Abstract
Enteroendocrine (EE) cells are the most abundant hormone-producing cells in the human body and are vital for metabolism, as well as intestinal and pancreatic function. They have been implicated in the pathogenesis of multiple diseases including diabetes mellitus. Although recent studies have identified multiple signaling pathways (including Wnt, MAPK, BMP and Notch) that can induce low levels of EE cell differentiation, the production of functional human EE cells in vitro remains challenging, making their study and therapeutic utilization difficult. To improve this, we employed the human intestinal organoid culturing system, as it mimics intestinal epithelial homeostasis, allowing for differentiation of multiple epithelial cell types. Using a small scale, directed screen, we targeted multiple transcriptional regulators, using small molecules known to control pancreatic and intestinal development, and hormone production. We chose small molecules instead of gene editing tools to avoid the potential pitfall of off-target mutagenesis.
We found that inhibition of FoxO1 in our organoid culture led to an increase in EE cell differentiation as assessed by EE-specific gene expression, with a 5-10 fold upregulation in expression of ChgA, NeuroD1, and Neurog3 compared to whole mucosal biopsies (P<0.01 for all targets, n=3 per group). Flow cytometry data showed 6-8% of cells produced CHGA, compared to 0.2% in undifferentiated organoids (P<0.0001, n=3 per group), and the 1% typically seen in the duodenum. We also noted a corresponding increase in the production of EE hormones, including glucose-dependent insulinotropic peptide (GIP), serotonin and somatostatin, by qPCR and immunofluorescence. Analysis of conditioned media using ELISA, compared to undifferentiated organoids, revealed increased serotonin (362.6±52.3 vs 167.5±5.1 ng/mL, P=.0037, n=3 per group) and GIP (5.76±1.31 pg/mL vs undetectable, n=3 per group). Independently, upregulation of GATA4-Nkx2.5 also induced EE cell differentiation and hormone production, although to a lesser extent than FoxO1 inhibition. The exception to this was GIP, which showed increased expression and production with GATA4-Nkx2.5 compared to FoxO1 inhibition (20.8±7.4 vs 5.8±1.3 pg/mL, n=3 per group), with a much larger increase when FoxO1 inhibition was followed by GATA4-Nkx2.5 activation (53.4±4.8 pg/mL, n=3). Of note, all experiments were performed in a minimum of three human lines.
Taken together, our data have identified multiple factors, including inhibition of FoxO1 and activation of GATA4-Nkx2.5, that can drive ex vivo human EE cell differentiation, with unique hormone production profiles, when targeted via small molecules. This is a critical first step towards understanding the role of enteroendocrine cells in disease and the development of EE cell-based therapies.
Robust differentiation of human enteroendocrine cells from intestinal stem cells
