Adeno-associated virus: integration at a specific chromosomal locus

Targeted adeno-associated virus vector transduction of nonpermissive cells mediated by a bispecific F(ab'gamma)2 antibody

We have developed a system for the targeted delivery of adeno-associated virus (AAV) vectors. Targeting is achieved via a bispecific F(ab')2 antibody that mediates a novel interaction between the AAV vector and a specific cell surface receptor expressed on human megakaryocytes. Targeted AAV vectors were able to transduce megakaryocyte cell lines, DAMI and MO7e, which were nonpermissive for normal AAV infection, 70-fold above background and at levels equivalent to permissive K562 cells. Transduction was shown to occur through the specific interaction of the AAV vector-bispecific F(ab')2 complex and cell-associated targeting receptor. Importantly, targeting appeared both selective and restrictive as the endogenous tropism of the AAV vector was significantly reduced. Binding and internalization through the alternative receptor did not alter subsequent steps (escape from endosomes, migration to nucleus, or uncoating) required to successfully transduce target cells. These results demonstrate that AAV vectors can be targeted to a specific cell population and that transduction can be achieved by circumventing the normal virus receptor.

Human adeno-associated virus type 5 is only distantly related to other known primate helper-dependent parvoviruses

We have characterized 95% (4,404 nucleotides) of the genome of adeno-associated virus type 5 (AAV5), including part of the terminal repeats and the terminal resolution site. Our results show that AAV5 is different from all other described AAV serotypes at the nucleotide level and at the amino acid level. The sequence homology to AAV2, AAV3B, AAV4, and AAV6 at the nucleotide level is only between 54 and 56%. The positive strand contains two large open reading frames (ORFs). The left ORF encodes the nonstructural (Rep) proteins, and the right ORF encodes the structural (Cap) proteins. At the amino acid level the identities with the capsid proteins of other AAVs range between 51 and 59%, with a high degree of heterogeneity in regions which are considered to be on the exterior surface of the viral capsid. The overall identity for the nonstructural Rep proteins at the amino acid level is 54.4%. It is lowest at the C-terminal 128 amino acids (10%). There are only two instead of the common three putative Zn fingers in the Rep proteins. The Cap protein data suggest differences in capsid surfaces and raise the possibility of a host range distinct from those of other parvoviruses. This may have important implications for AAV vectors used in gene therapy.

The adeno-associated virus type 2 regulatory proteins rep78 and rep68 interact with the transcriptional coactivator PC4

The adeno-associated virus type 2 (AAV-2) Rep78/Rep68 regulatory proteins are pleiotropic effectors of viral and cellular DNA replication, of cellular transformation by viral and cellular oncogenes, and of homologous and heterologous gene expression. To search for cellular proteins involved in mediating these functions, we used Rep68 as bait in the yeast two-hybrid system and identified the transcriptional coactivator PC4 as a Rep interaction partner. PC4 has been shown to mediate transcriptional activation by a variety of sequence-specific transcription factors in vitro. Rep amino acids 172 to 530 were sufficient and amino acids 172 to 224 were absolutely necessary for the interaction with PC4. The PC4 domains required for interaction were mapped to the C-terminal single-stranded DNA-binding domain of PC4. In glutathione S-transferase (GST) pull-down assays, in vitro-transcribed and -translated Rep78 or Rep68 proteins were bound specifically by GST-PC4 fusion proteins. Similarly, PC4 expressed in Escherichia coli was bound by GST-Rep fusion proteins, confirming the direct interaction between Rep and PC4 in vitro. Rep was found to have a higher affinity for the nonphosphorylated, transcriptionally active form of PC4 than for the phosphorylated, transcriptionally inactive form. The latter is predominant in nuclear extracts of HeLa or 293 cells. In the yeast system, but not in vitro, Rep-PC4 interaction was disrupted by a point mutation in the putative nucleotide-binding site of Rep68, suggesting that a stable interaction between Rep and PC4 in vivo is ATP dependent. This mutation has also been shown to impair Rep function in AAV-2 DNA replication and in inhibition of gene expression and inducible DNA amplification. Cytomegalovirus promoter-driven overexpression of PC4 led to transient accumulation of nonphosphorylated PC4 with concomitant downregulation of all three AAV-2 promoters in the absence of helper virus. In the presence of adenovirus, this effect was relieved. These results imply an involvement of the transcriptional coactivator PC4 in the regulation of AAV-2 gene expression in the absence of helper virus.

Redox-mediated gene therapies for environmental injury: approaches and concepts

Cellular redox state has been increasingly recognized as a critical component of stress-induced cellular responses and disease. Inherent in these responses are reactive oxygen species (ROS), which inflict direct cellular damage in addition to acting as intracellular second messengers modulating signal transduction pathways. These intracellular highways of communication are critical in determining cell fates and whole-organ responses following environmental injury. Although gene therapy for inherited and acquired disorders has exploded in the last decade, the application of gene therapeutic approaches for transient pathologic conditions resulting from environmental stress is just beginning to be recognized. This review will summarize the theoretical and practical applications of gene therapy for the treatment of environmental injury by modulating redox-activated cellular responses. Several approaches can be utilized to achieve this goal. These include the application of gene targeting to modulate the cellular redox state directly by expressing recombinant genes capable of degrading ROS at pathophysiologic important subcellular sites. The use of mitochondrial superoxide dismutase (MnSOD), which degrades superoxides arising from ischemia/reperfusion injury, is one example of this approach. MnSOD serves as a "garbage disposal" for potentially toxic ROS prior to cellular injury and the activation of signal transduction cascades important in whole-organ pathology and inflammation. In contrast, some ROS have been suggested to have beneficial effects on cellular responses following environmental injury. Hence, expressing the nitrogen oxygen synthetase gene (NOS) to enhance the levels of nitric oxide (NO.) and augment the beneficial effects of this compound has also been suggested as a useful redox-modulating gene therapy approach. Lastly, indirect intervention in signal transduction pathways following environmental stress by expressing dominant inhibitory proteins of redox-activated signal transduction cascades has also been useful in modulating cellular responses to redox stress. Two such examples have utilized dominant inhibitory forms of the retinoblastoma gene product (Rb) and IkappaBalpha which prevent activation of cyclin-dependent protein kinases and NF-kappaB, respectively. Ultimately, the most efficacious therapeutic approach or combination of approaches that alter the redox responsiveness of cells and organs to environmental injury will be determined through a comprehensive understanding of the relevant pathophysiologic processes.

Formation of adeno-associated virus circular genomes is differentially regulated by adenovirus E4 ORF6 and E2a gene expression

A central feature of the adeno-associated virus (AAV) latent life cycle is persistence in the form of both integrated and episomal genomes. However, the molecular processes associated with episomal long-term persistence of AAV genomes are only poorly understood. To investigate these mechanisms, we have utilized a recombinant AAV (rAAV) shuttle vector to identify circular AAV intermediates from transduced HeLa cells and primary fibroblasts. The unique structural features exhibited by these transduction intermediates included circularized monomer and dimer virus genomes in a head-to-tail array, with associated specific base pair alterations in the 5' viral D sequence. In HeLa cells, the abundance and stability of AAV circular intermediates were augmented by adenovirus expressing the E2a gene product. In the absence of E2a, adenovirus expressing the E4 open reading frame 6 gene product decreased the abundance of AAV circular intermediates, favoring instead the linear replication form monomer (Rfm) and dimer (Rfd) structures. In summary, the formation of AAV circular intermediates appears to represent a new pathway for AAV genome conversion, which is consistent with the head-to-tail concatemerization associated with latent-phase persistence of rAAV. A better understanding of this pathway may increase the utility of rAAV vectors for gene therapy.

Targeted integration of adeno-associated virus-derived plasmids in transfected human cells

Adeno-associated virus (AAV) integrates its genomic DNA into a defined region of human chromosome 19 (AAVS1). The specificity of integration is dependent on the presence of the inverted terminal repeats (ITR) and on expression of the rep gene. To develop vectors capable of targeting the insertion of a selected DNA sequence into a specific location of human chromosome, we determined whether the rep gene can mediate site-specific integration when cloned outside of an ITR-flanked transgene cassette. HeLa and Huh-7 cells were transfected with a plasmid containing the rep gene, as well as the green fluorescent protein (GFP) and neomycin (neo) resistance gene inserted between the ITRs of AAV. Southern blot analysis of individual clones detected Rep-mediated site-specific integration of the ITR-flanked DNA in 25% and 12% of the HeLa and Huh-7 clones, respectively. The localization of the GFP-Neo sequence on chromosome 19 also was confirmed by fluorescent in situ hybridization analysis of the transfected HeLa clones. Sequence analysis of the ITR-AAVS1 junction of one of the transfected Huh-7 clones indicated that the insertion of the ITR DNA fragment had occurred at nucleotide 1003. These results have implications for the development of AAV-derived vectors capable of directing the site-specific integration of a gene of interest.

Is DNA destiny? A cure for cystic fibrosis

Cystic fibrosis (CF) remains an attractive target for cure by gene therapy. Results from several trials are reviewed in this article and have shown that mature airway epithelial cells are relatively resistant to gene transfer, that host immune responses determine the duration of transgene expression and define the toxicity, and that the efficiency of transfection remains low. Significant hurdles to the development of gene therapy remain, including the definition of efficacy endpoints, the ability to produce enough material, and the ability to dose the entire lung. Nonetheless, invaluable insights into CF and pulmonary biology have been gained in the gene therapy research effort.

Lipofection of purified adeno-associated virus Rep68 protein: toward a chromosome-targeting nonviral particle

Adeno-associated virus (AAV) integrates very efficiently into a specific site (AAVS1) of human chromosome 19. Two elements of the AAV genome are sufficient: the inverted terminal repeats (ITRs) and the Rep78 or Rep68 protein. The incorporation of the AAV integration machinery in nonviral delivery systems is of great interest for gene therapy. We demonstrate that purified recombinant Rep68 protein is functionally active when directly delivered into human cells by using the polycationic liposome Lipofectamine, promoting the rescue-replication of a codelivered ITR-flanked cassette in adenovirus-infected cells and its site-specific integration in noninfected cells. The sequencing of cloned virus-host DNA junctions confirmed that lipofected Rep68 protein triggers site-specific integration at the same sites in chromosome 19 already characterized in cells latently infected with AAV.

Stomach Implant for Long-Term Erythropoietin Expression in Rats

To approach the goal of consistent long-term erythropoietin (Epo) expression in vivo, we developed an implantation procedure in which transduced autologous vascular smooth muscle was introduced into rats in a chamber created from a polytetrafluoroethylene (PTFE) ring placed under the serosa of the stomach. The implant became vascularized and permitted the long-term survival of smooth muscle cells expressing Epo. Hematocrits of treated animals increased rapidly and monitored over 12 months gave a mean value of 56.0 ± 4.0% (P < .001; n = 9), increased from a presurgery mean of 42.3 ± 1.6%. Hemoglobin levels rose from a presurgery mean of 15.2 ± 0.4 g/dL and for 12 months were significantly elevated with a mean value of 19.5 ± 1.3 g/dL (P < .001; n = 9). The hematocrit and hemoglobin levels of control animals receiving human adenosine deaminase (ADA)–expressing cells were not significantly different from baseline (P > .05; n = 5). In response to tissue oxygenation, kidney, and (to a lesser extent) liver are specific organs that synthesize Epo. Treated animals showed downregulation of endogenous Epo mRNA in kidney over a 12-month period. The PTFE implant provides sustained gene delivery, is safe, and is minimally invasive. It allows easy engraftment of transduced cells and may be applied generally to the systemic delivery of therapeutic proteins such as hormones and clotting factors.

© 1998 by The American Society of Hematology.

Stomach Implant for Long-Term Erythropoietin Expression in Rats

To approach the goal of consistent long-term erythropoietin (Epo) expression in vivo, we developed an implantation procedure in which transduced autologous vascular smooth muscle was introduced into rats in a chamber created from a polytetrafluoroethylene (PTFE) ring placed under the serosa of the stomach. The implant became vascularized and permitted the long-term survival of smooth muscle cells expressing Epo. Hematocrits of treated animals increased rapidly and monitored over 12 months gave a mean value of 56.0 ± 4.0% (P &lt; .001; n = 9), increased from a presurgery mean of 42.3 ± 1.6%. Hemoglobin levels rose from a presurgery mean of 15.2 ± 0.4 g/dL and for 12 months were significantly elevated with a mean value of 19.5 ± 1.3 g/dL (P &lt; .001; n = 9). The hematocrit and hemoglobin levels of control animals receiving human adenosine deaminase (ADA)–expressing cells were not significantly different from baseline (P &gt; .05; n = 5). In response to tissue oxygenation, kidney, and (to a lesser extent) liver are specific organs that synthesize Epo. Treated animals showed downregulation of endogenous Epo mRNA in kidney over a 12-month period. The PTFE implant provides sustained gene delivery, is safe, and is minimally invasive. It allows easy engraftment of transduced cells and may be applied generally to the systemic delivery of therapeutic proteins such as hormones and clotting factors.

© 1998 by The American Society of Hematology.

Membrane-associated heparan sulfate proteoglycan is a receptor for adeno-associated virus type 2 virions

The human parvovirus adeno-associated virus (AAV) infects a broad range of cell types, including human, nonhuman primate, canine, murine, and avian. Although little is known about the initial events of virus infection, AAV is currently being developed as a vector for human gene therapy. Using defined mutant CHO cell lines and standard biochemical assays, we demonstrate that heparan sulfate proteoglycans mediate both AAV attachment to and infection of target cells. Competition experiments using heparin, a soluble receptor analog, demonstrated dose-dependent inhibition of AAV attachment and infection. Enzymatic removal of heparan but not chondroitin sulfate moieties from the cell surface greatly reduced AAV attachment and infectivity. Finally, mutant cell lines that do not produce heparan sulfate proteoglycans were significantly impaired for both AAV binding and infection. This is the first report that proteoglycan has a role in cellular attachment of a parvovirus. Together, these results demonstrate that membrane-associated heparan sulfate proteoglycan serves as the viral receptor for AAV type 2, and provide an explanation for the broad host range of AAV. Identification of heparan sulfate proteoglycan as a viral receptor should facilitate development of new reagents for virus purification and provide critical information on the use of AAV as a gene therapy vector.

Cellular recombination pathways and viral terminal repeat hairpin structures are sufficient for adeno-associated virus integration in vivo and in vitro

The human parvovirus adeno-associated virus (AAV) is unique in its ability to target viral integration to a specific site on chromosome 19 (ch-19). Recombinant AAV (rAAV) vectors retain the ability to integrate but have apparently lost this ability to target. In this report, we characterize the terminal-repeat-mediated integration for wild-type (wt), rAAV, and in vitro systems to gain a better understanding of these differences. Cell lines latent for either wt or rAAV were characterized by a variety of techniques, including PCR, Southern hybridization, and fluorescence in situ hybridization analysis. More than 40 AAV-rAAV integration junctions were cloned, sequenced, and then subjected to comparison and analysis. In both immortalized and normal diploid human cells, wt AAV targeted integration to ch-19. Integrated provirus structures consisted of head-to-tail tandem arrays with the majority of the junction sequences involving the AAV inverted terminal repeats (ITRs). No complete viral ITRs were directly observed. In some examples, the AAV p5 promoter sequence was found to be fused at the virus-cell junction. Data from dot blot analysis of PCR products were consistent with the occurrence of inversions of genomic and/or viral DNA sequences at the wt integration site. Unlike wt provirus junctions, rAAV provirus junctions mapped to a subset of non-ch-19 sequences. Southern analysis supported the integration of proviruses from two independent cell lines at the same locus on ch-2. In addition, provirus terminal repeat sequences existed in both the flip and flop orientations, with microhomology evident at the junctions. In all cases with the exception of the ITRs, the vector integrated intact. rAAV junction sequence data were consistent with the occurrence of genomic rearrangement by deletion and/or rearrangement-translocation at the integration locus. Finally, junctions formed in an in vitro system between several AAV substrates and the ch-19 target site were isolated and characterized. Linear AAV substrates typically utilized the end of the virus DNA substrate as the point of integration, whereas products derived from AAV terminal repeat hairpin structures in the presence or absence of Rep protein resembled AAV-ch-19 junctions generated in vivo. These results describing wt AAV, rAAV, and in vitro integration junctions suggest that the viral integration event itself is mediated by terminal repeat hairpin structures via nonviral cellular recombination pathways, with specificity for ch-19 in vivo requiring additional viral components. These studies should have an important impact on the use of rAAV vectors in human gene therapy.

Transduction of the human immunodeficiency virus type 1 promoter into human chromosomal DNA by adeno-associated virus: effects on promoter activity

Transcription of the human immunodeficiency virus type 1 (HIV-1) genome takes place after integration of the provirus into human chromosomal DNA. HIV transcription is known to be modulated by viral and cellular factors but the influence of flanking chromosomal sequences on proviral gene expression has not been well defined. To investigate the activity of the integrated HIV promoter, we exploited the ability of recombinant adeno-associated virus (AAV-2) to transfer and stably integrate genes into the human genome at random or site-specifically. Chimeric AAV vectors were constructed containing an HIV-CAT reporter cassette; some vectors also contained the neomycin resistance gene to facilitate the isolation of positive clones. HeLa cells were infected with recombinant AAV, in some instances together with wild-type virus as a source of AAV rep function. We isolated 25 clones of G418-resistant cells which carried the integrated HIV-CAT cassette, generally occupying unique sites that did not correspond to the AAV-specific region of chromosome 19. The HIV promoter was transcriptionally active in most of the clones. Basal promoter activity varied substantially among the clones, and its responsivity to the HIV transactivator Tat was also variable. The integrated HIV promoter was transactivated to comparable degrees by the one-exon form and two-exon form of Tat. These findings provide evidence that the transcriptional activity of the HIV promoter can be greatly influenced by the site of proviral insertion.

Adeno-associated virus Rep78 protein and terminal repeats enhance integration of DNA sequences into the cellular genome

Two adeno-associated virus (AAV) elements are necessary for the integration of the AAV genome: Rep78/68 proteins and inverted terminal repeats (ITRs). To study the contribution of the Rep proteins and the ITRs in the process of integration, we have compared the integration efficiencies of three different plasmids containing a green fluorescent protein (GFP) expression cassette. In one plasmid, no viral sequences were present; a second plasmid contained AAV ITRs flanking the reporter gene (integration cassette), and a third plasmid consisted of an integration cassette plus a Rep78 expression cassette. One day after transfection of 293 cells, fluorescent cells were sorted by flow cytometry and plated at 1 cell per well. Two weeks after sorting, colonies were monitored for stable expression of GFP. Transfection with the GFP plasmid containing no viral sequences resulted in no stable fluorescent colonies. Transfection with the plasmid containing the integration cassette alone (GFP flanked by ITRs) produced stable fluorescent colonies at a frequency of 5.3% +/- 1.0% whereas transfection with the plasmid containing both the integration cassette and Rep78 expression cassette produced stable fluorescent colonies at a frequency of 47% +/- 7.5%. Southern blot analysis indicated that in the presence of Rep78, integration is targeted to the AAVSI site in more than 50% of the clones analyzed. Some clones also showed tandem arrays of the integrated GFP cassette. Both head-to-head and head-to-tail orientations were detected. These findings indicate that the presence of AAV ITRs and the Rep78 protein enhance the integration of DNA sequences into the cellular genome and that the integration cassette is targeted to AAVS1 in the presence of Rep78.

Characterization of the Rep78/adeno-associated virus complex

Adeno-associated virus (AAV) replication proteins Rep78 and Rep68 play major roles in the life cycle of AAV. We have recently provided in vivo evidence for the existence of a covalent association between Rep78 and virion single-stranded (ss) AAV DNA (K. M. R. Prasad and J.P. Trempe(1995) Virology 214, 360-370). In this work we have further characterized the Rep78 protein-AAV DNA covalent linkage. Exonuclease and primer extension analyses revealed that in the majority of isolated ssDNA, Rep78 protein is covalently linked to one of the 5' terminal thymidines. Pulse-chase experiments with radiolabeled methionine suggest that Rep protein remains associated with the virus particle for up to 8 hr after labeling in infected cells. Quantitative immunoprecipitation indicated that approximately 30% of the ssDNA remains associated with the Rep protein after cell fractionation and partial purification. When cells are infected with Rep-associated AAV particles, a significant proportion of viral DNA remains attached to Rep after entry into the nucleus. However, the linkage does not persist after nuclear entry. These observations suggest that covalently linked Rep78 protein may play a key role during wild-type AAV infections and AAV vector transductions.

L-Dopa and dopamine-producing gene cassettes for gene therapy approaches to Parkinson's disease

As an aid in the development of vector systems for use in gene therapy paradigms of central nervous system disorders such as Parkinson's disease, we have developed L-Dopa or dopamine-producing gene cassettes. Specifically, a human tyrosine hydroxylase cDNA (HTH-2) was rendered constitutively active by truncation of the N-terminal regulatory domain (tHTH). In addition, a bicistronic construct capable of directing the production of dopamine was created by inserting an internal ribosome entry site downstream of tHTH followed by the coding sequences of aromatic amino acid decarboxylase. All three constructs generated immunoreactive peptides of the predicted size, were enzymatically active, and produced L-Dopa (HTH-2, tHTH) or dopamine (bicistronic construct) following transient transfection of COS-7 cells. These constructs, in conjunction with viral or nonviral expression systems, may be efficacious in gene therapy approaches to Parkinson's disease.

Practical aspects of the development of ex vivo and in vivo gene therapy for Parkinson's disease

Current approaches to gene therapy of CNS disorders include grafting genetically modified autologous cells or introducing genetic material into cells in situ using a variety of viral or synthetic vectors to produce and deliver therapeutic substances to specific sites within the brain. Here we discuss issues related to the application of ex-vivo and in-vivo gene therapies as possible treatments for Parkinson's disease. Autologous monkey fibroblasts engineered ex-vivo to express tyrosine hydroxylase were grafted into MPTP-treated monkeys and found to express for up to 4 months. Adeno-associated (AAV) viral vectors expressing beta-galactosidase or tyrosine hydroxylase were introduced into monkey brains to determine the extent of infection and the types of cells infected by the vector at 21 days and 3 months. Gene expression was detected at both time points and was restricted to neurons in the striatum. These experiments demonstrate that two different approaches can be used to deliver proteins into the CNS. However, further technological advances are required to optimize gene delivery, regulation of gene expression, and testing in appropriate functional models before gene therapy can be considered for treating human disease.

Gene transfer by adeno-associated virus vectors into the central nervous system

Adeno-associated virus (AAV) vectors are derived from a nonpathogenic and defective human parvovirus. Although currently unable to display the integration specificity featured by its wild-type parent, the recombinant AAV (rAAV) system has continued to attract enormous interest primarily due to its unique features such as safety, high titers, broad host range, transduction of quiescent cells, and vector integration. Recently, rAAV-mediated in vivo gene transfers have demonstrated efficient long-term transduction (from 3 months to more than 1.5 years) and lack of cytotoxicity and cellular immune responses in the target tissues, especially in the CNS. Alternative approaches using rAAV plasmid DNA in nonviral gene delivery systems also generated promising results. Propelled by various efforts to improve the system, rAAV vectors will provide numerous opportunities to explore the potential therapeutic applications in humans.

Lack of site-specific integration of the recombinant adeno-associated virus 2 genomes in human cells

The adeno-associated virus 2 (AAV)-based vector system has been suggested for its potential use in human gene therapy because the wild-type (wt) AAV genome appears to integrate into the human chromosomal DNA in a site-specific manner. We systematically investigated the integration patterns of the recombinant AAV genomes lacking one or both the viral coding sequences. Four recombinant AAV genomes were constructed containing the genes for resistance to tetracycline (TcR) and the herpesvirus thymidine kinase (TK) promoter-driven gene for resistance to neomycin (neoR; vTc.Neo), the genes for resistance to ampicillin (ApR) and TK-neoR (vAp.Neo), the genes for AAV replication (rep) genes and TK-neoR (vRep.Neo), and the AAV capsid (cap) genes and TK-neoR (vCap.Neo). The integration pattern of each of the recombinant AAV genomes in individual clonal isolates of the human nasopharyngeal carcinoma cell line (KB) analyzed on Southern blots using a neo-specific DNA probe was distinctly different. In addition, in none of the clones examined was the proviral genome covalently linked to the previously described AAV right-junction (Rt.Jn.) human chromosomal DNA fragment, the putative specific-site of integration for the wt AAV genome. Furthermore, whereas a 276-bp DNA fragment could be readily amplified from each of these clones, using a neo-specific primer-pair by polymerase chain reaction (PCR), no amplified DNA product was obtained using the neo- and the Rt.Jn. primer-pair under identical conditions. Fluorescence in situ hybridization (FISH) analyses further revealed the lack of integration of the recombinant AAV into human chromosome 19, even in the presence of a functional rep gene as determined by rescue of the recombinant AAV genome in the presence of adenovirus. These data suggest that the recombinant AAV genomes integrate at sites that are different from that characterized for the wt AAV genome. These studies may have implications in the development of the AAV-based vector system for its potential use in human gene therapy.

HSV/AAV hybrid amplicon vectors extend transgene expression in human glioma cells

Novel hybrid vectors, which incorporate critical elements of both herpes simplex virus type 1 (HSV-1) amplicon vectors and adeno-associated virus (AAV) vectors, are able to sustain transgene expression in dividing glioma cells for over 2 weeks. These vectors combine the high infectibility and large transgene capacity of HSV-1 vectors with the potential for episomal amplification and chromosomal integration of AAV vectors. The hybrid vectors contain the HSV-1 origin of DNA replication, oriS, and the DNA cleavage/packaging signal, pac, which allow amplicon replication and packaging in HSV-1 virions. The lacZ reporter gene under control of the CMV IE1 promoter is flanked by AAV inverted terminal repeat (ITR) sequences, which facilitate replication and genomic integration of this cassette in the host cell nucleus. Constructs were generated with or without the AAV rep gene (rep+ and rep-) to assess its importance in extending transgene expression. Expression of Rep proteins was confirmed by Western blot analysis. An HSV-1 amplicon construct containing the reporter gene, but no AAV sequences, was used as a control. Constructs were packaged into HSV-1 virions with or without helper virus and these vector stocks were used to infect human U87 glioma cells in culture. The hybrid vectors supported transgene retention and expression for over 2 weeks, whereas the control amplicon vector lost the transgene after 10 days. Expression was somewhat longer for the rep+ as compared to the rep- hybrid vectors. Toxicity due to the HSV-1 helper virus was eliminated using helper virus-free amplicon vector stocks. Transgene constructs could also be packaged in AAV virions, using AAV and adenovirus or HSV-1 helper functions. These HSV/AAV hybrid vectors should allow long-term, nontoxic gene delivery of DNA constructs to both dividing and nondividing cells.

A novel 165-base-pair terminal repeat sequence is the sole cis requirement for the adeno-associated virus life cycle

Adeno-associated virus (AAV) replication is dependent on two copies of a 145-bp inverted terminal repeat (ITR) that flank the AAV genome. This is the primary cis-acting element required for productive infection and the generation of recombinant AAV (rAAV) vectors. We have engineered a plasmid (pDD-2) containing only 165 bp of AAV sequence: two copies of the D element, a unique sequence adjacent to the AAV nicking site, flanking a single ITR. When assayed in vivo, this modified hairpin was sufficient for the replication of the plasmid vector when Rep and adenovirus (Ad) helper functions were supplied in trans. pDD-2 replication intermediates were characteristic of the AAV replication scheme in which linear monomer, dimer, and other higher-molecular-weight replicative intermediates are generated. Compared to infectious AAV clones for replication, the modified hairpin vector replicated more efficiently independent of size. Further analysis demonstrated conversion of the input circular plasmid to a linear substrate with AAV terminal repeat elements at either end as an initial step for replication. This conversion was independent of both Rep and Ad helper genes, suggesting the role of host factors in the production of these molecules. The generation of these substrates suggested resolution of the modified terminal repeat through a Holliday-like structure rather than replication as a mechanism for rescue. Production of replicative intermediates via this plasmid substrate were competent not only for AAV DNA replication but also for encapsidation, infection, integration, and subsequent rescue from the chromosome when superinfected with Ad and wild-type AAV. These studies demonstrate that this novel 165-bp ITR substrate is sufficient in cis for the AAV life cycle and should provide a valuable reagent for further dissecting the cis sequences involved in AAV replication, packaging, and integration. In addition, this novel plasmid vector can be used as a substrate for both rAAV vector production and synthetic plasmid vector delivery.

Replication of rep-cap genes is essential for the high-efficiency production of recombinant AAV

Adenoassociated virus (AAV) has been developed as a vector for gene transfer because of its advantageous features: it is nonpathogenic, naturally replication-defective; it infects growth-arrested cells, and can transfer the therapeutic gene without co-delivery of any viral genes. However, a major obstacle in conducting systematic studies of AAV-mediated gene transfer in animal models is the difficulty of obtaining large quantities of recombinant virus. Recent development of AAV packaging cell lines has simplified the procedure of producing recombinant AAV (rAAV). However, the efficacy of producing large quantities of rAAV with these cell lines is yet to be demonstrated. In this study we have analyzed the difference between the replication of wild-type AAV and the production of rAAV. Using a combined single-plasmid system that carries both an AAV vector and the rep-cap genes, we have demonstrated that the AAV vector replicates to high number of copies whereas the rep-cap sequences remain unamplified in the virus-producing cells, When the copy number of rep-cap genes was increased by varying the vector/rep-cap ratio in the transfection mixture, the titer of rAAV increased proportionally. Thus, the titer of rAAV is limited by the low copy number of the rep-cap genes that results in an insufficient expression of the Rep and Cap proteins. We have also shown that generation of double-stranded replicating form of the vector DNA is accompanied by an amplified transgene expression. We propose that the increased gene expression from the accumulating double-stranded viral DNA is likely to be the mechanism by which wild-type AAV produces a large number of particles necessary to package the self-replicating AAV genomes. We conclude that mimicking this amplified expression of rep-cap genes may provide the key to produce high titers of rAAV.

Efficient long-term gene transfer into muscle tissue of immunocompetent mice by adeno-associated virus vector

Muscle-directed gene transfer is being considered for the treatment of several metabolic diseases, including hemophilia and Duchene's muscular dystrophy. Previous efforts to target this tissue for somatic delivery with various vector systems have resulted in transient expression due to silencing of the transgene or to an immune response against the vector-transduced cells. We introduced recombinant adeno-associated virus vector (rAAV) carrying a lacZ reporter into muscle tissue of immunocompetent mice. The lacZ reporter gene was efficiently transduced and expressed with no evidence of a cellular immune response. Moreover, gene expression persisted for more than 1.5 years. Molecular characterization of rAAV vector DNA suggests a mechanism for persistence, since vector episomes convert to high-molecular-weight genomic DNA. These data provide the first report for establishing long-term gene transduction into mammalian muscle cells in vivo without the need for immune modulation of the organism.

Cardiovascular gene therapy: current concepts

Gene therapy techniques are under development for many areas of medicine, including cardiovascular disease. Identifying appropriate gene targets will require more detailed knowledge of the molecular pathophysiology of these disorders, and choosing appropriate vectors and delivery systems will contribute significantly to the challenge of developing this approach for clinical use. The concepts of toxicology and therapeutic drug monitoring will need to be broadened to account for the unique chemical, biological, and genetic characteristics of gene therapeutic agents. This review will provide an overview of strategy development, currently available vectors, and examples of their application to cardiovascular gene transfer. Considerations of the potential toxicities associated with particular vectors and delivery systems, as well as the types of genetic modifications possible, will provide some guidelines regarding appropriate monitoring of their clinical application.

In vivo model of adeno-associated virus vector persistence and rescue

Gene therapy vectors based on human DNA viruses could be mobilized or rescued from individuals who are subsequently infected with the corresponding wild-type (wt) helper viruses. This phenomenon has been effectively modeled in vitro with both adenovirus (Ad) and adeno-associated virus (AAV) vectors but has not previously been studied in vivo. In the current study, we have developed an in vivo model to study the interactions of a recombinant AAV vector (AAV-CFTR) with wt AAV type 2 (AAV2) and a host range mutant Ad (Ad2HR405) for which monkey cells are permissive (D.E.Brough, S.A.Rice, S.Sell, and D.F.Klessig, J. Virol. 55:206-212, 1985). AAV-CFTR was administered to the respiratory epithelium of the nose or lung of rhesus macaques. Primary cells were harvested from the infusion site at time points up to 3 months after vector administration to confirm vector DNA persistence. Vector DNA was present in episomal form and could be rescued in vitro only by addition of wt AAV2 and Ad. In in vivo rescue studies, vector was administered before or after wt-AAV2 and Ad2HR405 infection, and the shedding of AAV-CFTR was examined. Ad2HR405 and wt-AAV2 infections were established in the nose with concomitant administration. wt-AAV2 replication occurred in the lung when virus was administered directly at a high titer to the lower respiratory tract. AAV-CFTR vector rescue was also observed in the latter setting. Although these studies were performed with small numbers of animals within each group, it appears that AAV-CFTR DNA persists in the primate respiratory tract and that this model may be useful for studies of recombinant AAV vector rescue.

High-level expression of adeno-associated virus (AAV) Rep78 or Rep68 protein is sufficient for infectious-particle formation by a rep-negative AAV mutant

Adeno-associated virus (AAV) codes for four closely related nonstructural proteins (Rep) required for AAV DNA replication and gene regulation. In vitro studies have revealed that either Rep78 or Rep68 alone is sufficient for AAV DNA replication. Rep52 and Rep40 are not required for DNA replication but have been reported to enhance the efficiency of accumulation of single-stranded progeny DNA. Previous studies on rep-expressing cell lines had indicated that only a subset of the four Rep proteins are required for the production of infectious AAV. We therefore set out to determine the minimal set of Rep proteins sufficient for the generation of infectious AAV. Transient cotransfections in HeLa cells of constructs for high-level expression of individual Rep proteins with a rep-negative AAV genome revealed that either Rep78 or Rep68 alone could complement for a full replication cycle yielding infectious virus. This result was confirmed by transfection studies in the cell line HeM2, which selectively expresses Rep78 at rather low levels under the control of the glucocorticoid-responsive mouse mammary tumor virus long terminal repeat (C. Hölscher, M. Hörer, J. A. Kleinschmidt, H. Zentgraf, A. Bürkle, and R. Heilbronn, J. Virol. 68:7169-7177, 1994). Increasing the level of Rep78 expression by transfection of a glucocorticoid receptor expression construct resulted in a higher level of DNA replication of a cotransfected rep-negative AAV genome and in the production of infectious rep-negative AAV particles. We further report on the generation of a new rep-expressing cell line, HeCM1, which was obtained by stable supertransfection of a construct for constitutive Rep40 expression into HeM1 cells (Hölscher et al., J. Virol. 68:7169-7177). Transfection of rather large amounts of rep-negative AAV DNA led to detectable virus production in HeCM1 cells even in the absence of the cotransfected glucocorticoid receptor expression construct, but higher yields were obtained after increasing the Rep78 level by coexpression of the glucocorticoid receptor. These data demonstrate that all Rep functions required for the productive replication of AAV in HeLa cells are contained within both Rep78 and Rep68.

Recombinant junctions formed by site-specific integration of adeno-associated virus into an episome

A model system using an episomal Epstein-Barr virus shuttle vector was recently developed to study the adeno-associated virus (AAV) site-specific integration event in chromosome 19q13.3-qter (C. Giraud, E. Winocour, and K.I. Berns, Proc. Natl. Acad. Sci. USA 91:10039-10043, 1994). In this study, we analyze the recombinant junctions generated after integration of the AAV genome into an Epstein-Barr virus shuttle vector carrying 8.2, 1.6, or 0.51 kb of the chromosome 19 preintegration sequence (AAVS1 locus). In most of the recombinants, one end of the viral genome was joined to a portion of the AAVS1 DNA previously shown to be a minimum target for AAV integration. Within this AAVS1 segment, the AAV insertion points were strikingly clustered around a binding site for the AAV regulatory protein. In all cases, the second junction with AAV occurred with vector DNA outside of the AAVS1 segment. With respect to the viral genome, one junction with the shuttle vector DNA occurred either within the AAV inverted terminal repeat (itr), or near the P5 promoter, approximately 100 nucleotides distal to a modified itr. The modified itr in 5 of 11 recombinants involved a head-to-tail organization. In one such instance, the AAV insert contained slightly more than one genome equivalent arranged in a head-to-tail manner with a junction close to the P5 promoter; the AAV insert in this recombinant episome could be rescued by adenovirus infection and replicated to virus particles. The significance of the head-to-tail organization is discussed in terms of the possible circularization of AAV DNA before or during integration.

Gene therapy for brain tumors

Gene therapy has opened new doors for treatment of neoplastic diseases. This new approach seems very attractive, especially for glioblastomas, since treatment of these brain tumors has failed using conventional therapy regimens. Many different modes of gene therapy for brain tumors have been tested in culture and in vivo. Many of these approaches are based on previously established anti-neoplastic principles, like prodrug activating enzymes, inhibition of tumor neovascularization, and enhancement of the normally weak anti-tumor immune response. Delivery of genes to tumor cells has been mediated by a number of viral and synthetic vectors. The most widely used paradigm is based on the activation of ganciclovir to a cytotoxic compound by a viral enzyme, thymidine kinase, which is expressed by tumor cells, after the gene has been introduced by a retroviral vector. This paradigm has proven to be a potent therapy with minimal side effects in several rodent brain tumor models, and has proceeded to phase 1 clinical trials. In this review, current gene therapy strategies and vector systems for treatment of brain tumors will be described and discussed in light of further developments needed to make this new treatment modality clinically efficacious.

Transduction of folate receptor cDNA into cervical carcinoma cells using recombinant adeno-associated virions delays cell proliferation in vitro and in vivo

Although folate receptors (FRs) mediate folate uptake into cells, the independent role of FRs in cell proliferation remains unclear. We tested the hypothesis that transduction of FR cDNA in sense or antisense orientation using recombinant adeno-associated virus modulated FR expression and altered proliferation of cervical carcinoma cells (which constitutively overexpress FR genes). We determined that the integration of recombinant adeno-associated virions was not site specific. When compared with untransduced cells, sense and antisense FR cDNA-transduced cells exhibited an increase and decrease in FR mRNA and FR expression on the cell surface, respectively. However, when compared with antisense FR cDNA-transduced and untransduced cells, sense FR cDNA-transduced cells exhibited statistically significant (a) increased in total FRs, (b) smaller colonies, (c) lowered cell proliferation in vitro, and (d) less tumor volume with dramatic prolongation of tumor doubling times (225.6 h vs. 96 h) after transplantation into nude mice. Finally, (f) using single cell-derived transduced clones, an inverse relationship between cell proliferation and FR expression was established (r = 0.90, P < 0.001). Thus, transduction of sense/antisense FR cDNA into cervical carcinoma cells modulated expression of FRs and had an impact on cell proliferation in vitro and in vivo.

AAV as a viral vector for human gene therapy. Generation of recombinant virus

Investigation of the adeno-associated virus (AAV) life cycle has enabled the establishment of methodology and identification of critical cis-acting sequences required for recombinant AAV production. Vectors derived from the defective human parvovirus (AAV) have been used for successful gene transfer and expression in many diverse mammalian cell types, such as erythroid, airway epithelium, and neuronal cells. One of the crucial steps in the continued case of AAV as a vector is the development of packaging systems that will allow efficient encapsidation of foreign genes into AAV virions. For this reason, the focus of this article will be generation of recombinant AAV vectors.

Approaches to gene transfer in keratinocytes

The introduction and expression of exogenous genetic material in cultured cells has provided a powerful tool for studying gene function and regulation. Immortalized cell lines have been useful for establishing gene transfer methodologies that are generally inefficient. For investigators of epidermal and mucosal biology, wishing to make use of the tissue architecture produced by primary keratinocytes in vitro, the limited life span of these cells presents a host of unique problems. Primary cells require the use of gene transfer methods that are highly efficient and will not significantly alter the cell's normal differentiation pathway. The purpose of this review is to evaluate gene transfer technology as it applies to keratinocytes.

Cell lines inducibly expressing the adeno-associated virus (AAV) rep gene: requirements for productive replication of rep-negative AAV mutants

The adeno-associated virus (AAV) rep gene codes for a family of nonstructural proteins which are required for AAV gene regulation and DNA replication. In addition, rep has been implicated in a variety of activities outside the AAV life cycle which have been difficult to study, since attempts to achieve separate and constitutive expression of rep in stable cell lines have failed so far. Here we report the generation of two cell lines which inducibly express Rep78 under the control of the glucocorticoid-responsive mouse mammary tumor virus promoter. In addition, one of the cell lines constitutively expresses relatively high levels of Rep52. Both cell lines showed similar plating efficiencies with and without induction of Rep78 expression, which rules out cytotoxic effects of Rep78. The cell lines efficiently support DNA replication of a rep-negative AAV genome and initiate the formation of AAV particles. However, despite the correct sizes and stoichiometry of the three capsid proteins, the AAV particles were noninfectious. This was found to be due to a defect in the accumulation of single-stranded AAV DNA. Transient transfection of single expression constructs for constitutive, high-level expression of individual Rep proteins (either Rep78, Rep68, Rep52, or Rep40) complemented this defect. Infectious rep-negative AAV progeny was produced at varying efficiencies depending on the rep expression construct used. These data show that functional expression of full-length Rep in recombinant cell lines is possible and that the state of Rep expression is critical for the infectivity of AAV progeny produced.

Phenotypic correction of Fanconi anemia in human hematopoietic cells with a recombinant adeno-associated virus vector

Fanconi anemia (FA) is a recessive inherited disease characterized by defective DNA repair. FA cells are hypersensitive to DNA cross-linking agents that cause chromosomal instability and cell death. FA is manifested clinically by progressive pancytopenia, variable physical anomalies, and predisposition to malignancy. Four complementation groups have been identified, termed A, B, C, and D. The gene for the FA complementation group C, FACC, has been cloned. Expression of the FACC cDNA corrects the phenotypic defect of FA(C) cells, resulting in normalized cell growth in the presence of DNA cross-linking agents such as mitomycin C (MMC). Gene transfer of the FACC gene should provide a survival advantage to transduced hematopoietic cells, suggesting that FA might be an ideal candidate for gene therapy. We demonstrated efficient transduction, expression, and phenotypic correction in lymphoblastoid cell lines derived from FA (C) patients using a recombinant adeno-associated virus (rAAV) vector containing the FACC gene. Molecular characterization of the transduced FACC gene showed an intact unrearranged proviral genome with expression sufficient to normalize cell growth, cell cycle kinetics and chromosomal breakage in the presence of MMC. These observations were extended by testing rAAV transduction in hematopoietic progenitor cells. Peripheral blood CD34+ cells isolated from a FA (C) patient and transduced with rAAV/FACC virus yielded 5-10-fold more progenitor colonies than mock-infected cells, consistent with genetic "rescue" of corrected cells. This is the first demonstration of rAAV gene correction in primary human hematopoietic progenitor cells and has important implications for gene therapy of hematopoietic disorders, specifically FA.

Adeno-associated virus vectors preferentially transduce cells in S phase

Vectors based on adeno-associated virus can stably transfer genes by chromosomal integration in recipient cells. In this study we have infected stationary and dividing primary human fibroblast cultures with adeno-associated virus vectors encoding alkaline phosphatase and neomycin phosphotransferase. We find that the transduction frequency of S phase cells is about 200 times that of non-S phase cells. However, neither S phase nor mitosis is essential for transduction. Single-stranded vector genomes survive in stationary cultures and can be recruited for transduction by stimulating these cultures to divide. Stable transductants contain randomly integrated vector sequences. These findings have important implications for the use of adeno-associated virus vectors in gene therapy.

Use of recombinant vectors derived from herpes simplex virus 1 mutant tsK for short-term expression of transgenes encoding cytoplasmic and membrane anchored proteins in postmitotic polarized cortical neurons and glial cells in vitro

We constructed three recombinant vectors derived from the herpes simplex virus type 1 mutant tsK, each of which contained a different transgene under the control of the herpes simplex virus type 1 immediate early 3 promoter inserted into the thymidine kinase locus: the prokaryotic enzymes beta-galactosidase and chloramphenicol acetyl transferase, and a fusion gene consisting of human tissue inhibitor of metalloproteinases linked to the last exon of Thy-1, which encodes for a glycosyl-phosphatidyl-inositol membrane anchor. Infection of postmitotic neocortical and hippocampal neurons in low-density primary cultures with these vectors, achieved reliable expression of all three foreign gene products in various neocortical cell types, e.g. pyramidal neurons, non-pyramidal neurons, and glial cells. The percentage of neurons expressing transgenes ranged from 1 to 46% depending on the multiplicity of infection (highest assayed = 5); the percentage of glial cells expressing transgenes ranged from 0.5 to 98% (highest multiplicity assayed = 3.4). Expression of transgenes could be detected for up to three days in approximately 20% of neurons infected at a multiplicity of infection of 1. Infection of neurons with tk K-derived recombinant vectors inhibited their protein synthesis by 40-50% at a multiplicity of infection of 10, but no effect was observed at a multiplicity of infection of 1. Infection of glial cells with the same vectors at a multiplicity of infection of 1 inhibited protein synthesis by more than 90%. Analysis of neuronal viability at different times post-infection indicated that more than 98% of neurons expressing transgenes 48 h post-infection were viable. Thus, low-density neuronal cultures can be used to assess the efficiency of herpes simplex virus type 1-derived gene transfer vectors and transgene expression in developing cortical postmitotic cells, before and after they establish polarity. In addition, we show that two cytoplasmic enzymes, beta-galactosidase and chloramphenicol acetyl transferase, are able to diffuse freely in the cytoplasm reaching even growth cones in young neurons, while the chimeric protein tissue inhibitor of metalloproteinases/Thy-1 is correctly targeted to the plasma membrane via a glycosyl-phosphatidylinositol anchor. This model system should be useful for investigation of cellular and molecular aspects of the development and establishment of neuronal polarity, as well as for analysis of signals involved in protein targeting in postmitotic neurons.