Search

Efficient production of insulin in response to changes in glucose levels has been a major issue for insulin gene therapy to treat diabetes. To express target genes in response to glucose specifically in hepatocytes, we generated a synthetic promoter library containing hepatocyte nuclear factor-1, CAAT/enhancer-binding protein (C/EBP) response element, and glucose-response element. Combinations of these three cis-elements in 3-, 6-, or 9-element configurations were screened for transcriptional activity and then glucose responsiveness in vitro. The most effective promoter (SP23137) was selected for further study. Intravenous administration of a recombinant adenovirus expressing furin-cleavable rat insulin under control of the SP23137 promoter into streptozotocin (STZ)-induced diabetic mice resulted in normoglycemia, which was maintained for >30 days. Glucose tolerance tests showed that treated mice produced insulin in response to glucose and cleared exogenous glucose from the blood in a manner similar to nondiabetic control mice, although the clearance was somewhat delayed. Insulin expression was seen specifically in the liver and not in other organs. These observations indicate the potential of this synthetic, artificial promoter to regulate glucose-responsive insulin production and remit hyperglycemia, thus providing a new method of liver-directed insulin gene therapy for type 1 diabetes.

Top of pageAbstract Type 1 diabetes is caused by insulin deficiency. Insulin gene therapy has been attempted to cure diabetes; however, there is no satisfactory system that results in the tight regulation of insulin production in response to physiological changes in glucose levels. To develop a liver-specific, glucose-responsive synthetic promoter that can tightly regulate insulin production in response to blood glucose levels, we employed 3 kinds of transcription factor-binding cis-elements: hepatocyte nuclear factor-1, CAAT/enhancer binding protein response element, and glucose-response element. We generated a library of synthetic promoters that contained 9 of these cis-elements in random combinations and screened each promoter for transcriptional activity. Among 821 promoters tested, we selected 3 with the highest transcriptional activity for further testing. To determine the glucose responsiveness of these promoters in vivo, we generated recombinant adenoviruses expressing furin-cleavable rat insulin under the control of selected synthetic promoters (rAd-SP-rINSfur) and the cytomegalovirus promoter (rAd-CMV-rINSfur) as a control for ubiquitous strong activity. rAd-SP-rINSfur (1010 viral particles) or rAd-CMV-rINSfur (109 viral particles) was administered into streptozotocin-induced diabetic NOD.scid mice (n=6/group) via the tail vein, and blood glucose levels were examined. We found that diabetic mice treated with rAd-SP-rINSfur became normoglycemic a few days after treatment, and normoglycemia was maintained for more than 1 month, whereas mice treated with rAd-CMV-rINSfur were dead within 10 days due to uncontrolled constitutive expression of the insulin gene. Glucose tolerance tests showed that rAd-SP-rINSfur-treated mice produced insulin in response to glucose and cleared exogenous glucose from the blood in a manner similar to non-diabetic control mice. Insulin expression in rAd-SP-rINSfur-treated mice was observed only in the liver, not in other tissues or organs, whereas insulin expression in rAd-CMV-rINSfur-treated mice was present in all tissues tested, indicating that the synthetic promoters were liver-specific. On the basis of these observations, we conclude that our newly constructed, liver-specific synthetic promoters can tightly regulate insulin production in response to glucose and may have therapeutic potential for the cure of diabetes in humans by liver-directed insulin gene therapy.