Assessing the potential for AAV vector genotoxicity in a murine model.

Gene transfer using adeno-associated virus (AAV) vectors has great potential for treating human disease. Recently, questions have arisen about the safety of AAV vectors, specifically, whether integration of vector DNA in transduced cell genomes promotes tumor formation. This study addresses these questions with high-dose liver-directed AAV-mediated gene transfer in the adult mouse as a model (80 AAV-injected mice and 52 controls). After 18 months of follow-up, AAV-injected mice did not show a significantly higher rate of hepatocellular carcinoma compared with controls. Tumors in mice treated with AAV vectors did not have significantly different amounts of vector DNA compared with adjacent normal tissue. A novel high-throughput method for identifying AAV vector integration sites was developed and used to clone 1029 integrants. Integration patterns in tumor tissue and adjacent normal tissue were similar to each other, showing preferences for active genes, cytosine-phosphate-guanosine islands, and guanosine/cytosine-rich regions. [corrected] Gene expression data showed that genes near integration sites did not show significant changes in expression patterns compared with genes more distal to integration sites. No integration events were identified as causing increased oncogene expression. Thus, we did not find evidence that AAV vectors cause insertional activation of oncogenes and subsequent tumor formation.

Pharmacological modulation of humoral immunity in a nonhuman primate model of AAV gene transfer for hemophilia B.

Liver gene transfer for hemophilia B has shown very promising results in recent clinical studies. A potential complication of gene-based treatments for hemophilia and other inherited disorders, however, is the development of neutralizing antibodies (NAb) against the therapeutic transgene. The risk of developing NAb to the coagulation factor IX (F.IX) transgene product following adeno-associated virus (AAV)-mediated hepatic gene transfer for hemophilia is small but not absent, as formation of inhibitory antibodies to F.IX is observed in experimental animals following liver gene transfer. Thus, strategies to modulate antitransgene NAb responses are needed. Here, we used the anti-B cell monoclonal antibody rituximab (rtx) in combination with cyclosporine A (CsA) to eradicate anti-human F.IX NAb in rhesus macaques previously injected intravenously with AAV8 vectors expressing human F.IX. A short course of immunosuppression (IS) resulted in eradication of anti-F.IX NAb with restoration of plasma F.IX transgene product detection. In one animal, following IS anti-AAV6 antibodies also dropped below detection, allowing for successful AAV vector readministration and resulting in high levels (60% or normal) of F.IX transgene product in plasma. Though the number of animals is small, this study supports for the safety and efficacy of B cell-targeting therapies to eradicate NAb developed following AAV-mediated gene transfer.

AAV-1-mediated gene transfer to skeletal muscle in humans results in dose-dependent activation of capsid-specific T cells.

In a clinical trial for adeno-associated virus serotype 1 (AAV-1)-mediated gene transfer to muscle for lipoprotein lipase (LPL) deficiency, 1 subject from the high-dose cohort experienced a transient increase in the muscle enzyme creatine phosphokinase (CPK) 4 weeks after gene transfer. Simultaneously, after an initial downward trend consistent with expression of LPL, plasma triglyceride levels returned to baseline. We characterized B- and T-cell responses to the vector and the transgene product in the subjects enrolled in this study. IFN-gamma enzyme-linked immunosorbent spot (ELISpot) and intracellular cytokine staining assays performed on peripheral blood mononuclear cells (PBMCs) from the subject who experienced the CPK elevation showed the activation of capsid-specific CD4(+) and CD8(+) T cells. Four of 8 subjects had detectable T-cell responses to capsid with dose-dependent kinetics of appearance. Subjects with detectable T-cell responses to capsid also had higher anti-AAV-1 IgG3 antibody titer. No subject developed B- or T-cell responses to the LPL transgene product. These findings suggest that T-cell responses directed to the AAV-1 capsid are dose-dependent. Whether they also limit the duration of expression of the transgene at higher doses is unclear, and will require additional analyses at later time points.

Diverse IgG subclass responses to adeno-associated virus infection and vector administration.

Humoral immune responses occur following exposure to Adeno-associated virus (AAV) or AAV vectors. Many studies characterized antibody responses to AAV, but human IgG subclass responses to AAV have not been previously described. In this study, IgG subclass responses were examined in serum samples of normal human subjects exposed to wild-type AAV, subjects injected intramuscularly with AAV vectors and subjects injected intravascularly with AAV vectors. A diversity of IgG subclass responses to AAV capsid were found in different subjects. IgG1 was found to be the dominant response. IgG2, IgG3, and IgG4 responses were also observed in most normal human subjects; IgG2 and IgG3 each represented the major fraction of total anti-AAV capsid IgG in a subset of normal donors. Subjects exposed to AAV vectors showed IgG responses to AAV capsid of all four IgG subclasses. IgG responses to AAV capsid in clinical trial subjects were inversely proportional to the level of pre-existing anti-AAV antibody and independent of the vector dose. The high levels of anti-AAV capsid IgG1 can mask differences in IgG2, IgG3, and IgG4 responses that were observed in this study. Analysis of IgG subclass distribution of anti-AAV capsid antibodies indicates a complex, non-uniform pattern of responses to this viral antigen. J. Med. Virol. 81:65-74, 2009. (c) 2008 Wiley-Liss, Inc.

High-throughput screening and biophysical interrogation of hepatotropic AAV.

We set out to analyze the fundamental biological differences between AAV2 and AAV8 that may contribute to their different performances in vivo. High-throughput protein interaction screens were used to identify binding partners for each serotype. Of the >8,000 proteins probed, 115 and 134 proteins were identified that interact with AAV2 and AAV8, respectively. Notably, 76 of these protein interactions were shared between the two serotypes. CDK2/cyclinA kinase was identified as a binding partner for both serotypes in the screen. Subsequent analysis confirmed direct binding of CDK2/cyclinA by AAV2 and AAV8. Inhibition of CDK2/cyclinA resulted in increased levels of vector transduction. Biophysical study of vector particle stability and genome uncoating demonstrated slightly greater thermostability for AAV8 than for AAV2. Heat-induced genome uncoating occurred at the same temperature as particle degradation, suggesting that these two processes may be intrinsically related for adeno-associated virus (AAV). Together, these analyses provide insight into commonalities and divergences in the biology of functionally distinct hepatotropic AAV serotypes.

AAV capsid CD8+ T-cell epitopes are highly conserved across AAV serotypes.

Adeno-associated virus (AAV) has become one of the most promising vectors in gene transfer in the last 10 years with successful translation to clinical trials in humans and even market approval for a first gene therapy product in Europe. Administration to humans, however, revealed that adaptive immune responses against the vector capsid can present an obstacle to sustained transgene expression due to the activation and expansion of capsid-specific T cells. The limited number of peripheral blood mononuclear cells (PBMCs) obtained from samples within clinical trials allows for little more than monitoring of T-cell responses. We were able to identify immunodominant major histocompatibility complex (MHC) class I epitopes for common human leukocyte antigen (HLA) types by using spleens isolated from subjects undergoing splenectomy for non-malignant indications as a source of large numbers of lymphocytes and restimulating them with single AAV capsid peptides in vitro. Further experiments confirmed that these epitopes are naturally processed and functionally relevant. The design of more effective and less immunogenic AAV vectors, and precise immune monitoring of vector-infused subjects, are facilitated by these findings.

Persistent expression of hF.IX After tolerance induction by in utero or neonatal administration of AAV-1-F.IX in hemophilia B mice.

The major complication associated with protein replacement therapy currently used in the treatment of hemophilia B (HB) is the development of antibodies to the infused human Factor IX (hF.IX). We hypothesized that vector-mediated expression of hF.IX, either at a prenatal stage or early in life may lead to tolerance to hF.IX and long-term transgene expression. Fetal, neonatal, and adult F.IX-deficient mice were injected with AAV-1-hF.IX, and the hF.IX levels as well as antibodies to hF.IX in the circulation were assayed. In utero injection followed by postnatal re-administration of adeno-associated virus 1 (AAV-1) vector achieved persistent expression of hF.IX in all animals, with no cellular or humoral immune response to F.IX. Similar results were seen after initial injection in neonatal mice followed by re-administration, whereas all mice injected at the adult stage developed antibodies to hF.IX. In contrast, after administration of AAV-2-hF.IX in the neonatal period, antibodies to hF.IX were formed in all the injected animals. We conclude that in utero or neonatal-stage injection of AAV-1-hF.IX can lead to long-term expression and absence of immune response. The differences in immune response between the AAV-1 and AAV-2 groups suggests that tolerance may be related to differences in bio-distribution, timing of expression, and/or the initial levels of hF.IX expression. This supports the concept of a narrow "window of opportunity" for tolerance induction.

Pre-existing AAV capsid-specific CD8+ T cells are unable to eliminate AAV-transduced hepatocytes.

The goal of these studies was to test whether adeno-associated virus (AAV) capsid-specific CD8(+) T cells cause loss of hepatic AAV-mediated gene expression in experimental animals. Mice immunized with adenoviral vectors expressing AAV capsid or with AAV vectors developed CD8(+) T cells in blood, lymphatic tissues, and liver to epitopes shared between AAV2 and AAV8, and serotype-specific neutralizing antibodies. At the height of the T cells' effector phase, mice were infused with a heterologous AAV vector expressing human factor IX under a hepatocyte-specific promoter. Despite the presence of lytic CD8(+) T cells in the liver, hepatic Factor IX expression was sustained and comparable in AAV-preimmune and naïve animals. These results suggest that, in mice, pre-existing CD8(+) T cells to AAV capsid do not affect the longevity of AAV-mediated hepatic gene transfer. These results are in contrast to the outcome of a recent gene therapy trial of hemophilia B patients who were treated by hepatic gene transfer of AAV2 vectors expressing Factor IX. The loss of Factor IX expression, accompanied by a rise in liver enzymes and detectable frequencies of circulating AAV capsid-specific T cells, suggested T-cell-mediated destruction of transduced hepatocytes following reactivation of AAV-specific T cells upon AAV transfer.

Modulation of tolerance to the transgene product in a nonhuman primate model of AAV-mediated gene transfer to liver.

Adeno-associated virus (AAV)-mediated gene transfer of factor IX (F.IX) to the liver results in long-term expression of transgene in experimental animals, but only short-term expression in humans. Loss of F.IX expression is likely due to a cytotoxic immune response to the AAV capsid, which results in clearance of transduced hepatocytes. We used a nonhuman primate model to assess the safety of AAV gene transfer coupled with an anti-T-cell regimen designed to block this immune response. Administration of a 3-drug regimen consisting of mycophenolate mofetil (MMF), sirolimus, and the anti-IL-2 receptor antibody daclizumab consistently resulted in formation of inhibitory antibodies to human F.IX following hepatic artery administration of an AAV-hF.IX vector, whereas a 2-drug regimen consisting only of MMF and sirolimus did not. Administration of daclizumab was accompanied by a dramatic drop in the population of CD4(+)CD25(+)FoxP3(+) regulatory T cells (Tregs). We conclude that choice of immunosuppression (IS) regimen can modulate immune responses to the transgene product upon hepatic gene transfer in subjects not fully tolerant; and that induction of transgene tolerance may depend on a population of antigen-specific Tregs.

The inhibitory effects of anticoagulation on in vivo gene transfer by adeno-associated viral or adenoviral vectors.

Identifying factors that influence gene transfer efficacy is critical for a successful gene-based clinical study. Here we demonstrate that in vivo AAV-2-mediated gene transfer is efficiently inhibited by unfractionated heparin, but not by a heparin preparation containing mainly low-molecular-weight forms (LMWH). Surprisingly, inhibitors of thrombin or factor Xa (F.Xa) significantly reduced AAV-2 transduction in a dose-dependent manner. These effects were independent of the vector promoter, transgene, or strain of mice. Expression by alternate AAV serotypes 5 and 8 was not affected by anticoagulant drugs, which suggests an AAV-2-specific effect. Moreover, AAV-2-mediated gene expression was diminished in mice with deficiency in thrombin generation (factor IX deficiency) and enhanced in mice with procoagulant phenotype due to factor V Leiden. In addition, inhibitors of F.Xa diminished adenovirus-mediated gene expression. These results demonstrated that coagulation activity itself is critical to ensure optimal viral vector transduction. Since intravascular delivery of vectors often requires the use of anticoagulants, the use of LMWH appears to be safe. These observations are of relevance for approaches using AAV-2 or adenoviral vectors, especially in early phase studies designed to identify the minimum therapeutic doses.

Novel hemophilia B mouse models exhibiting a range of mutations in the Factor IX gene.

Animal models have been critical to the development of novel therapeutics in hemophilia. A deficiency of current murine models of hemophilia B is that they are all due to gene deletions, a type of mutation that is relatively rare in the human hemophilia population. We generated mice with a range of mutations in the Factor IX (F.IX) gene; these more faithfully reflect the types of mutations that cause disease in the human population. Transgenic mice expressing either wild-type human F.IX (hF.IX), or F.IX variants with premature translation termination codons, or missense mutations, under the control of the murine transthyretin promoter, were generated and crossed with mice carrying a large deletion of the murine F.IX gene. Gene copy number, F.IX transcript levels in the liver, intrahepatocyte protein expression, and circulating levels of F.IX protein in the mice were determined and compared with data generated by transient transfection assays using the same F.IX variants. Mice were injected with a viral vector expressing hF.IX and displayed a range of immune responses to the transgene product, depending on the underlying mutation. These new mouse models faithfully mimic the mutations causing human disease, and will prove useful for testing novel therapies for hemophilia.