Search

Suicide gene therapy based on ganciclovir (GCV) metabolism by transgene herpes simplex thymidine kinase (HSV-1 TK) has been used to selectively kill proliferating cells in clinical settings such as cancer, vascular restenosis, and immunological disorders. We investigated whether encapsulation of ganciclovir (GCV) into liposomes would improve its efficacy, especially against hepatic tumors. Large unilamellar liposomes containing GCV were prepared by reversed-phase evaporation. Pharmacokinetic studies in rats showed that, compared with free GCV, the intravenous injection of liposome-encapsulated GCV (lip-GCV) led to a faster decrease in GCV plasma concentrations, but higher liver-blood ratios. After treatment of syngeneic HSV-1 TK+ liver metastases in rats, histologically active tumors were found in 95% of the transplanted lesions when physiological saline had been given and in 50% when free GCV had been given at 90.2 microM/kg twice daily. This dose is known to be insufficient for the eradication of HSV-1 TK+ tumors. In contrast, only 5% viable tumors were found in rats receiving lip-GCV at this same concentration. Average tumor volumes were 19 +/- 15, 7 +/- 9, and

Delivering retroviruses targeted to hepatocytes in vivo involves the injection of retroviruses directly into the blood stream of the portal vein. The aim of this work was to delineate the conditions for delivering retroviruses in vivo by perfusing in situ the bile duct of the regenerating rat liver, and to study the hepatocyte transgene expression. At 24 hr after partial hepatectomy, during the S phase of the cell cycle, regenerating livers were perfused for 2.8+/-0.5 hr through the bile duct with 36.2+/-6.8 ml (0.3+/-01 ml/min) of fresh culture supernatant containing amphotropic recombinant retroviruses encoding the beta-galactosidase gene. The virus total titer was 1.5 x 10(8) ffu (group I) or 6.5 x 10(8) ffu (groups II and III). The hepatic artery blood flow was either maintained (groups I and II) or interrupted (group III) during bile duct perfusion. Liver biopsies taken 7 days later showed that 31.4+/-24.2% (group I), 58.7+/-23.6% (group II), and 45.1+/-21.4% (group III) of hepatocytes expressed beta-galactosidase activity, predominantly in the periportal and mediolobular zones. This study demonstrates that hepatocytes of regenerating rat livers that have entered the S phase of the cell cycle as a result of partial hepatectomy can be transduced in vivo by retroviral vectors delivered in situ by bile duct perfusion. Furthermore, the number of transduced hepatocytes closely correlated with the viral total titer and was diminished by hepatic artery blood flow occlusion during perfusion.

Antitumor gene therapy using herpes simplex type 1 thymidine kinase (TKh) and ganciclovir (GCV) treatment has revealed an important intratumoral bystander effect. A whole tumor can be eliminated when only a fraction of its tumor cells express TKh. We now report that the bystander effect not only acts within a tumor, but also between distant tumors. One TKh+ tumor was generated simultaneously with one or multiple TKh- tumors in different rat liver lobes such that there was no contact between the resulting tumors. Both the TKh+ and the TKh- tumors regressed after GCV treatment and showed infiltration with macrophages and T lymphocytes. This distant bystander effect, which is likely immune mediated, should be of major importance for gene therapy of disseminated tumors.

The ex vivo approach to hepatic gene therapy involves several steps, which include the isolation and culture of hepatocytes, followed by their transduction with a retrovirus. Subsequently, autologous hepatocytes are transplanted. The number of hepatocytes that can be transduced by retroviruses bearing the therapeutic gene is one of the limiting steps that can impair the success of this strategy. We presently describe an experimental approach that leads to improved transduction efficiency in mouse and human hepatocytes in vitro. By using a recombinant retrovirus bearing the Escherichia coli beta-galactosidase gene, we show that addition of growth factors to the cells, namely human hepatocyte growth factor (HGF), allows marked increase in the transduction efficiency in mouse (up to 80%) and human (40%) hepatocytes. Familial hypercholesterolemia (FH) is due to mutation in the low-density lipoprotein (LDL) receptor gene and results in a deficiency in LDL receptors. Transduction of the human LDL receptor cDNA under the transcriptional control of the L-type pyruvate kinase promoter-activator into mouse hepatocytes led to an elevated tissue-specific expression of the human protein. These results suggest that the ex vivo approach remains a promising alternative for hepatic gene therapy.

Dogs were used as a large animal model to assess the feasibility and safety of a surgical method for gene transfer into hepatocytes in vivo. This method, which we previously described in rats, consists of a partial hepatectomy aimed at inducing liver regeneration, followed by the selective in situ perfusion of the remnant liver parenchyma with a retrovirus preparation. Isolation of the liver was obtained by clamping the afferent and efferent blood vessels, a procedure that prevented retroviral vector dissemination and genetic modification of nonhepatic organs. A helper-free retrovirus vector encoding beta-galactosidase targeted to the nucleus was perfused in the liver of 5 golden retriever dogs. Volumes up to 1,650 ml of fresh or concentrated vector stocks were perfused and the procedure was well tolerated. Gene transfer, observed in 3 of 5 treated dogs when documented on liver biopsy fragments obtained at day 4, involved 0.15-0.6% hepatocytes and persisted at equivalent levels at the time of sacrifice, 6 weeks later. No propagation of the vector to other tissues was detected. These observations suggest that the selective perfusion of the regenerating liver might be considered an alternative to liver transplantation for the treatment of certain severe genetic liver disorders, or for the delivery of a therapeutic protein into the serum.