Can Liver Cells Make Insulin?

Using the PDX-1 gene to covert liver cells into beta cells. “Adult liver cells can be redirected to produce insulin in response to glucose levels, and this approach may overcome the shortage in tissue availability from cadaver donors and the need for lifelong immune suppression," stated Dr. Sarah Ferber from The Endocrine Institute, Tel-Hashomer, Israel.

Dr. Ferber and colleagues investigated the possibility of using the pancreatic and duodenal homeobox gene 1 (PDX-1) to convert adult human liver cells into insulin-producing tissue.

As many as half the adult and fetal liver cells infected with a recombinant adenovirus expressing PDX-1 activated the insulin promoter, the authors report, thereby enabling them to store and secrete insulin.

Supplementing PDX-1 treatment with the soluble factors nicotinamide and EGF increased insulin gene expression by seven orders of magnitude, the report indicates, significantly increasing insulin content and secretion.

PDX-1 further induced the expression of numerous pancreatic transcription factors, the researchers note.

PDX-1-transdifferentiated adult liver cells not only produced, stored and secreted insulin in a glucose-regulated manner, but also ameliorated diabetes when implanted in diabetic mice.

"Our approach represents a novel approach in tissue engineering," Dr. Ferber said. "It allows inducing a developmental redirection of adult fully differentiated tissues, allowing these tissues to acquire a new identity and function, without altering the original genetic information."

"In our application of converting liver to endocrine pancreas, it allows the diabetic patient to be also the donor of his own therapeutic tissue," Dr. Ferber explained. "A similar approach could be implemented in treating additional degenerative or metabolic diseases that involve other tissues."

Dr. Ferber said that additional studies will examine whether transdifferentiated cells resist autoimmune attack, as in type 1 diabetes, and will simulate autologous implantation in larger animals. She anticipates initial experiments in humans in several years.

PNAS Early Edition 2005;doi:10.1073/pnas.0405277102