Effect of CpG Depletion of Vector Genome on CD8+ T Cell Responses in AAV Gene Therapy

Adeno associated viral (AAV) vectors have emerged as a preferred platform for in vivo gene replacement therapy and represent one of the most promising strategies to treat monogenetic disorders such as hemophilia. However, immune responses to gene transfer have hampered human gene therapy in clinical trials. Over the past decade, it has become clear that innate immune recognition provides signals for the induction of antigen-specific responses against vector or transgene product. In particular, TLR9 recognition of the vector's DNA genome in plasmacytoid dendritic cells (pDCs) has been identified as a key factor. Data from clinical trials and pre-clinical studies implement CpG motifs in the vector genome as drivers of immune responses, especially of CD8+ T cell activation. Here, we demonstrate that cross-priming of AAV capsid-specific CD8+ T cells depends on XCR1+ dendritic cells (which are likely the main cross-presenting cell that cooperates with pDCs to activate CD8+ T cells) and can be minimized by the elimination of CpG motifs in the vector genome. Further, a CpG-depleted vector expressing human coagulation factor IX showed markedly reduced (albeit not entirely eliminated) CD8+ T cell infiltration upon intramuscular gene transfer in hemophilia B mice when compared to conventional CpG+ vector (comprised of native sequences), resulting in better preservation of transduced muscle fibers. Therefore, this deimmunization strategy is helpful in reducing the potential for CD8+ T cell responses to capsid or transgene product. However, CpG depletion had minimal effects on antibody responses against capsid or transgene product, which appear to be largely independent of CpG motifs.

Engineering and In Vitro Selection of a Novel AAV3B Variant with High Hepatocyte Tropism and Reduced Seroreactivity

Limitations to successful gene therapy with adeno-associated virus (AAV) can comprise pre-existing neutralizing antibodies to the vector capsid that can block cellular entry, or inefficient transduction of target cells that can lead to sub-optimal expression of the therapeutic transgene. Recombinant serotype 3 AAV (AAV3) is an emerging candidate for liver-directed gene therapy. In this study, we integrated rational design by using a combinatorial library derived from AAV3B capsids with directed evolution by in vitro selection for liver-targeted AAV variants. The AAV3B-DE5 variant described herein was undetectable in the original viral library but gained a selective advantage upon in vitro passaging in human hepatocarcinoma spheroid cultures. AAV3B-DE5 contains 24 capsid amino acid substitutions compared with AAV3B, distributed among all five variable regions, with strong selective pressure on VR-IV, VR-V, and VR-VII. In vivo, AAV3B-DE5 demonstrated improved human hepatocyte tropism in a liver chimeric mouse model. Importantly, this variant exhibited reduced seroreactivity to human intravenous immunoglobulin (i.v. Ig), as well as individual serum samples from 100 healthy human donors. Therefore, molecular evolution using a combinatorial library platform generated a viral capsid with high hepatocyte tropism and enhanced evasion of pre-existing AAV neutralizing antibodies.

Type I IFN Sensing by cDCs and CD4+ T Cell Help Are Both Requisite for Cross-Priming of AAV Capsid-Specific CD8+ T Cells

Adeno-associated virus (AAV) vectors are widely used in clinical gene therapy to correct genetic disease by in vivo gene transfer. Although the vectors are useful, in part because of their limited immunogenicity, immune responses directed at vector components have complicated applications in humans. These include, for instance, innate immune sensing of vector components by plasmacytoid dendritic cells (pDCs), which sense the vector DNA genome via Toll-like receptor 9. Adaptive immune responses employ antigen presentation by conventional dendritic cells (cDCs), which leads to cross-priming of capsid-specific CD8⁺ T cells. In this study, we sought to determine the mechanisms that promote licensing of cDCs, which is requisite for CD8⁺ T cell activation. Blockage of type 1 interferon (T1 IFN) signaling by monoclonal antibody therapy prevented cross-priming. Furthermore, experiments in cell-type-restricted knockout mice showed a specific requirement for the receptor for T1 IFN (IFNaR) in cDCs. In contrast, natural killer (NK) cells are not needed, indicating a direct rather than indirect effect of T1 IFN on cDCs. In addition, co-stimulation by CD4⁺ T cells via CD40-CD40L was required for cross-priming, and blockage of co-stimulation but not of T1 IFN additionally reduced antibody formation against capsid. These mechanistic insights inform the development of targeted immune interventions.

Biodistribution of adeno-associated virus type 2 with mutations in the capsid that contribute to heparan sulfate proteoglycan binding

We compared the phenotypes of three mutant AAV2 viruses containing mutations in arginine amino acids (R585, R588 and R484) previously shown to be involved in AAV2 heparan sulfate binding. The transduction efficiencies of wild type and mutant viruses were determined in the eye, the brain and peripheral organs following subretinal, striatal and intravenous injection, respectively, in mice and rats. We found that each of the three mutants (the single mutant R585A; the double mutant R585, 588A; and the triple mutant R585, 588, 484A) had a unique phenotype compared to wt and each other. R585A was completely defective for transducing peripheral organs via intravenous injection, suggesting that R585A may be useful for targeting peripheral organs by substitution of peptide ligands in the capsid surface. In the brain, all three mutants displayed widespread transduction, with the double mutant R585, 588A displaying the greatest spread and the greatest number of transduced neurons. The double mutant was also extremely efficient for retrograde transport, while the triple mutant was almost completely defective for retrograde transport. This suggested that R484 may be directly involved in interaction with the transport machinery. Finally, the double mutant also displayed improved transduction of the eye compared to wild type and the other mutants.

Polymorphisms of genes involved in inflammation and blood vessel development influence the risk of varicose veins

Heredity plays an important role in the etiology of varicose veins (VVs). However, the genetic basis underlying this condition remains poorly understood. Our aim was to replicate top association signals from genome-wide association studies (GWASs) for VVs of lower extremities using 2 independent datasets-our sample of ethnic Russian individuals (709 cases and 278 controls) and a large cohort of British residents from UK Biobank (10 861 cases and 397 594 controls). Associations of polymorphisms rs11121615, rs6712038, rs507666, rs966562, rs7111987, rs6062618, and rs6905288 were validated in the UK Biobank individuals at a Bonferroni-corrected significance level. In Russian cohort, only rs11121615 reached a nominal significance level of P < .05. Results of original GWAS and replication studies were combined by a meta-analysis, and polymorphisms listed above as well as rs111434909 and rs4463578 passed a genome-wide significant threshold. Notably, the majority of these polymorphisms were located within or near genes involved in vascular development and remodeling, and regulation of inflammatory response. Our results confirm the role of these polymorphisms in genetic susceptibility to VVs and indicate the revealed genomic regions as good candidates for further fine-mapping studies and functional analysis. Moreover, our findings implicate inflammation and abnormal vascular architecture in VVs pathogenesis.

Evaluation of engineered AAV capsids for hepatic factor IX gene transfer in murine and canine models

Background

Adeno-associated virus (AAV) gene therapy vectors have shown the best outcomes in human clinical studies for the treatment of genetic diseases such as hemophilia. However, these pivotal investigations have also identified several challenges. For example, high vector doses are often used for hepatic gene transfer, and cytotoxic T lymphocyte responses against viral capsid may occur. Therefore, achieving therapy at reduced vector doses and other strategies to reduce capsid antigen presentation are desirable.

Methods

We tested several engineered AAV capsids for factor IX (FIX) expression for the treatment of hemophilia B by hepatic gene transfer. These capsids lack potential phosphorylation or ubiquitination sites, or had been generated through molecular evolution.

Results

AAV2 capsids lacking either a single lysine residue or 3 tyrosine residues directed substantially higher coagulation FIX expression in mice compared to wild-type sequence or other mutations. In hemophilia B dogs, however, expression from the tyrosine-mutant vector was merely comparable to historical data on AAV2. Evolved AAV2-LiC capsid was highly efficient in hemophilia B mice but lacked efficacy in a hemophilia B dog.

Conclusions

Several alternative strategies for capsid modification improve the in vivo performance of AAV vectors in hepatic gene transfer for correction of hemophilia. However, capsid optimization solely in mouse liver may not predict efficacy in other species and thus is of limited translational utility.

Electronic supplementary material

The online version of this article (doi:10.1186/s12967-017-1200-1) contains supplementary material, which is available to authorized users.

Dynamics of antigen presentation to transgene product-specific CD4+ T cells and of Treg induction upon hepatic AAV gene transfer

The tolerogenic hepatic microenvironment impedes clearance of viral infections but is an advantage in viral vector gene transfer, which often results in immune tolerance induction to transgene products. Although the underlying tolerance mechanism has been extensively studied, our understanding of antigen presentation to transgene product-specific CD4⁺ T cells remains limited. To address this, we administered hepatotropic adeno-associated virus (AAV8) vector expressing cytoplasmic ovalbumin (OVA) into wt mice followed by adoptive transfer of transgenic OVA-specific T cells. We find that that the liver-draining lymph nodes (celiac and portal) are the major sites of MHC II presentation of the virally encoded antigen, as judged by in vivo proliferation of DO11.10 CD4⁺ T cells (requiring professional antigen-presenting cells, e.g., macrophages) and CD4⁺CD25⁺FoxP3⁺ Treg induction. Antigen presentation in the liver itself contributes to activation of CD4⁺ T cells egressing from the liver. Hepatic-induced Treg rapidly disseminate through the systemic circulation. By contrast, a secreted OVA transgene product is presented in multiple organs, and OVA-specific Treg emerge in both the thymus and periphery. In summary, liver draining lymph nodes play an integral role in hepatic antigen presentation and peripheral Treg induction, which results in systemic regulation of the response to viral gene products.

Potential for cellular stress response to hepatic factor VIII expression from AAV vector

Hemophilia A and B are coagulation disorders resulting from the loss of functional coagulation factor VIII (FVIII) or factor IX proteins, respectively. Gene therapy for hemophilia with adeno-associated virus vectors has shown efficacy in hemophilia B patients. Although hemophilia A patients are more prevalent, the development of therapeutic adeno-associated virus vectors has been impeded by the size of the F8 cDNA and impaired secretion of FVIII protein. Further, it has been reported that over-expression of the FVIII protein induces endoplasmic reticulum stress and activates the unfolded protein response pathway both in vitro and in hepatocytes in vivo, presumably due to retention of misfolded FVIII protein within the endoplasmic reticulum. Engineering of the F8 transgene, including removal of the B domain (BDD-FVIII) and codon optimization, now allows for the generation of adeno-associated virus vectors capable of expressing therapeutic levels of FVIII. Here we sought to determine if the risks of inducing the unfolded protein response in murine hepatocytes extend to adeno-associated virus gene transfer. Although our data show a mild activation of unfolded protein response markers following F8 gene delivery at a certain vector dose in C57BL/6 mice, it was not augmented upon further elevated dosing, did not induce liver pathology or apoptosis, and did not impact FVIII immunogenicity.

Prevalence of neutralizing antibodies against liver-tropic adeno-associated virus serotype vectors in 100 healthy Chinese and its potential relation to body constitutions

Recombinant adeno-associated virus (rAAV) serotype 2, 3 and 8 vectors are the most promising liver-tropic AAV serotype vectors. Liver diseases are significant problems in China. However, to date, few studies on AAV neutralizing antibodies (Nabs) were working with the Chinese population or with the rAAV3 vectors. The present study aimed to determine the prevalence of Nabs in Chinese population against wild-type AAV2, AAV3 and AAV8 capsids as well as additional two AAV3 variants. In addition, we performed a preliminary analysis to investigate the potential influence of traditional Chinese medicine body constitutions on AAV Nabs. Our work demonstrated that the majority of healthy Chinese subjects were positive for AAV Nabs, with the order of AAV2>AAV3=AAVLK03>AAV8. There was no difference between: 1) AAV3 and its variants; 2) male and female subjects; and 3) different age cohorts (≤35, 36-50, and ≥51 years old). People in the Qi-deficiency constitution had significantly increased AAV8 Nabs than people in the Gentleness constitution. Our studies may have impact on the future clinical design of AAV-based gene therapy in the Chinese population.

Unique Roles of TLR9- and MyD88-Dependent and -Independent Pathways in Adaptive Immune Responses to AAV-Mediated Gene Transfer

The immune system represents a significant barrier to successful gene therapy with adeno-associated viral (AAV) vectors. In particular, adaptive immune responses to the viral capsid or the transgene product are of concern. The sensing of AAV by toll-like receptors (TLRs) TLR2 and TLR9 has been suggested to play a role in innate immunity to the virus and may also shape subsequent adaptive immune responses. Here, we investigated the functions of TLR2, TLR9 and the downstream signaling adaptor MyD88 in antibody and CD8+ T-cell responses. Antibody formation against the transgene product occurred largely independently of TLR signaling following gene transfer with AAV1 or AAV2 vectors, whereas loss of signaling through the TLR9-MyD88 pathway substantially reduced CD8+ T-cell responses. In contrast, MyD88 (but neither of the TLRs) regulated antibody responses to capsid. B cell-intrinsic MyD88 was required for the formation of anti-capsid IgG2c independently of vector serotype or route of administration. However, MyD88(-/-) mice instead produced anti-capsid IgG1 that emerged with delayed kinetics but nonetheless completely prevented in vivo readministration. We conclude that there are distinct roles for TLR9 and MyD88 in promoting adaptive immune responses to AAV-mediated gene transfer and that there are redundant MyD88-dependent and MyD88-independent mechanisms that stimulate neutralizing antibody formation against AAV.

Immune tolerance induction to factor IX through B cell gene transfer: TLR9 signaling delineates between tolerogenic and immunogenic B cells

A subset of patients with severe hemophilia B, the X-linked bleeding disorder resulting from absence of coagulation factor IX (FIX), develop pathogenic antibody responses during replacement therapy. These inhibitors block standard therapy and are often associated with anaphylactic reactions to FIX. Established clinical immune tolerance induction protocols often fail for FIX inhibitors. In a murine model of this immune complication, retrovirally transduced primary B cells expressing FIX antigen fused with immunoglobulin-G heavy chain prevented antibody formation to FIX and was also highly effective in desensitizing animals with preexisting response. In contrast, transplant of B cells that received the identical expression cassette via nucleofection of plasmid vector substantially heightened antibody formation against FIX, a response that could be blocked by toll-like receptor 9 (TLR9) inhibition. While innate responses to TLR4 activation or to retrovirus were minimal in B cells, plasmid DNA activated TLR9, resulting in CpG-dependent NF-κB activation/IL-6 expression and adaptor protein 3 dependent, CpG-independent induction of IFN-I. Neither response was seen in TLR9-deficient B cells. Therefore, TLR9 signaling in B cells, in particular in response to plasmid vector, is highly immunogenic and has to be avoided in design of tolerance protocols.

Role of the vector genome and underlying factor IX mutation in immune responses to AAV gene therapy for hemophilia B

Background

Self-complementary adeno-associated virus (scAAV) vectors have become a desirable vector for therapeutic gene transfer due to their ability to produce greater levels of transgene than single-stranded AAV (ssAAV). However, recent reports have suggested that scAAV vectors are more immunogenic than ssAAV. In this study, we investigated the effects of a self-complementary genome during gene therapy with a therapeutic protein, human factor IX (hF.IX).

Methods

Hemophilia B mice were injected intramuscularly with ss or scAAV1 vectors expressing hF.IX. The outcome of gene transfer was assessed, including transgene expression as well as antibody and CD8⁺ T cell responses to hF.IX.

Results

Self-complementary AAV1 vectors induced similar antibody responses (which eliminated systemic hF.IX expression) but stronger CD8⁺ T cell responses to hF.IX relative to ssAAV1 in mice with F9 gene deletion. As a result, hF.IX-expressing muscle fibers were effectively eliminated in scAAV-treated mice. In contrast, mice with F9 nonsense mutation (late stop codon) lacked antibody or T cell responses, thus showing long-term expression regardless of the vector genome.

Conclusions

The nature of the AAV genome can impact the CD8⁺ T cell response to the therapeutic transgene product. In mice with endogenous hF.IX expression, however, this enhanced immunogenicity did not break tolerance to hF.IX, suggesting that the underlying mutation is a more important risk factor for transgene-specific immunity than the molecular form of the AAV genome.

Improved Adeno-associated Viral Gene Transfer to Murine Glioma

Glioblastoma (GBM) is a deadly primary brain tumor. Current treatment, consisting of surgical removal of the tumor mass followed by chemotherapy and/or radiotherapy, does not significantly prolong survival. Gene therapies for GBM are being developed in clinical trials, for example using adenoviral vectors. While adeno-associated virus (AAV) represents an alternative vector system, limited gene transfer to glioma cells has hampered its use. Here, we evaluated newly emerged variants of AAV capsid for gene delivery to murine glioma. We tested a mutant AAV2 capsid devoid of 3 surface-exposed tyrosine residues, AAV2 (Y444-500-730F), and a "shuffed" capsid (ShH19, containing sequences from several serotypes) that had previously been selected for enhanced glial gene delivery. AAV2 (Y-F) and ShH19 showed improved transduction of murine glioma GL261 cells in vitro by 2- to 6-fold, respectively, over AAV2. While AAV2 gene transfer to GL261 cells in established tumors in brains of syngeneic mice was undetectable, intratumoral injection of AAV2 (Y-F) or ShH19 resulted in local transduction of approximately 10% of tumor cells. In addition, gene transfer to neurons adjacent to the tumor was observed, while microglia were rarely transduced. Use of self-complementary vectors further increased transduction of glioma cells. Together, the data demonstrate the potential for improved AAV-based gene therapy for glioma using recently developed capsid variants.

Identification of adeno-associated viral vectors suitable for intestinal gene delivery and modulation of experimental colitis

Effective gene transfer with sustained gene expression is an important adjunct to the study of intestinal inflammation and future therapy in inflammatory bowel disease. Recombinant adeno-associated virus (AAV) vectors are ideal for gene transfer and long-term transgene expression. The purpose of our study was to identify optimal AAV pseudotypes for transduction of the epithelium in the small intestine and colon, which could be used for studies in experimental colitis. The tropism and transduction efficiencies of AAV pseudotypes 1-10 were examined in murine small intestine and colon 8 wk after administration by real-time PCR and immunohistochemistry. The clinical and histopathological effects of IL-10-mediated intestinal transduction delivered by AAVrh10 were examined in the murine IL-10⁻/⁻ enterocolitis model. Serum IL-10 levels and IL-10 expression were followed by ELISA and real-time PCR, respectively. AAV pseudotypes 4, 7, 8, 9, and 10 demonstrated optimal intestinal transduction. Transgene expression was sustained 8 wk after administration and was frequently observed in enteroendocrine cells. Long-term IL-10 gene expression and serum IL-10 levels were observed following AAV transduction in an IL-10-/- model of enterocolitis. Animals treated with AAVrh10-IL-10 had lower disease activity index scores, higher colon weight-to-length ratios, and lower microscopic inflammation scores. This study identifies novel AAV pseudotypes with small intestine and colon tropism and sustained transgene expression capable of modulating mucosal inflammation in a murine model of enterocolitis.

High Efficiency Ex Vivo Gene Transfer to Primary Murine B Cells Using Plasmid or Viral Vectors

Primary autologous B-lymphocytes, following ex vivo gene transfer and re-implantation, have been successfully utilized to prevent autoimmune disease and adaptive responses to therapeutic proteins in several animal models. However, efficient gene transfer to primary B cells requires use of retroviral vectors, which increase the risk of insertional mutagenesis. Here, we evaluated several alternative gene transfer approaches. Resting splenic B cells were purified and activated with LPS, and ex vivo GFP gene transfer was performed by means of nucleofection, lipofectamine, adenoviral infection, or murine retroviral infection. The Adenoviral (Ad) vectors were added to B cell cultures with or without calcium phosphate precipitation. For transfection and nucleofection, naked plasmid DNA was utilized. Nucleofection technology represents a modified electroporation technique for effective transfer of nucleic acids to the nucleus and thus enhances the efficiency of transfer particularly for primary cells. Efficiency of ex vivo gene transfer was determined by flow cytometry using GFP, CD19, and a vital dye as markers. Nucleofection yielded the highest level of gene transfer with 60-65% of B cells being GFP+. Efficiencies were 30-35% for retrovirus, 20% for Ad5/11, 15% for Ad5/35, and 5% for lipofectamine-mediated transfection. Calcium phosphate precipitation increased efficiencies for Ad vectors to 30% (Ad5/11) and 25% (Ad5/35). Lipofectamin caused the greatest cell death at 80%, followed by nucleofection (35%), and viral vector (10-15% in each case). For all methods, gene transfer efficiencies were nearly identical for B cells from C57BL/6 or C3H/HeOuJ mice. In conclusion, recent advances in gene transfer technologies provide alternatives to retroviral vectors for primary B cells. If stable gene transfer is desired, non-integrating vector systems may be combined with transposon- or phage integrase-based systems or future site-specific systems to achieve integration into the host B cell genome.

High-efficiency transduction and correction of murine hemophilia B using AAV2 vectors devoid of multiple surface-exposed tyrosines

Elimination of specific surface-exposed single tyrosine (Y) residues substantially improves hepatic gene transfer with adeno-associated virus type 2 (AAV2) vectors. Here, combinations of mutations in the seven potentially relevant Y residues were evaluated for further augmentation of transduction efficiency. These mutant capsids packaged viral genomes to similar titers and retained infectivity. A triple-mutant (Y444+500+730F) vector consistently had the highest level of in vivo gene transfer to murine hepatocytes, approximately threefold more efficient than the best single-mutants, and ~30-80-fold higher compared with the wild-type (WT) AAV2 capsids. Improvement of gene transfer was similar for both single-stranded AAV (ssAAV) and self-complementary AAV (scAAV) vectors, indicating that these effects are independent of viral second-strand DNA synthesis. Furthermore, Y730F and triple-mutant vectors provided a long-term therapeutic and tolerogenic expression of human factor IX (hF.IX) in hemophilia B (HB) mice after administration of a vector dose that only results in subtherapeutic and transient expression with WT AAV2 encapsidated vectors. In summary, introduction of multiple tyrosine-mutations into the AAV2 capsid results in vectors that yield at least 30-fold improvement of transgene expression, thereby lowering the required therapeutic dose and potentially vector-related immunogenicity. Such vectors should be attractive for treatment of hemophilia and other genetic diseases.

AAV3-mediated transfer and expression of the pyruvate dehydrogenase E1 alpha subunit gene causes metabolic remodeling and apoptosis of human liver cancer cells

Most cancers rely disproportionately on glycolysis for energy even in the presence of adequate oxygen supply, a condition known as "aerobic glycolysis", or the Warburg effect. Pharmacological reversal of the Warburg effect has been shown to cause selective apoptosis of tumor cells, presumably by stimulating mitochondrial respiratory chain activity and production of reactive oxygen species that, in turn, induce a caspase-mediated series of reactions leading to cell death. We reasoned that a similar effect on tumor cells might result from up-regulation of the E1alpha subunit gene (pda1) of the pyruvate dehydrogenase complex (PDC) that catalyzes the rate-limiting step in aerobic glucose oxidation and thus plays a major role in the control of oxidative phosphorylation. To test this postulate, we employed a self-complementary adeno-associated virus (scAAV)-based delivery and expression system for targeting pda1 to the mitochondria of primary cultures of human hepatoblastoma (HB) and hepatocellular carcinoma (HCC) cells. Serotypes 1-10 scAAV vectors that included enhanced green fluorescent (egfp) reporter gene driven by either cytomegalovirus (CMV) or chicken beta-actin (CBA) promoters were analyzed for transduction ability of HB (Huh-6) and HCC (Huh-7 and HepG2) cell lines and primary cultures of normal human hepatocytes. Serotype 3 scAAV-egfp (scAAV3-egfp) vector was the most efficient and transduced up to 90% of cells. We limited the transgene expression primarily to liver cancer cells by generating scAAV3 vectors that contained the human alpha-fetoprotein promoter (AFP)-driven reporter gene (scAAV3.AFP-egfp) and the potentially therapeutic gene scAAV3.AFP-pda1. Infection of Huh-6 cells by the scAAV3.AFP-pda1 vector increased protein expression of E1alpha, PDC catalytic activity, and late-stage apoptotic cell death. Apoptosis was also associated with increased protein expression of Bcl-X/S, an early marker of apoptosis, and release of cytochrome c into the cytosol of infected HB cells. These data indicate that molecular targeting of mitochondrial oxidative metabolism in liver cancer cells by AAV3-mediated delivery of pda1 holds promise as a novel and effective therapeutic approach for human hepatic tumors.

Recombinant adeno-associated virus serotype 9 leads to preferential cardiac transduction in vivo

Heart disease is often the end result of inherited genetic defects, which may potentially be treatable using a gene-transfer approach. Recombinant adeno-associated virus (rAAV)-mediated gene delivery has emerged as a realistic method for the treatment of such disorders. Here, we demonstrate and compare the natural affinity of specific AAV serotype capsids for transduction of cardiac tissue. We compared the previously accepted optimal rAAV serotype for transduction of skeletal muscle, rAAV2/1, with rAAV2/8 and the newer rAAV2/9 vectors carrying the CMV-lacZ construct in their respective abilities to transcend vasculature and transduce myocardium following intravenous delivery of 1x10(11) vector genomes in neonatal mice. We found that both rAAV2/8 and rAAV2/9 are able to transduce myocardium at approximately 20- and 200-fold (respectively) higher levels than rAAV2/1. Biodistribution analysis revealed that rAAV2/9 and rAAV2/8 demonstrate similar behavior in extracardiac tissue. Vector genome quantification showed an increase in genome copy numbers in cardiac tissue for several weeks following administration, which corresponds to expression data. In addition, we intravenously administered 1x10(11) vector genomes of rAAV2/9-CMV-lacZ into adult mice and achieved an expression biodistribution profile similar to that found following delivery to newborns. Although higher doses of virus will be necessary to approach those levels observed following neonatal injections, adult myocardium is also readily transduced by rAAV2/9. Finally, we have demonstrated physiological disease correction by AAV9 gene transfer in a mouse model of Pompe disease via ECG tracings and that intravenous delivery of the same vector preferentially transduces cardiac tissue in nonhuman primates.

Evaluation of primitive murine hematopoietic stem and progenitor cell transduction in vitro and in vivo by recombinant adeno-associated virus vector serotypes 1 through 5

Conflicting data exist on hematopoietic cell transduction by AAV serotype 2 (AAV2) vectors, and additional AAV serotype vectors have not been evaluated for their efficacy in hematopoietic stem/progenitor cell transduction. We evaluated the efficacy of conventional, single-stranded AAV serotype vectors 1 through 5 in primitive murine hematopoietic stem/progenitor cells in vitro as well as in vivo. In progenitor cell assays using Sca1+ c-kit+ Lin- hematopoietic cells, 9% of the colonies in cultures infected with AAV1 expressed the transgene. Coinfection of AAV1 with self-complementary AAV vectors carrying the gene for T cell protein tyrosine phosphatase (scAAV-TC-PTP) increased the transduction efficiency to 24%, indicating that viral secondstrand DNA synthesis is a rate-limiting step. This was further corroborated by the use of scAAV vectors, which bypass this requirement. In bone marrow transplantation studies involving lethally irradiated syngeneic mice, Sca1+ c-kit+ Lin- cells coinfected with AAV1 +/- scAAV-TC-PTP vectors led to transgene expression in 2 and 7.5% of peripheral blood (PB) cells, respectively, 6 months posttransplantation. In secondary transplantation experiments, 7% of PB cells and 3% of bone marrow (BM) cells expressed the transgene 6 months posttransplantation. Approximately 21% of BM-derived colonies harbored the proviral DNA sequences in integrated forms. These results document that AAV1 is thus far the most efficient vector in transducing primitive murine hematopoietic stem/progenitor cells. Further studies involving scAAV genomes and hematopoietic cell-specific promoters should further augment the transduction efficiency of AAV1 vectors, which should have implications in the optimal use of these vectors in hematopoietic stem cell gene therapy.

Sustained correction of glycogen storage disease type II using adeno-associated virus serotype 1 vectors

Glycogen storage disease type II (GSDII) is caused by a lack of functional lysosomal acid alpha-glucosidase (GAA). Affected individuals store glycogen in lysosomes beginning during gestation, ultimately resulting in fatal hypertrophic cardiomyopathy and respiratory failure. We have assessed the utility of recombinant adeno-associated virus (rAAV) vectors to restore GAA activity in vivo in a mouse model of GSDII (Gaa(-/-)). A single systemic administration of a rAAV serotype 1 (rAAV1) vector to neonate animals resulted in restored cardiac GAA activity to 6.4 times the normal level (mean=641+/-190% of normal (Gaa(+/+)) levels with concomitant glycogen clearance) at 11 months postinjection. Greater than 20% of normal levels of GAA activity were also observed in the diaphragm and quadriceps muscles. Furthermore, functional correction of the soleus skeletal muscle was also observed compared to age-matched untreated Gaa(-/-) control animals. These results demonstrate that rAAV1 vectors can mediate sustained therapeutic levels of correction of both skeletal and cardiac muscles in a model of fatal cardiomyopathy and muscular dystrophy.

A new method for recombinant adeno-associated virus vector delivery to murine diaphragm

Genetically modified mice are important models for evaluation of potential gene therapies for human diseases. However, their small size often precludes the use of clinically feasible methods for vector delivery, therefore, alternative methods must be used. We have developed a gel-based method for delivery of recombinant adeno-associated virus vectors to the mouse diaphragm, an important target organ for many myopathic diseases. We hypothesized that delivery of vectors in a viscous solution would increase transduction by providing a longer exposure time to target cells. We demonstrate that gel-mediated delivery of rAAV serotypes 1, 2, and 5 results in higher transduction efficiencies than free vectors alone when administered in vivo to mouse diaphragms. We further establish greater tropism of rAAV1 vectors for the diaphragm compared to serotypes 2 and 5. This report describes a novel method for efficient delivery of rAAV vectors to the mouse diaphragm and is the first demonstration of gene transfer to the diaphragm using recombinant adeno-associated virus vectors.

Rescue of enzyme deficiency in embryonic diaphragm in a mouse model of metabolic myopathy: Pompe disease

Several human genetic diseases that affect striated muscle have been modeled by creating knockout mouse strains. However, many of these are perinatal lethal mutations that result in death from respiratory distress within hours after birth. As the diaphragm muscle does not contract until birth, the sudden increase in diaphragm activity creates permanent injury to the muscle causing it to fail to meet respiratory demands. Therefore, the impact of these mutations remains hidden throughout embryonic development and early death prevents investigators from performing detailed studies of other striated muscle groups past the neonatal stage. Glycogen storage disease type II (GSDII), caused by a deficiency in acid alpha-glucosidase (GAA), leads to lysosomal accumulation of glycogen in all cell types and abnormal myofibrillogenesis in striated muscle. Contractile function of the diaphragm muscle is severely affected in both infantile-onset and late-onset individuals, with death often resulting from respiratory failure. The knockout mouse model of GSDII survives well into adulthood despite the gradual weakening of all striated muscle groups. Using this model, we investigated the delivery of recombinant adeno-associated virus (rAAV) vectors encoding the human GAA cDNA to the developing embryo. Results indicate specific high-level transduction of diaphragm tissue, leading to activity levels up to 10-fold higher than normal and restoration of normal contractile function. Up to an estimated 50 vector copies per diploid genome were quantified in treated diaphragms. Histological glycogen staining of treated diaphragms revealed prevention of lysosomal glycogen accumulation in almost all fibers when compared with untreated controls. This method could be employed with disease models where specific rescue of the diaphragm would allow for increased survival and thus further investigation into the impact of the gene deletion on other striated muscle groups.

Dual vectors expressing murine factor VIII result in sustained correction of hemophilia A mice

Hemophilia A is a sex-linked disorder that results from a deficiency of functional factor VIII and is currently treated by protein replacement therapies. Within the past decade, gene therapy efforts have come to the forefront of novel therapeutics. In this work, a dual-vector approach was employed in which recombinant adeno-associated viral (rAAV) vectors expressing the heavy and light chains of the murine factor VIII gene were delivered either intramuscularly or intravenously to a mouse model of hemophilia A. From in vitro work, it was determined that coinfection with both vectors is required as heterodimerization of the heavy and light chains occurs intracellularly. In vivo, therapeutic levels of factor VIII expression were achieved throughout the duration of the study (22 weeks). Intravenous and intramuscular delivery resulted in a maximal average expression of 31.4 +/- 6.4 and 29 +/- 6.5% of normal murine factor VIII levels, respectively. Western blots of cryoprecipitate as well as immunostaining of injection sites with an anti-murine factor VIII light chain antibody also confirmed the expression of factor VIII. Because the murine form of the gene was used in the mouse model, less than 1 Bethesda unit of inhibitors was noted. This work demonstrates the feasibility of using rAAV vectors for the long-term treatment of hemophilia A.

Production and purification of serotype 1, 2, and 5 recombinant adeno-associated viral vectors

Recombinant adeno-associated viral (rAAV) vectors based on serotype 2 are currently being evaluated most extensively in animals and human clinical trials. rAAV vectors constructed from other AAV serotypes (serotypes 1, 3, 4, 5, and 6) can transduce certain tissues more efficiently and with different specificity than rAAV2 vectors in animal models. Here, we describe reagents and methods for the production and purification of AAV2 inverted terminal repeat-containing vectors pseudotyped with AAV1 or AAV5 capsids. To facilitate pseudotyping, AAV2rep/AAV1cap and AAV2rep/AAV5cap helper plasmids were constructed in an adenoviral plasmid backbone. The resultant plasmids, pXYZ1 and pXYZ5, were used to produce rAAV1 and rAAV5 vectors, respectively, by transient transfection. Since neither AAV5 nor AAV1 binds to the heparin affinity chromatography resin used to purify rAAV2 vectors, purification protocols were developed based on anion-exchange chromatography. The purified vector stocks are 99% pure with titers of 1 x 10(12) to 1 x 10(13)vector genomes/ml.

Improved method of recombinant AAV2 delivery for systemic targeted gene therapy

A major hurdle in most current gene therapy modalities is the ability to transduce target tissues at very high efficiencies that ultimately lead to therapeutic levels of transgene expression. We have developed a novel method of recombinant adeno-associated virus 2 (rAAV) delivery that results in increased vector transduction efficiencies using microspheres reversibly conjugated to rAAV vectors. We hypothesize that conjugation to microspheres should result in a higher effective concentration of vector as well as longer relative exposure time of vector to target cells as it moves through the tissue vasculature. In vitro experiments demonstrate that the same level of transduction seen with free vector can be achieved using 1% of vector when conjugated to microspheres. In addition, using magnetic microspheres, the region of infection can be targeted. In vivo, we demonstrate that microsphere-mediated delivery of rAAV vector results in higher transduction efficiencies than delivery with free vector alone when administered either intramuscularly or intravenously. Furthermore, we demonstrate targeting of transgene expression to specific tissues by retention of microsphere-bound vector in the capillary bed. These studies demonstrate a novel method to deliver rAAV vectors more effectively that could prove to be a successful alternative mode of virus-mediated human gene therapy.

Correction of the enzymatic and functional deficits in a model of Pompe disease using adeno-associated virus vectors

Pompe disease is a lysosomal storage disease caused by the absence of acid alpha-1,4 glucosidase (GAA). The pathophysiology of Pompe disease includes generalized myopathy of both cardiac and skeletal muscle. We sought to use recombinant adeno-associated virus (rAAV) vectors to deliver functional GAA genes in vitro and in vivo. Myotubes and fibroblasts from Pompe patients were transduced in vitro with rAAV2-GAA. At 14 days postinfection, GAA activities were at least fourfold higher than in their respective untransduced controls, with a 10-fold increase observed in GAA-deficient myotubes. BALB/c and Gaa(-/-) mice were also treated with rAAV vectors. Persistent expression of vector-derived human GAA was observed in BALB/c mice up to 6 months after treatment. In Gaa(-/-) mice, intramuscular and intramyocardial delivery of rAAV2-Gaa (carrying the mouse Gaa cDNA) resulted in near-normal enzyme activities. Skeletal muscle contractility was partially restored in the soleus muscles of treated Gaa(-/-) mice, indicating the potential for vector-mediated restoration of both enzymatic activity and muscle function. Furthermore, intramuscular treatment with a recombinant AAV serotype 1 vector (rAAV1-Gaa) led to nearly eight times normal enzymatic activity in Gaa(-/-) mice, with concomitant glycogen clearance as assessed in vitro and by proton magnetic resonance spectroscopy.

High-titer recombinant adeno-associated virus production utilizing a recombinant herpes simplex virus type I vector expressing AAV-2 Rep and Cap

Recombinant adeno-associated virus type 2 (rAAV) vectors have recently been used to achieve long-term, high level transduction in vivo. Further development of rAAV vectors for clinical use requires significant technological improvements in large-scale vector production. In order to facilitate the production of rAAV vectors, a recombinant herpes simplex virus type I vector (rHSV-1) which does not produce ICP27, has been engineered to express the AAV-2 rep and cap genes. The optimal dose of this vector, d27.1-rc, for AAV production has been determined and results in a yield of 380 expression units (EU) of AAV-GFP produced from 293 cells following transfection with AAV-GFP plasmid DNA. In addition, d27.1-rc was also efficient at producing rAAV from cell lines that have an integrated AAV-GFP provirus. Up to 480 EU/cell of AAV-GFP could be produced from the cell line GFP-92, a proviral, 293 derived cell line. Effective amplification of rAAV vectors introduced into 293 cells by infection was also demonstrated. Passage of rAAV with d27. 1-rc results in up to 200-fold amplification of AAV-GFP with each passage after coinfection of the vectors. Efficient, large-scale production (>109 cells) of AAV-GFP from a proviral cell line was also achieved and these stocks were free of replication-competent AAV. The described rHSV-1 vector provides a novel, simple and flexible way to introduce the AAV-2 rep and cap genes and helper virus functions required to produce high-titer rAAV preparations from any rAAV proviral construct. The efficiency and potential for scalable delivery of d27.1-rc to producer cell cultures should facilitate the production of sufficient quantities of rAAV vectors for clinical application.

Recombinant adeno-associated virus purification using novel methods improves infectious titer and yield

Conventional methods for rAAV purification that are based on cesium chloride ultracentrifugation have often produced vector preparations of variable quality and resulted in significant loss of particle infectivity. We report here several novel purification strategies that involve the use of non-ionic iodixanol gradients followed by ion exchange or heparin affinity chromatography by either conventional or HPLC columns. These methods result in more than 50% recovery of rAAV from a crude lysate and routinely produce vector that is more than 99% pure. More importantly, the new purification procedures consistently produce rAAV stocks with particle-to-infectivity ratios of less than 100, which is significantly better than conventional methods. The new protocol increases the overall yield of infectious rAAV by at least 10-fold and allows for the complete purification of rAAV in 1 working day. Several of these methods should also be useful for large-scale production.

Biochemical characterization of adeno-associated virus rep68 DNA helicase and ATPase activities

The adeno-associated virus (AAV) nonstructural proteins Rep68 and Rep78 are site-specific DNA binding proteins, ATP-dependent site-specific endonucleases, helicases, and ATPases. These biochemical activities are required for viral DNA replication and control of viral gene expression. In this study, we characterized the biochemical properties of the helicase and ATPase activities of homogeneously pure Rep68. The enzyme exists as a monomer in solution at the concentrations used in this study (<380 nM), as judged by its mobility in sucrose density gradients. Using a primed single-stranded (ss) circular M13 substrate, the helicase activity had an optimum pH of 7 to 7.5, an optimum temperature of 45 degreesC, and an optimal divalent-cation concentration of 5 mM MgCl2. Several nucleoside triphosphates could serve as cofactors for Rep68 helicase activity, and the order of preference was ATP = GTP > CTP = dATP > UTP > dGTP. The Km values for ATP in both the DNA helicase reaction and the site-specific trs endonuclease reaction were essentially the same, approximately 180 microM. Both reactions were sigmoidal with respect to ATP concentration, suggesting that a dimer or higher-order multimer of Rep68 is necessary for both DNA helicase activity and terminal resolution site (trs) nicking activity. Furthermore, when the enzyme itself was titrated in the trs endonuclease and ATPase reactions, both activities were second order with respect to enzyme concentration. This suggests that a dimer of Rep68 is the active form for both the ATPase and nicking activities. In contrast, DNA helicase activity was linear with respect to enzyme concentration. When bound to ssDNA, the enzyme unwound the DNA in the 3'-to-5' direction. DNA unwinding occurred at a rate of approximately 345 bp per min per monomeric enzyme molecule. The ATP turnover rate was approximately 30 to 50 ATP molecules per min per enzyme molecule. Surprisingly, the presence of DNA was not required for ATPase activity. We estimated that Rep translocates processively for more than 1,300 bases before dissociating from its substrate in the absence of any accessory proteins. DNA helicase activity was not significantly stimulated by substrates that have the structure of a replication fork and contain either a 5' or 3' tail. Rep68 binds only to ssDNA, as judged by inhibition of the DNA helicase reaction with ss or double-stranded (ds) DNA. Consistent with this observation, no helicase activity was detected on blunt-ended ds oligonucleotide substrates unless they also contained an ss 3' tail. However, if a blunt-ended ds oligonucleotide contained the 22-bp Rep binding element sequence, Rep68 was capable of unwinding the substrate. This means that Rep68 can function both as a conventional helicase for strand displacement synthesis and as a terminal-repeat-unwinding protein which catalyzes the conversion of a duplex end to a hairpin primer. Thus, the properties of the Rep DNA helicase activity suggest that Rep is involved in all three of the key steps in AAV DNA replication: terminal resolution, reinitiation, and strand displacement.

Cellular proteins required for adeno-associated virus DNA replication in the absence of adenovirus coinfection

We previously reported the development of an in vitro adeno-associated virus (AAV) DNA replication system. The system required one of the p5 Rep proteins encoded by AAV (either Rep78 or Rep68) and a crude adenovirus (Ad)-infected HeLa cell cytoplasmic extract to catalyze origin of replication-dependent AAV DNA replication. However, in addition to fully permissive DNA replication, which occurs in the presence of Ad, AAV is also capable of partially permissive DNA replication in the absence of the helper virus in cells that have been treated with genotoxic agents. Limited DNA replication also occurs in the absence of Ad during the process of establishing a latent infection. In an attempt to isolate uninfected extracts that would support AAV DNA replication, we discovered that HeLa cell extracts grown to high density can occasionally display as much in vitro replication activity as Ad-infected extracts. This finding confirmed previous genetic analyses which suggested that no Ad-encoded proteins were absolutely essential for AAV DNA replication and that the uninfected extracts should be useful for studying the differences between helper-dependent and helper-independent AAV DNA replication. Using specific chemical inhibitors and monoclonal antibodies, as well as the fractionation of uninfected HeLa extracts, we identified several of the cellular enzymes involved in AAV DNA replication. They were the single-stranded DNA binding protein, replication protein A (RFA), the 3' primer binding complex, replication factor C (RFC), and proliferating cell nuclear antigen (PCNA). Consistent with the current model for AAV DNA replication, which requires only leading-strand DNA synthesis, we found no requirement for DNA polymerase alpha-primase. AAV DNA replication could be reconstituted with purified Rep78, RPA, RFC, and PCNA and a phosphocellulose chromatography fraction (IIA) that contained DNA polymerase activity. As both RFC and PCNA are known to be accessory proteins for polymerase delta and epsilon, we attempted to reconstitute AAV DNA replication by substituting either purified polymerase delta or polymerase epsilon for fraction IIA. These attempts were unsuccessful and suggested that some novel cellular protein or modification was required for AAV DNA replication that had not been previously identified. Finally, we also further characterized the in vitro DNA replication assay and demonstrated by two-dimensional (2D) gel electrophoresis that all of the intermediates commonly seen in vivo are generated in the in vitro system. The only difference was an accumulation of single-stranded DNA in vivo that was not seen in vitro. The 2D data also suggested that although both Rep78 and Rep68 can generate dimeric intermediates in vitro, Rep68 is more efficient in processing dimers to monomer duplex DNA. Regardless of the Rep that was used in vitro, we found evidence of an interaction between the elongation complex and the terminal repeats. Nicking at the terminal repeats of a replicating molecule appeared to be inhibited until after elongation was complete.

Identification of linear DNA sequences that specifically bind the adeno-associated virus Rep protein

We have used baculovirus-expressed Rep68 that has been purified to homogeneity to reexamine the binding properties of the Rep protein. We find that Rep68 is capable of binding to a linear DNA sequence that is contained within a 25-bp sequence of the A stem of the adeno-associated virus (AAV) terminal repeat proximal to the B and C palindromes. This has been shown conclusively by demonstrating that Rep68 could specifically bind to a synthetic oligonucleotide containing the 25-bp region in the absence of the other sequences within the terminal repeat. Rep78 was also capable of binding the A stem recognition element, as demonstrated by the fact that a DNA affinity column containing the 25-bp sequence can be used to purify Rep78. The ability to recognize the linear DNA sequence within the A stem provides a mechanism by which the Rep protein can be oriented on the terminal repeat so that only the correct strand is cut at the terminal resolution site (trs site) during terminal resolution. In addition, computer analysis suggests that sequences similar to the A stem element are present within the three AAV promoter regions. Electrophoretic mobility shift experiments clearly demonstrate that the p5 promoter contains a Rep binding sequence. DNase protection experiments indicate that the Rep binding sequence within the p5 promoter is located between the YY1 initiator sequence and the TATA binding site. This position immediately suggests a mechanism by which the Rep protein could act as a repressor or a transactivator of p5 transcription by interacting with either YY1 or TBP. In addition, gel shift experiments suggest that the p19 promoter also contains a Rep binding site. The presence of Rep binding sites upstream of both promoters suggests that these sites may be involved in coordinate regulation of AAV transcription. In addition, we have identified a heterologous Rep binding sequence within pBR322 DNA. A comparison of the sequences within the A stem, p5, and pBR322 binding sites suggests that a repeating GAGC motif is at least part of the Rep recognition sequence. In the accompanying report (D. M. McCarty, J. H. Ryan, S. Zolutukhin, X. Zhou, and N. Muzyczka, J. Virol. 68:4998-5006, 1994), we examine the relative affinity of Rep to the A stem site and the complete terminal repeat. Finally, we also have reexamined the ability of Rep68 and Rep78 to cut at the trs site in substrates that do not contain the B and C palindromes or any apparent secondary structure.(ABSTRACT TRUNCATED AT 400 WORDS)

In vitro replication of adeno-associated virus DNA

The study of eukaryotic viral DNA replication in vitro has led to the identification of cellular enzymes involved in DNA replication. Adeno-associated virus (AAV) is distinct from previously reported systems in that it is believed to replicate entirely by leading-strand DNA synthesis and requires coinfection with adenovirus to establish completely permissive replication. In previous work, we demonstrated that two of the AAV nonstructural proteins, Rep78 and -68, are site-specific endonucleases and DNA helicases that are capable of resolving covalently closed AAV termini, a key step in AAV DNA replication. We have now cloned the AAV nonstructural proteins Rep78, Rep68, and Rep52 in the baculovirus expression system. Using the baculovirus-expressed proteins, we have developed an efficient in vitro AAV DNA replication system which mimics the in vivo behavior of AAV in every respect. With no-end AAV DNA as the starting substrate, the reaction required an adenovirus-infected cell extract and the presence of either Rep78 or Rep68. Rep52, as expected, did not support DNA replication. A mutant in the AAV terminal resolution site (trs) was defective for DNA replication in the in vitro assay. Little, if any, product was formed in the absence of the adenovirus-infected HeLa cell extract. In general, uninfected HeLa extracts were less efficient in supporting AAV DNA replication than adenovirus-infected extracts. Thus, the requirement for adenovirus infection in vivo was partially duplicated in vitro. The reduced ability of uninfected HeLa extracts to support complete DNA replication was not due to a defect in terminal resolution but rather to a defect in the reinitiation reaction or in elongation. Rep78 produced a characteristic monomer-dimer pattern of replicative intermediates, but surprisingly, Rep68 produced little, if any, dimer replicative form. The reaction had a significant lag (30 min) before incorporation of 32P-deoxynucleoside triphosphate could be detected in DpnI-resistant monomer replicative form and was linear for at least 4 h after the lag. The rate of incorporation in the reaction was comparable to that in the simian virus 40 in vitro system. Replication of the complete AAV DNA molecule was demonstrated by the following criteria. (i) Most of the monomer and dimer product DNAs were completely resistant to digestion with DpnI. (ii) Virtually all of the starting substrate was converted to heavy-light or heavy-heavy product DNA in the presence of bromo-dUTP when examined on CsCl density gradients.(ABSTRACT TRUNCATED AT 400 WORDS)