MICROSCOPY RESEARCH AND TECHNIQUE 48:223–238 (2000) Developments in Gene Therapy for Muscular Dystrophy DENNIS HARTIGAN-O’CONNOR1,3 AND JEFFREY S. CHAMBERLAIN1,2,3,* 1Program in Cellular and Molecular Biology, University of Michigan Medical School, Ann Arbor, Michigan 48109-0618 of Human Genetics, University of Michigan Medical School, Ann Arbor, Michigan 48109-0618 3Center for Gene Therapy, University of Michigan Medical School, Ann Arbor, Michigan 48109-0618 2Department KEY WORDS dystrophin; Duchenne; limb-girdle; vector; adenovirus; gutted; gutless; helperdependent; AAV; lentivirus ABSTRACT Gene therapy for muscular dystrophy (MD) presents significant challenges, including the large amount of muscle tissue in the body, the large size of many genes defective in different muscular dystrophies, and the possibility of a host immune response against the therapeutic gene. Overcoming these challenges requires the development and delivery of suitable gene transfer vectors. Encouraging progress has been made in modifying adenovirus (Ad) vectors to reduce immune response and increase capacity. Recently developed gutted Ad vectors can deliver full-length dystrophin cDNA expression vectors to muscle tissue. Using muscle-specific promoters to drive dystrophin expression, a strong immune response has not been observed in mdx mice. Adeno-associated virus (AAV) vectors can deliver small genes to muscle without provocation of a significant immune response, which should allow long-term expression of several MD genes. AAV vectors have also been used to deliver sarcoglycan genes to entire muscle groups. These advances and others reviewed here suggest that barriers to gene therapy for MD are surmountable. Microsc. Res. Tech. 48:223–238, 2000. r 2000 Wiley-Liss, Inc. INTRODUCTION The muscular dystrophies (MDs) are a common set of genetic disorders. Since none of the underlying biochemical defects were described until 10 years ago, treatment for these disorders has been mostly palliative. In the absence of a biological explanation for muscle weakness, little progress toward identifying rational treatments could be expected. This situation changed with the identification of dystrophin in 1987 (Koenig et al., 1987), which cleared the way for biochemical studies of the defect underlying DMD and offered hope that other dystrophies might be similarly understood. The genetics of many types of MD have now been delineated in considerable detail. In addition to mutations of dystrophin and the dystrophin-associated glycoproteins (Bonnemann et al., 1996; Ozawa et al., 1995), mutations in genes including calpain (Richard et al., 1995), caveolin-3 (Minetti et al., 1998), laminin-alpha2 (Helbling-Leclerc et al., 1995), fukutin (Kobayashi et al., 1998), and emerin (Bione et al., 1994) have been shown to cause forms of MD (Fig. 1). Each of these molecules must play a functionally important role in muscle and so offers an opportunity for understanding the biology of muscle and the pathogenesis of dystrophy. The potential of gene therapy for treatment of genetic disease was understood long before the development of positional techniques for identification of disease genes (Freese, 1972). When the dystrophin gene was identified, many scientists hoped that gene therapy for DMD would follow quickly. However, the difficulties of using gene transfer to treat DMD soon became clear. The major obstacles included the large size of the dystrophin gene (Table 1), the large mass of post-mitotic muscle cells in the body, and the tendency of the immune system to reject novel antigens. It is now clear r 2000 WILEY-LISS, INC. that efficient, long-term transfer of dystrophin required significant new vector technologies beyond those available in 1987. Today there is room for cautious optimism that DMD and other muscular dystrophies will be treated with gene therapy. Scientists have achieved a new understanding of the requirements for, and obstacles to, successful gene transfer. There is greater respect for both the capabilities and the limitations of available viral vectors. By choosing the best vector for use in a particular setting, significant success has already been achieved. MUSCULAR DYSTROPHY AS A CANDIDATE FOR GENE THERAPY The earliest clinical gene therapy protocols targeted cells of the hematopoietic system (Culver et al., 1991; Rosenberg et al., 1990). There were at least two advantages to this choice. First, gene transduction could be performed ex vivo, which allowed for in vitro selection of genetically altered cells. Second, there are diseases of the hematopoietic system in which corrected cells should have a selective growth advantage over uncorrected cells, allowing for successful therapy even when a minority of target cells are transduced. In principle, it can be argued that the muscular dystrophies should share these advantages. Dystrophic myoblasts can be isolated and manipulated ex vivo. Transplantation of corrected myoblasts into dystrophic muscle could restore normal muscle function. Since myoblasts can proliferate considerably before fusing into myofibers, *Correspondence to: Jeffrey S. Chamberlain, Program in Cellular and Molecular Biology, University of Michigan Medical School, Ann Arbor, Michigan 48109-0618. E-mail: firstname.lastname@example.org. Received 20 October 1999; accepted in revised form 21 October 1999 224 D. HARTIGAN-O’CONNOR AND J.S. CHAMBERLAIN Fig. 1. The dystrophin-glycoprotein complex (DGC). The DGC forms a structural link between the actin cytroskeleton and laminin-2 in the extracellular matrix. Integral components of this link include dystrophin, the dystroglycans, the sarcoglycans (SGs), sarcospan (SP), and various proteins associated with the C-terminus of dystrophin, including syntrophin (SYN), dystrobrevin (Db), and MAST. Many forms of muscular dystrophy (MD) are caused by mutations in genes coding for DGC proteins. Shown are congenital MD, one form of which is caused by mutations in a laminin subunit; limb-girdle MD, forms of which are caused by mutations of each of the SGs; Duchenne MD, caused by mutations in the dystrophin gene; and Bethlem myopathy, caused by mutations in subunits of collagen type VI. TABLE 1. DNA sequences used for gene therapy of muscular dystrophy At the present time, direct gene transfer to muscle is a more promising approach for gene therapy of MDs, so this review is focused on transduction of muscle tissue in vivo. Muscle tissue has several advantageous characteristics for gene therapy. First, the bulk of skeletal muscle is easily accessible for experimental manipulation. In addition, muscle tissue is efficiently transduced by commonly used viral vectors, including adenovirus and AAV (see Fig. 4; Kessler et al., 1996; Xiao et al., 1996). Myofibers have long lifespans in vivo, which should facilitate long-term gene transfer. Finally, few alternative treatments are available for the muscular dystrophies, which justifies the development of potentially expensive gene replacement therapies. On the other hand, gene therapy of muscle presents at least one daunting challenge: muscle tissue comprises over 40% of body mass. Most muscular dystrophies—with the possible exception of merosin-deficient CMD (Vilquin et al., 1996)—are caused by cell-autonomous defects, which argues that the majority of muscle cells will have to be treated individually (Phelps et al., 1995; Rafael et al., 1994). Currently most gene delivery to muscle is accomplished by intramuscular injection of vector particles, an impractical approach for treatment of all muscles in the body. In the future, vascular delivery systems will have to be developed to make gene replacement practical (see Fig. 5; Greelish et al., 1999). The various forms of MD are not equally well suited to the development of gene replacement therapy. Duchenne muscular dystrophy has received the most attention because it is the most common type of dystrophy and its pathogenesis was the first to be elucidated at the molecular level (Fig. 1). Replacement of a single isoform of dystrophin in striated muscle is sufficient for elimination of the major symptoms of the disease, an important consideration given the widespread expression of dystrophin isoforms (Cox et al., 1993). Furthermore, low-level dystrophin expression in a simple majority of muscle fibers has been shown to suffice for elimination of symptoms (Phelps et al., 1995). One of the major difficulties with the development of gene Sequence Conventional Ad BMD minigene ⌬H2-R19 minigene Laminin-␣2 Gutted Ad Dystrophin Utrophin AAV ␣-sarcoglycan emerin Lentivirus CMV MCK (full length) MCK (truncated) MCK (synthetic) Human skeletal ␣-actin Sizea (approximate kb) 8–10 6.2 6.0 9.3 27–36 11.2 10.3 4.8 1.2 0.8 8.9 0.7 6.5 3.3 0.6 2.2 aFor viral vector systems, listed in bold, the approximate capacity for foreign DNA is given. For potentially therapeutic genes, the length of the coding region of the cDNA is given. The last five items in the table refer to frequently used promoters. an individual myoblast or myogenic stem cell could correct a large volume of muscle tissue (Engel and Franzini-Armstrong, 1994). Also, corrected muscle fibers would likely display a strong selective advantage over dystrophic fibers, which have a limited half-life (Morgan et al., 1993). Unfortunately, the replicative capacity of myoblasts from dystrophic humans is low, which greatly limits the possibilities for culture, correction, and reinfusion of these cells (Blau et al., 1983). Heterologous myoblast transfer, which could bring normal dystrophin genes into the body, has met with little success (Mendell et al., 1995). This latter scenario has been limited by immunologic rejection of donor myoblasts and by the loss of myogenic potential following large-scale culturing in vitro (Gussoni et al., 1996; Karpati et al., 1993). Nonetheless, recent advances in identifying myogenic stem cells suggest that ex vivo strategies might one day be developed for the muscular dystrophies (Ferrari et al., 1998; Gussoni et al., 1999). DEVELOPMENTS IN GENE THERAPY FOR MD replacement therapy for DMD has proven to be the large size of the dystrophin gene, which has required development of entirely new vector systems for efficient delivery (see Figs. 4, 6; Table 1). The sarcoglycanopathies, as targets for development of gene therapy, share some of the advantages of DMD and are caused by defects in relatively small genes (Fig. 1; see Fig. 7; Table 1). Since alpha- and gammasarcoglycan are expressed primarily or exclusively in striated muscle, gene transfer to striated muscle should correct most symptoms of limb-girdle muscular dystrophy (LGMD) 2D and 2C (Bonnemann et al., 1995; Coral-Vazquez et al., 1999; Jung et al., 1996; Lim et al., 1995; McNally et al., 1994; Nigro et al., 1996; Noguchi et al., 1995; Roberds et al., 1993, 1994; Straub et al., 1999). Beta- and delta-sarcoglycan, however, are expressed in both smooth and striated muscle, so complete correction of LGMD2E and 2F may require more widespread gene expression (Coral-Vazquez et al., 1999; Straub et al., 1999). Though the levels of gene transfer required for functional correction have not been systematically investigated, it is likely that correction of a majority of fibers would be sufficient, given the fact that the sarcoglycanopathies and DMD cause dystrophy through an effect on the same protein complex (Fig. 1; Straub and Campbell, 1997). Perhaps most importantly, the coding region of each sarcoglycan gene is less than 2 kb, enabling delivery using an adeno-associated virus (AAV) vector (Fig. 4; Table 1; Lim and Campbell, 1998). AAV has been shown to deliver neoantigens to muscle without elicitation of an immune response-the basic goal of all gene replacement therapies (Kessler et al., 1996; Xiao et al., 1996). The major drawback to sarcoglycanopathies as models for gene therapy of MD is their relatively low incidence, which is probably less than 5% of that of DMD (Ljunggren et al., 1995). Other muscular dystrophies have received limited attention as candidates for gene therapy. Merosindeficient CMD (MCMD), for example, appears to be caused by a defect that is not entirely cell-autonomous, which could allow phenotypic correction despite transduction of a lower proportion of fibers (Fig. 1; Vilquin et al., 1996). Correction of MCMD has been investigated in transgenic mouse models, where it was shown that expression of laminin-alpha2 from a muscle promoter corrected the muscle phenotype but failed to correct a relatively minor neurological phenotype (Kuang et al., 1998). Unfortunately, the laminin-alpha2 mRNA is about 9.5 kb (Kuang et al., 1998), too long for delivery using AAV or conventional Ad vectors (Table 1). EmeryDreifuss muscular dystrophy (EDMD) is caused by deficiency of emerin, a widely expressed protein of the inner nuclear membrane and intercalated disks (Bione et al., 1994; Bonne et al., 1999; Cartegni et al., 1997; Manilal et al., 1996; Nagano et al., 1996). Since the most lethal features of EDMD are caused by absence of the protein in heart tissue, rather than skeletal muscle tissue, this disease offers the opportunity for effective intervention through treatment of a relatively small mass of tissue (Emery, 1987). In addition, the emerin coding region is less than 1 kb, which allows for its delivery in an AAV vector (Table 1). To date, development of gene therapy for EDMD has been hampered by the lack of an animal model. 225 THERAPEUTIC GENES FOR DELIVERY TO DYSTROPHIC MUSCLE For a recessive genetic disease or a dominant disease caused by haploinsufficiency, delivery of the disease gene itself is an obvious therapeutic choice. Unfortunately, gene replacement will not be feasible in all cases. First, some muscular dystrophies are dominantly inherited and not amenable to gene replacement. Second, gene replacement could lead to an immune response against the therapeutic protein in patients who do not express any significant protein from their disease locus. Finally, the size of the intact disease gene may be too large for delivery with the viral vector of choice (Table 1). To overcome these difficulties, it may be necessary to deliver a minigene, a homologous gene, or a modulatory gene instead of the disease gene. The use of a minigene for therapy of DMD was originally suggested by mutation analysis in patients with mild forms of Becker muscular dystrophy (BMD; Love et al., 1990). A number of these patients had large deletions in their dystrophin genes, which nonetheless led to the accumulation of small proteins with substantial functional capacity (Arahata et al., 1991; Bulman et al., 1991; Monaco et al., 1988). Considerable attention has been focused on a patient with a deletion of 46% of the dystrophin coding sequence but very mild BMD (BMD minigene in Table 1; England et al., 1990). This allele functions well in transgenic animals (Figs. 2, 3; Phelps et al., 1995). Similar minigenes have been delivered to muscles using both adenoviral and retroviral vectors and some efficacy was observed in young or immunosuppressed animals (Deconinck et al., 1996; Dunckley et al., 1993; Ragot et al., 1994; Vincent et al., 1993). The obvious problem with the use of such minigenes is that the means of their identification, study of patients with BMD, ensures that they are less than completely effective (Figs. 2, 3). It is, therefore, hoped that creation of recombinant dystrophin minigenes in the laboratory might yield more effective minigenes. For example, we have observed that modification of some naturally occurring BMD minigenes to restore an integral number of spectrin-like repeats resulted in a smaller molecule with superior effectiveness (⌬H2-R19 in Table 1; Figs. 2, 3; Chamberlain et al., unpublished data). Several other promising alleles, shown to restore the dystrophin-associated glycoprotein complex and small enough to be delivered using AAV, have unfortunately failed to provide functional correction in transgenic mice (Cox et al., 1994, Chamberlain et al., unpublished data). Delivery of a homologous gene may also prove useful in some circumstances. This approach could eliminate concerns about an immune response triggered by delivery of a formerly missing protein. The best example is delivery of utrophin for therapy of DMD. Utrophin is a protein closely related to dystrophin, but which is normally localized to the neuromuscular and myotendinous junctions (Helliwell et al., 1992, 1994). Expression of utrophin uniformly on the sarcolemma using transgenic animal technologies prevented development of dystrophic pathology in mdx mice (Tinsley et al., 1998). Utrophin expression on the sarcolemma could in theory be achieved either by delivery of utrophin in a viral vector or by delivery of factors that stimulate transcrip- 226 D. HARTIGAN-O’CONNOR AND J.S. CHAMBERLAIN Fig. 2. Histology of wild-type, mdx, and transgenic mdx muscle. Top-left panel, marked ⌬H2-R19, shows muscle histology in an mdx mouse made transgenic for a synthetic dystrophin minigene lacking an integral number of spectrin-like repeats. The histology shown is normal, as can be seen by comparison with the wild-type panel at lower-right: variation in fiber size is minimal and myonuclei are located at the periphery of their fibers. The top-right panel, marked ⌬17–48, shows muscle histology in an mdx mouse made transgenic for a minigene isolated from a patient with mild Becker muscular dystrophy. The histology of this muscle is close to normal, but an occasional central nucleus is observed, indicating low-level muscle regeneration. The bottom-left panel shows typical muscle histology in mdx mice. Extreme variation in fiber size, frequent centrally nucleated fibers, and some fibrosis are observed. Prominent central nucleation and fiber size variation indicate ongoing muscle regeneration. DEVELOPMENTS IN GENE THERAPY FOR MD 227 Fig. 3. Functional correction of dystrophy in transgenic mdx mice. The specific force, or force per unit cross-sectional area, produced by dystrophic muscle (mdx) is reduced compared to that produced by healthy muscle (C57). Transgenic mdx mice bearing a Becker muscular dystrophy (BMD) dystrophin minigene, called ⌬17–48, display partial correction of this parameter. Transgenic mice bearing a synthetic minigene with an integral number of spectrin-like repeats, called ⌬H2R19, are fully corrected. tion of the endogenous gene (Gramolini et al., 1999; Khurana et al., 1999). This approach might be advantageous because the utrophin coding region is almost as large (10.3 kb) as that of dystrophin (Table 1; Tinsley et al., 1992). The possibility of using homologous gene products for therapy has also been raised in other forms of muscular dystrophy, though there is no evidence yet that this would work. For example, M-laminin or merosin appears later in muscle development than does an earlier isoform called A-laminin, which differs from merosin only in the identity of its heavy chain (Fig. 1; Leivo and Engvall, 1988; Sanes et al., 1990; Xu et al., 1994). If the A-laminin heavy chain could be delivered to muscle or expression of the endogenous locus could be forced, perhaps this approach would be sufficient to alleviate dystrophy. Similarly, epsilon-sarcoglycan is a widely expressed gene with similarity to alpha-sarcoglycan, and it has been suggested that high-level expression of epsilon-sarcoglycan might allow for restoration of the sarcoglycan complex in LGMD type 2D (Ettinger et al., 1997). Another approach to treating the muscular dystrophies would be to deliver a modulatory gene together with, or instead of, a replacement for the mutant gene. The MDs are characterized by constant cycles of muscle fiber degeneration and regeneration that eventually fail (Fig. 2; Blau et al., 1983; Emery, 1993; McArdle et al., 1995). If these cycles could be slowed, or if regeneration could be made more robust, it is reasonable to assume that the dystrophic process would be delayed or alleviated. Genes that might achieve these goals include trophic factors that stimulate muscle hypertrophy, such as IGF-1 (Barton-Davis et al., 1998); regulatory molecules that can stimulate formation of new myogenic precursors, such as MyoD (Lattanzi et al., 1998; Megeney et al., 1996); or inhibitors of necrosis or apoptosis, such as calpastatin or Bcl family members (Tidball et al., 1995). REQUIREMENTS FOR GENE THERAPY OF MUSCULAR DYSTROPHY Most muscular dystrophies are caused by absence of a protein in muscle cells. Although some dystrophies have less obvious effects on the central nervous system, dysfunction of heart or skeletal muscle leads to death. For DMD and MCMD, it has been proven that gene replacement in striated muscle alone can alleviate the major features of the disease (Cox et al., 1993; Kuang et al., 1998). For these reasons, gene delivery to muscle cells is the most likely route to an effective treatment. Achievement of this goal will require (1) a suitable vector that can be mass-produced cheaply and (2) an efficient means for delivery of the vector to the surface of muscle fibers. Vector delivery to the muscle surface is usually accomplished on a small scale by multiple injections throughout the muscle; however, a systemic vascular delivery system would be vastly superior if one could be developed. Dystrophin-positive fibers appear to have a survival advantage over negative fibers, which might allow for use of a less efficient delivery system in the long term, especially if this selective advantage also applies to the other forms of dystrophy (Morgan et al., 1993). A vector that can achieve long-term persistence will be required. Since many patients with MD are diagnosed early in life, and the goal of an ideal therapy is extension of lifespan into the normal range, the ideal therapy would retain its effectiveness for about 70 years. This goal could be achieved by multiple treatments throughout life, but each individual treatment should retain effectiveness for at least several years. Experience has shown that long-term effectiveness depends on avoidance of the host immune response, so an appropriate vector must evade the immune system or be administered together with immune suppression. A suitable vector must also be capable of carrying and delivering the therapeutic gene(s). In some cases the disease gene will be large, but even small genes may need to be regulated using large, tissue-specific regulatory elements or might need to be delivered together with a modulatory gene. Several approaches can be taken to facilitate delivery of long stretches of DNA. First, development of new viruses as gene delivery vectors should be pursued. Development of a herpes simplex virus (type 1) system, for example, is encouraging in this regard (Huard et al., 1995; Marconi et al., 1999). Second, conventional vector systems could be adapted so as to increase their capacity for foreign DNA, as has been recently achieved with the development of gutted Ad vectors (Fig. 4; see Fig. 6; Fisher et 228 D. HARTIGAN-O’CONNOR AND J.S. CHAMBERLAIN Fig. 4. Vectors for gene therapy. The genome of Wild-type Ad may be divided into early and late regions. Early regions, designated by gradient shading, are replaced by exogenous DNA in conventional Ad vectors. 1st generation vectors contain deletions in the E1 and E3 regions. 2nd generation vectors contain deletions in additional early regions-E2B in this example. Ad5␤dys is an example of a gutted Ad vector, in which virtually the entire genome can be replaced by exogenous DNA. AAV vectors also contain deletions of nearly the entire viral genome, except for the viral ITRs, which are required for replication and packaging of the vector. The ITRs are only 145 bp in size and are not shown in this diagram. Lentiviral vectors contain an initial short stretch of viral RNA required for export and packaging of full-length transcripts in the packaging line. The remainder of the genome may be replaced with exogenous sequences, except for a small region near the polyadenylation signal that is required for reverse transcription. al., 1996; Kochanek et al., 1996; Kumar-Singh and Chamberlain, 1996). Use of these vectors eliminates the host immune response against viral proteins and may reduce the response against the therapeutic protein (Schiedner et al., 1998). These vectors have recently allowed our laboratory and others to deliver dystrophin without elicitation of a major immune response (Fig. 4; see Fig. 6). and production of immunogenic late proteins occurs only if the early phase progresses through the onset of viral DNA replication. Conventional Ad vectors are created by the replacement of viral early genes with an exogenous expression cassette (Fig. 4). The vectors can be propagated in packaging cell lines that express the deleted early genes, but are essentially replication-defective in noncomplementing cell lines and in vivo. First-generation vectors lack E1, which is involved in transactivation of other viral genes, so immunogenic viral proteins are expressed at a greatly reduced level (Gaynor and Berk, 1983; Nevins, 1981; Yang et al., 1996b). Adenoviral vectors have some advantageous properties for gene therapy. The serotypes used in creation of gene delivery vectors usually cause only mild or subclinical disease in the wild (Brandt et al., 1969; Schmitz et al., 1983). As a result, inadvertently generated, replication-competent virus is unlikely to harm an immunecompetent patient. Adenoviruses can be readily grown to very high titer. Since Ad virions remain tightly associated with lysed cells, within whose nucleus they are packed into crystalline arrays, concentrated stocks are easily prepared through low-speed centrifugation and collection of lysed cells. Adenoviruses efficiently infect most human cell types, including immature muscle cells (see Fig. 7), though most lymphocytes are relatively resistant. Finally, the popularity of Ad vec- VIRAL VECTORS FOR MUSCLE GENE THERAPY Adenoviral Vector Technology Adenoviral replication is usually divided into two stages, early and late, which are divided by the onset of DNA replication. Only a subset of viral genes, called E1, E2, E3, and E4, are expressed during the early phase (Fig. 4). These genes prepare the host cell for viral replication by stimulating production of necessary precursors and helping to prevent a host immune response. During the late phase genes for structural components of the Ad virion are expressed. Obviously, these structural proteins must produced at high levels; perhaps because of this, they are also the principal targets of the host immune response (Jooss et al., 1998a). Expression of the late genes requires DNA replication, though the mechanism behind late gene activation is not entirely understood (Thomas and Mathews, 1980). As a result, expression of late genes DEVELOPMENTS IN GENE THERAPY FOR MD tors has led to development of simple means for their manipulation (Chartier et al., 1996; reviewed in Graham and Prevec, 1991). Unfortunately, Ad vectors have drawbacks that limit their usefulness for gene replacement therapy of muscular dystrophy. The cloning capacity of first-generation vectors is only about 8 kb, which is the amount of space made available by deletion of E1 and E3 (Fig. 4; Table 1; Bett et al., 1993). Obviously, first-generation Ad vectors cannot deliver full-length dystrophin, since the coding region of dystrophin alone is 11 kb. These vectors can, however, be used to deliver shortened forms of dystrophin that are generated in the laboratory. Ad vectors can also be used to deliver smaller genes involved in other forms of muscular dystrophy (see Fig. 7). A more serious difficulty with first-generation Ad vectors is leaky expression of immunogenic viral proteins in vivo. Despite deletion of E1, viral gene expression and even limited replication can occur in noncomplementing cells over a longer time scale. Expression of viral proteins leads to a host immune response and elimination of gene expression from transduced tissues (Dai et al., 1995; Dong et al., 1996; Gaynor and Berk, 1983; Nevins, 1981; Van Ginkel et al., 1995; Yang et al., 1994, 1996b; Yang and Wilson, 1995; Zsengeller et al., 1995). The resulting inflammatory process can even lead to muscle damage and exacerbation of weakness. Immunosuppressive drugs can partially overcome this problem (Lochmuller et al., 1996), but immunosuppression has its own risks and so development of an improved vector would be preferable. Second Generation Ad Vectors Several groups created new Ad vectors lacking additional early genes in an effort to address problems with first-generation vectors (Fig. 4; Amalfitano et al., 1998; Armentano et al., 1995; Gorziglia et al., 1996; Schaack et al., 1995; Wang et al., 1995; Weinberg and Ketner, 1983). The deletions target E2A, E2B, and E4, which are the remaining early regions of the Ad genome. For each deletion a corresponding packaging cell line must be generated, which expresses the missing proteins in trans. These second-generation vectors provide additional cloning capacity and should further attenuate the virus. Further inhibition of viral replication is desirable for two reasons. First, these highly modified second-generation vectors are less likely to generate replication-competent virus during large-scale, clinicalgrade vector preparation. Second, since expression of viral late genes requires replication, complete inhibition of Ad genome replication should abolish late gene expression, which would eliminate the host immune response against late proteins (Amalfitano et al., 1998). E2A, E2B, and E4 deletions introduced into Ad vectors have provided about 1.4 kb, 1 kb, and 1.9 kb of additional cloning capacity, respectively. Individually, none of the deletions yields a dramatic increase in cloning capacity; nonetheless, combining two or three of these additional deletions might allow delivery of fulllength dystrophin via an Ad vector. However, since many second-generation vectors grow to lower viral titers than first-generation vectors or wild-type Ad (Zhou et al., 1996), combining all these deletions into a single vector may prove difficult. 229 Second-generation vectors have been shown to elicit a reduced immune response and thereby allow prolonged transgene expression (Gao et al., 1996; Hu et al., 1999; O’Neal et al., 1998; Wang et al., 1997). For example, an E1-, E2B-, and E3-defective vector designed in our laboratory demonstrated prolonged transgene expression and persistence of vector genomes, along with reduced toxicity, in the liver (Hu et al., 1999). Similar experiments to test the properties of E2B-defective vectors in muscle are underway. Secondgeneration vectors therefore offer significant promise in some settings, but further investigation of their advantages for treatment of MD are required. Gutted Adenoviral Vectors Gutted, or helper-dependent, Ad vectors may overcome many drawbacks associated with conventional Ad technology (Fig. 4). These deleted genomes were the first Ad vectors to be developed (Solnick, 1981; Thummel et al., 1981), but interest in their use has expanded recently with several demonstrations of their advantages for gene transfer. Growth and purification of these viruses was extremely laborious until the recent development of new techniques (Hardy et al., 1997; Hartigan-O’Connor et al., 1999; Parks et al., 1996). Further improvements will still be required before routine, large-scale growth of clinical-grade gutted virus is feasible. Gutted vectors contain cis-acting DNA sequences that direct adenoviral replication and packaging but do not contain viral coding sequences (Fig. 4; Fisher et al., 1996; Kochanek et al., 1996; Kumar-Singh and Chamberlain, 1996). Theoretically, the vectors can accommodate up to about 37 kb of exogenous DNA, though 28–30 kb is more typical (Fig. 4; Table 1). Since gutted vectors do not contain any viral genes, expression of viral proteins is not possible. Gutted vectors are defective viruses produced by replication in the presence of a helper virus, which provides all necessary viral proteins in trans. Like other defective viruses, gutted viruses are normally prepared as a mixture with helper virus. The starting material for production of all gutted viruses is plasmid DNA (Fisher et al., 1996; Kochanek et al., 1996; Kumar-Singh and Chamberlain, 1996). These plasmids contain the viral origins of replication (ITRs), the packaging signal (psi), and therapeutic DNA to be carried by the gutted virus. Co-transfection of this plasmid and helper viral DNA into a packaging cell line leads to replication of the helper virus and concomitant replication of the gutted virus, as directed by the viral ITRs contained in the starting plasmid. Robust helper virus replication causes lysis of the transfected cells. The resulting lysate contains a large number of helper virions and a relatively small number of gutted virions. To increase the number and proportion of gutted virions in the lysate, the initial mixture must be serially passaged. During serial passage, for unknown reasons, gutted virus is amplified more quickly than helper virus and eventually substantial enrichment occurs. Particles containing gutted viral genomes, rather than helper genomes, must then be purified on the basis of their lower density (Fisher et al., 1996; Kochanek et al., 1996; Kumar-Singh and Chamberlain, 1996). 230 D. HARTIGAN-O’CONNOR AND J.S. CHAMBERLAIN A gutted vector can accommodate not only the fulllength dystrophin cDNA but also expression cassettes coding for marker or modulatory proteins. The gutted vector Ad5␤dys, for example, contains the full-length dystrophin cDNA driven by a 3.3 kb muscle creatine kinase promoter as well as a lacZ gene controlled by the CMV promoter for monitoring vector production in vitro and gene delivery in vivo (Fig. 4; Kumar-Singh and Chamberlain, 1996). We have developed other vectors that contain dystrophin, either murine or human, and combinations of other regulatory, reporter gene, and phenotypic modulatory sequences (see Fig. 6). The very large cloning capacity of gutted vectors opens a new world of therapeutic possibilities that were previously not possible. Ad5␤dys delivers full-length dystrophin (and ␤-galactosidase) to the skeletal muscles of mdx mice with an efficiency comparable to that of first-generation viruses (Kumar-Singh and Chamberlain, 1996; Hauser et al., unpublished data). Gene expression persists for over 1 year in immune-deficient SCID/mdx mice, which indicates that the complete absence of viral gene expression does not destabilize gutted vector genomes in vivo. In immune-competent mdx mice, by contrast, expression of dystrophin persists for less than 1 month, suggesting that dystrophin expression may be eliminated by an immune response. Removal of all viral coding sequences from Ad5␤dys was therefore insufficient, by itself, to eliminate a host immune response against all antigens. A newer version of this gutted vector in which ␤-galactosidase is expressed from an inducible promoter leads to dystrophin expression for at least several months in immune-competent mice (Hauser et al., unpublished data). Gutted vectors should be designed to provide targeted expression of therapeutic proteins in the affected tissue only, a principle that was not followed in construction of Ad5␤dys. As is discussed further below, limiting gene expression to the target tissue may allow avoidance of the immune response and long-term gene expression. Adeno-Associated Virus (AAV) Vectors AAV particles were first identified by electron microscopic examination of human and simian Ad preparations (Atchison et al., 1965; Melnick et al., 1965). The particles were soon recognized as defective viruses that normally replicate only in the presence of Ad (Hoggan et al., 1966). It is now known that several other viruses and a variety of genotoxic treatments can also provide helper functions (Schlehofer et al., 1986; Yakinoglu et al., 1988; Yakobson et al., 1987, 1989). In fact, no helper virus genes are directly involved in replication of AAV DNA; instead, the helper virus seems to maximize synthesis of cellular proteins involved in AAV replication (Muzyczka, 1992). AAV is a 4,680-bp parvovirus with a fascinating and unusual life cycle. In the absence of helper virus, AAV infection is nonproductive and no progeny AAV particles are produced. Instead, the ssDNA genome may be converted to a double-stranded form and become covalently associated with cellular DNA at a specific locus on chromosome 19 (Kotin et al., 1990). If a helper virus subsequently infects the cell, the AAV genome is excised and begins to undergo lytic growth. The growth of AAV reduces production of helper virus particles and lysis of the cell eventually produces more AAV particles than Ad particles (Atchison et al., 1965). Gene delivery vectors based on AAV are prepared by replacement of all viral coding sequences with therapeutic DNA (Fig. 4; Muzyczka, 1992). The total amount of exogenous DNA that can be carried by AAV vectors is currently less than 5 kb, which unfortunately eliminates many muscular dystrophy genes from consideration (Table 1). The only remaining viral sequences are the AAV ITRs, which direct replication and packaging of the vector construct. Both AAV proteins and helper virus proteins must be provided in trans, an arrangement that will be recognized as similar to that of gutted adenovirus vectors. Originally, AAV vector stocks were therefore produced as a mixture of vector particles and helper virus particles (Hermonat and Muzyczka, 1984). Within the last 2 years, however, helper-virus-free packaging systems have been developed based on a better understanding of the requirements for efficient AAV replication (Matsushita et al., 1998; Salvetti et al., 1998; Xiao et al., 1998). Using these systems, pure AAV vectors can be produced at titers nearly equivalent to those of wild-type Ad. AAV was originally developed as a gene therapy vector based on the hope that genomic integration would allow long-term expression of transgenes. AAV has, in fact, been a successful vector because in some tissues it provides very long-term expression of transgenes in the absence of an immune response. For example, Xiao et al. (1996) found that expression of ␤-galactosidase from an AAV vector was stable for more than 8 months in the muscle of adult, immunecompetent animals. Though humoral responses against the AAV particles were observed, no cellular immune response was apparent. This result is astonishing given the robust cellular response observed after transduction of the ␤-galactosidase gene using Ad vectors. It has turned out that lack of a cellular immune response, not genomic integration, is the most important factor in maintenance of gene expression in muscle cells. Jooss et al. (1998b) showed that administration of an Ad-lacZ vector can elicit a cellular immune response that destroys muscle fibers previously transduced by AAV-lacZ, which could otherwise survive and continue to express ␤-galactosidase. These authors also demonstrated that AAV-lacZ vectors fail to transduce antigenpresenting cells (APCs) whereas Ad vectors efficiently transduce such cells. These data suggest that AAV vectors achieve immune evasion and persistent gene expression through avoidance of antigen presentation by professional APCs such as dendritic cells. Furthermore, it seems that most of the AAV genomes in muscle tissue are present in the form of large circular multimers (Duan et al., 1998). Formation of such multimers may play a role in the persistence of vector DNA, but Ad genomes can also survive inside the nucleus for long periods, so physical persistence of episomal DNA can be achieved through several mechanisms. The striking success of AAV vectors in long-term transduction of muscle tissue has given rise to hope that muscular dystrophies caused by defects in small genes may soon be treated using this vector. Some of the exciting results achieved so far are described below. Just as importantly, our insight into the mechanisms used by AAV to achieve persistent gene expression may DEVELOPMENTS IN GENE THERAPY FOR MD be applicable to other vector systems that have not yet achieved comparable success. Lentiviral Vectors Murine retroviral vectors were used in some of the earliest efforts at gene replacement for DMD (Dunckley et al., 1992, 1993). As murine retroviral vectors are unable to stably infect non-dividing cells, post-mitotic muscle fibers were not targeted in these studies. Instead, investigators attempted to transduce proliferating myoblasts in vitro or in vivo. Although transduction of myoblasts with retroviral vectors is effective and does lead to gene transfer after myoblast transplantation, the significant technical hurdles to myoblast transplantation remain. Transduction of myoblasts in vivo might be expected to be effective, since muscle regeneration is such a prominent feature of dystrophic pathology (Fig. 2). Unfortunately, direct injection of a murine retrovirus carrying minidystrophin resulted in transduction of only a few percent of fibers at the site of injection (Dunckley et al., 1993). Injection of bupivicaine, which induces muscle necrosis and regeneration, doubled the number of transduced fibers. More recently, implantation of retroviral-producer cells resulted in transduction of greater than 10% of fibers (Fassati et al., 1997). Such an approach is, however, limited by problems associated with immunological rejection of the producer cells. Finally, the 7.8 kb cloning capacity of retroviral vectors prevents delivery of full-length dystrophin cDNA clones. As a result, the use of the murine retroviruses for gene replacement therapy in muscle will require significant technical innovation before it becomes a promising approach. Recently, however, new retroviral vectors have been introduced that can integrate into the genome of postmitotic cells (Carroll et al., 1994; Naldini et al., 1996; Poeschla et al., 1998). These vectors are based on human or feline lentiviruses, which infect nondividing cells as part of their normal life cycle (Lewis et al., 1992; Weinberg et al., 1991). Like the older murine vectors, these vectors are produced by expression of a packageable vector construct in a cell line that expresses viral proteins (Fig. 4). Vector particles bud from the surface of such cells continuously and can be harvested from the supernatant. Because the lentivirus life cycle is nonlytic and because the enveloped vector particles are not as stable as naked Ad or AAV particles, the vector titers obtained are relatively low (about 109 per ml) even after concentration. The size of the vector construct is limited to about 10 kb by constraints on retroviral packaging, which again prohibits transfer of the full-length dystrophin cDNA (Table 1). Lentiviral vectors efficiently transduce post-mitotic neurons, hepatocytes, and muscle fibers (Kafri et al., 1997; Naldini et al., 1996). Injection of a lentiviral vector expressing GFP into muscle tissue resulted in transduction of about half the muscle fibers at the site of injection (Kafri et al., 1997). Gene expression was maintained for at least 2 months without elicitation of a major immune response. It should be noted, however, that the number of particles injected was several orders of magnitude lower than the number of Ad or AAV particles that are usually injected. Elicitation of a significant cellular immune response may require attainment of a certain gene expression threshold, which 231 may not have been achieved in this case (Tripathy et al., 1996). Alternatively, perhaps the VSV-G envelope protein, with which the vector was pseudotyped, does not allow efficient infection of dendritic cells. Further development of this technology might someday allow for effective treatment of muscular dystrophy. DELIVERY OF VIRAL VECTORS TO MUSCLE TISSUE A single injection into muscle tissue typically results in transduction of cells within 0.5–1 cm of the injection site. This limited vector diffusion means that saturation of a single muscle group would require multiple injections. Saturation of many large human muscles, including the heart and diaphragm, would require hundreds of injections and is probably impractical on a large scale. Therefore, if viral vectors can be proven suitable for stable transduction of muscle on a small scale, a means for vector delivery to a large mass of muscle tissue needs to be developed. Arterial delivery holds promise as a means to overcoming this difficulty (Fig. 5; Greelish et al., 1999; Welling et al., 1996). Unfortunately, transport of viral vectors across normal vascular endothelium is poor, even when high hydrostatic pressure is applied (Greelish et al., 1999; Jejurikar et al., 1997). Greelish et al. (1999) found that arterial instillation of Ad-lacZ under elevated hydrostatic pressure led to efficient transduction of vasculature but resulted in limited muscle infection (Fig. 5A–C). After vasodilation with papaverine and endothelial permeabilization with histamine, however, widespread transduction of entire muscle groups was achieved (Fig. 5D, G, H). Transduction was strictly limited to muscle groups served by the perfused arteries, confirming that gene delivery was accomplished through arterial delivery (Fig. 5E, F). The treatment was effective in both hindlimb and heart using either Ad or AAV vectors (Fig. 5I). If arterial delivery can be performed safely and routinely, then transduction of the majority of human muscle tissue for gene replacement may be feasible. Effective gene delivery depends not only on transport of vector to target cells, but also on subsequent infection. Tropism of Ad vectors for muscle fibers has been observed to decline with muscle maturation, during which process myofibers down-regulate expression of cellular receptors for adenovirus (Acsadi et al., 1994a; Acsadi et al., 1994b; Nalbantoglu et al., 1999). As a result, infection of immature or regenerating muscle by Ad is more efficient than infection of mature muscle. AAV, in contrast, seems to infect mature muscle as efficiently as immature muscle (Snyder et al., 1997). It is unknown whether the infectability of dystrophic human muscle will present a serious impediment to gene therapy of MD with Ad vectors, as little is known about Ad tropism for human muscle. Some success has been attained in modifying the tropism of adenovirus to improve infection of macrophage, endothelial, smooth muscle, fibroblast, and T cells (Douglas et al., 1996; Stein et al., 1999; Wickham et al., 1997); however, engineering of an adenovirus with dramatically increased tropism for skeletal muscle has not yet been achieved. Combining a vascular delivery method with a tropism-modified Ad vector could result in efficient gene 232 D. HARTIGAN-O’CONNOR AND J.S. CHAMBERLAIN Fig. 5. Gene transfer across the endothelial barrier: histamine and papaverine increase permeability to viral particles. A–C: Pattern of gene transfer in the absence of inflammatory mediators demonstrates interference of microvascular barrier with adenovirus transport. A: ␤-galactosidase activity in whole-mount-stained leg of adult rat 4 days after arterial infusion with AdCMVlacZ. X-gal shows that virus uptake is limited to microvasculature; muscle fibers do not stain. B,C: Staining is seen in most capillaries but absent in muscle fibers of the tibialis anterior. D–I: Efficient gene transfer to adult skeletal muscle fibers after forced exudation with histamine and papaverine. D: Entire hindlimb from rat dissected before whole-mount-staining to expose multiple cross-sections; universal fiber uptake is visible on most cross-sections. E,F: Gross (E) and light microscopic (F) appearance of marker gene distribution in the quadriceps shows detail on adjacent rectus femoris and vastus medialis. A tourniquet at the level of the common femoral artery occluded blood supply to the rectus femoris, preventing virus delivery and infection. G: Semimembranosus and adductor brevis with adjacent saphenous artery. H: Nomarski micrograph of tibialis anterior shows unstained wall of arteriole (right center) against a backdrop of uniformly stained muscle fibers. I: Heterotopically transplanted heart after isolated perfusion with histamine and papaverine analogous to that used for isolated limb. Reprinted from Greelish J P, Su L T, Lankford E B, Burkman J M, Chen H, et al. 1999. Stable restoration of the sarcoglycan complex in dystrophic muscle perfused with histamine and a recombinant adenoassociated viral vector. Nature Med 5:439–443, with permission of the publisher. transfer to human muscle using a systemic delivery method. GENE REPLACEMENT FOR DMD Transgenic mice have proven the feasibility of gene replacement in muscle as a means for treatment of DMD (Figs. 2, 3; Corrado et al., 1996; Cox et al., 1993; Phelps et al., 1995; Rafael et al., 1994; Wells et al., 1995). Several investigators have created mdx mice expressing a variety of levels of full-length or truncated dystrophins in muscle tissue. More recently, similar studies have been performed using utrophin vectors (Tinsley et al., 1996, 1998). These animals have provided detailed information on the level of gene expression needed to prevent occurrence of dystrophy, the functional capacity of different therapeutic molecules, and the percentage of muscle fibers that must express the transgene to prevent disease progression. After generation of transgenic mice and confirmation that a particular molecule is effective in preventing dystrophy, transduction using viral vectors can be attempted. Obviously, it has been more difficult to demonstrate functional correction of dystrophy using viral vectors than to demonstrate prevention of dystrophy with transgenes that are uniformly expressed in all muscles. Since most viral vectors are unable to carry a fulllength dystrophin, many studies have focused on dystrophin minigene vectors. Transgenic mdx mice have been created that express a truncated dystrophin minigene of 6.3 kb in length (Fig. 2; Phelps et al., 1995). This molecule, missing about half of the coding region, was based on the mutation discovered in a mildly affected Becker muscular dystrophy patient with a deletion of exons 17–48 (BMD minigene in Table 1). Few signs of dystrophy were observed in mice expressing this minidystrophin, though low-level muscle regeneration was detected on careful histological analysis (Phelps et al., 1995). Muscles from these animals also display slight deficits in their ability to generate force and resist injury (Fig. 3). These data show that this minidystrophin does not completely prevent dystrophy; however, the observed lack of fibrosis and almost normal force development indicate that smaller proteins could be useful in a clinical setting. In fact, a recombinant adenovirus expressing this minidystrophin was shown to prevent muscle pathology if injected within one week of birth (Deconinck et al., 1996; Vincent et al., 1993). Older mdx mice and dystrophic humans represent more difficult challenges due to a functional immune system, advanced pathology, and down-regulation of adenovirus receptors. Despite these difficulties, the same virus was found to be reasonably effective when injected into older mdx mice that had been immunosuppressed with FK506 (Lochmuller et al., 1996; Yang et al., 1998). These data offer hope that more effective minidystrophin molecules might be designed in the laboratory using knowledge of dystrophin structure and function. The BMD deletion described above results in the loss of a non-integral number of spectrin-like repeats forming the rod domain of dystrophin. Since this truncation likely disrupts the structure of the dystrophin rod domain, a variety of modified truncations have been engineered in attempts to develop a highly functional dystrophin encoded by a cDNA less than 6 kb in size (Phelps et al., 1995; Rafael et al., 1994; Yuasa et al., 1998; Hauser and Chamberlain, unpublished data). Transgenic mdx mice expressing some of these modified alleles have recently been observed to be more effective than the exon 17–48 truncation in preventing muscular dystrophy in mdx mice (Figs. 2, 3). Experiments to assess the efficacy of delivery of these modified cDNAs to dystrophic muscle using Ad and retroviral vectors are currently in progress. As described earlier, sarcolemmal expression of utrophin, a dystrophin-related protein normally localized to DEVELOPMENTS IN GENE THERAPY FOR MD 233 Fig. 6. Delivery of full-length dystrophin using a gutted adenoviral vector. Ghumdys is a gutted adenoviral vector containing the fulllength human dystrophin cDNA driven by a muscle creatine kinase (MCK) promoter. The vector contains no reporter gene. Top: Expression of dystrophin in mdx mouse muscle 5 days after injection of the vector. Note cytoplasmic accumulation of dystrophin, indicating highlevel expression. Bottom left: Marked DYS, shows immunofluores- cence staining for dystrophin. Arrowheads have been placed within two weakly positive fibers. Bottom right: Marked Merge, shows accumulation of Evans blue dye in dystrophin-negative fibers. Evans blue is a dye that permeates fiber with sarcolemmal disruption, indicating severe injury. Note that the weakly dystrophin-positive fibers do not take up dye, despite their proximity to damaged fibers. neuromuscular and myotendinous junctions, prevents dystrophy in transgenic mdx mice (Tinsley et al., 1998). This observation is important because it is possible that DMD patients with null alleles could mount an immune response against dystrophin, which might be perceived as a neoantigen. Even mdx mice, which express a low level of dystrophin in revertant myofibers, have been shown to mount a CTL response against dystrophin expressed by transplanted C57/BL10 myoblasts (Ohtsuka et al., 1998). In contrast, no strong immune response has been noted in experiments using gutted vectors to deliver dystrophin to mdx mice, so concerns about immune response in human patients may prove unfounded. Considerable work has been done with a synthetic utrophin minigene analogous to the common exon 17–48 deleted minidystrophin (Deconinck et al., 1997; Tinsley et al., 1996). In transgenic mdx mice, this allele provides significant improvement in dystrophic pathology based on histological and functional criteria; in fact, the improvement in phenotype is roughly similar to that obtained with the 6.3 kb minidystrophin (Phelps et al., 1995). Undoubtedly the most direct and efficacious route to gene therapy of DMD would be delivery of full-length dystrophin itself-true gene replacement (Fig. 6). Expression of even low levels of dystrophin in transgenic mdx mice eliminated the major symptoms of dystrophy. Dystrophin expression at levels between 20 and 5,000% of wild-type prevented all signs of dystrophy without deleterious side effects (Cox et al., 1993; Phelps et al., 1995). Mouse and human dystrophin molecules were found to be equally effective in these studies (Phelps et al., 1995). Animals with expression levels below 20% of wild-type level displayed intermediate signs of dystrophy, indicating that even low-level dystrophin expression may have therapeutic benefit. Some lines of transgenic mice produced dystrophin in a variable pattern (Phelps et al., 1995; Rafael et al., 1994). These nonuniformly expressing lines provide data on the percentage of fibers that must be transduced to alleviate symptoms. Comparison of the dystrophic pathology of several such lines indicated that at least 50% of muscle fibers must accumulate moderate amounts of dystrophin to prevent a severe dystrophy. 234 D. HARTIGAN-O’CONNOR AND J.S. CHAMBERLAIN The major difficulty with delivery of full-length dystrophin is the fact that most gene delivery vectors cannot accommodate the large size of the gene (Table 1). Until recently, full-length dystrophin could only be delivered to muscle using myoblast transplantation or naked DNA injection. However, as discussed above, recent improvements in helper-dependent or ‘‘gutted’’ Ad vectors have allowed delivery of full-length dystrophin (Fig. 6; Kochanek et al., 1996; Kumar-Singh and Chamberlain, 1996). We found that a gutted vector containing both dystrophin and ␤-galactosidase expression cassettes, Ad5␤dys (Fig. 4; Kumar-Singh and Chamberlain, 1996), efficiently delivered full-length dystrophin to skeletal muscle, but expression persisted for less than 1 month. The presence of a lymphocytic infiltrate surrounding dystrophin- and ␤-galactosidasepositive fibers suggested that an immune response was responsible for loss of expression. Since Ad5␤dys does not contain any viral coding sequences, it seemed likely that the immune response was directed against either dystrophin or ␤-galactosidase, the two transgenes carried by the vector (Fig. 4). ␤-galactosidase, which plays no therapeutic role in the vector, is known to elicit a potent T-cell-mediated response when delivered to immune-competent mice using conventional Ad vectors (Tripathy et al., 1996; Yang et al., 1996a). To reduce expression of ␤-galactosidase, we created a new form of the gutted vector, GE␤dys, in which the inducible ecdysone promoter drives lacZ expression. Use of this promoter greatly limits expression in mouse and human tissues, which do not express the ecdysone receptor (EcdR), but allows for titering of gutted vector preparations in modified 293 cells expressing EcdR. Since provocation of a strong CTL response is thought to require gene expression in antigen-presenting cells, attenuation of gene expression by use of such a promoter should reduce the immune response. In fact, lymphocytic infiltration of mdx muscles injected with GE␤dys was dramatically reduced, indicating successful reduction of the CTL response despite robust dystrophin expression in myofibers. Muscles injected with GE␤dys expressed dystrophin for at least 4 months post-injection, which was the longest time point tested (Salvatori et al., unpublished data). These data suggest that gutted vectors will allow delivery and long-term expression of full-length dystrophin, although avoidance of immune response remains a critical issue. Additional studies addressing the persistence and functional capacity of gutted vectors lacking all reporter genes have yielded highly encouraging results, suggesting that phase I clinical trials of gutted vectors should proceed (Fig. 6). GENE REPLACEMENT FOR SARCOGLYCANOPATHIES The dystrophin-glycoprotein complex found in skeletal muscle includes four related proteins named sarcoglycans (Fig. 1; Campbell and Kahl, 1989; Ervasti et al., 1990; Yoshida and Ozawa, 1990). Sarcoglycan deficiency is now known to cause four forms of autosomal recessive LGMD-one for each sarcoglycan molecule (Bonnemann et al., 1995; Jung et al., 1996; Lim et al., 1995; Noguchi et al., 1995; Roberds et al., 1994). Fig. 7. Recombinant ␦-SG adenovirus mediates high efficiency gene transfer into BIO 14.6 hamster muscle. A: Age-matched quadriceps from F1B, BIO 14.6, and BIO 14.6 injected with 109 particles of ␣-SG adenovirus or ␦-SG adenovirus (tissue harvested 7 days postinjection) were analyzed by immunofluorescence using an antibody specific for ␦-SG. Bar ⫽ 50 µm. B: ␦-SG adenovirus particles (109 ) were injected into the quadriceps femoris muscle of BIO 14.6 hamster. Seven days later, muscle cryosections were prepared and subjected to immunofluorescence using antibodies specific for individual sarcoglycan proteins, as shown. Bar ⫽ 50 µm. Reprinted from Holt K H, Lim L E, Straub V, Venzke D P, Duclos F, et al. 1998. Functional rescue of the sarcoglycan complex in the BIO 14.6 hamster using delta-sarcoglycan gene transfer. Mol Cell 1:841–848, with permission of the publisher. Collectively these forms of muscular dystrophy are called sarcoglycanopathies. A naturally occurring hamster mutant, the BIO 14.6 cardiomyopathic hamster, is available for delta-sarcoglycan (delta-SG) deficiency, so gene replacement studies have focused on this gene (Fig. 7; Nigro et al., 1997; Sakamoto et al., 1997). It is likely that a gene replacement therapy effective in amelioration of delta-SG deficiency could be directly adapted to the other sarcoglycanopathies. Holt et al. were the first to explore the feasibility of gene replacement in the BIO 14.6 hamster (Fig. 7; Holt et al., 1998). Due to the small size of the sarcoglycan genes, these authors were able to deliver delta-SG in a conventional E1, E3-deleted adenovirus (Fig. 7A, Table 1). They found that injection of delta-SG Ad, but not control alpha-SG Ad, restored all four sarcoglycans to the sarcolemma (Fig. 7B). Injection of young (3-week-old) hamsters with the Ad vector resulted in expression for at least several months, which also occurs after injection of Ad into young, but not mature, mice. Delta-SG Ad injection also prevented the development of morphological abnormalities, as assessed by the percentage of muscle fibers with central nucleation, and maintained sarcolemmal integrity. These results are very similar to those achieved in young mdx mice after delivery of full-length dystrophin DEVELOPMENTS IN GENE THERAPY FOR MD or dystrophin minigenes (Deconinck et al., 1996; Vincent et al., 1993). Unlike the dystrophin gene, the sarcoglycan genes are all small enough to be delivered using AAV (Table 1). Since AAV has proven capable of gene delivery to adult animals without elicitation of an immune response, it should be possible to inject older BIO 14.6 hamsters and maintain expression for a long period. For example, Greelish et al. (1999) injected 9-week-old hamsters and evaluated injected muscle 13 weeks after gene transfer, when stabilization of sarcolemmal integrity was demonstrated through exclusion of procion orange dye. Such studies offer hope that sarcoglycan deficiency could be corrected in adult, immune-competent humans. This hypothesis will be the focus of a number of limited human clinical trials over the next few years (Mendell and Wilson, personal communication). CONCLUSION There has been encouraging progress towards establishing gene replacement as a viable therapy for muscular dystrophy. Understanding the limitations of available gene transfer vectors led to the development of new options including gutted Ad, AAV, and lentiviral vectors. Gutted Ad vectors have allowed delivery of full-length dystrophin to adult muscle for the first time. Even in immune-competent animals, gene expression can be maintained for at least several months. In the future, these vectors may provide the capacity necessary for tightly regulated expression of therapeutic genes or delivery of multiple expression cassettes. Recent advances in the growth and purification of these gutted vectors should allow evaluation of their usefulness in disparate settings. AAV vectors have had encouraging success in gene delivery to the muscles of immunecompetent animals. Even proteins that are known to be very immunogenic, such as ␤-galactosidase, can be expressed for long periods. It is hoped that expression of small muscular dystrophy disease genes using AAV will be comparatively easy. Many challenges remain before gene replacement therapy for MD can be declared a success. Production of viral vectors at very large scale and high purity remains problematic, so further improvements will be required before delivery of full-length dystrophin to humans is practical. Long-term functional correction, as opposed to prevention, of the dystrophic phenotype in mouse models has not yet been demonstrated. The immunologic consequences of delivering disease genes to patients with null mutations are still unknown. Finally, safe, consistent delivery of viral vectors to a large mass of muscle tissue has not been shown to be practical. Stable gene transduction that provides functional correction at the level of single cells has been demonstrated. In the future, we hope that viral vectors such as gutted Ad, AAV, and others being developed will allow meaningful correction of muscular dystrophy at the level of whole muscles and human patients. Making this leap in scale is the challenge of the years to come. REFERENCES Acsadi G, Jani A, Huard J, Blaschuk K, Massie B, et al. 1994a. Cultured human myoblasts and myotubes show markedly different 235 transducibility by replication-defective adenovirus recombinants. Gene Ther 1:338–340. Acsadi G, Jani A, Massie B, Simoneau M, Holland P, et al. 1994b. A differential efficiency of adenovirus-mediated in vivo gene transfer into skeletal muscle cells of different maturity. Hum Mol Genet, 3:579–584. Amalfitano A, Hauser M A, Hu H, Serra D, Begy C R, Chamberlain J S. 1998. Production and characterization of improved adenovirus vectors with the E1 E2b, and E3 genes deleted. J Virol 72:926–933. Arahata K, Beggs A H, Honda H, Ito S, Ishiura S, et al. 1991. Preservation of the C-terminus of dystrophin molecule in the skeletal muscle from Becker muscular dystrophy. J Neurol Sci 101:148–156. Armentano D, Sookdeo C C, Hehir K M, Gregory R J, St. George J A, et al. 1995. Characterization of an adenovirus gene transfer vector containing an E4 deletion. Hum Gene Ther 6:1343–1353. Atchison R W, Casto B C, Hammond W M. 1965. AdenovirusAssociated Defective Virus Particles. Science 149:754–756. Barton-Davis E R, Shoturma D I, Musaro A, Rosenthal N, Sweeney H L. 1998. Viral mediated expression of insulin-like growth factor I blocks the aging-related loss of skeletal muscle function. Proc Natl Acad Sci USA 95:15603–15607. Bett A J, Prevec L, Graham F L. 1993. Packaging capacity and stability of human adenovirus type 5 vectors. J Virol 67:5911–5921. Bione S, Maestrini E, Rivella S, Mancini M, Regis S, et al. 1994. Identification of a novel X-linked gene responsible for EmeryDreifuss muscular dystrophy. Nature Genet 8:323–327. Blau H M, Webster C, Pavlath G K. 1983. Defective myoblasts identified in Duchenne muscular dystrophy. Proc Natl Acad Sci USA 80:4856–4860. Bonne G, Di Barletta M R, Varnous S, Becane H M, Hammouda E H, et al. 1999. Mutations in the gene encoding lamin A/C cause autosomal dominant Emery-Dreifuss muscular dystrophy. Nature Genet 21: 285–288. Bonnemann C G, Modi R, Noguchi S, Mizuno Y, Yoshida M, et al. 1995. Beta-sarcoglycan (A3b) mutations cause autosomal recessive muscular dystrophy with loss of the sarcoglycan complex [published erratum appears in Nat Genet 1996 Jan;12(1):110]. Nature Genet11: 266–273. Bonnemann C G, McNally E M, Kunkel L M. 1996. Beyond dystrophin: current progress in the muscular dystrophies [published erratum appears in Curr Opin Pediatr 1997 Apr;9(2):196]. Curr Opin Pediatr 8:569–582. Brandt C D, Kim H W, Vargosko A J, Jeffries B C, Arrobio J O, et al. 1969. Infections in 18,000 infants and children in a controlled study of respiratory tract disease. I. Adenovirus pathogenicity in relation to serologic type and illness syndrome. Am J Epidemiol 90:484–500. Bulman D E, Murphy E G, Zubrzycka-Gaarn E E, Worton R G, Ray P N. 1991. Differentiation of Duchenne and Becker muscular dystrophy phenotypes with amino- and carboxy-terminal antisera specific for dystrophin. Am J Hum Genet 48:295–304. Campbell K P, Kahl S D. 1989. Association of dystrophin and an integral membrane glycoprotein. Nature 338:259–262. Carroll R, Lin J T, Dacquel E J, Mosca J D, Burke D S, St Louis D C. 1994. A human immunodeficiency virus type 1 (HIV-1)-based retroviral vector system utilizing stable HIV-1 packaging cell lines. J Virol 68:6047–6051. Cartegni L, di Barletta M R, Barresi R, Squarzoni S, Sabatelli P, et al. 1997. Heart-specific localization of emerin: new insights into EmeryDreifuss muscular dystrophy. Hum Mol Genet 6:2257–2264. Chartier C, Degryse E, Gantzer M, Dieterle A, Pavirani A, Mehtali M. 1996. Efficient generation of recombinant adenovirus vectors by homologous recombination in Escherichia coli. J Virol 70:4805– 4810. Coral-Vazquez R, Cohn R D, Moore S A, Hill J A, Weiss R M, et al. 1999. Disruption of the sarcoglycan-sarcospan complex in vascular smooth muscle: A novel mechanism for cardiomyopathy and muscular dystrophy. Cell 98:465–474. Corrado K, Rafael J A, Mills P L, Cole N M, Faulkner J A, et al. 1996. Transgenic mdx mice expressing dystrophin with a deletion in the actin-binding domain display a ‘‘mild Becker’’ phenotype. J Cell Biol 134:873–884. Cox G A, Cole N M, Matsumura K, Phelps S F, Hauschka S D, et al. 1993. Overexpression of dystrophin in transgenic mdx mice eliminates dystrophic symptoms without toxicity [see comments]. Nature 364:725–729. Cox G A, Sunada Y, Campbell K P, Chamberlain J S. 1994. Dp71 can restore the dystrophin-associated glycoprotein complex in muscle but fails to prevent dystrophy [see comments]. Nature Genet 8:333–339. 236 D. HARTIGAN-O’CONNOR AND J.S. CHAMBERLAIN Culver K W, erson W F, Blaese R M. 1991. Lymphocyte gene therapy. Hum Gene Ther 2:107–109. Dai Y, Schwarz E M, Gu D, Zhang W W, Sarvetnick N, Verma I M. 1995. Cellular and humoral immune responses to adenoviral vectors containing factor IX gene: tolerization of factor IX and vector antigens allows for long-term expression. Proc Natl Acad Sci USA 92:1401–1405. Deconinck N, Ragot T, Marechal G, Perricaudet M, Gillis J M. 1996. Functional protection of dystrophic mouse (mdx) muscles after adenovirus-mediated transfer of a dystrophin minigene. Proc Natl Acad Sci USA 93:3570–3574. Deconinck N, Tinsley J, De Backer F, Fisher R, Kahn D, et al. 1997. Expression of truncated utrophin leads to major functional improvements in dystrophin-deficient muscles of mice. Nature Med 3:1216– 1221. Dong J Y, Wang D, Van Ginkel F W, Pascual D W, Frizzell R A. 1996. Systematic analysis of repeated gene delivery into animal lungs with a recombinant adenovirus vector. Hum Gene Ther 7:319–331. Douglas J T, Rogers B E, Rosenfeld M E, Michael S I, Feng M, Curiel D T. 1996. Targeted gene delivery by tropism-modified adenoviral vectors. Nature Biotechnol 14:1574–1578. Duan D, Sharma P, Yang J, Yue Y, Dudus L, et al. 1998. Circular intermediates of recombinant adeno-associated virus have defined structural characteristics responsible for long-term episomal persistence in muscle tissue [published erratum appears in J Virol 1999 Jan;73(1):861]. J Virol 72:8568–8577. Dunckley M G, Love D R, Davies K E, Walsh F S, Morris G E, Dickson G 1992. Retroviral-mediated transfer of a dystrophin minigene into mdx mouse myoblasts in vitro. FEBS Lett 296:128–134. Dunckley M G, Wells D J, Walsh F S, Dickson G 1993. Direct retroviral-mediated transfer of a dystrophin minigene into mdx mouse muscle in vivo. Hum Mol Genet 2:717–723. Emery A E. 1987. X-linked muscular dystrophy with early contractures and cardiomyopathy (Emery-Dreifuss type). Clin Genet 32:360– 367. Emery AEH. 1993. Duchenne Muscular Dystrophy. 2nd ed. New York: Oxford University Press. 392 pp. Engel A G, Franzini-Armstrong C, eds. 1994. Myology: basic and clinical. 2nd ed. New York: McGraw-Hill England S B, Nicholson L V, Johnson M A, Forrest S M, Love D R, et al. 1990. Very mild muscular dystrophy associated with the deletion of 46% of dystrophin. Nature 343:180–182. Ervasti J M, Ohlendieck K, Kahl S D, Gaver M G, Campbell K P. 1990. Deficiency of a glycoprotein component of the dystrophin complex in dystrophic muscle. Nature 345:315–319. Ettinger A J, Feng G, Sanes J R. 1997. epsilon-Sarcoglycan, a broadly expressed homologue of the gene mutated in limb-girdle muscular dystrophy 2D [published erratum appears in J Biol Chem 1998 Jul 31;273(31):19922]. J Biol Chem 272:32534–32538. Fassati A, Wells D J, Sgro Serpente P A, Walsh F S, Brown S C, et al. 1997. Genetic correction of dystrophin deficiency and skeletal muscle remodeling in adult MDX mouse via transplantation of retroviral producer cells. J Clin Invest 100:620–628. Ferrari G, Cusella-De Angelis G, Coletta M, Paolucci E, Stornaiuolo A, et al. 1998. Muscle regeneration by bone marrow-derived myogenic progenitors [see comments] [published erratum appears in Science 1998 Aug 14;281(5379):923]. Science 279:1528–1530. Fisher K J, Choi H, Burda J, Chen S J, Wilson J M. 1996. Recombinant adenovirus deleted of all viral genes for gene therapy of cystic fibrosis. Virology 217:11–22. Freese E 1972. Prospects of gene therapy. Science, 175:1024–1025. Gao G P, Yang Y, Wilson J M. 1996. Biology of adenovirus vectors with E1 and E4 deletions for liver-directed gene therapy. J Virol 70:8934– 8943. Gaynor R B, Berk A J. 1983. Cis-acting induction of adenovirus transcription. Cell 33:683–693. Gorziglia M I, Kadan M J, Yei S, Lim J, Lee G M, et al. 1996. Elimination of both E1 and E2 from adenovirus vectors further improves prospects for in vivo human gene therapy. J Virol 70:4173– 4178. Graham F L, Prevec L 1991. Manipulation of Adenovirus Vectors. In: Murray EJ, editor. Gene transfer and expression protocols. Methods in molecular biology, Vol. 7. Clifton, NJ.: The Humana Press Inc. p 109–128. Gramolini A O, Angus L M, Schaeffer L, Burton E A, Tinsley J M, et al. 1999. Induction of utrophin gene expression by heregulin in skeletal muscle cells: role of the N-box motif and GA binding protein. Proc Natl Acad Sci USA 96:3223–3227. Greelish J P, Su L T, Lankford E B, Burkman J M, Chen H, et al. 1999. Stable restoration of the sarcoglycan complex in dystrophic muscle perfused with histamine and a recombinant adeno-associated viral vector. Nature Med 5:439–443. Gussoni E, Wang Y, Fraefel C, Miller R G, Blau H M, et al. 1996. A method to codetect introduced genes and their products in gene therapy protocols. Nature Biotechnol 14:1012–1016. Gussoni E, Soneoka Y, Strickland C D, Buzney E A, Khan M K, et al. 1999. Dystrophin expression in the mdx mouse restored by stem cell transplantation. Nature 401:390–394. Hardy S, Kitamura M, Harris-Stansil T, Dai Y, Phipps M L. 1997. Construction of adenovirus vectors through Cre-lox recombination. J Virol 71:1842–1849. Hartigan-O’Connor D, Amalfitano A, Chamberlain J S. 1999. Improved production of gutted adenovirus in cells expressing adenovirus preterminal protein and DNA polymerase. J Virol 73:7835– 7841. Helbling-Leclerc A, Zhang X, Topaloglu H, Cruaud C, Tesson F, et al. 1995. Mutations in the laminin alpha 2-chain gene (LAMA2. cause merosin-deficient congenital muscular dystrophy. Nature Genet 11:216–218. Helliwell T R, Man N T, Morris G E, Davies K E. 1992. The dystrophin-related protein, utrophin, is expressed on the sarcolemma of regenerating human skeletal muscle fibres in dystrophies and inflammatory myopathies. Neuromusc Disord 2:177–184. Helliwell T R, Nguyen T M, Morris G E. 1994. Expression of the 43 kDa dystrophin-associated glycoprotein in human neuromuscular disease. Neuromusc Disord 4:101–113. Hermonat P L, Muzyczka N 1984. Use of adeno-associated virus as a mammalian DNA cloning vector: transduction of neomycin resistance into mammalian tissue culture cells. Proc Natl Acad Sci USA 81:6466–6470. Hoggan M D, Blacklow N R, Rowe W P. 1966. Studies of small DNA viruses found in various adenovirus preparations: physical, biological, and immunological characteristics. Proc Natl Acad Sci USA 55:1457–1471. Holt K H, Lim L E, Straub V, Venzke D P, Duclos F, et al. 1998. Functional rescue of the sarcoglycan complex in the BIO 14.6 hamster using delta-sarcoglycan gene transfer. Mol Cell 1:841–848. Hu H, Serra D, Amalfitano A 1999. Persistence of an [E1-, polymerase-] adenovirus vector despite transduction of a neoantigen into immune-competent mice. Hum Gene Ther 10:355–364. Huard J, Goins W F, Glorioso J C. 1995. Herpes simplex virus type 1 vector mediated gene transfer to muscle. Gene Ther 2:385–392. Jejurikar S S, Welling T H, Zelenock J A, Gordon D, Burkel W E, et al. 1997. Induction of angiogenesis by lidocaine and basic fibroblast growth factor: a model for in vivo retroviral-mediated gene therapy. J Surg Res 67:137–146. Jooss K, Ertl H C, Wilson J M. 1998a. Cytotoxic T-lymphocyte target proteins and their major histocompatibility complex class I restriction in response to adenovirus vectors delivered to mouse liver. J Virol 72:2945–2954. Jooss K, Yang Y, Fisher K J, Wilson J M. 1998b. Transduction of dendritic cells by DNA viral vectors directs the immune response to transgene products in muscle fibers. J Virol 72:4212–4223. Jung D, Duclos F, Apostol B, Straub V, Lee J C, et al. 1996. Characterization of delta-sarcoglycan, a novel component of the oligomeric sarcoglycan complex involved in limb-girdle muscular dystrophy. J Biol Chem 271:32321–32329. Kafri T, Blomer U, Peterson D A, Gage F H, Verma I M. 1997. Sustained expression of genes delivered directly into liver and muscle by lentiviral vectors. Nature Genet 17:314–317. Karpati G, Ajdukovic D, Arnold D, Gledhill R B, Guttmann R, et al. 1993. Myoblast transfer in Duchenne muscular dystrophy [see comments]. Ann Neurol 34:8–17. Kessler P D, Podsakoff G M, Chen X, McQuiston S A, Colosi P C, et al. 1996. Gene delivery to skeletal muscle results in sustained expression and systemic delivery of a therapeutic protein. Proc Natl Acad Sci USA 93:14082–14087. Khurana T S, Rosmarin A G, Shang J, Krag T O, Das S, Gammeltoft S. 1999. Activation of utrophin promoter by heregulin via the etsrelated transcription factor complex GA-binding protein alpha/beta. Mol Biol Cell 10:2075–2086. Kobayashi K, Nakahori Y, Miyake M, Matsumura K, Kondo-Iida E, et al. 1998. An ancient retrotransposal insertion causes Fukuyamatype congenital muscular dystrophy. Nature 394:388–392. Kochanek S, Clemens P R, Mitani K, Chen H H, Chan S, Caskey C T. 1996. A new adenoviral vector: Replacement of all viral coding sequences with 28 kb of DNA independently expressing both full-length dystrophin and beta-galactosidase. Proc Natl Acad Sci USA 93:5731–5736. Koenig M, Hoffman E P, Bertelson C J, Monaco A P, Feener C, Kunkel L M. 1987. Complete cloning of the Duchenne muscular dystrophy DEVELOPMENTS IN GENE THERAPY FOR MD (DMD) cDNA and preliminary genomic organization of the DMD gene in normal and affected individuals. Cell 50:509–517. Kotin R M, Siniscalco M, Samulski R J, Zhu X D, Hunter L, et al. 1990. Site-specific integration by adeno-associated virus. Proc Natl Acad Sci USA 87:2211–2215. Kuang W, Xu H, Vachon P H, Liu L, Loechel F, et al. 1998. Merosindeficient congenital muscular dystrophy. Partial genetic correction in two mouse models [published erratum appears in J Clin Invest 1998 Sep 15;102(6):following 1275]. J Clin Invest 102:844–852. Kumar-Singh R, Chamberlain J S. 1996. Encapsidated adenovirus minichromosomes allow delivery and expression of a 14 kb dystrophin cDNA to muscle cells. Hum Mol Genet 5:913–921. Lattanzi L, Salvatori G, Coletta M, Sonnino C, Cusella De Angelis M G, et al. 1998. High efficiency myogenic conversion of human fibroblasts by adenoviral vector-mediated MyoD gene transfer. An alternative strategy for ex vivo gene therapy of primary myopathies. J Clin Invest 101:2119–2128. Leivo I, Engvall E 1988. Merosin, a protein specific for basement membranes of Schwann cells, striated muscle, and trophoblast, is expressed late in nerve and muscle development. Proc Natl Acad Sci USA 85:1544–1548. Lewis P, Hensel M, Emerman M 1992. Human immunodeficiency virus infection of cells arrested in the cell cycle [published erratum appears in EMBO J 1992 Nov;11(11):4249]. EMBO J 11:3053–3058. Lim L E, Campbell K P. 1998. The sarcoglycan complex in limb-girdle muscular dystrophy. Curr Opin Neurol 11:443–452. Lim L E, Duclos F, Broux O, Bourg N, Sunada Y, et al. 1995. Beta-sarcoglycan: characterization and role in limb-girdle muscular dystrophy linked to 4q12. Nature Genet 11:257–265. Ljunggren A, Duggan D, McNally E, Boylan K B, Gama C H, et al. 1995. Primary adhalin deficiency as a cause of muscular dystrophy in patients with normal dystrophin [see comments]. Ann Neurol 38:367–372. Lochmuller H, Petrof B J, Pari G, Larochelle N, Dodelet V, et al. 1996. Transient immunosuppression by FK506 permits a sustained highlevel dystrophin expression after adenovirus-mediated dystrophin minigene transfer to skeletal muscles of adult dystrophic (mdx) mice. Gene Ther 3:706–716. Love D R, Flint T J, Marsden R F, Bloomfield J F, Daniels R J, et al. 1990. Characterization of deletions in the dystrophin gene giving mild phenotypes. Am J Med Genet 37:136–142. Manilal S, Nguyen T M, Sewry C A, Morris G E. 1996. The EmeryDreifuss muscular dystrophy protein, emerin, is a nuclear membrane protein. Hum Mol Genet 5:801–808. Marconi P, Simonato M, Zucchini S, Bregola G, Argnani R, et al. 1999. Replication-defective herpes simplex virus vectors for neurotrophic factor gene transfer in vitro and in vivo. Gene Ther 6:904–912. Matsushita T, Elliger S, Elliger C, Podsakoff G, Villarreal L, et al. 1998. Adeno-associated virus vectors can be efficiently produced without helper virus. Gene Ther 5:938–945. McArdle A, Edwards R H, Jackson M J. 1995. How does dystrophin deficiency lead to muscle degeneration? Evidence from the mdx mouse. Neuromusc Disord 5:445–456. McNally E M, Yoshida M, Mizuno Y, Ozawa E, Kunkel L M. 1994. Human adhalin is alternatively spliced and the gene is located on chromosome 17q21. Proc Natl Acad Sci USA 91:9690–9694. Megeney L A, Kablar B, Garrett K, erson J E, Rudnicki M A. 1996. MyoD is required for myogenic stem cell function in adult skeletal muscle. Genes Dev 10:1173–1183. Melnick J L, Mayor H D, Smith K O, Rapp F 1965. Association of 20 millimicron particles with adenoviruses. J Bacteriol 90:271–274. Mendell J R, Kissel J T, Amato A A, King W, Signore L, et al. 1995. Myoblast transfer in the treatment of Duchenne’s muscular dystrophy [see comments]. N Engl J Med 333:832–838. Minetti C, Sotgia F, Bruno C, Scartezzini P, Broda P, et al. 1998. Mutations in the caveolin-3 gene cause autosomal dominant limbgirdle muscular dystrophy. Nature Genet 18:365–368. Monaco A P, Bertelson C J, Liechti-Gallati S, Moser H, Kunkel L M. 1988. An explanation for the phenotypic differences between patients bearing partial deletions of the DMD locus. Genomics 2: 90–95. Morgan J E, Pagel C N, Sherratt T, Partridge T A. 1993. Long-term persistence and migration of myogenic cells injected into preirradiated muscles of mdx mice. J Neurol Sci 115:191–200. Muzyczka N 1992. Use of adeno-associated virus as a general transduction vector for mammalian cells. Curr Topics Microbiol Immunol 158:97–129. Nagano A, Koga R, Ogawa M, Kurano Y, Kawada J, et al. 1996. Emerin deficiency at the nuclear membrane in patients with EmeryDreifuss muscular dystrophy. Nature Genet 12:254–259. Nalbantoglu J, Pari G, Karpati G, Holland P C. 1999. Expression of 237 the primary coxsackie and adenovirus receptor is downregulated during skeletal muscle maturation and limits the efficacy of adenovirus-mediated gene delivery to muscle cells. Hum Gene Ther 10:1009– 1019. Naldini L, Blomer U, Gallay P, Ory D, Mulligan R, et al. 1996. In vivo gene delivery and stable transduction of nondividing cells by a lentiviral vector [see comments]. Science,272:263–267. Nevins J R. 1981. Mechanism of activation of early viral transcription by the adenovirus E1A gene product. Cell 26:213–220. Nigro V, Piluso G, Belsito A, Politano L, Puca A A, et al. 1996. Identification of a novel sarcoglycan gene at 5q33 encoding a sarcolemmal 35 kDa glycoprotein. Hum Mol Genet 5:1179–1186. Nigro V, Okazaki Y, Belsito A, Piluso G, Matsuda Y, et al. 1997. Identification of the Syrian hamster cardiomyopathy gene. Hum Mol Genet 6:601–607. Noguchi S, McNally E M, Ben Othmane K, Hagiwara Y, Mizuno Y, et al. 1995. Mutations in the dystrophin-associated protein gammasarcoglycan in chromosome 13 muscular dystrophy [see comments]. Science 270:819–822. Ohtsuka Y, Udaka K, Yamashiro Y, Yagita H, Okumura K 1998. Dystrophin acts as a transplantation rejection antigen in dystrophindeficient mice: implication for gene therapy. J Immunol 160:4635– 4640. O’Neal W K, Zhou H, Morral N, Aguilar-Cordova E, Pestaner J, et al. 1998. Toxicological comparison of E2a-deleted and first-generation adenoviral vectors expressing alpha1-antitrypsin after systemic delivery. Hum Gene Ther 9:1587–1598. Ozawa E, Yoshida M, Suzuki A, Mizuno Y, Hagiwara Y, Noguchi S 1995. Dystrophin-associated proteins in muscular dystrophy. Hum Mol Genet 4 Spec No:1711–1716. Parks R J, Chen L, Anton M, Sankar U, Rudnicki M A, Graham F L. 1996. A helper-dependent adenovirus vector system: removal of helper virus by Cre-mediated excision of the viral packaging signal. Proc Natl Acad Sci USA 93:13565–13570. Phelps S F, Hauser M A, Cole N M, Rafael J A, Hinkle R T, et al. 1995. Expression of full-length and truncated dystrophin mini-genes in transgenic mdx mice. Hum Mol Genet 4:1251–1258. Poeschla E M, Wong-Staal F, Looney D J. 1998. Efficient transduction of nondividing human cells by feline immunodeficiency virus lentiviral vectors. Nature Med 4:354–357. Rafael J A, Sunada Y, Cole N M, Campbell K P, Faulkner J A, Chamberlain J S. 1994. Prevention of dystrophic pathology in mdx mice by a truncated dystrophin isoform. Hum Mol Genet, 3:1725– 1733. Ragot T, Stratford-Perricaudet L D, Vincent N, Chafey P, Vigne E, et al. 1994. Adenovirus-mediated transfer of a human dystrophin gene to skeletal muscle of mdx mouse. Gene Ther 1(Suppl 1):S53–54. Richard I, Broux O, Allamand V, Fougerousse F, Chiannilkulchai N, et al. 1995. Mutations in the proteolytic enzyme calpain 3 cause limb-girdle muscular dystrophy type 2A. Cell 81:27–40. Roberds S L, erson R D, Ibraghimov-Beskrovnaya O, Campbell K P. 1993. Primary structure and muscle-specific expression of the 50-kDa dystrophin-associated glycoprotein (adhalin). J Biol Chem 268:23739–23742. Roberds S L, Leturcq F, Allamand V, Piccolo F, Jeanpierre M, et al. 1994. Missense mutations in the adhalin gene linked to autosomal recessive muscular dystrophy. Cell 78:625–633. Rosenberg S A, Aebersold P, Cornetta K, Kasid A, Morgan R A, et al. 1990. Gene transfer into humans—immunotherapy of patients with advanced melanoma, using tumor-infiltrating lymphocytes modified by retroviral gene transduction [see comments]. N Engl J Med 323:570–578. Sakamoto A, Ono K, Abe M, Jasmin G, Eki T, et al. 1997. Both hypertrophic and dilated cardiomyopathies are caused by mutation of the same gene, delta-sarcoglycan, in hamster: an animal model of disrupted dystrophin-associated glycoprotein complex. Proc Natl Acad Sci USA 94:13873–13878. Salvetti A, Oreve S, Chadeuf G, Favre D, Cherel Y, et al. 1998. Factors influencing recombinant adeno-associated virus production. Hum Gene Ther 9:695–706. Sanes J R, Engvall E, Butkowski R, Hunter D D. 1990. Molecular heterogeneity of basal laminae: isoforms of laminin and collagen IV at the neuromuscular junction and elsewhere. J Cell Biol 111:1685– 1699. Schaack J, Guo X, Ho W Y, Karlok M, Chen C, Ornelles D 1995. Adenovirus type 5 precursor terminal protein-expressing 293 and HeLa cell lines. J Virol 69:4079–4085. Schiedner G, Morral N, Parks R J, Wu Y, Koopmans S C, et al. 1998. Genomic DNA transfer with a high-capacity adenovirus vector results in improved in vivo gene expression and decreased toxicity 238 D. HARTIGAN-O’CONNOR AND J.S. CHAMBERLAIN [published erratum appears in Nat Genet 1998 Mar;18(3):298]. Nature Genet 18:180–183. Schlehofer J R, Ehrbar M, zur Hausen H 1986. Vaccinia virus, herpes simplex virus, and carcinogens induce DNA amplification in a human cell line and support replication of a helpervirus dependent parvovirus. Virology 152:110–117. Schmitz H, Wigand R, Heinrich W 1983. Worldwide epidemiology of human adenovirus infections. Am J Epidemiol 117:455–466. Snyder R O, Spratt S K, Lagarde C, Bohl D, Kaspar B, et al. 1997. Efficient and stable adeno-associated virus-mediated transduction in the skeletal muscle of adult immunocompetent mice. Hum Gene Ther 8:1891–1900. Solnick D 1981. Construction of an adenovirus-SV40 recombinant producing SV40 T antigen from an adenovirus late promoter. Cell 24:135–143. Stein C S, Ghodsi A, Derksen T, Davidson B L. 1999. Systemic and central nervous system correction of lysosomal storage in mucopolysaccharidosis type VII mice. J Virol 73:3424–3429. Straub V, Campbell K P. 1997. Muscular dystrophies and the dystrophin-glycoprotein complex. Curr Opin Neurol 10:168–175. Straub V, Ettinger A J, Durbeej M, Venzke D P, Cutshall S, et al. 1999. epsilon-sarcoglycan replaces alpha-sarcoglycan in smooth muscle to form a unique dystrophin-glycoprotein complex. J Biol Chem 274: 27989–27996. Thomas G P, Mathews M B. 1980. DNA replication and the early to late transition in adenovirus infection. Cell 22:523–533. Thummel C, Tjian R, Grodzicker T 1981. Expression of SV40 T antigen under control of adenovirus promoters. Cell 23:825–836. Tidball J G, Albrecht D E, Lokensgard B E, Spencer M J. 1995. Apoptosis precedes necrosis of dystrophin-deficient muscle. J Cell Sci 108:2197–2204. Tinsley J, Deconinck N, Fisher R, Kahn D, Phelps S, et al. 1998. Expression of full-length utrophin prevents muscular dystrophy in mdx mice. Nature Med 4:1441–1444. Tinsley J M, Blake D J, Roche A, Fairbrother U, Riss J, et al. 1992. Primary structure of dystrophin-related protein. Nature 360:591– 593. Tinsley J M, Potter A C, Phelps S R, Fisher R, Trickett J I, Davies K E. 1996. Amelioration of the dystrophic phenotype of mdx mice using a truncated utrophin transgene [see comments]. Nature 384:349–353. Tripathy S K, Black H B, Goldwasser E, Leiden J M. 1996. Immune responses to transgene-encoded proteins limit the stability of gene expression after injection of replication-defective adenovirus vectors. Nature Med 2:545–550. Van Ginkel F W, Liu C, Simecka J W, Dong J Y, Greenway T, et al. 1995. Intratracheal gene delivery with adenoviral vector induces elevated systemic IgG and mucosal IgA antibodies to adenovirus and beta-galactosidase. Hum Gene Ther 6:895–903. Vilquin J T, Kinoshita I, Roy B, Goulet M, Engvall E, et al. 1996. Partial laminin alpha2 chain restoration in alpha2 chain-deficient dy/dy mouse by primary muscle cell culture transplantation. J Cell Biol 133:185–197. Vincent N, Ragot T, Gilgenkrantz H, Couton D, Chafey P, et al. 1993. Long-term correction of mouse dystrophic degeneration by adenovirus-mediated transfer of a minidystrophin gene. Nature Genet 5:130–134. Wang Q, Jia X C, Finer M H. 1995. A packaging cell line for propagation of recombinant adenovirus vectors containing two lethal gene-region deletions. Gene Ther 2:775–783. Wang Q, Greenburg G, Bunch D, Farson D, Finer M H. 1997. Persistent transgene expression in mouse liver following in vivo gene transfer with a delta E1/delta E4 adenovirus vector. Gene Ther 4:393–400. Weinberg D H, Ketner G 1983. A cell line that supports the growth of a defective early region 4 deletion mutant of human adenovirus type 2. Proc Natl Acad Sci USA 80:5383–5386. Weinberg J B, Matthews T J, Cullen B R, Malim M H. 1991. Productive human immunodeficiency virus type 1 (HIV-1. infection of nonproliferating human monocytes. J Exp Med 174:1477–1482. Welling T H, Davidson B L, Zelenock J A, Stanley J C, Gordon D, et al. 1996. Systemic delivery of the interleukin-1 receptor antagonist protein using a new strategy of direct adenoviral-mediated gene transfer to skeletal muscle capillary endothelium in the isolated rat hindlimb. Hum Gene Ther 7:1795–1802. Wells D J, Wells K E, Asante E A, Turner G, Sunada Y, et al. 1995. Expression of human full-length and minidystrophin in transgenic mdx mice: implications for gene therapy of Duchenne muscular dystrophy. Hum Mol Genet 4:1245–1250. Wickham T J, Tzeng E, Shears L L, 2nd Roelvink P W, Li Y, et al. 1997. Increased in vitro and in vivo gene transfer by adenovirus vectors containing chimeric fiber proteins. J Virol 71:8221–8229. Xiao X, Li J, Samulski R J. 1996. Efficient long-term gene transfer into muscle tissue of immunocompetent mice by adeno-associated virus vector. J Virol 70:8098–8108. Xiao X, Li J, Samulski R J. 1998. Production of high-titer recombinant adeno-associated virus vectors in the absence of helper adenovirus. J Virol 72:2224–2232. Xu H, Christmas P, Wu X R, Wewer U M, Engvall E 1994. Defective muscle basement membrane and lack of M-laminin in the dystrophic dy/dy mouse. Proc Natl Acad Sci USA 91:5572–5576. Yakinoglu A O, Heilbronn R, Burkle A, Schlehofer J R, zur Hausen H 1988. DNA amplification of adeno-associated virus as a response to cellular genotoxic stress. Cancer Res 48:3123–3129. Yakobson B, Koch T, Winocour E 1987. Replication of adeno-associated virus in synchronized cells without the addition of a helper virus. J Virol 61:972–981. Yakobson B, Hrynko T A, Peak M J, Winocour E 1989. Replication of adeno-associated virus in cells irradiated with UV light at 254 nm. J Virol 63:1023–1030. Yang L, Lochmuller H, Luo J, Massie B, Nalbantoglu J, et al. 1998. Adenovirus-mediated dystrophin minigene transfer improves muscle strength in adult dystrophic (MDX) mice. Gene Ther 5:369–379. Yang Y, Wilson J M. 1995. Clearance of adenovirus-infected hepatocytes by MHC class I-restricted CD4⫹ CTLs in vivo. J Immunol 155:2564–2570. Yang Y, Nunes F A, Berencsi K, Furth E E, Gonczol E, Wilson J M. 1994. Cellular immunity to viral antigens limits E1-deleted adenoviruses for gene therapy. Proc Natl Acad Sci USA 91:4407–4411. Yang Y, Haecker S E, Su Q, Wilson J M. 1996a. Immunology of gene therapy with adenoviral vectors in mouse skeletal muscle. Hum Mol Genet 5:1703–1712. Yang Y, Su Q, Wilson J M. 1996b. Role of viral antigens in destructive cellular immune responses to adenovirus vector-transduced cells in mouse lungs. J Virol 70:7209–7212. Yoshida M, Ozawa E. 1990. Glycoprotein complex anchoring dystrophin to sarcolemma. J Biochem 108:748–752. Yuasa K, Miyagoe Y, Yamamoto K, Nabeshima Y, Dickson G, Takeda S 1998. Effective restoration of dystrophin-associated proteins in vivo by adenovirus-mediated transfer of truncated dystrophin cDNAs. FEBS Lett 425:329–336. Zhou H, O’Neal W, Morral N, Beaudet A L. 1996. Development of a complementing cell line and a system for construction of adenovirus vectors with E1 and E2a deleted. J Virol 70:7030–7038. Zsengeller Z K, Wert S E, Hull W M, Hu X, Yei S, et al. 1995. Persistence of replication-deficient adenovirus-mediated gene transfer in lungs of immune-deficient (nu/nu) mice. Hum Gene Ther 6:457–467.