The reactive stroma in pancreatic diseases

Abstract

Stroma, or activated connective tissue, is a key component in both pancreatitis and pancreatic tumors. Acute pancreatitis (AP) is an inflammation of the pancreas which ranges from mild and local to systemic with severe complications and high mortality. In paper I, we sought out to identify new potential biomarkers of AP, by comparing gene expression in mice with caerulein induced AP and control mice injected with sodium chloride. Regulator of calcineurin 1 (Rcan1) arose as the most promising candidate for an early marker of AP and was found to be regulated by oxidative stress, as it was upregulated by caerulein and H2O2 and this upregulation was inhibited by the antioxidant N-acetylcystein. Rcan1 protein was also found upregulated in the blood of mice early on after AP induction, and in patients with AP, suggesting the potential use of RCAN1 as a marker of AP in e.g. post-ERCP-pancreatitis.

Pancreatic ductal adenocarcinoma (PDAC) is the most common type of pancreatic cancer and has a dismal prognosis, in part due to resistance to available treatments. A desmoplastic stroma has become one of the hallmarks of PDAC and is currently receiving more and more attention from researchers. Overexpression of High mobility group A2 (HMGA2) protein has been associated with many cancers as well as with epithelial mesenchymal transition (EMT) and cancer stem cell properties. In paper II HMGA2 was shown to be correlated with lower overall survival in a cohort of 253 PDAC patients, acting as an independent prognostic marker. Tumor cell HMGA2-positivity was also significantly correlated with the abundance of PDGFRB+ stroma. Co-injections of Panc1 cells and pancreatic stellate cells (PSCs) in vivo increased tumor growth and also HMGA2 and PGDFRB expression. In vitro, HMGA2 expression was increased in Panc1 cells grown in 3D spheroids, both by co-culture with PSCs or TGFB1 stimulation. In paper III, the 3D co-culture spheroid models of either Panc1 or HPAFII PDAC cells with human PSCs, were extensively characterized by immunohistochemistry (IHC) and real time PCR. CK19 staining was utilized to identify the tumor cells by IHC within co-cultures. A novel method of detecting cell type specific mRNA expression within the co-cultures without the need for physical separation of the cells was also developed, where cells of different species (human and mouse) were co-cultured and analyzed by real time PCR using species specific primers. Co-culturing of PDAC cells and PSCs was shown to activate the stellate cells and increase tumor cell proliferation. As presented in the additional preliminary data, PDGFB did not induce TGFB1 expression by the PSCs as a possible source for the HMGA2 upregulation in cancer cells. HMGA2 protein expression in the 3D co-culture spheroid model was inconsistent and displayed some HMGA2+ PSCs in addition to the positive tumor cells. HMGA2 was found to be differentially expressed also in 3 clinical PDAC samples and one duodenal cancer, where HMGA2-positivity was found in both stroma and tumor cells, close to each other as well as separated.

In conclusion, this thesis work identified RCAN1 as a novel marker for AP and HMGA2 as a prognostic marker of PDAC. A novel 3D co-culture spheroid model of PDAC as well as a novel method of identifying cell type specific gene expression in 3D cultures without the need for physical separation, were developed. In addition, a complex relationship between HMGA2 expression in tumor and stroma cells was revealed, although the mechanism is not yet clarified.