Objectives: The plasticity of pancreatic tissue is demonstrated in many pancreatic diseases. It has previously been shown that pancreatic islet-to-duct transformation and acinoductal metaplasia have been associated with both pancreatic regeneration and adenocarcinoma in various in vivo and in vitro settings. Understanding this inherent morphogenetic plasticity of the adult pancreas could lead to new therapeutic approaches to pancreatic disease.
Methods: Cadaveric human pancreases (n = 7) were digested, and purified acinar tissue, which was approximately 85% immunoreactive for amylase and approximately 15% immunoreactive for CK-19, was embedded in a type 1 collagen matrix and cultured in a differentiation medium (DM) consisting of Dulbecco modified Eagle/F12 medium supplemented with cholera toxin (100 ng/mL), epidermal growth factor (10 ng/mL), and insulin (24 mU/mL) for 8 days. After this initial period, the resulting tissues were cultured in DM without cholera toxin, supplemented with gastrin (50 nmol/L) and hepatocyte growth factor (HGF; 10 ng/mL), with islet neogenesis-associated protein (INGAP; 167 nmol/L) or with gastrin + HGF + INGAP for 6 days. Tissue samples were then analyzed for amylase, cytokeratin 19, pancreas duodenum homeobox 1, and endocrine hormone immunoreactivity as well as dithizione positivity.
Results: After 8 days of culture, approximately 90% of acini transformed into ductlike structures. This acinoductal transformation was characterized by a complete absence of amylase staining, with virtually all cells CK-19 immunoreactive. Addition of INGAP led to an approximately 18-fold increase in pancreas duodenum homeobox 1 immunoreactivity, although without an observed increase in insulin production as measured by dithizone positivity. However, when acinar-derived ductlike structures were cultured with gastrin + HGF + INGAP, the total incidence of dithizone-positive structures increased approximately 6-fold (10.9 +/- 2.9% vs 1.7 +/- 0.4%, P = 0.037). Treatment with gastrin + HGF alone led to no significant change in any of the measured parameters.
Conclusions: We have developed a novel in vitro model of adult human acinoductal metaplasia that will aid not only in developing new methods of expanding beta-cell mass but also provide insights into pancreatic carcinogenesis.