Identifying novel molecular mechanisms involved in the generation of new insulin-producing beta cells from adult stem / progenitor cells

Diabetes mellitus is a group of metabolic disorders characterised by chronic hyperglycaemia following a reduction in beta cell mass, impaired beta cell function or insulin insensitivity. Diabetic patients are at significantly higher risk of morbidity and mortality and current require multiple daily administrations, offer relatively poor glycaemic control and have their own associated risks. Islet transplantations are therefore a preferential alternative. The prospect of regenerating beta cells is very attractive as it would provide superior temporal glycaemic regulation, is free from allogeneic rejection and given the increasing incidence of diabetes, could provide a cell therapy that meets the clinical demand.

Efforts to generate beta cells in vitro from pluripotent stem cells have so far proved limited as the derived cells often possess an immature phenotype and lack full functionality. While iPSCs and ESCs hold great potential in the generation of new beta cells for transplantation, the full understanding of the molecular mechanisms controlling differentiation still remain incompletely understood. In addition, alternative cell sources have been suggested to avoid the unknown complications of iPSC and ESC therapy.

Pancreatic ductal cells have recently emerged as a potential source of progenitor-like cells in vivo, based on their highly plastic nature, lineage tracing studies and in silico analysis indicating they possess a high degree of plasticity. Following pancreatomy murine ductal cells rapidly lose their differentiated phenotype and revert to a progenitor like state. These progenitor-like cells are capable of recapitulating pancreatic embryonic development giving rise to both exocrine and endocrine pancreatic lineages.

Attempts to expand and differentiate pancreatic progenitors in traditional 2D cultures have proved limited due to a lack of cell-cell interactions provided by the endogenous 3D microenvironment. Therefore, we have established a 3D culture system in the lab to expand adult pancreatic ductal cells as organoids, which maintain key functional and structural properties especially important for differentiation.Pancreata were dissected from reporter mice and cultured in a 3D Matrigel culture system with media supporting the growth of ductal organoids. These ductal organoids have the potential to give rise to all lineages of the exocrine pancreas and self-renew indefinitely. In addition, the 3D culture system better represents the in vivo environment with mechanical stresses and improved cell to cell contact. A few days following cell seeding organoid structures began to develop – recapitulating the embryonic development of the ductal compartment. Cell doubling times remained constant by passage with organoids split at a 1:6 ratio weekly. In order to better understand the cellular composition of our organoid cultures we performed immunofluorescent staining of our non-differentiated organoids. Unsurprisingly, given their ductal origin, our organoid cultures displayed homogenous expression of the ductal markers KRT19 and HNF1a as well as heterogenous expression of SOX9, a marker of embryonic proliferative cells. Interestingly heterogenous expression of PDX1, usually restricted to the beta cell population in the adult pancreas, was also observed.

To direct ductal organoids towards a beta cell fate we utilise an adenovirus (Ad3) encoding Pdx1, Ngn3 and MafA as a self-cleaving polycistronic cassette. As part of our characterisation and to ensure the organoids weren’t acquiring expression of Ad3 factors or beta cell markers we performed gene expression analysis over high passage number. Without Ad3 infection ductal organoids maintain their ductal phenotype and do not display expression of Ngn3, MafA or Insulin mRNA over 11 passages. In addition to these three key transcription factors for beta cell conversion, Ad3 carries a monomeric Cherry (mCherry) reporter enabling identification of successfully infected cells. Following successful Ad3 infection, Insulin protein is detected in cells co-expressing mCherry. Analysis by flow cytometry indicates a total infection efficiency of 20.69% whilst the differentiation efficiency is 6.19% (n=3). This indicates that not all cells within the organoid culture are capable of giving rise to beta-like cells.

Gene expression analysis by RT-qPCR, shows a downregulation in ductal markers (Krt19 and Sox9) following transduction of the Ad3 factors Pdx1, Ngn3 and MafA compared to non-infected controls. Whilst Insulin expression and Nkx6-1 (a marker of mature beta cells) were upregulated in infected cells, Glut2 (a component of the glucose sensing machinery of beta cells) was not significantly increased. This may be due to the fact that only a proportion of cells are successfully differentiated leading to only a marginal increase in transcript levels from the bulk RNA. Conversely, these cells may still be immature, lacking the machinery to detect changing glucose levels and secrete Insulin in response. Profiling of single cells by scRNA-seq will provide clarification of this question.

Ultimately our aim is to functionally test our derived beta-like cells both in vitro and in vivo. Despite seeing an increase in both Insulin transcript and protein we so far have no evidence that our derived beta-like cells are capable of secreting insulin upon glucose challenge. To test the in vivo functionality, cells will be transplanted under the kidney capsule of NOD-SCID diabetic mice and their ability to recover glucose levels will be evaluated over 25 weeks. The mice will be sacrificed, and the kidney capsule extracted and subjected to immunohistochemical staining to identify β-cell differentiation efficiency of the pancreatic progenitors.

A limitation of our system is the loss of the 3D structure following the differentiation of our ductal organoids to beta-like cells. Although cells do form clusters in our system, this doesn’t represent the in vivo niche. Islets are highly organised structures and beta cells exhibit high basal-apical polarity. These cell-cell interactions and polarity are crucial for correct insulin secretion. Therefore, aggregation of derived beta-like cells into pseudoislets could help to recapitulate the in vivo islet microenvironment and result in improved glucose-induced insulin secretion.

Publications:

Stem/progenitor cells in normal physiology and disease of the pancreas.  M.E. Alvarez Fallas, S. Pedraza-Arevalo, A.M. Cujba, T. Manea, C. Lambert, R. Morrugares, R. Sancho. Mol Cell Endocrinol. 2021;538:111459.

Ductal Ngn3-expressing progenitors contribute to adult β cell neogenesis in the pancreas.  C. Gribben, C. Lambert, H.A.  Messal, E.L Hubber, C. Rackham, I. Evans, H. Heimberg, P. Jones, R. Sancho, A. Behrens. Cell Stem Cell. 2021:S1934-5909(21)00340-4.

Improving the laboratory diagnosis of pyruvate kinase deficiency.  C. Laas, C. Lambert, T. Senior McKenzie, E. Sheldon, P. Davidson, D. Rees, B. Clark. Br J Haematol. 2021;193(5):994-1000.