Hepatocyte growth/scatter factor (HGF/SF) is a pleiotropic cytokine originally identified as a potent mitogen for rat hepatocytes. Two HGF/SF knockout mouse models have been reported, both of which exhibit developmental abnormalities causing embryonic lethality. To circumvent this limitation, we created a mouse conditionally deficient in liver expression of HGF/SF to specifically investigate the role of this mitogen in the process of adult liver regeneration. Gene targeting technology was used to generate a mouse with loxP sites flanking exon 5 of the HGF/SF gene (ex5-flox). In the absence of cre recombinase activity, mice homozygous for ex5-flox were indistinguishable from wild-type littermates. To ablate HGF/SF gene expression in vitro, primary hepatocytes established from homozygous HGF(ex5-flox) mice were infected with a recombinant adenoviral vector coding for cre recombinase (AdCre1). PCR analyses of genomic DNA demonstrated greater than 90% ablation of the ex5-floxed gene sequence. In vivo, HGF(ex.5-flox) mice were administered AdCre1 vector and the ablation of the HGF gene confirmed by Southern blot analysis. To induce liver regeneration, mice were injected with the hepatotoxin carbon tetrachloride. The regenerative capacity of hepatocytes in mice administered cre recombinase was shown to be significantly reduced when compared with mice injected with an adenovirus expressing LacZ. A similar reduction in hepatocyte regeneration was observed in HGF(ex.5.flox) mice carrying the cre transgene under the control of the interferon-inducible (pI:pC) Mx1 promoter, as an alternative strategy to ablate the HGF/SF gene in liver. Our results confirm the mitogenic role of HGF/SF in liver regeneration.

There are a number of reasons for choosing ornithine transcarbamylase (OTC) deficiency as a candidate for gene therapy: the gene has been cloned; the disorder is relatively common; the current clinical outcome is poor; and there are authentic animal models. In considering the development of gene therapy for OTC deficiency, we focused on the use of in vivo gene therapy with an adenoviral vector. Using the partially OTC-deficient sparse fur mouse we found transduction and expression could be achieved using an intravenous infusion of a recombinant adenovirus containing the OTC cDNA. The results were transient as a result of immune activation in response to the vector and vector-transduced cells. By modifying the adenoviral construct, creating an E1 deletion–E2 temperature-sensitive mutation, we blunted the cytotoxic T lymphocyte immune response and achieved correction of biochemical abnormalities for 2–3 months. We also found that transduction and expression following gene transfer occurred sufficiently rapidly to protect against acute hyperammonaemia within 24 h. Subsequent preclinical studies in mice and non-human primates demonstrated that E1–E4-deleted vectors had a substantially improved safety profile and similar efficacy. With this evidence of efficacy and safety of adenoviral vectors, we are embarking on a phase I trial of intravascular gene transfer using an E1–E4-deleted vector in adults with partial OTC deficiency.

Making Space for Intestinal Gene Therapy(1) the problems with current adenoviral vectors, (2) the See article on page 1586.issue of stem cell delivery in the gut, and (3) the concept of attempting to tolerize vector recipients through estab-

Recombinant adenovirus accomplishes highly efficient gene transferin vivo.Adenoviral vectors would be more attractive vehicles for gene therapy if transgene expression was inducible and restricted to the target tissue. In these studies, we hypothesized that selective transgene expression of a recombinant adenovirus could be accomplished by using a tissue-specific promoter of transcription. A replication-defective adenoviral vector was engineered to express thelacZmarker gene under control of the murine pancreatic amylase promoter. Expression of this vector occurred exclusively in the pancreas in neonatal and adult mice, while a similar vector with a constitutive promoter accomplished transgene expression in several organs. Within the adenoviral construct, the amylase promoter retained its ability to be induced by dexamethasone and insulin. This model will serve as a paradigm for selective and inducible adenoviral transgene expression.

Gene therapy represents a promising new modality for the treatment of inherited and acquired liver diseases. Clinical trials of liver-directed gene therapy are underway for diseases such as familial hypercholesterolemia, ornithine transcarbamylase deficiency, and cancer. This review summarizes the approaches to gene therapy developed to date and discusses their advantages and limitations.

Investigators working in the area of gene therapy believe the potential for advances in all medical disciplines is enormous. It is humbling, however to appreciate how far we need to go, as the field is truly in its infancy. Gene transfer technologies currently under evaluation in clinical trials have major limitations. Vector systems used in the clinics by the year 2000 probably have yet to be discovered. An additional lesson learned is that efforts at gene therapy are hampered by a lack of knowledge of the basic biology of the target organ and pathogenesis of the underlying disease. Successful gene therapy programs will critically evaluate the field and through fundamental research move steadily forward toward the long-term goal of truly effective therapy for a wide spectrum of disease. In the near future, two liver diseases are the most likely to be treated with gene therapy. The evaluation of patients with familial hypercholesterolemia is ongoing, and once approved, more candidates will be enrolled for therapy. Progress also has been made in creating vectors for the treatment of ornithine transcarbamylase deficiency. It remains to be seen whether adenoviruses or retroviruses will be used first in attempts to control this disease. Although the inflammatory response noted with current recombinant viruses is a formidable problem, the efficiency of gene transfer into the liver with these vectors makes continued study worthwhile.

Cirrhotic livers are considered to regenerate less actively than normal livers after hepatic resection. Little is known about the mechanisms responsible for impaired capacity of regeneration in cirrhotic liver. In the present study, we investigated the effect of phorbol ester on hepatocyte proliferation in healthy and cirrhotic hepatocytes, using one of the phorbol esters, 12-O-tetradecanoyl-phorbol-13-acetate (TPA), which has a direct effect on activation of protein kinase C (PKC). Cirrhosis was established by the administration of carbon tetrachloride and phenobarbital to rats. Healthy and cirrhotic hepatocytes were isolated from Wistar male rats by a two-step collagenase perfusion technique. DNA synthesis was estimated by [3H]thymidine incorporation into DNA and by autoradiographic nuclear labeling index. [3H]Thymidine incorporation was measured 24 hr after hepatocytes were stimulated by appropriate reagents. TPA (50 nM) stimulated [3H]thymidine incorporation in healthy hepatocytes (control vs TPA, 991 +/- 247 vs 2569 +/- 766 mean +/- SEM cpm/microgram DNA; P < 0.05), whereas TPA (50 nM) failed to stimulate in cirrhotic hepatocytes (control vs TPA, 1144 +/- 184 vs 1304 +/- 187 cpm/microgram DNA; NS). Staurosporine, a specific PKC inhibitor, suppressed [3H]thymidine incorporation in TPA-stimulated healthy hepatocytes (806 +/- 263 cpm/microgram DNA; P < 0.05); however, it had no effect on cirrhotic hepatocytes (1295 +/- 180 cpm/microgram DNA; NS). An autoradiographic nuclear labeling index exhibited the same results with [3H]thymidine incorporation. We conclude that TPA stimulates hepatocyte proliferation in healthy rat hepatocytes but has no effect on cirrhotic hepatocytes.