Differentiation of Human Pluripotent Stem Cells Into Pancreatic Beta-Cell Precursors in a 2D Culture System

Diabetes is a complex metabolic disorder affecting millions of people globally. Supplementation of insulin is considered the only treatment option for diabetes. More advanced cases are treated with beta cell replacement therapy, achieved through transplantation of either whole cadaveric pancreas or islets^1,2. Several issues surround transplantation therapy, such as limitation with the availability and quality of the tissue, invasiveness of transplantation procedures in addition to the continuous need for immunosuppressants. This necessitates the need for discovering novel and alternative options for beta cell replacement therapy^2,3. Human pluripotent stem cells (hPSCs) have recently emerged as a promising tool for understanding human pancreas biology and as a non-exhaustive and potentially a more personalized source for transplantation therapy^4,5,6,7. hPSCs, including human embryonic stem cells (hESCs) and human-induced pluripotent stem cells (hiPSCs), have a high self-renewal capacity and give rise to any tissue type of the human body. hESCs are derived from the embryo's inner cell mass, and hiPSCs are reprogrammed from any somatic cell^4,8.

Directed differentiation protocols are optimized to generate pancreatic beta cells from hPSCs that sequentially direct hPSCs through pancreatic developmental stages invitro. These protocols generate hPSC-derived islet organoids. While they have greatly improved at increasing the proportion of pancreatic beta cells therein, the efficiency of protocols is highly variable. It does not increase to more than ~40% of NKX6.1+/INSULIN+ or C-PEPTIDE + cells^{5,9,10,11,12,13}. However, the generated beta cells are not entirely identical to the adult human beta cells in terms of their transcriptional and metabolic profiles and their response to glucose^4,5,14. The hPSC-derived beta cells lack gene expression of key beta cell markers such as PCSK2, PAX6, UCN3, MAFA, G6PC2, and KCNK3 compared to adult humans islets⁵. Additionally, the hPSC-derived beta cells have diminished calcium signaling in response to glucose. They are contaminated with the co-generated polyhormonal cells that do not secrete appropriate amounts of insulin in response to increasing glucose levels⁵. On the other hand, hPSC-derived pancreatic progenitors, which are islet precursors, could be generated more efficiently in vitro compared to beta cells and, when transplanted in vivo, could mature into functional, insulin-secreting beta cells^15,16. Clinical trials are currently focused on demonstrating their safety and efficacy upon transplantation in T1D subjects.

Notably, expression of the transcription factors PDX1 (Pancreatic and Duodenal Homeobox 1) and NKX6.1 (NKX6 Homeobox 1) within the same pancreatic progenitor cell is crucial for commitment towards a beta cell lineage⁵. Pancreatic progenitors that fail to express NKX6.1 give rise to polyhormonal endocrine cells or non-functional beta cells^17,18. Therefore, a high co-expression of PDX1 and NKX6.1 in the pancreatic progenitor stage is essential for ultimately generating a large number of functional beta cells. Studies have demonstrated that an embryoid body or 3D culture enhances PDX1 and NKX6.1 in pancreatic progenitors where the differentiating cells are aggregated, varying between 40%-80% of the PDX1+/NKX6.1+ population^12,19. However, compared to suspension cultures, 2D differentiation cultures are more cost-effective, feasible, and convenient for application on multiple cell lines⁵. We recently showed that monolayer differentiation cultures yield more than up to 90% of PDX1+/NKX6.1+ co-expressing hPSC-derived pancreatic progenitors^20,21,22. The reported method conferred a high replicating capacity to the generated pancreatic progenitors and prevented alternate fate specifications such as hepatic lineage²¹. Therefore, herein, this protocol demonstrates a highly efficient method for the differentiation of hPSCs to pancreatic beta-cell precursors co-expressing PDX1 and NKX6.1. This method utilizes the technique of dissociating hPSC-derived endoderm and manipulating the cell density, followed by an extended FGF and Retinoid signaling as well as Hedgehog inhibition to promote PDX1 and NKX6.1 co-expression (Figure 1). This method can facilitate a scalable generation of hPSC-derived pancreatic beta-cell precursors for transplantation therapy and disease modeling.