Diabetes therapy by lentiviral hepatic insulin gene expression without transformation of liver

To the editor:

We read with interest the paper by Ren et al. [1] published in Diabetologia. In this study the liver of diabetic rats was transduced by a lentiviral vector system with the furin-cleavable insulin analogue B10. An instantaneous long-term correction of blood glucose concentrations was achieved within 2–3 days after virus application. At the end of the experiments, 500 days after virus application, an induced transdifferentiation of the transduced hepatocytes into cells immunohistochemically positive for insulin, glucagon and somatostatin was documented. Sixty days after virus injection, the treated rats showed a physiological insulin response and blood glucose concentrations after IVGTTs. The lentivirally transduced livers were positive in RT-PCR analyses for several beta cell typical transcription factors, but not for insulin or insulin proconvertase 2 as a sign of a partial transdifferentiation of hepatocytes into beta cells. Interestingly, the transduction by an empty lentiviral vector led to an induction of the beta cell transcription factors pancreatic and duodenal homeobox 1 (Pdx1) and neurogenic differentiation 1 (Neurod1). The lentiviral vector system used encodes for additional HIV accessory genes that might contribute to the induction of transdifferentiation.

We performed similar experiments. In our own ongoing experiments in which streptozotocin (STZ)-diabetic rats were transduced by the latest lentiviral vector system (third generation, which provides the highest biosafety [2, 3]) via portal vein injection, no signs of transdifferentiation occurred (Fig. 1). In contrast to Ren et al. [1], in our experiments we used furin-cleavable insulin and not the furin-cleavable insulin analogue B10. In this insulin analogue the histidine in position 10 of the B chain of the insulin molecule is replaced by aspartic acid [4]. The mutation of this insulin analogue has been shown to induce mitogenic effects [5, 6, 7], and according to the authors’ [1] conclusion, could also be responsible for the observed liver transdifferentiation. Using the furin-cleavable insulin for the lentiviral transductions we detected a decrease in blood glucose concentration of STZ-diabetic rats from 22.3 ± 3.4 to 15.6 ± 2.1 mmol/l after 3 days and to 15.0 ± 2.8 mmol/l after 30 days. About 20% of the hepatocytes of the transduced liver was positive for insulin (Fig. 1).

Open image in new window

Thus we can conclude that it is possible to improve blood glucose homeostasis through lentiviral overexpression of the insulin gene in the liver without concomitant transdifferentiation of the liver cells, purely by establishing insulin production in the liver, thereby replacing the loss of insulin supply from the endocrine pancreas to the organism. Use of the native insulin for expression in the liver prevents potential undesirable mitogenic effects.