mice also showed endocrine defects: β cell numbers were significantly increased from 14.5 dpc. This was not due to a modification of the number of progenitors or to proliferation defects. Interestingly, these results suggest that this pathway needs to be downregulated in pre-β progenitors to allow β cell differentiation. Of interest, cystic kidney and associated diabetes have been observed in MODY5 patients and Bardet-Biedl syndrome suggesting that Inv is MODY gene candidates. Aberrant Wnt signaling has been observed in almost all pancreatic tumor types. We investigated whether activation of β-catenin in different mature cell types was sufficient to initiate tumors in the pancreas. Although it was previously shown that Wnt pathway activation during development was sufficient to initiates solid pseudopapillary tumors, we found that neither activation of β-catenin alone in mature acinar nor in mature ductal cells induced pancreatic tumor formation. However, the activation of the Wnt pathway in mature ductal cells triggered inflammation.