Engineering Vero cells to secrete human insulin.

Cell therapy may have the potential for the treatment of Type I diabetes. To date, cells suitable for this purpose have not been developed. This study investigates the feasibility of modifying Vero, a cell line that may be considered safe to implant into humans, for this purpose. Stable Vero transfectants containing full-length human preproinsulin complementary deoxyribonucleic acid (cDNA) were generated using a liposomal transfection reagent. Reverse transcriptase-polymerase chain reaction, immunocytochemistry, Western blotting, and enzyme-linked immunosorbent assays were used to assess the resulting cells. Proinsulin was expressed but was not processed to insulin by these cells. Proinsulin cDNA was genetically modified, resulting in a form that is furin sensitive. The resulting stably transfected Vero clones constitutively release approximately 34%/h (32.68 +/- 2.21 to 35.62 +/- 3.14%) of the product formed, approximately 62% (59.99 +/- 6.45 to 64.64 +/- 4.57%) of which is mature insulin. These Vero transfectants did not exhibit glucose-stimulated insulin secretion. As GLUT2 and glucokinase (GCK) are not constitutively expressed by these cells, human GLUT2 cDNA and GCK cDNA were cotransfected with furin-sensitive preproinsulin cDNA into Vero cells. Insulin and GCK proteins were detected in the cytoplasmic region of the resulting cells, whereas GLUT2 was predominantly expressed in the nucleus. Coexpression of GLUT2 and GCK did not result in glucose-stimulated insulin secretion. The results from this study demonstrate the feasibility of engineering a relatively "safe" nonbeta cell line to produce human insulin. Coexpression of GLUT2 and GCK, at the levels achieved here, is not adequate enough to induce glucose-stimulated insulin secretion in such cells; the subcellular location of transfected components may be relevant.