Analysis of JUN/AP1 function in Hedgehog/GLI-associated cancer

The Hedgehog (HH)/GLI signaling pathway is a developmental pathway involved in segment patterning of the Drosophila larval body or in A-P axis and pattern formation of the vertebrate embryo. In adult mammals, the Hedgehog pathway is essential for tissue homeostasis, while deregulation of this pathway is associated with the development and growth of cancers of the skin, brain, gastrointestinal tract, lung and prostate. A growing body of evidence suggests that control of oncogenic HH signaling involves interactions with other pathways frequently activated in human malignancies. In line with these findings, we have shown that Epidermal Growth Factor Receptor (EGFR) signaling synergizes with Hedgehog/GLI in oncogenic transformation of human keratinocytes in vitro. The integration of HH/GLI and EGFR signaling is thought to be mediated via an activation of the RAS/RAF/MEK/ERK cascade leading to the subsequent activation of the AP1 transcription factor JUN. Activated JUN can cooperate with GLI to activate the expression of selected EGFR-HH/GLI target genes thereby promoting transformation. Furthermore, it has been shown that the artificial GLI2 activator form (GLI2delN) can cooperate with active JUN in human keratinocytes, though these data were mainly based on overexpression and in vitro experiments. Given the crucial role of JUN in HH-EGFR-dependent cancer development and its putative cooperation with a dominant active GLI2delN form, the aim of my thesis was to address the in vivo role of cJun/Ap1 in Gli-driven basal cell carcinoma (BCC), a very common form of non-melanoma skin cancer. Using a combination of genetic tools including RNA interference and cre/loxP transgenic mouse technology for conditional inactivation of cJun/Ap1, I could show that inhibition of cJun/Ap1 in established Ptch deficient BCC cells decreases tumor growth in allograft experiments. By contrast, genetic deletion of cJun in GLI2delN expressing mice does not interfere with GLI2delN driven skin tumor development, but rather resulted in enhanced tumor formation and increased proliferation of basal epidermal cells. Further I extended the analysis of GLI/JUN cooperation to pancreatic cancer, which is one of the most malignant cancers with minimal survival rate. In this type of cancer cooperation between GLI and JUN/AP1 signaling is found to be important for the proliferation and self-renewal of tumor initiating pancreatic cancer stem cells in vitro and for tumor growth in vivo. To summarize my data I could demonstrate that cJun is dispensable for GLI2delN driven skin cancer formation in vivo. By contrast, cooperation between GLI and JUN is a critical determinant of the tumor-initiating properties of pancreatic cancer stem cells. (author)