Retroviral vectors (RV) were the first vectors used for gene therapy. They can efficiently integrate into the chromatin of target cells. However, they require the target cells to be mitotically active for an efficient transduction. Therefore, induction of liver division or liver regeneration through manipulations such as partial hepatectomy or hapatocyte growth factor treatment have been required for efficient hepatocyte transduction 91011. More recently, it has been shown that RV can transduce hepatocytes from newborn mice 12 and dogs 13 without an exogenous stimulation of cell division. However, as shown by the SCID trial experience, the risk of insertional mutagenesis is still a major consideration for RV.

Lentivirus vectors (LV) offer similar advantages to the RV, in that they mediate long-term integration of the therapeutic transgene, but unlike RV, they do not require cellular mitosis to gain access to the host genome for integration. They also are thought to share the potential for insertional mutagenesis with subsequent carcinogenesis, although this has not been observed yet in animal models. Following systemic LV delivery, the majority of transduced liver cells are of nonparenchymal origin and therefore, the efficiency of hepatocyte transduction is relatively low.

Adenovirus (Ad) vectors are well suited for liver-directed gene therapy because they can transduce hepatocytes with high efficiency. When tested in vivo, first generation of Ad (FGAd) vectors, which are replication-defective but can still express viral genes at low levels, cause acute and chronic toxicity. Helper-dependent adenoviral (HDAd) vectors, which are devoid of all viral genes, offer a better safety profile and can provide long-term transgene expression with negligible chronic toxicity 14. Several preclinical studies have shown that HDAd results in long-term phenotypic correction in several genetic diseases 14. However, similar to FGAd they can still cause an acute toxic reaction due to activation of the host innate immune system when they are administered at high dose systemically 15. A clinical trial for OTCD using an early generation Ad vector bearing the human OTC gene was interrupted when the second subject at the highest dose suffered fatal complications. The trial involved 18 subjects divided into 6 cohorts of 1/2 log dose escalations between cohorts until two subjects were enrolled at the highest dose 16. As predicted from preclinical models, clinical findings were mild and transient in the initial 17 patients 16. However, unlike the previous subjects, the last patient enrolled developed within 24 hours after vector infusion a lethal reaction characterized by acute respiratory distress syndrome, hepatitis, disseminated intravascular coagulopathy, hyperammonemia, and high levels of serum IL-6 6. Several mechanisms have been proposed to be responsible for the activation of this acute response 17. However, regardless of the multiple mechanisms involved, systemic administration of Ad results in an acute toxic reaction which is triggered by the Ad capsid proteins in a dose-dependent fashion 15. This acute toxicity is currently the main obstacle preventing clinical application of HDAd and strategies to overcome this problem are currently under investigation 181920.

AAV vectors are derived from a non-pathogenic human parvovirus that can infect non-dividing cells and remains latent for prolonged periods, predominantly in an episomal state. AAV vectors appear to persist in infected cells and do not trigger a robust innate response following in vivo administration. A wide repertoire of different AAV serotypes with different tissue tropisms is now available for several disease applications 21. AAV vectors have a limited packaging capacity which precludes applications in diseases requiring large therapeutic genes. However, novel AAV serotypes with larger cloning capacity are emerging and they may at least in part overcome this problem 22. In the clinical study for liver-directed gene therapy of hemophilia B, a recombinant AAV vector expressing human Factor IX (FIX) was infused through the hepatic artery in subjects with severe hemophilia B in an open label, dose-escalation study. Two subjects in the higher dose cohorts achieved measurable FIX levels at 2 weeks after vector infusion but, in contrast to the results generated in animal models, they exhibited a gradual decline in factor levels to < 1% by 10 weeks after vector infusion. This was accompanied in both subjects by an asymptomatic transaminase elevation beginning 4 weeks after vector infusion, with a gradual decline to baseline normal levels coinciding with the loss of FIX expression 23. This reaction is due to the rejection of transduced hepatocytes by AAV capsid-specific memory CD8(+) T cells reactivated by AAV 24 and intense investigations are currently ongoing to overcome this problem. Another problem of the AAV vectors is as yet theoretical risk of insertional mutagenesis in humans. Studies in mice suggest that AAV vectors are predominantly nonintegrating 25, and a wealth of experience in the field had failed to uncover any evidence of tumor formation as a result of AAV transduction, except for a mouse disease model of mucopolysaccharidosis type VII 26. A recent study has reported that, in the tumor tissue, the vector appeared to have integrated in a region rich in microRNA sequences on mouse chromosome 12 27. On the other hand, follow-up periods ranging up to 9 years in several hemophilic dogs have failed to reveal any evidence of tumor formation 28.

Nonviral vectors offer a number of advantages over viral-based strategies, including minimal toxicity from the vector, long-term transgene expression, lack of a humoral response against the vector, and the consequent ability to repeat dose 29. A major advance in the field has been the development of the hydrodynamic injection technique which involves, in mice, the rapid injection of a large volume of naked plasmid DNA (pDNA) and typically results in 10–15% of hepatocyte transfection 3031. Systemic hydrodynamic procedure as practiced in the rodents is neither safe nor practical in larger animals or humans. However, clinically relevant methods using balloon catheters for regional hydrodynamic delivery of pDNA have been developed 3233. These studies demonstrate the feasibility of intravascular delivery to the liver using minimally invasive approaches, and are a step in the direction of human clinical trials.

Crigler-Najjar syndrome is an autosomal recessive condition characterized by non-hemolytic unconjugated hyperbilirubinaemia due to mutations bilirubin-uridinediphosphoglucuronate glucuronosyltransferase (UGT1A1). Patients with Crigler-Najjar syndrome type I (MIM 218800) are refractory to phenobarbital treatment, have life-threatening elevations of bilirubin, and are generally managed with phototherapy throughout childhood and adolescence. Although effective, phototherapy is cumbersome, inconvenient, and its efficacy may diminish with age because of increased skin thickness and decreased surface/mass ratio. Moreover, despite this treatment, patients remain at risk of brain damage when intercurrent infections may increase production of bilirubin above that which can be controlled by the phototherapy 34. Therefore, patients with Crigler-Najjar type I are often advised to consider liver transplantation, most frequently in the range of 18 – 25 years of age. Crigler-Najjar syndrome has long been considered a paradigm for developing gene therapies for metabolic liver diseases for several reasons: (a) the underlying defect is well characterized at the biochemical and molecular level; (b) the fraction of corrected hepatocytes required for clinical benefit is small, as deduced from hepatocyte transplantation studies 35; (c) the UGT1A1 does not require strict gene regulation for normal activity; (d) an animal model, the Gunn rat, recapitulating the human disease is available; (e) the outcome of the experimental therapies can be easily determined by measuring bilirubin fractions in serum and bile; (f) the UGT1A1 can be produced from skeletal muscle other than liver, its natural production site, and still retain the ability to transform bilirubin into water-soluble derivatives 36. For these several reasons, Crigler-Najjar syndrome type I is very attractive as a gene therapy disease candidate and its correction has been the goal of several studies using different vector systems including RV, LV, Ad, AAV, and nonviral vectors. RV expressing UGT1A1 injected in newborns 37 or in conjunction with partial hepatectomy 38 have achieved long-term correction of the hyperbilirubinemia in the Gunn rats. As previously discussed, LV can also transduce nonproliferating cells and, in the Gunn rats, they resulted in stable reduction of bilirubin levels to near normal levels for over 1 year after treatment 39. Impressive lifelong correction of hyperbilirubinemia has been also reported in the Gunn rats following a single intravenous injection of HDAd vector encoding UGT1A1 with negligible chronic toxicity 40. Among different serotypes, AAV serotype 1 was found to be the most efficient in correcting the hyperbilirubinemia of the Gunn rats although large hepatic macroscopic lipid lesions of unclear etiology were found in AAV-treated animals 41. A reduction of hyperbilirubinemia has also been reported following hydrodynamic injection of pDNA 42. However, as discussed in previous sections, each of the vectors used in this disease model has some limitations which are currently preventing clinical applications.

Urea cycle disorders typically present in the first few days after birth with poor feeding, vomiting, lethargy, and coma due to hyperammonemia. Despite aggressive pharmacotherapy, patients are at high risk for repeated episodes of hyperammonemia and cumulative neurological morbidity and mortality 4344. Given these significant problems, gene-replacement therapy could represent a viable alternative to OLT for long-term correction. Several studies over the past decade have found the therapeutic effect of several different FGAd vectors to be transient in the OTCD mouse models and lasting no longer than 2 months 45. HDAd instead can mediate long-term correction of the OTCD animal model without chronic toxicity 4647. The novel AAV serotypes (AAV7, 8, 9), with higher efficiency of hepatocyte transduction, have also resulted in long-term phenotypic correction 48. The application of LV and nonviral vectors for OTCD has not been reported to date and these vectors are likely to be inefficient in these diseases due to the high percentage of hepatocyte correction required.