IntroductionHemophilia A and B are the most common bleeding disorders caused by deficiencies of clotting factors VIII and IX, respectively, both of which are X-linked with a recessive heredity.1 Replacement of the deficient factors with frequent intravenous injections of plasma concentrates or recombinant proteins is the standard treatment for these diseases.2 Great efforts have been made for nearly a decade toward developing experimental gene therapy for these diseases and aiming at the development of a medical intervention that is more effective and convenient than the currently available replacement therapies.3 Hemophilia is a suitable clinical model for the development of gene therapy products and has a number of advantages: 1) there is a simple and well defined cause-and-effect relationship between the protein deficiencies and bleeding symptoms; 2) tissue-specific expression and precise regulation of the transgenes are not necessary; 3) well characterized animal models are available for preclinical studies; 4) an unequivocal endpoint for product efficacy can be assessed in clinical trials; and 5) even 1% to 5% of the normal physiological levels of the proteins is therapeutic.For gene therapy of hemophilia, the most challenging hurdle, with respect to the long-term expression of the deficient proteins at adequate levels, is the development of a suitable gene delivery system. Technologies have been evolving from ex vivo to in vivo approaches, from initial use of retroviral vector to recent application of adenviral (Ad) or adeno-associated virus (AAV) vector, demonstrating progress from early results of transient low-level expression to more sustained high-level expression.3 For hemophilia A treatment, Ad vectors are particularly useful, since the liver naturally produces factor VIII, and following intravenous (i.v.) injection, Ad vectors concentrate in the liver. This makes the gene transduction efficiency to liver very high. Adenovirus vectors have been developed for gene therapy due to their high titer, broad infectivity, potential for large payload, and in vivo gene delivery capacity.4 Although the immunogenicity and cytotoxicity associated with the early-generation Ad vectors have been a concern with respect to their clinical application, newly developed vectors, in which the viral coding sequences have been deleted, have significantly reduced the side effects associated with the vectors. The “gutless” Ad vector, or so called helper-dependent, large-capacity, or mini- Ad vectors are the representative examples of these new-generation Ad vectors.5-15
The mini-Ad vector system described in this report was developed based on two major research findings. First, an Ad- SV40 hybrid virus discovered during attempts to grow human Ad in non-permissive monkey COS-7 cells.16 The hybrid virus had a genome structure in which only both ends of the Ad sequences were retained and almost all coding sequences of the Ad genome were replaced by symmetric, tandemly repeated SV40 genomes. The hybrid viruses replicated and were packaged in the presence of a wild-type Ad as a helper. This finding implied that total replacement of the Ad genome was possible to form a mini-Ad vector as long as proper helper function and selective pressure was provided. Secondly, it was discovered that Ad packaging can be attenuated by deleting portions of the packaging signal.17 This finding provided a means to put selective pressure on the helper Ad (referred to as ancillary Ad) by specifically limiting its packaging process and allowing a preferential packaging of the mini-Ad. The system, therefore, is designed to have three main components: the mini-Ad vector, the E1-deleted ancillary Ad, and a production cell line that provides AdE1 complementation.Based on the mini-Ad vector system, MiniAdFVIII was developed. The MiniAdFVIII vector carries a 27 kb expression cassette, in which the full-length human factor VIII cDNA is flanked by a human albumin promoter and cognate genomic sequences. Infection of MiniAdFVIII in vitro showed that the vector mediated expression of functional human factor VIII at levels of 100-200 ng/106 cells per 24 hours in HepG2 and 293 cells. With single-dose intravenous injection of 1011 viral particles in hemophilic mice, MiniAdFVIII produced a sustained high-level expression of human factor VIII (at 100-800 ng/ml for up to 369 days) that corrected the factor VIII-deficient phenotype. Safety studies of MiniAdFVIII showed that there were no significant toxicities in mice and dogs after a single intravenous dose of up to 3×1011 and 6×1012 viral particles, respectively. In this report, other studies for developing the MiniAdFVIII vector with a site-specific integration capability and the development of a human factor VIII-tolerized mouse model for preclinical studies of MiniAdFVIII are described.
Therapeutic factor VIII levels and negligible toxicity in mouse and dog models of hemophilia A following gene therapy with high-capacity adenoviral vectors