Ex Vivo Gene Therapy Using Myoblasts and Regulatable Retroviral Vectors

Clare R. Ozawa, Matthew L. Springer, and Helen M. Blau

Stanford University School of Medicine Stanford, California, U.S.A.

I. INTRODUCTION: SKELETAL MUSCLE AS A TARGET TISSUE FOR GENE THERAPY

Human gene therapy—clinical treatments aimed at introducing or repairing genes to provide long-term correction for a defect caused by acquired or inherited disease—has been rapidly shuttled from a realm of theory and speculation into one of impending reality. The first approved protocol for human somatic gene therapy entered clinical trials in 1990 (1); the number of clinical trials worldwide has since burgeoned into the hundreds. Although gene therapy is still in its infancy and has yet to overcome a variety of pitfalls and problems, significant progress has been made. In this chapter, we discuss some advances in developing efficient means of gene delivery and in regulating gene expression to achieve pulsatile protein delivery when desired and avoid toxic levels. Although many of the features of muscle described below make it ideally suited for adenoviral, adeno-associated viral (AAV), lentivi-ral, and naked DNA delivery, the primary focus of this is ex vivo gene delivery via intramuscular cell implantation. Ex vivo myoblast-mediated gene delivery results in robust, long-term production in skeletal muscle of recombinant proteins ranging from muscle-specific proteins to systemic circulatory factors. This approach holds the advantage over direct viral or DNA delivery that the genetic change occurs outside the body, and the transgenic cells can be screened for potentially deleterious consequences like tumorigenicity before being implanted into a patient. Moreover, problematic immunological effects currently associated with most other methods are avoided with autologous myoblast-mediated delivery.

Of the many preclinical studies past and present, a number of approaches have employed skeletal muscle for delivery of genes in attempts to treat muscle disorders and other types of diseases. Skeletal muscle has been a target of special interest for gene therapy because of inherent properties that set it apart from other tissue types. Among its advantages include the fact that it is a very well-studied and understood tissue. This knowledge is invaluable for engineering strategies of gene delivery, and for assessing and controlling therapeutic protein expression. Skeletal muscle comprises a large percentage of the total body mass and thus is easily accessible to gene delivery. Mature, differentiated myofibers are relatively long lived, providing a lasting substrate for the stable expression of recombinant genes. Myoblasts that are purified and genetically engineered in tissue culture can be reinjected and will then enter host fibers where they are nurtured, properly innervated, and in close proximity to the blood. The multinucleate nature of muscle cells facilitates delivery of 2 or more different vectors that encode products that can meet inside the cell. This is particularly advantageous for vectors with limited capacity, such as AAV, or for introducing gene regulatory systems (see below). Moreover, although rich in contractile apparatus and not obviously suited for secretion, genetically altered skeletal muscle tissue has proven to be a surprisingly efficient factory for the production and delivery of recombinant proteins to the circulation, allowing for the treatment of a broad array of muscle and nonmuscle disorders where cell type-specific expression is not required.

In addition to the ex vivo gene transfer approach, an in vivo method of gene delivery to muscle is currently being tested and developed. In the in vivo approach, a vector harboring a copy of a corrected gene or encoding a product that can remedy a patient's defect is introduced directly into the muscle tissue of the patient. The vector employed may be either viral or nonviral in nature. Viral vectors that have been examined for their ability to transduce nondividing cells characteristic of muscle tissues include adenovirus, AAV, lentivirus, and herpes simplex type-1 virus (2-4). Although promising, in vivo approaches using viral vectors face several challenges, such as immune reactions elicited against viral elements (2,3). Although the most persistent expression has been seen with AAV (5-9), small capacity limits cDNA size, and difficulties in achieving adequate viral titers necessary for clinical trials have yet to be overcome. New generations of viral vectors characterized by more efficient production, lower immunoge-nicity and more stable expression are currently being developed that may prove to be powerful tools for muscle-mediated gene therapy.

Nonviral in vivo approaches for gene delivery to skeletal muscle center on intramuscular injection of plasmid DNA vectors. In the early 1990s, studies by Wolff and coworkers demonstrated that direct injection of naked plasmid DNA directly into the muscle tissues of mice led to transfection of skeletal myocytes and persistence of expression for at least 19 months in vivo (10,11). Naked plasmid DNA has also been shown to be taken up and expressed by cardiac muscle and skin (12-15). Plasmid vectors have a number of advantages for muscle-based gene therapy, and efforts to improve the efficiency of delivery appear warranted. These include simplicity of preparation and introduction into the host, and ability to be produced and stored in large quantities. In addition, the vectors are nonviral and are unlikely to be transmitted to other tissues. They also do not integrate into the genomes of host cells, precluding the risk of cancer by activation of a neighboring oncogene. Because achieving high levels of transgene expression using this approach has been problematic, direct plasmid DNA injection has been applied mostly to applications where only very low levels of transgene expression are required. One such application is in using intramuscular injection of plasmid DNA for vaccination purposes (16-19). Plasmids encoding antigenic proteins may be used to generate host antibodies specific to the antigen; because only small amounts of antigen are needed to elicit an immune response, plasmid vectors are well suited for this purpose. Recently, new progress has been made in developing methods for attaining stable, high-level gene expression using plasmid vectors. These include modifying DNA sequences within plasmids to enhance transcriptional efficiency of the vector (20), and combining plasmid DNA injection with the delivery of electric pulses to increase efficiency of myofiber transfection (21,22). Although promising, these methods require further study and development before they can be effectively applied to the therapeutic realm.

Cell-mediated or ex vivo gene delivery may provide a method of drug delivery for the treatment of a wide range of diseases (23). Skeletal muscle cells can be maintained as either proliferating or differentiating cells. The proliferative cells, known as myoblasts, are mononucleate muscle progenitors capable of fusing with each other to form new muscle fibers, or with preexisting myofibers (Fig. 1A). Myoblasts can be readily isolated from muscle and expanded in cell culture.

Figure 1 Myoblasts fuse to form multinucleate muscle cells. (A) Myoblasts are mononucleate muscle progenitor cells that can be isolated from muscle tissue and grown in cell culture (left). When provided with appropriate growth conditions, myoblasts fuse with each other in culture to form long, cylindrical, multinucleate myotubes (right). (B) Myoblasts can either fuse with each other to form new muscle cells, or they can fuse with preexisting muscle cells. Myoblasts of one genotype (shown with a dark gray nucleus) can fuse with multinucleate muscle cells of another genotype (shown with a light gray nucleus), thus delivering new genetic information to the preexisting muscle cell.

Figure 1 Myoblasts fuse to form multinucleate muscle cells. (A) Myoblasts are mononucleate muscle progenitor cells that can be isolated from muscle tissue and grown in cell culture (left). When provided with appropriate growth conditions, myoblasts fuse with each other in culture to form long, cylindrical, multinucleate myotubes (right). (B) Myoblasts can either fuse with each other to form new muscle cells, or they can fuse with preexisting muscle cells. Myoblasts of one genotype (shown with a dark gray nucleus) can fuse with multinucleate muscle cells of another genotype (shown with a light gray nucleus), thus delivering new genetic information to the preexisting muscle cell.

In vitro, they may be genetically engineered and extensively characterized, and then reimplanted back into muscle, where they stably fuse with myofibers (Fig. 1B). This unique property of skeletal muscle tissue has allowed for the development of myoblast-mediated gene transfer (24-26). Although the ex vivo approach of gene transfer to muscle is currently more cumbersome and costly than the in vivo approach, it provides certain advantages not offered by in vivo methods. Genetically altered myoblasts may be fully characterized in vitro before in vivo injection to ensure secretion of recombinant products of correct size and function at physiologically useful levels. In addition, isolated myoblasts are engineered outside the body with retroviruses, a process that generally ensures that only the proper cell type is transduced. In contrast, introduction of viral or nonviral vectors directly into muscle could theoretically lead to inadvertent low-level transduction of cells other than those being targeted, for example cells of the germline. Finally, recent studies suggest that a major limitation to ex vivo gene delivery—the requirement that syngeneic my-oblasts isolated from 1 patient be reinjected into that same patient to avoid rejection of cells by the immune system—may be overcome. Encapsulation of myoblasts overrides the requirement for a ''tailor-made'' therapy, allowing allogeneic cells that are invisible to the immune system to be used. My-oblasts appear advantageous over other cell types for this purpose because they do not overgrow and die within capsules, but instead differentiate and persist (27-29). Although this procedure prevents myoblasts from fusing into preexisting muscle, it theoretically allows universal donor cells to be derived from muscles of a single patient and implanted at ectopic sites in different patients for delivery of diverse products. All these advantages make ex vivo gene delivery via myoblasts a promising candidate for human gene therapy in the future. The remainder of this chapter focuses on ex vivo gene delivery to skeletal muscle.

Was this article helpful?

0 0

Post a comment