TUFT cells have been described as strong modulators of inflammatory cells in several tissues including pancreas. TUFT cells, also known as DCLK1⁺ cells, are dependent of the transcriptional factor POU2F3. Several works report DCLK1⁺ cells in early stages of PDAC development suggesting an important role of TUFT cells in PDAC development. Therefore, we developed a mice model (PDX1-Cre;Kras^G12D;Ink4a^fl/fl), known as PKI model, deficient or not of POU2F3. In this animal model, deficiency of POU2F3 results in the absence of TUFT cells in PDAC as expected. Although, tumor development and growth are not significantly influenced, the development of liver metastasis was almost completely inhibited in POU2F3-deficient mice. Surprisingly, the absence of metastasis was associated with a higher expression of epithelial-to-mesenchymal transition markers, but to a lower inflammatory microenvironment suggesting that inflammation influences metastasis production more than epithelial-to-mesenchymal transition in this animal model. We can conclude that POU2F3 could be a new therapeutic target for control PDAC progression.

Synthetic ligands of peroxisome-proliferator-activated receptor beta/delta (PPARβ/δ) are being used as performance-enhancing drugs by athletes. Since we previously showed that PPARβ/δ activation affects T cell biology, we wanted to investigate whether a specific blood T cell signature could be employed as a method to detect the use of PPARβ/δ agonists. We analyzed in primary human T cells the in vitro effect of PPARβ/δ activation on fatty acid oxidation (FAO) and on their differentiation into regulatory T cells (Tregs). Furthermore, we conducted studies in mice assigned to groups according to an 8-week exercise training program and/or a 6-week treatment with 3 mg/kg/day of GW0742, a PPARβ/δ agonist, in order to (1) determine the immune impact of the treatment on secondary lymphoid organs and to (2) validate a blood signature. Our results show that PPARβ/δ activation increases FAO potential in human and mouse T cells and mouse secondary lymphoid organs. This was accompanied by increased Treg polarization of human primary T cells. Moreover, Treg prevalence in mouse lymph nodes was increased when PPARβ/δ activation was combined with exercise training. Lastly, PPARβ/δ activation increased FAO potential in mouse blood T cells. Unfortunately, this signature was masked by training in mice. In conclusion, beyond the fact that it is unlikely that this signature could be used as a doping-control strategy, our results suggest that the use of PPARβ/δ agonists could have potential detrimental immune effects that may not be detectable in blood samples.