Direct reprogramming of human fibroblasts into insulin-producing cells using transcription factors

Abstract

Direct lineage reprogramming of one somatic cell into another without transitioning through a progenitor stage has emerged as a strategy to generate clinically relevant cell types. One cell type of interest is the pancreatic insulin-producing β cell whose loss and/or dysfunction leads to diabetes. To date it has been possible to create β-like cells from related endodermal cell types by forcing the expression of developmental transcription factors, but not from more distant cell lineages like fibroblasts. In light of the therapeutic benefits of choosing an accessible cell type as the cell of origin, in this study we set out to analyze the feasibility of transforming human skin fibroblasts into β-like cells. We describe how the timed-introduction of five developmental transcription factors (Neurog3, Pdx1, MafA, Pax4, and Nkx2-2) promotes conversion of fibroblasts toward a β-cell fate. Reprogrammed cells exhibit β-cell features including β-cell gene expression and glucose-responsive intracellular calcium mobilization. Moreover, reprogrammed cells display glucose-induced insulin secretion in vitro and in vivo. This work provides proof-of-concept of the capacity to make insulin-producing cells from human fibroblasts via transcription factor-mediated direct reprogramming.

Forward programming of hiPSCs towards beta-like cells using Ngn3, Pdx1, and MafA

“…Gut stem cells are adult cells also of the endoderm lineage, and have been identi ed to also be hormone-secreting, including insulin in the fetus [52], while our broblast-derived hiPSCs could have been too early in development for the protocol to differentiate towards that lineage. Interestingly, Fontcuberta-PiSunyer, M., et al (2023) [53] directly reprogrammed human broblasts into glucose-responsive β-cells using a ten-day protocol similar to our approach. They transfected human broblasts with adenoviral constructs for the exogenous expression of NGN3, PDX1, and MAFA, followed by expression of PAX4 and NKX2.2 in the rst week of culture.…”

“…A key difference from these two protocols compared to ours is the timing of expression of the markers of interest. Rather than expression of all three markers continuously for the entire culture period, Huang, X., et al ( 2023) [50] only induced expression of NGN3 for the rst two days and then expressed PDX1 and MAFA, while Fontcuberta-PiSunyer, M., et al (2023) [53] introduced adenoviral constructs for NGN3, PDX1, and MAFA together and separate constructs for PAX4 and NKX2.2 were introduced later in the culture. This suggests that the continuous expression in our hiPSCs could have resulted in overload of cellular machinery for the continuous protein production.…”

“…Recent success of studies [50,53] using similar lentiviral constructs and culture conditions to reprogram adult (stem) cells into glucose-responsive insulin-producing β-cells provide hope that a few adaptations to our protocol can also achieve such results. Future studies looking to direct the differentiation of hiPSCs towards β-cells should use separate lentiviral constructs with different induction factors for each marker of interest to allow for the most e cient expression of the markers of interest and control over their temporal expression patterns.…”