3D Bioprinting of Primary and iPSC-derived Islets with retained phenotype and function

Introduction: Currently, the US alone has approximately 107,000 individuals waiting for a transplant1; while the total need for organs continues to rise each year, the availability remains relatively static. A potential solution to reduce the disparity between organ supply and demand is via organ fabrication from regenerative cell sources like induced pluripotent stem cells (iPSCs). Fabrication can be enhanced by implementing advanced manufacturing strategies, like 3D bioprinting. Successful 3D printing of regenerative organs requires an appropriate bioink which can (i) support organ survival and function (ii) protect organs from the stresses of printing while (iii) ensuring printing resolution. In this work we developed such a bioink which could support the high-resolution printing of primary human islets as well as iPSC-islets while retaining viability, phenotype, and function in extended culture. Our developed procedure could also retain printing resolution while at the same time protecting the cells from either physical or metabolic stresses.

Materials and

Methods: We tested multiple bioinks in this study, as detailed in Figure 1A. The most promising one was bioink alginate (alg)/methylcellulose (MC) which was tested at 2-4% and 5-7% weight by volume respectively. The ink is mixed over the course of 7 hours at a low heat and speed to prevent any microbubble formation. Prior to printing, cells are either included as single cell undifferentiated iPSCs, undifferentiated iPSC aggregates formed on Aggrewells, primary islets, or iPSC-derived islets. 0.1 mL of the cell suspension is slowly mixed with 0.5 mL of the bioink, in order to ensure a uniform suspension. The bio-ink is printed at varying pressures (20-35 kPa) and speeds (10-20 mm/s) using a Cellink BioX printer. Post printing, the structures are crosslinked for 5 min in 50mM CaCl2, washed with PBS, and then incubated in the desired media. Immunofluorescent staining is performed following standard protocol. Glucose Stimulated Insulin Secretion is performed with the islets/ iPSC-islets inside the printed construct by exposing the constructs to subsequent concentrations of low and high glucose and measuring the resulting insulin secretion by Insulin ELISA (Alpco). Two sequential low and high glucose exposures are conducted to test for diffusion limitations in the printed constructs. Single cell RNAseq library preparation is conducted on iPSC-derived printed and control islets, following the 10X Genomics protocol for a 3PrimeV3.1 experiment, with analysis done in Partek Flow and Ingenuity Pathway Analysis.

Results, Conclusions, and Discussions: Having tested multiple bioink compositions and printing parameters, we determined the optimal bioink for printing both primary and iPSC derived human pancreatic islets , to be 3/6% alg/MC bio-ink crosslinked with 50 mM CaCl2, which produced an average construct 1198 ± 273 µm in diameter at pressures 8-30 kPa and nozzle speeds 15-45 mm/s (Fig 1a). Isolated human islets when printed in this bioink retained high viability and function with Glucose Stimulated Insulin Secretion (GSIS), resulting in an average stimulation index (high insulin secretion/low insulin secretion) of 4.64. Printing of iPSCs as single cells also resulted in high viability with distinct cell aggregates (Fig 1B). However, excessive debris was observed in the printed constructs, likely resulting from non-aggregated single cells. Hence printing was shifted to the later phase of iPSC-islet differentiation, with iPSC-aggregates being differentiated to the premature islet stage, followed by printing and further maturation. Characterization of the iPSC-islets after 7 days of culture post-printing, showed high viability and excellent up-regulation of pancreatic specific gene and transcription factors, including hormone expression (Fig 1C) being comparable to that of primary islets. Importantly, the printed iPSC-islets showed excellent functionality of insulin secretion in response to glucose, comparable to that of primary islets (Fig 1D). Further scRNA-seq analysis confirmed that several key pathways related to islet functionality were upregulated in the printed iPSC-derived islets, however pathways associated with ER stress were comparable to control, indicating that the process of printing is not exposing the islets to deleterious stress condition.

Conclusion: In this work we have successfully designed a bioink and printing protocol to 3D print both undifferentiated iPSCs and iPSC-islets. The printed iPSCs retained high viability and function akin to that of primary human islets. To the best of our knowledge, this is the first demonstration of 3D printing pancreatic islets derived from a regenerative cell source with demonstrated functionality followed by thorough characterization.

Acknowledgements (Optional):