Role for cell-to-cell communication in stem cell specification toward pancreatic progenitors: relevance to the design of novel therapies for diabetes.

Thesis (Ph.D.)--University of Washington, 2016-08

University of Washington Abstract Role for cell-to-cell communication in stem cell specification toward pancreatic progenitors: relevance to the design of novel therapies for diabetes. Wendy Yang Chair of the Supervisory Committee: Vincenzo Cirulli Metabolism, Endocrinology & Nutrition Pancreatic islets of Langerhans, responsible for the production of hormones such as insulin and glucagon, develop from pluripotent pancreatic progenitors following their specification toward an endocrine phenotype. Based on the established role of cell-to-cell communication as an important mechanism regulating developmental decisions during embryonic life, I investigated the expression pattern of Connexins (Cxs), the building blocks of Gap Junction channels, in the developing human pancreas, and in an in vitro model of pancreatic progenitor differentiation from human embryonic stem cells (hESC). I also investigated the role of β1 integrins and an associated downstream effector, integrin-linked kinase (ILK), on islet development in mice. In a first series of experiments, I investigated the expression pattern of Cxs in the developing human pancreas. Results from these studies revealed that while Cx32 is predominantly expressed in the acinar tissue, Cx36 is primarily expressed in developing islet β-cells. Cx43 exhibited the most interesting expression pattern, being primarily detected in putative islet cell progenitors that delaminate from the pancreatic ductal epithelium and aggregate with developing islet cell clusters. Building on these exciting findings, and based on the growing interest in defining mechanisms that drive islet cell progenitor development from populations of progenitors and stem cells, I decided to focus my subsequent studies primarily on the role of Cx43 during stem cell differentiation toward the islet cell lineage. Consequently, I focused on the analysis of Cx43 expression and function in stem cells during their differentiation along the pancreatic cell lineage, with a specific emphasis on its functional requirement for the induction of early pancreatic islet progenitors, as defined by the acquisition of the transcription factor Pdx1. The results from these experiments showed that Cx43 is expressed at high levels in Definitive Endoderm (DE) cells, as defined by expression of E-cadherin, FoxA2, and Sox17. Building on these results, I adopted a gain-of-function approach using AAP10, a Cx43 specific peptide that promotes Cx43 phosphorylation and causes constitutive activation (i.e. opening) of Cx43-Gap Junction channels. Results from these experiments show that AAP10 treatment positively affects the propensity of hESCs to adopt a DE phenotype. Gene expression analysis revealed significantly higher levels of FoxA2 and Sox17 transcripts in DE populations treated with AAP10, compared to untreated controls. This effect was correlated with larger Cx43 Gap Junction plaques. Finally, to investigate the developmental consequences of increasing Cx43 Gap Junction plaques by AAP10 treatment, I monitored the expression of Pdx1 and Nkx6.1, two transcription factors required for islet cell development and differentiation, at the end of the differentiation protocol. My results indicated that AAP10 treatment during the early stages of hESC differentiation lead to a robust increase in the number of Pdx1+ pancreatic progenitors, when compared to untreated cells. Collectively, these results demonstrate that the purposeful activation of Cx43 Gap Junction channels at the early stages of hESC commitment toward DE cells leads to higher yields of pancreatic progenitors during the directed differentiation of hESCs toward pancreatic cell lineages. My next set of experiments focused on the knockdown of Cx43 in the derivation of pancreatic progenitors from hESCs. Since gain-of-function studies showed a role of Cx43 in increasing the derivation of DE cell types, I hypothesized that knockdown of Cx43 would have the opposite effect. A targeted siRNA approach did show knockdown of Cx43, however, I observed multiple off target effects in mock and scramble transfections with this approach. Therefore, I moved to an shRNA approach specifically targeting Cx43. In these studies, I observed that targeted knockdown of Cx43 led to inhibition of Pdx1, Nkx6.1 and Insulin gene expression. Hence, these studies validated the requirement for Cx43 expression at the DE and PGT stages for hESC to effectively progress toward pancreatic endocrine progenitors. In parallel studies, based on the notion that Cx43 can also affect integrin-mediated cell adhesion and migration1, and building on our laboratory’s established experience on the role of integrin receptors as regulators of pancreatic progenitors cell adhesion, growth and differentiation2, I studied the role of β1 integrins in islet cell development and function. For these investigations, I focused on a loss-of-function approach using a Cre-lox strategy that made use of an insulin promoter-driven DNA Cre recombinase, cross-bred to β1 integrin-floxed mice3. The results of these studies showed that ablation of the β1 integrin gene in developing β-cells resulted in a dramatic reduction of the pancreatic β-cell mass. Based on the observation that ablation of β1 integrin caused a reduced expression of genes regulating cell cycle progression, we interpreted these results as being caused by a defective β1 integrin signaling which in turn negatively impacted β-cell expansion. Surprisingly, these mice were not diabetic, and their islet clusters showed normal architectural organization, and normal expression of mature β-cell markers. Collectively, these results demonstrated that β1 integrin receptors function as crucial positive regulators of β-cell expansion. Building on these results, I investigated the role of downstream effectors of β1 integrin signaling, and focused my efforts on the function of integrin linked kinase (ILK) on pancreatic development. The first set of experiments determined the localization of ILK within pancreatic cells. Immunostaining showed co-expression of ILK and β1 integrins in the pancreatic epithelium. Next, I examined the impact of ILK inhibition by Cpd22 on pancreatic ductal cell proliferation using the G3LC and SU86 cell lines. ILK inhibition by Cpd22 resulted in a dramatic decrease in BrdU incorporation. Interestingly, inhibition of ILK by Cpd22 positively impacted the gene expression levels of insulin, Pdx1, and Cx36. Next, I conducted in vivo experiments focusing on the generation of knockout mice in which I targeted the ablation of ILK to early pancreatic progenitors using a Pdx1-Cre transgenic mice crossed with ILK-floxed mice. Preliminary results from these studies revealed that deletion of ILK in Pdx1+ pancreatic progenitors phenocopied our earlier animal model of β1 integrin deletion in β-cells3. I noticed that Pdx1-Cre/ILK-/- mice had a significantly reduced islet cell mass. Interestingly, Pdx1-Cre/ILK-/- mice exhibited a mild glucose intolerance in spite of a dramatic reduction in β-cell mass. This is at variance with the β1 integrin mutant mice which did not show a diabetic phenotype. These results open up an avenue to investigate if the Pdx1-Cre/ILK-/- phenotype is to be attributed to the reduced islet cell mass, or to a direct effect of loss on ILK function on insulin secretion. Together, these studies illuminate an important role for mechanisms of cell-to-cell communication in the development and function of pancreatic islet cells. Furthermore, the new approaches tested in our experiments of stem cell differentiation toward islet cells may contribute a significant improvement of current protocols for the derivation of islet β-cells to be used as a possible cell replacement therapy for type 1 diabetes. Bibliography 1. Kameritsch, P., Pogoda, K. & Pohl, U. Channel-independent influence of connexin 43 on cell migration. Biochim. Biophys. Acta - Biomembr. 1818, 1993–2001 (2012). 2. Cirulli, V. et al. Expression and function of alpha(V)beta3 and alpha(V)beta5 integrins in the developing pancreas: Roles in the adhesion and migration of putative endocrine progenitor cells. J. Cell Biol. 150, 1445–1459 (2000). 3. Diaferia, G. R. et al. Β1 Integrin Is a Crucial Regulator of Pancreatic Β-Cell Expansion. Development 140, 3360–72 (2013).