Hemophilia A and FOXP3 Transgenic Mouse. Models. Benjamin R. Harmeling,1 Steven Ziegler,2 Troy Torgerson,1,3. Liping Chen,1 Hans D. Ochs,1,3 David J.
VECTOR AND TRANSGENE IMMUNOLOGY 77. TLR and Myd88 Signaling Pathway Is Required for the Full Induction of Host Adaptive Immune Responses to Adenoviral Vectors and Transgenes Zhe Zhang, Yan Zhi, Joanita M. Figueredo, Roberto Calcedo, James R. Miller, Guang-Ping Gao, James M. Wilson. 1 Gene Therapy Program, Department of Pathology and Laboratory Medicine, University of Pennsylvania, Philadelphia, PA. Adaptive immune responses including T cell and B cell responses to both backbones of adenoviral vectors as well as the transgenes were studied in TLR and MyD88 signaling deficient mice. C57/B6, MyD88 knockout (KO) or TLR2 (KO) mice were intramuscularly injected with Ad5 carrying SARS membrane protein Spike, the frequency of INF-γ producing CD8+ T cells in response to a dominant CD8+ epitope of Spike was significantly reduced in both KO mice compared to C57B6 mice (INF-γ intracellular staining assay); T cell proliferation in response to Ad5 particles (H3 Thymidine incorporation assay) was also attenuated as observed in MyD88 KO mice. However, in C3H/HEN and its congenic TLR4-deficient C3H/HEJ mice who were intramuscularly injected Ad5 carrying membrane protein GP of Ebola virus, GP-specific CD8+ T cell response remained the same, suggesting selective TLRs such as TLR2 but not TLR4 are required for the full induction of T cell responses. At day 14 after Ad5-LacZ infusion, total immunoglobulins in the serum against the vector and transgene lacZ were all significantly attenuated in the KO mice (ELISA). Antibody isotype profile of IgG1, IgG2a, IgG2b, IgG3, IgM, and IgA shows that the production of selective but not all isotypes in MyD88 KO serum were reduced (ELISA). The blocking titer against Ad5-LacZ in MyD88 serum was diminished compared to that of C57B6 (Neutralizing antibody assay). Finally, LacZ staining on liver sections showed a prolonged transgene expression in MyD88 (100% at D28; 90% at D50) compared to C57B6 (1% at D28; 0% at D50) mice. Our data suggest that MyD88-dependent TLR signaling pathway is involved in the host induction of adaptive immune responses. Further understanding of the pathway may provide valuable opportunity in fine tuning the induction of immune responses in adenoviral vector-based vaccine applications.
78. Regulation of Immune Responses Against Factor VIII Following Nonviral Gene Transfer in Hemophilia A and FOXP3 Transgenic Mouse Models Benjamin R. Harmeling,1 Steven Ziegler,2 Troy Torgerson,1,3 Liping Chen,1 Hans D. Ochs,1,3 David J. Rawlings,1,3 Carol H. Miao.1,3 1 Pediatrics, Childrens Hospital and Regional Medical Center, Seattle, WA; 2Immunology, Benaroya Research Institute, Seattle, WA; 3Department of Pediatrics, University of Washington, Seattle, WA. The success of naked DNA gene transfer as a novel approach to the treatment of hemophilia A has been limited by the formation of inhibitory antibodies against factor VIII (FVIII). Approximately 25-30% of hemophilia A patients produce inhibitory antibodies in response to FVIII protein replacement therapy. Potential gene therapy techniques used to treat hemophilia A have also resulted in a significant humoral immune response in murine models (Ye et al., 2004). The critical role of CD4+CD25+ regulatory T (Treg) cells in controlling autoimmune and alloimmune responses has recently been recognized. We hypothesize that regulatory T cells could modulate the formation of inhibitory antibodies against FVIII. The development of CD4+CD25+ Treg cells is controlled by a transcription factor encoded by FOXP3. Increased FOXP3 expression directly Molecular Therapy Volume 13, Supplement 1, May 2006 Copyright The American Society of Gene Therapy
correlates with the suppressive activity of Treg cells. An adoptive transfer of antigen-specific FOXP3 Treg cells could potentially suppress the FVIII-specific immune response in hemophilia A patients. In order to evaluate whether FOXP3 expressing Treg cells can modulate the antigen-specific immune response in the Hemophilia A (HemA) mouse model, we have developed transgenic mice of a HemA phenotype with overexpressed FOXP3 (HemA/ FOXP3-Tg). The mice were created by crossbreeding HemA mice and transgenic FOXP3 (FOXP3-Tg) mice to produce constitutively high levels of human FOXP3 in addition to their endogenous murine Foxp3 encoded protein. Four different groups of genetically altered C57BL/6 mouse strains including HemA, FOXP3-Tg, HemA/ FOXP3-Tg, and wild-type control, were injected intravenously with a 100 µg dose of FVIII plasmid (pBS-HCRHPI-FVIIIA) using a hydrodynamics-based gene delivery method. Clotting time of treated mouse plasma was significantly shortened one-day post treatment in all cohorts, as evaluated by a modified APTT assay, indicating similar responses to the plasmid treatment. Approximately two weeks following infusion, HemA mice developed a high-titer FVIII inhibitory antibodies and a concomitant increase in APTT clotting time. Plasmid-injected wild-type mice responded with a considerably lower titer of inhibitory antibodies, likely due to the presence of tolerizing endogenous murine FVIII. In contrast, both groups of FOXP3-Tg mice and HemA/ FOXP3-Tg mice failed to develop FVIII specific inhibitory antibodies, and retained a shorter APTT clotting time. These results strongly suggest that Treg cells induced by FOXP3 gene expression can regulate the formation of inhibitory antibodies against human FVIII. We are currently assessing the potential of adoptively transferred antigen-specific Treg cells to modulate this inhibitory immune response to gene transfer in HemA mice. If transferred Treg cells from FVIII exposed HemA/FOXP3Tg mice are as successful in controlling immune responses in recipients, then FOXP3 expressing FVIII-specific Treg cells could provide an effective therapeutic approach of regulating FVIII-specific inhibitory antibodies.
79. Systemic Administration of Lentiviral Vectors Triggers Innate Host Responses Brian D. Brown,1 Ehud Hauben,1 Giovanni Sitia,3 Anna Zingale,1 Lucia Sergi Sergi,1 Lucca Guidotti,3 Maria Grazia Roncarolo,1,2 Luigi Naldini.1,2 1 Telethon Institute for Gene Therapy, San Raffaele Scientific Institute, Milano, Italy; 2San Raffaele Vita Salute University, San Raffaele Scientific Institute, Milano, Italy; 3Immunopathogenesis of Liver Infections Unit, San Raffaele Scientific Institute, Milano, Italy. The initial interaction with host tissues of systemically administered viral vectors is a critical event for stable gene transfer. It is now well established that injection of adenoviral vectors leads to an inflammatory response, mediated by elevations in IL-6 and TNFα. This response triggers the adaptive immune response, and leads to anti-vector and transgene immunity. Systemic gene transfer by lentiviral vectors (LV) can lead to antitransgene immunity, however, proinflammatory cytokines within the circulation have not been observed, and in vitro transduction of monocyte-derived dendritic cells (DC) does not trigger DC maturation. Thus, it remains unclear whether the induction of antitransgene immunity is related to an adjuvant affect of the vector itself. In recent years there has been a growing understanding of how wildtype HIV, a single-stranded RNA (ssRNA) virus, interacts with cells of the innate immune system. However, little is known about the host response to pseudotyped LVs, which do not contain HIV accessory factors and have a broader cell tropism then the wildtype virus. Here, we set out to gain a better comprehension of host/LV interactions following systemic vector delivery. S33
OLIGONUCLEOTIDE THERAPIES: NON-VIRAL TARGETS C57BL/6 and Balb/c mice were treated with (VSV)LV by intravenous injection. The content of viral RNA and vector provirus within the liver and spleen was monitored by real-time PCR. By 4 hours post-injection the majority of vector uptake and reverse transcription had occurred. RNase Protection Assay revealed that concomitant to vector entry there was a >10-fold induction of type I interferons (IFNs), TNFα, and the chemokine MIG in both the liver and spleen. This response was not due to contaminants in the vector preparation, as heat-inactivated and bald (envelope negative) LV particles did not result in innate activation. Cytokine expression following LV delivery was transient and largely subsided within 72 hours. However, clearance of vector genomes had already begun to occur by 24 hours. This effect may have been due to the antiviral activity of the IFNs, a possibility we will validate in IFN receptor knock-out mice. We next set out to determine the factors involved in mediating the response to LV. Initially we sought to identify if particular DC subsets were involved. Indeed, in vitro studies revealed that plasmacytoid, but not myeloid DCs became activated to produce IFNs following exposure to LV. Similar findings were also seen with human DCs. Moreover, experiments with the bald LV suggested that endosome uptake was required to induce the cytokine response. Taken together, these results suggest a role for toll-like receptor (TLR)-7, a TLR that recognizes ssRNA and is found predominantly within the endosomes of plasmacytoid DCs, in LV-mediated innate immune activation. Studies are now underway to determine whether antagonists of TLR-7 can prevent the innate response to LV, and enable stable gene transfer in the absence of adaptive immunity.
80. Pre-Existing Cytotoxic T-Lymphocyte (CTL) Responses to the AAV CAP Protein Do Not Interfere with AAV Transgene Expression in Mice William M. Siders,1 Johanne Kaplan,1 Michael Lukason,2 Jacqueline Shields,1 Lisa Woodworth,1 Sam Wadsworth,2 Abraham Scaria.2 1 Immunotherapy Research Group, Genzyme Corporation, Framingham, MA; 2Molecular Biololgy, Genzyme Corporation, Framingham, MA. The use of adeno-associated viral (AAV) vectors as a delivery method for gene replacement strategies is currently being explored in several clinical indications. AAV vectors do not express any viral proteins suggesting that they represent a less immunogenic delivery system. However, recent data have suggested that administration of AAV vectors may actually result in the stimulation of T-cell responses that result in a loss of transgene expression and may limit the utility of these vectors. By using computer prediction algorithms, we identified several AAV peptides as putative MHC I binders. Immunization of mice with peptides predicted to have the highest binding affinities did not result in a detectable cytotoxic T lymphocyte (CTL) response. In addition, intravenous administration of increasing doses of AAV serotype 2 vector also failed to induce a measurable CTL response. In contrast, immunization with either a plasmid or an adenoviral vector encoding the full length AAV2 CAP protein, to mimic exposure to wild type AAV, generated a potent CTL response. A pre-existing CTL response is likely to be present in the human population and C57Bl/6 IgH6 knock-out (KO) mice were used to determine whether such CTLs could affect AAV transduction and subsequent transgene expression. When immunized with two doses of CAP plasmid, the IgH6 KO mice mounted a robust CTL response without the generation of AAV neutralizing antibodies. The mice were then injected with an AAV2 vector encoding the alpha-galactosidase (a-gal) protein. AAV transduction was unimpeded and the serum levels of a-gal protein as well as the longevity of transgene expression were comparable to those observed in mice pre-immunized with empty plasmid as a control. In contrast, S34
wild-type C57Bl/6 mice pre-immunized with CAP plasmid, which developed both CTLs and neutralizing antibodies against AAV, did not show any evidence of a-gal expression after the injection of vector. Taken together, results of these experiments suggest that pre-existing AAV-specific CTLs may have little or no effect on AAV transduction or transgene expression and confirm that neutralizing antibodies significantly interfere with transduction by AAV vector. I am an employee of Genzyme Corporation.
OLIGONUCLEOTIDE THERAPIES: NON-VIRAL TARGETS 81. Hippocampal Gene Knockdown of BACE1 by Lentiviral Gene Transfer of siRNA Reduces Amyloid Pathology and Improves Memory Performance in APP Transgenic Mice Oded Singer,1 Robert A. Marr,1 Edward Rockenstein,2 Leslie Crow,2 Fred H. Gage,1 Inder M. Verma,1 Eliezer Masliah.2 1 Laboratory of Genetics, The Salk Institute for Biological Studies, La Jolla, CA; 2Department of Neurosciences, The University of California San Diego, La Jolla, CA. The neurodegenerative process in Alzheimer’s disease (AD) is associated with increased β-secretase (BACE1) activity that results in progressive accumulation of amyloid precursor protein (APP) Cterminal fragments (CTFs) and amyloid-β (Aβ). Therefore, developing strategies to inactivate BACE1 might present an important approach toward the treatment of AD. To assess this possibility in vivo, APP transgenic mice that develop some aspects of AD neuropathology were treated with a lentiviral vector (LV) expressing small interfering RNA (siRNA) targeting BACE1. The LV-siBACE1 vector was expressed at high levels in HEK293T cells and reduced both BACE1 expression and production of Aβ1-42 in vitro. Onemonth post intra-hippocampal delivery of the LV-siBACE1 vector in APP transgenic mice, levels of BACE1 expression were significantly reduced at the site of the injection. This was accompanied by reduced deposition of Aβ, reduced intracellular CTFs and amelioration of the neurodegenerative alterations (dendritic / synaptic density) in the hippocampus as evaluated by confocal microscopy. Furthermore, the neuroprotective effects of LVsiBACE1 were associated with improved spatial learning and memory, tested by a Morris water maze. Our results suggest that lentiviral vector delivery of BACE1 siRNA can specifically reduce the cleavage rate of APP and neurodegeneration in vivo and indicates the potential therapeutic value of this approach for treating AD. We will discuss the possible outcome of long-term siRNA expression on the progression of AD in APP mice.
82. Cftr Gene Targeting in Murine ES Cells Mediated by the SFHR Technique Federica Sangiuolo,1 Maria Favia,2 Lorenzo Guerra,2 Antonio Filareto,1 Maria Lucia Scaldaferri,3 Rosa Caroppo,2 Paola Spitalieri,1 Ruggiero Mango,1 Emanuela Bruscia,4 Dieter Gruenert,5 Valeria Casavola,2 Massimo De Felici,3 Giuseppe Novelli.1 1 Department of Biopathology, Tor Vergata University, School of Medicine, Rome, Italy; 2Department of General and Environmental Physiology and Cell Biology, University of Bari, Bari, Italy; 3Department of Public Health, Tor Vergata University, School of Medicine, Rome, Italy; 4Laboratory Medicine, Yale University, New Haven, CT; 5California Pacific Medical Center Research Institute, San Francisco, CA. Small Fragment Homologous Recombination (SFHR)-mediated targeting is a gene therapy strategy where a specific genomic locus is modified through a target exchange between a small DNA fragment (SDF) and genomic DNA. Here we demonstrate that SFHR can stably Molecular Therapy Volume 13, Supplement 1, May 2006 Copyright The American Society of Gene Therapy
