Cell-based insulin therapies can potentially improve glycemic regulation in insulin dependent diabetes patients and thus help reduce secondary complications. The long-term goal of our work is to engineer autologous insulin-secreting intestinal endocrine cells as a non-beta cell approach to alleviate donor cell shortage and immune rejection issues associated with islet transplantation. These cells have been chosen for their endogenous similarity to beta cells, but generating cell constructs with sufficient insulin secretion for therapeutic effect has proven challenging. Previous work in our lab showed that a tissue engineered pancreatic substitute (TEPS) based on an engineered insulin-secreting L cell line, GLUTag-INS, was insufficient in affecting blood glucose levels in streptozotocin-induced diabetic mice, but promising since human insulin was detected in the blood. The objective of this project was therefore to fabricate an improved TEPS based on GLUTag-INS cells and evaluate its suitability as a standalone diabetes therapy. To achieve this objective, the following specific aims were (1) to investigate gene incorporation as a strategy to enhance recombinant insulin secretion from GLUTag-INS cells; (2) to develop and characterize a TEPS in vitro based on a microcapsule system containing improved GLUTag-INS cells with bioluminescence monitoring capability; and (3) to assess therapeutic efficacy of the graft in a diabetic, immune-competent mouse model and use bioluminescence monitoring to elucidate in vivo transplant behavior. This thesis therefore reports on the progression of studies from the genetic and molecular levels for improved insulin secretion per-cell, to the tissue level for enhanced secretion per-graft, and lastly to the preclinical level for therapeutic assessment in a diabetic mouse model.